•^m^w^tmm^^ms^sm^.
""
••• ""^^ ^^^--W^^~ r^^-^^i^\ 'I ^^ '-"'"fi?-^^ B^% -^^fc I %~ JT ^^^ I f^ '^"J^'^k" •" •"*• 1^^-^^ '"^
l||v^
l
""
".Vi •''":,V •' .i' •.•'•• -'-v. ' '';'•; -'::'t' t... '---.' ';. '•••' !.'•'-• :!|"': ' ::'' • 1 .•'• "";: •• • !-/':::'..'j ..'':•! l!'- "•.•'k""'- !"-"''' "•".-. ••'-'• :'•'> . :;' .'yj' • '•'.-..'•."•'"" "''•''•..-"•::^''^:''^ 'a.'1. •".--•• •ivV''^'-.'! t'1'' •'•':-;.";'l-"!t^.-.!^i:-^^ - :r,..^tL
fir""
S^i^liirW^ ^""^'"S ii1/ Jtl*& W.jiJL.-/.'' .I'tA hfeiv. iaSKi«T.
-------
-------
Economic Analysis for the Proposed Revisions to the
National Pollutant Discharge Elimination System Regulation
and the Effluent Guidelines for
Concentrated Animal Feeding Operations
Carol M. Browner
Administrator
I. Charles Fox
Assistant Administrator, Office of Water
Sheila E. Frace
Director, Engineering and Analysis Division
Janet Goodwin
Project Manager
Renee Selinsky Johnson
Economist
Engineering and Analysis Division
Office of Science and Technology
U.S. Environmental Protection Agency
Washington, D.C. 20460
January 2001
-------
-------
ACKNOWLEDGMENTS AND DISCLAIMER
This document was prepared by the Office of Science and Technology with the support of
Eastern Research Group Incorporated.
Neither the United States government nor any of its employees, contractors, subcontractors, or
other employees makes any warranty, expressed or implied, or assumes any legal liability or
responsibility for any third party's use of, or the results of such use of, any information,
apparatus, product, or process discussed in this report, or represents that its use by such a third
party would not infringe on privately owned rights.
-------
-------
CONTENTS
Page
EXECUTIVE SUMMARY
ES.l Introduction ES-1 .
ES.2 Data and Methodology ES-1
ES.2.1 DataSources ..... ES-1
ES.2.2 Methodology -. ES-2
ES.3 Regulated Community ES-3
ES.4 Annual Incremental Costs : ES-5
ES.5 Economic Impacts ES-7
ES.5.1 CAFO Impacts ... ES-7
ES.5.2 Processor Impacts ...... ES-9
ES.5.3 Market Impacts .. ES-10
ES.6 Other Regulatory Requirements ES-10
ES.6.1 Small Business Analysis : ES-10
ES.6.2 Cost Benefit Analysis ES-12
ES.7 Other Information „.-'..- ES-13
SECTION ONE INTRODUCTION
1.1 Existing Regulatory Framework 1-1
1.1.1 NPDES Permit Regulation of CAFOs 1-2
1.1.2 Effluent Limitations Guidelines for Feedlots , 1-4 .
1.1.3 Industries Affected .by the Proposed CAFO Regulations 1-4
• 1.1.4 Reasons Why EPA is Revising the Existing CAFO Regulations ...1-6
1.2 Overview of Sources of Data 1-7
1.3 Organization of the Report ; 1-9
-------
SECTION TWO PROFILE OF THE LIVESTOCK AND POULTRY INDUSTRIES
2.1 Recent Trends in the Livestock and Poultry Industries 2-1
2.1.1 Increased Livestock and Poultry Production 2-2
2.1.2 Fewer, Larger, and More Industrialized Livestock and
Poultry Operations 2-3
2.1.3 Geographic Shifts in Animal Production 2-6
2.1.4 Increased Farmer-Processor Linkages 2-8
2.2 Characteristics of Animal Confinement Operations That May be Affected
by the Proposed CAFO Regulations 2-10
2.2.1 Identification and Number of Affected CAFOs 2-11
2.2.1.1 All Livestock and Poultry Operations 2-11
2.2.1.2 Animal Confinement Operations 2-13
2.2.1.3 CAFOs Subject to the Proposed Regulations 2-15
2.2.2 Financial Characteristics of Livestock and Poultry Farms 2-18
2.2.3 Manure and Manure Nutrients Generated Annually at CAFOs ... 2-22
2.3 Industrial Organization of Livestock and Poultry Industries 2-23
2.3.1 Contracting in Animal Agriculture 2-24
. 2.3.2 Degree of Affiliation between CAFOs and Processors 2-27
2.4 Characteristics of Processing Firms That May Be Affected by the
Proposed CAFO Regulations ... 2-32
2.4.1 Identification and Number of Potential Co-Permittees 2-32
2.4.1.1 All Livestock and Poultry Processors 2-32
2.4.1.2 Sectors with Potential for "Substantial
Operational Control" 2-35
2.4.1.3 Identification of Potential Co-Permittees based on
Facility Type and Size 2-36
2.4.2 Financial Characteristics of the Livestock and Poultry
Processing Sector 2-39
2.5 Other Market Characteristics of the Livestock and Poultry Industries ..... 2-44
2.5.1 Annual Marketing Receipts 2-44
2.5.1.1 Total Farm Receipts from Marketings 2-44
2.5.1.2 Total Manufacturing Value of Shipments — 2-45
2.5.2 Supply and Demand Conditions for Livestock and Poultry
Products 1 • • 2-45
2.5.2.1 Farm Production 2-46
2.5.2.2 Domestic Demand 2-46
2.5.2.3 Imports and Exports 2-48
2.5.3 Industry Employment 2-49
11
-------
2.5.3.1 Total Farm Employment 2-49
2.5.3.2 Total Manufacturing Employment 2-51
SECTION THREE THE PROPOSED CAFO REGULATIONS
3.1 Summary of the Proposed Revisions 3-1
3.1.1 Revised Scope Requirements under the Proposed Regulations .... 3-1
3.1.2 Other Revised Requirements under the Proposed Regulations .... 3-4
3.2 Summary of ELG Options and NPDES Scenarios Considered by EPA 3-7
3.2.1 Effluent Guidelines Options 3-7
3.2.2 NPDES Scenarios 3-10
SECTION FOUR METHODOLOGY FOR ESTIMATING COMPLIANCE
COSTS AND ECONOMIC IMPACTS
4.1 Annual Compliance Costs -. 4-1
4.1.1 Baseline Compliance Assumption 4-2
4.1.2 Method for Estimating CAFO Compliance Costs 4-2
4.1.2.1 Compliance Costs to CAFO Operators 4-2
4.1.2.2 Compliance Costs to Recipients of CAFO Manure ....4-5
4.1.3 Cost Annualization Methodology 4-6
4.2 CAFO Analysis 4-8
• 4;2.1 Overview of the Representative CAFO Approach 4-8
4.2.2 Construction of EPA's Model CAFOs 4-10
4.2.2.1 Livestock and Poultry Sectors 4-11
4.2.2.2 Farm Producing Regions 4-11
4.2.2.3 Facility Size 4-12.
4.2.3 Sources of Data for EPA's Model CAFOs 4-14
4.2.3.1 Overview of ARMS Financial Data 4-15
4.2.3.2 Special Compilation of Representative ARMS Data .. 4-20
4.2.3.3 ARMS Data Aggregations for Model CAFOs 4-22
4.2.4 Development of the Financial Characterization of Model CAFOs . 4-27
4.2.4.1 Key Financial Variables : 4-28
4.2.4.2 Calculation of Financial Variables on a
Per-Animal Basis 4-29
4.2.4.3 Calculation'of Present Value of Net Cash Flow 4-32
4.2.4.4 USD A'sDebt-to-Asset Ratios for Model CAFOs 4-36
4.2.4.5 Construction of Representative Model CAFOs 4-37
4.2.5 Criteria for Assessing Regulatory Impacts . . . . 4-44
ru
-------
4.2.6 Cost Passthrough 4-50
4.2.6.1 Methodology for Computing Cost Passthrough 4-53
4.2.6.2 Three CPT Scenarios 4-55
4.2.7 Potential Cost Offsets . 4-55
4.3 Processor Level Analysis 4-56
4.3.1 Overview of Methodology 4-57
4.3.2 Sources of Data 4-58
4.4 Market Level Analysis 4-59
4.4.1 Overview of Methodology .4-60
4.4.1.1 Market Model 4-61
4.4.1.2 Input-Output Analysis .* 4-62
4.4.2 Sources of Data and Parameters 4-63
4.4.2.1 Market-Model Data 4-63
4.4.2.2 Input-Output Model Data 4-65
4.4.3 Criteria for Assessing Regulatory Impacts .4-67
SECTION BTVE TOTAL COSTS AND ECONOMIC IMPACTS OF THE
PROPOSED CAFO REGULATIONS (ALL SUBCATEGORIES)
5.1 Annual Compliance Costs of the Proposed CAFO Regulations '.. 5-2
5.1.1 Annual Costs under Two-Tier and Three-Tier Structures 5-2
5.1.2 Costs to CAFOs of Alternative Regulatory Options and Scenarios .5-7
5.1.2.1 Annual Costs of the Alternative ELG Options ....... 5-7
5.1.2.2 Annual Costs of the Alternative NPDES Scenarios 5-8
5.2 CAFO Impacts. 5-10
5.2.1 Baseline Financial Health of Model CAFOs 5-11
5.2.2 Post-compliance Impacts to Existing Operations (BAT Analysis) . 5-11
5.2.2.1 Impacts under the Two-Tier and Three-Tier Structures .. 5-11
5.2.2.2 Impacts under Other Regulatory Alternatives ,.... 5-18
5.2.3 Post-compliance Impacts to Offsite Recipients of CAFO Manure . 5-20
5.2.4 Post-compliance Impacts to New Operations (NSPS Analysis) .. . 5-22
5.2.4.1 Impacts of the NSPS Options on the Beef and Dairy
Subcategories 5-23
5.2.4.2 Impacts of the NSPS Options on the Swine, Veal, and
Poultry Subcategories 5-25
5.3 Processor Impacts 5-26
5.4 Market Impacts 5-27
IV
-------
5.4.1 Changes in Commodity Price and Quantity Production . 5-28
5J4.2 Changes in Total National Employment 5-30
5.4.3 Changes in Total National Economic Output 5-32
5.4.4 Other Market Impacts ... , 5-34
5.4.4.1 Regional Employment .......r 5^34
5.4.4.2 International Trade 5-35
SECTION SIX SUMMARY OF ECONOMIC IMPACTS:
POULTRY SUBCATEGORIES
I
6.1 Profile of the Poultry Production Industry 6-1
6.1.1 Industry Definition 6-1
6.1.2 Overview of the Poultry Industry 6-2
6.1.2.1 Trends in the Number and Size 6-2
6.1.2.2 Geographic Distribution 6-6
6.1.2.3 Supply and Demand Conditions 6-10
6.1.2.4 FarmPriceTrends 6-12
6.1.3 Financial Characteristics of Poultry Operations . 6-14
6.1.3.1 Overview of Financial Characteristics 6-14
6.1.3.2 Income Statement and Balance Sheet Information 6-16
6.1.3.3 Baseline Conditions for Poultry Operations 6-19
6.2 Profile of Poultry Processing Sectors 6-30
*
6.3 CAFO Analysis :.... . 6-32
6.3.1 Overview of Cost Input Data 6-32
6.3.2 Estimates of National Annual Compliance Costs 6-35
6.3.3 Analysis of CAFO Financial Impacts 6-36
6.4 Processor Analysis • 6-44
6.5 Market Analysis • 6-46
SECTION SEVEN SUMMARY OF ECONOMIC IMPACTS:
HOG SUBCATEGORY
7.1 Profile of the Hog Production Sectors 7-1
7.1.1 Industry Definition 7-1
7.1.2 Overview of the Hog Industry 7-2
7.1.2.1 Trends in the Number and Size 7-2
7.1.2.2 Geographic Distribution 7-4
7.1.2.3 Supply and Demand Conditions 7-7
7.1.2.4 Farm Price Trends 7-7
-------
7.1.3 Financial Characteristics of Hog Operations 7-10
7.1.3.1 Overview of Financial Characteristics 7-10
7.1.3.2 Income Statement and Balance Sheet Information ...... 7-10
7.1.3.3 Baseline Conditions for Hog Operations 7-15
7.2 Profile of the Hog Processing Sectors 7-19
7.3 CAFO Analysis .... 7-21
7.3.1 Overview of Cost Input Data 7-22
7.3.2 Estimates of National Annual Compliance Costs 7-25
7.3.3 Analysis of CAFO Financial Impacts 7-27
7.4 Processor Analysis • • • • 7-32
7.5 Market Analysis • 7-33
SECTION EIGHT SUMMARY OF ECONOMIC IMPACTS:
BEEF AND DAIRY SUBCATEGORIES
8.1 Profile of the Beef and Dairy Production Sectors 8-1
8.1.1 Industry Definition 8-1
8.1.2 Overview of the Beef and Dairy Industry 8-3
8.1.2.1 Trends in the Number and Size , 8-4
8.1.2.2 Geographic Distribution • • 8-8
• • 8.1.2.3 Supply and Demand Conditions 8-11
8.1.2.4 Farm Price Trends 8-13
8.1.3 Financial Data Characteristics of Beef and Dairy Operations 8-15
8.1.3.1 Overview of Financial Characteristics 8-15
8.1.3.2 Income Statement and Balance Sheet Information 8-17
8.1.3.3 Baseline Conditions for Cattle and Dairy Operations 8-24
8.2 Profile of Beef and Dairy Processing Sectors 8-32
8.2.1 Structure of the Red Meat (Beef) Industry 8-33
8.2.2 Structure of the Milk and Dairy Foods Industry 8-36
8.3 CAFO Analysis 8-37
8.3.1 Overview of Cost Input Data 8-38
8.3.2 Estimates of National Annual Compliance Costs :... 8-42
8.3.3 Analysis of CAFO Financial Impacts 8-44
8.4 Processor Analysis 8-48
8.5 Market Analysis 8-48
vi
-------
SECTION NINE INITIAL REGULATORY FLEXIBILITY ANALYSIS
9.1 The Regulatory Flexibility Act (RFA) As Amended by the Small
Business Regulatory Enforcement Fairness Act (SBREFA) '.. . 9-1
9.2 Initial Assessment 9-1.
9.2.1 Definition of Small CAFO Businesses 9-2
9.2.2 Number of Small Businesses Affected by the Proposed CAFO
Regulations •. •. 9-6
9.2.2.1 Equating SBA Size Standardsiwith Animal Inventory .. 9-6
9.2.2.2 Total Number of Operations that Match SBA
Size Standards 9-8
9.2.2.3 Total Number of Small CAFOs Subject to the
Proposed Regulations 9-9
9.2.3 Results of the Initial Assessment 9-12
9.3 EPA Compliance with RFA Requirements 9-13
9.3.1 Outreach and Small Business Advocacy Review 9-13
9.3.2 EPA's Initial Regulatory Flexibility Analysis ,.. 9-14
9.3.2.1 Reason EPA is Considering the Proposed Rule ...... 9-14
9.3.2.2 Objectives and Legal Basis for the Proposed Rule .... 9-15
9.3.2.3 Description and Estimate of Number of Small
Entities Affected 9-15
9.3.2.4 Description of the Proposed Reporting,
Recordkeeping, and Other Requirements 9-17
9.3.2.5 Identification of Relevant Federal Rules that May
Duplicate, Overlap, or Conflict with the Proposed
Regulations 9-20
9.3.2.6 Significant Regulatory Alternatives 9-20
9.4 EPA's Analysis of Small Business Impacts 9-22
9.4.1 Data and Methodology 9-22
9.4.2 Economic Analysis Results 9-27
SECTION TEN OTHER REGULATORY ANALYSIS REQUIREMENTS
10.1 Additional Administrative and Regulatory 10-1
10.1.1 Requirements of Executive Order 12866 . ,. 10-1
10.1.2 Requirements of the Unfunded Mandates Reform
Act (UMRA) 10-2
10.2 Need for the Regulations 10-3
vu
-------
10.3 Total Social Costs 10-4
10.3.1 Costs to Industry (Regulated CAFOs and Offsite Recipients) 10-6
10.3.2 Costs to the Permitting Authority (States and Federal
Governments) 10-7
10.3.2.1 Total Number of Permits 10-8
10.3.2.2 Administrative Unit Costs .. '. 10-9
10.3.2.3 Total Administrative Costs 10-11
10.3.3 Other Social Costs 10-12
10.4 Pollutant Reductions 10-15
10.5 Benefits Assessment .--,•• 10-15
10.6 Comparison of Cost and Benefits Estimates 10-17
SECTION ELEVEN REFERENCES
vtu
-------
APPENDIX A
APPENDICES
COST ANNUALIZATION MODEL
A.1 Input Data Sources • • A-l
A.I.I Marginal Tax Rate A-2
A. 1.2 Depreciation Method A-5
A.2 Sample Cost Annualization Spreadsheet - - • • A-6
A.3 Annualized Compliance Costs A-10
APPENDIX B MARKET MODEL DESCRIPTION
B.I Introduction and Overview • B-l
B.2 Model Parameters and Data B-6
B.3 Model in Detail '. • B-8
B.3.1 Farm Production Sector B-9
B.3.1.1 Farm Product Imports Equation B-10
B.3.1.2 Farm Product Exports Equation B-l 1
B.3.1.3 Domestic Farm Product Supply Equation B-12
B.3.1.4 Trade-Adjusted Farm Product Supply Equation B-12
B.3.2 The Processing Sector < • • > • • B-13
B.3.3 Retail Product Sector • • B-14
B.3.3.1 . Retail Product Import Equation B-14
B.3.3.2 Retail Product Export Equation B-15
B.3.3.3 Domestic Retail Product Demand Equation . B-15
B.3.3.4 Trade-adjusted Retail Product Demand Equation ... B-16
B.3.4 The Long-Run Market Equilibrium B-16
B.4 Using the Market Model B-17
B.4.1 Measuring Changes in Prices and Quantities B-17
B.4.2 Industry Direct Impacts B-19
B.4.3 Input/Output Analysis • B-19
B.4.3.1 Employment .k B-20
B.4.3.2 National Output .*. B-21
B.5 Glossary of Notation •• B-21
IX
-------
APPENDIX C
SUMMARY OF DEMAND AND SUPPLY ELASTICITY
LITERATURE
APPENDIX D
SENSITIVITY ANALYSIS
D.I CAFO Model Sensitivity Analysis D-l
D.I.I Sales Test Analysis (Lower Livestock Revenue) D-3
D.I.2 Sales Test Analysis (Livestock Revenue Only) D-6
D.1.3 Sales Test Sensitivity (Pre-tax Compliance Cost Assumption) ... D-9
D.I.4 Discounted Cash Flow Analysis (Lower Net Cash Income) D-9
D.I.5 Debt-to-Asset Analysis (Higher Debt-to-Asset Levels) D-12
D.2 Market Model Sensitivity Analysis D-14
D.2.1 Price Elasticities D-17
D.2.2 Prices 'D-19
APPENDIX E COST-EFFECTIVENESS ANALYSIS
E.I Pollutants of Concern E-2
E.I.I Introduction E-2
E.I.2 Pollutant Concentrations in Animal Manure and Wastewater E-4
E.2 Estimated Pollutant Removals , E-6
E.3 Cost-Effectiveness Analysis: Toxic Pollutants E-9
E.3.1 Methodology E-9
E.3.2 Cost-Effectiveness Results E-14
E.3.3 Comparison of Cost-Effectiveness Values with
Promulgated Rules E-18
E.4 Cost-Effectiveness Analysis: Nutrients and Sediments • • • • • E-20
E.4.1 Review of Literature E-21
E.4.2 Cost-Effectiveness Results , - E-24
-------
TABLES AND FIGURES
Table
ES-1 Number of Potential Operations Defined as CAFO's by Select Regulatory
Alternative, 1997 ES-4
ES-2 Annual Pre-tax Cost of Co-Proposed Two-Tier and Three-Tier Structure, $1999 ES-6
ES-3 Annualized Pre-tax Costs for the Alternative NPDES Scenarios ($1999, million) ES-7
ES-4 Number of CAFOs Affected under the Co-Proposed Alternatives
(Zero Cost Passthrough) .. ES-9
ES-5 Results of EPA's Small Business Analysis Under the BAT Option/Two-Tier Structure ES-11
ES-6 Total Annual Social Costs and Monetized Benefits, $1999 ES-12
1-1 Number of Animal Feeding Operations (1997) 1-6
2-1 Number of Livestock and Poultry Operations (Year-end Animal Inventory), 1969-1997 .. 2-4
2-2 USDA's Farm Typology Groups 2-7
2-3 Number of AFOs and Animals On Site, by Size Group, 1997 2-14
2-4 Number of Potential Operations Defined as CAFOs by Select Regulatory
Alternative, 1997 2-15
2-5 Number of Potential Operations Designated as CAFOs by Select Regulatory
Alternative, 1997 2-18
2-6 Financial Performance by Farm Typology Group, All Crop and Livestock
Production, 1999 2-20
2-7 Manure and Manure Nutrients "Available for Land Application," 1997 2-22
2-8 Contracting Use in the Livestock and Poultry Sectors, 1993 2-28
2-9 Percent of Animals Owned and Not Owned by Farmers by Sales-Based Size
Categories, 1997 2-29
2-10 Processing Industry Statistics by Primary Product Class and Sector, 1997 2-34
2-11 Key Financial.Characteristics of Selected Publicly Held Processing Firms (1996-1998) . . 2-42
2-12 Published Industry Key Financial Characteristics 2-43
2-13 Farm Receipts and Manufacturing Value of Shipments (1992 and 1997) 2-45
2-14 Total Livestock and Poultry Production, Selected Years (1970-1997) 2-46
2-15 Per Capita Demand for Livestock and Poultry Products, Selected Years (1970-1997) .. . 2-47
2-16 Livestock and Poultry Product Trade, Selected Years (1970-1997) 2-47
2-17 Livestock and Poultry Industry Employment by Industry Segment (1997) . 2-49
3-1 Summary Description of Options/Scenarios Considered by EPA 3-8
4-1 Model CAFOs by Sector, Size, and Region (Size Ranges and Average Inventory) 4-13
4-2 EPA-Requested ARMS Data for Model CAFOs by Sector, Size, and Region 4-19
4-3 EPA-derived Per-Animal Financial Data from the 1997 ARMS Data 4-24
4-4 ARMS Data Aggregation for Model CAFOs by Sector, Size, and Region 4-26
4-5 Per-Animal Total Gross Revenue for Model CAFOs, 1997 4-30
4-6 Per-Animal Net Cash Income for Model CAFOs (One Year), 1997 4-31
• 4-7(a) USDA Baseline Projections, Returns Per Unit, 1997-2006 4-34
4-7(b) EPA-derived Equivalent Baseline Projections, Returns Per Animal, 1997-2006 4-34
XI
-------
4-8 Projected Cash Stream (1998-2006) based on USDA Projections of Per-Unit Returns ... 4-35
4-9 Debt-to-Asset Ratios for Model CAFOs, 1997 4-38
4-10 Total Estimated Gross Farm Revenues for Representative Model CAFOs 4-40
4-11 Present Value of Total Net Cash Farm Income for Model CAFOs 4-41
4-12 Comparison of Regional Coverage between EPA's Cost and Financial CAFO Models ... 4-42
4-13 Economic Achievabllity Criteria for the Proposed CAFO Regulations , 4-46
4-14 Estimated CPT Based on Elasticity Estimates Identified in Recent Literature Searches .. 4-54
4-15 1997 Estimated Delivered Cost for the Hog and Poultry Processing Sectors 4-59
4-16 MarketModel Baseline Values (1997) • • • 4-64
4-17 RIMS II Multipliers for Secondary Impact Analysis (Comparison with USFOOD) 4-66
5-1 Annualized Post-Tax Costs, Two-Tier (500 AU), BAT Option/Scenario 4a,
$1997 millions • • • 5'3
5-2 Annualized Post-Tax Costs, Two-Tier (750 AU), BAT Option/Scenario 5,
$1997 millions 5-4
5-3 Annualized Post-Tax Costs, Three-Tier Structure, BAT Option/Scenario 3,
$1997 millions • 5-4
5-4 Annualized Costs to Offsite Recipients of CAFO Manure, $1997 and $1999 millions 5-7
5-5 Annualized Post-Tax Costs for All ELG Options and NPDES Scenarios ($1997, millions) 5-8
5-6 Annualized Post-Tax Costs for All ELG Options ($1997, millions) , 5-9
5-7 Annualized Post-Tax Costs of Options Under Alternative NPDES Scenarios
($1997, millions) • 5-10
5-8 Impacted Operations Under the Two-Tier Structure (BAT Option/Scenario 4a) 5-13
5-9 Impacted Operations Under the Two-Tier Structure (BAT Option/Scenario 5) 5-14
5-10 Impacted Operations Under the Three-Tier Structure (BAT Option/Scenario 3) 5-15
5-11 Number of CAFOs Affected under the Co-Proposed Alternatives by Size
(Zero Cost Passthrough) • 5-17
5-12 Number of CAFOs Adversely Affected under Alternative Options
(Zero Cost Passthrough) • • • 5-19
5-13 Number of CAFOs Adversely Affected under Alternative Options
(Partial Cost Passthrough) 5'20
5-14 Number of CAFOs Adversely Affected under Alternative Scenarios.
(Zero Cost Passthrough) 5-21
5-15 Percent Difference in Costs between Option 8 NSPS and Option 3 BAT,
Beef and Dairy Sectors ,..'.. - 5-24
5-16 Estimated Costs and Impact to Broiler and Hog Processors, BAT Option
(500 AU Threshold) : 5-27
5-17 Post-Compliance Farm Level Price Changes, Selected Regulatory Alternatives 5-29
5-18 Post-Compliance Retail Level Price Changes, Selected Regulatory Alternatives 5-30
5-19 Post-Compliance Farm Production Changes, Selected Regulatory Alternatives 5-31
5-20 Post-Compliance Total National Employment Changes, Two-Tier Structure
(500 AU Threshold) - 5-32
5-21 Post-Compliance Total National Employment Changes, Three-Tier Structure 5-33
5-22 Total National Gross Output Reductions, Selected Regulatory Alternatives 5-34
5-23 Regional Distribution of Predicted National Employment Reductions 5-36
Xll
-------
5-24 Post-Compliance Retail Product Import and Export Changes,
Selected Regulatory Alternatives 5-37
6-1 Trends in Number of Poultry Operations and Birds, 1974-1997 6-3
6-2 EPA's Estimate of the Number of CAFOs Affected under the Co-Proposed
Tier Structures , 6-5
6-3 Geographic Distribution of Broiler Operations by Major Producing State, 1997 6-7
6-4 Geographic Distribution of Layer Operations by Major Producing State, 1997 6-8
6-5 Geographic Distribution of Turkey Operations by Major Producing State, 1997 .... 6-9
6-6 Total U.S. Poultry Supply and Demand, 1992-1997 6-11
6-7 Average Quarterly and Annual Poultry Prices Received by Farmers,
Total U.S., 1992-1997 6-12
6-8 Farm Revenue at Poultry Farms (>$50,000 in Annual Revenue) By Revenue Category and
Economic Class • 6-15
6-9 Income Statement and Balance Sheet for Poultry Farms (Sales >$50,000), 1993-97 6-17
6-10 Income Statements for Single-Contract Farms with Broilers, 1993 . 6-18
6-11 Distribution of Commercial Farms, by Net Farm Income, 1990-1995 ' 6-20
6-12 Typical Financial Characteristics of Broiler Operations, By Size of Operation 6-22
6-13 Income Statement and Balance Sheet for Broiler Operations, By Size of Operation 6-23
6-14 Typical Financial Characteristics of Layer Operations, By Size of Operation 6-25
6-15 Income Statement and Balance Sheet for Layer Operations, by Size of Operation 6-26
6-16 Typical Financial Characteristics of Turkey Operations, by Size of Operation 6-28
6-17 Income Statement and Balance Sheet Turkeys Operations, by Size of Operation 6-29
6-18 Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 5)
for Broilers • - - 6-33
6-19 Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 5)
for Layers 6-34
6-20 Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 5)
for Turkeys • "• 6-35
6-21 Summary of the Range of Post-Tax Annualized Compliance Costs Per Animal,
By Option • • 6-36
6-22 Total Estimated Post-Tax Compliance Costs 6-37
6-23 Impacted CAFOs Under ELG Options & NPDES Scenarios, Layer and
Turkey Operations 6-39
6-24 Impacted CAFOs Under ELG Options & NPDES Scenarios, Broiler Operations 6-40
6-25 Economic Achievabiliry Results for Broiler CAFOs ; 6-41
6-26 Economic Achievability Results for Layer and Turkey CAFOs 6-42
6-27 Number and Percentage of Affected Broiler CAFOs (Manure Sales Assumption) ...... 6-44
6-28 Impact of Passed Through Compliance Costs Under Co-proposed Alternatives,
Broiler Sector • • 6-45
6-29 Summary of Market Model Results for the Broiler Sector 6-48
6-30 Summary of Market Model Results for the Layer Sector 6-49
6-31 Summary of Market Model Results for the Turkey Sector 6-50
7-1 Number of Hog Operations and Animals, 1974-1997 7-3
7-2 EPA's Estimate of the Number of CAFOs Affected Under the Co-Proposed
Tier Structures 7-5
xui
-------
7-3 Geographic Distribution of Hog Operations by Major Producing State, 1997 7-6
7-4 Total U.S. Hog Supply and Demand (carcass weight basis), 1992-1997 7-7
7-5 Actual Average Quarterly and Annual Hog Prices Received by Fanners, Total U.S.,
1992-1997 • 7-9
7-6 Farm Revenue at Hog Farms (>$50,000 in Sales), by Revenue Category and
Economic Class • 7-11
7-7 Income Statement and Balance Sheet for Hog Farms (Sales>$50,000), 1993-97 7-12
7-8 Costs and Returns for Hog Farms by Facility Type, Average 1993-1997 .. 7-14
7-9 Typical Financial Characteristics of Hog Operations, By Size of Operation 7-16
7-10 Income Statement and Balance Sheet for Farms with Hogs and Pigs, by Size
of Operation, 1997 '•'- 7-17
7-11 Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 5) 7-23
7-12 Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 5A) 7-24
7-13 Summary of the Range of Post-tax Annualized Compliance Costs Per Hog, By Option . . 7-25
7-14 Total Estimated Post-Tax Compliance Costs 7-26
7-15 Impacted CAFOs under ELG Options & NPDES Scenarios, Grow-Finish
Hog Operations 7-28
7-16 Impacted CAFOs under ELG Options & NPDES Scenarios, Farrow-Finish
Hog Operations : 7-29
7-17 Economic Affordability Results for Hog CAFOs, Grow-Finish Operations 7-30
7-18 Economic Affordability Results for Hog CAFOs, Farrow-Finish Operations 7-31
7-19 Impact of Passed Through Compliance Costs under Co-proposed Alternatives,
Hog Sector 7-33
7-20 Summary of Market Model Results for the Hog Sector 7-35
8-1 Number of Beef and Dairy Operations and Animals, 1974-1997 8-5
8-2 EPA's Estimate of the Number of CAFOs Affected under the Tier Structures 8-7
8-3 Geographic Distribution of Cattle and Calf Feedlots by Major Producing State, 1997 8-9
8-4 Geographical Distribution of Dairy Operations by Major Producing State, 1997 8-10
8-5 Total U.S. Beef and Dairy Supply and Demand,' 1992-1997 8-12
8-6 Actual Average Quarterly and Annual Prices Received by Farmers, Total U.S.,
1992-1997 < 8-14
8-7 Farm Revenue at Beef Feedlots and Dairy Farms, By Revenue Category
and Economic Class , 8-16
8-8 Composite Income Statement and Balance Sheet in SIC 0211, Feedlots-
Beef Cattle, 1997 - • 8-18
8-9 Financial Information for Establishments in SIC 0211—Beef Cattle Feedlots,
1993-1997, 8-19
8-10 Financial Characteristics from NCBA Financial Survey (1994-1998) 8-20
8-11 Income Statement and Balance Sheet for Dairy Farms (Sales >$50,000), 1993-1997 8-23
8-12 Typical Financial Characteristics of Fed Beef Operations, By Size of Operation 8-25
8-13 Income Statement and Balance Sheet for Farms with Beef Cows,
By Size of Operation, 1997 8-26
8-14 Typical Financial Characteristics of Dairy Operations, by Size of Operation 8-30
8-15 Income Statement and Balance Sheet for Dairy Operations, by Size of
Operation, 1997 8-31
8-16 Per-Animal and Per Facility Post-Tax Annualized Compliance Costs (Option 3) ....... 8-40
xiv
-------
8-17 Per-Animal and Per Facility Post-Tax Annualized Compliance Costs (Option 3 A) 8-41
8-18 Summary of the Range of Post-Tax Annualized Compliance Costs Per Animal,
By Option ; • • • 8-42
8-19 Total Estimated Post-Tax Compliance Costs 8-43
8-20 Impacted CAFOs Under ELG Options & NPDES Scenarios, Beef, Veal, and
Heifer Operations 8-46
8-21 Impacted CAFOs Under ELG Options & NPDES Scenarios, Dairy Operations ........ 8-47
8-22 Economic Achievabiliiy Results for Beef/Heifer CAFOs (Option 3) and
Veal CAFOs (Option 5) ' 8-49
8-23 Economic Achievability Results for Dairy CAFOs (Option 3) 8-50
8-24 Economic Achievability Results for Beef, Heifer, and Dairy CAFOs (Option 3 A) 8-51
8-25 Summary of Market Model Results for the Beef Sector 8-52
8-26 Dairy Summary of Market Model Results for the Dairy Sector 8-53
9-1 SBA Revenue Size Standards for Small Livestock and Poultry Operations 9-3
9-2 Number of Small CAFOs That May Be Affected by the Proposed Regulations 9-7
9-3 Total Number of Small CAFO Businesses Subject to Regulation 9-11
9-4 EPA's Preliminary Assessment of Small Business Impacts using a Sales Test 9-13
9-5 Numbers of Small CAFO Businesses by Sector, Size, and Region, Two-Tier Structure .. 9-16
9-6 Numbers of Small CAFO Businesses by Sector, Size, and Region, Three-Tier Structure .9-17
9-7 Estimated Per-Head Facility Costs (BAT Option/Co-Proposed Scenarios)
for Model CAFOs - - - - - • 9-24
9-8 Estimated Per-Head Facility Revenues for Model CAFOs 9-26
9-9 Results of EPA's Small Business Analysis . 9-28
10-1 Annual Pre-Tax Costs of Proposed BAT Option under the
Co-Proposed Scenarios, $1999 .. ......'.". 10-5
10-2 Summary of the Number of CAFOs Required to Apply for a Permit, by Sector 10-8
10-3 Administrative Costs Associated with a General Permit, $1999 10-10
10-4 Administrative Costs Associated with an Individual Permit, $1999 10-11
10-5 State and Federal Administrative Costs, Two-Tier Structure (Scenario 4a) . . 10-13
10-6 State and Federal Administrative Costs, Three-Tier Structure (Scenario 3) 10-14
10-7 Total Annual Social Costs and Monetized Benefits, $1999 ., 10-16
A-l State Tax Income Rates A~3
A-2 IRS Asset Class Lives and Recovery Periods Relevant for the Annualization of
Capital Costs , • • • A'7
A-3 Cost Annualization Model A-8
A-4 Total Annualized Compliance Costs per Head for Option 1 - A-11
A-5 Total Annualized Compliance Costs per Head for Option 2 A-13
A-6 Total Annualized Compliance Costs per Head for Option 3 A-1-5
A-7 Total Annualized Compliance Costs per Head for Option 3 A A-17
A-8 Total Annualized Compliance Costs per Head for Option 4 A-19
A-9 Total Annualized Compliance Costs per Head for Option 5 A-21
A-10 Total Annualized Compliance Costs per Head for Option 6 . . A-23
A-l 1 Total Annualized Compliance Costs per Head for Option 7 A-25
xv
-------
B-l Elasticity Estimates in the Agricultural Economics Literature B-7
B-2 General Structure of the Model B-10
B-3 COSTBEN/EPA Model Variables , B-22
B-4 COSTBEN/EPA Model Coefficients • B-24
B-5 COSTBEN/EPA Model Parameters , - - B-25
C-1 Demand Elasticities for Beef Products Ranked from the Lowest Estimate to the
Highest Estimate : • C-2
C-2 Supply Elasticities for Beef Products Ranked from the Lowest Estimate to the
Highest Estimate C-3
C-3 Demand Elasticities for Milk Ranked from the Lowest Estimate to the Highest Estimate .. C-4
C-4 Supply Elasticities for Milk Ranked from the Lowest Estimate to the Highest Estimate ... C-4
C-5 Demand Elasticities for Pork Ranked from the Lowest Estimate to the Highest Estimate .. C-5
C-6 Supply Elasticities for Pork Ranked from the Lowest Estimate to the Highest Estimate .. . C-7
C-7 Demand Elasticities for Broilers/Chickens Ranked from the Lowest to the
Highest Estimate C-8
C-8 Supply Elasticities for Broilers/Chickens Ranked from the Lowest to the
Highest Estimate C-9
C-9 Demand Elasticities for Eggs Ranked from the Lowest Estimate to the Highest
Estimate • C-10
C-10 Supply Elasticities for Eggs Ranked from the Lowest Estimate to the Highest Estimate .. C-l 1
C-l 1 Demand Elasticities for Turkey Ranked from the Lowest Estimate to the
Highest Estimate C-ll
C-12 Supply Elasticities for Turkey Ranked from the Lowest Estimate to the
Highest Estimate C-l 1
D-l Baseline Revenues in Main Analysis and Sensitivity Analysis
(Lower Livestock Revenues) '. • • • • D-4
D-2 Number of CAFOs Affected Assuming Alternative Assumption
(Lower Livestock Revenues) D-5
D-3 Baseline Revenues in Main Analysis and Sensitivity Analysis
(Livestock Revenues Only) D-7
D-4 Number of CAFOs Affected Assuming Alternative Assumption
(Livestock Revenues Only) D-8
D-5 Number of CAFOs Affected Assuming Alternative Assumption (Pre-Tax Costs) D-10
D-6 Baseline Net Cash Income in Main Analysis and Sensitivity Analysis
(Lower Net Cash Income) D-ll
D-7 Number of CAFOs Affected Assuming Alternative Assumption
(Lower Net Cash Income) • D-13
D-8 Baseline Debt-to-Asset Ratios in Main Analysis and Sensitivity Analysis
(Higher Debt-to-Assets) D-15
D-9 Number of CAFOs Affected Assuming Alternative Assumption
'(Higher Debt-to-Asset Ratios) D-16
D-10 Elasticity Sensitivity Test Sets D-18
D-l 1 Range of Postregulatory Farm Product Price Results with Different
Elasticity Assumptions D-18
D-12 Range of Total Employment Change Results with Different Elasticity Assumptions ... D-19
xvi
-------
D-13 Range of Price Changes with Different Baseline Price Assumptions D-20
E-l Leading Sources and Pollutants of Water Quality Impairment in the United States, 1998 . . E-3
E-2 Summary of Statistics from the National U.S. Water Quality Impairment Survey, 1998 .. E-4
E-3 Nutrients, Metals, and Pathogens in "Livestock and Poultry Manures E-5
E-4 Nutrients Generated from CAFOs and Loadings that Discharge to U.S. Waters (Baseline) E-7
E-5 Estimated Metals Generated and "Edge-of-Field" Loadings from CAFOs E-8
E-6 Total Metal Removals "At Stream" by Regulatory Option Considered E-10
E-7 Cost-Effectiveness Results by Select Regulatory Option/Scenario ($1981) E-15
E-8 Cost-Effectiveness Results by Sector under the Two-Tier Structure
(Scenario 4a) ($1981) ' .........:... E-17
E-9 Cost-Effectiveness Results by Sector under the Three-Tier Structure
(Scenario 3) ($1981) E-18
E-10 Industry Comparison of BAT Cost-Effectiveness for Direct Dischargers E-l 9
E-l 1 Summary of Pollutant Removal Cost Estimates and Benchmarks ... , E-22
E-12 Cost-Effectiveness Results by Select Regulatory Option/Scenario, Nutrients ($1999) ... E-25
E-13 Cost-Effectiveness Results by Select Regulatory Option/Scenario, Sediments ($1999) . .. E-27
E-14 Cost-Effectiveness Results by Sector under the Two-Tier Structure
(Scenario 4a) ($1999) E-28
E-15 Cost-Effectiveness Results by Sector under the Three-Tier Structure
(Scenario 3 ) ($1999) E-29
xvu
-------
Figure
2-1
2-2
4-1
B-l
B-2
E-l
Share of Farms and Value of Production, by Typology Group, 1997 2-6
Flow of Activities and Sharing of Responsibilities in a Contractual System 2-26
. USDA Farm Producing Regions •. 4-12
Livestock and Poultry Products Market Model B-4
Direct Impacts B-5
Cost-Effectiveness E-14
xvui
-------
EXECUTIVE SUMMARY
ES.l INTRODUCTION
The U.S. Environmental Protection Agency (EPA) is revising and updating the two
primary regulations that" ensure that manure, wastewater, and other process waters generated by
concentrated animal feeding operations (CAFOs) do not impair water quality. EPA's proposed
regulatory changes affect the existing National Pollutant Discharge Elimination System (NPDES)
provisions and the existing effluent limitations guidelines (ELG) for "feedlots." The NPDES
provisions define and establish permit requirements for CAFOs and the ELG establish the
technology-based effluent discharge standard that is applied to CAFOs. Both of these existing
regulations were originally promulgated in the 1970s. ,
EPA is revising the regulations to address changes that have occurred in the animal
industry sectors over the last 25 years, to clarify and improve implementation of CAFO
requirements, and to improve the environmental protection achieved under these rules. The
revisions EPA is proposing would affect who must apply for a permit under the NPDES program,
who is subject to the ELG, and what the ELG requires. A summary of the current and the
proposed NPDES and ELG regulations for CAFOs are presented in Sections 1 and 3 of this
report, respectively. More detailed information on the current and proposed regulations is
presented in Sections E, VII, and VIE of the preamble.
This Economic Analysis (EA) summarizes EPA's analysis of the estimated annual
compliance costs and the economic impacts that may be incurred by affected operations that are
subject to the proposed revisions. EPA also provides additional material on the proposed CAFO
regulations in the Development Document for the Proposed Revisions to the National Pollutant
Discharge Elimination System Regulation and the Effluent Guidelines for Concentrated Animal
Feeding Operations, which discusses how EPA estimated compliance costs of the proposed
regulations. EPA's benefits analysis, titled Environmental and Economic Benefit Analysis of the
Proposed Revisions to the National Pollutant Discharge Elimination System Regulation and the
Effluent Guidelines for Concentrated Animal Feeding Operations, provides information about
existing water quality impairments associated with animal production operations and estimates the
extent to which these impairments may be mitigated by the proposed CAFO regulations.
ES.2 DATA AND METHODOLOGY
ES.2.1 Data Sources
EPA did not conduct an industry-wide survey of all CAFOs. Rather, EPA is relying on
existing data sources and expertise provided by numerous government agencies, state agricultural
extension services, land grant universities, and information from industry trade associations and
ES-1
-------
agricultural professionals. Major data sources are discussed in detail where they are used to
conduct the analyses presented in this report. Two major sources of primary USDA data were
instrumental to the economic analysis. These include USDA's 1997 Census of Agriculture and
data from the 1997 Agricultural Resource Management Study (ARMS).
The 1997 Census of Agriculture is conducted by the National Agricultural Statistical
Service (NASS) and provides information on the number of feedlots, their geographic
distributions, the amount of cropland available to land apply animal manure generated from animal
confinement operations, and other information. These data are compiled by NASS, with the
assistance of personnel at USDA's Natural Resources Conservation Service (NRCS) who
developed a methodology to identify information specific to animal confinement operations. All
Census data provided to other government agencies, including EPA, are aggregated to preserve
confidential business information. EPA uses these data to develop its model CAFOs and to
extrapolate CAFO level costs to all operations nationwide. A discussion of the Census data used
for this analysis is provided in Section 4 of this report; more detailed information is provided in
the Development Document.
The 1997 ARMS data, compiled by NASS and the Economic Research Service (ERS),
provide complete financial accounting data for U.S. farms for each of the major commodity
sectors affected by the proposed CAFO regulations. These data are used to depict farm financial
conditions to evaluate regulatory impacts. Data for representative farms were obtained by ERS
through special tabulations of the ARMS data, conducted by ERS, that differentiate the financial
conditions among operations by commodity sector, facility size (number of animals on site), and
major farm producing region. As with the Census data, these data were aggregated by USDA in a
manner to preserve both the statistical representativeness and confidentiality of the respondent
survey data. Section 4 discusses the ARMS data in more detail.
ES.2.2 Methodology
EPA estimates the economic impacts of the proposed CAFO regulations using a
representative farm approach. A representative farm approach is consistent with past research
conducted by USDA and many land grant universities to assess a wide range of policy issues,
including environmental legislation pertaining to animal agriculture. This approach provides a
means to assess average impacts across numerous facilities by grouping facilities into broader
categories to account for differences among operations.
EPA developed two sets of models for determining economic impacts at animal
confinement operations-cost models and financial models. EPA evaluated compliance costs
based on more than 170 farm level cost models that were .developed to depict conditions at and to
evaluate compliance costs for select representative CAFOs. EPA's cost models are differentiated
by commodity sector, farm production region, facility size, and land availability for application of
manure. EPA's cost models provide the estimated compliance costs that are compared to
ES-2
-------
corresponding financial models that characterize financial conditions across different types of
operations. (Similar to the cost models, the financial models are also differentiated by sector,
facility size, and production region.) Economic impacts under a post-regulatory scenario are
approximated by extrapolating the average impacts for a given model CAFO across the larger
number of operations that share similar production characteristics and are identified by that CAFO
model.
For this analysis, EPA evaluates the economic achievability of the proposed regulatory
options at existing animal feeding operations based on changes in representative financial
conditions across three criteria. These criteria are: a comparison of incremental costs to total
revenue (sales test), projected post-compliance cash flow over a 10-year period, and an
assessment of an operation's debt-to-asset ratio under a post-compliance scenario. To evaluate
economic impacts to CAFOs in some sectors, impacts are evaluated two ways—assuming that a
portion of the costs may be passed on from the CAFO to the consumer and assuming that no
costs passthrough so that all costs are absorbed by the CAFO.
Additional information on how EPA developed the cost models is provided in the
Development Document. Section 4 of this report discusses how EPA developed the financial
models and addresses additional methodological issues.
ES.3 REGULATED COMMUNITY
The animal sectors covered in this analysis include the cattle, veal, heifer, dairy, hog,
broiler, egg layer, and turkey sectors. Not all confinement operations (or animal feeding
operations, AFOs) in these sectors may be CAFOs and thus subject to the proposed regulations.
Table ES-1 presents the estimated number of operations that would be defined as a CAFO under
each of the co-proposed alternatives, as well as other regulatory scenarios considered by EPA.
The two co-proposed alternatives include the "two-tier structure" that would define as CAFOs all
AFOs with more than 500 AU and the "three-tier structure" that would define as CAFOs all
AFOs with more than 1,000 AU and any operation with more than 300 AU, if they meet certain
"risk-based" conditions, as defined in the preamble (also summarized in Section 3 of this report).
EPA estimates that both proposed alternative structures would regulate about 12,660
operations with more than 1,000 AU, accounting for operations with more than a single animal
type. The two-tier structure would also regulate an additional 12,880 operations with between
500 and 1,000 AU, for a total of 25,540 operations. Under the three-tier structure, an estimated
39,330 operations would be subject to the proposed regulations (10 percent of all AFOs),
estimated as the total number of animal confinement operations with more than 300 AU. See
Table ES-1. Of these, EPA estimates that a total of 31,930 AFOs would be defined as CAFOs (9
percent of all AFOs) and would need to obtain a permit (Table ES-1), while an estimated 7,400
operations would certify that they do not need to obtain a permit.
ES-3
-------
Table ES-1. Number of Potential Operations Defined as CAFOs by Select Regulatory Alternative, 1997
Sector/Size
Category
Cattle
Veal
Heifers
Dairy
Hogs: GF •>
Hogs: FF *
Broilers
Layers: wet w
Layers: dry w
Turkeys
Total ^
"Two-Tier"
>300
AU
>500
AU
>750
AU
(# operations)
4,080
210
1,050
7,140
4,920
9,450
14,140
360
1,690
2,100
39,320
3,080
90
800
3,760
2,690
5,860
9,780
360
1,280
1,280
25,540
2,480
40
420
2,260
2,300
3,460
7,780
210
1,250
740
19,100
>300
AU
>500
AU
>750
AU
(% total)
4%
25%
84%
6%
9%
15%
41%
12%
2%
15%
10.5%
3%
10%
64%
3%
5%
9%
28%
12%
2%
9%
6.8%
2%
4%
34%
2%
4%
5%
22%
7%
2%
5%
5.1%
"Three-Tier"
>300 AU
(#)
3,210
140
980
6,480
2,650
5,700
13,740
360
1,650
2,060
31,930
(% total)
3%
16%
78%
6%
5%
9%
39%
12%
2%
15%
8.5%
Source: See Section 2, Table 2-3.
''"Hogs: FF' are farrow-finish (includes nursery and breeder pigs); "Hogs: GF" are grow-Fmish only.
"•""Layers: wet" are operations with liquid manure systems. "Layers: diy" are operations with dry systems
^Total" eliminates double counting of operations with mixed animal types.
In addition to being defined as CAFOs, AFOs can be designated as CAFOs, even if they
have less than 500 AU (two-tier structure) or 300 AU (three-tier structure). EPA estimates that
designation may bring an additional 50 operations under the proposed two-tier structure (500 AU
threshold) and 10 operations under the proposed three-tier structure each year nationwide.
The proposed regulations may also affect businesses that contract out the raising or
finishing production phase to a CAFO but exercise "substantial operational control" over the
CAFO (described in Section 3 of this report). EPA estimates that 94 meat packing plants that
slaughter hogs and 270 poultry processing facilities may be subject to the proposed co-permitting
requirements. Other types of processing firms, such as further processors, food manufacturers,
dairy cooperatives, and Tenderers, are not expected to be affected by the co-permitting
requirements since these operations are further up the marketing chain and do not likely contract
with CAFOs to raise animals. Fully vertically integrated companies (e.g., where the packer owns
the CAFO) are not expected to require a co-permit since the firm as the owner of the CAFO
would require only a single permit. *
EPA also expects that crop farmers who receive manure from CAFOs would be affected
under one of the two co-proposed options relating to offsite management of manure. EPA's
ES-4
-------
Development Document documents how EPA estimated the number of potentially affected crop
producers. These estimates are presented in Section 5 as part of EPA's overall analysis.
Section 2 of this EA presents more detailed information on the regulated community,
including a profile of the various CAFO sectors and meat and poultry processors.
ES.4 ANNUAL INCREMENTAL COSTS
EPA estimates the annual incremental costs of compliance using the capital and recurring
costs derived in the Development Document. EPA converts these costs to incremental annualized
costs, as described in Section 4 of this report. Annualized costs better describe the actual
compliance costs that a model CAFO would incur, allowing for the effects of interest,
depreciation, and taxes. EPA uses these annualized costs to estimate the total annual compliance
costs and to assess the economic impacts of the proposed requirements to regulated CAFOs by
taking the annualized costs for each CAFO model and aggregating them on the basis of the
number of affected CAFOs represented by each model. Section 4 and Appendix A provide more
details on the cost annualization methodology and results.
EPA calculates two types of compliance costs-pre-tax and post-tax. The post-tax costs
reflect the fact that a CAFO would be able to depreciate or expense these costs, thus generating a
tax savings. Post-tax costs thus are the actual costs the CAFO. would face. Pre-tax costs reflect
the estimated total social cost of the proposed regulations, including lost tax revenue to
governments. Pre-tax dollars are used when comparing estimated costs to monetized benefits that
are estimated to accrue under the proposed regulations (see Section 10). In this Executive
Summary, EPA presents aggregate incremental costs on a pre-tax basis (and in 1999 dollars).
The preamble to this rulemaking also presents all costs on a pre-tax basis. Throughout this
report, aggregate costs presented and those that are used to assess financial impacts on CAFOs
are presented as post-tax costs and in 1997 dollars, since 1997 is the base year of the analysis.
Table ES-2 summarizes the total annualized 1999 pre-tax compliance costs to CAFOs
attributed to the proposed two-tier structure and three-tier structure. The table shows these costs
broken out by sector and by broad facility size group. These costs represent the cost of the
proposed BAT Option under the ELG and reflect a combination of two options that vary
depending on regulated sector. The "BAT Option" refers to EPA's proposal to require nitrogen-
based and, where necessary, phosphorus-based land application controls of all livestock and
poultry CAFOs (Option 2), with the additional requirement that all cattle and dairy operations
must conduct groundwater monitoring and implement controls, if the groundwater beneath the
production area has a direct hydrologic connection to surface water (Option 3 BAT), and with the
additional requirement that all hog, veal, and poultry CAFOs must also achieve zero discharge
from the animal production area with no exception for storm events (Option 5 BAT).
ES-5
-------
Under the two-tier structure and the proposed BAT Option, EPA estimates that the
incremental annuaHzed compliance cost to CAFO operators would be approximately $831 million
annually (Table ES-2). Under the three-tier structure and proposed BAT option, EPA estimates
that the total cost to CAFO operators would be $930 million annually (Table ES-2). Estimated
total annualized costs for the three-tier structure include the cost to permitted CAFOs as well as
the estimated cost to operations to certify to the permit authority that they do not meet any of the
conditions and are thus are not required to obtain a permit. EPA estimates certification costs at
about $80 million annually.
Estimated total annuaHzed costs shown in Table ES-2 include costs to animal confinement
operations that may be designated as CAFOs. Total annualized cost to designated facilities is
estimated at about $5 million or less than $1 million annually, depending on tier structure. More
information on these costs and how they are calculated is provided in Section 5.
Estimated costs in Table ES-2 also include costs to offsite recipients of manure. EPA is
proposing that offsite recipients of CAFO manure certify that manure will be land applied in
accordance with proper agriculture practices. EPA estimates the annualized cost of this
requirement to offsite recipients to be $9.6 to $11.3 million across the co-proposed alternatives.
This analysis is provided in the Development Document. See Section 5 for more details.
Table ES-2 Annual Pre-tax Cost of Co-ProDOsed Two-Tier and Three-Tier Structure, $1999
Sector
AH
Scenarios
>1000 AU
Two-Tier Structure (500 AU)
500 - 1000
AU
<500 AU
Total
Three-Tier Structure (Scenario 3)
300 - 1000
AU
<300 AU
Total
($1999, millions, pre-tax)
Regulated CAFOs
Beef
Veal
Heifer
Hog
Broiler
Layer
Turkey
Subtotal
$191.5
$0.03
$3.7
$108.6
$225.5
$55.4
$9.9
$10.4
$605.0
$24.7
$0.3
$7.9
$65.4-
$67.0
$41.6
$4.3
$9.2
S220.2
$0.1
NA
NA
$3.6
$1.5
$0.1
NA
NA
$5.4
$216.4
$0.3
$11.6
$177.6
$294.0
$97.1
$14.2
$19.6
$830.7
$36.2
$0.8
'$10.7 '
$115.3
$80.4
$61.2
$5.4
$14.5
$324.5
$0.0
$0.0
$0.0
$0.7
$0.2
$0.0
$0.0
$0.0
$0.8
$227.7
$0.8
$14.4
$224.6
$306.1
$116.6
$15.3
$24.9
$930.4
Other Farming Operations
Offsite
Recipients
TOTAL
NA
NA
NA
NA
NA
NA
$9.6
$840.3
NA
NA
NA
NA
$11.3
$941.7
Source: USEPA. See Section 5.
ES-6
-------
Table ES-3 summarizes the total annualized (pre-tax, $1999) costs of alternative
technology options for each NPDES scenario and ELG technology basis considered by EPA. As
shown in the table, the total estimated costs across these options range from $355 million (Option
I/Scenario 1) to $1.7 billion annually (Option 5, applicable to all the animal sectors, and Scenario
4b). By scenario, this reflects the fact that fewer CAFOs would be affected under Scenario 1 (a
total of about 16,400 operations) as compared to Scenario 4b (about 39,300 operations affected).
EPA's estimate of the number of CAFOs and corresponding compliance costs does not adjust for
operations with mixed animal types and may be overstated. By technology option, with the
exception of Options 1 and 4, costs are evaluated incremental to Option 2 (see Table ES-3).
Compared to Option 2, Option 5 costs are greatest.
Table ES-3. Annualized Pre-tax Costs for the Alternative NPDES Scenarios (S1999, million)
Option/
Scenario
# CAFOs
Option 1
Option 2
Option 3
Option 4
Option 5
Option 6
Option 7
BAT Option
Scenario 4a
"Two-Tier"
25,540
$432.1
$548.8
$746.7
$903.9
$1,515.9
$621.6
. $671.3
$830.7
Scenario 2/3
"Three-Tier"
31,930
$459.1
$578.7
$859,7
$1,087.1
$1,629.6
$706.6
$756.6
$930.4
Scenario 1
16,420
$354.6
$444.4
$587.0
$707.0
$1,340.9
$501.5
$542.4
$680.3
Scenario 5
>750AU
19,100
$384.3
• $484.0
$649.5
$768.0
$1,390.4
$541.3
$585.1
$720.8
Scenario 4b
>300 AU
39,320
$493.6
$633.3
$883.6
$1,121.2
$1,671.3
$706.6
$756.6
$979.6
Source: See Section 5.
ES.5 ECONOMIC IMPACTS
ES.5.1 CAFO Impacts
EPA uses the financial criteria described previously to divide the impacts of the proposed
regulations into three impact categories: affordable, moderate, and financial stress. EPA does not
consider impacts under the affordable or moderate categories to result in closure of the CAFO as
: a result of compliance. Impacts under the stress category, however, may result in CAFOs being
vulnerable to closure post-compliance, taking other factors into consideration. More information
on these criteria and how they are used to determine economic achievability is provided in
Section 4.
ES-7
-------
Table ES-4 presents the estimated CAFO level impacts in terms of the number of
operations that are estimated to incur affordable, moderate, or stress impacts for each of the co-
proposed alternatives by sector. Based on results shown in Table ES-4, EPA proposes that the
regulatory alternatives are economically achievable for all representative model CAFOs in the
veal, turkey and egg laying sectors. The proposed requirements under the two-tier structure are
also expected to be economically achievable by all affected heifer operations. Furthermore,
although operations across most sectors may experience moderate impacts, EPA does not expect
moderate financial impacts to result in closure and considers this level of impact to be
economically achievable.
In the beef cattle, heifer, dairy, hog and broiler sectors, however, EPA's analysis indicates
that the proposed regulations will cause some operations to experience financial stress, assuming
nocostpassthrough. These operations may be vulnerable to closure by complying with the
proposed regulations.. Across all sectors, an estimated 1,890 operations would experience
financial stress under the two-tier structure and an estimated 2,410 operations would experience
stress under the three-tier structure. For both tier structures, EPA estimates that the percentage
of operations that would experience impacts under the stress category represent 7 percent of all
affected CAFOs or 8 percent of all affected operations in the sectors where impacts'are estimated
to cause financial stress (cattle, dairy, hog, and broiler sectors).
Table ES-4 shows results for the two-tier structure at the 500 AU threshold. By sector,
EPA estimates that 1,420 hog operations (17 percent of affected hog CAFOs), 320 dairies
(9 percent of operations), 150 broiler operations (2 percent), and 10 beef operations (less than
one percent) would experience financial stress. The broiler and hog operations with these impacts
have more than 1,000 AU on-site (i.e., no operations with between 500 and 1,000 AU fall in the
stress category). The dairy and cattle operations with stress impacts are those that have a ground
water link to surface water.
Table ES-4 also presents results for the three-tier structure. By sector, EPA estimates that
1,420 hog operations (17 percent of affected hog CAFOs under that alternative), 610 dairies
(9 percent of operations), 330 broiler operations (2 percent), and 50 beef and heifer operations
(1 percent) will experience financial stress. Hog operations with stress impacts all have more than
1,000 AU. Affected broiler facilities include operations with more than 1,000 AU, as well as
operations with less than 1,000 AU. Dairy and cattle operations in the stress category are
operations that have a hydrologic link from ground water to surface water. Based on these
results, EPA is proposing that the proposed regulations are economically achievable.
In the hog and broiler sectors, EPA also evaluates financial impacts with an assumption of
cost passthrough. For the purpose of this analysis, EPA assumes that the hog sector could
passthrough 46 percent of compliance costs and the broiler sector could passthrough 35 percent
of compliance costs. EPA derives cost passthrough estimates from price elasticities of supply and
demand for each sector reported in the academic literature (Section 4). Assuming these levels of
cost passthrough, the magnitude of the estimated impacts decreases to the affordable or moderate
ES-8
-------
impact category. Even in light of the uncertainty of cost passthrough (both in terms of whether
the operations are able to pass cost increases up the marketing chain and the amount of any cost.
passthrough), EPA proposes that the proposed regulations will be economically achievable to all
hog and broiler operations. Although EPA's analysis does not evaluate impacts assuming cost
passthrough in other sectors, EPA expects that long-run market and structural adjustment by
producers will diminish the estimated impacts in all sectors.
Table ES-4. Number of CAFOs Affected under the Co-Proposed Alternatives (Zero Cost Passthrough)
Sector
Fed Cattle
Veal
Heifer
Dairy
Hogs
Broilers
Layers - Wet
Layers - Dry
Turkeys
Total
Two-Tier Structure (500 AU Threshold)
No. of
CAFOs
3,080
90
800
3,760
8,550
9,780
360
1,280
1,280
28,970
Aff.
Mod.
(number)
2,830
90
680
3,240
6,920
1,960
360
1,280
1,230
18,580
240
0
120
200
210
7,670
0
0
50
8,490
Stress
10
0
0
320
1,420
150
0
0
0
1,890
Three-Tier Structure (Scenario 3)
No. of
CAFOs
3,210
140
980
6,480
8,360
13,740
360
1,660
•2,060
37,000
Aff.
Mod.
(number)
2,540
140
800
5,300
6,720
1,850
360
1,660
1,950
21,300
650
0
150
. 560
220
11,560
0
0
110
13,250
Stress
20
0
30
610
1,420
330
0
0
0
2,410
Source: USEPA. See Section 5.
ES.5.2 Processor Impacts
EPA did not conduct a detailed estimate, of the costs and impacts that would accrue to
individual co-permittees due to lack of data and market information. However, EPA believes that
the framework used to estimate costs to CAFO provides a means to evaluate the possible upper
bound of costs that could accrue to potential co-permittees, based on the potential share of (pre-
tax) costs that may be passed on from the CAFO.
Using this approach, the potential magnitude of costs to co-permittees is derived from the
amount of cost passthrough assumed hi the CAFO level analysis. Based on the results of this
analysis, EPA estimates that the range of potential annual costs to hog processors is $135 million
ES-9
-------
(partial cost passthrough, $1999) to $306 million (full cost passthrough). EPA estimates that the
range of potential annual costs to broiler processors as $34 million (partial cost passthrough) to
$117 million (full cost passthrough).
To assess the magnitude of impacts that could accrue to processors using this approach,
EPA compares the passed through compliance costs both to aggregate processor production costs
and to revenues (a sales test). The results of this analysis indicate that, even under full cost
passthrough, incremental cost changes to processors in these industries are estimated at less than
two percent of total costs and less than one percent of total revenues. Additional information is
provided in Section 5.
ES.5.3 Market Impacts
EPA's market analysis evaluates the effects of the proposed regulations on commodity
prices and quantities at the national level. EPA expects that predicted changes in animal
production may raise producer prices, as the market adjusts to the proposed regulatory
requirements. For most sectors, EPA estimates that producer price changes will rise by less than
one percent compared to the pre-regulation baseline price. At the retail level, EPA estimates that
poultry and red meat prices will rise about one cent per pound. EPA also estimates that egg
prices will rise by about one cent per dozen and that milk prices will rise by about one cent per
gallon.
This analysis also presents EPA's estimate of employment changes associated with the
proposed regulations. Within the farming sector, EPA predicts that the proposed CAFO
regulations will result in employment losses ranging from 2,700 to 3,000 jobs, depending on tier
structure. This estimated reduction compares to an estimated total farm level employment of over
one million jobs in these sectors nationwide. Estimated employment and job losses in the
agricultural sector include CAFO owner-operators and employed family members, as well as hired
farm labor.
ES.6 OTHER REGULATORY REQUIREMENTS
ES.6.1 Small Business Analysis
Table ES-5 shows EPA's estimate of the number of small businesses in the livestock and
poultry sectors and the number of small businesses that are expected to be affected by the
proposed regulations. Under the two-tier structure, EPA estimates that 10,550 operations that
will be subject to the proposed requirements are small businesses. Under the three-tier structure,
an estimated 14,630 affected operations are small businesses. Under the two-tier structure, EPA
estimates also include an additional 330 designated operations that meet the small business
definition, projected over a 10-year time frame (i.e., 33 AFOs designated annually). Under the
ES-10
-------
three-tier structure, EPA estimates that 100 operations with fewer than 300 AU that meet the
small business definition may be designated over a 10-year period (i.e., 10 AFOs designated
annually). Additional information is provided in Section 9.
EPA's assessment of the financial impacts of the proposed rule on small entities uses the
same approach as that used to evaluate the impact to CAFOs under the proposed regulations (see
Section 4). EPA is proposing that the proposed regulations are economically achievable by small
businesses in the livestock and poultry sectors. The results of this analysis are presented in Table
ES-5. As defined for this analysis, EPA's analysis indicates that the proposed requirements are
economically achievable to all affected small businesses in the beef, veal, heifer, dairy, hog, and
egg laying sectors ("Affordable" and also "Moderate"). Under the two-tier structure, EPA
expects that there are no small CAFO businesses in the turkey sector. Under the three-tier
structure, EPA expects that there are.an estimated 500 small CAFO businesses that confine
turkeys (Table ES-5).
Table ES-5. Results of EPA's Small Business Analysis Under the BAT Option/Two-Tier Structure
Sector
Fed Cattle
Veal
Heifer
Dairy
Hogs
Broilers
Layers
Turkeys
TOTAL
Number
of Small
CAFOs
.1,390
90
800
50
300
9,470
200
0
10,550
Aff.
Mod.
Stress
(Number of Operations)
1,130
90
680
40
300
1,860
200
0
4,300
250
0
120
10
0
7,460
0
0
7,840
10
0
0
0
0
150
0
0
160
Number
of Small
CAFOs
1,490
140
980
50
300
13,410
590
500
14,630
Aff.
Mod.
Stress
(Number of Operations)
1,100
140
800
40
300
1,910
590
460
5,340
380
0
150
10
0
11,220
0
40
11,800
10
0
30
0
0
280
0
0
320
Source: USEPA. See Section 5.
EPA's analysis indicates that the proposed requirements will not result in financial stress
to any affected small businesses in the veal, heifer (two-tier only), hog, dairy, egg laying, and
turkey sectors. In the beef, heifer (three-tier only), and broiler sectors, however, EPA's analysis
indicates that proposed regulations could result in financial stress to some small CAFO businesses,
making these businesses vulnerable to closure. Overall, these operations comprise about 2
percent of all affected small CAFO businesses. For the two-tier structure, EPA estimates that 10
small beef operations and 150 small broiler operations will experience financial stress. For the
ES-11
-------
three-tier structure, EPA estimates that 40 small beef and heifer operations and 280 small broiler
operations will experience financial stress. Small broiler facilities witii stress impacts are larger
operations with more than 1,000 AU under both tier structures. Small cattle and heifer operations
with stress impacts are those that have a ground water link to surface water. This analysis is
conducted assurning that no costs are passed through between the CAFO and processor segments
of these industries. Based on the results of this analysis, EPA is proposing that the proposed
regulations are economically achievable to small businesses in these sectors.
ES.6.2 Cost Benefit Analysis .
EPA estimated and compared the costs and benefits attributed to the proposed
regulations. The cost and benefit categories that EPA is able to quantify and monetize for the
proposed regulations are shown in Table ES-6. Total social costs of the proposed regulations in
1999 dollars range from $847 million to $949 million annually, depending on the co-proposed
approach. These costs include compliance costs to industry, costs to recipients of CAFO manure,
and administrative costs to States and Federal governments. EPA estimates that the monetized
benefits of the proposed regulations range from $146 million to $182 million annually, depending
on the co-proposed approach (Table ES-6). Section 10 provides additional information on these
analyses.
Table ES-6. Total Annual Social Costs and Monetized Benefits, S1999
Total Social Costs
Industry Compliance Costs (pre-tax):
NPDES Permitting Costs:
Offsite Recipients of CAFO Manure:
Total Social Costs
Monetized Benefits
Improved surface water quality
Reduced shellfish bed closures
Reduced fish kills
Improved water quality in private wells
Total Monetized Benefits
"Two-Tier" Structure
(500 AU threshold)
$830.7 million
$6.2 million
$9.6 million
$846.5 million
$108.5 million
$0.2 - 2.4 million
$0.2 - 0.4 million
$36.6-53.9 million
$145.5 - 165.1 million
"Three-Tier" Structure
(Scenario 3)
$930.4 million
$7.7 million
$11. 3 million
$949.4 million
$127.1 million
$0.2 - 2.7 million
$0.2 - 0.4 million
, $35.4 - 52.1 million
$163.0 - 182.3 million
Source: See Section 10.
ES-12
-------
ES.7 OTHER INFORMATION
This report contains a detailed industry profile of the affected regulated livestock and
poultry sectors and meat and poultry processors (Section 2). It also presents a summary of
estimated per-animal and per-facility costs by animal sector (Sections 6 through 8 and
Appendix A). Additionally, it presents an overview of the cost annualization approach (Appendix
A), details on the model used to estimate economic impacts on CAFOs and national level markets
(Section 4 and Appendix B), results of sensitivity analyses (Appendix D), and results of cost-
effectiveness analyses (Appendix E). See Section 1 for a guide to the contents of this report.
ES-13
-------
-------
SECTION ONE
INTRODUCTION
The U.S. Environmental Protection Agency (EPA) is revising and updating the two
primary regulations that ensure that manure, wastewater, and other process waters generated by
concentrated animal feeding operations (CAFOs) do not impair water quality. EPA's proposed
regulatory changes affect the existing National Pollutant Discharge Elimination System (NPDES)
provisions and the existing effluent limitations guidelines (ELG) for "feedlots." The NPDES
provisions define and establish permit requirements for CAFOs and the ELG establish the
technology-based effluent discharge standard that is applied to CAFOs. Both of these existing
regulations were originally promulgated in the 1970s. EPA is revising the regulations to address
changes that have occurred in the animal industry sectors over the last 25 years, to clarify and
improve implementation of CAFO requirements, and to improve the environmental protection
achieved under these rules. Proposed revisions to the NPDES and ELG regulations are referred
to in this report as the proposed CAFO regulations.
This Economic Analysis (EA) summarizes EPA's analysis of the estimated annual
compliance costs and the economic impacts that may be incurred by affected operations that are
subject to these requirements. It examines hi detail EPA's regulatory proposal and several
regulatory alternatives that were considered by EPA. The report covers financial impacts to
CAFOs, potential impacts on processors of livestock and poultry products, and market and other
secondary impacts such as impacts on prices, quantities, trade, employment, and output. It also
responds to requirements for small business analyses under the Regulatory Flexibility Act (RFA)
as amended by the Small Business Regulatory Enforcement Fairness Act (SBREFA) and for cost-
benefit analyses under Executive Order 12866 and the Unfunded Mandates Reform Act (UMRA).
This section begins with a discussion of the current regulatory framework and, in the
course of this discussion, defines and describes animal feeding operations and CAFOs
(Section 1.1). The reasons why EPA is revising these regulations are also discussed. Sources of
data that are critical to the analyses presented in this EA are then briefly described (Section 1.2)
and the section concludes with an outline of the report (Section 1.3).
1.1 EXISTING REGULATORY FRAMEWORK
In 1972, Congress passed the Federal Water Pollution Control Act, also known as the
Clean Water Act (CWA), to "restore and maintain the chemical, physical, and biological integrity
of the nation's waters." 33 U.S.C. § 125 l(a). The CWA establishes a comprehensive program
for protecting our nation's waters. Among its core provisions, the CWA prohibits the discharge
of pollutants from a point source to waters of the U.S. except those authorized by a NPDES
1-1
-------
permit. The CWA also provides for the development of technology-based effluent limitations that
are imposed through NPDES permits to control direct discharges of pollutants.
In response to the CWA, EPA established,several regulatory programs, of which two
pertain to livestock and poultry operations that confine animals (commonly referred to by EPA as
"animal feeding operations," or AFQs). These regulations include the requirements for discharge
permits for CAFOs under the NDPES program (40 CFR Part 122.23) (see Section 1.1.1) and the
ELG for animal feeding operations ("feedlots") (40 CFR Part 412) (see Section 1.1.2).
1.1.1 NPDES Permit Regulation of CAFOs
The NPDES permit program controls pollution from identifiable discharge points or
sources (e.g., discharge pipes or ditches). Under the NPDES permit program, all point sources
that directly discharge pollutants to waters of the U.S. must apply for a NPDES permit and may
only discharge pollutants under the terms of that permit. Such permits must include nationally
established effluent discharge limitations. In the absence of national effluent limitations, NPDES
permit writers must establish limitations and standards on a case-by-case basis, based on their
"best professional judgement (BPJ)." Effluent limitations guidelines and BPJ provide the basis for
technology-based effluent limits in NPDES permits.
Under the CWA, CAFOs are defined as point sources of pollution and are thus subject to
NPDES permitting requirements (33 U.S.C. § 1362). The existing NPDES provisions that define
which operations are CAFOs and that establish permit requirements for CAFOs (40 C.F.R. §
122.23) were promulgated on March 18,1976 (41 FR 11458).
Before an operation may be defined as a CAFO, it must first meet the definition of an
AFO. AFOs are agricultural enterprises where animals are kept and raised in confined situations
for a specified time during the year and congregate animals, feed, manure, dead animals, and
production operations on a small land area. As defined by federal regulation, AFOs are lots or
facilities where animals:
"...have been, are, or will be stabled or confined and fed or maintained for a total
' of 45 days or more in any 12 month period and crops, vegetation forage growth,
or post-harvest residues are not sustained in the normal growing period over any
portion of the lot or facility." (40 CFR 122.23(b)(l)).
As discussed in more detail in the preamble to the CAFO regulations, this definition is being
broadened to include as AFOs all operations that confine animals unless crops or other vegetation
are grown at all times over all parts of the lot or facility. •
In 1976, EPA issued regulations defining which AFOs met the definition of a CAFO under
the NPDES permit program. CAFOs are AFOs that confine a specified number of animals and in
1-2
-------
some cases meet specific discharge criteria. The specified number of animals is determined using
the concept of an "animal unit" (AU). The term AU is a metric established in the 1970
regulations that attempted to equate the characteristics of the wastes produced by different animal
types. For each animal type, EPA's regulations identify the number of animals that is equivalent
tooneAU.
As defined in the existing regulation (40 CFR 122), one AU is equivalent to:
• 1 slaughter or feeder beef cattle
• 0.7 mature dairy cows
• 2.5 swine weighing more than 55 pounds
• 55 turkeys
• 100 laying hens or broilers (facility with continuous flow watering systems); 30
hens or broilers (facility with liquid manure handling system)
• 0.5 horses
« 10 sheep or lambs
• 5 ducks
As discussed in more detail in the preamble to the CAFO regulations, EPA is proposing to change
how an AFO is defined as a CAFO and, therefore, subject to regulation.
The existing NPDES regulation defines AFOs with 1,000 AU or more as CAFOs. These
facilities are not CAFOs if they discharge only in the event of a 25-year, 24-hour storm. The
regulation also states that AFOs with between 300 and 1000 AU are CAFOs if they meet certain
conditions. These conditions include the discharge of pollutants into waters through a ditch,
flushing system, or other man-made device. An AFO with between 300 and 1000 AU may also
be defined as a CAFO if pollutants are discharged to waters that originate outside of and pass
over, across, or through the facility or come in to contact with confined animals. The state
agency or other authority that issues NPDES permits may also designate AFOs with fewer that
300 AU as CAFOs, if they are considered to have discharges that could significantly impair
surface water.
As discussed in more detail hi the preamble to the proposed CAFO regulations, EPA is
proposing to revise these criteria for determining when an AFO is a CAFO, as well as other key
provisions in the existing regulation, including the exemption under a storm event. Section 3 of
1-3
-------
this report provides a brief summary of the recommended changes that are being proposed for the
NPDES requirements affecting CAFOs.
All NPDES permits for CAFOs with more than 1,000 AU must include requirements
equivalent to or more stringent than the established ELG. As noted above, certain smaller
operations can also be defined or designated as CAFOs, but the ELG does not apply to these
CAFOs. In these cases, the permit writer must develop technology-based limitations based on
BPJ for inclusion in the NPDES permit.
1.1.2 Effluent Limitations Guidelines for Feedlots
The CWA authorizes EPA to establish restrictions on the types and amounts of pollutants
discharged from various industrial, commercial, and public sources of wastewater. Effluent
guidelines define the type and amount of pollutants an NPDES permitted facility is allowed to
discharge. Direct dischargers must comply with effluent limitations guidelines and new source
performance standards (NSPS). These limitations and standards are established by regulation for
categories of industrial dischargers and are based on the degree of control that can be achieved
using various levels of pollution control technology. These guidelines base the discharge (or
effluent) amount of the best available technology, or BAT, that is economically achievable.
The existing national effluent limitations guidelines for the feedlots category, including
beef, dairy, swine, and poultry subcategories (40 CFR §412), were established on February 14,
1974 (39 FR. 5704). The feedlot ELG allow for no discharge of process wastewater pollutants
into the Nation's waters except when chronic or catastrophic storm events cause an overflow
from a facility designed, constructed, and operated to hold process-generated wastewater plus
runoff from a 25-year, 24-hour storm event. As a result, the current effluent guidelines for
feedlots are usually referred to as "zero discharge" requirements. Many feedlots meet the "zero
discharge" requirement by containing wet manure in lagoons and by land applying manure.. The
current ELG are applicable to NPDES permits issued to CAFOs with more than 1,000 AU.
Discharge limits for facilities with fewer than 1,000 AU are established using BPJ.
As discussed in more detail in the preamble to the CAFO regulations, EPA is proposing to
revise the applicability of the ELG along with other key provisions in the existing regulation.
Section 3 of this report provides a brief summary of the recommended changes that are being
proposed for the ELG affecting CAFOs.
1.1.3 Industries Affected by the Proposed CAFO Regulations
In this EA, information is organized by sector rather than by subcategory. For example,
information is presented on fed cattle, heifer, and veal operations (often referred to as the "cattle"
sector), because heifer and veal operations are more similar to beef operations than to dairy
1-4
-------
operations. Under the regulation, however, the beef subcategory includes fed cattle and heifers,
but veal operations are covered under a separate subcategory. The dairy and swine subcategories
are discussed separately. Information on the types of operations in the poultry sector (i.e., broiler,
egg layer, and turkey operations) is also presented together, although the chicken and turkey
subcategories are distinct in the regulation. The proposed NPDES revisions do affect other types
of animal confinement operations, such as operations that raise sheep, lambs, goats, horses, and
other miscellaneous animal species. These sectors are not covered in this analysis.
This report focuses on the major livestock and poultry industries affected by the ELG and
the NPDES program requirements. By North American Industry Classification System (NAICS)1
code, these include:
• Cattle feedlots, NAICS 112112 [includes veal] (SIC 0211, beef cattle feedlots).
• Beef cattle ranching and farming, NAICS 112111 (SIC 0241, dairy heifer
replacement farms).
• Dairy cattle and milk production, NAICS 11212 (SIC 0241, dairy farms).
• Hog and pig farming, NAICS 11221 (SIC 0213, hogs).
• Broilers and other meat-type chickens, NAICS 11232 (SIC 0251, broiler, fryer,
and roaster chickens).
• Turkey production, NAICS 11233 (SIC 0253, turkey and turkey eggs).
• Chicken egg production, NAICS 11231 (SIC 0252, chicken eggs).
USDA reports that in 1997 there were 1.1 million livestock and poultry operations in the
United States, corresponding to these affected industry sectors. This number includes both
confinement and non-confinement operations (i.e., grazing and range fed) production as well as
commercial and noncommercial operations. Of these operations, EPA estimates that there are
about 376,000 AFOs that raise or house animals in confinement, as defined by the existing
regulations. Table 1-1 summarizes the estimated total number of AFOs of all sizes in each of the
four major livestock categories in 1997. EPA estimates that only a small subset of these AFOs
would be regulated as CAFOs, since most would not meet the size definitions or other criteria.
More information is provided in Section 2 of this report.
'NAICS recently replaced the SIC (Standard Industrial Classification) system.
1-5
-------
Sector
Beef operations, including both cattle, veal, and heifer operations
Dairy operations (milk production operations only)
Hog operations, including both "farrow-finish" and "grow-fmish" operations ^
Poultry operations, including broilers, layers (both wet and dry operations) and turkeys b/
Sum Total
Total AFOs rf
Total AFOs
108,180
116,870
117,880
123,750
466,670
375,700
Source: USEPA. See Section 2.
^Grow-finish operations finish more mature pigs while farrow-finish operations handle all stages of production
from breeding to finishing.
"Utilize either liquid or dry manure handling systems present at the facility.
''"'Total AFOs" eliminates double counting of operations with more than a single animal type (see Section 2).
1.1.4 Reasons Why EPA is Revising the Existing CAFO Regulations
EPA is proposing to revise the existing regulations to meet the following goals:
m Address continued and persistent reports of discharge and runoff of manure and
manure nutrients from livestock and poultry operations.
• Update the existing regulations to reflect major structural changes in these
industries over the last few decades.
B Improve the effectiveness of the existing regulations at protecting or restoring
water quality.
Each of these stated goals is briefly discussed below and in other sections of this report
(Sections 9 and 10). More detail is presented in Section IV and V of the preamble to this
proposed rulemaking. These reasons are summarized briefly below:
First, despite more than twenty years of regulation, there are persistent reports of
discharge and runoff of manure and manure nutrients from confined animal operations. The
proposed revisions to the existing ELG and NPDES regulations for CAFOs are expected to
mitigate future water quality impairment and the associated human health and ecological risks by
reducing pollutant discharges from the animal production industry.
Evidence that manure discharge and runoff is resulting in water quality impairment is
presented in the National Water Quality Inventory: 1998 Report to Congress (USEPA, 2000h).
In this report, the agriculture sector is identified as the leading cause of water quality impairments
1-6
-------
in the nation's rivers, streams, lakes, ponds, and reservoirs and also a major contributor to
identified water quality impairments in estuaries. The animal production industry is associated
with impairments caused by nutrients, pathogens, oxygen-depleting substances, and solids
(siltation). Animal production facilities also may contribute metals, pesticides, priority toxic
organic chemicals, and oil and grease and can contribute to the growth of noxious aquatic plants
due to the discharge of excess nutrients. Other documents that support this rulemaking, including
EPA's Environmental and Economic Benefit Analysis of the Proposed Revisions to the National
Pollutant Discharge Elimination System Regulation and the Effluent Guidelines for
Concentrated Animal Feeding Operations ("Benefits Analysis" [USEPA, 2000d]) and EPA's
Environmental Assessment of the Proposed Revisions to the National Pollutant Discharge
Elimination System Regulation and the Effluent Guidelines for Concentrated Animal Feeding
Operations ("Environmental Assessment" [USEPA, 2000b]), provide additional information
about existing water quality impairments associated with animal production operations.
Second, the existing regulations are being updated to reflect structural changes in these
industries over the last few decades. Periodic review and revision of existing regulations is
envisioned in the CWA. More animals are produced annually at fewer AFOs, leading to a
increasing share of animal production at larger operations that concentrate more animals (and thus
manure and wastewater) at a single location. This trend has coincided with increased reports of
large-scale discharges from these facilities. Furthermore, farming in traditional rural, agricultural
areas where the manure nutrients generated could be readily incorporated as a fertilizer in crop
production is giving way to animal production operations located close to slaughtering and
processing plants and near end-consumer markets, where land is often unavailable for land
application of manure and wastewater. These geographic shifts in fanning operations may be
shifting the flow of manure nutrients away from areas where these nutrients can be effectively
used to areas where they cannot be easily absorbed. Section 2 of this EA discusses these industry
changes in more detail. .
Finally, EPA believes it must improve the effectiveness of the regulations, first, by making
the regulations simpler and easier to understand and implement, second, by clarifying the
conditions under which an AFO is a CAFO and thus subject to NDPES regulatory requirements,
and third, by removing provisions that are no longer appropriate. Section 3 of this EA discusses
these changes, as well as the changes proposed to address industry trends and the need for
increased control of manure nutrients and discharges from CAFOs.
1.2 OVERVIEW OF SOURCES OF DATA
EPA is undertaking an expedited approach to this rulemaking effort and did not conduct
an industry-wide survey of all CAFOs using a CWA Section 308 questionnaire. Rather, EPA is
relying on existing data sources and expertise provided by numerous government agencies, state
agricultural extension services, land grant universities, and information from industry trade
associations and agricultural professionals. Major data sources are discussed in detail where they
1-7
-------
are used to conduct the analyses presented in this report. This section provides an overview of.
these sources and their importance to the economic analysis of the proposed CAFO regulations.
The key sources of data to conduct analyses presented in this EA include those used to
estimate compliance costs and economic impacts to the regulated community. The data that EPA
uses to estimate compliance costs are discussed in detail in EPA's Development Document for the
Proposed Revisions to the National Pollutant Discharge Elimination System Regulation and the
Effluent Guidelines for Concentrated Animal Feeding Operations (referred to as the
"Development Document"[USEPA, 2000a]). EPA's compliance cost estimates are based on
information compiled through EPA site visits to over 100 animal confinement operations and on
information developed by various agencies within the U.S. Department of Agriculture (USDA),
including the National Agricultural Statistics Service (NASS), the Animal and Plant Health
Inspection Service (APHIS), the Natural Resources Conservation Service (NRCS), and the
Economic Research Service (ERS).
Two major sources of primary USDA data are instrumental to the economic analysis.
These include USDA's 1997 Census of Agriculture and data from the 1997 Agricultural
Management Study (ARMS).
The 1997 Census of Agriculture is conducted by the National Agricultural Statistical
Service (NASS) and provides information on the number of feedlots, their geographic
distributions, the amount of cropland available to land apply animal manure generated from animal
confinement operations, and other information. These data are compiled by NASS, with the
• assistance of personnel at USDA's Natural Resources Conservation Service (NRCS) who
developed a methodology to identify information specific to animal confinement operations. All
Census data provided to other government agencies^ including EPA, are aggregated to preserve
confidential business information. EPA uses these data to develop its model CAFOs and to
extrapolate CAFO level costs to all operations nationwide. A discussion of the Census data used
for this analysis is provided in Section 4 of this report; more detailed information is provided in
the Development Document (USEPA, 2000a).
The 1997 ARMS data, compiled by NASS and ERS, provide complete financial
accounting data for U.S. farms for each of the major commodity sectors affected by the proposed
CAFO regulations. These data are used to depict farm financial conditions to evaluate regulatory
impacts. Data for representative farms were obtained by ERS through special tabulations of the
ARMS data, conducted by ERS, that differentiate the financial conditions among operations by
commodity sector, facility size (number of animals on site), and major farm producing region. As
with the Census data, these data were aggregated by USDA in a manner to preserve both the
statistical representativeness and confidentiality of the respondent survey data. Section 4
discusses the ARMS data in more detail. . _
Industry and the associated trade groups also provided information for the economic and
financial analyses. In particular, the National Cattlemen's Beef Association (NCBA) conducted a
1-8
-------
survey of their membership to obtain financial statistics specific to cattle feeding operations.
Industry data were also obtained from the National Milk Producers Federation (NMPF) and the
National Pork Producers Council (NPPC). Much of the information on food processors is from
the Census of Manufactures data series. Section 4 and other sections of this report provide more
information on how these and other data sources contribute to EPA's analyses.
Other key sources of data and information to conduct analyses presented in this EA
include information from university experts and the published agricultural literature, as well as
information from the Natural Resources Defense Council, Clean Water Network, state
cooperatives and extension services, and state and EPA regional representatives and information
from previous EPA studies of animal feeding operations. Section 11 of this EA presents the
bibliographic citation of all materials used in this EA. Additional citations on data used for this
analysis are provided in the Development Document (USEPA, 2000a).
1.3 ORGANIZATION OF THE REPORT
This report is organized to allow those interested in the impacts on a specific industry
sector to find information easily. The sections of the report are as follows:
• Section 2 provides an industry profile of the animal feeding industries, including
livestock and poultry farms and food processors.
• Sections describes the proposed CAFO regulations, including a description of the
various technology options considered for the ELG and regulatory scenarios
• considered for the NDPES regulation.
• Section 4 presents the methodology that EPA is using to estimate compliance
costs and economic impacts and reviews the data used for this analyses in more
detail.
• Section 5 presents a summary of the estimated national, annual costs and the
economic impacts to regulated facilities of the proposed CAFO regulations.
• Sections 6, 7, and 8 provide additional industry profile data and economic analyses
of the regulatory impacts to the poultry, hog, and cattle and dairy sectors.
• Section 9 presents the results of EPA's Initial Regulatory Flexibility Analysis and
describes the possible impacts on small businesses.
• Section 10 presents a discussion of the regulatory costs and benefits pursuant to
Executive Order 12866 and UMRA.
1-9
-------
• Section 11 presents the references used throughout the report and its appendices.
• Appendix A presents a description of EPA's method to annualize costs and more
detailed information on the annualized costs used as inputs to EPA's CAFO level
economic analysis.
• Appendix B presents background information on EPA's market level model to
estimate supply and quantity changes, as well as changes in national level economic
output and total employment (direct and indirect).
» Appendix C presents a summary of a literature review of the price elasticity of
supply and demand for the livestock and poultry sectors.
• Appendix D presents the results of EPA's sensitivity analysis.
• Appendix E presents EPA's analysis of the cost-effectiveness of the proposed
regulatory revisions, in terms of pollutant removal effectiveness for nutrients and
other priority pollutants, and background information on the methods EPA used
for this analysis.
This report does not include a detailed presentation of the economic benefits that are expected to
accrue as a result of the proposed regulations. This analysis is provided in the Benefits Analysis
(USEPA, 2000d) that supports this rulemaking. The Development Document (USEPA, 2000a)
provides more detailed information on EPA's farm level costs estimated for this analysis.
1-10
-------
SECTION TWO
PROFILE OF THE LIVESTOCK AND POULTRY INDUSTRIES
The proposed revision of the CAFO regulations would affect operations that confine cattle
and calves, milking cows, hogs and pigs, and poultry, including broilers, egg laying chickens, and
turkeys.1 Businesses that contract out the raising or finishing phase of production might also be
affected by the proposed co-permitting requirements in the proposed CAFO regulations. Affected
businesses may include meat packing plants and poultry processing firms.
The purpose of this industry profile is to provide an overview of the livestock and poultry
industries in terms of their current activities, structure, and key trends since the promulgation of
the original CAFO regulations in the 1970s. Section 2.1 reviews the trends that have influenced
the perceived need to revise the existing regulations. Section 2.2 profiles livestock and poultry
producers at the farm production level. Key topics include the number and size of potentially
affected operations, characterization of the financial performance of the farming sector, and an
overview of the amount of manure and nutrients generated. Section 2.3 describes the industrial
organization and structure of these industries, focusing on the role of vertical integration and
coordination between the animal feeding and the processing sectors. Section 2.4 profiles
livestock and poultry processors within the manufacturing segments of these industries. Key
topics include the number and types of potentially affected operations and an overview of the
financial characteristics of these sectors. Finally, Section 2.5 further characterizes these industries
in terms of their economic output, overall supply and demand conditions, and industry
employment.
This overview covers all of the livestock and poultry sectors together. Other sections of
this report provide a detailed profile of each production sector, including an overview of the
poultry sectors (Section 6), the hog sector (Section 7), and the beef and dairy sectors (Section 8).
The information in this Section 2 profile is used to provide a baseline description of these sectors
and to aid in understanding the methodology used to analyze the potential economic impacts
associated with the proposed CAFO regulations.
2.1 RECENT TRENDS IN THE LIVESTOCK AND POULTRY INDUSTRIES
Major structural changes in the livestock and poultry industries have occurred since the
1970s, when the regulatory controls for CAFOs were first instituted. These changes are
discussed in more detail in the sections that follow and include, but are not limited to:
^ther types of animal confinement operations are not covered in this report, including operations that
confine sheep, lambs, goats, horses, and.other nontraditional animal species.
2-1 •
-------
• Increased number of animals produced annually.
» Fewer, larger, more industrialized operations that concentrate more animals and
also more manure in a single location.
• Geographical shifts in where animals are raised.
• Increased coordination between animal feeding operations and processing firms.
The continued trend toward fewer but larger operations, coupled with greater emphasis on
more intensive production methods and specialization, is concentrating more manure nutrients and
other animal waste constituents within some geographic areas. This trend has coincided with
increased reports of accidental large-scale spills from these facilities and has fueled concern that
manure runoff is contributing to the ovemutrification of certain vulnerable U.S. waterways.
Additional information on water quality impairment and risks associated with manure discharge
and runoff is provided in the Environmental Assessment that supports this proposed rulemaking
(USEPA, 2000b).
2.1.1 Increased Livestock and Poultry Production
Since the 1970s, when the existing regulations for CAFOs were first instituted, total .
consumer demand for meat, eggs, milk, and dairy products has continued to increase. To meet
this demand, U.S. livestock and poultry production has risen sharply, resulting in an increase in
the number of animals produced and the amount of manure and wastewater generated annually.
Increased sales from U.S. farms is particularly dramatic in the poultry sectors, as reported
in the Census of Agriculture (various years). In 1997, turkey sales totaled 299 million birds
(USDA/NASS, 1999a). This compares to 1978 when 141 million turkeys were sold for slaughter
(USDC, 1980). Broiler sales totaled 6.4 billion in 1997, up from 2.5 billion broilers sold in 1974
(USDC, 1976; USDA/NASS, 1999a). The existing CAFO regulations effectively do not cover
broiler operations due to the exclusion of operations that use dry manure management systems,
which comprise the majority of operations (USEPA, 2000a). Red meat production also rose
during the 1974-1997 period. The number of hogs and pigs sold hi 1974 totaled 79.9 million,
compared to 142.6 million by 1997 (USDC, 1976; USDA/NASS, 1999a). Sales data of fed cattle
(i.e., USDA's data category on "cattle fattened on grain and concentrates") for 1975 show that
20.5 million head were marketed. By 1997, fed cattle marketings totaled 22.8 million head
(USDA/NASS, 1999a). The total number of egg laying hens rose from 0.3 million birds hi 1974
to 0.4 million birds in 1997. The number of dairy cows on U.S. farms, however, dropped from
more than 10.7 million cows to 9.1 million cows over the same period (USDA/NASS, 1999a).
(Data that show increases in farm level production are also presented in Section 2.2.3).
2-2
-------
Not only are more animals produced and sold each year, but the animals are larger in size.
Efficiency gains have raised animal yields in'terms of higher average slaughter weight Likewise,
production efficiency gams at egg laying and dairy operations have resulted in higher per-animal
yields of eggs and milk. The average number of eggs produced per egg laying hen was 218 eggs
per bird in 1970 compared to 255 eggs per bird in 1997 (USDA/ERS, 1996c; USDA/NASS,
1999a). Average annual milk production rose from under 10,000 pounds per cow in 1970 to
more than 16,000 pounds per cow in 1997 (NMPF, 1999). hi the case of milk production, these
efficiency gains have allowed farmers to maintain or increase production levels with fewer
animals. Although animal inventories at dairy farms may be lower, however, this may not
necessarily translate to reduced manure volumes generated because higher yields are largely
attributable to improved and often more intensive feeding strategies that may exceed the animal's
ability for uptake, This excess is not always incorporated (i.e.5 digested) by the animal and may
be excreted (Smith, 1998).
2.1.2 Fewer, Larger, and More Industrialized Livestock and Poultry Operations
Recent trends across the U.S. livestock and poultry sectors are marked by a decline in the
number of operations due to ongoing consolidation in the animal production industry
(MacDonald, et al., 2000; McBride, 1997). Increasingly, larger, more industrialized, and highly
specialized operations now account for a greater share of all animal production. This
concentrates more animals, and thus more manure and wastewater, in a single location, and raises
the potential for significant environmental damages unless manure is properly handled.
USDA reports that there were 1.1 million livestock and poultry farms in the United States
in 1997, about 50 percent fewer than the 2.3 million farms reported in 1974 (USDA/NASS,
1999a; USDC, 1976). (See Table 2-1.) In general, farms are closing, especially smaller
operations that cannot compete with large-scale, highly specialized, often lower cost producers.
USDA reports that in a normal year, 3 percent to 4 percent of all farm operators discontinue
farming for a variety of financial and personal reasons (Stam, et al., 1991).2 Of these, USDA has
estimated that about 2 percent to 3 percent of farm exits are "involuntary" (i.e., due to
bankruptcies, foreclosures, debt repayment problems, or inadequate farm incomes (Bentley, et al.,
1989). Involuntary farm exits caused by financial stress vary considerably by farm size and
production region, and commodity produced (Bentley, et al., 1989).
Livestock and poultry production are increasingly dominated by larger operations,
although these operations constitute a small share of the total number of operations. As shown in
Figure 2-1, based on USDA data that are grouped by farm typology described in Table 2-2, about
2The mid-1980s, however, were a period of peak farm financial stress and were associated with a farm
exit rate of 5-6 percent per year (Stam, et al., 1991).
2-3
-------
Table 2-1. Number of Livestock and Poultry Operations (Year-end Animal Inventory), 1969-1997
Year
AH Beef
Fed
Cattle
Dairy
Hogs
Layers
Pullets
Broilers
Turkeys
Total Number of Operations (based on year-end inventory)
1969
1974
1978
1982
1987
1992
1997
%69-97
1969
1974
1978
1982
1987
1992
1997
$69-97
1,076,200
1,024,900
954,400
957,700
841,800
803,200
804,600
-25%
NA
NA
NA
240,000
190,000
147,200
110,600
NA
568,200
403,800
312,100
277,800
202,100
155,300
116,900
-79%
686,100
470,300
445,100
329,800
243,400
191,300
109,800
-84%
471,300
316,200
240,900
215,800
144,400
88,200
. 72,600
-85%
33,800
34,300
31,700
30,100
27,600
' 23,900
23,900
-29%
NA
12,800
18,900
25,400
19,000
13,800
12,100
NA
Sum
Total *
2,835,500
2,262,300
2,003,100
1,836,600
1,478,400
1,275,900
1,139,900
-60%
Average Number of Animals per Operation (based on year-end inventory)
NA
40
36
36
38
41
42
5%
NA
NA
NA
115
146
179
247
• NA
20
26
33
39
50
61
78
290%
81
97
130
168
215
301
558
588%
787
1,100
1,500
1,700
2,600
4,000
5,100
548%
71,987
73,300
96,500
116,800
157,800
226,700
281,700
291%
NA
2,100
1,900
1,800
3,900
6,400
8,600
NA
NA
NA
NA
.NA
NA
NA
NA
NA
NA=Not available
"'Sum Total" is across all operations shown (less 'Fed Cattle") and may double count some operations with more
than a single animal type. The average number of animals per operation is calculated from corresponding data o
the number of animals on site at year-end (inventory). The number of beef operations in 1969 is not available an
is estimated from the total number of cattle and calf operations less dairies and estimated number of steer and
heifer operations. Numbers include both grazing and confinement operations, as well as commercial and
noncommercial operations.
^Percentage change from the 1969 to the 1997 Census, except for All Beef, which is calculated from 1974 to 1997.
Source: USDA7NASS, 1999a, 1997 Census of Agriculture (Table 1: Historical Highlights). Data for turkeys and
fed cattle ("cattle fattened on grains and concentrates") are from USDA/NASS (1999a) and USDC, Census of
Agriculture (Census years 1969,1974,1978,1982, 1987, and 1992).
2-4
-------
10 percent of all farms—both crop and livestock operations with more than $250,000 in annual
sales— accounted for about 60 percent of the total value of production in 1997 (USDA/ERS,
2000h). Economies of size account for much of the growth in farm size (MacDonald, et al.,
2000; McBride, 1997). At the same time, cost and efficiency considerations are pushing farms to
become more specialized and intensive. Steep gains in production efficiency have allowed farmers
to produce more with fewer animals because of higher per-animal yields and quicker turnover of
animals between farm production and consumer market. As a result, annual production and sales
have increased, even though the number of animals on farms at any one time has declined (i.e., an
increase in the number of marketing cycles over the-course of the year allows operators to
maintain production levels with fewer animals at any given time, although the total number of
animals produced by the facility over the year may be greater). (Table 2-1.)
The increase in animal densities at operations is evident by comparing the average number
of animals per operation between 1974 and 1997, as derived from Census of Agriculture data.
These data are shown in Table 2-1. In the poultry sectors, the average number of birds across all
operations is four to five times greater in 1997 compared to 1974. In 1997, the number of
broilers per operation averaged 281,700 birds, up from 73,300 birds in 1974 (USDC, 1976;
USDA/NASS, 1999a). The average number of egg laying hens per operation rose from 1,100
layers to 5,100 layers per farm, and the average number of turkeys per operation rose from 2,100
turkeys to 8,600 turkeys over the 1974 to 1997 period (USDC, 1976; USDA/NASS, 1999a).
The average number of hogs raised per operation rose from under 100 hogs to more than 500
hogs between 1974 and 1997 (USDC, 1976; USDA/NASS, 1999a). The average number of fed
cattle and dairy cows per operation more than doubled during this period, rising to nearly 250 fed
cattle and 80 milking cows by 1997 (USDC, 1976; USDA/NASS, 1999a). .
The trend toward fewer, larger, and more industrialized operations has contributed to
large amounts of manure being concentrated within a single geographic location. The greatest
potential risk is from the largest operations with the most animals, given the sheer volume of
manure generated at these facilities. Larger, specialized facilities often do not have an adequate
land base for manure disposal through land application. A USD A analysis of 1997 Census data
shows that animal confinement operations with more than 1,000 AU account for-more than 42
percent of all confined animals but hold only -3 percent of all cropland on these operations (Letson
and Gollehon, 1996). As a result, large facilities need to store significant volumes of manure and
wastewater that have the potential, if not properly handled, to cause significant water quality
impacts. By comparison, smaller operations manage fewer animals and tend to concentrate less
manure nutrients at a single farming location. Smaller operations also tend to be more diversified,
engaging in both animal and crop production. These operations often have sufficient cropland
and fertilizer needs to land apply manure nutrients generated by the farm's livestock or poultry
business.
2-5
-------
Limited resource
Small
family Retirement
farms
-/safes
less than Residential
$250flOO)
_ Lowewalss
Farming
occupation
Higher-sales
' OfterfemBK rge
farms
, Veiy iaiye
Monfamily farms
^O^/ffff!!^^
J
.,, _l^
. .. , . , ..-. -.,:: |
• '-•- ":- •-•••-:. :,t-^'i'^',^r\
;-*•'.. .'-. . • ••-. '-- V^T...- V-^.i '.::••''• Jv- '^%!|
«
BR
: i — i : 1
HI
farms
F"~l Value of
['jl production
10
20
30
40 50
Percent of total
Figure 2-1. Share of Farms and Value of Production, by Typology Group, 1997
Source: USDA/ERS, 2000h. Farm typology groups are described in Table 2-2.
Another recent analysis from USDA confirms that as animal production operations have
become larger and more specialized, the opportunity to jointly manage animal waste and crop
nutrients has decreased, because these larger operations typically have inadequate land available
for utilizing manure nutrients (Gollehon and Caswell, 2000). Estimates by USDA indicate that
the amount of manure nitrogen produced by confinement operations increased about 20 percent
between 1982 and 1997, while average acreage on livestock and poultry farms declined (Gollehon
and Caswell, 2000). Overall, USDA estimates that cropland controlled by operations with
confined animals have the assimilative capacity to absorb about 40 percent of the calculated
manure nitrogen generated by these operations (Gollehon and Caswell, 2000). EPA expects that
these excess manure nutrients will need to be transported offsite for use as a fertilizer substitute
by other agricultural producers.
2.1.3 Geographic Shifts in Animal Production
During the 1970s, the majority of farming operations were concentrated in rural,
agricultural areas, and manure nutrients generated at animal feeding operations were readily
incorporated as a fertilizer in crop production. In an effort to reduce transportation costs and
streamline distribution between the animal production and food processing sectors, livestock and
2-6
-------
Table 2-2. USDA's Farm Typology Groups
Typology Groups
Limited-resource
Retirement
Residential/lifestyle
Farming occupation
/low sales
Farming occupation
/high sales
Description
Any small farm with gross sales less than $100,000, total farm assets
less than $150,000, and total operator household income (from all
sources) less than $20,000. Limited resource farmers may report
farming, a nonfarm occupation, or retirement as their major
occupation.
Small farms whose operators report they are retired (excludes limited-
resource farms operated by retired farmers).
Small farms whose operators report their major occupation as other
than farming, again excluding those with limited resources.
Small farms with sales less than $100,000, whose operators report
farming as their major occupation (excludes limited-resource farms
whose operators report farming as their major occupation).
Small farms with sales between $100,000 and $249,000, whose
operators report farming as their major occupation.
Additional Farm Groupings
Large family farms
Very large
family farms
Nonfamily farms
Farms with sales between $250,000 and $499,999.
Farms with sales of $500,000 or more
Farms organized as nonfamily corporations or cooperatives, as well as
farms operated by hired managers.
1999
Share /a
5.9%
13.9%
43.4%
22.4%
8.2%
3.6%
2.7%
<3% *
Source: USDA farm typology groups (USDA/ERS, 2000b).
'"Share of farms and households in 1999 for all farms, except non-family farms (USDA/ERS, 2000g)
"'Information for non-family is for 1997 (USDA/ERS, 2000h).
poultry operations have tended to cluster near slaughtering and manufacturing plants as well as
near end-consumer markets (McBride, 1997; USGAO, 1995; Kohls and Uhl, 1998). Ongoing
structural and technological change in these industries is also influencing where facilities operate
and is contributing to locational shifts between the more traditional production regions and the
more emergent regions (Schrader, 1998; Lawrence et al., 1998; Kohls and Uhl, 1998; McBride,
1997; MacDonald, et al., 2000).
Operations in more traditional producing states tend to grow both livestock and crops and
tend to have adequate cropland for land application of manure. Operations in these regions also
tend to be smaller in size (McBride, 1997; Outlaw et al., 1996; USDA/NASS, 1999a). In
contrast, confinement operations in more emergent areas, such as hog operations in North
Carolina or dairy operations in the Southwest, tend to be larger in size and more intensive types of
operations (USDA/NASS, 1999a; Schrader, 1998; Lawrence et al., 1998; Outlaw, et al., 1996;
2-7
-------
McBride, 1997). These operations tend to be more specialized and often do not have adequate
land for application of manure nutrients (McBride, 1997; Letson and Gollehon, 1999; Gollehpn
and Caswell, 2000). Production is growing rapidly in these regions due to competitive pressures
from more specialized producers who face lower per-unit costs of production (McBride, 1997).
This may be shifting the flow of manure nutrients away from more traditional agricultural areas,
often to areas where these nutrients cannot be easily absorbed.
As reported by Census data, shifts in where animals are grown is especially pronounced in
the pork sector (McBride, 1997; Iowa State University, 1998; Martinez, 1999; MacDonald, et al.,
2000). Traditionally, Iowa has been the top ranked pork producing state. Between 1982 and
1997, however, ,the number of hogs raised in that state remained relatively constant with a year-
end inventory average of about 14.2 million pigs (USDC, 1984; USDA/NASS, 1999a). In
comparison, year-end hog inventories in North Carolina increased from 2.0 million pigs in 1982 to
9.6 million pigs in 1997 (USDC, 1984; USDA/NASS, 1999a). This locational shift has coincided
with reported nutrient enrichment of the waters of the Pamlico Sound in North Carolina (USEPA,
2000b). Growth in hog production also occurred in other emergent areas, including South
Dakota, Oklahoma, Wyoming, Colorado, Arizona, and Utah. Meanwhile, production dropped in
Illinois, Indiana, Wisconsin, and Ohio (USDC, 1994, and 1989; USDA/NASS, 1999a).
The dairy industry has seen similar shifts in where milk is produced, moving from the more
traditional Midwest and Northeast states to the Pacific and Southwestern states (McBride, 1997;
El-Osta and Johnson, 1998). Between 982 and 1997, the number of milk cows hi Wisconsin
dropped from 1.9 million to 1.3 million (USDC, 1984; USDA/NASS, 1999a). Milk cow
inventories have also declined in other traditional states, including Illinois, Indiana, Iowa,
Minnesota, Missouri, New York, Pennsylvania, Ohio, Connecticut, Maryland, and Vermont.
During the same period, milk cow inventories in California rose from 0.9 million in 1982 to 1.4
million in 1997. In 1994, California replaced Wisconsin as the top milk producing state
(USDA/NASS, 1999c). Milk cow inventories have also increased in Texas, Idaho, Washington,
Oregon, Colorado, Arizona, Nevada, and Utah (USDC, 1994, and 1989; USDA/NASS, 1999a).
These locational shifts have coincided with reported nutrient enrichment of waters, including the
Puget Sound and Tillamook Bay in the northwest, the Everglades in Florida, and Erath County in
Texas, and also elevated salinity levels due to excess manure near milk production areas in
southern CaUfornia's Chino Basin (USEPA, 2000b).
2.1.4 Increased Farmer-Processor Linkages
Over the past few decades, closer ties have been forged between growers and various
industry middlemen, including packers, processors, and cooperatives (USDA/ERS, 1996c).
Increased integration and coordination is being driven by the competitive nature of agricultural
production and the dynamics of the food marketing system, in general, as well as seasonal
fluctuations of production, perishability of farm products, and inability to store and handle raw
farm output. Closer ties between the animal production facility and processing firms—either
2-8
-------
through contractual agreement or through corporate ownership of CAFOs—raises questions of
who is responsible for ensuring proper manure disposal and management at the animal feeding
site. This is especially true given current trends toward an increasing number of large animal
confinement operations and the resultant need for increased animal waste management. As farms
become larger and more specialized, they may contract out some phases of the production
process. The contract farm might be less able to make decisions about environmental concerns
and to cover the costs of addressing these concerns (USDA/ERS, 1996c). Since environmental
controls may raise the cost of production to the farmer, the farmer may wait to implement best
management practices until the contractor .(processor) specifies and/or compensates the farmers
for the additional costs (USDA/ERS, 1996c).
Farmers and ranchers have long used contracts to market agricultural commodities.
However, increased use of production contracts is changing the organizational structure of the
individual industries and may raise policy questions regarding ownership responsibility as well as
environmental concerns (USDA/ERS, 1996c). Under a production contract, a business other than
the feedlot where the animals are raised and housed, such as a processing firm, feed mill, or
animal feeding operation, may own the animals and may exercise further substantial operational
control over the operations of the feedlot. In some cases, the processor may specify in detail the
production inputs used, including the genetic material of the animals, the types of feed used, and
the production facilities where the animals are raised (USDA/ERS, 1996c; Martinez, 1999; Ogishi
and Zilberman, 1999). The processor may also influence the number of animals produced at a site
(Ogishi and Zilberman, 1999). In general, these contracts do not deal with management of
manure and waste disposal (USDA/ERS, 1996c; Martinez, 1999). Recently, however, some
processors have become increasingly involved in how manure and waste is managed at the animal
production site. This information is available in the rulemaking record (for example: Associated
Press, 2000; Huslin, 2000a and 2000b; Montgomery, 2000; Goodman, 1999).
The use of production contracts in the livestock and poultry industries varies by
commodity group. Information from USDA indicates that production contracts are widely used
in the poultry industry and dominate broiler production (USDA/ERS, 1996c and 1999a),
Production contracting is becoming increasingly common in the hog sector, particularly for the
finishing stage of production in regions outside the Corn Belt (USDA/ERS, 1999a; Schrader,
1998; Lawrence et al., 1998). In the poultry sectors, vertical integration has progressed to the
point where large, multifunction producer-packer-processor-distributor firms are the dominant
force in poultry and egg production and marketing (Aust, 1997; Kohls and Uhl, 1998).
Production contracting has played a critical role hi the growth of integrators in the poultry
sectors. Data from USDA on animal ownership at U.S. farms demonstrates the use of production
contracts in these sectors. In 1997, USDA reported that 97 percent of all broilers raised on U.S.
farms were, not owned by the farmer. In the turkey and egg laying sectors, use of production
contracts is less extensive, since 70 percent and 43 percent of all birds in these sectors,
respectively, were not owned by the farmer (USDA/ERS, 1999a). In the hog sector, data from
USDA indicate that production contracting may account for 66 percent of hog production among
larger producers in the Southern and Mid-Atlantic states (USDA/ERS, 1999a). This differs from
2-9
-------
the Midwest, where production contracting accounted for 18 percent of hog production in 1997
(USDA/ERS, 1999a).
By comparison, production contracts are not widely used in the beef and dairy sectors.
Data from USDA indicate that less than 4 percent of all beef cattle and 1 percent of all milking
cows were not owned by the farmer in 1997 (USDA/ERS, 1999a). However, production
contracts are used in these industries that specialize in a single stage of livestock production, such
as to "finish" cattle prior to slaughter or to produce replacement breeding stock. However, this
use constitutes a small share of overall production across all producers. More detailed
information on contracting use in animal agriculture is discussed in Section 2.3.
To further examine the linkages between the animal production facility and the food
processing firms, and to evaluate the geographical implications of this affiliation, EPA conducts
an analysis that shows a relationship between areas of the country with an excess of manure
nutrients from animal production operations and areas with a large number of meat packing and
poultry slaughtering facilities. This manure—if land applied—would be in excess of crop uptake
needs and result in over application and enrichment of nutrients. Across the pork and poultry
sectors, this relationship is strongest in northwest Arkansas, where EPA estimates a high
concentration of excess manure nutrients and a large number of poultry and hog processing
facilities. By sector, EPA's analysis shows that there is excess poultry manure nutrients and a
large number of poultry processing plants in the Delmarva Peninsula in the mid-Atlantic, North
Carolina, northern Alabama, and also northern Georgia. In the hog sector, the analysis shows
excess manure nutrients and a large number of meat packing plants in Iowa, Nebraska and
Alabama. The analysis also shows excess manure nutrients from hogs in North Carolina, but
relatively fewer meat packing facilities, which is likely explained by continuing processing plant
closure and consolidation in that state. More information on this analysis is provided in the
rulemaking record. The results of this research is contained in the record (DCN 20786).
2.2 CHARACTERISTICS OF ANIMAL CONFINEMENT OPERATIONS THAT
MAY BE AFFECTED BY THE PROPOSED CAFO REGULATIONS
This report examines the economic effects of the proposed CAFO regulations on
agricultural operations that confine livestock and poultry. This section provides an overview of
the animal production sectors, focusing on facilities that confine animals and identifying those
production facilities that are likely to be affected by the proposed requirements.
Section 2.2.1 discusses how EPA identifies the number of operations that may be affected
by the proposed CAFO regulations. This section first identifies all livestock and poultry
operations, then estimates which ones are likely to be confined operations and thus potentially
affected operations, and finally estimates the number of such operations likely to be considered
CAFOs, given the size definitions in the proposed CAFO regulations. A brief summary of the
financial characteristics of U.S. livestock and poultry operations is provided in Section 2.2.2 (and
2-10
-------
is discussed in more detail in other sections of this report). Section 2.2.3 characterizes the
amount and nutrient content of manure generated at confined livestock and poultry operations.
2.2.1 Identification and Number of Affected CAFOs
There are three steps involved in determining the number of CAFOs that may be affected
by the proposed regulations. First, EPA identifies all livestock and poultry operations using
USDA data. Second, EPA estimates the total number of operations that confine animals, using
available Census data and other supplemental information from .USD A and industry. Third, based
on the regulatory thresholds being proposed, EPA identifies the number of operations that meet
the CAFO definition or may be designated as CAFOs by the Permitting Authority, using farm size
distribution data from USDA and other information. These steps are described in the following
sections.
2.2.1.1 All Livestock and Poultry Operations
The agricultural sectors that are the focus of the proposed CAFO regulations, identified by
North American Industry Classification System (NAICS)3 code, include:
• Cattle feedlots, NAICS 112112 [includes veal] (SIC 021.1, beef cattle feedlots).
• Beef cattle ranching and farming, NAICS 112111 (SIC 0241, dairy heifer
replacement farms).
• Dairy cattle and milk production, NAICS 11212 (SIC 0241, dairy farms).
• Hog and pig farming, NAICS 11221 (SIC 0213, hogs).
• Broilers and other meat-type chickens, NAICS 11232 (SIC 0251, broiler, fryer,
and roaster chickens).
• Turkey production, NAICS 11233 (SIC 0253, turkey and turkey eggs).
• Chicken egg production, NAICS 11231 (SIC 0252, chicken eggs).
As shown in Table 2-1, USDA reports that there were 1.1 million livestock and poultry
farms in the United States in 1997 (USDA/NASS, 1999a). This number includes all operations
3NAICS recently "replaced the SIC (Standard Industrial Classification) system.
2-11
-------
that raise beef, dairy, pork, broilers, egg layers, and turkeys, and includes both confinement and
non-confinement (grazing and rangefed) production. These data on the number of farms include
both commercial and non-commercial operations. (USDA generally defines a commercial
operation to include operations with annual farm revenues of more than 50,000 per year
(USDA/ERS, 1997b)).
The primary source of data on the number of livestock and poultry operations is USDA's
1997 Census of Agriculture (Census). The Census is a complete accounting of United States
agricultural production and is the only source of uniform, comprehensive agricultural data for
every county in the nation (USDAMASS, 2000a). NASS conducts the Census every 5 years.4
The Census is a mail questionnaire that is sent to a list of known U.S. operations from which
$1,000 or more of agricultural products were produced and sold or normally would have been
sold during the census year (USDA/NASS, 2000b). Census survey respondents are on their own
to interpret its questions.
Aggregated Census data are readily available from USDA. In general, the published
compendium provides summary inventory and sales data for the nation and for states. The
Census database itself, however, contains more detailed information that can be aggregated by
more precise facility size groupings, such as the regulatory thresholds specified by EPA for the
proposed CAFO regulations. USDA periodically publishes summary data from its databases and
provides customized analyses of the data to the public and other government agencies. EPA
obtained these data through a formal request to USDA. The requested data summaries that EPA
uses for its analyses are compiled by NASS, with the assistance of staff at USDA's NRCS who
developed a methodology for identifying farms likely to be CAFOs based on reported survey
information and developed estimates of animal units on these operations based on reported data
(described in the Development Document (USEPA, 2000a)). This methodology allows USDA to
perform special tabulations of the data to obtain information on the characteristics of facilities at
specific size thresholds for each sector, which were provided to the EPA and other government.
agencies. All data provided to EPA is sufficiently aggregated to ensure the confidentiality of an
individual farming operation.
In some cases, data obtained by USDA on the number of farms by inventory size
distribution do not always correspond with the facility size definitions examined by EPA. Where
data were not available in the desired size ranges, the data were linearly interpolated from
available data (USEPA, 2000a). USDA data also are not available on the number and distribution
of poultry operations with wet manure management systems and are instead estimated by EPA
using available data and supplemental information from industry experts and agricultural extension
agency personnel (USEPA, 2000a). Also, limited information on the number of farms that raise
more than a single animal type is available. To the extent that combinations of animal types are
located at livestock and poultry operations, these counts may be overstated.
4In prior years, the Census was conducted by the Department of Commerce's Bureau of the Census.
2-12
-------
USDA Census data report the number and size of livestock and poultry operations as of
year-end (December 31) and may not adequately reflect seasonal fluctuations in inventory and the
number and timing, of animals sold. EPA used the Census data to calculate average herd size over
the year. The resultant industry summary data are based on both reported inventory and sales,
adjusted by expected turnovers. This approach is consistent with that developed by USDA to
estimate potential manure nutrient loadings from animal agriculture (Lander, et al., 1998). More
information is provided in the Development Document (USEPA, 2000a).
2.2.1.2 Animal Confinement Operations
Census data on the number of livestock and poultry farms include both confinement and
non-confinement operations. However, only operations that confine animals are subject to the
proposed CAFO regulations.5 For many of the animal sectors, it is not possible to precisely
determine what proportion of the total livestock operations are confinement operations and what
proportion are grazing operations only. EPA has estimated the number of AFOs using available
data and other information from the Census as well as other USDA and industry publications
'(USDA/NASS, 1999b and 1999c and 1999d and 1998b; USDA/APHIS, 1995b; NPPC, 1998).
Data on the number of beef and hog operations that raise animals in confinement are available
from USDA. Since most large dairies have milking parlors, EPA assumes that all dairy operations
are potentially confinement operations. In the poultry sectors, there are few small non-
confinement operations and EPA assumes that all poultry operations confine animals. EPA's
analysis focuses on the largest facilities in these sectors only. The data and assumptions used to
derive EPA's estimates of the number of affected confinement operations have undergone
extensive review by USDA personnel. Detailed information on how EPA estimated the number
of AFOs that may be subject to the proposed CAFO regulations can be found in the Development
Document (USEPA, 20QOa).
Using available 1997 data from USDA, EPA estimates that there are about 376,000 AFOs
that raise or house animals in confinement, as defined by the existing regulations (Table 2-3).
Table 2-3 presents the estimated number of AFOs and the corresponding animal inventories for
1997 across select size groupings. These estimates are based on the number of "animal units"
(AU) as defined in the existing regulations at 40 CFR 122 (Appendix B), with the addition of the
revisions that are being proposed for immature animals and chickens.6 Data shown in Table 2-3
are grouped by operations with more than 1,000 AU and operations with fewer than 300 AU.
5Under the existing regulations, confinement operations are considered to be CAFOs where animals have
been, are, or will be stabled or confined and fed and maintained for a total of at least 45 days in any 12-month
period. These confinement areas must not sustain crops, vegetation, forage growth, or post-harvest residues in th
normal growing season.
6 As defined for the proposed CAFO regulations, one AU is equivalent to one slaughter or feeder cattle,
calf or heifer;.0.7 mature dairy cattle; 2.5 hogs (over 55 pounds) or 5 nursery pigs; 55 turkeys; and 100 chickens
regardless of the animal waste system used.
2-13
-------
As shown in Table 2-3, there were an estimated 12,660 AFOs with more that 1,000 AU in
1997 that accounted for about 3 percent of all confinement operations. In most sectors, these
larger-sized operations account for the majority of animal production. For example, in the beef,
turkey and egg laying sectors, operations with more than 1,000 AU accounted for more than 70
percent of all animal inventories in 1997; operations with more than 1,000 AU accounted for
more than 50 percent of all hog, broiler, and heifer operations (Table 2-3). In contrast, operations
with fewer than 300 AU accounted for 90 percent of all operations, but a relatively smaller share
of animal production.
USDA data on the total number of operations, shown in Table 2-3, include corporate-
owned operations. Despite industry trends toward larger facility size, increased specialization and
Table 2-3. Number of AFOs and Animals On Site, by Size Group, 1997
Sector/
Size Category
Cattle
Veal
Heifers
Dairy
Hogs: GF b/
Hogs: FF w
Broilers
Layers: wet d
Layers: dry "
Turkeys
Total *
All AFOs
>1,000 AU "
<300AU
number of operations
106,080
850
1,250
116,870
53,620
64,260
34,860
3,110
72,060
13,720
375,700
2,080
10
300
1,450
1,670
2,420
3,940
50
590
370
12,660
102,000
640
200
109,740
48,700
54,810
20,720
2,750
70,370
12,020
336,590
Total
>1000AU
<300 AU
number of animals on site, 1,000s
26,840
270
850
9,100
18,000
38,740
1,905,070
392,940
112,800
, NA
22,790
10
450
2,050
9,500
21,460
1,143,040
275,060
95,880
NA
2,420
210
80
5,000
2,700
5,810
476,270
58,940
2,260
NA
Source: USEPA, 2000a. Derived by USDA from published USDA/NASS data, including 1997 Census of
Agriculture. In some cases, available data are used to interpolate data for some AU size categories. Data for vea
and heifer operations are estimated by EPA (USEPA, 2000a). May not add due to rounding. NA=Not available.
"'As defined for the proposed CAFO regulations, one AU is equivalent to one slaughter or feeder cattle, calf or
heifer; 0.7 mature dairy cattle; 2.5 hogs (over 55 pounds) or 5 nursery pigs; 55 turkeys; and 100 chickens
regardless of the animal waste system used.
w"Hogs: FF" are farrow-finish (includes breeder and nursery pigs); "Hogs: GF" are grower-finish only.
'"'Layers: wef' are operations with liquid manure systems; "Layers: dry" are operations with dry systems.
^"Total AFOs" eliminates double counting of operations with mixed animal types. Based on survey level Census
data for 1992, operations with mixed animal types account for roughly 25 percent of total AFOs.
2-14
-------
scale efficiencies, and closer farmer-processor linkages (discussed in Section 2.1.4), the majority
of facilities remain independently owned and operated. USDA reports that there were roughly
3,000 incorporated livestock and poultry operations that were not family held in 1997,
constituting less than one-half of one percent of all operations (USDA/NASS, 1999a). Other
USDA data indicate that non-family farms (both crop and livestock operations) account for less
than 3 percent of all U.S. farms (see Table 2-2 and Figure 2-1).
2.2.1.3 CAFOs Subject to the Proposed Regulations
Table 2-4 presents the estimated number of operations that would be defined as a CAFO
under each of the two alternative regulatory alternatives. The "two-tier structure" would define
as CAFOs all animal feeding operations with more than 500 AU. The "three-tier structure"
.would define as CAFOs all animal feeding operations with more than 1,000 AU and any operation
with more than 300 AU, if they meet certain "risk-based" conditions, as defined in Section VT1 of
the preamble (also summarized in Section 3 of this report). Table 2-4 presents the number of
AFOs in terms of number of operations with more than 1,000 AU and operations for each co-
proposed middle category (operations with between 500 and 1,000 AU and between 300 and
1,000 AU, respectively). .
Table 2-4 Number of Potential Onerations Defined as CAFOs by Select Regulatory Alternative, 1997
Sector/Size .
Category
Cattle
Veal
Heifers
Dairy
Hogs: GF
Hogs: FF
Broilers
Layers: wet
Layers: dry
Turkeys
Total
"Two-Tier"
>300
AU
>500
AU
>750
AU
(# operations)
4,080
210
1,050
7,140
4,920
9,450
14,140
360
1,690
2,100
39,320
3,080
90
800
3,760
2,690
5,860
9,780
360
1,280
1,280
25,540
2,480
40
420
2,260
2,300
3,460
7,780
210
1,250
740
19,100
>300
AU .
4%
25%
84%
6%
9%
15%
41%
12%
2%
15%
10.5%
>500
AU
(% total)
3%
10%
64%
3%
5%
9%
28%
12%
2%
9%
6.8%
>750
AU
2%
4%
34%
2%
4%
5%
22%
7%
2%
5%
5.1%
"Three-Tier"
>300 AU
(#)
3,210
140
980
6,480
2,650
5,700
13,740
360
1,650
2,060
31,930
(% total)
3%
16%
78%
6%
5%
9%
39%
12%
2%
15%
8.5%
Source: See Table 2-3.
2-15
-------
Based on available USDA data for 1997, EPA estimates that both proposed alternative
structures would regulate about 12,660 operations with more than 1,000 AU, accounting for
operations with more than a single animal type (USEPA, 2000a).7 The two alternatives differ in
the manner in which operations with less than 1,000 AU would be defined as CAFOs and,
therefore, subject to regulation, as described in Section 3 of this report. As shown in Table 2-4,
in addition to the 12,660 facilities with more than 1,000 AU, the two-tier structure would regulate
an additional 12,880 operations with between 500 and 1,000 AU (USEPA, 2000a).
Under the three-tier structure, an estimated 39,330 operations would be subject to the
proposed regulations (10 percent of all AFOs), estimated as the total number of animal
confinement operations with more than 300 AU. See Table 2-4. Of these, EPA estimates that a
total of 31?930 AFOs would be defined as CAFOs (9 percent of all AFOs) and would need to
obtain a permit (Table 2-4), while an estimated 7,400 operations would certify that they do not
need to obtain a permit. Among those operations needing a permit, an estimated 19,270
operations have between 300 to 1,000 AU.
As shown in Table 2-4, the three-tier structure would regulate an additional 6,400
operations, compared to the two-tier structure. Overall, the three-tier structure would define as
CAFOs more operations in all sectors, with the exception of the hog sector. Including mixed
operations, the three-tier structure would define as CAFOs 5,100 more poultry operations, about
2,700 more dairies, and 400 more cattle operations, but 200 fewer hog operations (Table 2-4),
than the two-tier structure.
EPA estimates the number of operations that may be defined as CAFOs under the three-
tier structure using available information and compiled data from USDA, State Extension experts,
and agricultural professionals. These estimates rely on information about the percentage of
operations in each sector that would be impacted by the "risk-based" criteria described in Section
VTI. In some cases, this information is available on a state or regional basis only and is
extrapolated to all operations nationwide. EPA's estimates reflect information from a majority of
professional experts in the field. Greater weight is given to information obtained by State
Extension agents, since they have broader knowledge of the industry in their state. More detailed
information on how EPA estimated the number of operations that may be affected by the
proposed regulations under the three-tier structure is available in the Development Document
(USEPA, 2000a).
EPA is also requesting comment on two additional options for the scope of the rule. One
of these is an alternative two-tier structure with a threshold of 750 AU. Under this option, an
estimated 19,100 operations, adjusting for operations with more than a single animal type, would
be. defined as CAFOs (Table 2-4). This represents about 5 percent of all CAFOs, and would
affect an estimated 2,930 beef, veal, and heifer operations, 2,260 dairies, and 5,750 swine and
71992 Census data indicate that approximately 200 operations with more than 1,000 AU have multiple
animal types, so a corresponding reduction in these operations is calculated to adjust for mixed operations.
2-16
-------
9,980 poultry operations (including mixed operations). Under the other alternative—a variation
of the three-tier structure that is being co-proposed—the same 39,320 operations with 300 AU or
greater would potentially be defined as CAFOs. However, the certification conditions for being
defined as a CAFO would be different for operations with 300 to 1,000 AU (described later in
Section "VH of the preamble). EPA has not estimated how many operations would be defined as
CAFOs under this alternative three-tier approach, although EPA expects that it would be fewer
than the 31,930 estimated for the three-tier approach being proposed today. If after considering
comments, EPA. decides to further explore this approach, it will conduct a full analysis of the
number of potentially affected operations.
EPA does not anticipate that many AFOs with less than 500 AU (two-tier structure) or
300 AU (three-tier structure) will be subject to the proposed requirements. In the past 20 years,
EPA is aware of very few AFOs that have been designated as CAFOs. Based on available USDA
analyses that measure excessive nutrient application on cropland in some production areas and
other farm level data by sector, facility size and region, EPA estimates that designation may bring
an additional 50 operations under the proposed two-tier structure (500 AU threshold) each year
nationwide. EPA assumed this estimate to be cumulative such that over a 10-year period
approximately 500 AFOs may become designated as CAFOs and therefore subject to the
proposed regulations. EPA expects these operations to consist of beef, dairy, farrow-finish hog,
broiler and egg laying operations that are determined to be significant contributors to water
quality impairment (Table 2-5). Table 2-5 also shows EPA's estimates of designated facilities
under an alternative two tier structure at the 750 AU threshold. Under the three-tier structure,
EPA estimates that fewer operations would be designated as CAFOs, with 10 dairy and hog
operations may be designated each year, or 100 operations over a 10-year period (Table 2-5).
For the purpose of this analysis, EPA assumes that all potentially designated operations 'are
located in the more traditional farm production regions in the Midwest and Northeast.
EPA expects that the proposed CAFO regulations would mainly affect livestock and
poultry operations that confine animals. In addition to CAFOs, however, the proposed
regulations may also affect businesses that contract out the raising or finishing production phase
to a CAFO but exercise "substantial operational control" over the CAFO (described in Section 3
of this report). As discussed later in Sections 2.3 and 2.4, EPA estimates that 94 meat packing
plants that slaughter hogs and 270 poultry processing facilities may be subject to the proposed co-
permitting requirements. Other types of processing firms, such as further processors, food
manufacturers, dairy cooperatives, and Tenderers, are not expected to be affected by the co-
permitting requirements since these operations are further up the marketing chain and do not
likely contract with CAFOs to raise animals. Fully vertically integrated companies (e.g., where
the packer owns the CAFO) are not expected to require a co-permit since the firm as the owner of
the .CAFO would require only a single permit.
2-17
-------
2-5 Number of Potential Oberations Designated as CAFOs by Select Regulatory Alternative, 1997
Sector/Size
Category
Cattle
Veal
Heifers
Dairy
Hogs: GF
Hogs: FF
Broilers
Layers: wet
Layers: dry
Turkeys
Total
"Two-Tier"
>300 AU
>500 AU
>750 AU
"Three-Tier"
>300AU
(# operations)
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
40
0
0
220
0
200
20
20
0
0
500
60
0
0
380
0
350
30
30
0
0
850
0
0
o
50
0
50
0
0
0
0
100
Source: See Table 2-3. Estimates are shown projected over a 10-year period. ND=Not Determined.
EPA also expects that crop fanners who receive manure from CAFOs would be affected
under one of the two co-proposed options relating to offsite management of manure. EPA's
Development Document (USEPA, 2000a) documents how EPA estimated the number of
potentially affected crop producers. These estimates are presented in Section 5 as part of EPA's
overall analysis.
2.2.2 Financial Characteristics of Livestock and Poultry Farms
Table 2-6 provides a summary of USDA's financial performance classification by farm
typology groups across all crop and livestock producers for 1999 (USDA/ERS, 2000g). This
compilation is based on USDA's Agricultural Resource Management Study (ARMS) data. (This
database in described in greater detail in Section 4 of this report.) Farm financial information
presented in this section and used for this analysis does not include off-farm income, which may
account for between 16 percent of total household income at USDA-recognized "very large
family farms" (operations with more than $500,000 in annual sales) to 130 percent of total
household income at "small family farms," as defined in Table 2-2 (USDA/ERS, 2000d).
USDA classifies farm financial performance as "favorable" or "vulnerable", using two.
measures (USDA/ERS, 1997e and 1997a). The first measure is the debt-to-asset ratio, a ratio
2-18
-------
that compares total debt to total assets and is useful for determining the relative debt burden a
farm might be under. The second is net farm income, which if negative, could indicate relatively
poor financial performance. Used together, these terms define a farm's vulnerability to closure.
Vulnerable farms are those with both negative net farm income and high levels of debt, indicated
by a debt-to-asset ratio over 40 percent. (See Section 4.2.5 for more information on USDA's
farm performance classification.)
Using this performance classification, Table 2-6 shows that about 4 percent of all farms
are considered by USDA to be in a financially vulnerable position (USDA/ERS, 2000g). These
farms are highly leveraged and may experience income deficiencies (USDA/ERS, 1997b; Sommer,
et al., 1998).
Across U.S. commercial farms—defined by USDA as operations with revenues of
$50,000 or more per year—the share of operations that are considered financially vulnerable is
lower, estimated at less than one percent (USDA/ERS, 1997b). Additional information across all
commercial farms (both crop and livestock enterprises) indicate that net farm income averaged
$57,600 per operation (USDA/ERS, 2000J). Including non-commercial operations, farm income
tends to average much lower (Table 2-6). On average, net farm income was relatively stable for
commercial farms during the early 1990s (USDA/ERS, 1997b). Average debt-to-asset ratio
across all commercial farm businesses was 14 percent in 1999, indicating relatively low debt
burdens (USDA/ERS, 2000J). Average return on assets at commercial farm businesses was 3
percent; operating profit margins averaged 11 percent (USDA/ERS, 2000J). Including non-
commercial operations, average financial statistics for the farming sector as a whole (shown in
Table 2-6) tend to be more variable and generally lower overall (USDA/ERS, 1993 and 1996a).
The following summarizes the financial conditions of operations in the livestock and
poultry sectors. Financial conditions vary among these sectors, as well as by size and region
within these sectors, in some cases. Overall, available USDA ARMS data on the baseline
financial conditions of the livestock and poultry sectors that will be affected by the proposed
CAFO regulations indicate that operations in these sectors, on average, are not in a vulnerable
financial position. As is discussed below, on average, these operations have positive net farm
income and acceptable debt-to-asset ratios (less than 40 percent) indicating a low potential for
cash flow problems and a low relative risk of insolvency. Additional information is provided in
Sections 6, 7 and 8 of this report.
Beef operations experienced poor profitability in the 1990s due to falling prices and
declining receipts, in spite of increasing production.8 In 1997, net farm income averaged $10,100
8ARMS data do not differentiate between fed cattle operations and cow-calf operations. All data presented
here are for all beef operations only.
2-19
-------
Table 2-6 Financial Performance bv Farm Tvpolosv Group, All Crop and Livestock Production, 1999
Item
•
Share farms/
households
Small family farms
Limited-
resource
Retire-
ment
5.9
13.9
Resid./
lifestyle
43.4
Farming
occupation
Low
High
Large
family
farms
Very
Large
family
farms
(percent)
22.4
8.2
3.6
2.7
AH
100.0
Percent Distribution by Financial Performance
Favorable
Marginal
Income
Marginal
Solvency
Vulnerable
54.2
38.4
d
d
71.4
,27.8
d
d
57.9
34.1
*2.2
5.9
62.0
3-2.2
*3.4
2.3
71.4
15.2
9.2
4.3
71.1
14.1
8.6
6.2
60.6
13.0
22.4
3.9
Percent Distribution by Typology Group
Favorable
Marginal
Income
Marginal
Solvency
Vulnerable
5.2
7.5
d
d
15.9
12.8
d
d
40.4
49.0
*27.3
60.9
22.3
23.9
*21.7
12.3
9.4
4.1
21.5
8.4
4.1
1.7
8.8 .
5.3
2.7
1.2
17.4
2.5
Other performance measures
Net farm
income (S)
Debt-to-
asset (%)
Return
on assets (%)
Operating
profit margin
(%)
** -800
7.8
-12.2
-96.3
4,400
2.0
-1.2
-25.2
2,100
9.5
-1.6
-24.7
5,400
6.4
-2.5
-29.9
33,700
13.5
*0.7
*3.5
64,700
16.7
3.2
11.8
220,500
20.0
8.7
18.8
62.1
30.2
3.5
4.2
100.0
100.0
100.0
100.0
13,800
10.5
**0.2
** 1.4
Source: USDA/ERS, 2000g. Dollars are rounded to nearest hundred.
Farm types are defined in Table 2-2. Financial performance classification is defined in Section 4.2.5.
d=data suppressed due to insufficient observations.
*=standard error is between 25-50% of the estimate. **=standard error is between 50-75% of the estimate.
2-20
-------
per operation, 20 percent lower than 1994, not accounting for inflation (USDA/ERS, 1999a and
1997b). Average debt-to-asset ratios in this sector ranged from 9 percent to 13 percent in 1997,
depending on operation size—a reduction in debt burden since 1994 when the average farm had a
debt-to-asset ratio of 16 percent (USDA/ERS, 1999a and 1997b). Average returns on assets
(derived by EPA using 1997 ARMs data on average revenues to average assets) ranged from 2
percent to 4 percent, depending on facility size. Additional financial information on the beef
sector is provided in Section 8.
Among dairy operations, net farm income remained relatively stable during the mid- to late
1990s. USDA reports in 1997 net dairy farm income averaged $36,600 per operation, slightly
lower than in 1994 (USDA/ERS, 1999a and 1997b). Average debt-to-asset ratios ranged from 17
percent to 26 percent, depending on size (USDA/ERS, 1999a). Differences in debt-to-asset ratio
region were generally modest. Dairy operations, however, are among the most profitable of the
affected industry sectors. Returns on assets at dairy farms in 1997 averaged from 4 percent to 9
percent, depending on facility size (calculated by EPA from ARMS data as average revenues to
average assets). Additional financial information on the dairy sector is provided in Section 8.
Hog operations experienced low profitability and high debt in the mid-1990s, which
caused substantial financial stress among producers, though not reaching critical levels. In 1997,
debt and profitability improved on average. Net farm income averaged $21,700 per operation in
1997, a 10 percent increase from the 1994 average, not accounting for inflation (USDA/ERS,
1999a and 1997b). Average debt-to-asset ratios among hog operators ranged from 15 percent to
39 percent, depending on size, and also vary by production region, tending to be lower in the
Mid-Atlantic compared to the more traditional Midwest region (USDA/ERS, 1999a). High debt
levels were prevalent in 1994, with mean debt-to-asset ratio among cpmmercial hog farms
averaging 24 percent (USDA/ERS, 1997b). By 1997, the average debt-to-asset ratio was 18
percent (USDA/ERS 1999a). This decrease in debt levels, along with relatively high returns on
assets, indicates some upturn in the hog sector between 1994 and 1997. This may have buffered
the period of severe stress in the hog sector during 1997-1998 period when hog and wholesale
pork prices plummeted due to expanded production and decreased export demand (Southard,
1999). In 1997, the largest farms had debt levels near 40 percent and could be nearing financial
vulnerability; however, these operations had relatively high net income, averaging $157,700 per
operation in 1997 (USDA/ERS, 1999a). Average return on assets at larger-sized hog operations
(more than 300 AU) ranged from 4 percent to 10 percent, varying by facility size and region
(USDA/ERS 1999a). Additional financial information on this sector is provided in Section 7.
In 1997, average net farm income at poultry operations was about $11,000 at broiler and
egg laying operations and $23,500 at turkey operations. Debt-to-asset ratios at poultry
operations ranged from 7 percent to 30 percent, depending on operation size and sector
(USDA/ERS 1999a). Across all sized operations, broiler, layer, and turkey operations had
average debt-to-asset ratios of 19 percent, 11 percent, and 15 percent, respectively (USDA/ERS
1999a), Regionally, there is little difference in the average debt-to-asset ratios operations across
these sectors compared to the national level. Reported average return on assets at larger-sized
2-21
-------
broiler and turkey operations (more than 300 AU) ranged from 2 percent to 6 percent across
sectors (USDA/ERS 1999a). Additional financial information on the broiler, egg, and turkey
sectors is provided in Section 6.
2.2.3 Manure and Manure Nutrients Generated Annually at CAFOs
USDA's Natural Resources Conservation Service (NRCS) estimates that 128.2 billion
pounds of manure are "available for land application from confined AU" from the major livestock
and poultry sectors (Klellogg, et al, 2000). EPA believes these estimates equate to the amount of
manure that is generated at animal feeding operations, since USDA's methodology accounts for
all manure generated at confinement facilities. USDA reports that manure nutrients available for
land application totaled 2.6 billion pounds of nitrogen and 1.4 billion pounds of phosphorus in
1997 (Kellogg, et al., 2000). (See Table 2-7.) USDA's estimates do not include manure
generated from other animal agricultural operations, such as sheep and lamb, goats, horses, and
other farm animal species. ,
TaWe 2-7. Manure and Manure Nutrients "Available for Land Application," 1997
Sector
Cattle w
Dairy
Hogs
All Poultry
Total
USDA Estimates: "Available for
Application" from Confined AU" "
Total
Manure
bill. Ibs
32.9
45.5
16.3
33.5
128.2
Total
Nitrogen
Total
Phosphorus
million pounds
521 '
636
274
1,153
2,583
362
244
211
554
1,437
EPA Estimates:
Percentage Share by Facility Size Group b/
>1,000
AU
>750
AU
>500
AU
>300
AU
percent of total manure nutrients applied
83%
23%
55%
49%
49%
85%
31%
63%
66%
58%
86%
37%
69%
77%
64%
90%
43%
78%
90%
72%
Source: USDA and EPA, as indicated.
"'Nutrients are from USDA/NRCS using 1997 Census of Agriculture and procedures documented developed by
USDA. Numbers are "dry state" and reflect the amount of manure nutrient "available for application from
confined AU" and are assumed by EPA to coincide with manure generated at confined operations).
'"'Percentage shares are by EPA based on the share of animals within each facility size group for each sector (shown
in Table 2-3) across three facility size groups.
''"Cattle" is the sum of USDA's estimate for livestock operations "with fattened cattle" and "with cattle other than
fattened cattle and milk cows." ^Cattle" is the sum of USDA's estimate for livestock operations "with fattened
cattle" and "with cattle other than fattened cattle and milk cows" (Kellogg, et al., 2000).
The contribution of manure and manure nutrients varies by animal type. Table 2-7 shows
that the poultry industry was the largest producer of CAFO manure nutrients hi 1997, accounting
for 45 percent (1.2 billion pounds) of all nitrogen and 39 percent (0.6 billion pounds) of all
2-22
-------
phosphorus available for land application that year (Kellogg, et al., 2000). Among the poultry
sectors, EPA estimates that approximately 55 percent of all poultry manure was generated by
broilers, while layers generated 20 percent and turkeys generated 25 percent. The dairy industry
was the second largest producer of CAFO manure nutrients, generating 25 percent (0.6 billion
pounds) of all nitrogen and 17 percent (0.2 billion pounds) of all phosphorus (Table 2-7).
Together, the hog and beef sectors accounted for about one-fourth of all nitrogen and nearly 40
percent of all phosphorus from CAFO manure.
Table 2-7 shows EPA's estimate of the relative contribution of manure generated by select
major facility size groupings, including coverage for all operations with more than 1,000 AU, all
operations with more than 750 AU or 500 AU (two-tier structure), and all operations with more
than 300 AU (three-tier structure). EPA estimates these shares based on the share of animals
within each facility size group for each sector, as shown in Table 2-3. Given the number of AFOs
that may be defined as CAFOs and subject to the proposed regulations (Table 2-4), EPA
estimates that the proposed effluent guidelines and NPDES regulations will regulate 5 to 7
percent (two-tier structure) to 10 percent (three-tier structure) percent of AFOs nationwide.
Coverage in terms of manure nutrients generated will vary by the proposed regulatory approach.
As shown in Table 2-7, under the 500 AU two-tier structure, EPA estimates that the proposed
requirements will capture 64 percent of all CAFO manure; under the 750 AU two-tier structure,
EPA estimates that the proposed requirements will capture 58 percent of all CAFO manure.
Under the three-tier structure, EPA estimates that the proposed requirements will capture 72
percent of all CAFO manure generated annually (Table 2-7). The majority of this coverage (49
percent) is attributable to regulation of operations with more than 1,000 AU.
Additional information on the constituents found hi livestock and poultry manure and
wastewater is described in Environmental Assessment (USEPA, 2000b). Information on
USDA's estimates of nutrients available for land application and on the relative consistency of
manure (e.g., wetter or drier) for the main animal types is provided in the Development Document
(USEPA, 2000a).
2.3 INDUSTRIAL ORGANIZATION OF LIVESTOCK AND POULTRY
INDUSTRIES
This section presents a discussion of the industrial organization and structure of the
livestock and poultry industries, focusing on the role of vertical integration and coordination
between the animal feeding and the processing sectors. Section 2.3.1 describes the use of
contracts in animal agriculture and Section 2.3.2 discusses the role that contracting plays in
determining the degree of affiliation between a CAFO and the processing sector. These issues
have implications for this rulemaking and may determine whether producers are able to pass on
increases in production costs through higher prices and also whether processors are subject to the
proposed CAFO regulations as co-permittees.
2-23
-------
2.3.1 Contracting in Animal Agriculture
As briefly discussed in Section 2.1.4, over the past few decades, closer ties have been
forged between fanners and their respective manufacturing operations. Increased integration and
coordination are being driven by the competitive nature of agricultural production and the
dynamics of the food marketing system, as well as seasonal fluctuations of production,
perishability of farm products, and limited resources among farmers to handle raw farm output.9
"Contracting" in U.S. agriculture refers to arrangements between farmers and companies
or other farmers that specify conditions of producing and/or marketing an agricultural product.
Contracts can specify price, quantity, and/or quality requirements. There are two basic types of
contracts (USDA/ERS, 1996c):
• Marketing contracts refer to verbal or written agreements between contractors
and growers that set the price (or pricing mechanism) and an outlet for the
commodity before harvest or before the commodity is ready to be marketed. Most
management decisions remain with farmers, since ownership is retained while the
commodity is produced. The contractee (grower) also assumes all the risks .of
production but shares price risk with the contractor (processor). Types of
marketing contracts include forward sales of a growing crop, price setting after ^
delivery, and pre-harvest polling arrangements.
B Production contracts10 specify in detail the production inputs supplied by the
contractor (processor, feed mill, or other farm operations), the quality and quantity
of a particular commodity, and the type of compensation to the grower for services
rendered. The contractor (processor or integrator) may either own the animals
and/or may exercise substantial operational control over the type of production
practices used. The contractor may specify in detail the production inputs supplied
by the contractor, the quality and quantity of a particular commodity, and the type
of compensation to the grower for services rendered. In general, these contracts
do not deal with management of manure and waste disposal, although, the contract
may require the grower to be in compliance with all relevant environmental
requirements.
Martinez (1999) notes that, in the broiler industry, most major processors control the
vertical stages of production, from breeders to market-ready products, through vertical
integration and production contracts. These processor-integrators, such as Tysons, breed the
'Ownership of all aspects of production is the key to vertical integration (Aust, 1997). Vertical
coordination differs from vertical integration in that it includes "all means of vertically harmonizing production,
processing, and distributive activities (Aust, 1997) and covers the use of contracting in livestock and poultry
production, which occurs when an operation contracts with a packer, processor, or feed company to feed animals.
10Also known as "resource-providing contract" (Martinez, 1999).
2-24
-------
parent stock, produce hatchling eggs, and hatch the eggs; they provide chicks, feed, veterinary
services, but contract out the rasing of the chicken to farmers or growers (Martinez, 1999). The
growers provide the chicken houses and labor. The contract specifies a payment per pound of
live broiler produced, depending on the grower's relative performance. In such an integrated
marketing system, price discovery is at the interface between the processor and the retailer
(Martinez, 1999).
Increased use of contracts may be changing the organizational structure of the individual
industries and may raise policy questions regarding ownership responsibility as well as
environmental concerns (USDA/ERS, 1996c). This is especially true given current trends toward
an increasing number of large confined animal feeding operations and the resultant need for
increased animal waste management. Farmers raising animals under production contract with
processors face different risks and decisions than farmers raising animals under their own
ownership or under marketing contracts because farm management decisions are often tightly
controlled by the processor (Bastian et al., 1994). See Section 2.1.4. As farms become larger
and more specialized, they may contract out some phases of the production process. The contract
farm might be less able to make decisions about environmental concerns and to cover the costs of
addressing these concerns (USDA/ERS, 1996c). Since environmental controls may raise the cost
of production to the farmer, the farmer may wait to implement best management practices until
the contract specifies and/or compensates the farmers for the additional costs (USDA/ERS,
1996c). .
Continued growth in contract farming is anticipated because of its benefits to both the
processor and the contract farmer. Contracts reduce farmer exposure to price risk by combining
market functions and allowing them to secure a constant price and buyer (Kohls and Uhl, 1998).
Factors other than income stability include improved efficiency, market security, and access to
capital and new technologies (USDA/ERS, 1996c). The farmer also frequently receives
production inputs from the processor, including animals (chicks, breeder pigs, replacement
heifers), feed, technical support, and transportation of animals. Receipt of feed supplies from
integrators reduces uncertainty and risk in feed costs, often the single largest cost component of
livestock and poultry production (USDA/ERS, 1996c).
Processing firms benefit from contracting since they can realize rapid growth without the
need to own and maintain land and facilities while exerting control over the production process,
input supply, and genetics (Hayenga et al., 1996). Among hog processors, a recent survey
indicated that the top three advantages to processors of contracting is increased financial leverage,
reduced environmental/regulatory problems, and access to motivated labor (i.e., farmer growers)
(Hennessey and Lawrence, 2000).
Figure 2-2 depicts the flow of activities and typical sharing of responsibilities in a
contractual agricultural relationship, as summarized by Ogishi and Zilberman (1999).
2-25
-------
Integrator
Producer
Consumer
Figure 2-2 Flow of Activities and Sharing of Responsibilities in a Contractual System
Source: Ogishi and Zilfaerman, 1999
2-26
-------
2.3.2 Degree of Affiliation between CAFOs and Processors
EPA uses information on the degree of affiliation between CAFOs and processing firms
within a sector, based on the types of contractual arrangements used within a sector, to support
its assumptions on whether processors within an industry may be subject to the proposed CAFO
regulations as co-permittees (Section 2.4). These data are from USDA that report the use of
production contracts and the degree of animal ownership by the farm operator in these sectors.
Use of production contracts in the livestock and poultry industries varies by commodity
group. Information from USDA indicate that production contracts are widely used in the poultry
industry and dominate broiler production (Aust, 1997; USDA/ERS, 1999a and 1996c; Heffernan,
et al., 1999; Perry, et al., 1999; Martinez, 1999). Production contracts are also common in the
hog sector, and use in Hie finishing stage of production is rising rapidly in some regions (Pritchett
and Lui, 1998). By comparison, production contracts are.not widely used in the beef and dairy
sectors (USDA/ERS, 1996c and 1999a; Heffernan, et al., 1999). Additional details on vertical
coordination and contracting for each sector are presented in Sections 6, 7, and 8 of this report.
/
Production contracting in the poultry sectors differs from that in the other livestock -
sectors, since it is dominated by near vertical integration between a farmer ("grower") and a
processing firm ("integrator"). In the poultry sectors, vertical integration has progressed to the
point where large, multifunction producer-packer-processor-distributor firms are the dominant
force in poultry and egg production and marketing (Kohls and Uhl, 1998). Production
contracting has also played an important role hi the growth of integrators in the poultry sectors
(Aust, 1997).
In a 1993 study, USDA showed that almost 90 percent of all poultry farms use contracts,
most of which are production contracts (Table 2-8). Across all poultry sectors, production
contracting accounted for 86 percent of the value of all production in 1993 (USDA/ERS, 1996c).
Information from USDA on animal ownership at U.S. farms further indicates the potential degree
of processor control in these sectors (USD A/ERA, 1999a). These data indicate that production
contracting accounts for virtually all (98 percent) of U.S. broiler production in 1997 (USDA/ERS,
1999a). (See Table 2-9.) This indicates that nearly all broiler production may be under the
ownership or control of processing firms that are affiliated with broiler operations. Production
contracting accounts for a smaller share of turkey and egg production, accounting for 70 percent
and 37 percent, respectively (USDA/ERS, 1999a). See Table 2-9.
Production contracts are also common in the hog sector, but still account for a smaller
share of total production. In 1993, 6 percent of all hog farms grew animals under a production
contract, accounting for 11 percent of all hog marketings during that year (USDA/ERS, 1996c).
(See Table 2-8.) However, the hog sector is rapidly evolving into an industry of larger firms that
are specialized and vertically coordinated through production contracting, particularly in
production regions outside the Corn Belt (Pritchett and Lui, 1998). More recent data indicate
2-27
-------
that contracting use is rising rapidly. In 1997, an estimated 40 percent of the hogs farrowed and
44 percent of the finished hogs were produced by farms with production contracts (Lawrence, et
al., 1998).11 Most contract hogs are produced by large farms in emerging regions, such as North
Carolina (Hayenga et al.51996).
2-8 Contracting Use in the Livestock and Poultry Sectors, 1993
Commodity Group
All"
Cattle
Dairy b/
Hogs
Number of Farms
Farms (number)
Farms with Contracts (number)
Percentage under Contract
Farm with Production Contracts (number)
Farm with Marketing Contracts (number)
2,063,300
225,308
11%
43,609
185,736
740,138
13,278
2%
2,827
10,625
125,408
34,903
28%
2,661
32,441
82,132
9,232
11%
4,701
4,749
Value of Production
Total Value of Production (Smillion)
Production Value under Contract (Smillion)
Percentage under Contract
Value under Production Contract
(Smillion)
Value under Marketing Contract
(Smillion)
$150,493
$47,454
32%
$17,706
$27,748
$33,870
. $7,787
23%
$6,038
$1,749
$23,833
$10,185
43%
$178
$10,007
$8,436
$1,155
14%
$958
$197
Poultry
27,589
24,500
89%
23,379
2,081
$11,237
$9,642
86%
$8,845
$796
Source: USDA7ERS, 1996c.
^Includes all crops and livestock farms.
wMost milk is produced under marketing orders. Milk producers usually have a verbal agreement with their buyer
or cooperative. Because a quantity and final price are not specified before the sale, producers do not consider this
"contract." Farmers may have production contracts with other operators to produce a stream of replacement
heifers.
uUse of marketing contracts has also risen sharply, accounting for 57 percent of 1997 hog marketings
(Lawrence, et al., 1998).
2-28
-------
Table 2-9. Percent of Animals Owned and Not Owned by Farmers by Sales-Based Size Categories, 1997
Number/Percent
All
Farms
Small Family Farms
Low Sales
High
Sales
Large
. Family
Farms
Very
Large
Family
Farms
Non-
Family
Farms "
All Cattle and Calves b/
No. of Animals (millions)
Percent not owned (%)
99.3
7.4%
21.0
4.1%
17.9
6.3%
9.3
5.6%
15.0
7.0%
8.9
26.5%
Beef Cows
No. of Animals (millions)
Percent not owned (%)
39.6
4.7%
10.7
3.9%
6.8
7.4% d
2.5
7.0%
2.5
3.5% d
2.2
3.6% *
Heifers and Heifer Calves "
No. of Animals (millions)
Percent not owned (%)
25.7
8.8%
4.7
3.7%
4.9
5.1%
2.9
. 4.9% **
4.6
10.2%
2.8
33.9%
Milking Cows
No. of Animals (millions)
Percent not owned (%)
10.5
1.4%
1.2
0.3% "
2.8
1.7%*
1.4
1.6%*
3.7
0.4% "
0.8
3.4% .
Hogs and Pigs
No. of Animals (millions)
Percent not owned (%)
40.6
27.6%
3.4
31. 1%^
6.5
13.3% ^
9.8 .
29.8% d
13.4
40.2%
5.1
7.4% *
Broilers, Fryers and other Meat-type Chickens
No. of Animals (millions)
Percent not owned (%)
763.8
97.6%
50.6
93.7%
62.2*
78.9% d
233.9 "
100.0%
385.5
99.6%
ND
ND
Turkeys
No. of Animals (millions)
Percent not owned
154.0
70.0%
ND
94.5% ^
1.5*
90.2%
7.8*
93.3%
99.9 *
78.3% "
37.2*
36.2% ^
Layers
No. of Animals (millions)
Percent not owned (%)
163.6
37.3%
10.7
94.0%
4.5
89.9%
18.1
85.2% •
89.4
34.3% d
38.6*
0.6%
Source: USDA/ERS, 1999a. See Table 2-2 for USDA farm typology group definitions.
ND= not disclosed; data insufficient for disclosure (fewer than 30 observations).
"'Nonfamily farms include nonfamily corporations or cooperatives, as well as farms operated by hired
^Includes all feedlots as well as cow-calf operations and stacker or backgrounding operations.
^Relative standard estimate (RSE) of estimate >25% but <50%.
*RSE of estimate >50% but <75%.
•"RSE of estimate >75%.
managers.
2-29
-------
Information from USDA on animal ownership at U.S. hog farms in 1997 provides an
indication of the potential degree of processor control in this sector. These data indicate that
nationally production contracts accounted for about 30 percent of total hog production in 1997
(Table 2-9). Production contract use in the hog sector varies by production region. Production
contracts accounted for 66 percent of hog production in the Southern and Mid-Atlantic states in
1997, with use concentrated among the larger producers (USDA/ERS, 1999a). In comparison,
production contracting accounted for 18 percent of hog production in the Midwest (USDA/ERS,
1999a). (See Table 2-5.) This indicates that a large share of hog production may be under the
ownership or control of processing firms in the Southern and Mid-Atlantic states. Production
contracting is less common in the Midwest because coordination efforts are more diversified and
because independent producers enter into networks that are similar in many ways to highly
integrated systems (Kliebenstein and Lawrence, 1995). Production contracting in the hog sector
is becoming more like that in the poultry sector in that it is increasingly focused on the finishing
stage of production, with the farmer ("grower") entering into an agreement with a meat packing
or processing firm ("integrator"). Production contracts are also used in the hog sector to raise
immature animals between two independent animal feeding operations (USDA/ERS, 1996c).
Because of the frequency of production contract use and the increased probability of
animal ownership by processors in the poultry and hog sectors, EPA concludes that processors
within these industries may be subject to the proposed CAFO regulations as co-permittees (as
described in Section 2.4), and therefore assumes this for the purpose of this analysis. This is
because, under a production contract, many farm management decisions may remain with the
processor, who controls the level of production and the types of production practices used by the
contract grower. In many cases, the processor also retains ownership of the animals. The nature
of this relationship is believed to trigger the co-permittee requirements due to "substantial
operational control." (Additional information on EPA's proposed co-permitting requirements are
provided in Section YE of the preamble to the proposed CAFO regulations.).
By contrast, production contracts account for only a small share of beef and milk
production (USDA/ERS, 1999a and 1996c; Heffernan, et al., 1999). Contractual agreements in
these sectors are typified by marketing contracts (Table 2-8). Animal ownership on beef and dairy
farms is mostly by the farm operator (Table 2-9). Accordingly, EPA concludes that processors
within the beef and dairy industries would not be subject to the proposed CAFO regulations as
co-permittees. .
The majority of cattle and calves are sold through private arrangements and spot market
agreements (USDA/GIPSA, 1997). Production contracting is not common in the beef sector. As
shown in Table 2-8, less than two percent of all beef operations produced under contract in 1993,
accounting for 23 percent of the value of all beef produced (USDA/ERS, 1996c). Most beef
sector contracts are marketing-based, in which operations agree to sell packers a certain amount
of cattle on a predetermined schedule. As shown by 1997 USDA data of animal ownership,
production contracting accounts for a relatively small share (5 percent) of beef production
(USDA/ERS, 1999a). See Table 2-9. In recent years, however, the relationship between cattle
2-30
-------
producers and processors has become more interconnected, either through contractual
arrangements or through actual ownership (integration) by processors, particularly for some
phases of production (Bastian, et al., 1994).
Milk and dairy food production has become increasingly specialized, but has not
experienced vertical integration in the same way as other livestock industries. Most farm milk is
produced by independent, privately owned facilities (Manchester and Blaney, 1997). Production
contracting is uncommon in milk production (USDA/ERS, 1996c). In part, this is attributable to
the large role of farmer-owned, farmer-controlled dairy cooperatives, which handle about 80
percent of the milk delivered to plants and dealers. Milk is generally produced under marketing-
type contracts through verbal agreement with their buyer or cooperative. About a quarter of all
dairy farms were involved in contracting arrangements in 1993, accounting for 43 percent of all
milk produced (USDA/ERS, 1996c). Data from USDA further indicate that little more than one
percent of milk was produced under a production contract in 1997 (USDA/ERS, 1999a). See
Table 2-9.
Production contracts are used in the beef and dairy industries to establish a contractual
relationship to specialize in one stage of livestock production (USDA/ERS, 1996c). However,
this constitutes a small share of overall production in these sectors, as discussed above (also see
Table 2-9). Production contracts are used in these industries for custom feeding, cattle
backgrounding, and heifer replacement. For example, a beef feedlot operation may agree to fatten
or "finish" cattle not owned by the operation for a fee, based on weight gain prior to slaughter;
these custom feeding operations provide finish feeding under contract. Backgrounding or stocker
operations raise cattle under contract from the time the calves are weaned until they are on a
finishing ration in a feedlot This arrangement allows operators to increase business volume with
limited facilities through specialization. In the dairy sector, farms may use production contracts
with other operators to produce a stream of replacement heifers (USDA/ERS, 1996c). Additional
information is provided in Section 8 of this report
Among non-family farms, production contracts account for a relatively larger proportion
of total production in the beef and dairy sectors. See Table 2-9. USDA data on animal
ownership for 1997 indicate that 26 percent of total cattle arid calf production and 34 percent of
heifer and heifer calf production by non-family farms is not owed by the farmer (USDA/ERS,
1999a). Despite the limited use of contracts for the finishing and raising phase of production,
EPA expects that no businesses, other than the CAFO where the animals are raised, will be -
subject to the proposed co-permitting requirements in the beef and dairy sectors.
2-31
-------
2.4 CHARACTERISTICS OF PROCESSING FIRMS THAT MAY BE AFFECTED
BY THE PROPOSED CAFO REGULATIONS
EPA expects that the proposed CAFO regulations will mainly affect livestock and poultry
operations that confine animals. In addition to CAFOs, however, the proposed regulation may
also affect businesses that contract out the raising or finishing production phase to a CAFO under
the proposed co-permitting requirements. These proposed co-permitting requirements are
described in Section 3; more detailed information is provided in Section VII of the preamble. The
proposed co-permitting requirements could affect meat packing plants and poultry slaughtering
facilities that exercise "substantial operational control" over a CAFO. This section provides an
overview of the livestock and poultry processing sectors, focusing on processing firms that enter
into contract with CAFOs and are likely to be affected by the proposed rule.
Section 2.4.1 builds on the farmer-processor discussion in Section 2.3 by identifying the
types and number of livestock and poultry processing firms that may be affected by the proposed
CAFO regulations. This section first identifies all livestock and poultry processors, then estimates
which ones are likely to contract out to CAFOs and thus potentially be affected operations. A
summary of available financial data and information on the livestock and poultry processing
sectors is provided in Section 2.4.2. (An overview of the supply and demand conditions for
livestock and poultry products is provided in Section 2.5, expressed in terms of farm level
production.)
2.4.1 Identification and Number of Potential Co-Permittees
There are three steps involved in determining the number of processing firms that may be
affected by the proposed co-permittee requirements. First, EPA identified all livestock and
poultry processors, using data from the Department of Commerce. Second, EPA examined the
potential for "substantial operational control" between CAFOs and processing firms within
different livestock and poultry sectors. Third, once a sector was identified as having the potential
for "substantial operational control," EPA identified the number of facilities that are likely to be
affected by the proposed co-permitting requirements, using available industry information. These
steps are described in the following sections.
2.4.1.1 All Livestock and Poultry Processors
The food processing sectors supporting the animal production sectors that are the focus of
the proposed CAFO regulations encompass a broad group of businesses that process red meat,
milk and dairy products, and poultry products. These processors can be further subdivided into
meat packers and slaughterers, processors, further processors, and Tenderers. By NAICS code,
these operations include:
2-32
-------
• Animal (except poultry) slaughtering, NAICS 311611 (SIC 2011, meat packing
plants).12
« Meat processed from carcases, NAICS 311612 (SIC 2013, sausages and other
prepared meats),
• Poultry processing, NAICS 311615 (SIC 2015, poultry slaughtering and
processing).
• Liquid, dried, and frozen eggs, NAICS 311999(G) (SIC 2015, poultry slaughtering
and processing).
• Fluid milk manufacturing, NAICS 311511, (SIC 2026, fluid milk manufacturing).
• Creamery butter, NAICS 311512 (SIC 2021, creamery butter).
• Cheese manufacturing, NAICS 311513 (SIC 2022, cheese manufacturing).
• Dry, condensed and evaporated dairy manufacturing, NAICS 311514 (SIC 2023,
processed milk products).
• Ice cream and frozen dessert manufacturing, NAICS 31152 (SIC 2024, ice cream
and frozen desserts).
• Rendering and meat by-product processing, NAICS 311613 (SIC 2077, animal
and marine fats and oils).
The U.S. Department of Commerce's 1997 Census of Manufactures reports that there
were a total of 1,393 meat packing establishments (NAICS 311611) and 1,297 red meat
processing facilities (NAICS 311612) in the United States (USDC, 1999a). Another 474
establishments were involved in poultry slaughter and processing (NAICS 311615), as well as 84
establishments engaged in poultry and egg processing (NAICS 311999(G)) (USDC, 1999a).
There were also 240 rendering plants (NAICS 311613) in 1997 (USDC, 1999a).
Table 2-10 presents the key NAICS product class codes for the broader group of
establishments initially considered likely to be affected by the proposed CAFO regulations.
Detailed data are not disclosed for pork processors. More information on the market and
financial characteristics of these industry sectors is described in more detail in Section 2.4.2.
12Covers the slaughter of cattle, calves, steer, heifers, pork, sheep, and lamb.
2-33
-------
Table 2-10. Processing Industry Statistics by Primary Product Class and Sector, 1997
NAICS Industry or Product
Class Code and Primary Class
All
Est.
All
Employ.
Value
Added
($MM)
Cost of
Materials
($MM)
Value of
Shipments
(SMM)
Total
Capital
Expend.
(SMM)
Beef
311611(1),
Fresh and frozen beef, not
canned or made into sausage a
' 228
71,070
$4,174.3
$29,220.9
$33,347.7
$0.3
Dairy
31 1511, Fluid milk
311512, Creamery butter
31 1513, Cheese
311514, Dry, condensed, & evap.
dairy product
311520,
Ice cream & frozen dessert
612
34
524
213
451
58,217
1,827
36,918
15,325
19,818
$6,311.3
$241.9
$4,411.7
$4,015.9
$2,533.4
$15,887.8
$1,133.9
$15,918.2
$4,991.4
$3,312.8
$22,212.1
$1,367.5
$20,326.3
$9,021.6
$5,863.5
$428.3
$8.5
$485.9
$261.7
$159.4
Hog
311611(A),
Fresh and frozen pork
311611(0),
Pork, processed or cured ^
311612(1),
Pork, processed or cured b/
83
11
143
d
d
21,501
d
d
$1,838.6
d
d
$3,778.1
d
d
$5,635.5
d
d
$177.3
Poultry
31 1615(1), Young chickens
311615(4), Hens and-or fowl .
31 1615(7), Turkeys
311999(0),
Liquid, dried, and frozen eggs b/
212
15
39
44
137,674
3,527
27,339
4,651
$6,228.8
$130.4
$1,756.0
$433.3
$11,150.0
$179.7
$2,501.1
$991.5
$17,347.8
$309.2
$4,252.2
$1,426.2
$310.8
$11.2
$102.9
$30.6
Source: USDC, 1999a. d = data not disclosed.
''Made from animals slaughtered at this plant.
'"'Identified as further processors and excluded from processor impact analysis.
2-34
-------
2.4.1.2 Sectors with Potential for "Substantial Operational Control"
Not all of the livestock and poultry processors identified in Section 2.4.1.1 are expected to
be subject to the proposed CAFO regulations as co-permittees. However, there are no publicly
available data on the number of farms or processing firms that enter into a production contract
arrangement. Market information is not available on the number and location of firms that
contract out the raising of animals to CAFOs and the number and location of contract growers,
and the share of production, that raise animals under a production contract. EPA also does not
have data on the exact terms of the contractual agreements between processors and CAFOs to
assess when a processor would be subject to the proposed co-permitting requirements, nor does
EPA have financial data for processing firms or contract growers that utilize production contracts.
This is proprietary business information.
To estimate the number of potentially affected operations, EPA uses readily available
published data from USD A and the Department of Commerce. To distinguish potentially affected
operations, EPA examined all types of commodity processors and facility types in order to
methodologically eliminate those groups of processors that are unlikely to be affected by the
proposed regulations. To do this, EPA examined the potential for "substantial operational
control" between CAFOs and processing firms at the sector level based on the type of contractual
agreements used in that industry.
EPA expects that the proposed co-permitting requirements under the proposed CAFO
regulations will affect processing firms that have substantial operational control of a CAFO.
Generally, substantial operational control is found under a situation of vertical integration (where
the packer owns the CAFO) or vertical coordination (where production is controlled under a
production contract).. Vertically integrated companies are not expected to be required to obtain a
co-permit, since the firm would be required to obtain only a single permit as owner of the CAFO.
Companies that contract out to CAFOs may be required to obtain a co-permit under some types
"of contracting agreements.
As described in Section 2.3, under a production contract, most farm management
decisions are dictated by the contractor (processing firm) who controls the level of production
and the types of production practices used by the contract grower (fanner). In many cases, the
processor also retains ownership of the animals. In circumstances where production contracts are
widely used between CAFO and processor, EPA expects that the processor may be subject to co-
permitting requirements because of operational control and/or ownership by the processor. Thus,
a permit would be issued at a CAFO and another would be issued at the affiliated processing firm
that controls production at the CAFO. As noted previously, marketing contracts are unlikely to
allow the processor to exercise operational control over a CAFO and are thus not considered
likely to trigger the co-permit requirements under the proposed CAFO regulations.
Only the pork and poultry sectors are identified as sectors where use of production
contracts is common. Production contracting is uncommon in the beef and dairy sectors; also,
2-35
-------
most beef and dairy operations own their own animals (USDA/NASS, 1999a; USDA/ERS,
1996c), as discussed in Section 2.3. EPA believes that this supports its conclusions that
processors in the beef and dairy sectors are unlikely to be affected by the proposed co-permitting
requirements. These same data show that the vast majority of poultry operations and a sizeable
portion of pork production operations are under production contracts. Based on this information,
EPA concludes that only pork and poultry processors would be potentially affected by the
proposed CAFO regulations as possible co-permittees. This narrows the scope of potentially
affected facilities identified by EPA to these two sectors.
To further narrow the scope, EPA eliminates broad groups of processors as potential co-
permittees, including most food processing firms, such as further processors, food manufacturers,
and Tenderers. These operations are not expected to be affected by the co-permitting
requirements because these operations are further up the marketing chain and do not likely
contract with CAFOs to raise animals. In general, these types of establishments do not directly
purchase product from or have direct relations with CAFOs but rather obtain products from other
processors. Since EPA believes mat these establishments do not generally establish contractual
relationships with CAFOs, these facilities are excluded from the count of potentially affected
processors. For similar reasons EPA also excludes in its count of potential co-permittees any
establishments whose primary classification indicates wholesale or retail trade (egg processors,
e.g., egg graders and packers, are classified under wholesale trade). Given the types of contract
arrangements that are common in the hog and poultry industries, EPA expects that only packers
and slaughterers in these industries will be subject to the proposed co-permitting requirements.
2.4.1.3 Identification of Potential Co-Permittees based on Facility Type and Size
As discussed in the preceding section, EPA expects that the proposed co-permitting
requirements will primarily affect meat packers and poultry slaughters/processors in the pork and
poultry sectors only. This narrows the number of potential co-permittees to two NAICS industry
classifications for "animal (except poultry) slaughtering" (NAICS 311611) and "poultry
processing" (NAICS 311615).
To further eliminate broad categories of processors, EPA identifies the number of
potential co-permittees among pork and poultry processors based on available information about
facility type and operation size.
The meat packing industry is composed of slaughterhouses (where livestock are
slaughtered and further processed) and specialized meat processors (who do not slaughter but
instead manufacture sausage, luncheon meats, and other prepared products) (Kohls and Uhl,
1998). In 1997, there were 1,308 meat packing companies (NAICS 311611) in the United States
(USDC, 1999a). These companies cover all meat product types, including cattle, calves, steer
2-36
-------
and heifers, pork, and sheep and lambs. These 1,308 meat packing companies comprise 1,393
establishments.13
Although 1,393 meatpacking establishments are identified, far fewer of these are
considered likely to be affected by the proposed regulations. First, a majority of meat processors
were excluded because of facility size considerations. Hayenga et al. (1996) indicates that
production contracts in the hog sector are generally associated with the largest processors and
producers. Hayenga's study focused on the largest packers, feed companies, and hog
producers/contractors, "which both embody and transmit the driving forces toward change in
pork sector coordination and organization linkages" (Hayenga et al. (1996)). Accordingly, EPA
believes that only the largest firms are likely to have the capacity to process larger volumes of raw
farm product that are generated at CAFOs with more than 300 or 500 AU; also, only the largest
firms are likely to have the necessary resources and organizational structure that would allow
them to fully realize the benefits of production contracting. By contrast, smaller-sized processors
are unlikely to contract out to larger CAFOs or be able to handle the production volumes
generated at a regulated CAFO.
The 1997 Economic Census (USDC, 1999a) provides a simple way of identifying the
larger processors through their product class breakdowns, which identify numbers of
establishments by product class specialization (e.g., NAICS 311611A, Fresh and frozen pork, not
canned or made into sausage, made from animals slaughtered at this plant). Product class
specialization is determined on the basis of a detailed survey. USDC provides detailed surveys
only to a select group of mostly larger firms in the manufacturing industry. Included in this group
are large- and medium-sized establishments that are surveyed in the Annual Survey of
Manufactures (ASM). Historical information is used for product classifications for about 30
percent of additional large and medium non-ASM establishments, and about 15 percent of smaller
establishments with more than 20 employees are provided a short survey form, which includes
product information. Those not receiving surveys are not broken out by product class but are
included in the overall count of 1,393 establishments, with some additional information on
employment and value of shipments obtained by USDC through administrative records. The
establishments not sent surveys are those with fewer than 20 employees. EPA believes, as
discussed above, that these small-sized processors are unlikely to enter into production contracts
with CAFOs and thus assumes they are not directly affected by the proposed CAFO regulations.
Among meat packing firms (NAICS 311611), 437 establishments provided information on
product class (USDC, 1999a). Among these 437 establishments, however, only 94 were
13For the purpose of this analysis, EPA assumes that one establishment identified as a possible co-
permittee equals one affected entity, since data to determine which processors are most likely to have production
contracts are only available at the establishment level. Although firms may own multiple establishments, these
establishments generally are sited in different geographic locations and would be associated with different CAFOs.
Thus each establishment with production contracts could be required to obtain a co-permit. The number of
establishments, therefore, is a reasonable estimate of the number of co-permits.
2-37
-------
identified as pork processors (USDC, 1999a).14 Of these, Department of Commerce's 1997
product class specialization identifies 83 establishments that process fresh and frozen pork and 11
establishments that process or cure pork. These data generally account for larger processing
facilities that have more than 20 employees. EPA believes that processing firms that may be
affected by the proposed co-permitting requirements will mostly be larger facilities that have the
administrative and production capacity to take advantage of various contract mechanisms. This
estimate does not include other processors under NAICS 311611, including sausage makers and
facilities that "further process" hog hides and other by-products because these operations are
considered to be further up the marketing chain and likely do not contract out to CAFOs.
Thus, of the 437 meat processors, EPA considers that 94 pork processors are potentially
subject to the proposed co-permitting requirements. Not all of these firms are expected to engage
in production contracting. Survey data from Hayenga et al. (1996) indicate that only 5 of 19
(about a quarter) of the largest pork packers used production contracts or produced hogs
themselves in 1995. Assuming only a quarter of the 94 pork processors have contracts, only.
about 25 pork processors might be subject to co-permitting requirements. However, the use of
production contracts is increasing rapidly in the hog sector; at the same time, there is ongoing
consolidation and plant closures are occurring. As a conservative measure, EPA assumes that all
the 94 identified pork processors may be potential co-permittees. This estimate is likely an
overestimate of the number of potential co-permittees since the top six processing firms account
for 75 percent of all pork production (as cited in Heffeman, et al., 1999).
The Department of Commerce reports that there were a total of 558 poultry and egg
slaughtering and processing facilities in 1997. Among poultry processors (NAICS 311615), EPA
identified 259 firms with 474 establishments in the poultry processing industry using the 1997
Economic Census, consisting of 212 establishments that process young chickens, 15 that process
hens or fowl, and 39 that process turkeys, for a total of 266 chicken and turkey establishments
(USDC, 1999a). EPA's estimate excludes 44 reported egg processing facilities (NAICS 311999)
since these are considered to be further up the marketing chain and likely do not contract out to
CAFOs (see Table 2-10).15 Because production contracts are so widespread in the poultry
industry, EPA assumes that about 270 poultry processors identified in this manner might be
subject to co-permitting requirements. This estimate is likely an overestimate of the number of
potential co-permittees: Heffeman et al. (1999) note that 95 percent of all broilers are produced
under production contracts with fewer than 40 firms. As discussed previously, these data
generally account for larger-sized processing facilities with more than 20 employees to account
for the expectation that processing firms that may be affected by the proposed co-permitting
requirements have adequate administrative and production capacity to contract out to CAFOs.
14EPA did not include beef and veal-processors, since these processors are considered unlikely to have
production contracts, as discussed in Section 2.3. Other meat packers that slaughter other animals (lamb, sheep,
etc.) under this NAICS code were also not included.
15EPA did not include establishments that process other poultry (such as ducks and small game).
2-38
-------
2.4.2 Financial Characteristics of the Livestock and Poultry Processing Sector
The meat and poultry processing industries are characterized by the dominance of a few,
very large firms, many with worldwide influence. For example, Smithfield Foods, Inc., is the
world's largest pork processor, with record sales of $3.9 billion in 1997 (Hoovers, 1998).
Farmland Foods, Inc., also one of the top four pork processors, is expanding its global market in
South Korea, Japan, and Mexico. Tyson Foods, Inc., is the world's largest poultry producer,
with revenues of $4.7 billion in 1997 (Hoovers, 1998).
In 1997, the top ten meat and poultry processors had combined sales of $59.0 billion
compared to a total value of shipments of $86.2 billion for the meat processing and poultry
slaughtering sectors (NAICS 311611 and 311615) (IBP, Inc., 1997; USDC, 1999a), or nearly 70
percent of the market. Employment by these top 10 processors totals approximately 220,000
workers compared to total industry employment of 366,885 (TOP, Inc., 1997;~ USDC, 1999a).
The top 10 poultry processors controlled over 60 percent of the broiler market in 1997
(Thornton, 1999); an estimated 95 percent of all contract broiler production was controlled by
fewer than 40 firms (Heffeman, et al., 1999). In 1998, the top four broiler processors (Tyson,
Gold Kist, Perdue, and Pilgrim's Pride) comprise almost 50 percent of total production16
(Heffeman, et al., 1999; IBP, Inc., 1999). The four largest turkey processors (Jennie-O,
Butterball, Wampler, and Cargill) controlled more than 40 percent of the market17 (Heffeman, et
al., 1999; Heffeman, 1999). In 1998, the top four pork meat packers (Smithfield, IBP, Inc.,
ConAgra, Cargill) controlled nearly 60 percent of all hog production18 (Heffeman, et al., 1999;
USDA/GIPSA, 1998). Daily hog slaughter capacity of the top 10 pork packers is 83 percent of
the industry (NPPC, 1998).
Geographical distribution of processing plants mirrors distribution of farms. The top 10
pork packing companies in 1997, ranked by estimated daily slaughter capacity, operated 30
plants. Almost 40 percent of these plants were located in Iowa, with another 10 percent in North
Carolina, 10 percent in Nebraska, and 10 percent in Illinois (NPPC, 1998). The country's largest
sow farms are clustered in the Com Belt region and in North Carolina (Freese, 1997). Broiler
processing plant distribution among the top 10 broiler processors is most dense in the
southeastern states of Georgia and Arkansas (Thornton, 1999). As discussed later in Section 6,
Georgia and Arkansas were the largest broiler producing states in 1997. Plant distribution among
the top 10 turkey processors is greatest in Minnesota (24 percent), North Carolina (18 percent),
16Commonly measured by the "four-firm concentration ratio" or CR4. For broilers, the CR4 was 49
percent in 1998; the CR6 was 58 percent (Feedstuffs, as referenced in Heffeman, et al., 1999).
17For turkeys, the CR4 was 42 percent in 1998 (Turkey World, as referenced in Heffeman, et al., 1999). ,
18For hog meat packer, the CR4 was 57 percent in 1998 (National Hog Farmer, as referenced in
Heffeman, et al., 1999). The CR6 was 75 percent (NYTimes, as referenced in Heffeman, et al., 1999).
2-39
-------
and Virginia (12 percent) which corresponds to higher turkey farm distribution in these states (see
Section 6).
The financial health of red meat processors relies to a great extent on foreign and domestic
demand, as well as various production factors (USDC, 1999b). Generally, product shipments of
red meat and poultry products have risen in recent years. Product shipments of beef and pork
processors increased 3.4 percent between 1997 and 1998, despite a decrease in beef production
between 1997 and 1998 (USDC, 1999b). Product shipments of poultry products were expected
to increase by 3 percent in 1998; net returns were also expected to increase among poultry
producers as lower feed costs offset low prices caused by increased broiler supply in 1998
(USDC, 1999b). However, the costs of assembling large quantities of raw farm products and
transporting final products long distances to consumers tend to be high and may not be offset by
operating efficiency gains of large, central processing facilities (Kohls and Uhl, 1998).
EPA uses available company level data for certain publicly owned slaughterers and
processors to characterize the financial conditions in these industries. The results of some of this
research, Preliminary Profile of Poultry, Hog, Beef and Dairy Processors and Integrators, is
contained in the record (ERG, 1999c—see, DCN 70232). The majority of these companies are
privately owned and financial information is not readily available. To the extent that data are
available, EPA evaluates the general financial health of the processing sectors using data from the
Security and Exchange Commission (SEC) (ERG, 2000e) and information on publicly held firms
reported by Robert Morris Associates (RMA, 1996,1997, and 1998) and Dun & Bradstreet (Dun
& Bradstreet, 1996,1997,-and 1998). .
For this brief summary, EPA examines three financial variables, where available. These
include pre-tax net return on assets (ROA), interest coverage ratios (ICR), and debt-to-asset
ratios. The pre-tax net return on assets among these firms is a measure of financial health that
indicates profitability and whether the "investment" in terms of the firms' assets is providing an
adequate return on that investment, or whether investment elsewhere would be more profitable.
EPA has utilized ROA to assess the financial health of manufacturing firms in the Pulp, Paper and
Paperboard (USEPA, 1993), Pharmaceutical Manufacturing (USEPA, 1995a), Metal Products .
and Machinery (USEPA, 1995d), and the Pesticide Formulating, Packaging and Repackaging
(USEPA, 1996) industries.
The interest coverage ratio, which measure earnings before interest and taxes divided by
interest (also referred to as EBIT/Interest), indicates the ability of a firm to take on additional
debt. EPA has previously used ICRs, in conjunction with ROAs, to assess the financial health of
firms in the Pharmaceutical Manufacturing (USEPA, 1995a) and Metal Products and Machinery
(USEPA, 1995d) industries.19 Generally, ICRs over 3 indicate the ability of a firm to take on
I9In addition, EPA used the closely-related times-interest-eamed ratio (TIE) for the Pulp and Paper
(USEPA, 1993) and Transportation Equipment Cleaning (USEPA 1998) industries. TIE is defined as the sum of
EBIT and depreciation divided by interest payments. The terms ICR and TIE are sometimes used interchangeably.
2-40
-------
additional debt (Van Home, 1986). ICR is also sometimes compared to a lowest quartile ICR
based on public data such as that published by RMA to determine vulnerability (USEPA, 1995a
and!995d).
Debt-to-asset ratios also measure a firm's ability to take on additional debt. EPA has used
the debt-to-asset ratio to assess the financial health of firms in the Pulp, Paper and Paperboard
(USEPA, 1993) industry, although no benchmark was developed for that analysis.
Well-defined benchmarks for financial ratios do not, in general, exist for use in analyzing
the financial health of a firm. Instead analysts compare ratios over time and between firms to
assess financial health (Brigham and Gapenski, 1997). For this analysis-—as in analyses conducted
for previous effluent guidelines—EPA uses the value of ROA for the lowest quartile of firms (as
published by RMA) as a benchmark for competitive financial performance (USEPA, 1995a,
1995d, 1996). For debt-to-asset ratios, EPA uses a benchmark of 0.40 to indicate vulnerability
among U.S. farms, as developed by USDA (USDA/ERS, 1997e; Sommer et al., 1998). This
benchmark is similar to the average debt-to-asset ratio for all manufacturing firms, reported by
Brigham and Gapenski (1997).
To evaluate the financial health of processing firms, EPA uses SEC 10K filings for 10
firms that operate in the pork and/or poultry processing industries (ERG, 2000e). Table 2-11
presents summary data derived from balance sheets and income statements for these 10 firms,
averaging each firm's data over the years 1996,1997, and 1998. These average data are then
used to develop ranges and medians across all the firms in the data set.20 Of the 10 pork and
poultry processing firms EPA investigated, 9 firms show an average positive net income before
taxes during this period. The median net income for this group was $30.1 million. Assets ranged
from $182.3 million to $11.3 billion with a median of $831.4 million. Liabilities ranged from
$178.5 million to $11.4 billion with a median of $775.6 million. Estimated ROA among these
firms, shown in Table 2-11, indicates that 60 percent of the firms listed generated returns that
exceeded 4 percent; the median was 5.7 percent. The debt-to-asset ratios among these firms
ranged from 0.52 to 0.78, with a median of 0.64, which is greater than the average debt-to-asset
ratio for manufacturing firms as a whole (ERG, 2000e). Of the 10 firms investigated, seven had
ICRs that could be used informatively, and of these, all but two had an ICR greater than 3. These
latter two firms, however, had ICRs near 3 (2.37 and 2.58).
Thus, despite the fairly high debt-to-asset ratio, the ICR indicator suggests that debt levels
among these firms might not be so high as to prohibit additional debt acquisition. The sheer size
of these firms, however, gives them the ability to absorb fairly large costs before any real effect on
their financial condition would be felt. For example, an additional debt of $5 million would be
only 0.7 percent of the $755 million debt at the median firm.
20Generally, 1997 and 1998 were better years for these firms than 1996, in which 4 firms reported •
negative pre-tax income. _ ' ~
2-41
-------
Table 2-11. Key Financial Characteristics of Selected Publicly Held Processing Firms (1996-1998)
Firm
Net Income
(Post-tax)
Total Assets
Total
Liabilities
$1,000, average of 1996, 1997, and 1998
% Pre-tax
Profit to
Total Assets
ICR
Debt-to-
Asset Ratio
average of 1996, 1997, and 1998
Pork Only
One firm
$38,074
$978,839
$675,169
5.9
6.6
0.69
Pork and Poultry
Low
High
Median
$109,397
$472,367
$290,882
$1,506,855
$11,392,167
$6,449,511
$706,499
$8,890,133
$4,798,316
7.1
11.3
' 9.2
3.81
— *>
3.81 *
Poultry Only
Low
High
Median
($20,741)
$99,267
$23,208
$178,491
$1,732,533
$473,277
$134,429
$3,021,333
$286,711
-8.8
9.7
5.8
2.37
8.14
3.19
All Firms
Low
High
Median (all
firms)
($20,741)
$472,367
$30,086
$178,491
$11,392,167
$775,634
$111,128
$8,890,133
$531,045
-8.8
11.3
5.7
2:37
8.14 ^
4.24
0.47
0.78
0.63
0.52
0.68
0.64
0.52
0.78
0.64
Source: 1996,1997, and 1998 SEC 10-K filings for the following firms: Cal Maine Foods, Inc., ConAgra, Inc.,
Hormel Foods Corp., Michael Foods, Inc., Pilgrims Pride Corp., Sanderson Farms, Inc., Seaboard Corp.,
Smithfield Foods, Inc., Tyson Foods, Inc., WLR Foods, Inc. See ERG, 2000e for excerpted filings.
"^Three firms reported negative interest payments averaged over the 3 year time frame, resulting in negative ICRs,
were not used in the ranges nor are they used to calculate medians, since these firms may still be able to take on
additional debt.
2-42
-------
Table 2-12. Published Industry Key Financial Characteristics
SIC Code
Number of
Observations ^
Quartile
Lower
Median
ICR (EBIT/Interest)
2011 -Meatpacking
2015 - Poultry slaughtering and processing
26
' 13
1.1
1.2
4.2
3.6
Upper
12.6
10.0
ROA (% Profit Before Taxes/Total Assets)
201 1 - Meat packing
2015 - Poultry slaughtering and processing
28
14
0.8
-1.1
7.8
10.5
17.3
12.0
Source: RMA, 1997.
"'Within the S10-50MM asset group.
Table 2-12 presents information for 1997 on 28 firms in the red meat packing (hogs and
beef) industry and 14 firms in the poultry industry. These published financial data are from Robert
Morris Associates (RMA) on pork and poultry processing firms (RMA, 1996,1997,1998). The
largest group of firms reported in the RMA data are those in the $10 million to $50 million asset
class, which represents firms that are smaller in size than those represented by the data in
Table 2-11. RMA reports pre-tax return on assets and ICR, among other financial ratios, but
does not report the debt-to-asset ratio. As Table 2-12 shows, the median firm had an ICR above
3 and a ROA of over 7 percent in 1997. This indicates substantially better financial performance
than the lowest quartile firms, which may be vulnerable financially given considerably lower
estimates for these indicators compared to the median value.21
EPA also reviewed available Dun & Bradstreet data for all red meat packers and poultry
slaughterers (Dun & Bradstreet, 1996,1997, and 1998). This source does not provide ICR or
debt-to-asset ratios, but does provide ROA in the form of post-tax returns. Among meat packing
firms, the median ROA was 6.0 percent for 142 meat packing establishments in 1996, 5.3 percent
for 83 meat packing establishments in 1997, and 5.1 percent for 59 establishments in 1998.
Among poultry processors, the median ROA was 2.2 percent for 55 poultry processing
establishments in 1996,2.0 percent for 33 establishments in 1997, and 4.2 percent for 27
establishments in 1998 (Dun & Bradstreet, 1996,1997, and 1998). Average asset levels across
1996-1998 for this group were $2.0 million for meat packers and $13.5 million for poultry
processors, which is generally smaller in size than the firms represented by both the SEC and the
RMA data.
21EPA also reviewed RMA data for 1996 and 1998 to see if there were any major differences by year and
concluded that the 1997 data for meat processors was not substantially different from data in the other years. For
poultry processors, however, 1997 median ROA was somewhat higher than the medians for 1996 and 1998,
reported at 4.2 percent for 1996 and 7.3 percent for 1998, compared to 10.5 for 1997 (RMA, 1996, 1997,1998).
2-43
-------
Based on these readily available data, EPA concludes that livestock and poultry processing
firms that could be affected by the proposed co-permitting requirements, particularly the larger
firms, are generally not financially vulnerable, showing some ability to take on additional debt and
showing reasonable returns on investment.
2.5 OTHER MARKET CHARACTERISTICS OF THE LIVESTOCK AND
POULTRY INDUSTRIES
The following sections discuss annual farm receipts and manufacturing value of shipments
(Section 2.5.1), supply and demand conditions (Section 2.5.2), and employment (Section 2.5.3) in
the livestock and poultry industries. Other market characteristics are discussed in more detail in
the individual sector sections of this report (Sections 6,7, and 8).
2.5.1 Annual Marketing Receipts
USDA estimates that total economic output from the farm and processing sectors was
$231 billion in 1997 and accounted for nearly 3 percent of gross domestic product (estimated at
$8.7 trillion during that year) (Council of Economic Advisors, 2000). Table 2-13 summarizes
these data.
2.5.1.1 Total Farm Receipts from Marketings
The 1997 Census of Agriculture reports that farm revenue across each of the regulated
sectors totaled $95.6 billion in 1997 (Table 2-13). This represents nearly one-half of total farm
marketing across all the livestock and crop sectors, estimated at $196.9 billion in 1997
(USDA/NASS, 1999a). Farm revenues from cattle farming make up the largest portion of total
U.S. farm receipts. In 1997, revenues from farm cattle sales totaled $40.5 billion and accounted
for almost one-half of the total value of U.S. livestock and poultry farming. Of this, fed cattle
operations accounted for $20.4 billion (USDA/NASS, 1999a). Farm milk sales totaled $19.0
billion in 1997 (USDA/NASS, 1999a). The poultry industry had combined revenues of $22.3
billion in 1997, up nearly 50 percent compared to 1992 (USDA/NASS, 1999a). The bulk of
poultry revenues is attributable to broiler sales (estimated at $14.2 billion in 1997); eggs and
turkeys accounted for $8.1 billion. EPA estimates these shares based on reported state level
revenues for the three main poultry sectors (USDA/ERS, 1996b). Hog farm revenues also rose
from 1992 levels, reaching $13.8 billion in 1997 (USDA/NASS, 1999a).
2-44
-------
2.5.1.2 Total Manufacturing Value of Shipments
The 1997 Census of Manufactures reports that the value of shipments across each of the
manufacturing businesses in.the livestock and poultry industries totaled $159.5 billion in 1997
(Table 2-13). Total receipts across each of the livestock and poultry industries examined totaled
$147.9 billion in 1992. Table 2-13 also shows estimated total sales of livestock and poultry
products in the U.S. economy, estimated at more than $230 billion in 1997. EPA calculates this
total using farm-to-retail price spreads reported by USDA in its monthly Agricultural Outlook
reports (USDA/ERS, 2000f), adjusted by farm level revenues reported by USDA (USDA/NASS,
1999a).
Table 2-13. Farm Receipts and Manufacturing Value of Shipments (1992 and 1997)
Sector
Beef
Fed Cattle
Dairy
Hogs
Poultry
Layers
Broilers
Turkeys
Total
Farm
1992
(Smillion)
$41,700
$21,100
$17,800
$10,000
$15,400
$3,300
$9,800
$2,300
$84,900
1997
($million)
$40,500
$20,400
$19,000
$13,800
$22,300
$4,800
$14,200
$3,300
$95,600
Processing
1992
($ million)
$44,500
$53,800
$25,900
$23,800
$147,900
1997
($million)
$50,300
$48,300
$29,200
$31,700
$159,500
Total
1997
($billion)
$82.7
$58.8
$35.4
$54.0
$230.9
Source: Farm revenues for beef, fed cattle, dairy, hogs, and all poultry are from the 1997 Census (USDA/NASS,
1999a). Shares for each of the three poultry sectors are estimated from shares reported for states (USDA/ERS,
1996c). Manufacturing shipment values are from USDC, 1999a. Total is estimated to avoid .possible double
counting from vertically integrated facilities. "Total" is calculated using farm-to-retail price spreads reported by
USDA (USDA/ERS, 2000f), adjusted by farm level revenues reported by USDA (USDA/NASS, 1999a). :
2.5.2 Supply and Demand Conditions for Livestock and Poultry Products
Livestock and poultry operations are part of a production chain that includes suppliers,
meat packing plants, food processors, cooperatives, and retailers. These operations provide the
raw materials to slaughterers, packers, and processors in the form of live animals, raw milk, and
eggs. These raw materials are converted into cuts of meat and various processed foods, milk, and
dairy products, which are eventually sold to consumers at retail establishments.
2-45
-------
This section includes a summary of recent trends in domestic supply and demand and net
trade in the U.S. livestock and poultry sectors. Selected years that approximate Census of
Agriculture years are shown. For this discussion, production and utilization (demand and traded
volumes) are expressed in terms of farm level production equivalents (i.e., animal carcass weight,
milk equivalents, number of eggs). These data are summarized in Tables 2-14,2-15, and 2-16.
2.5.2.1 Farm Production
Production of red meat, milk, and poultry products increased over the period from 1970
to 1997. Table 2-14 shows these trends for selected years. The largest gains were in poultry
meat production: both turkey and broiler production more than tripled from 1970 to 1997
(Putnam and Allshouse, 1999). Egg production rose by more than 10 percent over this period.
Pork and beef production has increased nearly 20 percent since 1970. Milk production rose more
than 30 percent from 1970 to 1997.
Table 2-14. Total Livestock and Poultry Production, Selected Years (1970-1997)
Year
1970
1974
1978
1982
1987
1992
1997
%70-97
Beef
Pork
Chicken
Turkey
(billion pounds carcass weight)
21.7
23.1
24.2
22.5
23.6
23.1
25.5
18%
14.7
14.3
13.4
14.2
14.4
L7.2
17.3
18%
8.5
8.7
10.4
12.6
16.0
21.4
27.6
225%
1.7
1.9
2.0
2.5
3.7
4.8
~ 5.4
218%
Eggs
(billion dozen)
5.7
5.5
5.6
5.8
5.9
5.9
6.4
12%
Milk
(billion Ibs)
117.0
115.6
121.5
135.5
142.7
150.9
156.1
33%
Source: Putnam and Allshouse (1999).
2.5.2.2 Domestic Demand
Americans are among the highest per capita consumers of poultry, meat, and dairy
products in the world. Since the 1980s, per capita U.S. demand for poultry meat, in particular,
increased dramatically, outpacing the rate of population growth. As shown in Table 2-15,
domestic demand for poultry meat has more man doubled: demand for chicken meat rose from 40
pounds per person in 1970 to more than 80 pounds per person in 1997 and turkey demand rose
from 8 pounds per person in 1970 to nearly 18 pounds per person in 1997 (Putnam and
Allshouse, 1999). Per capita egg demand dropped over the period from more than 300 eggs to
about 240 eggs per person per year. Per capita red meat demand also dropped from 1970 to
2-46
-------
1997: beef demand declined from 114 pounds per person to 96 pounds per person and pork
demand dropped from 72 pounds to 63 pounds per person (Putnam and Allshouse., 1999). Per
capita demand for fluid milk and dairy products, however, rose from an aggregate of 531 pounds
per person in 1970 to 569 pounds per person in 1997 (NMPF, 1999).
Table 2-15. Per Capita Demand for Livestock and Poultry Products, Selected Years (1970-1997)
Year
1970
1974
1978
1982
1987
1992
1997
%70-97
Beef
114.1
115.5
117.7
103.9
103.8
94.7
95.2
-17%
Veal
3.0
2.3
2.9
2.0
1.8
1.2
1.2
-60%
Pork
Chicken
Turkey
(Ibs./person)
72.1
68.2
60.2
62.6
62.7
67.8
62.5
-13%
,40.1
39.6
• 44.8
51.5
62.0
76.9
83.8
109%
8.1
8.7
8.7
10.6
14.7
17.9
17.6
117%
' Eggs
(doz./person)
308.1
283.0
271.5
264.1
253.8
235.0
242.4
-21%
Milk
(Ibs.,
ME/person)
• 530.9
520.5
531.1
519.4
550.2
569.6
568.6
7%
Source: Putnam and Allshouse (1999), except data on dairy utilization (demand and traded volumes), which are
expressed in terms of milk equivalent (ME), total solids basis (NMPF, 1999). .
Table 2-16. Livestock and Poultry Product Trade, Selected Years (1970-1997)
Year
1970
1974
1978
1982
1987
1992
1997
%70-97
Beef
Imports
Exports
Pork
Imports
Exports
Chicken
Turkey
Eggs
Exports
(million Ibs.)
1,792
1,615
2,297
1,939
2,269
2*440
2,343
31%
101
115
214
305
656
1,400
2,136
2015%
491
488
495
612
1,195
645
633
29%
194
204
421
365
236
552
1,044
438%
183
235
505
674
920
1,732
5,048
2658%
43
43
57
' 56
37
186
598
1291%
(mill, doz.)
45
56
121
185
136
175
220
389%
Milk
Imports
Exports
(mill. Ibs., ME)
3,165
4,937
3,902
4,184
4,206
4,245
4,383
38%
884
1,158
760
10,640
5,597
7,032
5,244
493%
Source: Putnam and Allshouse (1999), except data on dairy utilization (demand and traded volumes), which are
expressed in terms of milk equivalent (ME), total solids basis (NMPF, 1999). Annual poultry (chicken, turkey,
egg) imports are low and are not shown. - • •
2-47
-------
2.5.2.3 Imports and Exports
Despite its position as one of the largest agricultural producers in the world, historically
the U.S. has not been a major player in world markets for red meat (beef and pork) or dairy
products. In fact, until recently, the U.S. was a net importer of these products (Putnam and
Allshouse, 1997 and 1999; USDA/WAOB, 1999 and 2000; NMPF, 1999). The presence of a
large domestic market for value-added meat and dairy products has limited U.S. reliance on
developing export markets for its products (NDB, 1995; USDA/ERS, various). In recent years,
however, slowing growth and/or saturation in domestic demand has forced U.S. industries to step
up efforts to export products abroad. As the U.S. has taken steps to expand export markets for
red meat and dairy products, one major obstacle has been that it remains a relatively high cost
producer of these products, particularly of milk and dairy products, as compared to other
established net exporters, such as New Zealand, Australia, and Latin America, as well as other
more established and government-subsidized exporting countries, including the European Union
and Canada (NDB, 1995; USITC, 1998a). .
Increasingly, however, continued efficiency gains and low-cost feed is making the U.S.
more competitive in world markets for these products, particularly for red meat (Iowa State
University, 1998; USDA/WAOB, 1999 and 2000). The U.S. is among the world's lowest cost
producer of pork, but still ranks close to competitors Australia and Argentina in terms of cost of
production (Iowa State University, 1998). The U.S. is currently the world's second largest beef
exporter, after Australia, and is among Hie world's top producer of high grade beef
(USDA/WAOB, 1999 and 2000). While the proposed CAFO regulations may raise production
costs and potentially reduce production quantities that would otherwise be available for export,
EPA believes that any quantity and price changes resulting from the proposed requirements will
not significantly alter the competitiveness of U.S. export markets for red meat or dairy foods.
U.S. poultry products account for a controlling share of world trade and exports account
for a sizable and growing share of annual U.S. production (Putnam and Allshouse, 1997 and
1999; USDA/WAOB, 1999 and 2000; USITC; 1998b). The U.S. is among the world's lowest
cost producer of poultry products due to higher feed efficiency and lower overall feed costs
(USITC, 1998b). Given the established presence of the U.S. in world poultry markets and the
relative strength in export demand for these products, EPA does not expect that the predicted
quantity and price changes resulting from today's proposed regulations will have a significant
impact on the competitiveness of U.S. poultry exports.
Table 2-16 shows recent trends in U.S. livestock and poultry trade. As shown, U.S.
exports of meat and dairy products have grown dramatically. U.S. beef exports more than
doubled and pork exports have nearly doubled since the early 1990s (Putnam and Allshouse, 1997
and 999). Chicken and turkey exports also rose sharply and nearly tripled over the period
(Putnam and Allshouse, 1997 and 1999). Historically, dairy product exports have been more
variable year-to-year; however, in recent years commercial exports have been steadily increasing,
while subsidized exports and foreign aid have dropped (NMPF, 1999).
2-48
-------
2.5.3 Industry Employment
Employment figures presented in this report include total farm labor (hired labor only and
total, which also includes self-employed and family) and employment in the processing sector.
Where published data are not available, EPA estimates employment in terms of full-time
equivalents (FTEs) using available information from USD A survey data.
Combined, total employment .within the farm and processing sectors is estimated at 1.7
million in 1997 and accounted for more than one percent of national civilian employment
(estimated at 129.6 million employed during that year) (Council of Economic Advisors, 2000).
This estimate reflects total direct employment only, representing the number of jobs related to the
production and processing of these products. This estimate does not include indirect or induced
employment, or workers throughout the economy that provide additional support to the industry.
These data are summarized in Table 2-17. ,
Table 2-17. Livestock and Poultry Industry Employment by Industry Segment (1997)
Animal
Commodity
Group
Fed Cattle "
Pork
Dairy .
Poultry
Total
Total Farm
Labor *
(FTEs)
336,700
195,900
483,800
71,800
1,088,200
Hired
Labor as
% Total "'
(percent)
NA
8%
17%
39%
NA
Total
Processing
Labor6'
Total Farm
and
Processing
(FTEs)
145,617
84,723
141,400
204,200
575,940
482,317
280,623
625,200
276,000
1,664,140
Farm %
Agric.
Labor*
Farm&Mfg
% Total
Labor *
(percent)
10%
6%
14%
2%
32%
0.4%
0.2%
0.5%
0.2%
1.3%
"• Total farm employment is updated by EPA from 1990 estimates by Abel, Daft & Barley (1993) to account for
changes between 1990 and 1997 (Council of Economic Advisors, 2000). Estimates are allocated by sector based on
its share of annual farm revenue (USDA/NASS, 1999a) and exclude employment at cattle grazing operations.
b/ Hired labor is expressed as a percent of total estimated farm employment from labor estimates provided by
USDA from its Farm Labor Survey database (Milton, 1999 and 1998).
d Processing sector employment is from the 1997 Census of Manufactures (USDC, 1999a).
* Shows total farm employment as a share of civilian (agricultural) employment and total farm and processing
level employment as a share of civilian (total) employment, seasonally adjusted annual averages (Council of
Economic Advisors, 2000).
2.5.3.1 Total Farm Employment
EPA estimates total farm sector employment by updating 1990 estimates developed by an
agricultural consultancy group, Abel, Daft & Barley (1993). The Abel, Daft & Barley study
estimated total farm sector employment by calculating FTEs based on labor hours worked ,
2-49
-------
reported by USDA converted to person-year equivalents.22 EPA is unable to duplicate this
approach using more recent data because similar farm labor information is not regularly updated
by USDA. Instead, EPA adjusts these 1990 estimates to account for changes in agricultural
employment between 1990 and 1997. This is done by projecting USDA's 1990 farm labor hours
based on changes in civilian employment reported by the Department of Labor (Council of
Economic Advisors, 2000).
EPA's calculated changes in agriculture sector FTEs from 1990 to 1997 are allocated to
the livestock and poultry sectors using Able, Daft and Barley's assumption that one-half of total
farm level employment is in these sectors (with the remaining employed by the crop production
sectors). The additional FTEs are allocated across each of the livestock and poultry sectors based
on each sector's share of annual livestock and poultry receipts (USDA/NASS, 1999a). In
addition, EPA adjusts the 1990 labor estimates for the beef sector to exclude farm employment by
rangeland operations, also based on the share of farm level sales hi mat subsector (USDA/NASS,
1999a).
The resulting farm labor estimates for 1997 are shown in Table 2-17. This table shows
that total farm level FTEs in the livestock and poultry sectors listed are estimated at about 1.1
million in 1997. As' a share of total agricultural employment, total farm employment by the
regulated sectors accounts for roughly 30 percent, excluding employment at cattle grazing
facilities. Across EPA's estimate of 376,000 AFOs (see Table 2-3), this translates to an average
of nearly 3 FTEs per operation. Because farm labor is seasonal and because independent owners
might work more than 40 hours per week, these figures may not represent the actual number of
individuals who work on farms.
This estimate of total farm employment includes operator labor, unpaid family labor, and
hired labor. The bulk of farm employment consists of owner-operators and other family members.
USDA defines these as self-employed -workers, including operators or partners who complete
unpaid agricultural work (including the .contractee); and unpaid workers, including workers, other
than self-employed workers, who did at least 15 hours of unpaid agricultural work (e.g., family
members) (USDA/NASS, 1998c). Limited available information indicate that hired farm labor
accounts for roughly 40 percent of total farm employment in the poultry sectors but a small share
in the hog and dairy sectors. Hired labor numbers are not available for confinement beef
operations only. Hired workers may be full-time or part-time, and seasonal or year-round
employees.
•^USDA-reported labor hours were converted to person-year equivalents by dividing by 2,080 (i.e., 52
forty-hour weeks). Data used for these estimates are from various ERS publications, including Production and
Efficiency Statistics, 1990 and State Financial Summary, 1990.
2-50
-------
2.5.3.2 Total Manufacturing Employment
Employment in the processing sector is available from U.S. Census of Manufactures for
1997 (USDC, 1999a). As shown in Table 2-17, processing level employment totaled 0.6 million
in 1997. For the red meat sectors, the Census of Manufactures data are expressed across all
plants engaged in NAICS 311611, Animal (except poultry) slaughtering, andNAICS 311612,
Meat processed from carcasses, which encompass both the beef and pork meat industries as well
as other miscellaneous sectors, including lamb and sheep.23 As reported by the Department of
Commerce, the pork processing sector employed 84,700 persons and the beef processing sector
employed 145,600 persons in 1997 (Table 2-17).
Dairy manufacturing employment is an aggregate of reported employment across the dairy
product codes (NAICS 311511, Fluid milk manufacturing; NAICS 311512, Creamery butter
manufacturing; NAICS 311513, Cheese manufacturing; NAICS 311514, Dry, condensed and
evaporated dairy product manufacturing; and NAICS 311520, Ice cream and frozen dessert
manufacturing). In 1997, there were 141,400 people employed in dairy processing (corrected by
the Department of Commerce to avoid double counting).
Poultry employment is the aggregate of reported employment for NAICS 311615, Poultry
processing, andNAICS 311999G, Liquid, dried, and frozen eggs. In 1997, employment in
poultry processing totaled at 204,200 jobs (Table 2-17).
23 Employment by firms engaged in the manufacture of "Prepared Feeds" (SIC 2048) were not included
for this analysis since these data cannot be proportioned out across each of the beef, dairy, pork, and poultry
sectors. Typically, however, it is recognized that employment in the feed grains sector constitutes part of direct
employment in the livestock and poultry sectors, since it includes workers engaged in the manufacture of
agricultural inputs and their supplies and employment in supplier industries (Abel, Daft & Barley, 1993).
2-51
-------
-------
SECTION THREE
THE PROPOSED CAFO REGULATIONS
Section 1 provides a summary of the existing NPDES permit regulations and technology-
based pollutant limits affecting CAFOs that have been in place since the 1970s. This section
summarizes the proposed revisions to the CAFO regulations (Section 3.1) and describes the ELG
Options and NPDES Scenarios that EPA is proposing, as well as the regulatory alternatives that
were considered by EPA during the development of this rulemaking (Section 3.2).
3.1 SUMMARY OF THE PROPOSED REVISIONS
The major regulatory revisions being proposed by EPA include changes to the scope of
the regulations (i.e., which operations are subject to the proposed regulations) and other changes
to the regulatory requirements for CAFOs. This section summarizes these changes briefly. More
detailed information on the proposed regulatory changes is available in the preamble (see, Section
VII, "What Changes to the NPDES CAFO Regulations Are Being Proposed?" and Section VEI,
"What Changes to the Feedlot Effluent Limitations Guidelines Are Being Proposed?").
EPA is proposing to make the provisions of this rule effective three years from
promulgation for facilities newly defined as CAFOs (approximately December, 2005). For
facilities that are currently defined as CAFOs, the effluent guideline requirements are immediately
applicable upon permit renewal or permit application.
3.1.1. Revised Scope Requirements under the Proposed Regulations
The existing NPDES regulation uses the term "animal unit" (AU) to describe sizes of
facilities that are CAFOs and therefore subject to the regulations. The metric AU was established
in the 1970 regulations to equate the wastes produced by different animal types based on waste
characteristics. The existing regulation defines facilities with 1,000 AU or more as CAFOs.
Facilities with 300 AU to 1,000 AU are CAFOs if they meet certain conditions, or if they are
designated as a CAFO by the permit authority. Those facilities with fewer than 300 AU are
CAFOs only if designated by the permit authority.
EPA's proposal clarifies the definition of a CAFO to include both the production areas
(animal confinement areas, manure storage areas, raw materials storage areas and waste
containment areas) and the land application areas that are under the control of the CAFO owner
or operator.
3-1
-------
EPA's proposal also broadens the applicability of the existing permit regulation with
regard to the size of a facility that is defined as a CAFO. EPA is co-proposing two alternatives
for determining who is affected by the NPDES program for CAFOs, including the two-tier
structure and the three-tier structure. The alternatives offer comparable environmental benefits
but differ in administrative approach.
The proposed two-tier structure simplifies the definition of which facilities are CAFOs by
establishing a single threshold for each animal sector at the equivalent of 500 AU. Facilities with
more than 500 AU would be defined as CAFOs; facilities with fewer than 500 AU are CAFOs
only if designated by the permit authority. EPA believes this two-tier approach will simplify the
regulation by making it clear which CAFOs are subject to the proposed requirements and will also
enhance compliance and facilitate enforcement. EPA is also soliciting comment on whether to
propose an alternative AU threshold under the two-tier structure, that would establish a single
threshold for each animal sector at the equivalent of 750 AU. Facilities with more than 750 AU
would be defined as CAFOs; facilities with less than 750 AU are CAFOs only if designated by the
permit authority.
The proposed three-tier structure retains the framework of the existing regulation: all
confinement operations with more than 1,000 AU are defined as CAFOs; operations with between
300 AU and 1,000 AU are CAFOs if they meet certain conditions or if designated by the permit
authority; and operations with fewer than 300 AU are CAFOs only if they are designated by the
permit authority. However, EPA's proposal would alter the conditions for defining which
operations in the middle tier are CAFOs and would require all facilities with 300 AU to 1,000 AU
either to apply for a permit or to certify that they do not meet the conditions for being defined as a
CAFO.
Under the three-tier structure, EPA is proposing certain "risk-based" conditions, among
which are included: mere is direct contact of animals with waters of the U.S. at the operation;
there is insufficient storage and containment at the production area to prevent discharge from
reaching waters of the U.S.; there is evidence of discharge in the last five years; the operation's
production area is located within 100 feet of waters of the U.S.; the operator does not have, or is
not implementing, a Permit Nutrient Plan; and manure of more than twelve tons annually is
transported to off-site recipients without following proper manure management. Additional
information on the set of conditions for defining which operations in the middle tier are CAFOs is
provided in Section VII of the preamble. The proposed "risk-based" conditions are described in
Part 122.23(a)(3)(ii)(B) Option 2 of the proposed regulation.
Also under the three-tier structure, EPA is soliciting comment on whether to propose an
alternative AU threshold that would alter the conditions for defining which operations in the
middle tier are CAFOs. This scenario would use a similar but less inclusive set of conditions that
, would result in fewer AFOs meeting the definition of a CAFO. Information on the set of
conditions for defining which operations in the middle tier are CAFOs under this alternative is also
provided in Section VII of the preamble.
3-2
-------
EPA's proposal also expands the regulatory definition of CAFOs to include all types of
poultry operations regardless of the type of manure handling system or watering system they use.
EPA's proposal would also include stand-alone immature swine and heifer operations. The
existing regulation only applies to chicken operations which use either a specific type of drinking
water delivery system or liquid manure handling systems. Most chicken operations use neither of
these technologies, therefore, EPA is proposing to redefine CAFOs with respect to chickens
eliminating any distinctions about how the birds are raised or how the manure is handled. The
existing regulations also applies to swine weighing more than 55 pound or mature dairy cows
only. For the purpose of identifying CAFOs, EPA is proposing to redefine CAFOs with respect
to immature pigs and heifers. These proposed changes account for industry trends over the past
25 years toward specialization, including the practice of confining immature swine that weigh less
than 55 pounds in separate nursery facilities
The 300 AU, 500 AU, 750 AU and 1,000 AU equivalent number of animals for each
sector would be as follows:
Animal Type
Cattle37
Veal
Mature Dairy Cattle
Mature Swine
Immature Swine
Chickens
Turkeys
Ducks
Horses
Sheep or Lambs
1,000 AU
1,000
1,000
700
2,500
10,000
100,000
55,000
5,000
500
10,000
750 AU 500 AU
(number of animals)
750
750
525
1,875
7,500
75,000
41,250
3,750
375
7,500
500
500
350
1,250
5,000
50,000
27,500
2,500
250
5,000
300 AU
300
300
200
750
3,000
30,000
16,500
1,500
150
3,000
"'Other than mature dairy cattle or veal.
EPA's estimate of the number of animal confinement operations that would be defined or
designated as CAFOs is presented in Section 2 of this report.
EPA's proposal retains the permit writer's discretion to "designate" a confinement facility
as a CAFO, even if below the AU threshold that defines CAFOs. EPA is proposing to simplify
the criteria, however, by eliminating two criteria that have proven difficult to implement—the
"direct contact" criterion and the "man made device" criterion. EPA is proposing to eliminate
these criteria for the 300 to 1,000 AU tier in the proposed three-tier alternative and, for
simplicity's sake, for the less than 500 AU tier in the proposed two-tier alternative. The proposal
retains the existing requirement for the permit authority to consider a number of factors when
determining whether the facility is a significant contributor of pollution to waters of the U.S., as
well as the requirement for an on-site inspection in order to make that determination. EPA is also
3-3
-------
proposing to clarify its authority to designate facilities in States with NPDES authorized
programs.
EPA's proposal also broadens the applicability of the existing effluent guidelines affecting
feedlot operations. The existing effluent guidelines regulations apply to only those CAFOs with
more than 1,000 AU. EPA is proposing to broaden the applicability of the effluent guidelines to
be consistent with the definition of a CAFO for key sectors. The proposed effluent guidelines
revisions would apply to beef, dairy, swine, poultry and veal operations that are defined or
designated as CAFOs under either of the co-proposed structures and that are above the threshold
for the effluent guideline. For those operations below the threshold defining them as CAFOs
(designated CAFOs), the permit writer would use best professional judgment (BPJ) to develop
the site-specific permit conditions. .
EPA's proposed effluent guidelines revisions do not alter the existing effluent guideline
regulations for horses, ducks, sheep or lambs. In these sectors, only facilities with more than
1,000 AU are subject to the effluent guidelines. Permits for operations in these subcategories
with fewer than 1,000 AU would continue to be developed based on the best professional
judgement of the permit writer. •
EPA is further proposing to clarify that entities that exercise "substantial operational
control" over the CAFO are "operators" of the CAFO and thus would need to obtain a permit
along with the CAFO owner or operator. The trend toward specialized animal production under
contract with processors, packers, and other integrators has increasingly resulted in
concentrations of excess manure beyond agricultural needs in certain geographic areas. Especially
in the poultry and swine sectors, the processor provides the animals, feed, medication and/or
specifies growing practices. EPA believes that clarifying that both parties are liable for
compliance with the terms of the permit as well as responsible for the excess manure generated by
CAFOs will lead to better management of manure. EPA's estimate of the number of processing
firms that may be subject to the proposed regulations as co-permittees is presented in Section 2 of
this report.
3.1.2 Other Revised Requirements under the Proposed Regulations
Additional regulatory requirements that are being proposed for CAFOs are summarized as
follows. Additional information on these proposed changes, along with EPA's justification for
proposing these requirements is presented in Sections VTI and VIH of the preamble.
3-4
-------
Eliminate the 25-year, 24-hour Storm Event Permit Exemption
Under the current rule, an operator, whose facility is designed, constructed and operated
to contain a 25-year, 24-hour storm .event is not required to apply for a permit if discharges occur
only as a result of such an event. EPA is proposing to eliminate the 25-year, 24-hour storm event
permit exemption and to impose a broader, more explicit duty for all CAFOs to apply for a
permit. However, EPA is proposing to retain the 25-year, 24-hour storm standard as a design
standard in the effluent guidelines for certain sectors (specifically, the beef and dairy sectors).
CAFOs in those sectors would need to obtain permits, but the permits would allow certain
discharges as long as the facility meets the 25-year, 24-hour storm design standard.
Include Land Application Area as Part of the CAFO
The proposal clarifies the-definition of a CAFO as including the production area (animal
confinement area, manure storage area, waste containment area) as well as the land application
area that is under the control of the CAFO owner or operator. EPA is also co-proposing options
for off-site handling of manures. Under the first co-proposal, EPA would require that the CAFO
operator obtain a certification from off-site recipients of CAFO manure that the manure is being
land applied according to proper agricultural practices. Under the second co-proposal, EPA
would require that the CAFO operator provide off-site recipients of CAFO manure with
information about the nutrient content of the manure and proper agronomic use of the manure,
and that the CAFO operator maintain records on the identity of manure recipients, the volumes
received, and the dates the manure was received.
Increase the Stringency of the Effluent Guidelines.
EPA is proposing to establish Best Practicable Control Technology Currently Available
(BPT), Best Available Technology Economically Achievable (BAT) and Best Control Technology
for Conventional Pollutants (BCT), and New Source Performance Standards (NSPS) for CAFOs.
EPA is proposing to establish requirements for discharges from the production area and land
application of manure, including the requirement that all CAFOs develop a Permit Nutrient Plan
(PNP). More detailed information on the proposed revisions to the effluent guidelines and
standards is provided in Section VIE of the preamble. .
The BPT requirements in the existing regulations apply to beef, dairy, swine, veal, and
poultry operations with more than 1,000 AU. These requirements establish a zero discharge
requirement from the production area with a design standard or the 10-year, 24 hour storm event.
The proposed revisions to BPT limitations for these subcategories requires zero discharge from
the CAFO production area with a design criteria of the 25-year, 24-hour storm event. BPT
requires that all CAFOs obtain a PNP to control the discharge of pollutants from the land
3-5
-------
application fields. The PNP defines the rate at which manure can be applied to land owned or •
controlled by the CAFO.
EPA is proposing to restrict manure application to a phosphorus basis where necessary
due to soil conditions. In situations where the build up of phosphorus in the soil is excessive,
manure application would be prohibited. In all other areas, application of manure to meet the
nitrogen requirements of the crop will be allowed. In addition, manure and wastewater
application is prohibited within 100 feet of surface water, tile drain inlets, agricultural drainage
wells, and sinkholes. EPA is proposing that manure must be applied to cropland at rates not to
exceed the crop requirements for nutrients and the ability of the soil to absorb any excess
phosphorus. BPT establishes specific recordkeeping requirements associated with ensuring the
achievement of the zero discharge limitation for the production area and that the application of
manure and wastewater is done in accordance with land application requirements. EPA is also
proposing to require the CAFO operator to maintain records of any excess manure that is
transported off-site.
EPA is also proposing to establish BCT limitations that are equivalent to the BPT
limitations for the control of conventional pollutant discharges from CAFOs.1
EPA is proposing to revise BAT requirements for the beef and dairy subcategories based
on the BPT requirements with the additional requirement that the CAFO achieve zero discharge
to ground water beneath the production area in locations where the ground water has a direct
hydrologjcal connection to surface water. The land application requirements of the proposed
BAT requirements are the same as BPT. The existing regulation requires zero discharge from the
production area with the 25-year, 24-hour storm design standard, which allows for discharges
when catastrophic or chronic storms exceed this design criteria. The preamble also solicits
comment on basing BAT on BPT without the additional ground water controls. For the hog, veal
and poultry subcategories, the proposed BAT requirements include elimination of the current
exemption for overflows in Ihe event of a chronic or catastrophic storm at the production area.
The preamble also solicits comment on basing BAT on BPT without the additional controls
precluding overflow due to rain events.
EPA's proposal also revises standards for new operations (New Source Performance
Standards, or NSPS). For the beef and dairy subcategories, EPA is proposing new source
standards based on the same technology requirements as BAT. NSPS for the hog, veal and
poultry subcategories is based on the proposed BAT requirement with the additional requirement
that there be no discharge from the production area of pollutants through ground water that has a
direct hydrological connection to surface water. Both the BAT and NSPS requirements have the
same land application and record keeping requirements as proposed for BPT.
'Conventional pollutants are defined as Biochemical Oxygen Demand (BOD), Total Suspended Solids
(TSS), fecal coliform, oil and grease and pH.
3-6
-------
Additional information and EPA's justification for proposing these requirements is
presented in Section Vm of the preamble.
Other Regulatory Changes
EPA's proposal would make several other changes to the existing regulations, which
would require permit authorities to include the following conditions in permits. These conditions
would: require retention of a permit until proper facility closure; establish the method for
operators to calculate the allowable manure application rate; specify restrictions on timing and
methods of application of manure and wastewater to assure use for an agricultural purpose (e.g.,
certain applications to frozen, snow covered or saturated land) to prevent impairment of water
quality; address risk of contamination via groundwater with a direct hydrological connection to
surface water for existing swine, poultry and veal CAFOs; address the risk of improper manure
application off-site by either requiring that the CAFO operator obtain from off site recipients a
certification that they are land applying CAFO manure according to proper agricultural practices
or requiring the CAFO to provide information to manure recipients and keep appropriate records
of off-site transfers, or both; and establish design standards to account for chronic storm events.
More detailed information on the proposed revisions to the NPDES permit requirements
and the effluent guidelines affecting CAFOs is provided in the preamble.
3.2 Summary of ELG Options and NPDES Scenarios Considered by EPA
This section describes the guidelines' technology options ("ELG Options ") and NPDES
alternative scenarios ("NPDES Scenarios") that EPA is proposing, as well as the regulatory
alternatives that were considered by EPA during the development of this rulemaking. These
proposed and alternative options and scenarios are summarized in Table 3-1.
3.2.1 Effluent Guidelines Options
The technology options considered during this rulemaking include the following:
Option 1. This option is equivalent to Option 1 previously described under BPT
(Section 3.1.2). It requires zero discharge from the production area with liquid storage designed,
constructed, and maintained to handle all process wastewater and storm water runoff from the 25-
year, 24-hour storm event. In addition, Option 1 requires management practices to ensure that
the production area, including manure and wastewater storage areas, are being adequately
maintained. Option I also establishes a requirement to develop a PNP which establishes the
proper land application rate for manure and wastewater to meet the nitrogen requirements for the
crops being grown by the CAFO.
3-7
-------
Table 3-1. Summary Description of Options/Scenarios Considered by EPA
Technology Options (E]LG)
Option 1
Option 2 (BPT- all
subcategories)
Option 3 (BAT -
Beef/Heifers^Dairy)
Option 4
Option 5 (BAT -
Swine/Poultry/Veal)
Option 6
Option 7
N-basedland application controls and inspection and recordkeeping requirements
for the production area (described in Section VIII.C.3 of the preamble)
Same as Option 1, but restricts the rate of manure application to a P-based rate
where necessary (depending on specific soil conditions at the CAFO)
Adds to Option 2 by requiring all operations to determine whether the groundwater
beneath the production area has a direct hydrologic connection to surface water; if
so, requires groundwater monitoring and controls
Adds to Option 3 by requiring sampling of surface waters adjacent to production
area and/or land under control of the CAFO to which manure is applied
Adds to Option 2 by establishing a zero discharge requirement from the production
area that does not allow for an overflow under any circumstances
Adds to Option 2 by requiring that large hog and dairy operations install and
implement anaerobic digestion and gas combustion to treat their manure
Adds to Option 2 by prohibiting manure application to frozen, snow covered or
saturated ground
Regulatory Scope Options (NPDES) ,
Scenario 1
Scenario 2
Scenario 3
"Three-Tier"
Scenario 4a
"Two-Tier" (500 AU)
Scenario 4b
Scenario 5
"Two-Tier" (750 AU)
Scenario 6
Retains existing 3-tier structure and establishes additional requirements would
remove the 25-year,24-hour storm exemption
Same as Scenario 1; operations with 300-1,000 AU would be subject to the
regulations based on certain "risk-based" conditions (described in Part
122.23(a)(3)(ii)(B) Option 2 of the proposed regulation)
Same as Scenario 2, but requires all operations with 300-1,000 AU to either apply
for a NPDES permit or to certify to the permit authority that they do not meet any
of the conditions and thus are not required to obtain a permit
Establishes 2-tier structure and applies ELG standard to all operations with more
than 500 AU
Establishes 2-tier structure and applies ELG standard to all operations with more
than 300 AU
Establishes 2-tier structure and applies ELG standard to all operations with more
than 750 AU
Retains existing 3-tier structure and establishes a simplified certification process
(described in Section VII.C.2 of the preamble)
3-8
-------
Option 2. This option is equivalent to Option 2 previously described under BPT. Option
2 builds on the requirements established under Option 1 retaining the same requirements for the
production area. Option 2 further restricts the amount of manure that can be applied to crop land
owned or controlled by the CAFO. Manure and wastewater would have to be applied at the
appropriate rate, taking into account the nutrient requirements of the crop and soil conditions.
Option 2 requires that manure be applied to meet the phosphorus requirements of the crops
grown if soil conditions warrant and if soils have a very high level of phosphorus build-up, no
manure or wastewater could be applied to the crop land owned or controlled by the CAFO.
Option 3. This option is the proposed BAT Option for the beef and dairy subcategories
and the proposed NSPS option for all subcategories (described in Section 3.1.2). Option 3 adds
to the requirements for Option 2 by requiring that all CAFO operations perform an assessment to
determine whether the ground water beneath the feedlot and manure storage area has a direct
hydrological connection to surface water beneath the production area. If a link is established, the
facility must monitor ground water upstream and downstream of the production area to ensure
that they are achieving zero discharge to ground water. EPA assumes that CAFOs will comply
with the zero discharge requirement by installing liners of synthetic material beneath lagoons and
ponds and impervious pads below storage of dry manure stockpiles. EPA's costs for liners reflect
both a synthetic liner and compacted clay to protect the liner and prolong its useful life.
Additional information on why EPA is proposing mis option for BAT and NSPS is presented in
Section Vin of the preamble.
Option 4. Option 4 adds to the requirements for Option 3 by requiring sampling of
surface waters adjacent to feedlots and/or land under control of the feedlot to which manure is
applied. This option requires CAFOs to sample surface water both upstream and downstream
from the feedlot and land application areas following a one half inch rain fall not to exceed 12
samples per year. The samples would be analyzed for concentrations of nitrogen, phosphorus and
total suspended solids (TSS). These pollutants are believed to provide an adequate indication of
whether a discharge is occurring from the operation. Any difference in concentration between the
upstream and downstream samples would be noted. This monitoring requirement could provide
some indication of discharges from the land application or feedlot areas.
Option 5. This option is the proposed BAT Option for the swine, poultry, and veal
subcategories and the proposed NSPS option for Ihese sectors (with the addition of the
groundwater requirements under Option 3, as described in Section 3.1.2). Option 5 establishes a
zero discharge requirement from the production area that does not allow for an overflow under
any circumstances. By keeping precipitation from coming in contact with the animals, raw
materials and waste handling and storage areas, CAFOs could operate the confinement areas and
meet zero discharge regardless of rainfall events. Option 5 includes the same land application
requirements as Option 2, which would restrict the rate of manure and wastewater application to
a phosphorus based rate where necessary depending on the specific soil conditions at the CAFO.
Additionally, application of manure and wastewater would be prohibited within 100 feet of
3-9
-------
surface water. Additional information on why EPA is proposing this option for BAT and NSPS
is presented in Section VIE of the preamble.
Option 6. Option 6 adds to the requirements of Option 2 by requiring that large hog and
dairy operations (EPA would apply Option 6 to hog operations and dairies with more than 2,000
AU) install and implement anaerobic digestion and gas combustion to treat their manure. With
proper management, such a system can be used to generate additional on-farm revenue.
Anaerobic digestion requires that the treatment be performed in an enclosed system to allow for
the capture, collection and transmission of the methane gas, to a combustion device (i.e., engine,
generator, boiler, and/or absorption cooler). The enclosed system will reduce air emissions,
especially odor and hydrogen sulfide, and potentially reduce nitrogen losses from ammonia
volatilization. The treated effluent will also have less odor and should be more transportable
relative to undigested manure, making off-site transfer of manure more economical.
Option 7. Option 7 adds to the requirements of Option 2 by prohibiting manure
application to frozen, snow covered, or saturated ground. This prohibition requires that CAFOs
have adequate storage to hold manure for the period of time during which the ground is frozen, or
saturated. The necessary period of storage ranges from 45 to 270 days depending on the region.
In practice, this may result in some facilities needing storage to hold manure and wastes for 12
months. EPA is requesting comment on whether there are specific conditions which warrant a
national standard that prohibits application when the ground is frozen, snow covered, or
saturated. "
3.2.2 NPDES Scenarios
Under the NPDES permit program, all point sources that directly discharge pollutants to
waters of the U.S. must apply for a NPDES permit and may only discharge pollutants in
compliance with the terms of that permit. NPDES permits may be issued by EPA or a State,
Territory, or Tribe authorized by EPA to implement the NPDES program.
A NPDES permit may be either an individual permit tailored for a single facility or a
general permit applicable to multiple facilities within a specific category. General NPDES permits
are available to address a category of discharges that involve similar operations with similar
wastes. The general permit specifies the type or category of facilities that may obtain coverage
under the permit. Those facilities that fall within this category then must submit a "notice of
intent" (NOI) to be covered under the general permit to gain permit coverage. EPA anticipates
that the Agency and authorized States will use general NPDES permits to a greater extent than
individual permits to address CAFOs.
EPA's NPDES Scenarios differ in terms of the number of operations that would be
affected by the proposed regulations. EPA's estimate of the number of animal confinement
3-10
-------
operations that would be defined or designated as CAFOs is presented in Section 2 of this report.
The NPDES scenarios considered during this rulemaking include the following:
Scenario 1. This scenario retains the existing three-tier structure and the conditions for
defining the middle tier AFOs as CAFOs. That is, any AFO that meets the size condition
(operations with between 300 AU to 1,000 AU) would be defined as a CAFO if it also meets one
of the two specific criteria governing the method of discharge, namely, pollutants are discharged
through a man-made ditch, flushing system, or other similar man-made device, or pollutants are
discharged directly into waters of the United States that originate outside of the facility and pass
over, across, or through the facility or otherwise come into direct contact with the confined
animals. EPA is not proposing this scenario because these conditions have proven to be difficult
to interpret and implement for AFOs in the 300 to 1,000 AU size category and have not facilitated
compliance or enforcement. For more information, see Section VTI of the preamble.
Scenario 2. This scenario also retains the existing three-tier structure but modifies the
conditions for defining the middle tier AFOs as CAFOs. That is, any AFO that meets the size
condition (operations with between 300 AU to 1,000 AU) would be defined as a CAFO if it met
one or more of the risk-based conditions, described in Section 3.1.1. In this scenario, owners or
operators of AFOs in the middle tier would not be required to certify to the permit authority that
the facility is not a CAFO. However, all facilities that do meet one or more of the conditions
would have a duty to apply for an NPDES permit. This scenario is not being proposed because of
concerns that there would be no way for the permit authority to know which operations were
taking the exemption and which should, in fact, be applying for a permit. For more information,
see Section VII of the preamble.
Scenario 3. This scenario is the co-proposed three-tier structure that retains the existing
three-tier framework but modifies the conditions for defining AFOs in the middle tier as
CAFOs. That is, any AFO that meets the size condition (operations with between 300 AU to
1,000 AU) would be defined as a CAFO if it met one or more of the risk-based conditions, briefly
described in Section 3.1.1. (More detailed information is available in the proposed regulation and
in Section VII of the preamble). Under this co-proposal, EPA would require all middle tier AFOs
to either apply for an NPDES permit or to certify to the permit authority that they do not meet
any of the conditions which would require them to obtain a permit. Additional information on
why EPA is co-proposing this scenario is presented in Section "VTI of the preamble.
Scenario 4a~ This scenario is the co-proposed two-tier structure that establishes which
operations are defined as CAFOs based on size alone (described in Section 3.1.1). In this
alternative, EPA is proposing that the threshold for defining operations as CAFOs be equivalent
to 500 animal units (AU). All operations with more .than 500 AU would be defined as CAFOs.
Operations with fewer than 500 AU would be CAFOs only if designated by EPA or the State
permit authority. Additional information on why EPA is co-proposing this scenario is presented
in Section VII of the preamble.
3-11
-------
Scenario 4b. This scenario is an alternative to Scenario 4a under the co-proposed two-
tier that establishes which operations are defined as CAFOs based on size alone. In this
alternative, EPA would define operations as CAFOs be equivalent to 300 AU. All operations
with 300 or more animal units would be defined as CAFOs. Operations with fewer than 300
animal units would be CAFOs only if designated by EPA or the State permit authority.
Scenario 5. This scenario is an alternative to Scenario 4a under the co-proposed two-tier
structure that establishes which operations are defined as CAFOs based on size alone. In this
alternative, EPA would define operations as CAFOs be equivalent to 750 AU. All operations
with 750 or more animal units would be defined as CAFOs. Operations with fewer man 750
animal units would be CAFOs only if designated by EPA or the State permit authority. After
considering each of these alternatives (Scenarios 4a and 5), EPA is proposing 500 AU as the
appropriate threshold for a two-tier structure, but is also requesting comment on a threshold of
750 AU.
Scenario 6. This scenario is an alternative to Scenario 3 under the co-proposed three-tier
structure that retains the existing tiers but amends the conditions under which middle tier AFOs
with between 300 AU to 1,000 AU are defined as CAFOs. These operations would be required
to obtain an NPDES permit unless they can certify that they do not meet the conditions for
definition as a CAFO, as is being proposed in Scenario 3. However, the conditions for making
this certification would be different than those under Scenario 3, and the substantive permit
requirements for operations between 300 and 1,000 AU that do not certify would be different
than those for CAFOs with more than 1,000 AU. For more information, see Section Vn of the
preamble.
3-12
-------
SECTION FOUR
METHODOLOGY FOR ESTIMATING
COMPLIANCE COSTS AND ECONOMIC IMPACTS
This section presents the data and methodologies EPA uses to estimate the total annual
incremental costs and the economic impacts that would be incurred by the livestock and poultry
industry as a result of the proposed revisions to the ELG and NPDES regulations affecting
CAFOs ("proposed CAFO regulations"). EPA's regulatory impact analysis examines potential
effects across three industry segments: CAFOs (e.g., feedlots and feedyards), processors (e.g.,
meat packers and slaughtering facilities), and consumer markets.
Section 4.1 briefly summarizes the methods and assumptions used to estimate annual
CAFO level compliance costs (described in greater detail in other rulemaking support
documents). Section 4.2 presents the development and characterization of EPA's representative
"model CAFOs" that are differentiated by commodity sector, facility size, and major production
region, among other factors. These model CAFOs are the basis for calculating the total annual
costs of the proposed CAFO regulations and are used to evaluate potential financial impacts to
regulated CAFOs. The approach EPA uses to evaluate impacts to CAFOs provides a general
framework to assess the potential upper bound of costs and impacts that could accrue to
processors and meat packers, as discussed in Section 4.3. Finally, Section 4.4 discusses the
methodology EPA uses to assess additional market impacts, including national level changes in
prices and available quantities, as well as changes in national aggregate employment and economic
output.
4.1 ANNUAL COMPLIANCE COSTS
The sections that follow briefly summarize the methodology EPA uses to develop the
incremental compliance costs that are used to estimate total annual regulatory costs and economic
impacts. Topics include: (!) EPA's assumptions regarding the degree of compliance with the
existing regulatory program; (2) the methodology EPA uses to develop the incremental
compliance costs for each of the regulatory options under consideration; and (3) the inputs and
assumptions EPA uses to annualize those compliance costs. More detailed information on EPA's
baseline assumptions and the data used to calculate costs are provided in EPA's Development
Document (USEPA, 2000a). Appendix A of this report provides additional detail on EPA's
approach for annualizing compliance costs for input into the economic model.
4-1
-------
4.1.1 Baseline Compliance Assumption
For the purpose of this analysis, EPA assumes that all CAFOs that would be subject to the
proposed regulations are currently in compliance with the existing regulatory program (including
the NPDES regulations and the effluent limitations guidelines and standards for feedlots) and
existing state laws and regulations. As a practical matter, EPA recognizes that this is not true,
since only 2,500 operations out of an estimated 12,700 CAFOs with more than 1,000 AU have
actually obtained coverage under an NPDES permit and the remainder may in fact experience
additional costs to comply with the existing requirements. EPA has not estimated these additional
costs in the analysis that is presented in the preamble because the Agency did not consider these
costs part of the incremental costs of complying with the proposed CAFO rule.
To assess the incremental costs attributable to the proposed rules, EPA's analysis
evaluates current federal and state requirements for animal feeding operations and calculates
compliance costs of the proposed requirements that exceed the current requirements. Operations
located in states that currently have requirements that meet or exceed the proposed regulatory
changes would already be in compliance with the proposed regulations and would not incur any
additional cost These operations are not included as part of the cost analysis. A review of
current state waste management requirements for determining baseline conditions is included in
the Development Document and also in other sections of the record. See State Compendium:
Programs and Regulatory Activities Related to Animal Feeding Operations compiled by EPA
(USEPA, 20001).
EPA's analysis also accounts for current structures and practices that are assumed to be
already in place at operations that may contribute to compliance with the proposed regulations.
Additional information is also provided in the following section (Section 4.1.2). This information
is also provided in the Development Document (USEPA, 2000a).
4.1.2 Method for Estimating CAFO Compliance Costs
4.1.2.1 Compliance Costs to CAFO Operators
For the purpose of estimating total costs and economic impacts, EPA calculates the costs
of compliance for CAFOs to implement each of the regulatory options being considered. (Section
3 of this report describes each of the regulatory technology options considered in this proposed
rulemaking.) EPA estimates costs associated with four broad cost components: nutrient
management planning, facility upgrades, land application, and technologies for balancing on-farm
nutrients. Nutrient management planning costs include manure and soil testing, record keeping,
monitoring of surface water and groundwater, and plan development. Facility upgrades reflect
costs for manure storage, mortality handling, storm water and field runoff controls, reduction of
fresh water use, and additional farm management practices. Land application costs address
agricultural application of nutrients and reflect differences among operations based on cropland
4-2
-------
availability for manure application. Specific information on the capital costs, annual operating and
maintenance costs, start-up or first year costs, and also recurring costs that EPA uses to estimate
costs and impacts of the proposed regulations is provided in the Development Document
(USEPA, 2000a). EPA uses these model CAFOs to estimate aggregate incremental costs to the
CAFO industry.
EPA evaluates compliance costs using a representative facility approach based on more
than 170 farm level models that were developed to depict conditions and to evaluate compliance
costs for select representative CAFOs (USEPA, 2000a). The major factors used to differentiate
individual model CAFOs include the commodity sector, the farm production region, and the
facility size (based on herd or flock size or the number of animals on-site). EPA's model CAFOs
primarily reflect the major animal sector groups, including beef cattle, dairy, hog, broiler, turkey,
and egg laying operations. Practices at other subsector operations are also reflected in the cost
models, such as replacement heifer operations, veal operations, flushed caged layers, and hog
grow- and farrow-finish facilities (USEPA, 2000a).1 EPA uses model facilities with similar waste
management and production practices to depict operations in regions that were not separately
modeled.
Another key distinguishing factor incorporated into EPA's model CAFOs is information
on the availability of crop and pasture land for land application of manure nutrients. For this
analysis, nitrogen and phosphorus rates of land application are evaluated for three categories of
cropland availability: Category 1 CAFOs are assumed to have sufficient cropland for all on-farm
nutrients generated, Category 2 CAFOs are assumed to have insufficient cropland, and Category
3 CAFOs are assumed to have no cropland (USEPA, 2000a). EPA uses 1997 information from
USDA to determine the number of CAFOs within each category. This information takes into
account which nutrient (nitrogen or phosphorus) is used as the basis to assess land application and
nutrient management costs.
For Category 2 and Category 3 CAFOs, EPA evaluates additional technologies that may
be necessary to balance nutrients. EPA evaluates additional technologies that reduce off-site
hauling costs associated with excess on-farm nutrients, as well as to address ammonia
volatization, pathogens, trace metals, and antibiotic residuals. These technologies may include
Best Management Practices (BMPs) and various farm production technologies, such as feed
management strategies, solid-liquid separation, composting, anaerobic digestion, and other
retrofits to existing technologies. EPA considers all these technologies for identification of "best
. available technologies" under the various options for BAT described in Section VIE of the
preamble.
EPA uses soil sample information compiled by researchers at various land grant
universities to determine areas of phosphorus and nitrogen saturation, as described in the
'Grow-finish operations finish more mature pigs while farrow-finish operations handle all stages of
„ production from breeding to finishing.
4-3
-------
Development Document (USEPA, 2000a). This information provides the basis for EPA's
assumptions of which facilities would need to apply manure nutrients on a phosphorus- or
nitrogen-based standard.
EPA's cost models also take into account other production factors, including climate and
farmland geography, land application and waste management practices and other major
production practices typically found in the key producing regions of the country. Model facilities
reflect major production practices used by larger confined animal farms, generally those with more
than 300 AU. Therefore,.the models do not reflect pasture and grazing type farms, nor do they
reflect typical costs to small farms. EPA's cost models also take into account practices required
under existing state regulations and reflect cost differences within sectors depending on manure
composition, bedding use, and process water volumes. More information on the development of
EPA's cost models is provided in the Development Document (USEPA, 2000a).
To estimate aggregate incremental costs to the CAFO industry from implementing a
particular technology option, EPA first estimates the total cost to a model facility to employ a
given technology, including the full range of necessary capital, annual, start-up, and recurring
costs. Additional detailed information on the baseline and compliance costs attributed to model
CAFOs across all sectors and across all the technology options considered by EPA is provided in
foe Development Document (USEPA, 2000a).
After estimating the total cost to an individual facility to employ a given technology, EPA
then weights the average facility level cost to account for current use of the technology or
management practice nationwide. This is done by multiplying the total cost of a particular
technology or practice by the percent of operations that are believed to use this particular
technology or practice in order to derive the average expected cost that could be incurred by a
model CAFO. EPA refers to this adjustment factor as the "frequency factor" and has developed
such, a factor for each individual cost (i.e. each technology) and cost component (i.e. capital and
annual costs) in each of its CAFO models (USEPA, 2000a). The frequency factor reflects the
percentage of facilities that are, technically, already in compliance with a given regulatory option
since they already employ technologies or practices that are protective of the environment. The
frequency factor also accounts for compliance with existing federal and state regulatory
requirements as well as the extent to which an animal sector has already adopted or established
management practices to control discharges.
EPA has developed its frequency factors based on data and information from USDA's
NRCS andNAHMS, state agricultural extension agencies, industry trade groups and industry-
sponsored surveys, academic literature, and EPA's farm site visits. More detailed information on
how EPA developed and applied these weighting factors is provided in the Development
Document (USEPA, 2000a). To identify where farm level costs-may be masked by this weighting
approach, EPA evaluates costs with and without frequency factors. The results of this sensitivity
analysis indicate that the model CAFO costs used to estimate aggregate costs and impacts, as
presented in this preamble, are stable across a range of possible frequency factor assumptions.
4-4
-------
The data and information EPA vises to develop its model CAFOs are compiled by USD A,
in combination with other information collected by EPA from extensive literature searches, more
than 100 farm site visits, and numerous consultations with industry, universities, and agricultural
extension agencies. Additional detailed information on the data and assumptions EPA uses for its
model CAFOs is provided in the Development Document (USEPA, 2000a).
4.1.2.2 Compliance Costs to Recipients ofCAFO Manure
To calculate the cost to offsite recipients of CAFO manure under the proposed
regulations, EPA builds upon the cropland availability information in the CAFO models, focusing
on the two categories of farms that have excess manure nutrients and that need to haul manure
offsite for alternative use or to be spread as fertilizer (i.e., Category 2 and Category 3 CAFOs,
where facilities are assumed to have insufficient or no available cropland to land apply nutrients,
respectively). EPA also uses this information to determine the number of affected operations
under select regulatory alternatives, discussed in Section 2, as well as to determine the number of
offsite recipients affected under select regulatory alternatives (USEPA, 2000a).
USD A defines farm level "excess" of manure nutrients on a confined livestock farm as
manure nutrient production less crop assimilative capacity (Gollehon and Caswell, 2000). USDA
estimates manure nutrient production using the number of animals by species, standard manure
production per animal unit, and nutrient composition of each type of manure. Recoverable
manure is the amount that can be collected and disposed by spreading on fields or transporting off
the producing farm (Gollehon and Caswell, 2000).
Depending on the nutrient used to determine the rate of manure application (nitrogen or
phosphorus), EPA estimates that approximately 7,500 to 10,000 CAFOs with more than 300 AU
are expected to generate excess manure. This includes about 2,600 animal feeding operations that
have no major crop or pasture land. These estimates are derived from a USDA analysis of
manure nutrients relative to the capacity of cropland and pastureland to assimilate nutrients.
EPA's estimate does not account for excess manure that is already disposed of via alternative uses
such as pelletizing or incineration. More detailed information is provided in the Development
Document (USEPA, 2000a). .
For the purpose of this analysis, EPA assumes that affected offsite facilities are field crop
producers who use CAFO manure as a fertilizer substitute. Information on crop producers that
currently receive animal manure for use as a fertilizer substitute is not available. Instead, EPA
approximates the number of operations that receive CAFO manure and may be subject to the
proposed regulations based on the number of acres that would be required to land apply manure
nutrients generated by Category 2 and Category 3 CAFOs. EPA assumes that offsite recipients
will only accept manure when soil conditions allow for application on a nitrogen basis. Therefore,
the manure application rate at offsite acres in a given region is the nitrogen-based application rate
for the typical crop rotation and yields obtained in that region. EPA then estimates the number of
4-5.
-------
farms that receive CAPO manure by dividing the acres needed to assimilate excess manure
nitrogen by the national average farm size of 487 acres, based on USDA data. The results of this
analysis indicate that 18,000 to 21,000 offsite recipients would receive excess CAFO manure
(USEPA, 2000a).
EPA's estimated costs to manure recipients include the costs of soil testing and
incremental recordkeeping. EPA evaluates these costs using the approach described in Section
4.1.2.1. Excess manure hauling costs are already included in costs assessed to CAFOs with
excess manure. For the purpose of this analysis, EPA assumes that crop farmers already maintain
records documenting crop yields, crop rotations, and fertilizer application, and that crop farmers
already have some form of nutrient management plan for determining crop nutrient requirements.
EPA estimates, on average, per-farm incremental costs of approximately $540 to CAFO manure
recipients for complying with the offsite certification requirements. This analysis is provided in
the Development Document (USEPA, 2000a).
4.1.3 Cost Annualization Methodology
EPA develops CAFO compliance costs from estimated startup (first-year) costs, recurring
3- and 5-year costs, and annual operating and maintenance costs (described in Section 4.1.2). To
compare estimated costs to available farm financial data in any one year, EPA annualizes these
costs using the approach described below. A 1997 time frame is used for comparison with
available USDA data on livestock and poultry farms, as reported in the 1997 Census of
Agriculture and other related databases. The inputs and assumptions of the cost annualization
model are summarized below.
Additional detail on the input costs used for this analysis are provided in the Development
Document (USEPA, 2000a). A summary of the annualized costs by CAFO model is provided
later in this EA in Sections 6, 7, and 8, depending on commodity sector. Additional details on the
annualization model are provided in Appendix A of this report. Annualized compliance costs are
also presented in Appendix A (expressed in 1997 dollars).
Annualized costs account for the time value of money and reflect the annual repayment
amount of an on-farm capital investment by spreading the initial costs over the expected life of the
structure. EPA calculates both pre-tax and post-tax annual costs. Pre-tax costs are used to
estimate the total cost of the proposed CAFO regulations to society (social costs). Post-tax costs
are used to measure the economic impacts at CAFOs and to account for the reduction in a
CAFO's tax liability. This reduction in taxes paid ("tax shield") acts to offset the expected
compliance costs incurred by a facility. This portion of costs is borne by federal and state
governments through a reduction in tax revenue. Accordingly, the economic impacts of the
proposed CAFO regulations are measured as the impact of the expected compliance cost incurred
by a CAFO minus an appropriate tax shield.
4-6
-------
The major inputs to and assumptions used in the cost annualization model are: (1) the
discount rate; (2) the life of the asset; and (3) tax rates.
EPA uses the discount rate to calculate the present values of the cash flows and is
analogous to an interest rate used to compute a mortgage payment. The annualization model uses
a real discount/interest rate of 7 percent, as recommended by the Office of Management and
Budget (OMB, 1992), and does not have to be adjusted for inflation.
The life of the asset is the time period over which the costs are to be annualized (like a
mortgage time period) and is determined according to the Internal Revenue Code's classes of
depreciable property. The time period over which the annualization is made depends on the
serviceable life of the structure and also on the depreciable life, which affects what portion of a
capital cost can be used each year to reduce taxable income. IRS rules govern the designation of
depreciable life, which is assigned on the basis of serviceable life. Most of the types of capital
investments that will be required under these proposed regulations are typically depreciated over
10 years (IRS, 1999).2 The cost annualization model thus incorporates a 10-year annualization
period to compute both pre-tax and post-tax annual costs. The equation EPA uses to calculate
annual cost operates from mid-year to mid-year (mid-year convention); thus the entire time frame
of the analysis begins in Year 1 and concludes in Year 11 but is discussed here as a 10-year time
frame (see Appendix A).
The assumed tax rates are used to determine a facility's tax benefit or tax shield.
Estimated tax savings are subtracted from the actual outlay in each year and are used to calculate
the annual post-tax cost of compliance. The tax shield calculation uses three inputs: marginal tax
rates (composed of the federal tax rates and an assumed state tax rate); an estimate of CAFO
taxable income (net cash minus depreciation plus value of inventory change, assuming the CAFO
is using the accrual method of accounting); and a depreciation schedule.
Appendix A provides a description of how the tax rates are assigned to each model
CAFO. Revenue estimates are derived from USDA data for representative model CAFOs
(discussed in Section 4.2.3). The depreciation schedule is dictated by IRS rules, but there are
several choices of depreciation schedule within those rules. For reasons outlined in Appendix A,
EPA has chosen the Modified Accelerated Cost Recovery System (MACRS), which is a
commonly used and generally advantageous depreciation schedule for tax minimization purposes.
The tax shield is calculated using the depreciable capital cost in each year plus any recurring
expenditures that are allowed to be expensed in each year. Each model CAFO is assigned a tax
rate (marginal federal rate plus an assumed state tax rate) based on the estimated amount of
2Many of the types of investments would best be classified as single-purpose agricultural structures, which
are defined by IRS as any enclosure or structure specifically designed, constructed, and used for housing, raising,
and feeding a particular kind of livestock, including structures to contain produce or equipment necessary for
housing, raising, and feeding livestock..
4-7
-------
taxable income. The tax rate times the depreciated and expensed compliance costs in each year
equals the estimated tax shield.
Appendix A presents a sample spreadsheet that shows how all of the computations are
made and also documents all annualized incremental costs estimated by EPA. Section 5 of this
report presents the aggregate, national level annualized compliance costs (post-tax), by regulatory
option, that are calculated by the cost annualization model. Sections 6, 7. and 8 present the per-
head and per-CAFO annualized costs and the economic impacts (post-tax) for selected model
CAFOs for each of the major livestock and poultry sectors.
4.2 CAFO ANALYSIS
4.2.1 Overview of the Representative CAFO Approach
EPA estimates incremental costs and economic impacts of the proposed CAFO
regulations using a representative farm approach. For this analysis, EPA developed "model
CAFOs" for each sector to assess the average costs and economic impacts of the proposed
regulations across differently sized, differently managed, and geographically distinct operations.
Using a representative farm approach, EPA constructs a series of model facilities that
reflect the EPA's estimated compliance costs and available financial data. EPA uses these model
CAFOs to develop an average characterization for a group df operations. EPA's cost models are
described in Section 4.1.2.1. From these models, EPA estimates total annualized compliance
costs by aggregating the average facility costs across all operations that are identified for a
representative group. As with EPA's cost models, EPA's financial models are grouped according
to certain distingiushing characteristics for each sector, such as facility size and production region,
that may be shared across a broad range of facilities. Economic impacts under a post-regulatory
scenario are approximated by extrapolating the average impacts for a given model CAFO across
the larger number of operations that share similar production characteristics and are identified by
that CAFO model. EPA compares its estimated compliance costs at select model CAFOs to
corresponding financial conditions at these model facilities.
The representative farm approach is consistent with past research conducted by USDA
and the land grant universities and then- affiliated research organizations, including the Food and
Agricultural Policy Research Institute (FAPRI), the Center for Agriculture and Rural
Development (CARD), Texas A&M's Agriculture and Food Policy Center, the Texas Institute for
Applied Environmental Research (TIAER) and the University of Missouri's Commercial
Agriculture Program. These organizations and others have widely adopted the representative
farm approach to assess a broad range of policy issues, including changes in federal agricultural
policy and pricing programs, domestic food programs, environmental legislation, and international
trade. This approach has been used to assess agricultural market changes for both livestock and
crop commodities (AFPC, 1999 and 2000; Skinner, 1981). Studies conducted by Heimlich and
4-8
-------
Barnard (1995), Outlaw et al. (1993), Christensen et al. (1981), and DPRA (1995) have used the
representative farm approach to evaluate the financial impact of implementing management
measures and installing animal waste systems on livestock CAFOs. A representative approach has
been used to evaluate financial impacts in rulemakings where actual facility level data were not
available (DPRA, 1995).
A representative approach provides a means to assess average impacts across numerous
facilities by grouping facilities into broader categories to account for the multitude of differences
among animal confinement operations. Use of this approach is considered necessary to account
for differences in performance among farming operations.
Costs to manage animal waste are site specific and depend on the waste disposal and
storage requirements of an individual operation. Estimates of the range of potential costs to
operations show that costs may vary with respect to equipment and maintenance procedures
necessary to meet regulatory requirements (USGAO, 1995; Fleming et al.,,1997; Bennet et al.,
1992). The ability of an operation to absorb these costs may vary across producers. For
example, the number of animals raised or housed on site directly affects the magnitude of total
costs; however, a facility's size can also influence an operation's ability to pay. Many studies
suggest that there is a statistically positive relationship between farm size and per-unit production
costs, such that as farm size (number of animals) increases, costs per-animal are generally lower
(Perry and Green, 1999; Van Arsdall and Nelson, 1985; Kumbhakar et al., 1989; Purdy et al.,
1997; Weersink and Howard, 1991; Lazarus, et al., 1999). This may result in lower per-unit
capital costs and create a relative advantage among larger operations.
Previous studies have also noted that on-farm improvement costs may vary by farm
production region (Westerbarger and Letson, 1995; Outlaw et al., 1993). Regional differences
may also affect farm financial performance and may influence an operation's ability to pay for
these improvements (Outlaw et al., 1993; El-Osta and Johnson, 1998). Specific requirements for
handling livestock and poultry manure may vary because conditions differ across farms and among
producing regions (NMPF, 1996). Compliance costs may be higher in some regions than in
others, depending on the types of technologies that may be required to manage waste and on
various site-specific farmland characteristics. Such factors include topography, climate, average
rainfall, soil type and conditions, underlying rock formations, and local evaporation rates. For
example, producers in regions with limited cropland on which to apply manure may face higher
disposal costs. Producers that are located near targeted or vulnerable waterways may also face
higher costs.
A literature review of the research that examines the potential for size and scale economies
in these industries, among other factors that contribute to differences in farm performance among
producers, is available in the rulemaking record (ERG, 2000d—see DCN 70641).
At various stages in the proposed rulemaking, EPA has presented its proposed
methodological approach to USDA personnel and to researchers at various land grant universities
4-9
-------
for informal review and feedback (Foster, 2000a; Johnson et al., 1999a and 2000a; USDA, 1999;
USEPA, 1999b and 1999c; Vukina, 2000). The Development Document (USEPA, 2000a)
documents additional review of EPA's cost models. (See Section XII of the preamble for a
summary of EPA's outreach activities; additional information is available in the rulemaking
record:)
4.2.2 Construction of EPA's Model CAFOs
EPA's model CAFOs are developed to conform to a representative farm approach. These
model CAFOs reflect average conditions for selected groups of CAFO operations. EPA's model
CAFOs. consist of two major modeling components: cost models and financial models.
Three factors are recognized to have a major impact on the way CAFOs operate and
therefore are assumed as key factors to differentiate the individual model CAFOs. These factors
include: (1) commodity sector (beef, veal, heifers, dairy, hogs, broilers, layers, turkeys), (2) farm
production region, and (3) facility size (based on herd or flock size or the number of animals on
site). Both EPA's cost models and financial models are differentiated by sector and select region
and size group.
The CAFO models EPA uses for this analysis represent the interface between a larger
number of cost models and a smaller number of financial models. Fewer financial models are
developed because of data availability issues (discussed in detail in Section 4.2.3). The cost
models are able to accommodate a number of additional distinguishing factors compared to the
financial models. EPA's cost models can divide sectors into various subsectors (e.g., hogs are
divided into grow-finish and farrow-finish operations, and egg layers are divided into operations
that utilize liquid and dry manure management systems). The cost models also provide more
refined production regions (e.g., five regions as opposed to iwo) and additional facility size
representations. Furthermore, the cost models can account for the type of animal production
facility, the availability of crop and pasture land to land apply manure nutrients, farmland
geography, and existing state regulatory requirements, as well as reflect cost differences within
sectors based on manure composition, bedding use, and process water volumes. For more
detailed information on the cost models EPA uses for this analysis, see the Development
Document (USEPA, 2000a).
• The sections below highlight the three primary factors (sector, region, and size group) that
can be differentiated by both the cost models and the financial models and briefly discuss how
model CAFOs are distinguished by these factors. Additional details on the development of the
model CAFOs are presented in Sections 4.2.3 and 4.2.4.
4-10
-------
4.2.2.1 Livestock and Poultry Sectors
EPA's model CAFOs cover the following livestock and poultry sectors:
• Fed beef cattle operations
» Veal operations
• Heifer operations
• Dairy farms
• Hog farms, including grow-finish and farrow-finish operations
• Egg laying operations, including facilities with both liquid and dry manure
management systems
• Broiler operations
• Turkey operations
EPA's cost models distinguish all these sectors and subsectors. The financial models,
however, are based on USDA data that are collected across all beef, dairy, hog, egg laying,
broiler, and turkey sectors. These data do not distinguish finances at various subsector levels
within these sectors, for example, at the level of grow- versus farrow-finish operations in the hog
sector, or fed cattle versus stand-alone veal or heifer operations in the cattle sector.
4.2.2.2 Farm Producing Regions
As a starting point for determining the relevant farm production regions for its model .
CAFOs, EPA refers to USDA's ten farm regions (Figure 4-1). EPA's cost models aggregate
USDA's regions into five broad production regions for the purposes of estimating costs: Midwest
(MW); Central (CE); Pacific (PA); Mid-Atlantic (MA); and South (SO).3 The financial models, •
however, cannot distinguish this many regions per sector because available USDA financial data
do not allow for aggregation at this level of detail.4 Therefore, two regions are represented by the
model CAFOs per sector, representing the major production regions for each sector.
3Information on these regional groupings is provided in the Development Document (USEPA, 2000a).
4 Aggregation of data into broader regional groups is necessary to ensure the confidentiality of USDA's
respondent level survey data and to maintain a statistically representative sample of survey data.
4-11
-------
Northern! lake Slates m
• gauss §
———I%=B=B
Figure 4-1. USD A Farm Producing Regions
Source: USDA/ERS, 2000a.
Table 4-1 shows the regions represented by the model CAFOs for each sector.
Section 4.2.4 discusses in more detail how the model CAPO regions relate to the ten USDA
regions. CAFOs in regions other than the two major production regions "hi each sector are
allocated to the two major regions on an equally weighted basis (i.e., within each sector, 50
percent of CAFOs in regions other than the two primary production regions are assigned to one
major production region and the other 50 percent are assigned to the other). 'The only exceptions
are the veal, heifer, and wet layer sectors, which are modeled using only one region each due to
limited data and information on these subsectors.
4.2.2.3 Facility Size
EPA establishes its model CAFOs based on three broad facility size groups: CAFOs with
more than 1,000 AU;5 CAFOs with between 500 and 1,000 AU; and CAFOs with less than 500
AU. For most sectors, EPA's model CAFOs generally distinguish between two facility sizes
within each group, i.e., "Large 1" and "Large 2" operations among CAFOs with more than 1,000
AU; and "Medium 1" and "Medium 2" operations among CAFOs with between 500 and 1,000
As defined for the proposed CAFO regulations, one AU is equivalent to one slaughter or feeder cattle,
calf or heifer; 0.7 mature dairy cattle; 2.5 hogs (over 55 pounds) or 5 nursery pigs; 55 turkeys; and 100 chickens
regardless of the animal waste system used.
4-12
-------
Table 4-1: Model CAFOs by Sector, Size, and Region (Size Ranges and Average Inventory)
Sector
Fed
Cattle
Veal
Dairy
Heifers
Hog: FF
Hog:
GF
Layers:
Wet
Layers:
Dry
Broilers
Turkeys
Region
MW,
CE
MW
PA,
MW
MW
MA
MW
MA'
MW
so
MW
SO
MA
SO
MA,
SO
MA,
MW
Medium l(a)
Medium l(b)
Medium 2
300-1000AU
300-500
(455)
300-500 (400)
200-350
(235)
300-500 (400)
750-1,250
(814)
750-1,250
(846)
750-1,250
(900)
750-1,250
(963)
-T-"^ * ,/
*• f. ^ j w *
*v
*"" ^ 5~"*/
30,000-41,700
(32,375)
30,000-41,700
(37,906)
30,000-40,000
(36,634)
30,000-40,000
(36,796)
16,500-25,700
(18,539)
16,500-25,700
(18,092)
1,250-1,875
(1,460)
1,250-1,875
(1,518)
1,250-1,875
(1,422)
1,250-1,875
(1,521)
•y. ~> " *" * ""
•*•*-,•*-„ j^
•V ! \ ?*• *
9 ""' % -r >
41,700-62,500
(44,909)
41,700-62,500
(52,582)
40,000-60,000
(51,362)
40,000-60,000
(51,590)
25,700-38,500
(31,267)
25,700-38,500
(30,514)
500-1,000
(777)
>750 (540)
525-700
(460)
750-1,000'(750)
1,875-2,500
(2,152)
1,875-2,500
(2,165) .
1,875-2,500
(2,124)
1,875-2,500
(2,184)
9,000-30,000
(3,624)-
62,500-180,000
(97,413)
62,500-180,000
(97,484)
60,000-90,000
(73,776)
60,000-90,000
(73,590)
38,500-55,000
(45,193) '
38,500-55,000
(45,469)
Large 1
Large 2
>1,000 AU
1,000-8,000
(1,877)
^iSSite^"
*fe^v-V'' ^I^Sfe ^•^"^t' *&$**
>700
(1,419)
>1,000 (1,500)
2,500-5,000
(3,444) .
2,500-5,000
(3,509)
2,500-5,000
(3,417)
2,500-5,000
(3,554)
>30,000
(86,898)
180,000-600,000
(293,512)
180,000-600,000
(279,202)
90,000-180,000
'(117,581)
90,000-180,000
(115,281)
>55,000
(97,111)
>55,000
(158,365)
>8,000
(30,003)
rf „ ^ ^
* t^ ~ »*>
^^ &*f*. ~ . ^<^
„<-„_,£?„' Jt /
J "">" ,^S
~*s" ^ "* ^.v.
& r<^ ^"^f
>5,000
(13,819)
>5,000
(17,118)
>5,000
(10,029)
>5,000
(8,893)
&~^et ^ ^- J^^ ^
^^ ^ % ,,
fff " c -"" „ . 'f
*^*~ *™ sH n ^ ^*
-j». ~ -*-.•• ^
>600,000
(884,291)
>600,000
(1,229,095)
>180,000
(281,453)
> 180,000
(303,155)
* -^ r />•
}&* „. "° ^ff {*
V ^ »s-*.t
* ^ "4 •*
^ - «^ ^^
j*' "i *"
Source: USEPA, 2000a. Shaded cells indicate model CAFO not developed. Values shown in parentheses are the average
number of animals (inventory) assumed for each model CAFO. EPA's model regions cover USDA farm regions (Figure 4-1):
BeefiVeal/Dairv/Heifer: Midwest (MW)-N. Plains, Lake States, Corn Belt, Northeast, Appalachia; Central (CE)- Mountain, S.
Plains, Delta States, Southeast; Pacific (PA)-Mountain, S. Plains, Delta States, Southeast.
Hogs: Mid-Atlantic (MA)-Northeast, Appalachia; MW-Northem Plains, Lake States, Corn Belt.
Layers/Broilers: South (SO)-Northeast, Appalachia, Mountain, S. Plains, Delta States, Southeast, Mountain, Pacific; Midwest
(MW)-N. Plains, Lake States, Corn Belt; Mid-Atlantic (MAHNortheast, Appalachia, Delta States, Pacific; S. Plains,
Mountain.
Turkeys: MW-N. Plains, S. Plains, Pacific, Mountain, Lake States, Cora Belt; MA-Appalachia, Delta States, NE and SE.
4-13
-------
AU ("Medium" operations are further differentiated between CAFOs with between 300 and 500
AU). Facility size categories vary by sector; the average number of animals represented by each
model CAFO is based on typical inventory estimates that are common for that size range in a
particular sector. Table 4-1 shows the range of facility sizes and the average number of animals
associated with this range that EPA assumes for this analysis for each model CAFQ. See the
Development Document (USEPA, 2000a) for more information on the development of the model
sizes shown in the table.
Data limitations restrict the number of facility size categories available for EPA's financial
models. However, as will be discussed later, EPA uses the available data to derive estimates of
financial data for the given sizes of model CAFOs and to match financial models to cost models
for each model CAFO shown in Table 4-1.
EPA develops the costs to confinement operations with less than 300 or 500 AU that may
be designated as CAFOs by scaling the estimated compliance costs for the available "medium" and
"large" CAFO models. (See Section 2 for information on expected designated facilities under
each co-proposed alternative.) The resulting costs-derived on a per-head basis—are adjusted
by the average head counts at operations with fewer than 500 AU or 300 AU to derive the
annualized per-facitity compliance cost. Costs for CAFOs with fewer than 500 AU or 300 AU
assume that these operations have sufficient cropland for all on-farm nutrients generated
(identified in the cost model as Category 1 costs). More detailed cost information is provided in
the Development Document (USEPA, 2000a).
4.2.3 Sources of Data for EPA's Model CAFOs
For this analysis, EPA is relying on existing data sources and expertise provided by
USDA, industry, state agriculture extension agencies, and various land grant universities.6 Two
major sources of farm level data include USDA's Census of Agriculture and USDA's ARMS.
These databases provide farm level descriptive and financial data and are used to develop
representative model CAFOs for EPA's cost and economic impact assessment. USDA's Census
of Agriculture and ARMS databases are compiled from farm surveys conducted by USDA that
contribute to long-standing data summaries of the U.S. farm sector.
The Census and ARMS databases are maintained by USDA, which periodically publishes
aggregated data from these databases and also compiles customized analyses of the data for
members of the public and other government agencies. The data EPA uses for this analysis are
produced as a customized analysis, and are compiled with the assistance of staff at USDA's
National Agricultural Statistics Service (NASS) and USDA's Economic Research Service (ERS).
6In past ELG rulemakings, EPA has exercised its legal authority under Section 308 of the Clean Water
Act and has conducted an industry-wide survey of all affected facilities to obtain actual facility level business and
financial information.
4-14
-------
In providing such analyses, USDA maintains a sufficient level of aggregation to ensure the
confidentiality of an individual operation's activities or holdings.
The Census is the primary data source used to depict general (nonfinancial) farm
conditions for EPA's model CAFOs and serves as the basis for estimating the range of compliance
costs that may be incurred by CAFOs under the proposed regulations.7 In subsequent stages of
the economic analysis, EPA uses estimates of the number of affected CAFOs to extrapolate the
results of impact analyses to the national level and to aggregate costs across all sectors and
nationally. A more detailed description of the Census data and how these data are used to
develop EPA's model CAFOs are provided in the Development Document (USEPA, 2000a).
The ARMS database is the primary source of farm financial data used to assess the
potential CAFO level economic impacts. Other sources of data also contribute to this analysis as
described hi this report.
The following sections focus on EPA's use of the ARMS data, since these data are the
primary source of financial data used to analyze CAFO level economic impacts in EPA's
representative farm approach. Section 4.2.3.1 provides an overview of the ARMS data,
Section 4.2.3.2 discusses the special compilation of data ERS made available to the public, and
Section 4.2.3.3 presents a general discussion of which level of ARMS data aggregation was used
to develop each of the model CAFOs.
4.2.3.1 Overview of ARMS Financial Data
The ARMS is USDA's primary method for collecting data covering a broad range of
issues about agricultural production practices and costs. These data provide the only national
perspective on annual changes in the financial conditions of production agriculture (USDA/ERS,
2000a). The ARMS is an annual survey conducted using a probability sample, which in 1997
included information from 11,724 surveyed farms nationwide. The sample survey is hand-
enumerated by trained personnel. USDA extrapolates the data from the sample survey to
represent farming and ranching operations in the 48 contiguous states. These national level data
are published in a series of annual and periodic reports, such as USDA ERS's annual
compendiums on farm cost of production and also farm financial performance (data formerly
identified as the Farm Costs and Returns Survey data).
7EPA uses available.Census data to characterize the animal feeding sectors and provide information on a
range of factors, including the number of feedlot operations, how these operations are distributed geographically,
and general size ranges for these operations. Other farm production data reported in the Census are used to
compare the amount of cropland available to land-apply manure nutrients generated relative to crop need. These
latter calculations were provided with the assistance of researchers at USDA's Natural Resources Conservation
Service (NRCS). For more information, see the Development Document (USEPA, 2000a).
4-15
-------
To develop a reasonably complete financial model of a representative CAFO, mformation
on farm assets and liabilities (balance sheet information) and revenues, costs, earnings, taxes, and
net income (income statement information) is required at a minimum. The ARMS data generally
provide breakouts of critical financial line items that are required to construct a financial profile of
the representative CAFOs. ARMS data provide USDA with the means to compile complete
income statements and balance sheets from surveys of operations in each of the major farming
sectors.8 This full financial accounting is a major advantage, since no other sources of farm
financial data are as complete.
The Census does collect some financial data, but they are not sufficiently detailed to
construct financial models. Complete income statement information is not available from this
source, since most survey respondents do not fully report revenue and cost information, thus
requiring that USDA impute missing values. Complete balance sheet data are also not available.
Information available in the ARMS data includes:
" Income statement information. Revenue and expense data include:
— Gross cash income—includes livestock and crop sales (including net
Commodity Credit Corporation loan proceeds), government payments, and
other farm-related income.
— Cash operating expenses—includes Variable expenses such as livestock,
feed, seed, fertilizer, and hired labor, and fixed expenses including land,
insurance, and property taxes.
- Net cash income—gross cash income minus variable and fixed cash
operating expenses.
— Net farm income—includes gross cash income minus variable cash
expenses, fixed expenses, depreciation charges, and labor noncash benefits,
plus the value of inventory change and non-money income.
» Balance sheet information. Asset and liability data include:
— Current assets—such as farm inventory and purchased inputs, excluding
noncurrent assets (such as breeding animals and farm equipment).
— Current liabilities—such as accounts payable, excluding noncurrent
liabilities (such as real estate).
8Farms are classified by sector based on the source of 50 percent of its annual farm revenue (McElroy,
1993). For example, a farm is defined as a hog operation if it generates 50 percent of its farm sales from hogs.
4-16
-------
- Total farm assets—includes current and noncurrent assets (such as
investment in cooperatives, land and buildings, breeding herd, and farm
equipment).
- Total farm liabilities—includes current and non-current liabilities (such as
real estate and debt other than real estate).
• Financial ratios. Indicators of farm financial health include measures of liquidity
(current ratio), solvency (debt-to-asset ratio), profitability (return on assets and
return on equity), and other measures of financial efficiency:
A copy of the income statement and balance sheet summary information received from
ERS is contained hi the rulemaking record (USDA/ERS, 1999a—-see DCN 70063). See Sections
6, 7, and 8 for examples of the available data by sector.
With the ARMS data, farm operations can be classified (grouped or typed) in many ways
based on the information collected on the survey. ERS usually reports the data by farm typology,
which divides farms into small farms (limited-resource, retirement, residential or lifestyle, low
sales, and high sales) and other farms (large family farms, very large family farms, and non-family
farms). An alternative method of typing is by farm business (legal) organization as reported by
the respondent. The mutually exclusive categories for this item include individual proprietorship,
partnerships, family corporations, non-family, corporations and other. Farms can also be grouped
by the occupation of the operator. The categories for operator occupation include farming, hired
manager, non-farm work, and retired. The ARMS data also provide other farm typology data
useful for financial characterization. This information includes a farm classification, developed by
USDA's ERS, that separates U.S. farms into mutually exclusive and homogeneous groups based
on legal organization of family farms by proprietorships, partnerships, and family corporations
that are not operated by a hired manager. USDA's farm typology groups are shown in Table 2-2.
Data on both family and nonfamily farms are included in the ARMS data (USDA/ERS,
2000b and 1999a). (See Table 2-2 and Figure 2-1.) USDA's farm typology data also provide
information on animal ownership based on the percentage of animals raised at the farm site that
are owned or not owned by the farmer (e.g., raised under contract for another business entity)
(USDA/ERS, 1999a). This information is useful for understanding the average data that include
all the different farm structure types that are captured in the aggregated ARMS data for each
livestock and poultry sector.
An advantage of the ARMS data is that data collected for 1997 include information on the
total number of farms and the total number of animals in the sample set. Data of the total number
of animals and farms in the 1997 ARMS data set correspond to the financial data compiled by
ERS for each aggregated data grouping by commodity, production region, and facility size
groups, described in Table 4-2 (USDA/ERS, 1999a—DCN 70063). The match between average
financial information and total numbers of farms and animals allows EPA to calculate average
4-17
-------
financial data on a per-animal basis for each sector. Section 4.2.4.2 discusses how EPA converts
the ARMS financial data onto a per-animal basis to construct its model CAFOs.
The ARMS revenue data are separable by enterprise (livestock and crops), as well as other
farm-related income and government payments. Off-farm revenue is not included in the ARMS
data used by EPA for this analysis, as described previously under revenue definitions for "Income
statement information." However, the ARMS operating expense data are not separable by
enterprise (e.g., buildings, labor and equipment for crop versus livestock production), but
represent average production costs for an operation as a whole. This generally limits the types of
analyses that EPA may conduct using these data for its regulatory impact analysis.
The main limitation of survey data sets, including the ARMS data, is that they reflect
average financial conditions across an entire sector and may not be representative of certain
subsectors within that sector. For example, as discussed in Section 2.2, ARMS financial data are
mixed across farms that graze animals and ones that grow animals under confinement. The cost
structures between these types of operations may vary depending on differences in expenditures
for production inputs, such as feed, labor, equipment, land, and buildings. USDA's data are also
mixed across operations where farming is not the major occupation of farm operator (e.g.,
noncommercial, part-time operations). The inclusion of all types of operations—noncommercial,
higher cost producers, as well as smaller scale production units—may result in the average
financial statistics being too low to be representative. In particular, the inclusion of
noncommercial farms, when expressed across all farms, will result in a lower average annual gross
and net income than would be the case if noncommercial farms were excluded from the data
(USDA/ERS, 1996a). The ARMS data are representative of the population, which contains more
smaller-sized operations and fewer larger-sized operations. As a result, the data may be less
representative of the types of larger scale operations that would be affected by the proposed
CAFO regulations.
In general, the inclusion of financial information on the vast range of operations may result
in lower average data such that the revenue and cost data that are used for this analysis are lower
even higher. This possibility is not very problematic since this would mean that EPA's analysis is
overly conservative. Of more concern are those cases where use of average data may understate
the effects of compliance by overstating a CAFO's baseline financial conditions and its ability to
pay for on-farm improvements under the proposed regulation (i.e., concluding that a model
CAFO is able to afford the estimated regulatory costs when, in fact, it cannot). Examples where
financial conditions may be overstated may include facilities that may perform below the USDA
average (e.g., contract growers) and operations with little available land that are represented by
the average debt-to-asset ratios reported by USDA. Section 4.2.4.5 discusses how EPA matched
up the number of CAFO cost models to the limited number of available financial models. EPA
addresses concerns about the limitations of the financial data, in part, by undertaking extensive
sensitivity analysis of the model CAFO results, which is presented in Appendix D. Variance
analysis of the ARMS financial data is not available.
4-18
-------
Table 4-2. EPA-Requested ARMS Data for Model CAFOs by Sector, Size, and Region
Sector
Beef
Dairy
Hogs
Layers/
Broilers
Turkeys
Number of Animals
"Small"
<200
<100
<800
<90,OQO
(layers)
<30,000
(broilers)
<10,000
"Medium"
200 to 800
100 to 500
800 to
2,500
90,000
to 120,000
30,000
to 90,000
10,000
to 40,000
"Large"
>800
>500
>2,500
>120,000
>90,000
>40,000
EPA-Requested
Aggregations
( Regions)
U.S.
S. Plains (SP)
Ml/Plains (MP)
All Other
U.S.
South (SO)
West(W)
Mid-Atlantic .
(MA)
Midwest (MW)
U.S.
Mid-Atlantic
(MA) ,
Midwest (MW)
U.S.
Southwest (SW)
East (E)
Midwest (MW)
U.S.
West/Midwest
(W/MW)
East (E)
USDA Farm Producing
Regions Represented
Average U.S.
Southern Plains
Northern Plains + Mountain
U.S. less other regions
Average U.S.
Southern Plains + Delta +-SE
Pacific + Mountain
Northeast + Appalachian
Lake St. + Corn Belt + N.
Plains
Average U.S.
Northeast + Appalachian
Lake St. + Cora Belt + N.
Plains
Average U.S.
S. Plains + Pacific + Mountain
NE + Appalachian + Delta + SE
Lake St. + Com Belt + N.
Plains
Average U.S.
Pacific + Mountain + Lake St. +
Corn Belt + N. Plains + S .
Plains
NE + Appalachian + Delta + SE
Cost
Model
Match
*• ;'"A;t',, -o ,"•
'"/Tw''"t^
CE
MW
v~ "t
tftvf
„ 75'*
*-JsJsS~ r~-S£.
t£^'
6
PA
*•
fJif *
MW
MW
MA
?** _*. "* t* Sv
•*> %
so
MW
MW
SO
Source: USEPA, 2000a and USDA/ERS, 1999a. USDA's Farm Producing Regions shown in Figure 1; region
names for EPA's CAFO cost models are defined in Table 4-1. EPA aggregations of USDA regions (as indicated)
are used to avoid disclosure of ARMS data. Shaded cells indicate model CAFO not developed.
4-19
-------
EPA investigated several other potential sources of farm level financial information before
deciding to use the ARMS data. As already discussed, cost and revenue data in the Census are
unsuitable for developing EPA's model CAFOs since they do not provide complete financial
information. Other frequently used sources of financial data, such as Robert Morris and
Associates (RMA) and Dunn & Bradstreet, are not useful for EPA's analysis because they do not
provide detailed farm financial statistics and corresponding information on facility size, such as the
number of animals raised on site, which is needed to present revenue data on a per-animal basis
(which is essential for scaling the financial models to match cost models as explained in Section
4.2.4). An alternative approach that is commonly used by the land grant universities is to consult
an "expert panel" or "focus group" that would consist of state or regional extension specialists to
compile representative, regional financial budgets for each sector. EPA decided not pursue this
approach since it is common practice at the Agency to rely on statistically validated data obtained
from survey questionnaires; also, such an alternative approach would have been time and resource
intensive.
EPA also obtained financial information from industry. In particular, the National
Cattlemen's Beef Association (NCBA) provided aggregated summary information on financial
conditions at cattle feeding operations based on responses to a survey questionnaire of its
membership. Although not intended as a statistical and conclusive financial analysis, these data do
provide a summary of information gathered and submitted on a voluntary basis by individual
feedyards throughout the nation. This information is useful to EPA since.it allows the Agency to
evaluate how well the ARMS data for cow and calf operations represent conditions at cattle
feedyards. More information on these data and how EPA uses these data is provided in Section 8
of this report.
Finally, EPA also explored a modeling approach that calculates the amount of livestock
revenue that is likely to be generated for each representative CAFO based on key market data,
including the USDA-reported price received by producers for raw farm oulput and average yield,
expressed either as animal weight at slaughter or the volume of milk or number of eggs produced
annually. For farms producing meat animals, the model also accounts for the number of
"turnovers" or annual marketing cycles, representing the total number of meat animals produced
and sold for slaughter in a full year cycle. (More information on this approach is provided in the
rulemaking record, see USEPA, 2000e and Appendix J of docket item DCN 93080) EPA did not
use this approach because it most likely understates financial conditions at CAFOs. This
approach cannot account for income from crops or other secondary livestock raised on site, nor
can it account for other supplemental income, such as other farm-related revenue and government
payments.
4.2.3.2 Special Compilation of Representative ARMS Data
Aggregated ARMS data are readily available through periodic compendiums published by
USDA. In general, these publications provide financial information on a total, national basis
4-20
-------
across all farms for four aggregated sectors only: beef, dairy, hogs, poultry (which combines all
broiler, turkey, and egg laying operations). However, the underlying ARMS database compiled
during 1997, in conjunction with USD A efforts to collect the Census, contains more detailed
information. These data can be grouped to show differences among farms by factors such as
facility size and production region. Data are also available on average end-of-year farm
inventories for a particular size and regional group. The 1997 ARMS database also allows for the
breakout of data for the poultry category across the broiler, turkey, and egg laying sectors.
To depict financial conditions for selected representative farm groups, EPA requested that
USD A provide the 1997 ARMS data on a more disaggregated basis than that found in the
published data. As previously noted, USDA periodically publishes summary data from the ARMS
database and provides customized analyses of the data to the public and other government
agencies. The requested ARMS data summaries were compiled with the assistance of staff at
USDA's ERS, who performed special tabulations of the data to differentiate the financial
conditions among farms by selected facility size categories and by primary producing region for
each sector. ERS developed a methodology for identifying farms likely to be CAFOs based on
reported survey information, and developed estimates of animal units on these operations based
on reported data. Given these estimates, farms were grouped into animal unit size categories and
provided to the EPA and other government agencies.
All data provided to EPA are sufficiently aggregated to ensure the confidentiality of an
individual farming operation and to maintain the statistical representativeness of the sample data.
Initially, EPA requested these representative data for the U.S. and for each of the USDA
farm producing regions shown in Figure 4-1. However, data are not available at this level of
aggregation for each of the sectors and size groups requested. Table 4-2 shows the facility size
and production regions ultimately requested by EPA. The table also shows that the ARMS
database supports data disaggregation across three broad size categories only, roughly but not
exactly matching EPA's model CAFO groups. With some exception, data obtained by USDA fall
into three facility size groups: CAFOs with more.than 1,000 AU, CAFOs with between 300 and
1,000 AU, and CAFOs with fewer than 300 AU (Table 4-1). Even at this level of aggregation,
there are some sectors for which ERS is not able to release data for all sizes and regions requested"
because of concerns about confidentiality and statistical representativeness. Where data at one
level of aggregation are not available, EPA uses data at higher levels of aggregation, as discussed
below hi Section 4.2.3.3. '
It is because of ARMS data disclosure limitations that EPA's financial models do not
exactly match EPA's cost model across the range of regions and facility sizes represented in the
costs models. In spite these limitations, the average U.S. (published) data and the representative
level data (the disaggregated data provided by ERS and used as the basis for the representative
farm approach) are instrumental in the development of EPA's model CAFOs. EPA uses these
data, along with accompanying farm descriptive data, to construct the model CAFOs as discussed
in Section 4.2.4.
4-21
-------
4.2.3.3 ARMSData Aggregations for Model CAFOs
Generally, EPA uses the ARMS data associated with the applicable sector, size, and
region to characterize a model CAFO's financial situation. In some cases, however, USDA did
not disclose representative farm data for a particular size or region for a sector. In other cases,
EPA judged certain disclosed data as potential outliers that would likely understate the impact to
a facility.
Table 4-2 lists the available representative farm breakouts of the ARMS data. For those
sectors where representative financial data are not disclosed by USDA, EPA uses national level
data for its model CAFOs, usually specified for a particular size group. Sectors for which
representative financial data (gross and net cash income) are not disclosed include (USDA/ERS,
1999a):
• Beef operations with >800 head in SP and "All Other" regions.
• Dairy operations with >500 head in MW, MA, & SO; dairies <100 head in SO
region.
» Hog operations with between 800-2,500 head in MA region.
» Egg operations with >120,000 and between 90,000 and 120,000 birds (US, E,
SW,MW).
• Broiler operations, all facility sizes in SW region.
• Turkey operations with >40,000 (US, SO, MW .regions); operations with between
10,000 and 40,000 birds (MW region); operations with <10,000 birds (MW, SO
regions).
•
In a few cases, EPA has judged the disclosed representative level data (data for a
particular sector, region, and size) to be potential outliers, based on a determination that the data
substantially differed from the national average. In these cases, EPA substitutes the reported data
with national level values, usually specified for a sector as a whole. As a result, these data may be
more reflective of national level conditions and less reflective of representative financial
conditions characterizing model CAFOs. In most cases, EPA's decision to substitute available
representative data with national level data for some sectors results in the use of a more
conservative input value, such that estimates of farm impacts would be higher than those
estimated using the representative level data (e.g., egg laying and turkey operations in the
Midwest and some hog and beef operations).
Table 4-3 summarizes the per-animal financial data derived by EPA using the 1997 ARMS
data and describes the criteria EPA used for selecting/orriitting variables used for this analysis. As
4-22
-------
shown, in most cases data are substituted due to non-disclosure of data. In some cases data are
substituted because the ARMS data for a particular aggregated variable resulted in values with a
relative standard error (RSE) of more than 75 percent (indicated in Table 4-3 as "RSE >75%").
In general, data with a high RSE are not used. The only case where EPA uses ARMS data
despite an RSE of more than 75 percent was for turkey operations (U.S., All sizes); EPA uses
this data because limited useful data were disclosed for this sector. In the beef sector, national
data are substituted for regional data because either data are not disclosed or available data have a
RSE of more than 75 percent. National data are also used because of nondisclosure of data for
some regions in the dairy, hog and broiler sectors. National data are used for the layer and
turkeys sectors because a complete set of regional values at the different size categories are not
available; some of these data also have a high RSE (Table 4-3).
Table 4-4 also summarizes the data EPA uses, highlighting those data that are substituted
for this analysis. This table identifies the level of USD A data aggregation used for each model
CAFO and indicates, where footnoted, which data are substitutes. Any other cells where "US" or
"All" is shown indicate that substitutions are made to use available average national data or
average data across all facility size categories. In some cases, EPA uses sector level data to
depict conditions for a subsector within that commodity group.
In some cases, EPA has decided to use national average available data despite limited
available regional data because of notable differences between regions. For example, in the turkey
sector, total entity revenues average $26 per animal across all turkey operations in the Midwest
compared to $7 in the East region (Table 4-3). In the layer sector, total entity revenues average
$45 per animal across all egg operations in the Midwest compared to $13 in the East region
(Table 4-3).9 In some cases these omitted data have an RSE of more than 75 percent. For both
the turkey and egg laying sectors, EPA substitutes national level data for regional data. Use of
the national level-data for egg laying and turkey operations in the Midwest may overstate the
impact to a facility; conversely, use of national data for egg laying and turkey operations in the
East may understate the impact to a facility.
In the case of hogs, EPA justifies sizable regional per-animal differences because of
additional information on the reported percentage of animal (Table 4-3). For example, in the
Mid-Atlantic, total entity per animal at hog operations in the Midwest are estimated at $80 to
$170 per animal, depending on facility size, compared to $230 to 300 per animal in the Midwest
region. Animal ownership by hog operators is more than 60 percent at operations in the Mid-
Atlantic (Table 4-3) where overall revenues are likely lower due to lower prices paid to contract
growers who comprise the bulk of production. This is compared to under 20 percent at
operations in the Midwest where overall revenues are likely higher due to higher prices paid to
9One explanation is that operations in the East may have lower overall farm revenues due to a larger
number of contract growers compared to the Midwest where average revenues are higher and where contract
growers constitute a small share of production (as indicated by data in Table 4-3).
4-23
-------
Table 4-3. EPA-derived Per-Animal Financial Data from the 1997 ARMS Data
Sector
Beef
Dairy
Hogs
Size
AH
L
M
S
All
L
M
S
All
L
M
S
All
L
M
S
AH
L
M
S
All
L ,
M
S
All
L
M
S
All
L
M
S
All
L
M
S
All
L
M
S
Region
US
US
us
us
us
MP
MP
MP
SP
SP
SP
SP
US
us
us
us
w
w
w
w
MW
MW
MW
MW
MA
MA
MA
MA
US
US
US
US
MW
MW
MW
MW
MA
MA
MA
MA
Total
Revenue
$1,060
$862
$535
$1,074
$1,012
$854
$502
$1,097
$718
d
$461
$710
$2,573
$2,613
$2,524
$2,584
$2,672
$2,567
$2,343
$6,786
$2,584
d
$2,498
$2,620
$2,561
d
$2,866
$2,283
$363
$185
$297
$644
$377
$229
$304
$606
$174
$84
d
$383
livestock
Revenue
$503
$512
$329
$463
$513
$479
$340
$470
$440
d
$357
$398
$2,279
$2,470
$2,239
$2,172
$2,473
$2,412
$2,166
$5,540
$2,207
d
$2,250
$2,148
$2,218
d
$2,358
$2,036
$213
$141
$187
$328
$228
$186
$193
$307
$102
$47
d
$232
Total
Expenses
$916
$606
$455 •
$947
$821
$532
$421
$918
$683
d
$417
$687
$2,086
$2,178
$2,058
$2,041
$2,300
$2,166
$2,118
$6,221
$2,048
d
$2,054
$2,022
$1,965
d
$2,138
$1,793
$293
$144
$233
$533
$301
$182
$238
$487
$144
$52
d
$352
NetCash
$144
$256
$79
$127
$190
$322
$81
$178
$35
. d
$44
$23
$487
$435
$466
$543
$372.
$402
$225
$565
$535
d
$444
$598
• $595
d
$728
$490
$70
$41
. $64
. $110
$76
$47
$66
$119
$31
$31
$0
$31
%Not
Own
4.7%
3.9%
6.9%
4.2%
7.8%
d
6.9%
8.7%
4.6%
d
10.6%
2.9%
1.4%
0.1%
1.7%
2.2%
0.3%
0.0%
1.4%
0.5%
2.0%
d
0.6%
3.0%
,2.5%
d
4.5%
1.1%
27.6%
37.6%
20.-5%
20.9%
17.9%
16.8%
17.9%
19.1%
66.3%
80.3%
d
36.5%
Explantion
Substitute data (Table 4-5 and 4-6)
Substitute data (Table 4-5 and 4-6)
Data used (Table 4-5 and 4-6)
Data used (Table 4-5 and 4-6)
Data not used; RSE >75%
Data not disclosed
Data not used; RSE >75%
Data not used; RSE >75%
Substitute data (Table 4-5 and 4-6)
Data used (Table 4-5 and 4-6)
Data used (Table 4-5 and 4-6)
Data not disclosed
Data used (Table 4-5 and 4-6)
Data used (Table 4-5 and 4-6)
Data not disclosed
Data used (Table 4-5 and 4-6)
Data used (Table 4-5 and 4-6)
Data used (Table 4-5 and 4-6)
Substitute data (Table 4-5 and 4-6)
Data used (Table 4-5 and 4-6)
Data not disclosed
Data not used; RSE >75%
4-24
-------
Table 4-3. EPA-derived Per-Animal Financial Data from the 1997 ARMS Data (continued)
Sector
•roilers
Layers
Turkeys
Size
All
L
M
S
All
T,
M
S
All
L
M
S
All
L
M
S
All
T,
M
S
All
L
M
S
All
L
M
S
US
L
M
S
All
L
M
s
Region
US
US
us
us
East
East
East
East
MW
MW
MW
MW
US
us
us
us
East
East
East
East
MW
MW
MW
MW
US
US
US
us
East
East
East
East
MW
MW
MW
MW
Total
levenue
$1.9
$1.1
$1.5
$10.9
$1.5
$1.2
$1.4
$5.2
d
d
d
d
$24.6
d
d
$46.3
$13.0
d
d
$16.5 .
$45.3
d
d
$140.8
$20.1
d .
$11.2
$4.9
$5.8
d
$6.5
d
$26.3
d
d
d
Jvestock
Revenue
$0.5
$0.2
$0.2
$4.4
$0.2
$0.2
$0.1
$1.6
d
d
d
d
$17.0
d
d
$27.7
$8.9
d
d
$9.3
$33.2
- d
d
$97.7
$16.6
d
$6.5
$1.6
$2.0
d
$1.4
. d
$23.0
d
d
d
Total
Expenses
$1.4
$0.6
$0.9
$10.9
$1.0
$0.7
$0.9
$5.0
d
d
d
d
$20.6
d
d
$41.1
$11.7
d
d
$17.1
$33.9
d
d
$110.1
$18.3
d
$8.7
$3.5
$4.5
d
$3.5
d
$24.3
d
d
d
NetCash
$0.5
$0.5
$0.6
$0.0
$0.5
. $0.5
$0.6
$0.2
d
d
. d
d
$4.1
d
d
$5.1
$1.3
d
d
($0.5)
$11.4
d
d
$30.7
$1.8
d
$2.5
$1.5
$1.2-
d
$2.9
d
$2.0
d
d
d
%Not
Own
97.6%
92.2%
96.3%
99.6%
97.9%
99.5%
96.1%
98.7%
d
d
d
d
43.3%
d
d
83.5%
61.3%
d
d
85.6%
28.7%
d
d
65.3%
70.0%
d
83.1%
85.5%
99.8%
d
100.0%
d
56.9%
d
d
d
Explantion
Substitute data (Table 4-5 and 4-6)
Substitute data (Table 4-5 and 4-6)
Data not used; RSE >75%
„
Data used (Table 4-5 and 4-6)
Data used (Table 4-5 and 4-6)
Data not used; RSE >75%
Data not disclosed
Data not disclosed
Data not disclosed
Data not disclosed
Substitute data (Table 4-5 and 4-6)
Data not disclosed
Data not disclosed
Data not used; RSE >75%
Data not disclosed
Data not disclosed
Data not used; RSE >75%
Data not disclosed
Data not disclosed
Substitute data; RSE >75%
Data not disclosed
Substitute data (Table 4-5 and 4-6)
Data not used; RSE >75%
Data not disclosed
Value substituted w/ average US
Data not disclosed
Data not used; RSE >75%
Data not disclosed
Data not disclosed
Data not disclosed
Source: Derived by EPA using 1997 ARMS data. "%Not Own"= Percentage of animals not owned by the farmer.
4-25
-------
Table 4-4. ARMS Data Aggregation for Model CAFOs by Sector, Size, and Region
Sector
Fed
Cattle
Dairy
Heifer
Veal
Hog:
FF
Hog:
GF
Layer:
Wet
Layer:
Dry
Broiler
Turkey
Regions ^
MW
CE
MW
PA
MW
MW
MA
MW
MA
MW
SO
MW
SO
MA
SO
MA
MW
CAFOs
<300
AU
; "/>,-r yJ--j\
I;,: ,-.M'^
Dairy
MW
Small
•
X ,
ff •*. *•
Hog
MW
Small
"
-
-
—_
_
•v
CAFOs CAFOs
Medium l(a) Medium l(b)
300-1000AU
CAFOs
Medium 2
Beef US Medium
Beef MP Medium
Dairy MW Medium
Dairy West Medium
Beef US Medium
Beef U.S.
Medium
Hog MA All
Hog MW Medium
Hog MA All
Hog MW Medium
CAFOs
Large 1
CAFOs
Large 2
>1,000 AU
Beef US Large
Beef MP Large
Dairy MW
Large
Dairy W
Large
BeefUS
Large
r'.^rV^
i"1
r;f<:~
i**" t j, „
~ ^.f
Hog MA Large
Hog MW Large
Hog MA Large
Hog MW Large
t: ' Layer US All "
Layer US All *
Broiler US Medium
a/
Broiler Bast Medium
Turkey US Medium ^
Broiler US Large ^ '
Broiler East Large
Turkey US
All"7
r ,f-
"•>/*-->'
Source: USEPA, 2000a and USDA/ERS, 1999a. Descriptions of model CAFOs used for.EPA's CAFO Model.
Region names defined in'Tables 4-1 and 4-2. Size groups by AU are approximate (see Table 4-2).
Shaded cells indicate model CAFO not developed. Cells with "US" or "All" indicate substitutions made based on
available average national data or average data across all facility size categories. The commodity sector indicated
in the cell is, in some cases, used to depict conditions at model CAFOs for a subsector of that commodity group.
Higher level of data aggregation substituted for suspected outlier. ,
4-26
-------
independent owner-operators. Contract hog growers comprise a smaller share of production in
the .Midwest, as shown in Table 4-3.
Table 4-3 shows the reported ARMS revenue expressed both as average total entity
revenue and revenue from primary livestock sales only. The ARMS operating expenses data are
not distinguishable by business enterprise, but are total for the business (Table 4-3). EPA
recognizes that the mix of enterprises at operations and EPA's use of total entity revenue for this
analysis may affect the results of its analysis in some sectors and regions. To address this
concern, EPA conducted additional sensitivity analyses related to the use of ARMS livestock
revenue data only versus total entity revenue, which is provided in Appendix D of this report.
Although annual revenue sources are distinguishable in the ARMS data, it is not clear that this
precisely reflects an entity's different business enterprises. Producers generally make business
decisions taking into consideration all components of the business, some of which are
interdependent. For example, some livestock producers grow crops to offset feed costs in
addition to revenue considerations. EPA typically conducts analyses of regulated entities using
data for a business entity as opposed to an individual product line at an entity.
ARMS financial data are not available for the different types of beef feedlot operations
(fed cattle, veal, heifer operations), hog operations (farrow-finish and grow-finish hog
operations), and egg laying operations based on manure management system used (liquid versus
dry). For this analysis, EPA uses available ARMS data for the broad commodity sector
categories: beef sector data are used for fed cattle, heifers, and veal; hog sector data are used for
farrow-finish and grow-finish operations; and egg sector data are used for layer operations with
both wet and dry manure systems. Refer to Table 4-2 for sector, size, and region designations
shown in the model CAFO cells.
For the purpose of this analysis and because of lack of other statistically validated survey
data, EPA uses the ARMS data for beef operations to depict conditions at regulated cattle feeding
operations. Additional information on EPA's use of ARMS data for this sector is provided in
Section 8 and in the rulemaking record (Stott, 2000a; USEPA, 2000n and 2000m; ERG, 2000b;
NCSA, 1999).
4.2.4 Development of the Financial Characterization of Model CAFOs
The following sections discuss the key financial variables EPA uses to assess post-
compliance impacts (Section 4.2.4.1), calculation of the key variables on a per-animal basis
(Section 4.2.4.2), calculation of present value of cash flow (Section 4.2.4.3), use of USDA's
debt-to-asset ratios (Section 4.2.4.4), and construction of EPA's financial models used to depict
representative CAFOs (Section 4.2.4.5). Section 4.2.4.6 discusses some of the general limitations
of the farm financial data used to characterize model CAFOs for this analysis.
4-27
-------
When the individual components of the financial model are derived, EPA matches the
financial models to the cost models (described in Section 4.2.4.5). This intersection of financial
models and cost models forms the basis for the analysis of economic impacts using this
representative farm approach.
4.2.4.1 Key Financial Variables
For this analysis, EPA focuses on three financial measures that are used to assess the
affordability of the proposed CAFO regulations. These are: (1) total gross revenue; (2) net cash
income; and (3) debt-to-asset ratio. All are taken or derived from data in the 1997 ARMS data
summaries prepared by ERS and form the basis for the financial characterization of me model
CAFOs. These reported measures (at the level of aggregation shown in Table 4-4) are used
directly by EPA to represent baseline conditions at the model CAFOs.
Gross cash income is used to measure the ratio of compliance costs to sales (sales test).
EPA and others frequently have used a sales test to evaluate post-compliance impacts in previous
regulatory analyses (USEPA, 1987,1994 and 1996; DPRA, 1995; USGPO, 1999; Heimlich and
Barnard, 1995). For this analysis, EPA uses average USDA-reported gross cash income for
representative farms, defined in Section 4.2.3.1, to evaluate post-compliance costs at model
CAFOs. This measure includes total farm income from both an operation's livestock business as
well as income from other sources, such as sales of crops and other secondary livestock on site.
Other farm-related revenue and government payments are also included. USDA's farm revenue
data are available for 1997 only. To account for potential changes in farm revenue since 1997,
EPA conducted additional sensitivity analyses that are presented Appendix D.
The financial position of a farm can be calculated using either net cash income or net farm
income. Net farm income is an effective measure of long-term profitability because it accounts
for capital replacement costs and non-cash business income (i.e., land, capital, and labor services).
Net cash income reflects current or short-term cash earnings. For this analysis, EPA uses cash-
based measures, for reasons cited hi Section 4.2.5. The reported net cash income, or gross
income less operating expenses, is used as an estimate for cash flow to construct baseline per-
animal annual cash flow (Section 4.2.4.1) and baseline and post-compliance discounted cash flow
(Section 4.2.4.2).
The debt-to-asset ratio is a measure of a company's solvency and its' ability to finance
regulatory costs through additional debt. This measure is calculated by USDA as the ratio of
business debt to business assets and reflects the share of assets owed to lenders. USDA uses a
combination of a farm's net income and debt-to-asset ratio to classify the overall financial .position
of a farm based on annual earnings and solvency (USDA/ERS, 1997a and 1997e). EPA also uses
the reported total farm assets and liabilities (assets divided by liabilities equal the USDA reported
debt-to-asset ratio) to construct post-compliance debt-to-asset ratios, as discussed hi
Section 4.2.4.4.
4-28
-------
Section 4.2.5 provides additional information on .how these financial variables are
evaluated to demonstrate the economic achievability of the proposed CAFO regulations.
4.2.4.2 Calculation of Financial Variables on a Per-Animal Basis
ARMS financial data obtained from ERS include representative farm financial data and
corresponding summary information that match the reported average revenue to the total number
of farms and total number of animals in the sample set. From these data, EPA converts the
relevant financial statistics in the ARMS data to per-animal values for each model farm. Financial
data derived on a per-animal basis include gross cash income and net cash income.
Per-animal financial data are calculated by multiplying the average value of the reported
financial data per farm by the total number of farms (also obtained form the 1997 ARMS data)
and then dividing by the total number of animals, as follows:
Avg. Value ($) =
Animal
Avg. Value ($) per Farm^ * Total # Farms(
Total # Animals,
'(U)
Where:
i = animal sector
j = selected representative size/region for that sector
The resulting per-animal calculations for the key financial variables are presented in
Tables 4-5 and 4-6, EPA derives these data from the ARMS database for calendar year 1997.
Descriptive farm data on the total number of farms and the total number of animals on those farms
reflect total farm inventory during that year. These data correspond exclusively to the average
income statement and balance sheet information for representative farms compiled by ERS. In
some cases, the data shown in the table are at a higher level of aggregation than the sector,
region, and size a CAFO model is representing because of data collection or disclosure issues or
because EPA judges the data to be possible outliers, as discussed in Section 4.2.3.3.
To establish consistency with the cost model outputs, EPA scales the financial data using
the same number of animals assumed for each of EPA's cost models, whose outputs also are
expressed on a per-animal basis. This approach allows for greater accuracy by aligning the the
compliance costs incurred and revenues generated at a facility. This approach also allows for
more flexibility. Since the models are initially expressed on a per-unit basis, EPA can readily use
financial data for one CAFO model as a substitute for another simply by adjusting the number of
animals at a facility (assuming that per-unit costs and revenues between these representative
facilities may be considered similar).
4-29
-------
4-5 Pcr-Animal Total Gross Revenue for Model CAFOs, 1997
Sector
Beef
Dairy
Heifer
Veal
Hog: FF
Hog: GF
Layer: Wet
Layer: Dry
Broiler
Turkey
Region
MW
CE
MW
PA
MW
MW
MA
MW
MA
MW
SO
MW
SO
MA
SO
MA
MW
Revenue per Animal ($)
CAFOs
<300
AU
-
$2,620
*~ t
r»
1
$606
r ~
-
*"
CAFOs CAFOs
Medium l(a) Medium l(b)
300-1000AU
CAFOs
Medium 2
$535
$502
$2,498
$2,343
$535
$535
$174
$304
$174
$304
. _ "*" •ri"*" "" " ~^f~* 'tf »•**
CAFOs
Large 1
CAFOs
Large 2
>1,000 AU
$862
$854
$2,613
$2,567
$862
,~ * i^
* •*-« ** v*. ?
$£.4«- Cy "-- -v
" *" v &M *""\ f
<."* * ™ ff ^
$84
$229
$84
$229
$24.6
1 * 5 'i.* /
•V * JS^'
$24.6
$1.5
$1.4
3>ll.z
$1.1
$1.2
$20.0
- v* ° T -v
-V" . '
£JUm^(;B JL/Cll VCU AIVJ11I *J iJ-Lx^lV AJJLVUj i ^ J^^d. kjll**v*'^^* */^*ii»j i««u*wu-vw ii»k^»*»-* -^»* -.-. — —w— —.— . ^ —
level of data aggregation used for each model. Rounded to nearest dollars (beef, dairy, hogs) or nearest ten cents
(poultry). Size groups by AU are approximate (see Table 4-2).
4-30
-------
4-fi Per-Animal Net Cash Income for Model CAFOs (One Year), 1997
Sector
Beef
Dairy
Heifer
Veal
Hog: FF
Hog: GF
Layer: Wet
Layer: Dry
Broiler
Turkey
Region
MW
CE
MW
PA
MW
,MW
MA
MW
MA
MW
SO
MW
SO
MA
SO
MA
MW
Net Cash Income
CAFOs
<300
AU
,' 3
^ " t'
$598
'^«*
- -, ,
"n*~ V-
$119
„ ,£^
"*' rf
'^*V
»\ t£
« ,,
-S *
* * f ""<•
-"• ~-
^ *-%,„"
CAFOs CAFOs
Medium l(a) Medium l(b)
300-1000AU
per Animal ($)
CAFOs
Medium!
$79
$81
. $444
$225
$79
^^f --*f x,*^ • \
$31
$79
$66
$31
$66
CAFOs
Large 1
CAFOs
Large 2
>1,000 AU
$256
$322
$435
$402
$256
*i?"**^I T -^
/'
•P %^" *
s*
$31
$47
$31
$47
$4.1
$4.1
$0.6
$0.6
$2.6
$0.5
$0.5
$1.8
r*"^ ,*-
* *• '
3; Derived from USDA/ERS, 1999a. Shaded cells indicate model CAFO not developed, bee labie 4-4 tor
level of data aggregation used for each model. Rounded tonearest dollars (beef, dairy, hogs) or nearest ten cents
(poultry). Size groups by AU are approximate (see Table 4-2).
4-31
-------
In addition to providing modeling flexibility, agriculture sector analyses routinely express
costs and revenues as a unit metric, such as dollars per animal or dollars on a unit weight basis.
Expressing an increase in production costs on a per unit basis provides a quick assessment of the
potential effects of such increases on a farming operation, often without further analysis. For
example, in an analysis of the economic impacts on livestock producers from wastewater and
runoff control requirements in coastal areas, incremental costs were reported on a per-animal
basis (Heimlich and Barnard, 1995; DPRA, 1995). The range of estimated costs—$17 to $49 per
dairy cow and $2 to $5 per hog—were determined to be affordable for producers (DPRA, 1995).
These assessments agree with other studies. In evaluating a range of technology options to
regulate hog producers in North Carolina, researchers at North Carolina State University (NCSU)
have established a cutoff of $3 to $5 per marketed hog as being "economically feasible," implying
that per-head Costs in excess of that threshold might be prohibitive (NCSU, 1999). Similarly,
researchers at Cornell University surveyed milk producers in New York who indicated that they
would likely stay in business if they had to pay up to $50 per cow for environmental
improvements (Poe et al., 1999). EPA's estimates of per-animal and per-facility costs for the
proposed regulations are provided in Section 6, 7, and 8 for each of the commodity sectors.
4.2.4.3 Calculation of Present Value of Net Cash Flow
The 1997 ARMS data are limited by the fact that they represent a snapshot of conditions
in a single year and do not account for the expected variability of farm financial conditions over
time. To account for changes in an operation's cash flow post-compliance, EPA estimates the
present value of projected farm earnings, measured as a future cash flow stream hi 1997 dollars.
The present value of cash flow, also known as the discounted cash flow, or DCF, represents the
value in terms of today's dollars of a series of future receipts. EPA calculates baseline cash flow
as the present value of a 10-year stream of farm cash flow.
EPA projects future earnings over the period of the analysis (1997 to 2006) using net cash
flow values derived on a per-animal basis from the ARMS data, USDA's projections of farm level
earnings from 1997 through 2006, and other market data to translate USDA's projected per-unit
returns into per-animal returns. EPA approximates future earnings using USDA's projected
changes (expressed in dollars per unit) by applying the equivalent incremental change (expressed
in dollars per animal) for each year during the forecast period to the 1997 ARMS data for that
variable.(Table 4-6). EPA uses the resultant time series data to calculate the present value of net
cash income used for this analysis. This approach is outlined below.
Future earnings at livestock and poultry operations are projected annually by USD A. For
this analysis, EPA forecasts out future earnings from the 1997 ARMS baseline data based on
USDA's Agricultural Baseline Projections to 2009 (USDA/WAOB, 1999. and 2000). USDA's
projections are shown in Table 4-7(a) and are expressed on a per-unit basis (i.e., cash returns per
animal or per- unit output).
4-32
-------
To translate USDA-projected changes shown in Table 4-7(a) on a per-animal basis, EPA
uses available market information, such as average per-animal yields reported by USDA and/or
annual marketing cycles based on industry data. .For dairy, the 1997 average output of 16,781
pounds of milk per cow is assumed (USDA/NASS, 1999c). An average weight of 257 pounds
per finished hog is assumed, based on reported weights for "swine for slaughter" and "farrow-
finish" (NPPC, 1998), averaged according to the market share of each (USDA/APHIS, 1995b).
The average number of 255 eggs per egg laying chicken in 1997 is assumed (USDA/NASS,
1998b and 1998f). An average broiler weight of 5.5 pounds per bird is assumed, derived from the
total number of animal and pounds of production, reported in 1997 (USDA/NASS, 1999d). For
turkeys, EPA assumes an average weight of 23.8 pounds per bird derived from weight estimates
and market share information for turkey toms and hens (USEPA, 1999J; USDA/NASS, 1999d).
Equivalent per-animal values that represent USDA's projected changes are also shown in
Table 4-7(b) for the dairy, hog, and poultry sectors. No translation is needed for projections for
the beef cattle sector, since USDA already reports projected returns on a per bred cow basis.
Once USDA's projections are expressed on a per-animal basis, future earnings are
approximated by applying the incremental national average change (dollars per animal) between
each year during the forecast period to the 1997 baseline data for each representative model
CAFO. These revised cash streams over the forecast period are shown in Table 4-8.
EPA then calculates adjusted per-animal values reflecting future earnings (Table 4-8) in
terms of present value using a 7 percent discount rate. The equation EPA uses to calculate the
present value of cash flow is :
NPV =
n
I
Where:
r
n
i
= series of cash flows
= interest rate
= number of cash flow periods
= current iteration
EPA calculates the present value of the capital and annual pollution control costs for the same
10-year period with the same discount rate (7 percent) as for the cash flow analysis (Section
4.1.3).
The adjustment of the DCF analysis to represent post-compliance conditions for each
model CAFO is as follows:
Post-regulatory Status = PVCF - [PVCC x (1 - CPT)]
4-33
-------
Table 4-7(aX USDA Baseline Projections, Returns Per Unit, 1997-2006
Sector/
Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
Beef
($/cow)
-0.91
-16.80,
18.45
45.58
34.23
25.76
19.24
25.79
32.63
38.98
Dairy
($/cwt)
11.62
12.41
9,85
9.83
10.46
10.89
11.05
11.28
11.48
11.69
Hogs
($/cwt)
3.26
-5.75
-0.91
4.28
8.51
7.30
5.69
5.17
4.39
3.86
Broilers
(0/lb)
3.14
10.60
11.74
8.97
6.62
9.29
10.42
13.29
15.75
15.20
Egg Layers
(eVdoz)
8.69
10.15
5.30
1.89
6.58
2.85
-0.66
1.08
0.50
-0.14
Turkeys
(0/lb)
-3.83
-3.3
9.8
4.76
1.45
-2.13
-3r70
-1.90
-1.66
-0.07
Source: USDA/WAOB, 1999 and 2000., Values shown for 1997 are back-calculated from 1998, reported in the
2000 Baseline Projections based on the percentage change projected for 1997 to 1998, as reported in 1999
Baseline Projections. Time series data projected by USDA include: Beef (Costs and returns to cow-calf enterprise,
$/cow); Dairy (Returns above feed concentrate costs, $/cwt; Hogs (Returns above cash costs, $/cwt); Broilers (Net
returns, cents/pound); Layers (Net returns, cents/dozen); Turkeys (Net returns, cents/pound).
Table 4-7. EPA-derived Equivalent Baseline Projections, Returns Per Animal, 1997-2006
Sector/
Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
Beef ,
Dairy
Hogs
Broilers
Egg Layers
Turkeys
($/animal)
($0.9)
$(16.8)
$18.5
. $45.6
$34.2
$25.8
$19.2
$25.8
$32.6
$39.0
$1,960
$2,094
$1,662
$1,658
. $1,765
$1,837
$1,864
$1,903
$1,937
$1,972
$8.4
($14.8)
($2.3)
$11.0
$21.9
$18.8
$14.6
$13.3
$11.3
$9.9
$0.17
$0.58
$0.65
$0.49
$0.36
$0.51
$0.57
$0.73
$0.87
$0.84
$1.85
$2.16
$1.13
$0.40
$1.40
$0.61
$(0.14)
$0.23
$0.11
$(0.03)
$(0.91)
$(0.79)
$2.33
$1.13
$0.35
$(0.51)
$(0.88)
$(0.45)
$(0.40)
$(0.02)
Source: Derived by EPA from USDA/WAOB, 1999 and 2000 (Table 4-6). Per-animal equivalents are calculated
using USDA and industry information on per-animal yields and marketing turns (NPPC, 1998; USD A/APHIS,
1995b; USEPA, 1999J; USDA/NASS, 1999c, 1999d, 1998b and 1998f). Original USDA/WAOB values are used
for beef cattle, since USDA reports projected returns on a per-bred-cow basis. Rounded to nearest dollar (dairy) or
nearest ten cents (beef, hogs).
4-34
-------
ce
C
3
£
•*•»
•1
h<
4)
04
«*"«
o
.2
1
'5s
£
^
P
8
0
•8
CM
ea
o
•u
*•*
C5
oo
OS
OS
cs
co
J3
Cfl
u
"S
1
o5
•«.
P*
00
rs
H
G1
S
5
1
Ss
¥
o
u
-*"»
IK
CO
S
"8
•g
>
j
05
f-
os
OS
c
£
1
!
VO
^
O
(M
0
o
TT
0
«N!
o
o
rH
CO
69
O
s
69
CO
69
r~
CO
69
OS
VO
69
to
CO
69
I-J
$
fr
&
VO
69
O
CM
•el-
69
r-
00
CO
69
00
CO
69
_
co
69
OO
69
CS
69
•O
»— i
69
r--
69
.^j.
•3-
69
^
|
fr
P
O
T— 1
VO
69
to
C--
69
69
CM
O
69
to
69
CO
O
69
VO
69
0
69
CM
CO
r-
69
OO
OS
to
69
CO
1
•b
ca
Q
69
OO
f-
CO
69,
CO
69
VO
O
CO
69
O\
69
vo
O
CM
69
O
o
69
co
o
69
to
co
>o
69
CM
69
hJ
£
fr
C3
Q
r-
co
CM
69
i— i
O
CM
69
OO
VO
T— 1
69
O\
CM
69
CM
O
69
OS
CM
69
C--
69
, — ,
69"
oo
to
co
69
to
\^^»/ VQ .
.S3 4
'£ fe.
A C3
ID CS
CUTS
ca co
& 3
o g
CO f~-
o> o\
S,0^
"1J J?
& o H
o Js o
S *o •&
CM ID 7T
*n —3 *2
a OH c
ca 0, o
OS "*
5^* £H o>
us ca •*•'
•™-H _ ( .
"^ ^ Q>
'ca I5 "
Si S Ctf
^ ° *§
13 "S "w
^ s
II 1
ti .» 2
gp^-O
g £, u
"co C-- *°
<: "cS i
Q^ »— 4 ^X
r^ _o "-^
rT_| *™~
j^ti-vo'
•a .S3 o
g 8^
"= -° S
03 ~* OS
I'll
^r h w
O ID ID
CO CU &4
•-
-------
Where:
PVCF = present value of future model CAFO cash flow
PVCC = present value of after-tax incremental pollution control costs for the CAFO
model
CPT = cost pass-through assumed for the CAFO model (selected sectors only)
EPA's cost passthrough estimates (the percentage of costs that CAFOs are expected to be
able to pass through to higher levels of the marketing chain in the form of higher product prices)
are presented later in Section 4.2.6 of this report.
This methodology estimates the long-term reduction to earnings that result from the costs
of complying with the proposed CAFO regulations. If the post-regulatory status is less than or
equal to zero, it does not make economic sense for the owner to upgrade the farm. Under these
circumstances, the average farm represented by that model might be vulnerable to closure.
Unlike previous regulatory analyses conducted by EPA, fihis approach does not consider
salvage value of liquidated assets at a CAFO. There are a number of reason why using salvage
value in this analysis might be inappropriate. First, identifying the true market value of a facility
using the value of fixed assets is a very uncertain means of determining true salvage value. The
assets of livestock and poultry farms include land, animal housing structures, and waste disposal
operations. All of these are highly specific, immobile assets without alternative uses (Aust, 1997).
Second, the accuracy of the components of salvage value from the ARMS survey is questionable,
with some valuations based on the personal judgement of individual farmers. Also the appropriate
markdowns from market value of the value realized post-closure must be generated. Third, the
salvage value analysis ignores liquidation costs (e.g., legal fees, real estate broker fees, etc.),
which can be difficult to estimate. Finally, small family-owned farms may not base a decision of
whether to remain in business on salvage value and projected cash flow. Other less tangible
reasons for staying in business might include the satisfaction of working for oneself and the ability
to employ family members. Furthermore, in many cases, the farm is not just a business but is the
family home. Decisions to liquidate are complicated, and many farmers would try to stay in
business under nearly all adversities unless forced to close under circumstances of persistent
negative cash flows that undermine their ability to make payments on loans or to survive
financially.
4.2.4.4 USDA 's Debt-to-Asset Ratios for Model CAFOs
To model the regulatory baseline, EPA uses the debt-to-asset ratio reported by USDA for
1997 (USDA/ERS, 1999a) for each model CAFO. These ratios are identical to those calculated
4-36
-------
using the reported assets and liabilities on the same balance sheet summary. The baseline debt-to-
asset ratios EPA uses for this analysis are shown in Table 4-9.
EPA calculates the postregulatory debt-to-asset ratio as:
_ ',. . . T, .. debt + capital compliance cost
Postcomphance Debt-to-Asset Ratio = *-— s—
asset + capital compliance cost
4.2.4,5 Construction of Representative Model CAFOs
The final step in the creation of model CAFOs is the linking of the cost models to the
financial models. Using the ARMS data, EPA develops more than 40 unique financial models that
differentiate by sector, facility size, and producing region. EPA then matches these financial
models to more than 170 individual cost models that are more finely differentiated. The cost
models break out sector, facility size, and producing region, as well as land availability and facility
type.
Facility Size
EPA's cost models provide a greater number of breakouts by facility size than do the
financial models. Specifically, the cost models estimate regulatory costs for two medium and two
large facility financial models are expressed on a per-animal basis, EPA believes that this will limit
misrepresentation when the models are matched up.
As noted previously, the cost models are able to differentiate more facility sizes and
producing regions than the financial models, which are limited by the level at which ARMS data
must be aggregated to avoid disclosure of financial data. The ARMS data also cannot be used to
distinguish the differences in financial conditions associated with the amount of land available for
manure applications. Furthermore, the representative ARMS data do not reflect differences in
financial conditions associated with specific enterprises (i.e., data are for an average entity and do
not distinguish between livestock enterprise and other farm-related business) and with facility
types (e.g., the data cannot differentiate hog farrow-finish and hog grow-finish operations, while
the cost models are developed to account for potential compliance cost differences between these
two types of operations). . ,
The following discussion presents how EPA uses the available data, accounting for
limitations, to link the financial models and the costs models to create representative CAFO
models that reflect varying facility sizes, producing regions, land availability, and facility types. In
some cases EPA has identified a potential for model CAFOs to underestimate impacts due to the
4-37
-------
Table 4-9. Debt-to-Asset Ratios for Model CAFOs, 1997
Sector
Beef
Dairy
Heifer
Veal
Hog: FF
Hog: GF
Layer: Wet
Layer: Dry
Broiler
Turkey
Region
MW
CE
MW
PA
MW
MW
MA
MW
MA .
MW
SO
MW
SO
MA
SO
MA
MW
Debt-to-Asset Ratio (%)
CAFOs
<300
AU
"-
i
20%
*- •**•-*•
t -r _^SCL^_ Jft-
.-„ «
17%
"**- Z-Zl&^r ^
'
„-'-*-
_
i^
_,
__ '
CAFOs
Medium l(a)
CAFOs
Medium l(b)
CAFOs
Medium 2
300-1000AU
13%
17%
23%
23%
13%
v ~
13%
13%
25%
13%
25%
CAFOs
Large 1
>1,000 AU
• 9%
9%
26%
24%
9%
^ ' ^ -„
•<-> v •» . - •
,?
« **" *<# '
? 't^ v t
'"^V^''
»^ « A
f * ™. *x
31%
40%
31%
40%
11%
'**tp|fe'5;-,ay5?».--li
t ^4,j5»H%*5p
^.•mi^ *•" c-%s*vf-? ^^3j"
^ &%fe, ^o^«ft^x^5
6
11%
21%
19%
23%
30%
26%
15% '
x ? >
•*• * *i t 1
i.**"
Source: USDA/ERS, 1999a. Shaded cells indicate model CAFO not developed. See Table 4-4 for level of data
aggregation used for each model. Size groups by AU are approximate (see Table 4-2).
use of a limited number of financial models as estimators for conditions at operations with size
categories (Table 4-1), but financial data are available only for a single size per group (see
Table 4-2). EPA is able, however, to account for the differences in financial conditions between
Medium 1 and Medium 2 CAFOs and between Large 1 and Large 2 CAFOs using the per-animal
financial data and the average number of animals associated with the cost models developed for
these model sizes to create an estimate of financial conditions at each model CAFO. The number
of animals at an operation drives not only the magnitude of expected compliance costs, but also
directly affects financial conditions such as revenues at a CAFO. EPA can match the expected
4-38
-------
financial conditions at an operation of the same size as that assumed in estimating compliance
costs for each model CAFO. This is accomplished by multiplying the number of animals
(Table 4-1) and per-animal financial data (shown in Tables 4-5 and 4-6) for each respective model
CAFO. The result is an estimate of CAFO level revenues and net present .value of cash flow at
each model CAFO. Tables 4-10 and 4-11 show EPA's estimates of total CAFO revenues and
total net present value of cash flow for each model CAFO that is assumed in the baseline.
EPA's cost models evaluate costs for some of the largest facilities, which are often much
larger than what is reflected in ihe ARMS financial data. In part, EPA can account for this
difference, since all cost and revenue data are converted to a per-animal basis. 'Per-animal
financial data readily allow EPA to use USDA financial data aggregated over a broader facility
size grouping as an estimator for financial data at model CAFOs within that size range. For
example, EPA uses USDA data for a larger-sized farm (as shown in Table 4-2) to develop
estimated financial conditions at both the Large 1 and Large 2 model CAFOs through the use of
the per-animal conversion. Since the ARMS data are more reflective of smaller-sized operations
within each facility size group, EPA's use of these data likely overstate impacts since financial
conditions at the largest operations are likely more favorable than the reported average for that
size group.
Production Regions
EPA's cost models also provide greater breakout of farms across production regions than
EPA's financial models. Table 4-12 compares the regional breakouts for the financial and cost
components of the model CAFOs. As shown in the table, many of the same key producing states
are represented within both the comparable financial and cost model regions, and EPA is
confident that the major performance differences between the regions are captured when the
financial and cost models are linked. For example, the ARMS data show significantly lower per-
animal revenues among hog farms in the Mid-Atlantic region compared to those in the Midwest
(see Table 4-5). Higher revenues in the Midwest most likely reflect additional income from crop
sales, as well as generally higher per-animal farm prices received by independent operators
compared to contract growers. Contract growers are more common among hog operations in the
Mid-Atlantic than they are among those in the Midwest.
Land Availability
As described in Section 4.1, estimated compliance costs distinguish among three
categories of land availability: Category 1 CAFOs (sufficient cropland for all manure nutrients
generated), Category 2 CAFOs (insufficient cropland), and Category 3 CAFOs (no cropland).
However, the ARMS data EPA uses for its financial models are averages by animal sector, facility
size, and producing region only. Data are not available to represent differential land availability
among operations within a sector. EPA recognizes that there may be some financial differences
4-39
-------
Table 4-10. Total Estimated Gross Farm Revenues for Representative Model CAFOs
Sector
Beef
Dairy
Heifer
Veal
Hog: FF
Hog: GF
Layer:
Wet
Layer:
Dry
Broiler
Turkey
Reg.
MW
CE
MW
PA
MW
PA '
MW
MA
MW
MA
MW
SO
MW
SO
MA
SO
MA
MW
Annual Gross Revenue per CAFO (S)
CAFOs
<300
AU
•^^•'"•'y^fc
*"''--'• * .' •ST",",: „
$523,949
;feS8II
*?-^'iJr~f:££&
,/v,7.u^.x ''4^4 A"^
', S";-r^*;,r"ljit^|;
.v?-.'. '> i Vf^/':
$454,594
"* r
r
-
-
,"
-
- -
CAFOs
Medium l(a)
CAFOs
Medium l(b)
CAFOs
Medium 2
300-1000AU
$243,363
$228,405
$587,055
$550,552
$213,945
$200,796
$213,945
$147,601
$247,363
$168,014
$273,497
$933,528
$797,314
$54,051
$52,203
$208,440
$203,414
$264,845
,$443,673
$265,369
$432,125
„' ^
$1,294,961
$1,105,995
$75,783
$73,190
$351,545
$343,079 '
$415,589
$390,045
$1,149,129
$1,077,676
$401,148
$376,492
$288,826
$377,727
$653,962
$381,042
$645,453
$89,989
$2,400,783
$2,399,035
$108,100
$105,130
$508,119
$511,223
CAFOs
Large 1
CAFOs
Large 2
>1,000 AU
$1,617,656
$1,602,960
$3,707,777
$3,462,984
$1,292,746
$1,281,002
i 4^-k
**» ^
•w~Sf- ^
r ""SI**
V
> , ^^S?l/*5--,
» r <~ *"
^ ^ ^ -*~
f " *•
'A -.
~" >
Is
$1,431,002
$3,164,473
$743,422
$2,296,584
^St^fffffJ^SsffS
A-,^Mw^i8£^y&
°*>*V'.£-'t fe^&$x&3 «^. .
3-^^l^^t-.-
#'?:i^^;£-*K-S$%
$30,269,485
$21,777,839
$342,679
$325,349
./•-4 r* **
* ^
> H-. s ' *""
Source: Derived by USEPA from USDA/ERS, 1999a. Total farm revenues are estimated based on per-unit
revenues derived from USDA data, shown in Table 4-5, multiplied by the average number of animals shown for
each model CAFO in Table 4-1. Shaded cells indicate model CAFO not developed. Size groups by AU are
approximate (see Table 4-2).
4-40
-------
Table 4-11 Present Value of Total Net Cash Farm Income for Model CAFOs
Sector
Beef
Dairy
Heifer
Veal
Hog: ..
FF
Hog:
GF
Layer:
Wet
Layer:
Dry
Broiler
Turkey
Reg.
CE
MW
PA
MW
PA
MW
MA
MW
MA
MW
SO
MW
SO
MA
SO
MA
MW
Annual Net Cash Income per CAFO ($)
CAFOs
<300
AU
$119,612
""• ;\'?-i
^'jTiT
^;"H - ^
\? fv
— 7 ** * •£"
$88,980
^-^ .T"^
„*- „•,..,
^ :
f^ ? ' -;
» <, i. ,~
r*' ""^ «»
^r** »— , ij-
"^ ^ " f
^^ -,t_'*°>*v L
CAFOs
Medium l(a)
CAFOs
Medium l(b)
CAFOs
Medium 2
300-1000AU
$36,723 '
$104,272
$52,182
. $31,748
$32,284
$31,748
$25,921
$53,740
$29,506
$59,418
„> ***•*- -^
* & & ** -.
$153,898
$131,443
$20,974
$20,881
$47,274
$46,135
$46,512
$96,389
$46,603
$93,880
:-' ^ T.;<1
$213,483
$182,331
$29,406
$29,276
.$79,731
$77,811
$62,712
$204,107
$103,376
$59,527
$60,533
$42,860
$66,336
$142,075
$66,918
$140,226
$14,835
$395,785
$395,497
$41,946
$42,052
$115,242
$115,946
CAFOs
Large 1
CAFOs
Large 2
>1,000 AU
$604,075
$617,506
$570,069
$384,420
$482,745
f ;:r~t^
$109,340
$160,180
$110,743
$158,925
$352,806
$1,133,560
$1,191,659
. $55,335
$57,615
$171,886
$280,306
$9,655,865
_^^ ^ -v
fC^%^
4 1~ £ 'fya'
^' ^
$533,397
$642,722
$277,106
$466,449
$4,990,126
$3,590,221
$145,514
$137,912
t& £ *>>
.-v ' ^ j-'
7^' "^ '
Source: Derived from USDA/ERS, 1999a. Present value calculations of forecasted earnings are based on per-umt
cash flow (Table 4-6) and USDA projections (Table 4-7) and are used to estimate total net present value of cash
flow at each model CAFO using the average number of animals assumed for each model (Table 4-1). Shaded cells
indicate model CAFO not developed. Size groups by AU are approximate (see Table 4-2).
4-41
-------
Table 4-12. Comparison of Regional Coverage between EPA's Cost and Financial CAFO Models
Sector
Fed
Cattle,
Veal&
Heifers
Dairy
Hogs
Layers
Broiler
Turkey
Region
MW
CE
MW
PA
MA
MW
SO
MW
MA
SO
MA
MW
ARMS Regional Data Aggregation
Used for the Financial Models
Not disclosed by USDA. Average,
national data used as proxy.
Mountain/Plains (Northern Plains +
Mountain)
Midwest (Lake States + Corn Belt +
Northern Plains)
West (Pacific + Mountain)
Mid-Atlantic (Northeast +
Appalachian)
Midwest (Lake States + Corn Belt +
Northern Plains)
Full coverage not disclosed by USDA.
Average, national data used as proxy.
Full coverage not disclosed by USDA.
Average, national data used as proxy.
Not disclosed by USDA. Average,
national data used as proxy.
East (Northeast + Appalachian + Delta
+ Southeast)
Full coverage not disclosed by USDA.
Average, national data used as proxy.
Full coverage not disclosed by USDA.
Average, national data used as proxy.
USDA Regional Data Aggregation
Used for the Engineering Cost Models
Lake States + Corn Belt + Northern Plains +
Northeast + Appalachian
Pacific + Mountain + Southern Plains +
Delta + Southeast
Lake States + Corn Belt + Northern Plains +
Northeast + Appalachian
Pacific + Mountain + Southern Plains +
Delta + Southeast
Northeast + Appalachian
Lake States + Corn Belt + Northern Plains
Northeast -+ Appalachian + Delta +
Southern Plains + Pacific + Mountain
Lake States + Corn Belt + Northern Plains
Northeast + Appalachian + Delta +
Southern Plains + Pacific + Mountain
Lake States + Corn Belt + Northern Plains
Northeast + Appalachian + Delta +
Southeast
Pacific -i- Mountain + Lake States + Corn
Belt + Northern Plains + South Plains
between CAFOs with land available for land application of manure nutrients and those with little
or no land available for land application, both in terms of CAFO level assets and also expenditures
to manage animal waste. For this analysis, EPA uses data that incorporate an assumption that
asset levels are the same for a given model CAFO across all land-use categories (Table 4-9).
In general, EPA expects that compliance costs are lower at operations that grow crops in
addition to livestock, since these operations often tend to have sufficient cropland to land apply
manure nutrients as a means of waste disposal. ARMS data, however, cannot be used to
differentiate the financial conditions at operations with varying land availability. Specifically,
EPA's engineering cost models for Category 1 CAFOs and some Category 2 CAFOs assume
4-42
-------
these operations have more land for manure application and, thus, generally lower compliance
costs compared to Category 3 and some Category 2 CAFOs which EPA assumes will incur
additional compliance costs in lieu of land application. EPA's financial models are not able to
reflect the higher asset levels for Category 1 CAFOs and some Category 2 CAFOs due to data
limitations. Therefore the financial impact on these farms using the financial ratios might be
overstated. Likewise, the financial models are not able to reflect the lower asset levels for
Category 3 CAFOs and some Category 2, and impacts might be understated.
For the purpose of this analysis, EPA assumes that the available financial data for the
sector as a whole (with appropriate breakouts for size and producing region) roughly approximate
the financial conditions at all three CAFO categories. Thus each model CAFO is associated with
one financial model and three cost models representing the three land availability categories.
Facility Type
As described in Section 4.1, EPA's estimated compliance costs distinguish among the
types of facilities within a sector. However, the ARMS data available for the financial models are
averages by animal sector, facility size, and producing region only. Data are not available to
represent subsectors within a sector. For example, in the hog and poultry sectors, the ARMS data
are mixed across farms where the farmer owns the animals (e.g., independently owned and
operated farms) and farms where the farmer raises animals on behalf of a corporate entity that
typically retains ownership of the animals (e.g., contract grower). However, the financial
conditions across these two types of farms differ. Gross farm revenues generated by contract
growers tend to be lower compared to independent operation since the contract price is often
lower than the market price received by independent operators; however, the contract grower
typically faces lower production costs since the processor supplies much of the contract grower's
production inputs. In addition, a grower operation's current assets may be lower since these
operations do not own the chickens they grow, nor produce crops requiring storage (Perry et al.,
1999).
The average ARMS data also do not differentiate between different types of operations in
some sectors and between multiple enterprises within an operation. For example, ARMS data are
available across all hog operations, but profits tend to be higher at the more specialized grow-
finish operation compared to a farrow-finish operation (Yeske, 1996; USDA/ERS, 2000c). The
ARMS data span all beef operations, including cow-calf and grazing, as well as cattle feeding
operations that are company-owned and custom feedyards. Returns tend to be more favorable at
more specialized feedlot operations; among those, financial conditions may vary depending on
whether the operation is company-owned and a custom feedyard or is a contract heifer operation.
More information on the differences among facility types is provided by commodity sector in
Sections 6, 7, and 8 of this report.
4-43
-------
As a result, EPA's financial models do not distinguish among financial conditions by
facility type in the cattle sector (for which cost models differentiate fed cattle, veal, and heifer
operations), in the hog sector (for which cost models differentiate between farrow-finish and
grow-finish operations), and in the egg laying sector (for which cost models differentiate wet
manure management system arid dry manure management system operations). A more
homogenous facility type was modeled for costing purposes in the dairy, broiler, and turkey
sectors. EPA uses available ARMS financial data for the beef sector to approximate conditions
across the varying operations of cattle feedyards, veal operations, and contract heifer operations.
EPA uses financial data for the hog sector as a whole to depict conditions at both farrow-finish
and grow-finish operations. Among egg laying operations, EPA uses available financial data for
facilities with either wet or dry manure management systems.
Finally, as noted previously, EPA's analysis uses data for an average entity. The ARMS
data count among an operation's total farm income the revenue from the sales of both livestock,
crops, and other farm-related income, as well as government payments. Off-farm revenue is not
included in this analysis. Although the ARMS revenue data are roughly distinguishable by
enterprise, the ARMS cost data are not. Due to similar data limitations in previous rulemakings,
EPA routinely examines economic impacts using financial data that reflect an entity's total
revenue and costs across all enterprises at a business entity. Accordingly, EPA measures
economic achievability at the entity level in terms of potential closure of the facility and not as a
potential product line closure.
4.2.5 Criteria for Assessing Regulatory Impacts
EPA uses its model CAFOs, described in Section 4.2.4, to assess the economic impacts of
the proposed CAFO regulations across differently sized, differently managed, and geographically
distinct operations. EPA evaluates the economic achievability of the proposed regulatory options
at existing animal feeding operations based on changes in representative financial conditions
across three criteria. These criteria are: a comparison of incremental costs to total gross farm
revenue (sales test), projected post-compliance cash flow over a 10-year period, and an
assessment of an operation's debt-to-asset ratio under a post-compliance scenario. EPA evaluates
economic impacts to CAFOs in some sectors two ways—assuming that a portion of the costs may
be passed on from the CAFO to the consumer and assuming that no costs passthrough so that all
costs are absorbed by the CAFO.
EPA uses the financial criteria to divide the impacts of the proposed regulations into three
impact categories. The first category is the affordable category, which means that the regulations
have little or no financial impact on CAFO operations. The second category is the moderate
impact category, which means that the regulations will have some financial impact on operations
at the affected CAFOs, but EPA does not consider these operations to be vulnerable to closure as
a result of compliance. The third category is the financial stress category, which means that EPA
considers these operations to be vulnerable to closure post-compliance.
4-44
-------
The basis for EPA's economic achievability criteria for this rulemaking is as follows.
USDA's financial classification of U.S. farms identifies an operation with negative income and a
debt-asset ratio in excess of 40 percent as "vulnerable." An operation with positive income and a
debt-asset ratio of less than 40 percent is considered "favorable." EPA adopts this classification
scheme as part of its economic achievability criteria, using net cash flow to represent income.
This threshold and cash flow criterion is established by USDA and other land grant universities
(discussed later). The threshold values EPA uses for the cost-to-sales test (3 percent, 5 percent
and 10 percent) are those EPA has determined to be appropriate for this rulemaking and are
consistent with threshold levels used by EPA to measure impacts of regulations for other point
source dischargers (also discussed later).
'For this analysis, EPA uses all three criteria to determine economic achievability. EPA
considers the proposed regulations to be economically achievable for a representative model
CAFO if the average operation has a post-compliance sales test estimate within an acceptable
range, positive post-compliance cash flow over a 10-year period, and a post-compliance debt-to-
asset ratio not exceeding 40 percent. If the sales test shows that compliance costs are less than 3
percent of sales, or if post-compliance cash flow is positive and the post-compliance debt-to-asset
ratio does not exceed 40 percent and compliance costs are less than 5 percent of sales, EPA
considers the options to be "Affordable" for the representative CAFO group. A sales test of
greater than 5 percent but less than 10 percent of sales with positive cash flow and a debt-to-asset
ratio of less than 40 percent is considered indicative of some impact at the CAFO level, but at
levels not as severe as those indicative of financial distress or vulnerability to closure. These
impacts are labeled "Moderate" for the representative CAFO group. EPA considers both the
"Affordable" and "Moderate" impact categories to be economically achievable by the CAFO.
If (with a sales test of greater than 3 percent) post-compliance cash flow is negative or the
post-compliance debt-to-asset ratio exceeds 40 percent, or if the sales test shows costs equal to or
exceeding 10 percent of sales, EPA considers the proposed regulations to be associated with
potential financial stress for the entire representative CAFO group. In such cases, each of the
operations that are represented by that group may be vulnerable to closure. These impacts are
labeled as "Stress." EPA considers the "Stress" impact category to indicate that the proposed
requirements may not be economically achievable by the CAFO, subject to other considerations.10
Table 4-13 shows a summary of these criteria.
10Commonly used measures of "farm financial stress" include bankruptcies, foreclosures, and net exits
(Stam, et al., 1991). Indicators of stress in agriculture as reported by farm banks include delinquent loans,
discontinued financing, farm closures, liquidation, bankruptcy (Stam, et al., 1991 and 2000)
4-45 . .
-------
Table 4-13. Economic Achievability Criteria for the Proposed CAFO Regulations
Criteria
Sales test < 3%
Sales test > 3% AND negative cash flow OR debt-to-asset > 40%
Sales test < 5% AND positive cash flow AND debt-to-asset < 40%
Sales test > 5% but < 10% AND positive cash flow AND debt-to-
asset < 40%
Sales test > 10%
Economically Achievable
"Affordable"
X-
X
'
X
"Moderate"
X
X
"Stress"
X
X
EPA's choice of criteria in any economic assessment of a regulation is variable and highly
dependent on the industry being regulated. EPA recognizes that each industry has its own special
attributes and each requires an individual assessment of appropriate financial criteria. As such, the
Agency does not advocate a "one size fits all" benchmark for all industries but assesses each
industry's general conditions and uses generally accepted analytical approaches for identifying
economic impacts in that industry, if available, among other factors. Where appropriate, these
other factors include, but are not limited to, what criteria have been developed to analyze other
industries.
Federal agencies such as EPA, USD A, and others have been analyzing the impacts of
regulatory requirements on regulated communities for many years. For example, the CWA, with
its requirement to assess economic achievability, has prompted EPA since the early 1970s to
analyze the economic and financial impacts of effluent guidelines on affected industries.
Generally, EPA measures impacts using a variety of approaches that attempt to assess changes in
key financial variables post-compliance, hi many cases a benchmark is developed. This
benchmark may be based on, for example, the lowest quartile performance of firms in the industry
(e.g., USEPA, 1998b) or on an assessment of what has been generally accepted in past analyses
or by the financial community, tempered by any specifics of the industry. Usually EPA uses more
than one financial variable in an assessment since a single variable is rarely sufficient to fully
describe the relative financial health of an affected entity.
For this rulemaking, EPA has selected criteria based on those commonly used in the
agricultural sector to measure financial stress, in conjunction with criteria that EPA has used in
the past to determine the affordability of effluent guidelines that have been developed for other
industries.
The basis for EPA's economic achievability criteria for this rulemaking (Table 4-11) is as
follows. In its analyses of the financial performance of U.S. commercial farms, USDA uses a
combination of a farm's net income and debt-to-asset ratio to classify the overall financial position
4-46
-------
of a faim based on annual earnings and solvency (USDA/ERS, 1997e). Net farm income is an ^
effective measure of long-term profitability; the debt-to-asset ratio is a useful measure of a farm's
financial risk. Together these two measures provide an indicator of the farm's long-term financial
health and viability (Sommer et al., 1998). For example, if a farm earns enough income to service
debt and meet its other financial obligations, then a high debt-to-asset ratio may be acceptable,
while a farm carrying a low debt load may be able to weather periods of low or negative farm
income (Sommer et al., 1998).
USDA considers net income and debt-to-asset ratio jointly to classify farm performance
into one of the following financial positions (Sommer et al., 1998; USDA/ERS, 2000g, 1997a and
1997e):
• Favorable—Farms with positive net farm income and a debt-to-asset ratio of less
than 0.40. USDA considers these farms to be financially stable.
• Marginal Income—farms with negative net farm income and a debt-to-asset ratio
of 0.40 or less. USDA feels that periods of negative income may not pose
financial difficulties if the farm is carrying a low debt load and can either borrow
against equity or obtain income from off-farm sources.
• Marginal Solvency—Farms that generate positive net farm income, despite higher
debt service requirements. USDA states that a high debt-to-asset ratio may be
acceptable if .the farm can generate enough income to service its debt and meet
other financial obligations.
• Vulnerable—Farms with negative net farm income and a debt-to-asset ratio above
6.40. These farms do not generate sufficient income either to meet current
expenses or to reduce existing indebtedness (USDA/ERS, 1997e). USDA
considers these farms to be financially unstable and least likely to survive an
economic shock.
EPA has adopted this classification scheme as part of its economic achievability criteria,
but uses net cash income instead of net farm income in order to conduct a cash flow analysis.
USDA's debt-to-asset ratio threshold of 0.40 that defines whether a farm is highly leveraged,
which is consistent with other recommendations (Ohio State University, 1999). While a higher
ratio generally indicates financial risk, debt-to-asset ratios tend to be higher for large farms and
for those specialized in livestock feeding (Iowa State University, 1999b). For example, ratios of
0.30 to 0.40 are common among Iowa farms, although many operate with little or no debt (Iowa
State University, 1999b). Another caution when considering debt-to-asset ratios is that a high
debt load does not make farms less efficient; high efficiency farms are able to service a higher debt
load and maintain a higher debt-to-asset ratio with less risk than a low efficiency farm (Iowa State
University, 1999b). Therefore, the range of acceptable values for an operation's debt-to-asset
ratio will vary depending on income variability, the proportion of owned land (or other assets)
4-47
-------
used in the farming operation, risks associated with normal production, and fluctuations in farm
asset values that may occur due to changes in demand for agricultural assets (FFSC, 1997).
EPA also frequently uses negative post-compliance cash flow calculated over the period of
the impact analysis to identify regulated entities that are vulnerable to closure. Several economic
analyses for other effluent guidelines have used this measure as the only criterion or as part of a
group of criteria for a closure analysis (see USEPA, 2000c, 1999g, and 1998b). In these analyses,
EPA considers negative cash flow over the period of analysis as at least one indicator that a
facility might be likely to close post-compliance.
In addition to USDA's measure of debt-to-asset ratio in conjunction with net income,
EPA uses a "sales test" that compares estimated compliance costs to total revenues. For this
analysis, EPA evaluates the ratio of costs to sales using post-tax cost estimates since this more
accurately reflect the impact on a business' bottom line. Previous ELG analyses have evaluated
cost-to-sales ratios using both post-tax and pre-tax costs.11
EPA frequently uses a sales test to evaluate post-compliance impacts in previous
regulatory analyses (USEPA, 1987,1994 and 1996; DPRA, 1995; USGPO, 1999). EPA's use of
a sales test is also common practice to evaluate small business impacts for most regulatory
development. Other agencies also use a sales test, including OSHA (1999). USDA has also
considered cost-to-sales impacts to evaluate impacts to animal confinement operations (Heimlich.
and Barnard, 1995).12 m general, however, the sales test is not widely used to measure impacts in
the agricultural sector (Foster, 2000a).
The threshold values EPA uses for this analysis (cost-to-sales in excess of 3 percent, 5
percent, and 10 percent) are those determined to be appropriate for this rulemaking. EPA has
used 1 percent and 3 percent sales test benchmarks to screen for the potential for impacts in many
small business analyses (e.g., USEPA, 1997b and 1999g). These benchmarks are only screening
tools, but do support EPA's contention that a sales test of less than 3 percent generally indicates
minimal impact (Snyder, 2000). Heimlich and Barnard (1995) do not define a threshold where the
management measures would not be considered economically achievable since "...there are no
hard and fast guidelines for what is economically achievable, any appraisal of overall
achievability... is subjective" (Heimlich and Barnard, 1995).
1 JFor example, ELGs that were evaluated using post-tax costs include the Landfills, the Commercial
Hazardous Waste Combustors, and the Centralized Waste Treatment industries. Pre-tax costs were used to
evaluate the Pulp, Paper, and Paperboard, the Pesticide Formulating, Packaging and Repackaging, and the Metal
Products and Machinery industries. EPA used both pre- and post-tax costs to evaluate the Transportation
Equipment Cleaning industry.
12Heimlich and Barnard (1995) also compared costs with cash operating expenses and net farm income
(i.e., a profit-test).
4-48
-------
The 5 percent benchmark is consistent with threshold values established by EPA in
previous regulations for other point source dischargers. Generally, EPA's analyses have assumed
that sales tests less than 5 percent indicate compliance costs that are achievable (see, for example
USEPA 1987 and 1994). Other analyses have assumed the same threshold but have further
assumed that ratio values in excess of 5 percent may constitute moderate impacts, taking into
consideration other factors (USEPA 2000o, 1999o, and 1996). This analysis adopts this
framework to analyze regulatory impacts to CAFOs. In another analysis, a sales test result of
greater than 5 percent was labeled a "sales impact" (USEPA, 1987). Sales impacts were assessed
separately from impacts that could make a facility vulnerable to closure.
EPA uses an upper limit on a sales test result of 10 percent (whereby this result alone
indicates financial stress), rather than to assume that there is no upper limit on a sales test
percentage if other financial variables are also analyzed (such as in USEPA, 1996). EPA believes,
in this case, that if a sales impact is very high, this result should be considered a substantial impact
and might make an operation vulnerable to closure in spite of positive cash flow and an adequate
debt-to-asset ratio. EPA thus uses the 10 percent benchmark to ensure that potential vulnerability
will not be underestimated. In fact, relatively few operations could incur costs greater than 10
percent of revenues while continuing to show positive cash flow.
Because EPA does not use a sales test alone as a measure of financial vulnerability, use of
a sales test operates more as a screening tool. In most cases (results between 3 percent and 10
percent), a finding of financial stress is driven by cash flow and debt burden considerations. Thus,
over the key range of sales impacts, EPA's methodology is consistent with many USDA analyses,
which very frequently use net income and debt-to-asset ratios to assess impacts (e.g.,
USDA/ERS, 1997e). The exception is that EPA does not consider noncash income and
depreciation (that is, EPA uses a net cash income-type analysis rather than a net farm income
analysis). This is consistent with current views on the use of cash flow analysis in preference to
net income analysis among financial analysts (Brigham and Gapenski, 1997; Jamagin, 1996).13
Use of a sales test is one of the more common metrics used in regulatory analyses.
Another common metric is an examination of earnings before taxes as a percentage of gross
income or revenues (called a "profit test"), where the change in ratio post-compliance is used as
an indicator of the impact that compliance costs may have on profits. Profit tests or net income
analyses are frequently used by federal agencies.
Heimlich and Barnard (1995) measured economic achievabiliry by comparing estimated
regulatory costs with gross cash income (i.e., a sales-test) as well as cash operating expenses and
net farm income (i.e., a profit-test)". In recent years, however, EPA has tended to move away
13Net cash income corresponds to total farm cash revenues .minus cash expenses; i.e., it is the agricultural
term corresponding to the financial term "cash flow." Net farm income includes both non-cash income and non-
cash expenses (e.g., depreciation) and corresponds to the term "net income" in accounting.
4-49
-------
from using net income analysis and profit tests as an indicator of the financial strength of a
regulated entity for a number of reasons. First, many financial analysts are now acknowledging .
that net income analysis is a less accurate measure of financial health than a cash flow analysis,
since net income includes depreciation as a cost even though depreciation is not a cash outlay
(see, e.g., Brigham and Gapenski, 1997). Also, for valuing corporate loans, the Financial Analysis
Standards Board considers discounted cash flow the best estimator for assessing fair value for
enterprises lacking a quoted market price (Jarnagin, 1996). Second, profitability can be highly
variable since a firm has a certain amount of leeway in calculating earnings in any given year in
order to minimize tax liability. Privately held entities (which predominate in the livestock and
poultry industries, for example) have few incentives to show large profits, but every incentive to
show minimal earnings for tax purposes. Third, if a large segment of an industry is showing
negative net income, it is difficult to assess the impact of a regulation on profitability.
Typically, EPA considers that if a regulated entity is not profitable before pollution control
investments are made, the entity "may not claim that substantial impacts would occur due to
compliance...." (USEPA, 1995c). EPA has, in some cases, used the concept of "baseline closure"
if only a few entities have negative net income and cannot be analyzed (see, for example, USEPA,
1997b). These entities may not necessarily close prior to implementation of a rule, yet they
cannot be analyzed within a profit test or net income analysis framework.
The problem with analyses that are based on profitability is especially an issue in the
agricultural sector. Heimlich and Barnard (1995) point out that many farms "may be motivated
by noneconomic considerations and should be considered hobby or recreational activities, rather
than businesses, particularly when net farm income is negative." They further contend that the
majority of farm operators reporting negative net income have nonfarm sources of income and
that they may be using the farm losses to offset off-farm income to reduce income tax liability.
They also find, as does EPA, that when net farm income is negative, costs as percentages of net
income are difficult to interpret.
EPA, therefore, considers a sales test to be a more analytically useful tool than a profit test
for assessing impacts in the livestock and poultry industries for three major reasons. First, EPA
has concerns that profit-based measures may overstate vuhierabihty. Second, revenues are
generally not as sensitive to incentives to show minimum values for tax purposes as profits and
thus are not as likely as profits to be understated. Third, sales are never negative and thus a
comparison between costs and sales can be adequately interpreted.
4.2.6 Cost Passthrough
EPA generally measures the economic impacts of the proposed CAFO regulations on the
basis of the estimated compliance costs incurred at the CAFO (discussed in Section 4.1.3). Even
when post-tax costs are considered, however, there may be other mitigating factors that influence
what costs the CAFO ultimately incurs. For the purpose of this analysis, EPA assumes that
4-50
-------
producers in the livestock and poultry sectors will be able to pass on some portion of costs
through the market levels. Passthrough of compliance costs will occur either in the long-run
through market adjustment (i.e., higher prices through changes in supply and demand) or in the
shorter term as processors take steps to ensure a steady and continued supply of raw farm input
(i.e., raise production prices for live animals or animal products produced by CAFOs).
Individual farms may be considered competitive and may not individually be able to raise
prices. Collectively, however, if production costs rise across the industry as a whole (as it would
under the proposed CAFO regulations), economic theory indicates that prices will rise except in
highly unusual circumstances. Prices will rise unless demand for a product is perfectly elastic
(that is, consumers, including processors or packers, are not willing to pay more for a product
even in a scarcity situation, an unusual scenario) or unless supply of a product is perfectly inelastic
(no matter what the price, the producer would not change the quantity supplied, another unusual
situation). In the real world, even nearly perfectly elastic demand or nearly perfectly inelastic
supply at the sector or industry level for most products is rare." For this analysis, cost passthrough
due to market adjustment assumes that the demand for agricultural commodities— including
meat, milk and dairy products, and eggs—is generally considered price inelastic (i.e., food is a
necessity and demand will not decrease at a rate proportional to an increase in price).
Whether the price increase is small or large depends on the relative elasticities of supply
and demand. When demand is more elastic than supply (which means consumer demand has more
effect on price than supplier production), prices tend to rise less than when demand is less elastic
than supply. The relative elasticities of supply and demand and their role in computing the extent
to which prices would be likely to rise are discussed in more detail hi Section 4.2.6.1.
When farmers can raise their prices, the increased costs of compliance are covered in some
part by these price increases, hi such cases, compliance costs thus are effectively "passed
through" to the next economic level (e.g., processor or packer). EPA has undertaken a number of
approaches for determining an appropriate assumption for cost passthrough (CPT). The
following sections discuss the ways in which EPA computed a point estimate for CPT and also
document the decisions EPA made to construct several scenarios that can be used to bound the
estimates of CPT and thus economic impacts at the farm (and, as discussed hi Section 4.3, the
processor) level.
Actual cost passthrough from the farm to the processing sectors is difficult to predict.
Some potential scenarios include the following. First, an affiliated processing firm (integrator)
may raise the contract price paid to its growers, thus offsetting increased production costs due to
compliance. Second, an integrator may pay for all or a portion of on-farm disposal costs as part
of the production inputs typically supplied to contract growers. Alternatively, an integrator may
pay for all or some portion of off-site manure disposal costs, such as transportation and disposal
at a centralized waste treatment facility. Finally, an integrator may offset costs by funding
research or market development of alternative uses of manure by-products, such as pelletization,
etc., that will benefit producers by offsetting costs. Some states have proposed to mandate shared
4-51
-------
responsibility for manure and CAFO waste, including Kentucky (Associated Press, 2000) and
Maryland (Huslin, 2000a). Other examples of how processors are sharing the cost of
environmental regulations with the agricultural sector is available in the rulemaking record (see,
for example: Huslin, 2000b; Montgomery, 2000; Goodman, 1999).
EPA believes that the assumptions of cost passthrough are appropriate for this analysis,
particularly for the pork and poultry sectors. As discussed in Section 2, EPA expects that meat
packing plants and slaughtering facilities in the pork and poultry industries may be affected by the
proposed co-permitting requirements in the proposed CAFO regulations. Given the efficiency of
integration and closer producer-processor linkages, the processor has an incentive to ensure a
continued production by contract growers. EPA expects that these operations will be able to pass
on a portion of all incurred compliance costs and will, thus, more easily absorb the costs
associated with the proposed CAFO regulations. This passthrough may be achieved either
through higher contract prices or through processor-subsidized centralized off-site or on-site
waste treatment and/or development of marketable uses for manure.
EPA recognizes, however, that some industry representatives do not support assumptions
of cost passthrough from contract producers to integrators, as also noted by many small entity
representatives during the SBREFA outreach process as well as by members of the Small
Business Advisory Review (SBAR) Panel. These commenters have noted that integrators have a
bargaining advantage in negotiating contracts, which may ultimately allow them to force
producers to incur all compliance costs as well as allow them to pass any additional costs down to
growers that may be incurred by the processing firm.
To examine this issue, EPA has conducted an extensive review of the agricultural
literature on market power and price transmission in each of the livestock and poultry sectors and _
concluded that there is little evidence to suggest that increased production costs would be
prevented from being passed on through the market levels. EPA believes that this literature
generally supports a determination that agricultural producers in the livestock and poultry sectors
will be able to pass on compliance costs. A summary of this research is provided in the
rulemaking record (ERG, 2000c—DCN 70640). As is discussed in this literature summary, while
there is a potential for market power in each of the animal products industries, and while vertical
integration has squeezed the rents out of many farming activities, the general conclusion in the
literature is that there is little evidence mat these factors will prevent increased production costs
from being passed through the marketing chain of these industries.
Given the uncertainty of whether costs will be passed on, EPA presents the results of this
analysis assuming some degree of cost passthrough and also no cost passthrough (i.e., the highest
level of impacts projected). EPA requests comment on its cost passthrough assumptions.
Although EPA does consider the results of both of these analyses in making its determination of
economic achievability, EPA's overall conclusions do not rely on assumptions of cost
passthrough.
4-52
-------
4.2.6.1 Methodology for Computing Cost Passthrough
The first step in computing CPT is to identify appropriate demand and supply elasticities
for the products in question. Elasticities measure how markets respond to changes in price. In
this case, supply elasticities measure how, in theory, farmers respond to changes in the price they
receive: generally, rising farm prices will encourage producers to expand production, whereas
falling prices will result in production cutbacks. Demand elasticities, in this case, measure the
price response at the processor and consumer levels. Generally, an increase in the price paid for a
good—whether it is raw farm product purchased by processors or is retail foods purchased by
consumers—usually results in a decrease in aggregate demand. A decrease in prices will generally
boost the amount demanded and thus raise the amount purchased of that good.
Economists use elasticities to explain market behavior and also to predict price and
quantity changes in markets. For the purpose of this analysis, EPA uses estimates of elasticities of
supply and demand obtained through an extensive search of the agricultural economics literature
and consultation with leading experts in the field. Much of this research is conducted by the
various land grant universities and is published in the leading academic journals. Before selecting
an estimate for use in the analysis, EPA compiled a range (i.e., the lowest estimate and highest
estimate) of these published supply and demand elasticities to estimate a CPT range for each
sector. Since cost passthrough usually reflects longer term market adjustment, elasticities that are
specified in the long run are the most appropriate for this analysis. In particular, estimates of
supply elasticities are highly dependent on time frame. Generally, the longer the time frame, the
more elastic is supply because farms have time to change, either expanding or contracting their
operations. In the short-term, however, farms have less flexibility. The supply elasticities
identified in the literature, however, include short-, intermediate- and long-run estimates. The
demand elasticities identified generally do not specify a time period.
The results of this literature search are summarized in Table 4-14. The low and high
values shown in the table correspond to the range of values found in the literature (see Appendix
C for a complete listing of studies and values found). EPA uses these values to compute the low
and high values estimated for CPT in Table 4-14. The low supply elasticity estimates in
Table 4-14 are generally short-run and result in lower CPT estimates, For comparison, EPA also
uses the high supply elasticity values, which tend to reflect long-run conditions of highly elastic
supply.14 In those cases, EPA estimates CPT to be almost complete (i.e., approaches 100
percent). The "selected" elasticity values represent a consensus of expert opinion on a reasonable
estimate of supply and demand elasticities for each sector (Vukina, 2000, and Foster, 2000a) and
are considered to reflect long-run conditions. EPA uses these "selected" values to compute the
"selected" CPT values, shown in Table 4-14, as discussed below.
I4In discussions with USD A, ERS staff have indicated that they generally assume that supply is less than
perfectly elastic in the short-run and almost perfectly elastic in the long run (Hahn, 1999).
. 4-53
-------
Table 4-14. Estimated CPT Based on Elasticity Estimates Identified in Recent Literature Searches
Sector
Beef
Dairy
Hogs
Broiler"
Layer
Turkey
Range of Estimated
Price Elasticity of Supply ^
Low
Value
-0.170
-0.322
0.007
0.064
0.031
0.210
Selected
Value b/
1.020
1.527
0.628
0.200
0.942
0.200
High
Value
3.240
6.690
0.628
0.587
0.942
0.518
Range of Estimated
Price Elasticity of Demand
Low
Value
-2.590
-0.650
-1.234
-1.250
-0.780
-0.680
Selected
Value b/
-0.621
-0.247
-0.728
-0.372
-0.110
-0.535
High
Value
-0.150
-0.050
-0.070
-0.104
-0.022
-0.372
Estimated CPT "
Low
Value
NE
NE
1%
5%
4%
24%
Selected
Value
62%
86%
46%
35%
90%
27%
High
Value
96%
99%
90%
85%
98%
58%
Sources: Various, see Appendix C.
"'Estimated elasticities as identified in Tables C-l through C-12, Appendix C.
^Elasticities representing a consensus of expert opinion (Vukina, 2000, and Foster, 2000a).
'Values for supply elasticities less than zero not estimated (NE).
Includes elasticity estimates for both broilers and chickens because studies vary between the two terms when
analyzing the markets for meat from chickens.
EPA employs a simple method for calculating estimated CPT percentages using supply
and demand elasticities. The price elasticity of supply is divided by the difference of the price
elasticity of supply and the price elasticity of demand for each sector, as shown below:
CPT =
price elasticity of supply
price elasticity of supply - price elasticity of demand
The resulting "selected" CPT estimate for each sector, shown in Table 4-14, is the "Partial CPT"
estimate that provides the basis for EPA's CPT analysis.15 This approach is consistent with that
used by EPA in past regulatory analyses (see, for example: USEPA, 2000c, 1999g, and 1997a).
I5The CPT calculated in this manner is an average over all farms and may not reflect the actual CPT of
any one farm.
4-54
-------
Given the wide range of supply and demand elasticities in the literature, the resulting CPT
estimates cover a wide range (Table 4-14). As a conservative measure, EPA selects elasticities
that represent a consensus of expert opinion for a reasonable estimate of elasticities for its
analysis. These "selected" sets of elasticities result in CPT of 27 percent to 90 percent.
4.2.6.2 Three CPT Scenarios
To fully address the entire range of possibilities with regard to cost passthrough, EPA
evaluates impacts of the proposed CAFO regulations under three CPT scenarios. Two of these
scenarios are bounding scenarios that define the limits of worst-case and best-case from both the
farm and market perspectives. In the two scenarios, EPA assumes either that all costs at the
CAFO level can be passed through (100% CPT) or that none of the costs at the CAFO level can
be passed through (Zero CPT). The 100% CPT scenario leads to worst-case impacts at the
market level, but is associated with no impacts at the CAFO level. Therefore, CAFO level results
for this scenario are not reported in subsequent sections, since there are no impacts at the CAFO
level when 100% CPT is assumed. Alternatively, the Zero CPT scenario leads to worst-case
impacts at the CAFO level, but is associated with no impacts at the market level. Under this
scenario, all costs are assumed to be absorbed at the CAFO and no costs are passed on to
consumers.
( ' ~ '
To provide a more reasonable estimate of likely impacts at both the CAFO and other
levels in the marketing chain, EPA uses the point estimates of CPT values discussed in
Section 4.2.6.1. This scenario is the Partial CPT scenario, shown in Table 4-14 as the "selected
value" CPT estimate. These CPT values result in impacts that fall between the numbers seen
under the 100% and Zero CPT scenarios. EPA believes that the results obtained under the Partial
CPT scenario are likely to be more realistic than those obtained under the two bounding scenarios
in most cases.
4.2.7 Potential Cost Offsets
Available cost-sharing and technical assistance, as well as manure sales, particularly for
more valuable poultry litter, may provide potential offsets to compliance costs incurred by CAFOs
under the proposed CAFO regulations. As a conservative measure, EPA does not to consider
such offsets as part of its analysis. As a result, the impacts of the economic impact analysis are
likely overstated. -
EPA has investigated the potential for compliance costs to be offset by cost-share and
technical assistance from various federal and state conservation programs (ERG, 2000a—DCN
70130). In these programs, cost-sharing dollars are provided for animal waste management
practices that are included in the regulation. However, certain eligibility requirements may limit
program availability. Eligibility criteria can include size of operation, location in a geographic
4-55
-------
priority area, and high pollution potential. Because it is not certain which operations could take
advantage of cost-share assistance or to what degree, EPA determined that cost-share assistance
could not be reliably used as a cost offset in the impact analysis and thus did not incorporate any
assumptions of cost share assistance into the economic impact methodology.16
Generally, EQIP funds are not available to operations with more than 1,000 AU; however,
applicability based on animal units is not straightforward because USDA*s AU definitions differ
for some sectors compared to EPA's regulatory thresholds (Featherston and Atwood, 1999).
Differing AU definitions may allow some larger-sized facilities to access to EQEP funds for
environmental improvements. For example, broiler and egg operations with more than 100,000
birds are defined by EPA to have more than 1,000 AUs; the EQIP program covers operations
with up to 250,000 layers and up to 455,000 broilers (USDA's definition of 1,000 AU for these
sectors). Because of this AU definition discrepancy, a large proportion of poultry operations will
actually be eligible for EQIP funding for waste storage and treatment (ERG, 2000a). EQIP
funding would be available to eligible operations with fewer than 1,000 AU. However, EPA does
not incorporate estimates of cost-sharing because program funds from these programs are subject
to funding limitations, and current allocations may not be able to cover all new applicants that
may be affected by the proposed regulation (ERG, 2000a; Featherston and Atwood, 1999).
EPA has also evaluated the potential for some operations, particularly poultry operations,
to offset costs based on poultry litter sales. This analysis is conducted based on the typically
higher value of dry poultry litter. A summary of this analysis is provided in Section 6 of this
report but the results are not incorporated into the main impact analysis. EPA did not estimate
the value of manure sold for the wetter manures that are common in the hog and dairy sectors.
4.3 PROCESSOR LEVEL ANALYSIS
As discussed in Section 2.4 of this report, EPA estimates that 94 meat packing plants that
slaughter hogs and 270 poultry processing facilities may be subject to the proposed co-permitting
requirements. This section presents an overview of the modeling framework (Section 4.3.1) and
the data used by EPA (Section 4.3.2) to assess potential national level aggregate costs to the
processing sectors in these industries. EPA does not evaluate the potential magnitude of costs to
egg and turkey processors because the compliance costs to CAFOs in these industries is projected
to be easily absorbed by CAFOs, as presented in Section 5. EPA expects that no meat packing or
processing facilities in the cattle and dairy sectors will be subject to the proposed co-permitting
requirements, for reasons outlined in Section 2.
'^Previous research has included cost-share program dollars as an offset to compliance costs associated
with environmental regulation (DPRA, 1995; Heimlich and Barnard, 1995).
4-56
-------
4.3.1 Overview of Methodology
EPA did not precisely estimate the costs and impacts that would accrue to individual co-
permittees. Information on contractual relationships between contract growers and processing
firms is proprietary and EPA does not have the necessary market information and data to conduct
such an analysis. Market information is not available on the number and location of firms that
contract out the raising of animals to CAFOs and the number and location of contract growers,
and the share of production, that raise animals under a production contract. EPA also does not
have data on the exact terms of the contractual agreements between processors and CAFOs to
assess when a processor would be subject to the proposed co-permitting requirements, nor does
EPA have financial data for processing firms or contract growers that utilize production contracts.
EPA, however, believes that the framework EPA uses to estimate costs to GAFO does
provide a means to evaluate the possible upper bound of costs that could accrue to processing
facilities in those industries where production contracts are more widely utilized and where EPA
believes the proposed co-permitting requirements may affect processors. EPA's CAFO level
analysis examines the potential share of costs that may be passed on from the CAFO, based on
market information for each sector. Assuming that a share of the costs that accrue to the CAFO
are eventually borne by processors, EPA is proposing that this amount approximates the
magnitude of the costs that may be incurred by processing firms in those industries that may be
affected by the proposed co-permitting requirements. To assess the impact of the regulations on
processors, EPA compares the passed through compliance costs to both aggregate processor
costs of production and to revenues (a sales test), using .cost and revenue data described in
Section 4.3.2.
This approach does not assume any addition to the total costs of the rule as a result of co-
permitting, yet it does not assume that there will be a cost savings to contract growers as result of
a.contractual arrangement with a processing firm. This approach merely attempts to quantify the
potential magnitude of costs that could accrue to processors that may be affected by the co-
permitting requirements. Due to lack of information and data, EPA does not analyze the effect of
relative market power between the contract grower and the integrator on the distribution of costs,
nor the potential for additional costs to be imposed by the integrator's need to take steps to
protect itself against liability and perhaps to indemnify itself against such liability through its
production contracts. EPA also does not specifically analyze the environmental effects of co-
permitting.
EPA has conducted an extensive review of the agricultural literature on market power in
each of the livestock and poultry sectors and concluded that there is little evidence to suggest that
increased production costs would be prevented from being passed on through the market levels.
This information is provided in the docket (ERG, 2000c—DCN 70640). However, as discussed
in Section 4.2.6, EPA recognizes that some industry representatives do not support assumptions
of cost passthrough from contract producers to integrators and requests comments on its cost
4-57
-------
passthrough assumptions in general and as they relate to the analysis of processor level impacts
under the proposed co-permitting requirements.
EPA's processor analysis does not specifically account for the few large corporate
operations that are vertically integrated, to the extent that the corporation owns and operates all
aspects of the operation from animal production to final consumer product. These operations are
covered by EPA's CAFO analysis to the extent that they are captured by USDA's farm survey
and are included among EPA's model CAFOs. While the ARMS data may include data on
CAFOs that are owned by corporate operations, these data cannot be broken out to create a
model specifically designed to represent these operations. Since EPA's analysis uses farm
financial data and not corporate data, this analysis does not reflect the ability of corporations to
absorb compliance costs that may be incurred at CAFOs that are owned by a higher corporate
entity. EPA expects that its analysis overestimates me impact to corporate entities since revenues
of corporate entities are, in most cases, no less and likely exceed those at independently owned
and operated CAFOs.
*
EPA believes, therefore, that impacts on corporate operations that manage CAFOs would
be minimal. Impacts at the CAFO level in these operations would generally not be felt at the
higher corporate level because the farms in such a vertically integrated corporate structure are
typically operated as cost centers. Because each farm owned by the corporate entity is a vital part
of the whole structure of the operation, the corporation does not necessarily expect profitability at
the farm level; the corporation tends to look at the whole operation and judge financial conditions
and make financial decisions at higher levels of the organization. As long as profitability at the
last stage of processing appears adequate, there is no point to closing a farm or farms because of
increased costs or lack of profitability at the farm production level.
At the corporate level, these operations have the advantage of size as well. Large size
generally provides a corporation with many more financial and personnel resources for dealing
with regulatory costs and minimi zing the impacts of those costs. Furthermore, large corporations
can often benefit from economies of scale, allowing for smaller per-unit costs of construction or
equipment than might be the case in smaller operations. For these reasons, EPA believes that
these large corporations will have the resources to comply with the proposed CAFO regulations
without undue financial strain.
4.3.2 Sources of Data
EPA evaluates processor level impacts using deh'vered costs and revenues from the 1997
Census of Manufactures for the pork and poultry processing industries (USDC, 1999a). .
Deh'vered costs reflect the raw materials, parts, scraps, and supplies consumed or put into
production by processing companies, but do not include the cost of fuels consumed, electricity
purchased, or work performed under contract. Revenues reflect values of shipments at the sector
level. Deh'vered costs and revenues are estimated by industry and classified by NAICS material
code for major raw materials consumed.
4-58
-------
NAICS codes used for the hog processing industry are 311611 (Animal, except poultry,
slaughtering) and 311612 (Meat processed from carcasses). The NAICS codes used for Hie
poultry processing industry are 311615 (Poultry processing) and 311999 (All Other Misc. Food
Manufacturing). NAICS material codes were used to break poultry processing into the broiler,
egg, and turkey sectors. Table 4-15 presents the 1997 cost and revenue data for the hog and
broiler processing sectors EPA uses for this analysis.
======
NAICS
Material.
Code"'
11221003
31161113
11221003
31161113
Total
Materials Consumed
Value of
Shipments
($ Millions)
Hog Processing
Hogs slaughtered (NAICS Code 311611, Animal
(Except Poultry) Slaughtering)
Fresh and frozen pork (NAICS Code 311611,
Animal (Except Poultry) Slaughtering)
Hogs slaughtered (NAICS Code 311612, Meat
Processed from Carcasses)
Fresh and frozen pork (NAICS Code 311612, Meat
Processed from Carcasses)
-
--'
—
—
$38,510.9
Broiler Processing
11232001
11232003
Total
Young chickens slaughtered
Hens and other chickens slaughtered
—
—
$17,656.9
Delivered Cost
($ Millions)
$10,607.2
$1,146.9
$29.5
$3',928.5
$15,712.2
$8,946.7
$188.3
$9,135.0
=====!]
Source: Derived from USDC, 1999a.
"'NAICS Material Code denotes the industry group that produced the raw. materials consumed by the hog and
poultry processing industries.
4.4 MARKET LEVEL ANALYSIS
EPA's CAFO and processor analyses measure the effects of the proposed regulations on
CAFOs and the manufacturing sectors. Asi these effects influence the decisions of farmers,
processors and packers, and ultimately consumers, they translate into changes in the price and
quantity of farm commodities and retail foods and generate changes throughout the national
economy. To better understand the potential impacts, EPA conducts a market level analysis.
4-59
-------
EPA's market analysis evaluates the effects of the proposed regulations on national
markets. The analysis uses a linear partial equilibrium model adapted from the COSTBEN model
developed by researchers at USD A. The modified EPA model provides a means to conduct a
long-run static analysis to measure the market effects of the proposed regulations in terms of
predicted changes in farm and retail prices and product quantities. Once price and quantity
changes are predicted by the model, EPA uses input-out multipliers that relate changes in sales to
other national level market changes.
Measured impacts include changes in price and available quantities as well as changes in
national employment and economic output. Other market changes examined by EPA include
changes in regional employment and changes in U.S. livestock and poultry trade (imports and
exports). This section presents an overview of the model (Section 4.4.1) and the data sources for
the analysis (Section 4.4.2).
4.4.1 Overview of Methodology
The market model predicts and quantifies the broader market changes that may result from
regulatory compliance costs imposed on CAFOs. The mechanisms that produce these effects are
relatively simple. Compliance costs increase farmers' costs of production. The supply function
represents the amount of a product a producer is willing to supply at a given price. When
producers' costs go up the supply function shifts up, indicating producers require a higher price to
supply a given quantity. A new market equilibrium is reached when supply eventually equals
demand. EPA estimates the impact of the regulations by how that shift in the supply function
changes the overall market equilibrium.
Individual farmers generally have a limited ability to pass on increased costs associated
with regulations because of the competitive nature of livestock and poultry production and the
dynamics of the food marketing system. The marketing system is often characterized by local
oligopsony conditions, or "few buyers" (Rogers and Sexton, 1994). Since farmers may be
considered "price-takers," farms with unusually high costs or those requiring major investment to
meet the regulatory standards may be forced to drop out of the industry. The decline in the
number of farms and their lost production is what shifts the supply function. To the extent that
lower cost producers step in to fill the void as other firms exit, the supply function will shift less
than might otherwise be expected.
An economy is a tightly woven web of interactions at many levels. A change hi any one
sector (a direct impact) results in changes in others (indirect impacts). The supply shifts in farm
production that are predicted by EPA's market model would cause manufacturers to alter their
production patterns. As some industries increase production and others decrease, employees are
hired and fired, in turn changing their income and spending patterns (induced impacts).
4-60
-------
Changes in prices, trade, and social welfare are measured as differences between the
baseline equilibrium before the regulation—preregulation—and the shocked equilibrium after the
regulation has been implemented—postregulation. As discussed in the following sections, EPA's
market model provides a means to measure the direct impact of a policy change on affected
markets. Indirect and induced impacts can also be estimated by applying the results of the.market
model to an input-output model of the national economy.
4.4.1.1 Market Model
EPA's market model is adapted from the COSTBEN model developed by the USDA's
Economic Research Service (Hahn, 1996). EPA's model is designed to estimate the results of the
policy once it is fully implemented, so it does not include the additional short-run results that
COSTBEN calculates. Long-run static analysis is appropriate for measuring the final market
effects of the proposed CAFO regulations because the compliance costs being considered are
long-run costs that include both annualized capital investments and ongoing maintenance costs.
The EPA model assesses only the final outcome and does not address the steps taken to reach that
outcome. Appendix B provides a more detailed discussion of the market model's structure and
operation.
The market model analysis uses a simple, linear partial equilibrium model to predict the
effects of the proposed CAFO regulations on national markets. It measures conditions at two
market stages: farm and retail. These markets are interconnected. The supply of farm products
influences the supply of retail products, and the demand for retail products results in a derived
demand for farm products. The model puts farm level supply and retail demand functions into the
same units and sets them equal to estimate a long-run equilibrium. EPA models the effect of the
estimated compliance costs as an additive shift in the domestic supply function, and their impacts
are measured as the change from the baseline equilibrium to the new post-regulatory equilibrium.
The model finds the equilibrium price and quantity in the two markets and calculates the related
production, imports, exports, and economic impacts.
The model assumes perfect competition in all markets and is similar to models typically
found in agricultural economics literature (Kohls and Uhl, 1998; Pearce and Turner, 1990; Tomek
and Robinson, 1972).17 EPA defines each supply and demand function by its elasticity and the
baseline price and quantity values. Each animal sector is modeled separately, so interaction
effects between products are not included. For example, substitution of pork for beef when beef
prices rise is not included in the model. The implications of this limitation are unclear, as prices in
several sectors would be affected at the same time.
17,
7Although EPA recognizes the possibility of oligopsony in certain sectors, data are insufficient to develop
a more.complex model of market behavior.
" 4-61
-------
EPA estimates market level changes in terms of changes in consumer and farm level prices
for the selected products and also changes in the amount of commodities produced and traded
internationally (expressed as changes in U.S. product exports and imports).
4.4.1.2 Input-Output Analysis
The market model assesses the estimated direct impacts associated with complying with
the proposed regulations, measured in terms of dollars of industry output per year. EPA uses
these model results in an input-output analysis framework to estimate the affect of the proposed
CAFO regulations on national total employment and regional agricultural employment (farm and
processors) and also national economic output.
Changes in economic output are measured in terms of changes in Gross Domestic Product
(GDP). Changes in employment are measured in terms of full-time equivalents. Predicted
changes in aggregate employment are measured in terms of both direct and indirect/induced
employment. Direct employment measures the number of jobs related to the production and
processing including workers engaged in the manufacture of agricultural inputs and their supplies.
Other employment provides a broader measure of industry-related employment and includes
workers throughout the economy that provide support to the industry. Indirect employment
covers veterinary service providers, feed suppliers, agricultural supplies and farm services, and
trucking and transportation industries. Induced employment covers other local goods and
services, such as bank tellers, grocery store clerks, restaurant employees and gas station
attendants. EPA's analysis does not adjust for offsetting increases in other parts of the economy
and other sector employment that may be stimulated as a result of the proposed regulations, such
as the construction and farm services sectors.
Input-output analysis uses multipliers that forecast how much more or less output the
whole economy would produce as a result of each dollar increase or decrease in spending by a
given industry. Once the change in output (price times quantity) is estimated using EPA's market
model, EPA evaluates these changes using input-output" multipliers to estimate the ripple effects
as changes in one industry pass through its suppliers and the rest of the economy. Multiplying the
original change by the multiplier gives a measure of the total direct effects (on immediate
suppliers), indirect effects (on the suppliers' suppliers and all other industries), and induced effects
(on households' spending and labor decisions). These estimates may be driven by an original
change in final demand, output, earnings, or employment and yield results in terms of final
demand, output, earnings, tax revenue, or employment changes.
Although the application of multipliers is as simple as multiplication, the multipliers
themselves embody a great deal of information. Basically,, they are a synopsis of all the
interactions of an industry with the rest of the economy. The change in spending by each industry
affected by the proposed CAFO regulations is multiplied by a multiplier unique to that industry.
Multipliers for the processing industries are adjusted to avoid double counting the effects of
4-62
-------
changes in those industries with changes in the agricultural sectors that supply them. Consumer
spending is treated as separate industry. When consumers must adjust their spending patterns to
pay higher meat prices, they have less money to spend on other things. Because the market model
predicts that much of the cost of the proposed CAFO regulations would be shifted to consumers,
this shift in consumer spending is often a significant portion of the economy-wide impact of a
regulation.
4.4.2 Sources of Data and Parameters
The market model requires specification of only a small number of parameters. A base
year provides the starting conditions for the model. EPA uses 1997 as the base year, which is
consistent with the engineering cost estimates that reflect 1997 conditions. Detailed citations for
each base year value appear in Tables 4-16 and 4-17 and Appendix B.
4.4.2.1 Market Model Data
Data and parameter inputs to the market model include prices and quantities of supply and
demand, as well as various elasticities. All the data used for the market model are from published
historical series. A summary of the key input data is presented in Table 4-16. More information
on the data sources and variables used for this analysis are described in Appendix B.
Price and quantity data for each market are mostly from USD A publications that
summarize national market conditions. ERS collates data from USDA/NASS, U.S. Department
of Commerce's Bureau of Labor Statistics (BLS), and other sources, as well as its own research,
to develop consistent price and quantity series for many agricultural activities. These series are
published in many formats such as USDA's Agricultural Outlook and also the Livestock, Dairy,
and Poultry Situation and Outlook reports (USDA/ERS, various dates). Much of this
information is readily available on-line at ERS' website. The USDA World Agricultural Outlook
Board (WAOB) bases many of its projections on these series and provides a concise summary of
the information in its annual series, titled USDA Agricultural Baseline Projections
(USDA/WAOB, 1999 and 2000). Another compendium series EPA uses for this analysis is
USDA/ERS' Food Consumption, Prices and Expenditures, 1970-1997 (Putnam and Allshouse,
1999). Other source material is from NASS statistical bulletins for these sectors, Agricultural
Prices Annual Summary, and USDA's Foreign Agricultural Service (FAS).
Where necessary, EPA supplements USDA market data with information from other
sources. Prices for choice fed steers and veal and choice retail beef were compiled with the
assistance of the National Cattlemen's Beef Association (NCBA) through their membership-with
Cattle-Fax, a member-owned information organization. Milk utilization (domestic demand and
traded volumes) is reported on a milk equivalent, total solids basis, as calculated by the National
Milk Producers Federation (NMPF, 1999). This measure aggregates the milk content across a
4-63
-------
Table 4-16. Market Model Baseline Values (1997)
Variable
Beef
Dairy b/
Hog
Broiler "
Turkey *
Layer.*
Farm Products
Price
Quantity
Produced
Quantity
Exported
Quantity
Imported
$66.09/cwt
40,893
thous. head
282
•thous. head
2,046
thous. head
$13.38/cwt
156,100
mil. Ibs
0
0
$54.30/cwt
91,960
thous. head
55
thous. head
3,178
thous. head
370/lb
.
'
-
400/lb
—
~
~
70£/doz.
6,473
mil. doz.
895
mil. doz.
0
Retail Products
Price
Quantity
Demanded
Quantity
Exported
Quantity
Imported
$2.80/lb
25,824
mil. Ibs
2,136
mil. Ibs
2,343
mil. Ibs
145.5
156,100
mil. Ibs
5,244
mil. Ibs
4,383
mil. Ibs
$2.45/lb
17,274
mil. Ibs
. 1,044
mil. Ibs.
633
mil. Ibs
151jzS/lb
27,551
mil. Ibs
5,048
mil. Ibs
5
mil. Ibs
105jzS/lb
5,412
mil. Ibs
598
mil. Ibs
0
1060/doz.
5,578
mil. doz.
228
mil. doz.
7
mil. doz.
Elasticities
Price Elasticity
ofDemand
Price Elasticity
of Supply
-0.621
1.020
-0.247
1.527
-0.728
0.628
-0.372
0.200
-0.535
0.200
-0.110
0.942
Sources:
Prices: Beef (NCBA, 2000); Dairy (USDA/ERS, 1998b); Hog (USDA/ERS, 1999c); Poultry (USDA/WAOB,
1999).
Quantities: BeefiVeal (USDA/ERS, 1998b and 1999d; USDA/NASS, 1998d; USDA/WAOB, 1999; Putnam and
Allshouse, 1999); Dairy (USDA/ERS, 1998; NMPF, 1999); Hogs (USDA/ERS, 1998b; USDA/NASS, 1998d;
USDA/WAOB, 1999); Poultry (USDA/ERS, 1998a).
Price Elasticities: Demand, all sectors (Huang, 1993); Beef (Foster and Burt, 1992; Foster, 2000a); Dairy (Chavas,
Kraus, and Jesse, 1990); Hog (Holt and Johnson, 1988); Chicken and Turkey (Vukina, 2000); Eggs (Chavas and
Johnson, 1981).
"'Includes veal and heifer.
w Consumer Price Index for Dairy Products as a composite good. Output is on a milk equivalent, total solids basis.
''Includes various forms of chicken meat, e.g. broilers, mixed poultry meat. Because of coordination between the
broiler feeding and processing sectors, it is not possible to determine prices per animal at the farm level. The
chicken and turkey markets are modeled in terms of pounds of production.
^gg quantities are in million dozens. Layer "Exports" represent eggs diverted to hatcheries, which is a fixed
number in the model.
4-64
-------
wide range of fluid .milk and dairy foods and allows for easy comparison with annual farm milk
supplies.
Another model input includes elasticities that measure the change in quantity relative to a
change in price. As described in Section 4.2.6 on cost passthrough, the elasticity measures that
EPA uses are the same as the "selected values" in Table 4-14. These values are selected based on
an extensive search of the agricultural economics literature and consultation witih leading experts
in the field. Price response research is conducted by various land grant universities and is
published in the leading academic journals. Limited information is available on the effects of price
on traded quantities. Although there is a general consensus that overseas markets are more
sensitive to prices than domestic markets (Foster, 2000a), elasticities for animal product imports
and exports have not been widely studied. EPA assumes that elasticities for imports and exports
are the same as domestic elasticities. Because price elasticities are critical to the outcome of the
model, EPA has performed a sensitivity analysis to see how alternative assumptions affect the
results. These sensitivity analyses are reported in Appendix D.
4.4.2.2 Input-Output Model Data
EPA uses multipliers from the Regional Input-Output Modeling System, version 2
(RIMS IT) (USDC, 1997b) to estimate the impact on the national economy of changes in the
animal products industry. RIMS n was developed to facilitate the use of input-output analysis in
regional planning. It is widely used to assess the impacts of changes in economic activity, such as
military base closings and economic development projects. A summary of the multipliers that
• EPA uses for the input-output analysis are presented in Table 4-17. Estimated effects include
changes in national employment (measured in terms of full-time equivalents) and changes in
economic output (measured in terms of changes in Gross Domestic Product). These estimated
changes are based on the estimated direct impacts, described in the preceding section, which are
measured in terms of dollars of industry output per year. More information is provided in
Appendix B.
Table 4-17 also illustrates the differences between RIMS n and another set of multipliers
from USFOOD, which is published by Ohio State University. USFOOD multipliers focus on the
agricultural sector and are based on the IMPLAN (Impact Analysis and Planning) input-output
software, developed by researchers at the U.S. Forest Service and University of Minnesota.
Multipliers from the IMPLAN model are commonly used for input-output analyses of agricultural
commodities (see, for example: Boggess, et al., 1997; Seidl and Weiler, 2000; Gray, et al., 1999;
Able, Daft & Early, 1993).
EPA has selected the RIMS H model for this analysis because it offers a more detailed
breakout of processing and nonagricultural industries, as well as household multipliers. The
version of USFOOD available at the time of this research was based on 1977 production patterns
4-65
-------
Table 4-17. RIMS II Multipliers for Secondary Impact Analysis (Comparison with USFOOD)
Sector Name
Total Output
RIMSH
USFOOD
Total Employment"7
RIMS II
USFOOD
Own Industry
Employment ^
RIMSH
Farm Products .
Poultry and Eggs
Meat Animals
Hogs
Cattle Feeding
Dairy Farm Products
2.8217
2.1692
—
• —
2.3215
2.5882
...
2.6823
2.7146
1.8188
26.3665
20.8698
—
—
23.5130
35.5598
.
32.2572
32.1979
42.3332
12.1043
10.2072
-
—
11.5267
Food and Kindred Products"'
Meat Packing Plants
Meat/Egg Processing
Poultry Processing
Fluid Milk Processing
Dairy Processing
2.4755
—
2.1822
2.3968
-
—
3.0777
'
—
2.6917
18.2332
—
18.0150
16.3311
—
—
32.1910
"
•
32.1408
3.3511
-
6.1750
2.6981
—
Other
Construction
Households d
3.0851
2.1642
1.9332
—
32.4816
23.8483
26.8497
—
10.9000
0.2823
Sources: RIMS H: USDC, 1996, RIMS II 6/19/96 based on 1987 benchmark I-O accounts and 1992 employment
and salary data, region is entire United States, adjusted to be output-driven. USFOOD: USDC, 1996, and
Sporleder and Liu, 1992.
''Employment multipliers adjusted for inflation to 1997 values based on CPI-U.
^Processing sector multipliers adjusted to eliminate CAFO level impacts.
''USFOOD does not include household multipliers. The RIMS II household multiplier used in the USFOOD
results.
updated to 1985 output and prices while RIMS II is based on 1987 production patterns updated
with 1992 data. RIMS n is based on more up-to-date information as well as providing greater
detail in the households and processing industries. Other available models provide the ability to
perform more detailed input-output analysis, which is not warranted given the likely uses and level
of detail needed for this analysis.
As Table 4-17 shows, RIMS n and USFOOD have slightly different industry definitions
and so are not exactly comparable. Only those multipliers with the same component standard
industrial classifications (SIC codes) are shown on the same line in the table. While the output
multipliers are similar, RIMS n employment multipliers are 25 to 50 percent smaller than
4-66
-------
USFOOD employment multipliers. Some of the difference may be attributable to changes in farm
and food processing technology between 1977 and 1987 and employment changes between 1985
and 1992. Methodological differences between the two approaches may also cause multipliers to
differ. Rickman and Schwer (1995) found that when the IMPLAN and RMS H models were
adjusted to control for differing closure methods, the multipliers did not differ significantly.
Once national level employment impacts are determined, these estimated impacts are used
to estimate potential regional impacts at the community level. EPA distributes national
employment estimates onto a regional basis using state level data that reflect livestock and
poultry production by the largest facilities in each sector. The data EPA uses to allocate national
employment impacts in the agriculture sectors are from the 1997 Census (USDA/NASS, 1999a),
which are supplemented with other state level data from USDA (USDA/NASS, 1998b) and with
imputed values to correct for omitted values (Westat, 2000).
4.4.3 Criteria for Assessing Regulatory Impacts
Impacts at the market level must be carefully assessed. Unless a rule has a profound effect
on all levels of the U.S. economy, market level effects at the national level are usually quite small.
Conversely, however, small market level effects do not preclude the possibility of large impacts on
smaller units of the economy-at the sector level, regional level, or regulated entity level (in the
latter case, at individual CAFOs). Therefore, EPA closely scrutinizes the results of its input-
output analysis to avoid summary judgments regarding the affordability of the CAFO regulations,
given the likely modest reductions in overall economic output and employment that EPA
calculates (even when the gains in employment and output associated with compliance cost
expenditures are not considered).
The proposed CAFO regulations could result in disproportionate effects in some
producing areas and could induce out-migration or relocation among producing regions. These
changes could affect rural communities that may depend on the farming sectors in an area to
sustain regional employment and support local businesses (such as input supply industries and
other supporting infrastructure) as well as to contribute to local tax revenues (CARD, 1993).
These regional impacts may not be readily apparent in a review of the aggregate market level
impacts. To the extent that data are available, EPA's analysis examines regional effects that may
occur under the proposed CAFO regulations (as described in Section 4.4.2.2).
Furthermore, affordability judgments may require some consideration of who bears the
impacts. It may not be equitable to assume that because few impacts can be seen at the highest
levels of the economy, a rule is affordable if major dislocations are expected to occur at the
regulated-entity level (e.g., a large number of CAFOs becoming vulnerable to closure).
With these concerns in mind, EPA assesses the affordability of the regulatory options and
the proposed CAFO regulations from a market standpoint in a more qualitative way than it does
4-67
-------
affordability at the CAFO or processor level. EPA evaluates the predicted market changes by
comparing these to reported actual changes over the last decade. This comparison provides a
benchmark for evaluating the affordability of predicted changes and also takes into account year-
to-year variability of the affected markets.
This comparison alone, however, is overly simplistic. Agricultural markets constantly
respond to change, whether it be year-to-year volatility in producer prices or input prices (such as
the cost of feed), or even reductions in output due to severe weather events. At the same time,
consumer markets for these products are large and relatively stable: a constant stream of suppliers
is almost certain and retail prices, when expressed in real terms, have remained fairly flat or have
actually been decreasing. While it is true that these markets must constantly adapt to change, this
does not address the fact that the proposed CAFO regulations would result in a sustained increase
in production costs (and result in a permanent upward shift in the supply curve).
Although these markets as a whole appear able to absorb most any market shock, large-
scale disruption within individual segments of these markets, such as the farm and processing
sectors, can occur (as discussed above) and also warrant consideration. Thus, to the extent
possible, EPA considers affordability at all three levels-CAFO, processor, and national level
markets-when assessing the overall affordability of the proposed CAFO regulations.
4-68
-------
SECTION FIVE
TOTAL COSTS AND ECONOMIC IMPACTS OF THE PROPOSED
CAFO REGULATIONS (ALL SUBCATEGORIES)
This section presents the national level aggregate compliance costs and economic impacts
to regulated facilities under the proposed CAFO regulations. Section 5.1 presents EPA's analysis
of the expected post-tax costs to industry as a result of the proposed CAFO regulations. Sections
5.2 through 5.4 present the results of EPA's analysis that evaluates the financial impacts across
three industry segments: CAFOs, processors, and national markets. Section 5.2 presents the
results of EPA's analysis at the CAFO level in terms of the expected costs to both existing and
new facilities, as well as the cost to other farming operations that use CAFO manure as a fertilizer
substitute. Section 5.3 presents an analysis of the potential costs and impacts to the processing
sectors in some industries. Finally, Section 5.4 presents the results of EPA's market level
analysis, focusing on the potential secondary impacts of the proposed regulations on both
consumer and farm level prices and quantities, and changes in national level employment and
economic output.
The results presented in this section span various technology options and also different
scope scenarios considered by EPA during the development of the proposed revisions. These
ELG Options and NPDES Scenarios are described in Section 3 of this report.1 A summary
overview of the ELG Options and NPDES Scenarios is provided in Table 3-1.
For the purpose of this analysis, the "BAT Option" refers to EPA's proposal to require
nitrogen-based and, where necessary, phosphorus-based land application controls at all livestock
and poultry CAFOs (Option 2), with the additional requirement mat all cattle and dairy operations
(except veal) must conduct groundwater monitoring and implement controls, if the groundwater
beneath the production area has a direct hydrologic connection to surface water (Option 3), and
with the additional requirement that all hog, veal, and poultry CAFOs achieve zero discharge from
the animal production area with no exception for storm events (Option 5). "
EPA is jointly proposing two NPDES Scenarios that differ in terms of the manner in
which operations are defined as a CAFO. Scenario 4a is the two-tier approach that defines as
CAFOs all animal feeding operations with more than 500 AU; facilities with fewer than 500 AU
are CAFOs only if designated by the permit authority. (Alternatively, Scenario 5 is the two-tier
alternative that defines all animal feeding operations with more than 750 AU as CAFOs.)
'More detail of the technology options considered by EPA is provided in Section VIII of the preamble.
Section VII of the preamble provides additional information on the alternative scope scenarios considered by EPA.
The preamble presents the Agency's rationale for each regulatory decision.
5-1
-------
Scenario 3 is the three-tier approach that defines as CAFOs all animal feeding operations
with more than 1,000 AU and any operation with more than 300 AU, if they meet certain "risk-
based" conditions. Facilities with fewer than 300 AU are CAFOs only if designated by the permit
authority. Under Scenario 3, EPA would require all confinement operations with between 300
and 1,000 AU to either apply for a NPDES permit or to certify to the permit authority that they
do not meet certain conditions and thus are not required to obtain a permit. EPA is also soliciting
comment on an alternative to co-proposed three-tier structure (Scenario 6). EPA did not evaluate
the costs and economic impacts under this alternative, however, EPA expects mat the numbers of
CAFOs affected under Scenario 6 would be fewer than the 31,930 estimated for Scenario 3;
therefore, costs and impacts should be no more than those for Scenario 3. If after considering
comments, EPA decides to further explore this approach, it will conduct a full analysis of this
scenario. Scenario 6 will not be further addressed in this section.
For the purpose of this discussion, the "two-tier structure " refers to the combination of
BAT Option 3 (beef and dairy subcategories, except veal) and BAT Option 5 (swine, veal, and
poultry subcategories) and NPDES Scenario 4a, which covers all operations with more than 500
AU. Where indicated, the two-tier structure may refer to the alternative threshold at 750 AU.
The "three-tier structure " refers to the combination of proposed BAT Option by subcategory and
NPDES Scenario 3 that covers operations down to 300 AU based on certain conditions.
More detail of me technology options and scoping options that are being proposed by
EPA as well as alternatives that were considered by EPA during the development of this
rulemaking is provided in Section 3 of this report. More detailed information is provided in
Sections VTI and Vin of the preamble.
5.1 ANNUAL COMPLIANCE COSTS OF THE PROPOSED CAFO REGULATIONS
This section presents EPA's estimates of the compliance costs that would be incurred by
existing sources under the regulatory options being considered for the beef, veal, heifer, dairy,
pork, broiler, turkey, and egg laying sectors under both the two-tier (Scenario 4a and Scenario 5)
and three-tier (Scenario 3) structure. Section 5.1.1 presents EPA's estimate of the annual
incremental costs of the proposed BAT Option under both tier structures. Section 5.1.2 presents
EPA's estimate of the annual costs of other ELG Options and NPDES Scenarios considered by
EPA during the development of this rulemaking.
5.1.1 Annual Costs under Two-Tier and Three-Tier Structures
Tables 5-1 through 5-3 summarize the total annualized compliance costs to CAFOs
attributed to the proposed BAT Option/Scenario 4a (two-tier structure at 500 AU threshold), the
proposed BAT Option/Scenario 5 (two-tier structure at 750 AU threshold) and the proposed
BAT Option/Scenario 3 (three-tier structure). The tables show these costs broken out by sector
5-2
-------
and by broad facility size group. EPA calculates these costs using the data, methodology, and
assumptions described in Section 4 and Appendix A. More detailed cost data are provided in the
Development Document (USEPA, 2000a).
Under the two-tier structure, EPA estimates that 25,540 CAFOs with more than 500 AU
may be defined as CAFOs and subject to the proposed regulations (Table 5-1). EPA estimates
that 19,100 CAFOs may be defined as CAFOs under the alternative two-tier threshold of 750 AU
(Table 5-2). Under the three-tier structure, an estimated 31,930 CAFOs would be defined as
CAFOs (Table 5-3) and an additional 7,400 operations in the 300 to 1,000 AU size range would
need to certify that they do not need to apply for a permit. Additional operations would be
affected by the proposed regulations if designated as CAFOs by the permitting authority. More
information on EPA's estimate of the number of affected CAFOs is provided in Section 2.
Table 5-1. Annualized Post-Tax Costs, Two-Tier (500 AU), BAT Option/Scenario 4a, $1997 millions
Sector
Beef
Veal
Heifer
Dairy
Hog
Broiler
Layer
Turkey
Total
No. of
Operations ^
3,080
90
800
3,760
8,550
9,780
1,640
1,280
25,540
Total
$135.0
$0.2
$8.6
$111.4
$198.9
$74.4
$9.1
$13.3
$550.9
>1000 AU
$118.5
$0.03
$2.8
$65.7
$148.8
$41.8
$6.3
$6.8
$390.7
500 - 1000 AU
$16.5
$0.2
$5.8
$43.3
$48.9
$32.5
$2.8
$6.5
$156.4
<500 AU
$0.1
NA
NA
$2.4
$1.2
$0.1
' • NA
NA
$3.8
Source: USEPA. Options/Scenarios are defined in.Table 3-1. May not add due to rounding. NA=Not Applicable.
"Total number of affected facilities adjusts for operations with more than a single animal type and includes
expected defined CAFOs only (excludes designated facilities). However, estimated costs include costs to
designated CAFOs. Section 2 provides additional information on the number of affected facilities.
5-3
-------
Table 5-2. Annualized Post-Tax Costs, Two-Tier (750 AU), BAT Option/Scenario 5, S1997 millions
Sector
Beef
Veal
Heifer
Dairy
Hog
Broiler
Layer
Turkey
Total
No. of
Operations ^
2,480
40
420
810
5,750
7,780
1,460
740
20,920
Total
$125.3
$0.1
$4.2
$86.9
$171.1
$66.l'
$9.0
$9.6
$472.2
>1000 AU
$118.5
$0.03
$2.8
$65.7
$148.8
$41.8
$6.3
$6.8
$390.7
750 - 1000 AU
$6.7
$0.1
$1.4
$15.0
$21.5
$23.4
$2.6
$2.7
$73.3
<750 AU
$0.1
NA'
NA
$6.3
$0.9
$1.0
NA
NA
$8.2
Source: USEPA. Options/Scenarios are defined in Table 3-1. May not add due to rounding. NA=Not Applicable.
"Total number of affected facilities adjusts for operations with more than a single animaltype and includes
expected defined CAFOs only (excludes designated facilities). However, estimated costs include costs to
designated CAFOs. Section 2 provides additional information on the number of affected facilities.
Table 5-3. Annualized Post-Tax Costs, Three-Tier Structure, BAT Option/Scenario 3 ($1997 millions) *
Sector
Beef
Veal
Heifer
Dairy
Hog
Broiler
Layer
Turkey
Total
No. of
Operations b/
3,210
140
980
6,480
8,350
13,740
2,010
2,060
31,930
Total
- $143.5
$0.5
$10.6
$146.9
$214.9
$90.0
$9.8
$17.4
$633.7
>1000 AU
$118.5
$0.03
$2.8
$65.7
$148.8
$41.8
$6.3
$6.8
$390.7
300 - 1000 AU
$25.0
$0.5
$7.8
$81.0
$65,5
$48.1
$3.5
$10.5
$242.0
<300AU
NA
NA
NA
$0.4
$0.2
NA
NA
NA
$0.6
Source: USEPA. Options/Scenarios are defined in Table 3-1. May not add due to rounding. NA=Not Applicable.
'Total number of affected facilities adjusts for operations with more than a single animal type and includes expected
defined CAFOs only (excludes designated facilities). However, estimated costs include costs to designated CAFOs.
Section 2 provides additional information on the number of affected facilities.
5-4
-------
EPA's estimate of the number of affected CAFOs counts operations with more than a
single animal type only once.2 However, EPA's analysis computes total compliance costs (and also
financial impacts to facilities) based on the total number of CAFOs in each sector, including mixed
operations that have more than the threshold number of animals (300, 500, or 750 AU) of at least
one animal type. This approach avoids understating costs at operations with more than one animal
type. Such operations may incur costs to comply with the proposed requirements for each type of
animal raised on site that meets the size threshold for a CAFO or is designated as a CAFO by the
permitting authority. Therefore, EPA's compliance costs estimates likely represent the upper
bound, since costs at facilities with more than a single animal type may, in some cases, be lower
due to shared production technologies and practices across all animal types that are produced on
site.
EPA calculates two types of compliance costs-pre-tax and post-tax. The post-tax costs
reflect the fact that a CAFO would be able to depreciate or expense these costs, thus generating a
tax savings. Post-tax costs thus are the actual costs the CAFO would face. Pre-tax costs reflect
the estimated total social cost of the proposed regulations, including lost tax revenue to
governments. Pre-tax dollars are used when comparing estimated costs to monetized benefits that
are estimated to accrue under the proposed regulations (see Section 10). All costs presented in
this section are presented in terms of post-tax 1997 dollars and account for annual tax savings to
CAFOs. Post-tax costs are also used to evaluate impacts to regulated facilities, presented in
Section 5.2.). EPA's estimated compliance costs presented hi the Development Document are also
estimated in 1997 dollars, since 1997 is the base year of the analysis (USEPA, 2000a). Cost
results presented in the preamble to this rulemaking (and the Executive Summary to this report)
are converted from 1997 dollars to 1999 dollars using the Construction Cost Index (ENR, 2000).
Under the two-tier structure at 500 AU threshold, EPA estimates that the total incremental
compliance cost to CAFO operators would be approximately $551 million annually, 1997 post-tax
dollars (Table 5-1). Under the alternative two-tier structure at 750 AU threshold, EPA estimates
that the total incremental compliance cost to CAFO operators would be approximately $472
million annually (Table 5-2). Under the three-tier structure, EPA estimates that the total cost to
CAFO operators would be $634 million annually (Table 5-2). (Pre-tax costs are estimated at $831
million, $721 million, and $980 million annually, respectively, expressed in 1999 dollars.3) Most of
this cost (roughly 70 percent) is incurred by CAFOs with more than 1,000 AU. Overall, about
one-third of all estimated compliance costs are incurred within the hog sectors.
Estimated costs for the three-tier structure include the cost to permitted CAFOs and the
cost to operations to certify to the permit authority that they do not meet any of the "risk-based"
2Census data for 1992 indicate that operations with mixed animal types account for 25 percent of
operations with 300 and 1000 AU. Fewer operations are mixed among operations with more than 1,000 AU
(USEPA, 2000a).
3These pre-tax costs are presented in Section X.E of the preamble.
5-5 ,
-------
conditions and thus are not required to obtain a permit. Under Scenario 3, EPA estimates that
31,930 livestock and poultry operations with more than 300 AU are defined as CAFOs under the
proposed regulations and required to obtain a permit (Table 5-3). An estimated 7,400 operations
would certify that they do not need to obtain a permit (estimated as the difference between
permitted operations under Scenario 3 and Scenario 4b, as shown in Table 5-5). Certification
costs incurred by these operations are included in the total annual estimated costs and are
estimated at about $40 million annually (post-tax, 1997 dollars) for each of the technology options
that build on Option 2 (Options 3, 5,6, and 7). This amount is based on the estimated difference
between costs for Scenario 3 and Scenario 4b, under the ELG Option 2, for permitted CAFOs.4
EPA expects that this difference reflects the cost to operations to certify out of the permit program
associated with phosphorus-based PNP costs, facility upgrades, and letters of certification from
manure recipients. The cost to certify out of the program is assumed to be the same as the
estimated costs incurred under Option 2, since this option covers basic facility upgrades and
nutrient management planning.
Estimated costs shown in Tables 5-1 through 5-3 include costs to animal confinement
operations that may be designated as CAFOs. As discussed in Section 2, EPA assumes that
designation may bring an additional 50 operations each year under the two-tier structure at 500
AU threshold, an additional 85 each year under the two-tier structure at 750 AU threshold, and an
additional 10 operations each year under the three-tier structure. In this analysis, estimated costs
to designated facilities are expressed on an average annual basis over a projected 10-year period.
For the purpose of this analysis, EPA assumes that operations that may be designated as CAFOs
and subject to the proposed regulations would consist of dairy and farrow-finish hog operations
located in more traditional farming regions. EPA also expects that some beef, egg laying, and
broiler operations may also be designated as CAFOs under either two-tier structure (see
Table 2-5). Total annualized costs to designated facilities are estimated at under $4 million (Table
5-1), under $8 million (Table 5-2), under $1 million dollars annually (Table 5-3), depending on
scenario.
Table 5-4 shows EPA's estimated incremental costs to offsite recipients. Offsite recipients
include field crop producers who use CAFO manure as a fertilizer substitute. As described in the
preamble, EPA is proposing that offsite recipients of CAFO manure certify to the CAFO that
manure that will be land applied in accordance with proper agriculture practices. EPA estimates.
that 18,000 crop operations will receive manure and therefore be required to certify proper manure
utilization under the proposed two-tier structure (Scenario 4a). Under the three-tier structure, up
to 3,000 additional crop operations may be affected (Table 5-4). Annualized costs to offsite
recipients are estimated at $9.2 to $10.9 million annually across the two-tier (500 AU threshold)
and three-tier structures, respectively (1997 dollars) (Table 5-4).5 Costs to offsite recipients for
4Total costs to CAFOs that certify out are calculated as the difference in the estimated cost of Option 2
under Scenario 4b ($423 million) and the cost of Option 2 under Scenario 3 ($384 million). See Table 5-5.
5These costs do not account for possible tax savings associated with the expenditure.
5-6
-------
the two-tier structure at the 750 AU threshold costs have not been estimated. EPA's estimates the
costs to recipients of CAFO manure that cover incremental recordkeeping and soil tests every 3
years, the cost of soil and manure sampling at the CAFO site, training for manure applicators,
application equipment calibration, and the hauling cost of excess manure generated by the CAFO.
These costs do not include the costs of spreading manure at the offsite location or any additional
payments made to brokers or manure recipients in counties with excess manure. This analysis is
provided in the Development Document (USEPA, 2000a).
Table 5-4 Annualized Costs to Offsite Recipients of CAFO Manure, $1997 and $1999 millions
Sector
Number of Recipients
Total Costs ($1999 million)
>1000 AU
13,489
$7.2
$6.9
>500 AU
17,923
$9.6
$9..2
>300 AU
21,155
$11.3
$10.9
^3: Development DC
rounding. NA = Not Applicable. Costs are indexed to 1997 dollars using the Construction Cost Index (ENR,
2000).
5.1.2 Costs to CAFOs of Alternative Regulatory Options and Scenarios
Alternative regulatory options considered by EPA during the development of proposed
CAFO regulations include various technology options and also different regulatory scope
scenarios, as summarized in Table 3-1 (Section 3). The following sections provide additional
break out of these costs.
Table 5-5 summarizes the total annualized (post-tax) costs of alternative technology
options for each NPDES scenario and ELG technology basis considered by EPA. As shown in the
table, the total estimated costs across these options range from $230 million (Option I/Scenario 1)
to $1.1 billion annually (Option 5, applicable to all the animal sectors, and Scenario 4b). By
scenario, this reflects the fact that fewer CAFOs would be affected under Scenario 1 (a total of
about 16,400 operations) as compared to Scenario 4b (about 39,300 operations affected). Since
EPA's estimate of the number of CAFOs and corresponding compliance costs do not adjust for
operations with mixed animal types, costs may be overstated. By technology option, with the
exception of Options 1 and 4, costs are evaluated incremental to Option 2 (see Table 3-1).
Incremental to Option 2, Option 5 costs are greatest.
5.1.2.1 Annual Costs of the Alternative ELG Options
Table 5-6 summarizes the total annualized (post-tax) costs to CAFOs of the proposed BAT
Option along with six alternative technology options considered by EPA under the two-tier (500
AU threshold), two-tier (750 AU threshold), and three-tier structures, respectively. These costs
are broken out by each sector in the tables below.
5-7
-------
Table 5-5. Annualized Post-fax Costs for All ELG Options and NPDES Scenarios ($1997, millions)
Option/
Scenario
#CAFOs"
Option 1
Option 2
Option 3
Option 4
Option 5
Option 6
Option 7
BAT Option
Scenario 4a
"Two-Tier"
25,540
$286.7
$359.2
$490.8
$598.3
$940.7
$418.8
$438.8
$550.9
Scenario 2/3
"Three-Tier"
31,930
$308.6
$383.6
$576.7
$736.5
$1,027.5
$500.3
$517.9
$633.7
Scenario 1
16,420
$230.3
$284.2
$396.0
$460.0
$819.7
$331.9
$347.3 '
$444.7
Scenario 5
>750AV
19,100
$252.9
$313.2
$422.2
$502.3
$853.5
$361.2
$378.2
$472.2
Scenario 4b
>300 AU
39,320
$335.3
$423.1
$594.1
$760.9
$1,057.5
$482.9
$503.5
$664.4
Source: USEPA. Cost estimates shown include costs to designated operations. Numbers may not add due to
rounding. Option/Scenario definitions provided in Table 10-2.
As shown in Table 5-6, the total estimated costs across these options range from $287 million
(Option 1) to $941 million (Option 5, all subcategories) in annual 1997 dollars. Table 5-6 also
presents these same results for the two-tier structure at the 750 AU threshold. As shown, the total
estimated costs across these options range from $253 million (Option 1) to $854 million
(Option 5) in annual 1997 dollars, as applicable to all the animal sectors. For the three-tier
structure, the total estimated costs across these options range from $309 million (Option 1) to
$1,028 million (Option 5) in annual 1997 dollars.
5.1.2.2 Annual Costs of the Alternative NPDES Scenarios
Table 5-7 summarizes the total annualized post-tax compliance costs associated with the
alternative NPDES scenarios that were considered but not proposed by EPA, in 1997 dollars.
(Results for Scenario 2 are the same as those for Scenario 3 since each scenario affects the same
number of CAFOs and both costs and impacts are the same.) As shown, the estimated annual
post-tax compliance costs for the proposed BAT Option range from $445 million (Scenario 1) to
$664 million (Scenario 4b) in annual 1997 dollars. This outcome is consistent with expectations,
since fewer CAFOs are affected under Scenario 1 compared to Scenario 4b.
5-8
-------
Annnalired Post-Tax Costs AH ELG Ontions ($1997, millions)
Option 1 Beef
Veal
Heifer
Dairy
Hog
Broiler
Layer
Two-Tier Structure (500 AU)
Option 1
Option 2
Option 3
Option 4
Option 5
Option 6
Option 7
BAT Option
$47.8
$77.2
$135.0
$146.4
$532.5
$77.2
$85.9
S135.0
$0.1
$0.2
$0.2
$0.6
$0.2
$0.2
$0.2
$0.2
$3.6
$5.7
$8.6
$12.0
$6.7
$5.7
$5.7
$8.6
$64.6
$60.7
$111.4
$125.5
$105.6
$83.2
$120.4
$111.4
$88.8
$118.7
$131.1
$154.7
$198.9
$155.7
$129.8
$198.9
$66.2
$74.4
$80.7
$124.0
$74.4
$74.4
$74.4
$74.4
$7.3
$9.1
$9.6
$15.7
$9.1
$9.1
$9.1
$9.1
Turkey
$8.3
$13.3
$14.3
$19.4
$13.3
$13.3
. $13.3
$13.3
Total
$286.7
$359.2
$490.8
$598.3
$940.7
$418.8
$438.8
$550.9
Two-Tier Structure (750 AU)
Option 1
Option 2
Option 3
Option 4
Option 5
Option 6
Option 7
BAT
$43.3
$71.5
$125.3
$134.4
$513.6
$71.5
$79.6
$125.3
$0.1
$0.1
$0.1
$0.2
$0.1
$0.1
$0.1
$0.1
$1.2
$2.5
$4.2
$6.0
$3.1
$2.5
$2.5
$4.2
$53.2
$49.0
$86.9
$96.3
$81.1
$59.9
$95.7
$86.9
$83.1
$105.5
$114.9
$131.6
$171.1
$142.5
$115.6
$171.1
$58.7
$66.1
$71.2
$106.2
$66.1
$66.1
$66.1
$66.1
$7.1
$8.9
$9.4
$14.7
$9.0
$8.9
$8.9
$9.0
$6.3
$9.6
$10.2
$12.9
$9.6
$9.6
$9.6
$9.6
$252.9
$313.2
$422.2
30VJ.V
$361.2
$378.2
$472.2
Three-Tier Structure (Scenario 3) ,
Option 1
Option 2
Option 3
Option 4
Option 5
Option 6
Option 7
BAT
$48,6
$77.8
$143.5
$160.4
$544.4
$90.7
$98.9
$143.5
$0.2
$0.3
$0.6
$1.2
$0.5
$0.5
$0.5
$0.5
$4.5
$6.9
$10.6
$15.1
$8.8
$7.6
$7.6
$10.6
$73.5
$72.2
$146.9
$177.8
$141.7
$102.1
$137.6
$146.9
$85.8
$111.9
$147.6
$176.9
$214.9
$182.3
$156.1
$214.9
$78.1
$87.8
$98.4
.$159.7
$90.0
$90.0
$90.0
$90.0
•• $7.7
$9.6
$10.5
$18.1
$9.8
$9.8
$9.8
$9.8
$10.2
$17.1
$18.8
$27.4
$17.4
$17.4
$17.4
$17.4
$308.6
$383.6
$576.8
$736.7
$1,027.5
$500.3
$517.9 1
$633.6 ||
Source: USEPA. Options/Scenarios are defined in Table 3-1. Numbers may not add due to rounding.
Estimated costs include costs to designated CAFOs.
5-9
-------
Across all other alternative technology options and scope scenarios, EPA estimates that
costs range from a low of $230 million (Option I/Scenario 1) to a high of over $1 billion annually
(Option 5/Scenario 4b). Under the proposed BAT Option and alternative NPDES scenarios, the
hog sector bears the largest costs among the regulated subcategories, followed by the beef
subcategory. Lowest costs are associated with the poultry sector (Table 5-7).
Table 5-7. Annualized Post-Tax Costs of Options Under Alternative NPDES Scenarios (S1997, millions)
Sector
Option 1
Option 2
Option 3
Option 4
Option 5
Option 6
Option 7
BAT
Option
($1997 millions)
Scenario 1
Cattle
Dairy
Hog
Poultry
Total
$43.7
$57.0
$80.3
$49.3
$230.3
$72.8
$53.7
$100.7
$57.1
$284.2
$127.1
$98.8
$109.6
$60.5
$396.0
$138.0
$112.5
$125.7
$83.8
$460.0
$509.7
$91.2
$161.6
$57.2
$819.7
$72.7
$63.4
$138.6
$57.1
$331.9
$80.5
$98.3
$111.3
$57.1
$347.3
$127.1
$98.8
$161.6
$57.2
$444.7
Scenario 4b
Cattle
Dairy
Hog
Poultry
Total
$60.1
$77.3
$98.1
$99.9
$335.3
$93.6
$76.4
$136.2
$116.9
S423.1
$159.8
$152.4
$153.8
$128.0
$594.1
$180.7
$180.4
$192.5
$207.2
$760.9
$564.2
$141.1
N $235.3
$116.9
$1,057.5
$93.6
$99.2
$173.3
$116.9
$482.9
$102.4
$134.9
$149.3
$116.9
$503.5
$159.8
$152.4
$235.3
$116.9
$664.4
Source: USEPA. Options/Scenarios are defined in Table 3-1. Numbers may not add due to rounding.
5.2 CAFO IMPACTS
This section presents the CAFO level impacts under each of the ELG options and many of
the NPDES scenarios considered by EPA during the development of the proposed regulations.
Section 5.2.1 discusses the pre-regulatory, or baseline, financial health of EPA's model CAFOs
developed for this analysis. Section 5.2.2 examines the impact to existing facilities to comply with
the proposed ELG requirements for Best Available Technologies Economically Achievable (BAT).
Section 5.2.3 examines the affect of the proposed offsite requirements to recipients of CAFO
manure. Section 5.2.4 examines the impact to new facilities to comply with the proposed ELG
requirements for New Source Performance Standards (NSPS).
5-10
-------
Results are shown for the two-tier (500 AU threshold) and three-tier structures. The
impacts of the two-tier structure at 750 AU threshold are also discussed. Results are also shown
for other alternative regulatory scenarios considered by EPA during the development of this
rulemaking. ,
5.211 Baseline Financial Health of Model CAFOs
As discussed in Section 4, EPA's CAFO analysis examines compliance cost impacts for
representative model CAFOs. All baseline model CAFOs, regardless of sector or size or
production region, are considered to be financially healthy in the baseline before the impacts of the
proposed CAFO regulations are considered. Based on these data, all model CAFOs currently are
estimated to have positive discounted cash flow and debt-to-asset ratios of approximately 40
percent or less. Post-regulatory impacts are measured against this baseline. EPA considers that
negative cash flow or debt-to-asset ratios greater than about 40 percent in the impact analysis can
be attributed to the compliance costs associated with the regulatory options considered.
5.2.2 Post-compliance Impacts to Existing Operations (BAT Analysis)
5.2.2.1 Impacts under the Two-Tier and Three-Tier Structures
Economic achievability is determined by applying the proposed criteria described in Section
4.2.5. These criteria.include a sales test and also analysis of post-compliance cash flow and debt-
to-asset ratio for an average model CAFO. EPA uses these financial criteria to divide the impacts
of the proposed regulations into three impact categories. The first category is the affordable
category, which means that the regulations have little or no financial impact on CAFO operations.
The second category is the moderate impact category, which means that the regulations will have
some financial impact on operations at the affected CAFOs, but EPA does not consider these
operations to be vulnerable to closure as a result of compliance. The third category is the financial
stress category, which means that EPA considers these operations to be vulnerable to closure post-
compliance. EPA considers the stress impact category to indicate that the proposed requirements
may not be economically achievable by the CAFO, subject to other considerations.
For this analysis, impacts under the affordable and moderate category are associated with
positive post-compliance cash flow over a 10-year period and a debt-td-asset ratio not exceeding
40 percent, in conjunction with a sales test result that shows that compliance costs are less than 5
percent of sales ("affordable") or between 5 and 10 percent ("moderate"). "Stress" .impacts are
associated with negative cash flow or if the post-compliance debt-to-asset ratio exceeds 40
percent, or sales test results that show costs equal to or exceeding 10 percent of sales. Additional
information on these criteria and a discussion of the basis for EPA choosing these criteria for this
analysis is provided in Section 4.2.5. Table 4-13 provides a summary of the proposed threshold
values for each of these impact categories used for this analysis.
5-11
-------
EPA's model analyzes impacts under two sets of conditions for ELG Option 3. Option 3
would impose additional requirements (such as liners, groundwater monitoring, and
recordkeeping) if a hydrologic connection from the confinement areas to surface water is present.
(See Section 3.2 for additional information.) To depict the possibility of either condition, EPA
estimates impacts under two sets of alternative assumptions: one alternative assumes that there is a
hydrologic connection from groundwater to surface waters at the CAFO (Option 3A); the other
alternative uses average costs conditions across all operations—including those operations with
and without a hydrologic link (Option 3) (USEPA, 2000a).6 For this analysis, based on available
data and information, EPA's analysis assumes mat 24 percent of the affected operations have a
hydrologic connection to surface waters (as described in the Development Document, [USEPA,
2000a]). These operations will incur costs associated with groundwater monitoring controls. This
affects results shown for the cattle (beef, veal, and heifer) and dairy sectors. Impacts for Options 3
and 3 A are aggregated hi the results tables. However, of the estimated impacts shown for the
cattle and dairy sectors, all impacts under the moderate and stress categories, along with a portion
of impacts under the affordable category, are attributable to Option 3A costs and assumptions.
Tables 5-8 through 5-11 present the estimated CAFO level impacts for the proposed BAT
Option under the two-tier and three-tier structures by sector. Tables 5-9 through 5-10 show these
results across all facility sizes; Table 5-11 breaks out these results by broad size categories.
Results are expressed in terms of the number of operations that fall within the affordable,
moderate, or stress impact categories for facilities that are defined as CAFOs. For some sectors,
impacts are shown for both the zero and the partial cost passthrough assumptions.
Based on these results, EPA proposes that the regulatory alternatives are economically
achievable for all representative model CAFOs in the veal, turkey and egg laying sectors. The
proposed requirements under the two-tier structure are also expected to be economically
achievable by all affected heifer operations. Furthermore, although operations across most sectors
may experience moderate impacts, EPA does not expect moderate financial impacts to result in
closure and considers this level of impact to be economically achievable.
In the beef cattle, heifer, dairy, hog and broiler sectors, however, EPA's analysis indicates
that the proposed regulations will cause some operations to experience financial stress, assuming
no cost passthrough. These operations may be vulnerable to closure by complying with the
proposed regulations. Across all sectors, an estimated 1,890 operations would experience financial
stress under the two-tier structure and an estimated 2,410 operations would experience stress
under the three-tier structure. Under the two-tier structure at the 750 AU threshold, EPA
estimates that 1,700 operations would experience financial stress. For both tier structures, EPA
estimates that the percentage of operations that would experience impacts under the stress
category represent 7 percent of all affected CAFOs (or 8 percent of all affected operations hi the
Alternatively, estimated costs for "Option 3JB" reflect representative facility level costs where no
hydrologic link is present. Option 3, 3 A, and 3B costs are provided in the Development Document.
5-12
-------
sectors where impacts are estimated to cause financial stress in the cattle, dairy, hog, and broiler
sectors).
'Results shown in Tables 5-8 through 5-11 do not include designated facilities. In addition
to impacts evaluated for operations that are defined as CAFOs, EPA estimates that the proposed
regulations could result in financial stress to 20 designated dairies under both the co-proposed
two-tier and the three-tier structures. EPA does not expect that designated operations in other
sectors will experience financial stress due to compliance. Designated dairies that are expected to
experience stress based on the results of this analysis are operations that are designated due to a
hydrologic link to surface waters, projected over a 10-year period. Under the alternative two-tier
structure (750 AU threshold), no designated operations would experience financial stress based on
the results of this analysis.
5-8 Tmnacted Onerations Under the Two-Tier Structure (BAT Option/Scenario 4a)
Sector
Fed Cattle
Veal
Heifer
Dairy
Hogs: GF
Hogs: FF
Broilers
Layers - Wet
Layers - Dry
Turkeys
Total
Number
of
CAFOs
3,080
90
800
3,760
2,690
5,860
9,780
360
1,280
1,280
28,970
Affordable
Moderate
Stress
Zero Cost Passthrough
Affordable
Moderate
Stress
Partial Cost Passthrough
(Number of Affected Operations)
2,830
90
680
' 3,240
1,710
5,210
1,960
360
1,280
1,230
18,580
240
0
120
200
180
30
7,670
0
0
50
8,490
10
0
0
320
810
610
150
0
0
. 0
1,890
ND
ND
ND
ND
2,690
5,860
8,610
. ND
ND
ND
26,840
ND
ND
ND
ND
0
0
1,170
ND
ND
ND
1,800
ND
ND
ND
ND
0
0
0
ND
ND
ND
330
Source: USEPA. Impact estimates shown include impacts to designated operations. Numbers may not add due to
rounding. ND = Not Determined. Option/Scenario definitions provided in Table 3-1. Category definitions
("Affordable," "Moderate" and "Stress") are provided in Table 4-13.
5-13
-------
Table 5-8 shows results for the two-tier structure at the 500 AU threshold. By sector,
EPA estimates that 1,420 hog operations (17 percent pf affected hog CAFOs), 320 dairies (9
percent of operations), 150 broiler operations (2 percent), and 10 beef operations (less than 1
percent) would experience financial stress. The broiler and hog operations with these impacts have
more than 1,000 AU on-site (i.e., no operations with between 500 and 1,000 AU fall in the stress
category). The dairy and cattle operations with stress impacts are those that have a ground water
link to surface water. The results of the two-tier structure at the 750 AU threshold are very similar
in terms of number of operations affected, although no cattle operations and 190 fewer dairy
operations would experience financial stress (Table 5-9).
Table 5-9. Impacted Operations Under the Two-Tier Structure (BAT Option/Scenario 5)
Sector
Fed Cattle
Veal
Heifer
Dairy
Hogs: GF
Hogs: FF
Broilers
Layers - Wet
Layers - Dry
Turkeys
Total
Number
of
CAFOs
2,480
40
420
2,260
2,300
3,460
' 7,780
210
1,260
740
20,920
Affordable
Moderate
Stress
Zero Cost Passthrough
Affordable
Moderate
Stress
Partial Cost Passthrough
(Number of Affected Operations)
2,370
40
390
2,070
1,310
2,820
1,650
210
1,260
720
12,830
100
0
30
50
180
30
5,980
0
0
10
6,390
0
0
0
130
810
610
150
0
0
0
1,700
ND
ND
ND
ND
2,300
, 3,460
6,740
ND
ND
ND
19,540
ND
ND
ND
ND
0
0
1,040
ND
ND
ND
1,230
ND
ND
ND
ND
0
0
0
ND
ND
ND
130
Source: USEPA. Impact estimates shown include impacts to designated operations. Numbers may not add due to
rounding. ND = Not Determined. Option/Scenario definitions provided in Table 3-1. Category definitions
("Affordable," "Moderate" and "Stress") are provided in Table 4-13.
Table 5-10 presents results for the three-tier structure, and show that 1,420 hog operations
(17 percent of affected hog CAFOs under that alternative), 610 dairies (9 percent of operations),
330 broiler operations (2 percent), and 50 beef and heifer operations (1 percent) will be adversely
impacted. Hog operations with stress impacts all have more than 1,000 AU. Affected broiler
facilities include operations with more than 1,000 AU, as well as operations with less than 1,000
5-14
-------
AU. Dairy and cattle operations in the stress category are operations that have a hydrologic link
from ground water to surface water. Based on these results, EPA is proposing that the proposed
regulations are economically achievable.
Table 5-10. Tmnacted Onerations Under the Three-Tier Structure (BAT Option/Scenario 3)
Sector
Fed Cattle
Veal
Heifer
Dairy
Hogs: GF
Hogs: FF
Broilers
Layers - Wet
Layers - Dry
Turkeys
Total
Number
of
CAFOs
3,210
140
980
6,480
2,650
5,710
13,740
360
1,660
2,060
37,000
Affordable
Moderate
Stress
Zero Cost Passthrough
Affordable
Moderate
Stress
Partial Cost Passthrough
(Number of Affected Operations)
2,540
140
800
5,300
1,660
5,070
1,850
360
1,660
1,950
21^00
650
0
150
560
190
30
11,560
0
0
110
13,250
20
0
30
610
810
610
330
0
0
0
2,410
ND
ND
ND
ND
2,650
5,710
12,320
ND
ND
ND
33,410
ND
ND
ND
ND
0
0
1,440
ND
ND
ND
2,930
ND
ND
ND
ND
0
0
0
ND
ND
ND
660
Source: USEPA. Impact estimates shown include impacts to designated operations. Numbers may not add due to
rounding. ND = Not Determined. Option/Scenario definitions provided in Table 3-1. Category definitions
("Affordable," "Moderate" and "Stress") are provided in Table 4-13.
In the hog and broiler sectors, EPA also evaluates financial impacts with an assumption of
cost passthrough. For the purpose of this analysis, EPA assumes that the hog sector could pass
through 46 percent of compliance costs and the broiler sector could pass through 35 percent of
compliance costs. EPA derives these estimates from price elasticities of supply and demand for
each sector reported in the academic literature. More detailed information is provided in
Section 4.2.6. Assuming these levels of cost passthrough in these sectors, the magnitude of the
estimated impacts decreases to the affordable or moderate impact category. Even in light of the
uncertainty of cost passthrough (both in terms of whether the operations are able to pass cost
increases up the marketing chain and the amount of any cost passthrough), EPA proposes that the
proposed regulations will be economically achievable to all hog and broiler operations.
5-15
-------
Although EPA's analysis does not consider cost passthrough among cattle or dairy
operations, EPA does expect that long-run market and structural adjustment by producers in this
sector will diminish the estimated impacts. However, EPA did determine that an evaluation of
economic impacts to dairy producers would require that EPA assume cost passthrough levels in
excess of 50 percent before operations in the financial stress category would, instead, fall into the
affordable or moderate impact category. In this analysis, EPA evaluates impacts under a 67
percent cost passthrough assumption, which indicates that no dairy operations would experience
financial stress under the co-proposed tier structures. Additional information on this analysis is
provide in Section 8. EPA did not conduct a similar evaluation of estimated impacts to beef cattle
and heifer operations.
Table 5-11 breaks out EPA's estimated CAFO level impacts by broad size categories,
including operations with more than 1,000 AU and operations with fewer than 1,000 AU, for each
of the co-proposed tier structures. Impacts are shown for the zero cost passthrough scenario only.
EPA believes its estimated impacts may be overstated since the analysis does not quantify
various cost offsets that are available to most operations, some of which are described in
Section 4.2.7. One source of potential cost offset is cost share and technical assistance available to
operators for on-site improvements that are available from various state and federal programs,
such as the Environmental Quality Incentives Program (EQIP) administered by USD A. Another
source of cost offset is revenue from manure sales, particularly of relatively higher value dry
poultry litter. EPA's analysis does not account for these possible sources of cost offsets because
the amount of cost offset is likely variable among facilities, depending on certain site-specific
conditions. If EPA were to quantify the potential cost offsets as part of its analysis, this would
further support EPA's proposed determination that the proposed requirements are economically
achievable to affected operations. This analysis is provided in Section 6.
Appendix D provides results of sensitivity analyses, conducted by EPA, to examine the
impact under differing model assumptions. This analysis examine the change in the modeling
results from varying the baseline assumptions on gross and net cash income, debt-to-asset ratios as
well as other variability factors for model CAFOs. These sensitivity analyses conclude that the
results presented here are stable across a range of possible modeling assumptions. EPA also
conducted sensitivity analysis of the compliance costs developed for the purpose of estimating
CAFO level impacts, as documented in the Development Document (USEPA, 2000a).
5-16
-------
Table 5-11. Number of CAFOs Affected under the Co-Proposed Alternatives by Size (Zero Cost Passthrough)
Sector
Two-Tier Structure (500 AU Threshold)
No. of
CAFOs *
Aff.
Mod.
Stress
(number)
Three-Tier Structure
No. of
CAFOs *
Aff.
Mod.
Stress
(number)
CAFOs >1,000 AU
Fed Beef
Veal
Seifer
Dairy
Hogs
Broilers
Layers - Wet
Layers - Dry
Turkeys
Total
2,080
10
.300
1,450
4,090
3,940
50
590
370
12,870
2,080
10
300
1,450
2,460
200
50
590
370
7,500
0
0
0
0
210
3,600
0
0
0
3,810
0
0
0
0
. 1,420
150
0
0
0
1,560
2,080
10
300
1,450
4,090
3,940
50
590
370
12,870
2,080
' 10
300
1,450
2,460
200
50
590
370
7,500
0
0
0
0
210
3,600
0
0
0
3,810
CAFOs <1,000 AU
Fed Beef
Veal
Heifer
Dairy
Hogs
Broilers
Layers - Wet
Layers - Dry
Turkeys
Total
1,000
80
500
2,310
4,460
5,840
310
690
910
16,100
760
80
380
1,790
4,460
1,760
310
690
860
11,080
240
0
120
200
0
4,080
0
0
50
4,690
10
0
0
320
0
0
0
0
0
330
1,140
130
680
5,030
4,270
9,800
310
1,060
1,690
24,100
460
130
500
3,850
4,260
1,650
310
1,060
1,580
13,810
680
0
150
560
10
: 7,970
0
0
110
9,450
0
0
0
0
1,420
150
0
0
0
1,560
20
0
30
610
0
180
0
0
0
850
Source: USEPA. Impact estimates shown include impacts to designated operations. Numbers may not add due to
rounding. ND = Not Determined. Option/Scenario definitions provided in Table 3-1. Category definitions
("Affordable," "Moderate" and "Stress") are provided in Table 4-13.
5-17
-------
5.2.2.2 Impacts under Other Regulatory Alternatives
Tables 5-12 through 5-14 present the CAFO level impacts across the alternative ELG
options and 1SIPDES scenarios considered but not proposed by EPA during the development of this
rulemaking. In some cases, the results for alternative options are shown for Options 1 through 5
only, along with the proposed BAT Option, but are not shown for Options 6 and 7. As shown in
Table 3-1, for all technology options, with the exception of Options 1 and 4, EPA evaluates costs
in relation to Option 2. Results for Option 6 and 7 are not shown because these costs are not
evaluated in relation the proposed BAT Option combination (Option 3 for beef/dairy and Option 5
for pork/veal/poultry). If the full cost of Options 6 and 7 are considered, in addition to estimated
impacts under the proposed BAT Option, impacts to facilities would likely be more severe than
those shown for the proposed BAT Option.
Table 5-12 presents EPA's estimate of CAFO level impacts for the alternative ELG
options, assuming zero cost passthrough, under both the two-tier (Scenario 4a) and three tier
structure (Scenario 3). The results shown are partially aggregated and combine impacts in the
cattle sector, including all beef, veal, and heifer operations, and in the hog sector, including all
farrow-finish and grow-finish operations. Results are broken out to show estimated impacts under
the two sets of Option 3 assumptions (i.e., Option 3 across all operations and Option 3 A for
operations with a hydrologic link to surface waters). As shown, the proposed BAT Option is
associated with the same or slightly higher impacts than some of the other options. Under
alternative Option 4, however, EPA estimates that a substantial number of affected poultry
operations would experience financial stress, as defined for this analysis, assuming no passthrough
of costs.
Table 5-13 shows the impacts of the proposed BAT Option and the alternative options
under a partial cost passthrough assumption for the hog and poultry sectors only. For both co-
proposed alternatives, assuming modest levels of cost passthrough, the model shows that no hog
and poultry operations would experience financial stress impacts under the proposed BAT Option.
Under some other ELG options, some operations would experience impacts under the stress
category, even assuming cost passthrough. Under the two-tier structure at 500 AU threshold,
stress impacts are estimated at hog operations under the alternative Option 6 and also at poultry
operations under Option 4. Under the three-tier structure, additional operations would experience
financial stress under these options. Based on these results, the number of potential closures is
likely greatest under Option 4, which would require all CAFOs to conduct both groundwater
monitoring and surface water sampling.
5-18
-------
Table 5-12. Number of CAFOs Adversely Affected under Alternative Options (Zero Cost Passthrough)
Option
Cattle
Dairy
Hog
Broiler
Layer
Turkey
Stress Impacts under the Two-Tier Structure (500 AU Threshold)
Option 1
Option 2
Option 3
Option 3A
Option 4
Option 5
Option 6
Option 7
BAT
Option
0
0
0
10
0
30
0
0
10
0
0
0
320
0
0
0
0
320
610
300
230
310
570
1,420
1,210
500
1,420
0
150
260
90
6,660
150
150
150
150
0
0
0
0
0
0
0
0
0
0
0
0
0
10
a
0
0
0
Stress Impacts under the Three-Tier Structure
Option 1
Option 2
Option 3
Option 3A
Option 4
Option 5
Option 6
Option 7
BAT
Option
0
0
0
50
20
100
0
0
50
0
0
0
610
0
0
0
0
610
610
300
230
320
570
1,420
1,210
500
1,420
0
330
470
360
10,750
-330
330
330
330
0
0
0
0
0
0
o
0
0
0
0
0
10
10
0
0
0
0
Source: USEPA. Options/Scenarios are defined in Table 3-1. Category definitions ("Stress") are provided in
Table 4-.13. Numbers may not add due to rounding. Option 3A impacts reflect operations where there is a
determined groundwater hydrologic connection to surface waters (assumed at 24 percent of affected operations).
Option 3 impacts reflect average costs conditions across all operation for this option.
5-19
-------
Table 5-13. Number of CAFOs Adversely Affected under Alternative Options (Partial Cost Passthrough)
Option
Option 1
Option 2
Options
Option 4
Option 5
Option 6
Option 7
BAT Option
Hog
Poultry
Stress Impacts-Two-Tier Structure
(500 AU Threshold)
Hog
Poultry
Stress Impacts-Three-Tier Structure
(Scenario 3)
(number of operations)
0
0
0
0
0
. 440
0
0
0
0
0
130
0
0
0
0
0
0
0
0
0
440
0
0
0
0
0
2,610
0
0
0
0
Source: USEPA. Options/Scenarios are defined in Table 3-1. Category definitions ("Stress") are provided in
Table 4-13. Numbers may not add due to rounding. Option 3 includes impacts to facilities with a hydrologic link.
Table 5-14 compares the CAFO level impacts across the proposed and the alternative
JNPDES scenarios. Results for Options 1 through 5 are shown, along with the proposed BAT
Option. Results also break out estimated impacts under the two sets of Option 3 assumptions.
As shown in the table, the number of potential closures range from 450 operations
(Option 2/Scenario 1) to nearly 12,000 potential closures (Option 4/Scenario 4b). Among
options, the number of possible closures is highest under the more stringent options, including
Options 3A (i.e., requires groundwater controls at operations where there is a determined
groundwater hydrologic connection to surface waters), Option 4 (groundwater controls and
surface water sampling), and Option 5 (i.e., zero discharge from the animal production area with
no exception for storm events). Differences across scenarios reflect differences in the number of
affected operations; accordingly, the number of potential closures is likely greatest under Scenario
4b, which would define as CAFOs all confinement operations with more than 300 AU.
5.2.3 Post-compliance Impacts to Offsite Recipients of CAFO Manure
As discussed in Section 4.1.2.2, EPA assesses the economic impact to offsite recipients of
CAFO manure by comparing the estimated cost of this requirement to both aggregate and average
per-farm production costs and revenues. For the purpose of this analysis, EPA assumes that these
regulatory costs would be bome by a non-CAFO farming operation that uses animal manures as a
fertilizer substitute.
5-20
-------
Table 5-14 Number of CAFOs Adversely Affected under Alternative Scenarios (Zero Cost Passthrough)
Sector
Cattle
Dairy
Hogs
Poultry
Total
Cattle
Dairy
Hogs
Poultry
Total
No. of
CAFOs '
Option 1
Option 2
Option 3
Option
3A
Option 4
Option 5
BAT
Option
(number of operations with stress impacts)
NPDES Scenario 1
2,860
3,480
5,480
5,500
17,320
0
0
610
0
610
0
0
300
150
450
0
0
230
120 '
340
10
110
310
40
480
0
0
570
2,530
2,980
NPDES Scenario 4a (>500 AU)
3,960
3,760
8,550
12,700
28,970
0
0
610
, 0
610
0
0
300
150
450
0
0
230
260
490
10
320
310
100
730
0
0
570
6,660
7,230
20
0
1,420
150
1,590
30
0
1,420
150
1,590
10
110
1,420
150
1,690
10
320
1,420
150
1,890
NPDES Scenario 3 (>300 AU with certification)
Cattle
Dairy
Hogs
Poultry
Total
Cattle
Dairy
Hogs
Poultry
Total
4,330
6,480
8,360
17,830
37,000
0
0
, 610
0
610
0
0
300
330
630
0
0
230
470
700
50
610
320
370
1,350
0
0
570
10,740
11,310
NPDES Scenario 4b (>300 AU)
5,330
7,140
14,370
18,300
45 140
0
0
610
0
610
0
0
300
320
620
0
0
230
470
700
90
700
330
380
1,500
30
0
570
11,030
11,630
100
0
1,420
330
1,850
180
0
1,420
320
1,910
50
610
1,420
330
2,410
90
700
1,420
320
2,530
Source: USEPA. Options/Scenarios are defined in Table 3-1. Category definitions ( Stress ) are provided in
Table 4-13. Option 3A impacts reflect operations where there is a determined groundwater hydrologic connection
to surface waters (assumed at 24 percent of affected operations). Option 3 impacts reflect average costs conditions
across all operation for this option.
5-21
-------
As shown in Table 5-4, EPA estimates that 17,900 to 21,200 farming operations would
incur $9.2 million to $10.9 million in costs associated with requirements for the offsite transfer of
CAFO manure, depending on the co-proposed alternative ($1997). This translates to a cost of
roughly $500 per recipient, calculated as the average cost across the number of recipients
(Table 5-4).7 As reported by USD A, farm production expenses in 1997 totaled $150.6 billion
nationwide. Revenue from farm sales totaled $196.9 billion. Averaged across the total number of
farms, average per-farm costs and revenues were $78,800 and $113,000 in 1997, respectively.
Using these data, the ratio of incremental costs to offsite recipients to average operating expenses
is less than one percent. Compliance costs as a share of average farm revenue are estimated to be
even lower. Total estimated compliance costs as a share of aggregate farm expenses and sales are
also estimated at well below one percent.
5.2.4 Post-compliance Impacts to New Operations (NSPS Analysis)
EPA's proposed rule requires existing CAFOs to meet the BAT requirements of Option 3
for the beef and dairy subcategories (except veal) and Option 5 for the swine, veal, and poultry
subcategories. For new beef and dairy sources, EPA proposes that operations meet the same
performance standards required under the proposed BAT Option for these subcategories (Option
3 BAT). For new hog, veal, and poultry sources, EPA proposes an option that combines Option
5 BAT with the additional requirement that if there is a hydrologic link to surface water, the new
operations will also implement groundwater controls. This combined option is referred to as
Option 5+3 NSPS for the swine, veal, and poultry subcategories.
During the development of this rulemaking, EPA considered a similar combined "zero
overflow" and groundwater control option for new cattle and dairy operations (Option 8 NSPS),
which would have required all animals to be confined within a .covered structure (see Section VHI
of the preamble for a description of this NSPS option). EPA rejected Option 8 NSPS for the beef
and dairy subcategory based on the results of a barrier to entry analysis.
EPA's economic analysis of the proposed NSPS options assesses whether the proposed
standards constitute a "barrier" for new businesses wishing to enter the animal production market.
This determination is based on whether new sources would be subject to higher costs than
existing sources. Higher compliance costs for new sources could be considered a barrier to entry,
since existing sources would have a cost advantage. Generally, an NSPS option will have similar
or lower costs than the corresponding BAT option for an existing operation. This is because new
sources do not need to undertake expensive retrofits when installing pollution controls. NSPS
options are considered by EPA to result in no barriers to entry for new sources if the costs are the
same as or no greater than the BAT costs for existing sources since existing operators do not gain
a cost advantage over new operators. A significant cost advantage would be a barrier to entry for
'However, EPA calculates the total cost to offsite recipients based on an estimated cost to recipients of
roughly $1,000-per facility, which is assessed across 54 percent of facilities that are assumed to incur incremental
costs. For more information, see the Development Document (USEPA, 2000a).
5-22
-------
new sources. For NSPS options that are more stringent (and more costly) than requirements for
existing sources, EPA must determine whether the additional costs constitute a barrier to entry.
This section reviews all the analyses conducted by EPA to assess the NSPS options that
are being proposed. Detailed results are presented in the rulemaking record (ERG, 2000f— see
DCN 70599).
5.2.4.1 Impacts of the NSPS Options on the Beef and Dairy Subcategories
For this proposed rulemaking, EPA has evaluated the proposed option for new sources for
the beef and dairy subcategories (Option 3 NSPS). EPA's analysis indicates that requiring Option
3 for new sources for the beef and dairy subcategories would not create any barriers to entry since
the estimated costs for new sources are the same as or less expensive than the BAT costs for
existing sources. This determination is based on a comparison of the costs of Option 3 BAT to
the costs of Option 3 NSPS on a model-by-model basis. For this comparison, EPA uses the
weighted average of Option 3A and Option 3B costs for both Option 3 BAT and Option 3 NSPS.
Estimated Option 3 NSPS costs for new beef and dairy operations are lower than Option 3 BAT
costs since they do not include retrofitting costs that would be incurred by existing sources.
EPA's comparison of the estimated NSPS and BAT costs shows that the new source costs for
some model facilities were estimated to be more than 10 percent lower than those for existing
facilities (ERG, 2000f). However, EPA assumes that new operations will not incur costs
estimated under Option 3 A, which includes groundwater controls, since they are not nicely to
establish a new operation where there is a hydrologic link to surface waters (and where operating
expenses would be more costly). Thus Option 3 NSPS costs are likely to be even lower
compared to Option 3 BAT costs than was determined in this analysis. Since the estimated costs
for new sources are the same as or lower than the costs to existing facilities, EPA concludes that
the proposed NSPS option for the beef and dairy subcategory poses no barriers to entry in the
beef and dairy sectors.
This section also presents EPA's evaluation of an alternative option for new sources for
the beef and dairy subcategories (NSPS Option 8). This option is evaluated by comparing the
estimated Option 8 NSPS costs to the Option 3 BAT costs for each sector on a model-by-model
. basis. The difference in costs is determined by calculating the percent difference between the
costs of the NSPS options and the BAT options for each CAFO model. This percent difference is
used to judge whether to conduct additional analysis to determine whether a barrier to entry might
exist. The results of this comparison are presented in Table 5-15. Where Option 8 NSPS costs
are lower than Option 3 BAT costs, a negative percentage is shown (Table 5-15). As shown in
the table, Option 8 NSPS costs are less expensive than Option 3 BAT for beef and heifer
operations. For the dairy sector, however^ costs of Option 8 are significantly higher, estimated at
more than 10 times higher than Option 3 BAT for most models. Because of this large cost
difference in the dairy sector, EPA conducted the following additional analysis.
5-23
-------
Table S-15. Percent Difference in Costs between Option 8 NSPS and Option 3 BAT, Beef and Dairy Sectors
Sector
Fed Cattle
Dairy
Heifers
Region
MW
CE
PA
MW
CE
MW
Facility
Size
Ml
M2
LI
L2
Ml
M2
LI
L2
Ml
M2
LI
Ml
M2
LI
Ml
M2
LI
Ml
M2
LI
Percent Difference
Category 1
-34%
-39%
-79%
-80%
-16% , ,
-17%
-31%
-35%
1220%
1604%
2346%
918%
1210%
1642%
-18%
-19%
-52%
-33%
-35%
-76%
Category 2
-29%
-34%
-42%
-47%
-11%
-13%
-16%
-13%
1095%
1073%
1143%
878%
967%
1160%
-16%
-16%
-34%
-30%
-36%
-64%
Category 3
-58%
-69%
-89%
-95%
-41%
-53%
-81%
-95%
1484%
1994%
2791%
1070%
1423%
1849%
-33%
-38%
-74%
-49%
-56%
-82%
Source: USEPA. Where percentages are negative, Option 3 NSPS is less expensive than Option 3 BAT. See
Table 4-1 (Section 4) for CAFO model definitions (region and size) and definitions of land availability categories.
5-24
-------
To assess the cost differences shown in Table 5-15, EPA examines the potential economic
impacts to new dairy operations using the same model CAFO approach EPA uses to examine
options for existing sources (described in Section 4). This approach compares NSPS compliance
costs to financial conditions for a representative model CAFO. Potential impacts to new sources
are determined based on the sales test, a post-compliance cash flow analysis, and post-compliance
debt-to-asset ratio analysis for each model CAFO. The results of this analysis indicate whether
the costs will result in affordable or moderate impacts to a facility (considered by EPA to not
result in facility closure) or whether the expected costs will result in financial stress impact (i.e.,
facility or product line closure). Section 4.2.5 provides additional information on this approach.
Where the resulting impacts are "affordable" or "moderate" for the BAT Options but
result in "stress" under the NSPS Option, EPA considers this to be an indication that barriers to
entry may exist (i.e., few, if any, new sources would enter the market). EPA assesses cost-to-
sales ratios to evaluate the magnitude of the cost differences and judge whether this cost
difference is significant in terms of the impact it might have on new operations. If the sales test
result for the NSPS option does not differ greatly from the sales test results for the BAT option,
EPA judges that the cost difference will not pose a barrier to entry.
The results of this analysis indicate that while Option 8 will not result in stress impacts at
beef or heifer operations, all (100 percent) of all dairy operations would experience financial stress
under Option 8 NSPS. The results of this analysis are expressed as percentages, since EPA has
not estimated numbers of new sources. This analysis is conducted assuming that all costs are
incurred at the facility under the zero cost passthrough scenario. Based on these results, EPA
determined that Option 8 would pose a barrier to entry in the dairy sector; therefore, EPA decided
not to propose Option 8 NSPS. More detailed information is available in the docket (DCN
70599).
5.2.4.2 Impacts of the NSPS Options on the Swine, Veal, and Poultry Subcategories
For this proposed rulemaking, EPA has evaluated the proposed option for new sources for
the swine, veal, and poultry subcategories (Option 5+3 NSPS). EPA's analysis indicates that
requiring Option 3 and Option 5 for new sources for the swine, poultry and veal subcategories
would not create any barriers to entry since the estimated costs for new sources are the same as or
less expensive than the BAT costs for existing sources. This determination is based on a
comparison of the costs of Option 5 BAT to the costs of Option 5+3 NSPS on a model-by-model
basis. For this analysis, EPA assumes that new operations will not incur costs estimated under
Option 3 A, which includes groundwater controls, since they are not likely to establish a new
operation where there is a hydrologic link to surface waters (and where operating expenses would
be more costly). Without Option 3A costs, the cost of Option 5+3 NSPS is identical to the cost
of Option 5 BAT. Since the estimated costs for new sources are the same as the costs to existing
facilities, EPA concludes that the proposed NSPS option for these subcategories poses no barriers
to entry in the hog, poultry and veal sectors.
5-25
-------
EPA also conducted additional analyses that evaluate other scenarios, including the
possibility that some new facilities might be sited in hydrologically sensitive areas and may incur
higher costs than those assumed in the analysis presented here. This analysis is provided in the
rulemaking record (ERG, 2000f—-see DCN 70599).
5.3 PROCESSOR IMPACTS
As discussed in Section 4, EPA did not conduct a detailed estimate of the costs and
impacts that would accrue to individual co-permittees due to lack of data arid market information.
However, EPA believes that the framework used to estimate costs to CAFOs provides a means to
evaluate the possible upper bound of costs that could accrue to co-permittees, based on the
potential share of (pre-tax) costs that may be passed on from the CAFO. EPA is proposing that
this amount approximates the magnitude of the costs that may be incurred by processing firms in
those industries that may be affected by the proposed co-permitting requirements. More
information on this approach is provided in Section 4.2.6.
Table 5-16 presents the results of EPA's analysis. This analysis focuses on the potential
magnitude of costs to co-permittees in the pork and poultry sectors only since these are the
sectors where the proposed co-permitting requirements are likely to affect processing facilities.
However, EPA did not evaluate the potential magnitude of costs to egg and turkey processors
because the compliance costs to CAFOs in these industries is projected to be easily absorbed by
CAFOs (see Tables 5-8 through 5-10). The results presented in Table 5-16 are for the pork and
broiler industries only. EPA also did not evaluate the potential costs to cattle and dairy
processors because EPA does not expect that the proposed co-permitting requirements to affect
meat packing and processing facilities in these industries, for reasons outlined in Section 2.4 of
this report.
The potential magnitude of costs to co-permittees is derived from the amount of cost
passthrough assumed in the CAFO level analysis. For this analysis, EPA evaluates two scenarios
of cost passthrough to processors: partial cost passthrough (greater than zero) and also 100
percent cost passthrough. EPA's partial cost passthrough scenario assumes that 46 percent of all
hog compliance costs and 35 percent of all broiler compliance costs are passed on through the
marketing chain. Based on the results of this analysis, EPA estimates that the range of potential
annual costs to hog processors is $135 million (partial cost passthrough) to $306 million (full cost
passthrough). EPA estimates that the range of potential annual costs to broiler processors as $34
million (partial cost passthrough) to $117 million (full cost passthrough). Table 5-16 presents
these results, expressed in 1999 pre-tax dollars. (EPA did not evaluate results for the two-tier
structure at 750 AU threshold, but results for this scenario would be lower than those for the two-
tier structure at 500 AU threshold.)
To assess the magnitude of impacts mat could accrue to processors using this approach,
EPA compares the passed through compliance costs to both aggregate processor costs of
production and to revenues. Table 5-16 presents the results of this analysis, which are presented
5-26
-------
in terms of the assumed costs that are assumed to be passed through, compared to 1997 data from
the Department of Commerce on the revenue and costs among processors in the hog and broiler
industries. As shown, EPA estimates that, even under full cost passthrough, incremental cost
changes are less than two percent, and passed through compliance costs as a share of revenue are
less than one percent. .
ahlp
-------
The following sections present a summary of the key results of EPA's market model and
present predicted changes in farm and retail prices, quantities, national and regional employment,
and national economic output. Results are shown for the two-tier (500 AU threshold) and three-
tier (Scenario 3) structures, along with the results for alternative ELG options considered by EPA
during the development of this rulemaking (expressed across a range). EPA did not evaluate
market impacts under the two-tier structure at the 750 AU threshold, but impacts under this
alternative should be less than those at the 500 AU threshold.
Appendix D provides results of sensitivity analyses, conducted by EPA, to examine the
impact under differing model assumptions. EPA examined variations in the price elasticities and
prices assumed for these industries, based on information reported in the agricultural literature
and statistical compendiums. These sensitivity analyses demonstrate that the results presented
here are stable across a range of possible modeling assumptions.
5.4.1 Changes in Commodity Price and Quantity Production
Tables 5-17 and 5-18 show predicted farm and retail price changes, as compared to pre-
regulation baseline price levels. Results across alternative NPDES scenarios are provided in the
record and are not presented here since they do not differ substantially from the results presented
for each co-proposed alternative. For comparison purposes, the average annual percentage
change in price from 1990 to 1998 is provided.
EPA expects that predicted changes in animal production may raise producer prices, as the
market adjusts to the proposed regulatory requirements. For most sectors, EPA estimates that
producer price changes will rise by less than one percent of the pre-regulation baseline price
(Table 5-17). The exception is in the hog sector, where estimated compliance costs slightly
exceed one percent of the baseline price. Predicted farm level price changes are modest when
compared to the historical year-to-year changes attributable to weather, feed costs, and other
factors.
At the retail level, EPA expects that the proposed regulations will not have a substantial
impact on overall production or consumer prices for value-added meat, eggs, and fluid milk and'
dairy products. EPA estimates that retail price increases resulting from the proposed regulations
will be under one percent of baseline prices in all sectors, averaging below the rate of general
price inflation for all foods (Table 5-18). In terms of retail level price changes, EPA estimates
that poultry and red meat prices will rise about one cent per pound. EPA also estimates that egg
prices will rise by about one cent per dozen and that milk prices will rise by about one cent per
gallon. Results of this analysis do not differ substantially across the range of alternative ELG
options (Tables 5-17 and 5-18).
Table 5-19 summarizes the forecast reductions in farm level production; following a shift
in the supply curve under post-compliance. As shown, predicted quantity reductions are under
one-half of one percent of pre-regulation production levels for all sectors. Other than export and
5-28
-------
import changes, quantity changes at the retail level (not shown) are expected to be directly
proportional to changes at the farm level because the model assumes a fixed proportions
production process. Results of this analysis do not differ substantially across the range of
alternative ELG options (Table 5-23). EPA uses these estimated production changes, multiplied
by the appropriate per-unit market price, to compute the overall change in market value
associated with complying with the proposed CAFO regulations. EPA uses these derived values
as inputs to its input-output analysis framework and allows EPA to compute changes in
employment and economic output under post-compliance.
-17 Pnst-Comnliance Farm Level Price Changes, Selected Regulatory Alternatives
Option
Pre-reg. Avg. Price (1997)
Beef
Dairy
($/cwt)
66.09
13.38
Hogs
54.30
Broilers
(cents/lb.)
37.00
Layers
(cents/doz.)
69.80
Turkeys
(cents/lb.)
40.10
Price Increases
Two-Tier Structure (500 AU Threshold)
BAT Option
Range of Alt. Options
0.21
0.08-0.80
0.06
0.03-0.07
0.59
0.26-0.59
0.19
0.17r0.31
0.13
0.11-0.23
Three-Tier Structure
BAT Option
Range of Alt. Options
0.23
0.08-0.82
0.08
0.04-0.10
0.63
0.25-0.63
0.22
0.19-0.40
Percent Change
Avg. Annual Change (%)
(1990-1998)
4.60
8.00
15.20
5.70
0.14
0.11-0.26
11.50
Two-Tier Structure (500 AU Threshold)
BAT Option
Range of Alt. Options
0.30
0.1-1.2
0.50
0.2-0.5
1.10
0.5-1.1
0.50
0.4-0.8
0.20
0.2-0.3
Three-Tier Structure (500 AU Threshold)
BAT Option ,
Range of Alt. Options
0.30
0.1-1.2
0.60
0.3-0.7
1.20
0.5-1.2
0.60
0.5-1.1
0.20
0.2-0.4
0.12
0.08-0.18
0.16
0.09-0.25
4.40
0.30
0.2-0.4
0.40
0.2-0.6
Source: USEPA, except historical data (pre-regulatory average price and average annual change data) that are from
USDA/ERS, 1999c and 1998b; USD A/WAOB, 1999; and NCBA, 2000.
5-29
-------
Table 5-18. Post-Compliance Retail Level Price Changes, Selected Regulatory Alternatives
Option
Pre-reg. Avg. Price (1997)
Beef
(Mb.)
2.80
Dairy
(index)
145.50
Hogs
($/lb.)
2.45
Broilers
(cents/lb.)
151.00
Layers
(cents/doz.)
106.00
Turkeys
(cents/lb.)
105.10
Price Increases
Two-Tier Structure (500 AU Threshold)
BAT Option
Range of Alt. Options
0.00
0.00-0.02
0.61
0.33-0.68
0.01
0.00-0.01
0.19
0.17-0.31
0.13
0.11-0.23
0.12
0.08-0.18
Three-Tier Structure
BAT Option
Range of Alt. Options
0.00
0.00-0.02
0.78
0.39-0.97
0.01
$0.00-0.01
0.22
0.19-0.40
0.14
0.11-0.26
0.16
0.09-0.25
Percent Change
Avg. Annual Change (%)
(1990-1998)
2.30
2.40
5.10
3.00
7.20
2.40
Two-Tier Structure (500 AU Threshold)
BAT Option
Range of Alt. Options
0.10
0.1-0.6
0.40
0.2-0.5
0'.30
0.2-0.3
0.10
0.1-0.2
0.10
0.1-0.2
0.10
0.1-0.2
Three-Tier Structure
BAT Option
Range of Alt. Options
0.20
0.1-0.6
0.50
0.3-0.7
0.40
0.1-0.4
0.10
0.1-0.3
0.10
0.1-0.2
0.20
0.1-0.2
Source: USEPA, except historical data (pre-regulatory average price and average annual change data)that are from
USDA/ERS, 1999c and 1998b; USDA/WAOB, 1999; and NCBA, 2000.
5.4.2 Changes in Total National Employment
The proposed CAFO regulations are not expected to result in significant changes in
aggregate employment (measured in terms of full-time equivalents (FTEs).8 Losses in
employment are associated with decreases in commodity production in response to higher
compliance costs (Table 5-19). Predicted changes in aggregate employment are measured in
terms of both direct and indirect/induced employment. Direct employment measures the number
of jobs related to production and processing including workers engaged in the manufacture of
agricultural inputs and their suppliers. Other indirect or induced employment provides a broader
measure of industry-related employment and includes workers throughout the economy. More
information is provided in Section 4.4.
sl FTE = 2,080 hours of labor.
5-30
-------
Table 5-19. Post-Comoliance Farm Production Changes, Selected Regulatory Alternatives
Option
Pre-regulatory
Quantity
Beef
47,967
Dairy
Hogs
Broilers
(million pounds)
156,100
23,542
27,551
Layers
(mil. doz.)
6,473
Turkeys
(mil. Ibs.)
5,412
Farm Level Quantity Reduction
Two-Tier Structure (500 AU Threshold)
BAT Option
Range of Alt. Options
65
23-244
193
105 - 217
70
31-70
13
11-21
1
1-1
Three-Tier Structure
BAT Option
Range of Alt. Options
70
24 - 250
248
125 - 308
75
30-75
15
13-27
1
1-2
3
2-5
4
3-7
Source: USEPA, except historical data (pre-regulatory quantity data) that are from USDA/ERS, 1998a, and 19y«b;
USDA/WAOB, 1999; and Putnam and Allshouse, 1999.
Absorption of compliance costs by the producers and declines in production and trade are
expected to result in fewer jobs in the livestock and poultry industries. Tables 5-20 and 5-21
present EPA's estimates of both the direct (i.e., farm and processor level) and total (i.e., national
level) employment losses across all sectors. Total direct farm level employment is expected to
drop by 2,700 FTEs to 3,000 FTEs, depending on tier structure. These estimates include CAFO
owner-operator job losses due to business closure. Farm level employment losses are greatest in
the hog sector, coinciding with EPA's estimate that this sector would incur over one-third of
estimated total compliance costs (Section 5.1.1). Predicted direct employment losses in the food
processing sector constitute a small share of overall employment losses, estimated at less than 500
FTEs under either co-proposed scenario (less than 3 percent), compared to pre-regulatory
baseline conditions. Compared to total employment in the farm and processing sectors of these
industries (see Section 2.5.3), employment losses are modest.
The remaining approximately 80 percent of predicted job losses are estimated to occur as
a result of indirect or induced effects on nationwide employment (Tables 5-20 and 5-21). These
additional losses occur outside the affected livestock and poultry sectors and include loses in
those industries that support the agricultural community. These predicted job losses are likely to
be offset by stimulated employment in other sectors throughout the economy, such as the
construction and farm services sectors. Potential offsets in terms of gains in employment to other
sectors were not evaluated by EPA.
The total reduction in aggregate national level employment in all sectors of the economy
(both direct and indirect employment) is estimated at 16,600 FTEs under the two-tier structure
and 18,100 FTEs under the three-tier structure (Tables 5-20 and 5-21). These predicted
employment losses represent shifts in employment requirements from one industry to another.
5-31
-------
Compared to the 129.6 million people employed in 1997 (Council of Economic Advisers, 2000),
this loss is less than one-hundredth of one percent of national level employment. Under the
alternative technology options considered by EPA, estimated total employment losses that range
from 8,400 FTEs (Option I/Scenario 4a) to 31,600 FTEs (Option 5/Scenario3), compared to pre-
regulatory baseline conditions (Tables 5-20 and 5-21).
Table 5-20. Post-Compliance Total National Employment Changes, Two-Tier Structure (500 AU Threshold)
Option
Beef
Dairy
Hogs
Poultry
Total *
(FTEs)
Direct CAFO Level Employment Reductions
BAT Options
Range of Alt. Options
793
284 - 2,969
492
268 - 554
931
416-931
449
373 - 730
2,666
1,358-4,816
Direct Wholesale/Processing Employment Reductions
BAT Options
Range of Alt. Options
109
39-410
19
11-22
250
111-250
69
59-114
448
221 - 747
Consumer/Indirect/Induced Employment Reductions
BAT Options
Range of Alt. Options
3,697
1,325 - 13,839
2,688
1,465 - 3,028
5,195
2,322-5,195
1,921
1,622-3,156
13,501
6,827 - 23,502
Total Employment Reductions
BAT Options
Range of Alt. Options
4,599
1,648 - 17,218
3,200
1,744-3,604
6,376
2,849 - 6,376
2,439
2,053 - 4,000
16,615
8,406 - 29,066
Source: USEPA's market model results, in conjunction with RIMS II multipliers (USDC, 1997b).
Totals may not add due to rounding and may include double counting since each sector is modeled separately.
Note: Total employment in 1997 was 129.6 million (Council of Economic Advisors, 2000).
5.4.3 Changes in Total National Economic Output
EPA does not expect that the proposed regulations will result in significant changes in
aggregate employment or national economic output, measured in terms of Gross Domestic
Product (GDP). EPA expects, however, that there will be losses in employment and economic
output associated with decreases in animal production due to rising compliance costs. These
losses are estimated throughout the entire economy, using available modeling approaches, and are
not attributable to the regulated community only. As with estimated employment losses, the
estimated changes in economic output do not account for any possible increases in spending in
those sectors of the economy supplying the goods and services needed to meet regulatory
requirements. Net output losses are expected to be minimal.
5-32
-------
Table 5-22 shows EPA's predicted changes in total gross output (not including gains in
other sectors) attributable to complying with the proposed regulations. Compared to 1997
baseline levels, EPA estimates reductions in national economic output that range from $1.7 billion
to $1.9 billion under the two-tier and three-tier structure, respectively (Table 5-26). This
projected change is modest when compared to total GDP, estimated at $8.3,trillion in 1997
(Council of Economic Advisors, 2000). Under the alternative technology options considered by
EPA, estimated GDP losses range from $0.8 billion (Option I/Scenario 4a) to $3.1 billion (Option
5/Scenario3) (if applicable to all sectors), compared to pre-regulatory baseline conditions.
Pnst-Comnliance Total National Employment Changes, Three-Tier Structure
Option
Beef
Dairy
Hogs .
Poultry
Total
(FTEs)
Direct CAFO Level Employment Reductions
BAT Options
Range of Alt. Options
BAT Options
Range of Alt. Options
850
294 - 3,045
633
319-784
1,005
402-1,005
552
442-954
Direct Wholesale/Processing Employment Reductions
117
41 - 421
25
13-31
269
108 - 269
84
70 -148
Consumer/Indirect/Induced Employment Reductions
BAT Options
Range of Alt Options
3,963
1,370 - 14,195
3,459
1,743-4,283
5,602
2,242 -,5,602
2,329
1,903 - 4,071
Total Employment Reductions
BAT Options
Range of Alt Options
4,929
1 704 - 17,661
4,117
2,075 - 5,099
6,876
2,752 - 6,876
2,966
2,415 - 5,173
3,040
1,462 - 5,235
496
231-799,
15,353
7,290 - 25,583
18,889
8,983 - 31,617
Source: USEPA's market model results, in conjunction with RMS II multipliers (USDC, 1997b).
Totals may not add due to rounding and may include double counting since each sector is modeled separately.
Note: Total employment in 1997 was 129.6 million (Council of Economic Advisors, 2000).
5-33
-------
Table 5-22. Total National Gross Output Reductions, Selected Regulatory Alternatives
Option
Beef
Dairy
Hogs
Poultry
Total
($1997 million)
Two-Tier Structure (BAT Option/Scenario 4a)
BAT Options
Range of AIL Options
458
164-1,715
296
161-333
655
293 - 655
242
. 204 - 397
1,651
832-2,893
Three-Tier Structure (BAT Option/Scenario 3)
BAT Options
Range of Alt Options
491
170-1,760
381
192- 472
707
283 - 707
295
240 - 513
1,874
888 - 3,141
Source: USEPA's market model results, in conjunction with RMS II multipliers (USDC, 1997b).
Totals may not add due to rounding and may include double counting since each sector is modeled separately.
Note: U.S. Gross Domestic Product in 1997 was $8,300.8 billion (Council of Economic Advisors, 2000).
5.4.4 Other Market Impacts
This section presents the results of EPA's analyses to examine other market effects that
may occur as a result of the proposed CAFO regulations. Estimated impacts include changes in
regional level employment (used as a measure to evaluate community level impacts) and also
changes in international trade (evaluated in terms of U.S. import and export volumes).
5.4.4.1 Regional Employment
For this analysis, EPA examines the potential impacts to the agricultural community by
assessing whether the proposed CAFO regulations could have community or regional level
impacts, particularly in the agricultural sectors. Such impacts could alter the competitive position
of livestock and poultry production across the nation or lead to growth or reductions in farm
production (in- or out-migration) in different regions and communities. Ongoing structural and
technological change in these industries has influenced where farmers operate and has contributed
to locational shifts between the more traditional production regions and the emerging,
nontraditional regions. Production is growing rapidly in the non-traditional regions due to
competitive pressures from more specialized, lower cost producers. This is especially true in hog
and dairy production (El-Osta and Johnson, 1998; McBride, 1999; Iowa State University, 1998;
Martinez, 1999).
EPA does not expect that the proposed CAFO regulations would have a significant impact
on where animals are raised. On the one hand, on-site improvements in waste management and
disposal, as required by this regulation, could accelerate recent shifts in production to more
5-34
-------
nontraditional regions (such as the South and Pacific and parts of the Southeast) as higher-cost
producers in some regions exit the market to avoid relatively higher retrofitting costs of existing
facilities.. On the other hand, the proposed regulations may favor more traditional production
systems where operators grow both livestock and crops, since these operations tend to have
available cropland for land application of manure nutrients. These types of operations tend to be
more diverse and generally smaller in size. Long-standing farm services and input supply
industries in these more traditional areas (such as the Midwest and Mid-Atlantic) could likewise
benefit from the proposed CAFO regulations given the need to support on-site improvements in
manure management and disposal. Local and regional governments, and other nonfarm
enterprises, could also benefit.
The market model is national in scope and cannot address specific subregions. For this
analysis, EPA approximates changes in regional employment by disaggregating the national
employment reductions (both direct and indirect) to indicate regions where impacts of the
proposed CAFO regulations are most likely to occur. For each animal sector, EPA distributes Hie
national direct employment impacts at the CAFO and processor/wholesale level onto a regional .
basis using state level data that reflect livestock and poultry production by the largest facilities in
each state. These data include the 1997 Census (USDA/NASS, 1999a) and other USDA data
(USDA/NASS, 1998b), with imputed values for omitted USDA data (Westat, 2000). EPA
allocates total national indirect and induced employment changes onto a regional basis according
to each state's share of 1997 total U.S. population as reported for 1997 from U.S. Census data
(U.S. Census Bureau, 1999). The direct and indirect/induced results are summed to show the
total impact for each state. EPA evaluates regional impacts from these' state level estimates,
aggregated to USD A's farm producing regions, shown in Figure 4-1.
• Table 5-23 breaks out the estimated regional employment impacts between direct (farm
and processing level) and indirect/induced (other economy-wide) job losses. As shown, EPA
estimates direct employment losses to be greatest in the Midwest region given the sheer volume of
animal production in "the region, which includes the Dakotas, Nebraska, and Kansas as well as the
Corn Belt and Lake states. In the Midwest region, estimated direct job loses range from 1,300 to
or 1,400 jobs, depending on tier structure, which is about one-half of the direct job losses
estimated nationwide following compliance with the proposed regulations. Estimated job losses
include CAFO owner-operator job losses due to business closure. Including the estimated
indirect and induced employment impacts, overall job losses are more evenly distributed among
the regions and are greatest in the Mid-Atlantic, which covers 'areas with both high consumer
populations and concentrated hog and poultry operations in North Carolina, Virginia, and the
Delmarva Peninsula. Based on these results, EPA concludes that more traditional agricultural
regions would not be disproportionately affected by the proposed CAFO regulations.
5.4.4.2 International Trade
As part of its market analysis, EPA evaluates the potential for changes in U.S. trade
(imports, exports) of meat, eggs, and dairy products. Foreign trade impacts are difficult to
predict, since exports are determined by economic conditions in foreign markets and changes in
5-35
-------
the international exchange rate for the U.S. dollar that cannot be incorporated into a simple
market model. Nevertheless, predicted higher prices could attract more imports and discourage
exports with the net effect of the proposed regulations resulting in an increase in U.S. imports and
a corresponding decrease in U.S. exports of meat, eggs, and dairy products.
Table 5-23. Regional Distribution of Predicted National Employment Reductions
Region ^
Agricultural
Sectors Direct
Indirect/
Induced
Total
Percent of
Labor Force
(FTEs)
Two-Tier Structure (BAT Option/Scenario 4a)
Pacific
Central
Midwest
South
Mid-Atlantic
Total
346
690
1,289
323
466
3,114
2,160
1,976
3,157
2,011
4,196
13,501
2,506
2,666
4,446
2,334
4,662
16,614
0.012%
0.013%
0.013%
0.012%
0.011%
0.012%
Three-Tier Structure (BAT Option/Scenario 3)
Pacific
Central
Midwest
South
Mid-Atlantic
Total
432
779
1,405
390
530
3,536
2,457
2,247
3,590
2,287
4,771
15353
2,888
3,026
4,995
2,677
5,302
18,888
0.013%
0.015%
0.015%
0.014%
0.013%
0.014%
Source: USEPA's market model results, in conjunction with RIMS II multipliers (USDC, 1997b). State level
employment data are from the U.S. Census Bureau (1999).
Totals may not add due to rounding and may include double counting since each sector is modeled separately.
Regions are based on the USD A Farm Production Regions (see Figure 4-1): Pacific=Pacific, Central=Mountain
and Southern Plains, Midwest?=Corn Belt, Lake States, and Northern Plains, South=Delta and Southeast, Mid-
Atlantic=Northeast and Appalachia.
Table 5-24 summarizes the impacts on retail level trade forecast from EPA's market model. The
results of this analysis show that U.S. trade will not be significantly impacted by the proposed
CAFO regulations. EPA estimates that U.S. imports (exports) will increase (decrease) by less
than one percent compared to baseline (pre-regulation) levels in each of the commodity sectors.
By sector and by co-proposed alternative, the potential change in imports compared to baseline
trade levels ranges from a 0.02 percent increase in broiler imports to a 0.82 percent increase in
dairy product imports. The predicted drop in U.S. exports ranges from a 0.01 percent reduction
5-36
-------
in turkey exports to a 0.27 percent reduction in hog exports. Baseline information on U.S.
imports and exports of livestock and poultry products is presented in Section 2.5.
Tahle f-24 Post-Corn oliance Retail Product Import and Export Changes, Selected Regulatory Alternatives
Option
Beef
Dairy
Hogs
Broilers
Layers
Turkeys
(percent)
Percentage Increase in Imports
Two-Tier Structure (500 AU Threshold)
BAT Option
Range of Alt. Options
0.15
0.05 - 0.56
0.64
0.35 - 0.72
0.21
0.09 - 0.21
0.02
0.02 - 0.04
0.12
0.09 - 0.20
Three-Tier Structure
BAT Option
Range of Alt. Options
0.16
0.06 - 0.58
0.82
0.41 - 1.01
0.23
0.09 - 0.23
0.03
0.03 - 0.05
0.13
0.10-0.23
NA
NA
NA
NA
Percentage Decrease in Exports
Two-Tier Structure (500 AU Threshold)
BAT Option
Range of Alt. Options
0.09
0.03 - 0.34
0.10
0.06 - 0.12
• 0.25
0.11-0.25
0.05
o;o4 - o.os
o.oi
0.01 - 0.02
Three-Tier Structure
BAT Option
Range of Alt. Options
0.10
0.03 - 0.35
0.13
0.07 - 0.16
0.27
0.11-0.27
0.06
0.05 - 0.10
0.01
0.01 - 0.03
0.06
0.04 - 0.09
0.08
0.05-0.13
Source: USEPA, except historical data that are from Putnam and Allshouse, 1999. NA = Not applicable.
5-37
-------
-------
SECTION SIX
SUMMARY OF ECONOMIC IMPACTS:
POULTRY SUBCATEGORIES
This section presents a profile of the poultry industry, including farmers in the broiler, egg,
and turkey sectors (Section 6.1) and also poultry processors (Section 6.2). Following the
industry profile, this section provides a detailed summary of EPA's economic analysis of the
proposed CAFO regulations as it affects regulated CAFOs (Section 6.3), poultry processors
(Section 6.4), and national markets (Section 6.5).
6.1 PROFILE OF THE POULTRY PRODUCTION INDUSTRY
This section presents a profile of poultry production operations (broiler, egg, and turkey
operations) and provides background information for analyzing the costs and benefits of the
proposed CAFO regulations. The purpose of this profile is to provide a baseline description of
the current activities, structure, and performance of the poultry production industries. The
following sections describe the types of operations in this sector and present an overview of the
industry, describing the number and size of operations (including the subset of regulated
operations), geographic distribution, supply and demand conditions, price trends, and the financial
conditions that characterize this sector.
6.1.1 Industry Definition .
Poultry operations can be classified into three individual sectors based on the type of
commodity in which they specialize. These sectors include operations that breed and/or raise:
• Broilers or young meat chickens that are raised to a live weight of 4 to 4.5 pounds
and other meat-type chickens, including roasters that are raised to 8 to 9 pounds.
Classification: NAICS 11232, broilers and other meat-type chickens (SIC 0251,
broiler, fryer and roaster chickens).
• Hens that lay shell eggs, including eggs that are sold for human consumption and
eggs that are produced for hatching purposes. Classification: NAICS 11231,
Chicken egg production (SIC 0252, chicken eggs) and NAICS 11234, poultry
hatcheries (SIC 0254, poultry hatcheries).
• Turkeys and turkey hens, including whole turkey hens that range from 8 to 15
pounds at slaughter, depending on market, and also turkey "canners and cut-ups"
that range from 22 to 40 pounds. Classification: NAICS 11233, turkey production
(SIC 0253, turkey and turkey eggs).
6-1
-------
Beyond the poultry production sector, manufacturing and further processing are
conducted by firms classified under NAICS 311615, Poultry Slaughtering, as well as an array of
food processing sectors. Egg processors are classified under NAICS 311999(G), Liquid, dried,
and frozen eggs. Almost 70 percent of all egg production is sold in fresh form to retail stores or
to institutional buyers. See Section 6.2 for more information about poultry processors.
6.1.2 Overview of the Poultry Industry
The poultry market is among the most robust of all the U.S. livestock industries, showing
increasing production and efficiency gains, ongoing strong growth in both domestic consumer
demand and exports, continued favorable farm prices and returns, and increasing economies and
enhanced efficiency within the poultry sector marketing chain. This is especially true of the
poultry meat markets. The U.S. poultry industry is characterized by its rapid rates of growth and
emphasis on increased industriaKzation, specialization, and consolidation. Historically, the poultry
industry has been a highly integrated industry, with operations combining breeding, hatching, and
growing functions, as well as grain farming and feed operations. Increased production efficiencies
have generally favored larger operations and have encouraged the continued emphasis on
specialization (Kohls and Uhl, 1998). Consequently, poultry meat and egg production has
become concentrated among fewer, larger producers, as evidenced by a reduction in the number
of poultry operations but an increase in the average size of these operations. Another major trend
in this industry is a trend away from traditional areas of operation to emerging areas of operation
where costs of production are lower.
6.1.2.1 Trends in the Number and Size
USDA reports that in 1997 there were a total of 63,200 commercial poultry farms in'the
United States, based on annual sales (USDA/NASS, 1999a). Table 6-1 shows USDA's estimate
of the number of farms in the broiler, turkey, and egg sectors. In 1997, there were nearly 24,000
broiler operations based on annual sales. At year-end 1997, there were more than 12,000 turkey
operations and 72,600 egg laying operations, based on inventories (Table 6-1). These data on the
number of farms include both commercial and non-commercial operations, as well as confinement
and non-confinement operations.-
Table 6-1 shows broad trends based on data on the number of poultry operations and
corresponding total number of animals by facility size categories for selected years between 1974
and 1997. As shown, the number of poultry farms is declining, hi the broiler sector, the number
of operations declined from 34,340 operations in 1974 to 24,000 operations in 1997, a 30 percent
decline in the number of operations (Table 6-1). In the turkey sector, the number of operations
decreased from 13,000 to 12,000 between 1974 and 1997. The number of egg laying operations
dropped from 316,200 operations in 1974 to 72,600 operations by 1997, a reduction of 77
percent (Table 6-1).
6-2
-------
Table 6-1. Trends in Number of Poultry Operations and Birds, 1974-1997
Year
Broiler
1974
1978
1982
1987
1992
1997
Layer
1974
1978 .
1982
1987
1992
1997
Turkey
1974 ^
1978 "
1982
1987
1992
Operations
34,340
31,743
30,100
27,645
23,949
23,937
316,243
240,891
215,812
144,438
88,235
72,616
12,787
18,936
25,366
19,031
13,766
Birds
(million)
2,519
3,062
3,517
4,362
5,429
6,742
336
354
362
374
351
367
Flock Size
(number)
73,400
96,500
116,800
157,800
226,700
281,700
1,100
1,500
1,700
2,600
4,000
5,100
27
36
47
74
88
104
2,100
1,900
1,900
3,900
6,400
8,600
Percent of
Operations
Percent of
Birds
(>100million birds [sales])
30%
38%
44%
52%
65%
68%
70%
82%
89%
93%
97%
98%
(>100K birds [inventory])
0.1%
0.1%
0.2%
0.3%
0.7%
0.8%
30.1%
31.9%
41.6%
54.0%
62.2%
69.2%
(>100K birds [inventory])
12%
10%
11%
17%
23%
28%
61%
66%
72%
76%
• 80%
83%
O/. ^rrai -y -f) .J - ~ - 5 . _. .
Average flock size per operation is computed from the USDA shown, rounded to the nearest hundred.
^Turkey data for 1974 and 1978 imputed from sales data.
Meanwhile, overall poultry production and sales have continued to rise steadily. Although
the number of broiler operations has decreased, total sales from all U.S. farms rose from an
estimated 2.4 billion broilers in 1974 to 6.7 billion broilers in 1997, nearly a three-fold increase.
Year-end inventories at turkey operations rose from 27 million turkeys in 1974 to 104 million
6-3
-------
turkeys in 1997, or nearly quadrupling over that time period.1 Year-end laying hen inventories
have increased from 336 million in 1974 to 367 million in 1997, a 9 percent increase.
Increasing production is due to increasing flock sizes and production efficiency gains in
these sectors. Across these sectors, average flock size per operation in 1997 was nearly four to
five times that in the 1970s. Among broiler operations, average flock size rose from 73,400 birds
in 1974 to 281,700 birds in 1997 (Table 6-1). At turkey operations, average flock size increased
from 2,100 birds in 1974 to 8,600 birds in 1997. Average flock size at egg laying operations rose
from 1,100 birds in 1974 to 5,100 birds in 1997.
The trend toward large farm size in these sectors is also indicated by data in Table 6-1. In
the broiler sector, the proportion of operations selling more than 100,000 birds has grown from
about 30 percent in 1974 to more than 68 percent in 1997. These larger sized operation
accounted for almost 98 percent of total broiler sales, up from 70 percent in 1974 (Table 6-1). In
the turkey sector, animal inventories and production are heavily concentrated among the larger
sized turkey operations: operations with more than 100,000 turkeys account for about 83 percent
of total inventory. In 1974, about 60 percent of turkey inventories were associated with larger-
sized operations. In the egg sector, operations account for almost 70 percent of overall animal
inventory in 1997, up from 30 percent in 1974 (Table 6-1).
For the purpose of this analysis, EPA estimates the number of confinement operations that
may be subject to the proposed CAFO regulations using 1997 Census data that are aggregated by
USDA's NASS. NASS developed a methodology for identifying farms likely to be CAFOs based
on the Census survey information and estimated animal units on these operations based on
reported data. A summary of these data are provided in the Development Document (USEPA,
2000a). These summary data reflect average flock size throughout the year, accounting for both
animals sales and inventories. Where applicable, data are adjusted for the average number of
marketing cycles (USEPA, 2000a). This avoids misrepresentation due to seasonal fluctuations in
inventory and the number and timing of animals sold. From these data, EPA has estimated the
number of confinement operations (referred to here as AFOs) using available data and other
information from the Census as well as other USDA and industry publications (USDA/NASS,
1999a, 1999d, and 1998b). These data may differ from that presented in Table 6-1.
Expressed on this basis, USDA estimates that there were the USDA reports that there
were 34,860 broiler operations that raised a total of 1.9 billion broilers during the year
(Table 6-2). There were also 13,720 turkey operations raising a total 112.8 million turkeys.
Operations with egg layers and pullets totaled 75,170 with an average annual inventory of 393
million egg layers on-site. Not all of these operations would be subject to the proposed
regulations. See Table 6-2.
Given that inventories are reported as of December 31 for each census year and given the proximity of
this reporting time frame to the Thanksgiving slaughter period, inventory numbers reported by U.S. Census may
be, in part, explained by market conditions for a particular year in addition to structural changes in the farm sector.
6-4
-------
Table 6-2 presents EPA's estimates of the number of operations that are CAFOs that
would be subject to the proposed regulations. Under the two-tier structure, EPA estimates that
mere are 9,780 broiler operations, 1,280 turkey operations and 1,640 egg laying and pullet
operations that have more than 500 AU (i.e., operations with more than 50,000 chickens or more
than 27,500 turkeys). Under the three-tier structure, EPA estimates that 13,740 broiler
operations, 2,060 turkey operations and 2,010 egg laving operations with more than 300 AU (i.e.,
operations with more than 30,000 chickens or more than 16,500 turkeys) would meet the "risk-
based" conditions described in Section VII of the preamble and thus require a permit (Table 6-2).
(More information on the co-proposed tier structures is provided in Section 3.)
Table 6-2. EPA's Estimate of the Number of CAFOs Affected under the Co-Proposed Tier Structures
Sector
Broilers
Layers-wet
Layers-dry
Turkeys
Sum Total
Total
Number
ofAFOs
34,860
3,110
72,060
13,720
123,750
Number of CAFOs
>1,000
AU." .
3,940
50
590
370
4,950
Two-Tier Structure
(500 AU Threshold)
500-1,000
AU
5,840
310
690
910
7,760
<500 AU
20
20
0
0
40
Total
CAFOs
9,800
380
1,280
1,280
12,740
Three-Tier Structure
(Scenario 3)
300-1,000
AU
9,800
310
1,060
1,690
12,860
<300AU
0
0
0
0
0
Total
CAFOs
13,740
360
1,660
2,060
17,820
Source: USEPA, 2000a. See Section 2 for more information. See Table 3-1 for definitions of the
options/scenarios.
"Layers: wet" are operations with liquid manure systems; "Layers: dry" are operations with dry systems. The
number of operations shown eliminates double counting of operations with mixed animal types.
"'As defined for the proposed regulations, one AU is equivalent to 55 turkeys and 100 chickens regardless of the
animal waste system used. •
' EPA expects few, if any, poultry AFOs with fewer than 500 AU will be subject to the
revised requirements. Most poultry operations have fewer than 500 AU (USDA/NASS, 1999a).
Under the two-tier structure, EPA expects that designation of broiler operations with fewer than
50,000 chickens will be limited to two broiler and two egg operations being designated annually,
or a total of 40 poultry operations over a 10-year period. EPA expects that no turkey operations
would be designated as CAFOs and subject to the proposed regulations. EPA expects that no
confinement poultry operations will be designated as CAFOs under the proposed requirements
under the three-tier structure (Table 6-2).
As shown in Table 6-2, EPA estimates that a total of 9,800 broiler operations, 1,660 layer
operations, and 1,280 turkey operations would either to be defined (>500 AU) or designated
6-5
-------
(<500 AU) as CAFOs under the two-tier structure. A total of 13,740 broiler operations, 2,020
layer operations, arid 2,060 turkey operations would be defined as CAFOs under the three-tier
structure. EPA does not expect poultry operations with fewer than 300 AU to be designated as
CAFOs under the three-tier structure. These estimates adjust for operations with more than a
single animal type.
More information on how EPA estimated the number of affected animal confinement
operations is presented in Section 2 of this report, along with additional estimates on the number
of affected poultry operations under other regulatory options considered by EPA.
6.1.2.2 Geographic Distribution
Poultry production is almost entirely managed through contract production, with
operations located near poultry integrators who continue to develop newer, larger, more
automated hatcheries, processing plants, distribution centers, and water treatment plants. Egg
production has expanded mostly in the North Atlantic states (Kohls and Uhl, 1998). Broiler and
turkey production has shifted from the northern to the southern states (Kohls and Uhl, 1998).
The Southeast offers a number of cost advantages compared with other producing regions,
including its relatively lower labor costs, proximity to end markets, lower housing and energy
costs, and milder weather contributing to greater feed efficiency (USGAO, 1995; NCSU, 1998;
Kohls and Uhl, 1998). Poultry production and feedgrain production are closely interrelated;
however, it has become cheaper to transport surplus grains from surplus-producing areas in the
north to low-cost poultry producing areas than to raise birds near grains. The seasonality of
poultry production has been reduced as production has shifted to warmer climates and as use of
confinement production has become more prevalent (Kohls and Uhl, 1998).
Nearly 60 percent of all broiler production is concentrated among the top five producing
states. In 1997, Georgia and Arkansas were the largest broiler producing states, each
representing about 15 percent of all broiler meat production (Table 6-3). Alabama accounted for
another 12 percent of production. Mississippi and North Carolina were also among the top five
producing states, each accounting for about 9 percent of U.S. production. Other top ten
producing states in 1997 included Texas, Maryland, Virginia, Delaware, and Missouri (Table 6-3).
Combined, the top ten producing states accounted for 79 percent of U.S. broiler production in
1997 (Table'6-3).
U.S. egg production is fairly evenly distributed among the top five producing states, with a
combined market share of 42 percent of all egg production in 1997. These five leading states
included Ohio, California, Pennsylvania, Indiana, and Iowa, with production shares ranging from
9.5 percent to 7.5 percent each (Table 6-4). Other top ten producing states in 1997 included
Georgia, Texas, Arkansas, Minnesota, and North Carolina. The top ten states accounted for 66
percent of egg production in 1997. The top 20 states represented 90 percent of all production
(Table 6-4).
6-6
-------
Table 6-3. Geoffranhic Distribution of Broiler Operations by Major Producing State, 1997
Major Producing
State
Georgia
Arkansas
Alabama
Mississippi
North Carolina
Texas
Maryland
Virginia
Delaware
Missouri
California
Oklahoma
South Carolina
Tennessee
Pennsylvania
Florida
Kentucky
West Virginia
Minnesota
Ohio
AllOther
Top 5 states
Top 10 states
Top 20 states
Total U.S.
Broilers Produced
(thousands)
1,182,800
1,164,600
906,200
720,300
665,000
455,100
295,300
259,400
256,900
250,000
237,300
197,400
182,800
138,600
135,200
132,400
110,600
90,800
46,300
45,800
287,000
4,638,900
6,155,600
7,472,800
7,760,200
(percent)
15%
15%
12%
9%
9%
6%
4%
3%
3%
3%
3%
3%
2%
2%
2%
2%
1%
1%
1%
1%
4%
60%
79%
96%
100%
Operations Reporting Sales
(number)
2,245
3^650
2,477
1,393
2,086
1,000
997
671
805
451
240
632
366
548
845
321
243
186
621
496
3,664
. 11,851
15,775
20,273
23,937
(percent)
9%
15%
10%
6%
9%
4%
4%
3%
3%
2%
1%
3%
2% .
2%
4%
1%
1%
1%
3%
2%
15%
50%
66%
85%
100%
Source: USDA/NASS, 1998f andUSDAMASS, 1999a. (Farms reporting sales).
6-7
-------
Table 6-4. Geographic Distribution of Layer Operations by Major Producing State, 1997
Major Producing
State
Ohio
California
Pennsylvania
Indiana
Iowa
Georgia
Texas
Arkansas
Minnesota
North Carolina
Alabama
Florida
Nebraska
Missouri
Mississippi
Maine
Washington
Michigan
South Carolina
Wisconsin
All Other
Top 5 states
Top 10 states
Top 20 states
Total U.S.
Eggs Produced
(million)
6,976
6,663
5,900
5,652
5,527
4,867
4,186
3,215
2,957
2,794
2,499
2,499
2,469
1,719
1,547
1,434
1,379
1,327
1,228
998
7,633
30,718
48,737
65,836
• ' 73,469
(percent)
9%
9%
8%
8%
8%
7%
6%
4%
4%
4%
3%
3%
3%
2%
2%
2%
2%
2%
2%
1%
10.4% .
42%
66%
90%
100%
Operations Reporting Inventory
(number)
3,190
2,731
3,259
1,846
1,892
1,295
6,473
1,835
1,964
1,726
1,250
1,203
1,506
3,707
941
554
1,543
2,276
730
2,543
30,152
12,918
26,211
42,464
72,616
(percent)
4%
4%
4%
3% , .
3%
2%
9%
3%
3%
2%
2%
2%
2%
5%
1%
1%
2%
3%
1%
4%
42%
18%
36%
58% -
100%
Source: USDA/NASS, 1998f and USDA/NASS, 1999a. (Farms reporting year-end inventory of layers and pullets
13 weeks old and older).
6-8
-------
The top five turkey producing states account for almost 60 percent of all turkeys sold
commercially. North Carolina was the largest producing state with 18 percent of the market in
1997 (Table 6-5). Minnesota was the second largest producer, accounting for 15 percent of sales.
Virginia, Arkansas, and California—each with roughly 8 percent of total sales—were also among
the top five producing turkey states. Other top ten producing states in 1997 included Missouri,
Indiana, South Carolina, Texas, and Pennsylvania. The top 10 producing states accounted for
more than 80 percent of turkey production (Table 6-5).
Table 6-5. Geosraohic Distribution of Turkey Operations by Major Producing State, 1997
Major Producing
State
STorth Carolina
Minnesota
Virginia
Arkansas
California
Missouri
Indiana
South Carolina
Texas
Pennsylvania
towa
Michigan
Ohio
West Virginia
South Dakota
Illinois
North Dakota
Oklahoma
Kansas
Maryland
AllOther
Top 5 states
Top 10 states
Top 20 states
Total U S
Turkeys Sold
(thousands)
56,471
47,185
26,031
25,454
23,552
21,085
13,685
13,504
12,767
10,702
7,280
6,481
6,469
4,468
3,566
3,160
2,624
1,749
1,617
751
18,986
178,693
250,435
288,600
307,587
(percent)
18%
15%
8%
8%
' 8%
7%
4%
4%
4%
3%
2%
2%
2%
1%
1%
1%
1%
1%
1%
0%
6%
58%
81%
94%
100%
Operations Reporting Sales
(number)
733
359
389
289
211
402
259
168
215
304
206
241
281
80
51
109
29
65
62
49
1,529
1,981
3,329
4,502
6,031
(percent)
12%
6%
6%
5%
3%
7%
4%
3%
4%
5%
3%
, 4%
5%
1%
1%
2%
0%
1%
1%
1% .
25%
33%
55%
75%
,100%
Source: USDA/NASS, 1999a. (Turkeys sold and farms reporting sales).
6-9
-------
6.1.2.3 Supply and Demand Conditions
The poultry industry experienced large gains in production in the past decade, especially in
broiler/chicken meat and turkey production. These gains were driven, in part, by continued
strong domestic and also international demand growth. Poultry meat consumption in the United
States totaled more than 100 pounds per person in 1997, more than double that reported for
1970. Increased demand for poultry meat products abroad has also helped boost production,
particularly turkey exports. The U.S. is a net exporter of poultry and the dominant world supplier
(USDA/WAOB, 1999). Table 6-6 shows terns in poultry meat and eggs from 1992 through
1997, as reported by USDA (Putnam and AUshouse, 1997 and 1999).
U.S. broiler and chicken meat production continues to increase steadily, rising nearly 30
percent between 1992 and 1997. Demand for broiler and other chicken is also increasing, rising
15 percent between 1992 and 1997 (Table 6-6). Expressed on a.per capita basis, demand for
broilers rose from 76.9 pounds per person in 1992 to 83.8 pounds per person in 1997. U.S.
broiler and chicken meat exports account for almost one-fifth of total production annually. From
1992 to 1997, broiler exports nearly tripled from 1.7 billion pounds to 5.0 billion pounds
(Table 6-6). U.S. imports of broiler and chicken meat are low and account for less than one
percent of total supplies.
U.S. egg production has increased steadily each year .since 1992, reaching 6.4 billion
dozen eggs produced in 1997 (Table 6-6). Aside from a spike in per capita demand to above 240
eggs per person per year in 1997 (reflects both whole shell egg demand and use in processed
foods), egg demand has more or less stabilized at about 235 eggs per person per year (with some
annual fluctuations). While the market conditions for eggs remain strong, aggregate demand for
eggs is down compared to 1970s when demand per person was 300 eggs per person per year
(Putnam and Allshouse, 1997 and 1999). The United States exports more than 200 million dozen
eggs annually, accounting for more than 3 percent of total production. Egg exports increased
more than 25 percent during the 1990s, rising from 175 million dozen eggs exported in 1992 to
220 million dozen eggs exported in 1997 (Table 6-6). Egg imports remain negligible.
Turkey production (carcass weight basis) rose 13 percent from 4.8 billion pounds to 5.4
billion pounds (Table 6-6). During the same period, total domestic demand for turkey meat
increased slightly to 4.7 billion pounds in 1997. Expressed on a per capita basis, however,
demand for turkey products dropped by nearly 2 percent from 1992 to 1997. Since 1990, annual
demand growth has been more or less flat, and the demand for turkey meat appears to have
reached a plateau at roughly 18 pounds per person annually (Table 6-6). Turkey exports
constitute an increasing share of U.S. turkey production. Turkey exports more than tripled
between 1992 and 1997, increasing from 186 million pounds to 598 million pounds (Table 6-6).
Currently, exports account for more than 10 percent of total annual production. U.S. turkey
imports are negligible, and almost all U.S. demand is supplied domestically.
6-10
-------
Table 6-6. Total U.S. Poultry Supply and Demand, 1992-1997
Year
Production
Imports
Exports
Total Demand
Per Capita
Demand
Broiler and Chicken Meat
1992
1993 '
1994
1995
1996
1997
%92-97
(million pounds ready-to-cook, carcass weight)
21,423
22,530
24,175
25,323
26,615
27,551
29%
—
—
1
4
4
5
NA
1,732 .
2,174
2,966
3,993
4,685
5,048
192%
19,624
20,368
21,103
21,238
21,854
22,541
15%
(lbs./person)
76.9
78.9
80.8
80.5
82.0
83.8
9%
Eggs
1992
1993
1994
1995
1996
1997
%92-97
(million dozen)
5,905
6,006
6,178
6,216
6,359
• 6,436
9%
4
5
1
4
5
6
48%
175
176
212
229
276
220
26%
5,002
5,068
5,160
514
5,228
5,325
7%
Turkey Meat
1992
1993
1994
1995
1996
1997
%92-97
(million pounds ready-to-cook, carcass weight)
4,777
4,798
4,937
5,069
5,401
5,412
13%
NA
NA
NA
.NA
NA
NA
NA
186
224
280
348
438
598
222%
4,584
4,596
4,652
4,706
4,906
4,727
3%
(eggs/person)
235
236
238
235
236
242
3%y
(IbsJperson)
17.9
17.8
17.8
17.8
18.4
17.6
-2%
Source: Putnam and Allshouse, 1997 and 1999. Supplemented with information from USDA/ERS, 1998c and
1997f. Excludes beginning and ending stocks, shipments to U.S. territories, and hatching. Per capita demand is
shown to depict real demand growth, adjusting for growth in U.S. population, which has grown, on average, at
about 1% per year. ,
6-11
-------
6.1.2.4 Farm Price Trends
The poultry sectors experience cyclical expansions and contractions in output. Price
cycles in the poultry meat and egg sectors are caused by a relatively elastic supply and the
tendency for producers to base future production plans on current prices and profits (Kohls and
Uhl, 1998). Perishability of poultry products raises the urgency of farmers to market their
products, thus limiting producer flexibility. Birds must be sold when they reach proper market
weight and maturity; eggs must be sold fresh, regardless of market conditions and prices (Kohls
and Uhl, 1998). Table 6-7 presents the average quarterly and annual broiler, turkey, and egg
prices received by producers from 1992 through 1997.
Table 6-7. Average Quarterly and Annual Poultry Prices Received b
Year
Average
Ql
Average
Q2
Average
Q3
f Farmers, Total U.S., 1992-1997
Average
Q4
Average
Annual
Broilers Monthly Prices Received by Farmers ($/pound)
1992
1993
1994
1995
1996
1997
0.299
0.323
0.345
0.325
0.355
0.392
0.314
0.347
0.369
0.325
0.378
0.370
0.335
0.361
0.355
0.367
0.403
0.396
0.326
0.346
0.329
0.369
0.405
0.338
0.319
0.344
0.350,
0.347
0.385
0.374
All Eggs Monthly Prices Received by Farmers ($/dozen)
1992
1993
1994
1995
1996
1997
1992
1993
1994
1995
1996
1997
0.558
0.650
0.640
0.606
0.777
0.744
0.530
0.663
0.592
0.577
0.721
0.633
0.552
0.581
0.598
0.633
0.732
0.663
0.617
0.622
0.604
0.736
0.811
0.750
Turkeys Monthly Prices Received by Farmers ($/pound)
0.363
0.363
0.376
0.384
0.416
0.376
0.375
0.376
0.398
0.384
0.433
- .0.407
0.377
0.398
0.419
0.417
0.445
0.410
0.391
0.422
0.437
0.455
0.446
0.404 ,
0.564
0.629
0.609
0.638 -
0.760
0.698
0.376
0.390
0.407 .
0.410
0.435
0.399
Source: USDA/NASS, 1998a.
6-12
-------
Broiler prices have been steadily increasing in response to growing market demand. From
1992 to 1997, average annual farm-gate prices rose more than 17 percent, up from $0.32 per
pound to $0.37 per pound (USDA/NASS, 1998a). Broiler prices quoted in Table 6-7 from
USDA/NASS are "equivalent liveweight returns to producers" and are derived from retail, ready-
to-cook prices by subtracting processing costs and multiplying by the dressing percentage. Egg
prices rose between 1992 and 1996, rising 35 percent from $0.56 per dozen to $0.76 per dozen.
Changing market conditions by 1997, however, dropped farm prices back down to $0.69 per
dozen (USDA/NASS, 1998a). Turkey farm prices increased steadily each year from 1992 to
1996, rising more than 15 percent from $0.37 per pound to $0.43 per pound. By 1997, however,
farm prices had dropped back down to $0.40 cents per pound.
The actual price a farmer receives will depend on whether the operation is an independent
owner-operator or whether the operation grows animals under a production contract with a
processor/integrator. The price that a contract grower receives tends to be lower than the market
price received by an independent operator. However, the contract grower's production costs are
oftentimes much lower than those incurred by independents because the contractor provides many
of the production inputs to the grower. In addition to reduced variable costs, contract growers
face less price risk because the integrator guarantees the grower a sales outlet and a certain range
of fees. In comparison, independent growers must cover all production costs, find a sales outlet,
and cope with market price fluctuations. Growers accept integrators' contracts to reduce risk and
gain access to inputs and outlets. Whether the grower is better or worse off with a contract
depends on the grower's attitude toward risk and business objectives, as well as the perceived
gains given the specific terms of the contract.
Sample broiler contract prices reported by the National Poultry Growers Association
range from 3.25 to 7 cents per pound (NPGA, 1998). This same source indicates an average
basic broiler payment of $200/thousand birds, or 3.6 cents per bird assuming a 5.5 pound average
bird. Perry et al. (1999) estimate contract fees of about 22 to 24 cents per bird (1995 data), or
about 4 cents per pound, assuming 5.5 Ibs. per bird. Compared to USDA-reported producer
-prices of about 37 cents per pound (Table 6-7), the price the grower receives is roughly 10
percent of the farm gate price.
The contract broiler grower's payment rate is compensation for the services provided hi
growing chicks to market weight. The integrator retains ownership of the birds and provides
feed, veterinary services, medicines, technical support, and transportation of the animals. These
items amount to approximately 60 percent of the variable expenses of raising broilers (Perry et al.,
1999). Easterling and Lasley (1985) estimate that feed, chicks, and veterinary and other items
cost the contractor a total of 20.6 cents per pound. As a result, average cash expenses are only
$53,446 compared to the nearly $200,000 average for farms with no poultry that generate sales of
$50,000 or more. Contract broiler growers' income statements show most of their income as
"Other farm-related income," which reflects contracting fees collected, rather than as livestock
sales (Perry et al., 1999). Livestock costs for broiler producers with sales of more than $50,000
averaged $1,754 hi 1995 while comparable farms, without poultry averaged $14,825. Broiler
6-13
-------
operations feed costs averaged only $3,725 compared to $26,742 at non-poultry operations
(Perry, et al., 1999). Thus, while the average farm, with sales of $50,000 or more, retains 21
cents for every dollar of sales, the average broiler producer retains 39 cents. The dollar value of
broiler producers' sales are lower because they receive only the contract rate rather than the full
price of the product. However, this relationship makes the average annual gross income for
broiler producers only $86,048 compared to the U.S. average of $250,478 for farms without
poultry that generate sales of $50,000 or more (Perry et al., 1999). Broiler producers' average
net farm income in 1995 was $15,969, about half that of farms without poultry that generate sales
of $50,000 or more (Perry et al., 1999).
Egg contract prices, based on layer production budgets by flock size are reported at
7 cents per dozen (DPRA, 1995) compared to the USDA-reported producer price of 70 cents per
dozen in 1997 (Table 6-7), or about 10 percent. A older USDA study (Easterling and Lasley,
1985) estimated contract payments of 6 cents per dozen, or about 9 percent of the estimated
wholesale egg price used in this study (68.6 cents per dozen). As for broilers, lower prices to
contract growers may be substantially offset by the contractor's provision of production inputs.
Easterling and Lasely (1985) estimate that feed, pullets, veterinary and other livestock production
items cost the contractor a total of 41 cents per dozen. These are items typically provided by
contractors.
Grower payments for turkeys are also estimated by Easterling and Lasley (1985) to be 4.5
cents per pound liveweight in 1984. Growers may also receive substantial production inputs from
contractors. Easterling and Lasley estimate the feed, poults, and veterinary costs and other
production items cost the contractor a total of 33.6 cents per pound. Again, the contract price,of
4.5 cents per pound appears to be about 10 percent of the USDA-reported producer price, which
averages about 40 cents per pound (Table 6-7).
6.1.3 Financial Characteristics of Poultry Operations
6.1.3.1 Overview of Financial Characteristics
USDA reports commercial poultry farms in the U.S. generated a total of $22.3 billion in
annual revenue in 1997 (USDA/NASS, 1999a).2 As shown in Table 6-8, virtually all (99 percent)
poultry farm revenues were from the sales of all poultry meat and eggs. Less than $0.3 billion of
all poultry farm revenue was generated from the sales of other livestock or crop production
(Table 6-8).
2USDA defines commercial farms as those with gross sales of $50,000 or more during a given year.
6-14
-------
Table 6-8. Farm Revenue at Poultry Farms (>S50,000 in Sales), by Revenue Category and Economic Class
Revenue Category/Economic Class
# Poultry Farms (1,000's)
Revenues ($1,000)
Sales by Revenue Category (reported and percentage share)
Primary Livestock
Secondary Livestock
Crop Sales
All Farms
Primary Livestock
Secondary Livestock
Crop Sales
All Farms
27,680
534
261
28,475
97%
2%
1%
100%
21,903,113
202,729
65,855
22,171,699
99%
1%
0%
100%
Sales by Economic Class (reported and percentage share)
>$1 million in revenue
Between $0.5-$ 1.0
Between $0.25-$0.50
Between $0.10-$0.25
Between $0.05-$0.10
All Farms
>$1 million in revenue
Between $0.5-$ 1.0
Between $0.25-$0.50
Between $0.10-$0.25,
Between $0.05-$0.10
AH Farms
5,380
8,454
7,421
5,364
1,855
28,474
19%
30%
26%
19%
7%
100%
12,852,259
5,709,457
2,619,496
857,460
133,026
22,171,698
58%
26%
12%
4%
1%
16667%
Source: USDA/NASS, 1999a (Table 50 and Table 51). Based on data for commercial farms with more than
$50,000 in annual revenues. Excludes non-commercial farms with revenues below $50,000.
Primary Livestock: Hogs (NAICS 1122) and Poultry (NAICS 1123), respectively.
Secondary Livestock: Beef (beef farming, NAICS 112111, and beef feedlots, NAICS 112112), Dairy (NAICS
11212), miscellaneous categories (NAICS 1122, NAICS 1124, NAICS 1125), along with Hogs (NAICS 1122) and
Poultry (NAICS 1.123), respectively.
Crop Sales: Oilseed/Grains (NAICS 1111), Vegetables (NAICS 1112), Fruits/Nuts (NAICS 1113), Greenhouse
(NAICS 1114) and other crops (NAICS 1119).
6-15
-------
As the Census data do not report farm revenues separately among the primary poultry
sectors, sales are estimated based on other available USDA data for these sectors (USDA/ERS,
1996c). Accordingly, the broiler sector captures the largest share of total U.S. poultry sales,
claiming nearly two-thirds of all farm sector poultry receipts. The farm value of all U.S. broiler
and chicken meat production is estimated $13.6 billion in 1997 (USDA/NASS, 1999a;
USDA/ERS, 1996c). Turkey, revenues account for under 15 percent of total poultry receipts
generated annually. In 1997, the farm value of all U.S. turkey production was estimated at $3.2
billion. Revenues from egg sales account for about one-fifth of total annual poultry receipts,
totaling an estimated $4.5 billion in 1997 (Table 6-8).3 The remaining 3 percent of all poultry
revenues ($0.7 billion) were generated from the sale of miscellaneous poultry products.
As shown in Table 6-8 almost one-half of all commercial poultry farms generate more than
$0.5 million in revenue annually (USDA/NASS, 1999a). The remaining share of farms generate
revenues below $0.5 million. These data do not distinguish among the primary poultry sectors
(broilers, turkeys, and layers). The $0.5 million threshold corresponds with the definition of a
"small business" in the broiler and turkey sectors established by the Small Business
Administration; SBA's definition of a "small" layer operation is one with less than $9.0 million in
annual revenues (SBA, 1998; USGPO, 2000). (Section 9 provides additional information on
EPA's small business analysis.)
6.1.3.2 Income Statement and Balance Sheet Information
Table 6-9 presents average income statement and balance sheet data for commercial
poultry farms from 1993 through 1997. (These data do not distinguish among the primary poultry
sectors—broilers, turkeys, and layers.) The average U.S. poultry farm was in a favorable financial
position from 1993 through 1997 with a positive net farm income and a debt-to-asset ratio that
from ranged from 0.19 to 0.30 from 1993 to 1997. (USDA's farm performance criteria are
described in Section 4.2.5.) While the national average income statement shows a positive net
income, additional .information show that between 1991 and 1994 about 21 percent of all poultry
farms experienced negative income (USDA/ERS, 1997b). Operations in the poorer performing.
category likely are smaller operations that are not affected by the proposed CAFO regulations.
Direct financial comparisons between poultry meat and egg production are difficult
because of the way USDA data are structured. However, there are cost and return characteristics
that can be compared across poultry sectors. For example, because feed costs comprise a large
share of total production costs, each subcategory's feed conversion ratio (i.e., the pounds of feed
per unit of production) is an important indicator of production efficiency. Broilers have a lower
(i.e., more efficient) feed conversion ratio than turkeys, requiring less feed per pound of product
produced (Easterling and Lasley, 1985). This translates into a lower feed cost per pound of meat
produced. Feed is also a smaller percentage of total cost for broilers than for turkeys or layers.
3 This does not include layer farms revenues generated from the sales of culled inventories for chicken
meat or the sales of below grade eggs for'use in processing.
6-16
-------
Table 6-9. Income Statement and Balance Sheet for Poultry Farms (Sales >S50,000), 1993-97
Item
1993
1994
1995
1996
1997
Dollars per Farm
[ncome Statement
Gross cash income
Livestock sales
Crop sales (incl. net CCC loans)
Government payments
Other farm-related income ^
Less: Cash expenses
Variable Cash expenses
Fixed Cash expenses
Equals: Net cash farm income
Less: Depreciation
Labor, non-cash benefits
Plus: Value of inventory change
Non-money income b/
Equals: Net farm income
$147,575
$79,102
$11,106
$1,156
$56,211
$108,528
$87,977
$20,550
$39,047
$25,943
$301
$3,007
$4,680
$20,490
$204,915
$130,438
$6,273
$653
$67,551
$166,355
$148,413
$17,941
$38,560
$20,017
$617
$5,317
$5,078
$28,322
$203,711
$122,364
$5,700
$507
$75,139
$161,491
$138,373
$23,118
$42,220
$20,866
$527
$8,197
$5,933
$34,957
$174,089
$74,105
$5,958
$350
$93,676
$116,614
$96,855
$19,759
$57,475
$19,895
$269
$494
$5,132
$42,937
$341,206
$238,281
$10,158
$1,213
$91,554
$271,349
$244,282
$27,066
$69,858
$20,417
$384
$5,603
,$5,043
$59,703
Balance Sheet
Farm assets
Current assets
Non-current assets
Farm liabilities
Current liabilities
Non-current liabilities
Farm equity
Debt/asset ratio
$537,351
$58,152
$479,199
$130,914
$33,339
$97,575
$406,437
0.24
$611,305
$90,618
$520,687
$114,579
$23,863
$90,715
$496,727
0.19
$578,753
$79,672
$499,082
$136,437
$51,424
$85,013
$442,316
0.24
$560,174
$41,207
$518,967
$168,055
$30,849
$137,206
$392,119
0.30
680,690
71,598
609,092
173,200
38,139
135,061
507,490
0.25
Source: USDA/ERS, 1999a and 1996e.
a Includes income from machine-hire, custom work, livestock grazing, land rental, contract production fees,
outdoor recreation, and any other farm-related source.
b Defined as home consumption and imputed rental value of farm dwellings owned by the farm operation.
Table 6-10 provides additional information from USDA (1993 data) that describes income
statement differences among broiler operations (USDA/ERS, 1996c). At the average single-
contract farm, fixed expenses account for about one-third of the cash expenses, with interest and
insurance as the largest components of fixed expenses. At contract operations, there is no one
dominant expense category for the variable expenses. Although the average poultry operation
shows a positive net cash income, the average contract broiler operation shows approximately
one-half the gross income ($77,452 versus $147,575), about 65 percent of the net cash income
($25,341 versus $39,047), and a higher proportion of fixed expenses (33 percent versus 19
6-17
-------
percent) compared to the average poultry operation. On the other hand, feed accounts for 38
percent of the variable costs for the poultry operation but only 6 percent of the variable costs for
the contract broiler operation.
Table 6-10. Income Statements for Single-Contract Farms with Broilers, 1993
Item
Gross cash income
Production fees
Cash expenses
Variable expenses
Feed purchase
Livestock related
Seed and plants
Fertilizer and chemicals
Labor
Fuels and oils
Repairs and maintenance
Machine work/custom hire
Utilities
Other variable expenses
Fixed expenses
Real estate/property taxes
Interest and insurance
Rent/lease payments
Net cash income
Farms with Contract Value of
Farm with One
Broiler Contract
$77,452
$53,552
$52,111
$34,682
S
$2,063
$11,495
$1,082
$2,916
$6,716
$5,476
$5,071
$865
$5,417
$2,963
$17,429
$1,756
$14,503
$1,170
$25,341
Under
$300,000
($ per cwt.
$33,625
$20,836
$25,119
$19,802
S
$1,393
$914
$215
$591
$4,034
$3,018
$3,460
$334
$3,734
$2,109
$5,317
$838
$4,065
$414
$8,506
$300,000 to
$599,999
*ain)
$73,957
$53,140
$47,510
$30,741
S
$1,404
$1,839
$959
$2,907
$5,199
$5,249
$4,657
$643
$4,917
$2,714
$16,769
$1,901
$13,535
$1,333
$27,477
More than
$600,000
$150,438
$102,840
$102,848
$65,972
S
$4,607
$1,539
$2,652
$6,470
$14,262
$9,882
$8,426
$2,175
$9,083
$4,813
$38,876
$2,769
$32,208
$1,899
$47,590
Source: USDA/ERS, 1996c. S = suppressed because the relative standard error exceeds 50 percent
Table 6-11 provides other financial information on the broiler, turkey, and egg laying
sectors that summarize farm and wholesale costs and estimated net returns from 1990 to 1995.
These data are compiled monthly by USD A, based on estimated costs and regional spot market
prices, and represent national average returns to producers and wholesalers (USDA/ERS, 1997d).
Feed accounts for between 60 percent and 64 percent of costs to raise broilers, turkeys, and egg
layers (see Table 6-11). Costs for broilers range from $0.26 to $0.27 per pound while prices
range from $0.31 to $0.35 per pound. Costs for turkeys range from $0.36 to $0.38 per pound,
while prices range from $0.38 to $0.41 per pound (USDA/ERS, 1997d). Costs for eggs range
6-18
-------
from $0.46 to $0.47 per dozen while prices range from $0.45 to $0.62"per dozen. In 1992, the
cost exceeded the price for eggs (Table 6-11).
Net returns as a percentage of price range from 14 percent to 24 percent for broilers. For
turkeys, net returns are much lower, i.e., 2 percent to 13 percent of price. Eggs show the largest
fluctuation in net returns as a percentage of price, from a 3 percent decrease in 1992 to a 25
percent gain in 1990 (USDA/ERS, 1997d). See Table 6-11.
6.1.3.3 Baseline Conditions for Poultry Operations
Tables 6-12 through 6-17 provide a summary of title financial baseline conditions assumed
for this analysis. These data are aggregated from the 1997 ARMS data set and are obtained by
USDA's ERS, as described in Section 4. These data are separated by select facility size and
production region groupings (see Table 4-4), but do not reflect conditions separately across the
different types of poultry operations within a sector (e.g., dry layers vs. wet layers or contract
operations vs. non-contract operations). Additional information on how these data differ by
region are provided in the record (USDA/ERS, 1999a, see DCN 70063).
According to the 1997 ARMS data the average poultry operation demonstrated a
favorable financial position in 1997 with positive net income and a debt-to-asset ratio that ranged
from 7 percent (small turkey operations) to 30 percent (large broiler operations), across select
operation sizes (USDA/ERS, 1999a). See Tables 6-12, 6-14, and 6-16. These debt-to-asset
ratios indicate that—on average—poultry operations are not in a vulnerable financial position and
have a low potential for cash flow problems and a low relative risk of insolvency. Based on these
data, EPA assumes that baseline (prior to regulation) net cash flow for all model types for the
poultry sector is estimated to be positive, and baseline debt-to-asset ratios for all model types are
40 percent or less. All poultry operations in this analysis, therefore, are considered financially
healthy, on average, in the regulatory baseline.
Broiler Operations
Data shown in Table 6-12 are distributed by broad facility size groups. As shown, more
than 75 percent of operations have fewer than 30,000 birds, however, these operations only
account for about 7 percent of all broilers raised annually (Table 6-12). There are fewer larger-
sized operations with more than 90,000 birds (7 percent of all farms), but these operations raise
over 49 percent of all broilers annually (Table 6-12). Smaller broiler operations with less than
30,000 birds are slightly more diversified than larger ones, with about 17 percent of all farm
revenue from crops. This compares to broiler operations with more than 90,000 birds, where
livestock comprises the bulk of all annual farm sales and only 3 percent of farm revenues are from
crops (Table 6-12). Overall, the average broiler operation does not have a large value of crop
production, regardless of size of operation.
6-19
-------
Table 6-11 Distribution of Commercial Farms, by Net Farm Income, 1990-1995
Sector
1990
1991
1992
1993 1994 1995
Broilers (cents per pound)
Farm price
Farm production costs
Feed costs
Feed costs as a % of farm costs
Farm net returns
Net returns as % of price
Wholesale price
Wholesale cost
Wholesale net returns ^
Net returns as % of price
32.38
26.87
16.59
62%
5.51
17%
54.77
49.24
5.53
10%
Turkeys
Farm price
Farm production costs
Feed costs
Feed costs as a % of farm costs
Farm net returns
Net returns as % of price
Wholesale price
Wholesale cost
Wholesale net returns ^
Net returns as % of price
38.37
37.10
23.40
63%
1.27
3%
62.35
62.60
-0.25
-0%
30.91
26.67
16.34
61%
4.24
14%
52.03
49.06
2.97
6%
31.85
26.68
16.33
61%
5.17
16%
52.57
49.32
3.25-
6%
34.43
26.39
16.04
61%
8.04
23%
55.18
48.83
. 6.35
12%
34.95
27.16
16.81
62%
7.79
22%
55.80
49.84
5.96
11%
34.66
26.38
16.03
61%
8.28
24%
56.23
48.81
7.42
13%
cents per pound)
37.65
36.42
22.72
62%
1.23
3%
60.79
61.83
-1.04
-2%
37.63
36.86
23.16
63%
0.77
2%
60.48
62.38 .
-1.90
-3%
38.99-
36.09
22.39
62%
2.90
7%
62.83
61.42
1.41
2%
40.74
37.68
23.98
64%
3.06
8%
65.53
63.40
2.13
3%
41.08
35.63
21.93
62%
5.45
13%
65.89
60.83
5.06
8%
Eggs (cents per dozen)
Farm price
Farm production costs
Feed costs
Feed costs as a % of farm costs
Farm net returns
Net returns as % of price
Wholesale price
Wholesale cost
Wholesale net returns ^
Mpf rfMiims aQ °/ rtf nrire*
62.00
46.61
28.41
61%
15.39
25%
83.81
67.33
16.48
70%
56.65
46.25
28.05
61%
10.40
18%
79.49
67.07
12.42
16%
45.00
46.23
28.03
61%
-1.23
-3%
68.43
66.73
1.70
2%
51.34
45.95
27.75
60%
5.39
10%
75.06
66.46
8.60
11%_
49.20
47.08
28.88
61%
2.12
4%
71.09
67.58
3.51
5%
53.58
47.05
28.85
61%
6.53
12%
76.36
67.55
8.81
1 ?%
Source: Derived from USDA/ERS, 1997d.
"'Average for 12-Metro area.
Table 6-12 also shows the percentage of broilers owned by farmers compared to those not
owned by farmers. EPA uses this information on animal ownership as an indication of the extent
of production contract use in these sectors (see Section 2.3). Across all broiler operations in
1997, nearly all (98 percent) broilers' and meat chickens were not owned by farmers (USDA/ERS,
6-20
-------
1999a). Percentages vary slightly across farm sizes, with up to 99.6 percent of birds not owned
by the farming operation for farms with more than 90,000 birds, compared to 92 percent among
smaller-sized operations (Table 6-12).
The data shown in Table 6-13 are also differentiated by selected size categories and reveal
differences among operations by size. The income statement data for broiler operations reflect
the prevalence of production contracting in this sector. These data also point to increasing
specialization as the size of an operation increases. A larger proportion of birds are not owned at
the largest operations than at the smallest operations and the smallest operations may have
proportionately larger expenditures on livestock-related expenses than large operations.
Operations with fewer than 30,000 birds have average feed expenditures of $4,275, while
operations with more than 90,000 birds have average feed expenditures of only $2,525
(Table 6-13). This is expected given that nearly 100 percent of broilers are not owned at large
operations and feed is most likely provided by the contractor.
Operating margins (measured as average net cash farm income as a percentage of average
gross cash income) indicate that there may be large economies of scale associated with broiler
sector operations (USDA/ERS, 1999a). Operating margins are negligible at operations with
fewer than 90,000 birds, as compared to 39 percent and 43 percent at operations with 30,000 to
90,000 birds, and more than 90,000 birds. The smallest operations also show the lowest return on
assets (measured as average net farm income to average farm assets); operations with less than
30,000 birds show average return on assets of 1.2 percent, as compared to 6.8 percent and 10
percent at operations with between 30,000 and 90,000 birds and more than 90,000 birds,
respectively (USDA/ERS, 1999a). See Table 6-13. The 1997 ARMS data include, among an
average farm's assets, the value of the owner's home when it is located on the farm. Since
smaller operations may be more likely to have the owner's dwelling located on the farm than
larger operations, if dwelling values were excluded, the returns on the "business" assets might be
higher for these smallest operations.
Section 4 of this report presents key financial data used for this analysis, shown in
Table 6-13, that are calculated onto a per-animal basis. For the broiler sector, total gross farm
revenues are estimated to range from $1.10 to $1.50 per bird (includes revenue from other farm-
related sources). Net cash income ranges from $0.50 to $0.60 per bird among CAFO models,
depending on facility size and region (see Tables 4-5 and 4-6).
Layer Operations
Data shown in Table 6-14 are distributed by broad facility size groups. As shown, nearly
100 percent of layer operations have fewer than 90,000 birds, however, these operations only
account for about 39 percent of all layers raised at these operations (Table 6-14). There are very
few larger-sized operations with more than 120,000 birds (0.2 percent of all farms), but these
operations are associated with approximately 56 percent of all layers (Table 6-14). The average
6-21
-------
Table fi-12 Tvnical Financial Characteristics of Broiler Operations, By Size of Operation
Item
Number of farms
Percent of farms
Percent of value of production
Livestock value of production
Crop value of production
All Farms
34,264
100.0%
100.0%
94.7%
5.3%
Less than
30,000 Birds
24,813
75.3%
19.4%
82.9%
17.1%a
30,000 to
90,000 Birds
6,167a
18.0%
46.9%
97.7%
2.3%a
Number of broilers and fryers
Distribution of broilers and fryers
Percent of broilers and fryers owned
Percent of broilers and fryers not owned
Number of sample farms with broilers/ fryers
763,830,283
100.0%
2.4%c
97.6%
, 275
50,40 l,533a
6.6%
7.8%c
92.2%
85
343,127,493
44.9%
3.7%
96.3%
102
Debt-to- Asset Ratios
All Regions
Mid-Atlantic
South
0.1930
d
0.1846
0.1476a
d
0.1207a
0.2076
d
0.1910
EPA Derived Gross Cash Income Per Animal "
All Regions
Mid-Atlantic
South
$1.93
d
$1.53
$10.95
d
$5.17
$1.47
d
$1.42
More than
90,000 Birds
2,284
6.7%
33.7%
97.4%
2.6%b
370,301,257
48.5%
0.4%
99.6%
88
0.3042
d
0.2640
$1.13
d
$1.16
EPA Derived Net Cash Income Per Animal "
All Regions
Mid-Atlantic
South
$0.49
d
$0.51
$0.01
d
$0.16
$0.57
d
$0.57
$0.48
c
$0.49
Source: USEPA and USDA/ERS, 1999a.
a = The relative standard error of the estimate exceeds 25 percent, but no more than 50 percent.
b — The relative standard error of the estimate exceeds 50 percent, but no more than 75 percent
c = The relative standard error of the estimate exceeds 75 percent.
d = Data insufficient for disclosure.
"EPA derived gross cash and net cash income per animal by dividing the average gross or net cash income line
items by the average number of animals as reported for each size group and region.
6-22
-------
Table 6-13. Income Statement and Balance Sheet for Broiler Operations, By Size of Operation
Item
All Farms
Less than
30,000 Birds
30,000 to
90,000 Birds
More than
90,000 Birds
Income Statement
Gross cash income
Livestock income
Crop sales (incl. net CCC loans)
Government payments
Other farm-related income y
Total variable expenses
Livestock purchases
Feed
Other variable expenses f
Total fixed expenses
Equals: Net cash farm income
Less: Depreciation and Other u
Plus: Value of inventory change
Plus: Non-money income "'
Equals: Net farm income
Farm assets
Current assets
Non-current assets
Land, buildings, and equipment 31
Farm liabilities
Current liabilities
Non-current liabilities
Farm equity
43,033
10,064a
7,985a
995a
23,989
23,228
l,199a
3,925a
18,104
8,825
10,979a
6,566a
3,298
3,886
11,598
21,378a
8,622a
6,258a
813a
5,684a
15,681a
991
4,275a
10,415a
5,680a
17c
3,901a
3,021b
3,508
2,645c
81,731
10,256b
7,7 16a
d
d
37,266a
d
2,980b
31,992a
12,598a
31,868
10,014a,b
l,768b
4,799
28,421
183,266
25,837a
28,229b
d
d
70,615
d
2,525s.
33,924
34,185
78,466
27,359b
10,560c
5,690
67,357
Balance Sheet
301,982
18,162
283,820
276,084
58,270a
14,260a
44,010a
243,712
226,233a
145,528a
211,705a
204,444a
33,385a
ll,728b
21,657a
192,847
396,477
17,388a
379,090a
373,408
82,296a
17,524a
64,773 a
314,181a
902,856
61,320
841,537
822,877
274,606
34,052
240,554
628,250
Source: Based on USDA/ERS, 1999a. Copies of these data are in the rulemaking record (DCN 70063).
"Machine-hire, custom work, livestock grazing, land rental, contract fees, and other farm-related sources.
^Incl. livestock leasing, custom feed processing, bedding, grazing, supply, transportation, storage, general business
expenses, and registration fees. Footnote a or b refers to an RSE on "other livestock-related" portion of the total.
^Includes labor, non-cash benefits. Footnote (a) refers to an RSE on "non-cash benefits" portion of the total.
Footnote (a,b) refers to an RSE on "depreciation" for "a" and "non-cash benefits" for "b."
4/The value of home consumption plus an imputed rental value of farm dwellings.
5/The value of the operator's dwelling and associated liabilities are included if the dwelling was located on the ;
farm, a = Relative standard error (RSE) of the estimate exceeds 25 percent, but no more than 50 percent.
b = RSE of estimate >50%, but <75%. c = RSE of estimate >75%. d = Data insufficient for disclosure.
6-23
-------
small layer operation does not have a high value of crop production; livestock value of production
for small operations is greater than 70 percent. Large layer operations typically have a livestock
value of production of almost 100 percent.
Table 6-14 also shows the percentage of layers owned by farmers compared to those not
owned by farmers. EPA uses this information on animal ownership as an indication of the extent
of production contract use in these sectors (see Section 2.3). .Across all layer operations in 1997,
about 43 percent of layers and pullets were not owned by farmers (USDA/ERS, 1999a).
Percentages vary across farm sizes, with only about 10 percent of birds not owned by the farming
operation for farms with more than 120,000 birds, compared to 84 percent among smaller-sized
operations (Table 6-14).
The data shown in Table 6-15 are also differentiated by selected size categories, however,
data are not disclosed for any but small operations with fewer than 90,000 birds and overall
average operations. The income statement data (as well as the data in Table 6-14) point to
increasing specialization as the size of an operation increases. A larger proportion of animals are
not owned at the smallest operations (compared to overall operations), and the smaller operations
have proportionately smaller expenditures on livestock-related expenditures than the overall
operation. Expenditures on livestock and feed average about 46 percent of total variable
expenses at an average layer operation; operations with fewer than 90,000 birds are associated
with expenditures on livestock and feed averaging 39 percent of total variable expenses
(Table 6-15).
Operating margins (measured as net cash farm income to gross cash income) are not
substantially different between the smallest operations and all operations. Operating margins
average 11 percent at the smallest operations and 16 percent for all layer operations. Return on
assets data (measured as average net farm income to average farm assets) are also similar. The
smallest layer operations average a 2.7 percent return on assets compared to a 3.6 percent return
on average for all layer operations (USDA/ERS, 1999a). See Table 6-15. The 1997 ARMS data
include, among an average farm's assets, the value of.the owner's home when it is located on the
farm. Since smaller operations may be more likely to have the owner's dwelling located on the
farm than larger operations, if dwelling values were excluded, the returns on the "business" assets
might be higher for these smallest operations.
Section 4 of this report presents key financial data used for this analysis, shown in
Table 7-10, that are calculated onto a per-animal basis. For the egg laying sector, total gross farm
revenues are estimated at $25 per bird (includes revenue from other farm-related sources). Net
cash income is estimated at about $4 per bird among CAFO models, depending on facility size
and region (see Tables 4-5 and 4-6).
6-24
-------
Table 6-14. Typical Financial Characteristics of Layer Operations, By Size of Operation
Item
Number of farms
Percent of farms
Percent of value of production
Livestock value of production
Crop value of production
All Farms
129,172
100.0%
100.0%
100.0%
20.5%
Less than
90,000 Birds
128,846
99.7%
71.8%
71.8%
28.2%
90,000 to
120,000 Birds
d
" d
d
d
d
More than
120,000 Birds
d
0.2%
25.3%
99.5%
d
Number of layers and pullets
Distribution of layers and pullets
Percent of layers and pullets owned
Percent of layers and pullets not owned
Number of sample farms with layers
and pullets
213,362,980
100.0%
56.7%
43.3%
409
82,179,082
38.5%
16.5%a
83.5%
378
d
d
d
d
11
d
56.2%
89.2%
d
20
Debt-to-Asset Ratios
All Regions
Mid-Atlantic
South
0.1059
0.1236
0.1047
0.0985
0.1173
0.0983
d
d
d
d
d
d
EPA Derived Gross Cash Income Per Animal "
All Regions
Mid-Atlantic
South
$24.63
$45.27
$13.02
$46.26
$140.84
$16.54
d
d
d
d
d
d
EPA Derived Net Cash Income Per Animal "
All Regions
Mid-Atlantic
South
$4.06
$11.39
$1.31
$5.12
$30.72
($0.54)
d
d
d
d
d
d
Source: USEPA and USDA/ERS, 1999a.
a = The relative standard error of the estimate exceeds 25 percent, but no more than 50 percent.
b = The relative standard error of the estimate exceeds 50 percent, but no more than 75 percent.
c = The relative standard error of the estimate exceeds 75 percent.
d = Data insufficient for disclosure.
"EPA derived gross cash and net cash income per animal by dividing the average gross or net cash income line
items by the average number of animals as reported for each size group and region.
6-25
-------
Tnrnmp Statement and Balance Sheet for Layer Operations, By Size of Operation
Item
All Farms
<90,000 Birds
90,000-
120,OOOBirds
Income Statement
Gross cash income
livestock income
Crop sales (incl. net CCC loans)
Government payments
Other farm-related income "
Total variable expenses
Livestock purchases
Feed
Other variable expenses "
Total fixed expenses
Equals: Net cash farm income
Less: Depreciation and Other 3/
Plus: Value of inventory change-
Plus: Non-money income 4/
Equals: Net farm income
40,679
28,039
6,130a
609-
5,902a
27,655
l,874a
10,952
14,829
6,323
6,701a
4,344a
3,745a
4,738
10,S40a
29,503
17,652
6,075a
608
5,168a
20,493
l,249a
6,827a
12,414
5,743
3,267b
3,977a
3,720a
4,740
7,750a
d
d
d
d
d
d
d
d
d
d
d
d
>120,000 Birds
d
d
d
d
d
d
d
d
d
d
d
d
Balance Sheet
Farm assets
Current assets
Non-current assets
Land, buildings, and equipment 5/
Farm liabilities
Current liabilities
Non-current liabilities
Farm equity
301,193
29,713
271,480
258,617
31,883
9,026
22,857
269,309
291,889
28,067a
263,822
251,478
28,739
8,392
20,347
263,151
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
"Machine-hire, custom work, livestock grazing, land rental, contract fees, and other farm-related sources.
^cl. livestock leasing, custom feed processing, bedding, grazing, supply, transportation, storage, general business
expenses, and registration fees.
'Includes labor, non-cash benefits. Footnote (a) refers to an RSE on "non-cash benefits" portion of the total.
4/The value of home consumption plus an imputed rental value of farm dwellings.
'"The value of the operator's dwelling and associated liabilities are included if the dwelling was located on the
farm, a = Relative standard error (RSE) of the estimate exceeds 25 percent, but no more than 50 percent
b - RSE of estimate >50%, but <75%. c = RSE of estimate >75%. d = Data insufficient for disclosure.
6-26
-------
Turkey Operations
Data shown in Table 6-16 are distributed by broad facility size groups. As shown,
approximately 82 percent of operations have fewer than 10,000 birds. The exact number of large
turkey operations and the distribution at these operations are not disclosed. The average small
turkey operation has a high value of crop production; livestock value of production for small
operations is only about 33 percent (Table 6-16). Large turkey operations typically have a
livestock value of production of almost 99 percent (Table 6-16).
Table 6-16 also shows the percentage of turkeys owned by farmers compared to those not
owned by farmers. EPA uses this information on animal ownership as an indication of the extent
of production contract use in these sectors (see Section 2.3). Across all turkey operations in
1997, about 70 percent of turkeys were not owned by farmers (USDA/ERS, 1999a). For small
operations, more than 85 percent of turkeys were not owned by farmers; the percentage of
turkeys not owned by large operations was not disclosed, but is probably less than the percentage
not owned among smaller operations, since the overall percentage not farmer-owned (70 percent)
is lower than the percentages associated with smaller and mid-size operations (86 percent and 83
percent).
Table 6-17 also presents average income statement and balance sheet information for
turkey operations in 1997, by size of operation, although data for the largest operations are not
disclosed. The income statement data (as well as the data in Table 6-16) point to increasing
specialization as the size of an operation increases. A larger proportion of birds are not owned at
the smallest operations than at largest operations based on the fact that operations with less than
10,000 birds and with 10,000 to 40,000 birds show equivalent percentages of turkeys not owned,
while the average for all turkey operations is much lower. However, the smallest operations have
proportionately smaller expenditures on livestock-related expenses than larger operations.
Expenditures on livestock and feed average about a quarter of total variable expenses at a turkey
operation with fewer than 10,000 birds; operations with between 10,000 and 40,000 birds are
associated with expenditures on livestock and feed averaging nearly two-thirds of total variable
expenses (Table 6-17). Explanations for these differences may include differences in the degree of
specialization and feeding strategies, and other factors.
Operating margins (measured as net cash farm income to gross cash income) decline with
size. Operations with fewer than 10,000 birds show an operating margin of 30 percent, as
compared with 23 percent and 9 percent for operations with between 10,000 and 40,000 birds
and all operations, respectively (USDA/ERS, 1999a). However, return on assets does not vary
much by size. The smallest operations show a return on assets (measured as average net farm
income to average farm assets) of 5.2 percent, as compared with 6.4 percent and 4.4 percent at
operations with between 10,000 and 40,000 turkeys and all turkey operations, respectively
(USDA/ERS, 1999a). See Table 6-17. The 1997 ARMS data include, among an average farm's
assets, the value of the owner's home when it is located on the farm. Since smaller operations
may be more likely to have the owner's dwelling located on the farm than larger operations, if
6-27
-------
Table 6-16 TvDical Financial Characteristics of Turkey Operations, By Size of Operation
Item
Number of farms
Percent of farms
Percent of value of production
Livestock value of production
Crop value of production
AH Farms
ll,266a
100.0%
100.0%
89.8%
10.2%c
Less than
10,000 Birds
9,267a
82.3%
12.2%b
32.9%b
67.1%a
10,000 to
40,000 Birds
l,406a
12.5%a
10.6%a
88.9%
ll.l%a.
More than
40,000 Birds
d
d
77.2%
98.9%
d
Number of turkeys
Distribution of turkeys
Percent of turkeys owned
Percent of turkeys not owned
Number of sample farms with turkeys
153,994,175a
100.0%
30.0%b
70.0%
146
d
d
14.5%c
85.5%a
40
26,923,570
17.5%
16.9%a
83.1%
83
Debt-to-Asset Ratios
All Regions
Mid-Atlantic
Midwest
0.1512a
0.2039
0.1258
0.0736a
d
d
0.2305a
0.2177
d
EPA Derived Gross Cash Income Per Animal "
All Regions
Mid-Atlantic
Midwest
$20.08
$5.78
$26.34
$8.60
d
d
$11.24
$6.47
d
EPA Derived Net Cash Income Per Animal "
All Regions •
Mid-Atlantic
Midwest
$1.77
$1.24
$2.00
$2.56
d
d
$2.55
$2.92
d
d
d
d
d
23
d
d
d
d
d
d
c
c
c
Source: USEPA and USDA/ERS, 1999a.
a - The relative standard error of the estimate exceeds 25 percent, but no more than 50 percent.
b - The relative standard error of the estimate exceeds 50 percent, but no more than 75 percent
c - The relative standard error of the estimate exceeds 75 percent.
d = Data insufficient for disclosure.
"EPA derived gross cash and net cash income per animal by dividing the average gross or net cash income line
items by the average number of animals as reported for each size group and region.
6-28
-------
Table 6-17. Income Statement and Balance Sheet for Turkey Operations, By Size of Operation
Item
All Farms
Less than
10,000 Birds
10,000 to
40,000 Birds
More than
40,000 Birds
Income Statement
Gross cash income
Livestock income
Crop sales (incl. net CCC loans)
Government payments
Other farm-related income "
Total variable expenses
Livestock purchases
Feed
Other variable expenses ^
Total fixed expenses
Equals: Net cash farm income
Less: Depreciation and Other 3/
Plus: Value of inventory change
Plus: Non-money income 4/
Equals: Net farm income
274,433c
227,043c
25,219c
859b
21,312c
236,362c
d
122,678c
76,253c
13,939b
24,131c
12,470b,c
6,022b
5,822
23,500c
43,020c
13,616c
d
d
2,546c
24,854c
943a
4,573c
19,338c
5,346c
12,819c
4,406b,d
5,570a
. d
19,743c
215,299a
123,810a
16,228a
l,553b
73,728
137,788a
15,862b
63,237b
48,689b
28,731a
48,780
17,492a,b
10,869a
4,885
47,043
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
Balance Sheet
Farm assets
Current assets
Non-current assets
Land, buildings, and equipment 5/
Farm liabilities
Current liabilities
Non-current liabilities
Farm equity
531,744b
57,496a
424,248a
454,787a
80,381c
27,65 Ic
52,731b
451,363a
380,924b
16,057b
364,867b
354,689b
28,048c
7,956c
20,092c
352,876a
734,905
72,363a
662,543
637,930
169,430a
40,855a
128,574b
565,476
d
d
d
d
d
d
d
d
Source: Based on USDA/ERS, 1999a. Copies of these data are in the rulemaking record (DCN 70063).
"Machine-hire, custom work, livestock grazing, land rental, contract fees, and other farm-related sources.
^Incl. livestock leasing, custom feed processing, bedding, grazing, supply, transportation, storage, general business
expenses, and registration fees. Footnote (c) or (b) refers to an RSE on "other livestock-related" portion of the
total.
^Includes labor, non-cash benefits. Footnote (b,c) refers to an RSE on "depreciation" for "b" and "non-cash
benefits" for "c." Footnote (b,d) refers to an RSE on "depreciation" for "b" and "non-cash benefits" for "d."
Footnote (a,b) refers to an RSE on "depreciation" for "a" and "non-cash benefits" for "b."
""The value of home consumption plus an imputed rental value of farm dwellings.
5/The value of the operator's dwelling and associated liabilities are included if the dwelling was located on the
farm, a = Relative standard error (RSE) of the estimate exceeds 25 percent, but no more than 50 percent.
b = RSE of estimate >50%, but <75%. c = RSE of estimate >75%. d = Data insufficient for disclosure.
6-29
-------
dwelling values were excluded, the returns on the "business" assets might be higher for these
smallest operations.
Section 4 of this report presents key financial data used for this analysis, shown in
Table 6-17 that are calculated onto a per-animal basis. For the turkey sector, total gross farm
revenues are estimated to range from $11 to $20 per bird (includes revenue from other farm-
related sources). Net cash income ranges from $1.80 to $2.60 per bird among CAFO models,
depending on facility size and region (see Tables 4-5 and 4-6).
6.2 PROFILE OF POULTRY PROCESSING SECTORS
Poultry farms represent the beginning of the chicken meat and egg products marketing
chain that also includes poultry slaughtering facilities, poultry and food processors, integrators,
and retailers. Farms provide the raw materials to slaughterers and processors in the form of live
birds and eggs, which then are converted into cuts of meat and processed foods. These products
are eventually sold to consumers at retail establishments.
Most broilers and turkeys are marketed as eviscerated, ice-packed, or frozen "ready-to-
cook" (RTC) poultry, which is available in many forms, including whole birds, cut-up birds,
poultry parts, and self-basting poultry. Chicken and turkey products may also be "further
processed," referring to breaded and pre-cooked parts, ready-made and frozen meals, and other
manufactured products. Egg processors are classified under NAICS 311999(G), Liquid, dried,
and frozen eggs. Almost 70 percent of all egg production is sold in fresh form to retail stores or
to institutional buyers. Another 29 percent of all shell eggs are sold to "breakers," which are
firms that process eggs into dried, frozen, and liquid egg products used as ingredients by
processors in numerous food products; the remaining one percent of egg output is exported
(American Egg Board, 1998). See Section 2 for more information about poultry processors.
Additional information on the processing sectors in these industries is provided in
Section 2, which also shows how EPA estimated the potential number of processors that may be
affected by the proposed regulations as co-permittees.
As presented in Section 2.4, EPA estimates that about 270 poultry processing
establishments would be subject to co-permitting requirements. EPA's determination is based
largely on the fact that production contracting accounts for a large share of poultry production
(USDA/ERS, 1999a and 1996c; Heffernan, et al., 1999). In the poultry sector, vertical
integration has progressed to the point where large, multifunction producer-packer-processor-
distributor firms are the dominant force in poultry and egg production and marketing (Kohls and
Uhl, 1998). Today's integrators are subsidiaries of feed companies, independents, cooperatives,
meat packers, or retailers, or are affiliates of conglomerate corporations. These firms may own
and/or direct the entire process from the production of hatching eggs through the merchandising
of ready-to-eat-sized broiler portions to restaurants (Hayenga et al., 1996). Coordination through
production contracting now dominates the poultry industry (Aust, 1997). Nearly all poultry is
6-30
-------
grown through contract production, with fully integrated production operations comprising a
small share of production. Spot market trading of poultry is insignificant (USDA/ERS, 1996c). .
Elements specific to the poultry sector—including assets at several levels (breeding flocks,
hatcheries, broiler houses, and processing plants), a short biological process, and a perishable
product—have led to a tightly coordinated flow of eggs, chicks, and broilers. This setup
minimizes transaction costs and risks and reduces production costs, allowing integrators to
supersede the market (Hayenga et al., 1996). Decentralization and direct sales have accompanied
integration of the poultry marketing channels, with most poultry meat and shell eggs moving
directly from packaging plants to retail buyers (Kohls and Uhl, 1998).
In a 1993 study, USDA showed that almost 90 percent of all poultry production is
produced under contract (USDA/ERS, 1996c). Contract production across all market segments
in the poultry industry accounted for 86 percent of the value of all production. Nearly all poultry
grown under contract is through production contracts. Under most poultry production contracts,
the integrator supplies some inputs (chicks, feed, medication, field supervision) and the farmer
provides other inputs (housing, water, and fuel, etc.) (Kohls and Uhl, 1998).
Across all broiler farms in 1997, nearly all (98 percent) broilers and meat chickens were
not owned by farmers (USDA/ERS, 1999a). This compares to the turkey and layer sectors where
an estimated 70 percent of turkeys and 37 percent of layers were not owned by the farmers that
raise them, respectively (USDA/ERS, 1999a). Percentages vary across farm sizes and indicate
large differences in the organization and ownership across these three sectors. For the most part,
the broiler sector appears to be dominated by production contracting since most operators— both
small and large—do not claim ownership of their flocks. By comparison, a greater share of
smaller-sized layer and turkey operations report that they do not own their flocks while larger
operations in these sectors do own their animals, indicating greater use of contract arrangements
among smaller operators (USDA/ERS, 1999a).
Turkey production, while dominated by integrators and year-round confinement buildings,
is not as industrialized as the broiler sector (Hayenga et al., 1996). Contract arrangements are the
primary means of procuring turkey meat, accounting for 65 percent of all output in 1990, most of
which consists of production contracts (Kohls and Uhl, 1998). Owner-integrated and non-
integrated (independent) enterprises accounted for 28 percent and 7 percent, respectively, in 1990
(Kohls and Uhl,. 1998).
Information on the organization of the U.S. egg industry is limited. Large regional
cooperatives dominate the U.S. egg industry. Because of spoilage, fragility, and corresponding
transportation costs, shell eggs are hot a national market. Shell egg processing is often managed
directly by retailers. Integrators have achieved economies of scale by concentrating both the
packaging and handling of eggs into larger, more automated facilities. Available information on
this sector from the 1980s indicate that contracting accounts for over one-half of all egg
production, but the arrangements are less formalized than in the poultry meat sectors (Kohls and
Uhl, 1998). Owner-integrated enterprises accounted for nearly 40 percent This form of
integration is one in which the producing and marketing firms are the same. One-tenth of annual
output was from independent businesses (Kohls and Uhl, 1998).
6-31
-------
6.3 CAFO ANALYSIS
This section presents the results of EPA's CAFO level analysis for the poultry sectors,
including broiler, turkey, and egg laying operations. As discussed in Section 4, EPA uses a
representative farm approach to estimate the impact of the proposed CAFO regulations on
affected operations. Each model CAFO differs by facility size groupings and key farm production
regions. For these sectors, the production regions reflected in this analysis are the Mid-Atlantic
(MA), Midwest (MW) and South (SO) regions, as defined in Table 4-1 (Section 4). Section 4
provides a summary of how EPA developed the various financial models used for this analysis.
The Development Document (USEPA, 2000a) provides additional information on the cost models
developed by EPA.
Results presented in this section focus on the "BAT Option" that refers to EPA's
proposed technology option for the CAFO regulations (described in Section 3). For the purpose
of this discussion, the "two-tier structure" refers to the combination of BAT Option 5 for the
poultry subcategory and NPDES Scenario 4a that covers all operations with more than 500 AU.
Where indicated, the two-tier structure may refer to the alternative threshold at 750 AU (Scenario
5). The "three-tier structure" refers to the combination of BAT Option 5 (poultry subcategory)
and NPDES Scenario 3 that covers operations down to 300 AU based on certain conditions.
Results for other technology options and scoping scenarios considered by EPA as part of this
rulemaking are also summarized. Table 3-1 summarizes EPA's proposed and alternative ELG
Options and NPDES Scenarios discussed in this section.
Section 6.3.1 presents a summary of the cost input data that EPA uses for this analysis,
including (post-tax) per-animal and per-facility costs for EPA's model CAFOs. Section 6.3.2
presents EPA's estimate of the aggregate, national level costs of the proposed CAFO regulations
for the poultry sector. Section 6.3.3 presents EPA's predicted financial impacts to this sector in
terms of the estimated number and percentage of CAFOs that are expected to experience financial
stress as a result of the proposed CAFO regulations. EPA evaluates economic impacts to CAFOs
in this sector two ways—assuming that a portion of me costs may be passed on from the CAFO
to the consumer (Partial CPT) and assuming that no costs pass through so that all costs are
absorbed by the CAFO (Zero CPT).
6.3.1 Overview of Cost Input Data
Tables 6-18 through 6-20 presents estimated input costs that EPA uses for this analysis to
assess costs and impacts to the broiler, turkey, and egg sectors. These data include the post-tax
annualized compliance costs, estimated on a per-animal and per-facility. These costs reflect
EPA's estimated capital costs, annual operating and maintenance costs, start-up or first year
costs, and also recurring costs (discussed in the Development Document, USEPA, 2000a). These
6-32
-------
facility costs are annualized using the approach described in Appendix A of this report. Appendix
A shows the individual sector costs by model across all technology options.4
Other input data for this analysis include EPA's estimate of the number of affected CAFOs
and baseline financial conditions at model CAFOs. EPA's estimate of the number of animal
confinement operations that would be defined or designated as CAFOs is presented in
Section 6.1.2.1 (see Table 6-2). Additional information is provided in Section 2 of this report.
The average baseline financial conditions for model CAFOs that EPA assumes for this analysis are
presented in Section 4. Tables 4-5 through 4-9 in that section present the financial data used in
this analysis and include gross farm revenues, net cash flow, and debt-to-asset ratios for this
sector, as derived by EPA using the 1997 ARMS data.
Table 6-18. Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 5) for Broilers
Sector
Broiler
Reg
SO
MA
Model
Ml(a)
Ml(b)
M2
LI
L2
Ml(a)
Ml(b)
M2
LI
L2
Average
Animals
Per
Facility
36,634
51,362
73,776
117,581
281,453
36,796
51,590
73,590
115,281
303,155
Cat. 1
Cat. 2
Cafe3
Per Animal
Cafcl
Cat. 2
Cat. 3
Per Facility
($1997)-
$0.15
$0.14
$0.13
$0.13
$0.11
$0.13
$0.13
$0.12
$0.12
$0.10
$0.10
$0.09
$0.08
$0.08
$0.06
$0.11
$0.10
$0.10
$0.09
$0.07
$0.08
$0.07
$0.07
$0.06
$0.05
$0.08
$0.07
$0.07
$0.07
$0.05
$5,341
$7,172
$9,924
$15,296
$29,715
$4,948
$6,612
$9,051
$13,744
$29,187
$3,614
$4,647
$6,177
$9,109
$16,77
6
$4,134
$5,353
$7,167
$10,74
6
$20,39
7
$2,826
$3,706
$4,953
$7,508
$14,834
$2,901
$3,784
$5,023
$7,522
$16,126
Source: USEPA. See Table 4-1 for model sizes and regions. Costs reflect the estimated capital costs, annual
operating and maintenance costs, start-up or first year costs, and also recurring costs assumed by EPA (see the
Development Document, USEPA, 2000a) that are annualized using the approach described in Appendix A.
4The estimated costs are the same across the NPDES Scenarios, i.e., technology option costs do not
change by scenario, although total costs change due to the difference in numbers of CAFOs affected under each
scenario.
6-33
-------
Tables 6-18 through 6-20 present the estimated post-tax annualized compliance costs per
animal (in 1997 dollars) for each of the poultry sector under the proposed BAT Option (Option
5). EPA estimates post-tax costs for these sectors to range from $0.05 per bird to $0.15 per bird
in the broiler sector, from $0.02 per bird to $0.60 per bird for the layer sector, and from $0.05 per
bird to $0.83 per bird for the turkey sector (Tables 6-18 through 6-20). The range in costs is
explained by differences in the assumed availability of land for manure applications (see definition
of Category 1,2, and 3 hi Section 4.1.2), region, and size of operation. In general, the annualized
post-tax compliance costs per representative CAFO increase with model size. These tables also
present the range of post-tax annualized compliance costs per CAFO. Per CAFO compliance
costs range from $2,830 to $29,720 in the broiler sector, from $660 to $53,090 for the layer
sector, and from $1,640 to $102,790, in the turkey sector (Tables 6-18 through 6-20). As
documented'in the Development Document (USEPA, 2000a), EPA believes that its estimated
costs are conservative.
fi-19 Per-Animal and Per-Facilitv Post-tax Annualized Compliance Costs (Option 5) for Layers
Sector
Wet
Layers
Dry
Layers
Reg
•
SO
SO
MW
Model
M2
LI
Ml(a)
Ml(b)
M2
LI
L2
Ml (a)
Ml(b)
M2
LI
L2 •
Average
Animals
Per
Facility
3,654
86,898
32,375
' 44,909
97,413
293,512
884,291
37,906
52,582
97,484
279,202
1,229,095
Cat. 1
Cat. 2
Cat. 3
Per Animal
Cat. 1
Cat. 2
Cat. 3
Per Facility
($1997)
$0.55
$0.27
$0.18
$0.14
$0.15
$0.18
$0.00 ^
$0.25
$0.27
$0.23
$0.19
$0.00 ^
$0.60
$0.27
$0.04
$0.04
$0.03
$0.02
$0.02
$0.05
$0.04
$0.04
$0.03
$0.02
$0.39
$0.15
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0:02
$0.02
$0.02
$0.02
$2,018
$23,885
$5,783
$6,241
$14,213
$51,613
$0a/
$9,418
$14,280
$22,246
$53,092
$0^
$2,197
$23,70
4
$1,372
$1,630
$3,009
$6,547
$16,98
3
$1,813
$2,166
$3,722
$7,476
$26,40
1
$1,420
$12,856
$664
$861
$1,602
$4,593
$13,582
$868
$1,158
$1,886
$5,201
$22,440
Source: USEPA. See Table 4-1 for model sizes and regions. Costs reflect the estimated capital costs, annual
operating and maintenance costs, start-up or first year costs, and also recurring costs assumed by EPA (see the
Development Document, USEPA, 2000a) that are annualized using the approach described in Appendix A.
"'EPA did not estimate costs for layer operations under Options 2 through 7.
6-34
-------
The costs presented here are those assumed to be incurred by the regulated CAFO and do
not account for the likelihood that some compliance costs will be passed on through the
marketing levels in the industry.
Table 6-20. Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 5) for Turkeys
Sector
Turkeys
Reg.
MA
MW
Model
Ml(a)
Ml(b)
M2
LI
Ml(a)
Ml(b)
M2
LI
Average
Animals
Per
Facility
18,539
- 31,267
45,193-.
97,111
18,092
30,514
45,469
158,365
Cat. 1
Cat. 2
Cat. 3
Per Animal
Cat. 1
Cat. 2
Cat. 3
Per Facility
($1997)
$0.71
$0,70
$0.61
$0.57
$0.83
$0.79
$0.65
$0.65
$0.29
$0.23
$0.18
. $0.16
$0.28
$0.22
$0.18
$0.15
$0.09
$0.07
$0.06
$0.05
$0.13
$0.12
$0.10
$0.10
$13,251
$21,774
$27,353
$55,756
$15,008
$23,972
$29,571
$102,793
$5,297
$7,066
$7,988
$15,38 .
0
$5,042
$6,702
$7,999
$23,60
9
$1,641
$2,265
$2,500
$4,513
$2,418
$3,679
$4,603
$15,278
Source: USEPA. See Table 4-1 for model sizes and regions. Costs reflect the estimated capital costs, annual
operating and maintenance costs, start-up or first year costs, and also recurring costs assumed by EPA (see the
Development Document, USEPA, 2000a) that are annualized using the approach described in Appendix A.
Table 6-21 presents the range of per animal post-tax compliance costs in 1997 dollars for
broiler, layer, and turkey operations. (The proposed and alternative ELG Option and NPDES
Scenarios considered by EPA during this rulemaking are defined in Table 3-1.) For broilers,
option costs range from $0.05 to $0.28 per animal. For layers, option costs range from $0.01 to
$1.95 per animal. For turkeys, option costs range from $0.05 to $0.98 per animal. As shown, the
proposed BAT Option (Option 5) is the second least expensive option for the poultry sectors.
6.3.2 Estimates of National Annual Compliance Costs
Table 6-22 presents EPA's estimate of the aggregate national level compliance costs for
the poultry sectors under the proposed BAT Option (Option 5) and the co-proposed two-tier
structure (Scenario 4a at 500 AU threshold) and the three-tier structure (Scenario 3). Costs
under the two-tier structure at the 750 AU threshold (Scenario 5) are also briefly discussed, along
6-35
-------
Table 6-21. Summary of the Range of Post-Tax Annualized Compliance Costs Per Animal, By Option
Option
Option 1
Option 2
Option 3
Option 4
Option 5
Option 6
Ootion 7
Broiler
Minimum
Maximum
Layer ^
Minimum
Maximum
Turkey
Minimum
Maximum
- ($1997)
$0.05
$0.05
$0.05
$0.07
$0.05
$0.05
$0 05
$0.09
$0.15
$0.18
$0.28
$0.15
$0.15
$0.15
$0.01
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.55
$0.60
$0.93
$1.95
$0.60
$0.60
$0.60
$0.05
$0.05
$0.05
$0.09
$0.05
$0.05
$0.05
$0.17
$0.83
$0.92
$0.98
$0.83
$0.83
$0.83
Source: USEPA.
"tPA did not estimate costs for layer operations under Options 2 through 7.
with, other regulatory alternatives considered by EPA during this rulemaking. The description of
the proposed BAT Option and the co-proposed NPDES Scenarios is provided in Section 3.
Across all the poultry sectors, EPA estimates total incremental costs (post-tax) of the
proposed BAT Option at $97 million per year under the two-tier structure and $117 million per
year under the three-tier structure (Table 6-22). About one-half of this total estimated cost is for
operations with more than 1,000 AU. The majority of compliance costs would be incurred by the
broiler sector, estimated at about 80 percent of total costs to these sectors (Table 6-23). Table 6-
22 shows estimated costs for each sector. (EPA estimates that the costs of the proposed BAT
option under the two-tier structure at 750 AU threshold will total $78 million per year across the
poultry sectors.)
6.3.3 Analysis of CAFO Financial Impacts
EPA's impact analysis uses a representative farm approach to estimate the number of
CAFOs that would experience affordable, moderate, or stress impacts as a result of the CAFO
regulations, as described in Section 4. Economic achievability is determined by applying the
proposed criteria, which include a sales test and also analysis of post-compliance cash flow and
debt-to-asset ratio for an average model CAFO. Impacts are extrapolated to all CAFOs in the
poultry sector using the estimated number of operations represented by each model CAFO.
As described hi Section 4.2.5, if an average model facility is determined to incur economic
impacts under regulation that are regarded as "Affordable" or "Moderate," then the proposed
regulations are considered economically achievable. ("Moderate" impacts are not expected to
result in closure and are considered to be economically achievable by EPA.) If an average
operation is determined to incur "Stress," then the proposed regulations are not considered to be
6-36
-------
Table 6-22. Total Estimated Post-Tax Compliance Costs
Scenario/Size
>1,000 AU
Total
Alternative
Two-Tier
Structure (>750
AU)
Total Two-Tier
Structure (>500
AU)
Total Two-Tier
Structure (>300
AU)
Total Three-
Tier Structure
(>300 AU)
Option
Number of CAFOs
Cost of Proposed
BAT Option
Cost of Alternative
Options
Number of CAFOs .
Cost of Proposed
BAT Option
Cost of Alternative
Options
Number of CAFOs
Cost of Proposed
BAT Option
Cost of Alternative
Options
Number of CAFOs
Cost of Proposed
BAT Option
Cost of Alternative
Options
Number of CAFOs
Cost of Proposed
BAT Option
Cost of Alternative
Options
Broilers
Wet Layers
Dry Layers
Turkeys
($1997 millions)
3,940
$41.8
$37.3 - $61.5
7,780 .
$60.8
$53.9 - $100.9
9,780
$74.4
$66.2 - $124.0
14,120
$89.7
$81.8-$161.5
13,740
$90.0
$78.1 - $159.7
50
$0.9
$0.9 -$1.1
320
$1.2
$1.2 - $2.4
360
$1.5
$1.4 -$3.2
360
$1.5
$1.4-$3.2
360
$1.4
$1.4 -$3.7
590
. $5.4
$4.7 - $7.6
1,140
$7.1
$5.6 -$11. 8
1,280
$7.6
$5.9 - $12.5
1,700
$8.4
$6.3-$15.0
1,660
$8.4
$6.0 -$14.4
370
$6.8
$4.7 -$8.6
740
$9.1
$5.9 - $12.6
1,280
$13.3
$8.3 - 19.4
2,100
$17.3
$10.3-$27.6
2,040
$17.4
$10.2 - $27.4
Source: USEPA. Numbers of CAFOs include defined CAFOs only. Costs include those for designated operations.
economically achievable. "Affordable" and "Moderate" impacts are associated with positive post-
compliance cash flow over a 10-year period and a debt-to-asset ratio not exceeding 40 percent, in
conjunction with a sales test result that shows that compliance costs are less than 5 percent of
sales ("Affordable") or between 5 and 10 percent ("Moderate"). "Stress" impacts are associated
with negative cash flow or if the post-compliarice debt-to-asset ratio exceeds 40 percent, or sales
test results that show costs equal to or exceeding 10 percent of sales.
6-37
-------
Using this classification scheme, EPA's analysis indicates that some poultry operations
would experience financial stress as a result of the proposed CAFO regulations under the
proposed BAT Option and both co-proposed scenario, assuming compliance costs cannot be
passed through the marketing chain.
Table 6-23 presents the results of EPA's analysis for layer and turkey operations. As
shown, none of the model facilities evaluated for these sectors are estimated to experience
financial stress as a result of the proposed regulation. Given these results, EPA did not conduct
further analysis to examine the economic impacts to these sectors under a cost passthrough
scenario.
Table 6-24 presents the results of EPA's analysis for broiler operations. As shown, EPA's
analysis projects that a total of 150 broiler operations (one percent of all poultry operations)
might experience financial stress under the two-tier structure. Alternatively, results for the two-
tier structure at 750 AU threshold also indicate that 150 broiler operations might experience
financial stress. Under the three-tier structure, EPA estimates that a total of 330 broiler
operations (2 percent of all poultry operations) might experience financial stress. Under the two-
tier structure, all broiler operations with stress impacts have more than 1,000 AU. Under the
three-tier structure, affected broiler facilities include operations with more than 1,000 AU, as well
as operations with less than 1,000 AU. EPA does not expect that any designated broiler
operations will be impacted under the stress category under either co-proposed tier structure.
These results assume that no cost are passed through to consumers. However, EPA expects that
long-run market and structural adjustment producers in these sectors will diminish the estimated
impacts to these sectors as costs are passed through to consumers.
For the broiler sector, EPA also evaluates financial impacts with an assumption of cost
passthrough. For the purpose of this analysis, EPA assumes that the broiler sector could
passthrough 35 percent of compliance costs. EPA derives these estimates from price elasticities
of supply and demand for each sector reported in the academic literature (see Section 4).
Assuming this level of cost passthrough, the magnitude of the estimated impacts decreases to the
affordable or moderate impact category under the proposed BAT Option and the co-proposed
scenarios (Table 6-24). When partial CPT is assumed in the broiler sector, no poultry operations
experience stress impacts under the proposed BAT Option and all regulatory scenarios
investigated. However, a total of 1,170 broiler operations (12 percent of poultry operations) are
estimated to experience moderate impacts under the two-tier structure (500 AU threshold), and a
total of 1,440 broiler operations (10 percent of all poultry operations) are estimated to experience
moderate impacts under the three-tier structure, using the partial CPT assumption. EPA does not
consider moderate impacts to adversely affect a finding of economic achievability. Based on the
results presented here, EPA proposes that the proposed CAFO regulations are economically
achievable under the co-proposed scenarios.
6-38
-------
Table 6-23. Impacted CAFOs Under ELG Options & NPDES Scenarios, Layer and Turkey Operations
Alternative
ELG Options
and
NPDES Scenarios
Two-Tier (>1000)
Number of CAFOS
BAT Option
Alt. ELG Options
Layers
Affordable
Moderate
Stress
Turkeys
Affordable
Moderate
Stress
(Number of Affected Operations)
640
640
640
0
0
0
0
370
370
370
0
0
0
0
Two-Tier (>750 AU, Scenario 5)
Number of CAFOS
BAT Option
Alt. ELG Options
1,460
1,460
1,310-1,460
0
0-160
0
0
740
720
720-740
20
0-20
0
0
Two-Tier (>500 AU, Scenario 4a)
Number of CAFOS
BAT Option
Alt. ELG Options
1,640
1,640
1,330-1,640
0
0-310
0
0
1,280
1,230
1,230-1,280
50
0-50
0
0
Two-Tier (>300 AU, Scenario 4b)
Number of CAFOS
BAT Option
Alt. ELG Options
2,060
2,060
1,750-2,060
0
0-310
0
0
2,100
1,990
1,990-2,100
110
0-110
0
0
Three-Tier (Scenario 3)
Number of CAFOS
BAT Option
Alt. ELG Options
2,010
2,010
1,700-2,010
0
0-310
0
0
2,060
1,950
1,950-2,060
110
0-110
0
0
Source: USEPA. Does not include impacts to designated CAFOs.
Compared to the alternative options, the proposed BAT Option results in stress impacts
that are generally greater than Option 1 impacts (the low end of the range shown in Tables 6-23
and 6-24) and substantially less than Option 4 impacts (the high end of the range shown in
Tables 6-23 and 6-24). Stress impacts for other options are similar to the BAT Option. Section 5
provides additional information that compares the co-proposed scenarios with other alternative
scenarios.
6-39
-------
Tahlp fi-24 Tmnacted CAFOs Under ELG Options & NPDES Scenarios, Broiler Operations
Alternative
ELG Options
and
NPDES Scenarios
Total
#
CAFOs
Aff.
Moderate
Stress
Zero Cost Passthrough
Affordable
Moderate
Stress
Partial Cost Passthrough
(Number of Affected Operations)
Two-Tier (>1000)
BAT Option
Alt. ELG Options
3,940
200
0-500
3,600
1,670-3,800
150
0-2,270
3,080
630-3,940
Two-Tier (>750 AU, Scenario 5)
BAT Option
Alt ELG Options
7,780
1,650
0-1,950
5,980
3,120-7,520
150
0-4,660
6,740
630-7,780
860
0-3,310
1,040
0-7,150
Two-Tier (>500 AU, Scenario 4a)
BAT Option
Alt ELG Options
9,780
1,960
0-2,270
7,670
3,120-9,390
150
0-6,660
• 8,610
630-9,780
Two-Tier (>300 AU, Scenario 4b)
BAT Option
Alt ELG Options
14,140
1,960
0-2,270
11,860
3,120-13,320
320
0-11,020
12,690
630-14,140
Three-Tier (Scenario 3)
BAT Option
Alt ELG Options
13,740
1,850
0-2,150
11,580
3,010-12,940
330
0-10,750
12,320
630-13,740
1,170
0-9,020
1,450
0-10,770
1,420
0-10,510
0
0
0
0
0
0-130
0
0-
2,740
0
0-
2,610
Source: USEPA. Does not include impacts to designated CAFOs.
Tables 6-25 and 6-26 present a more detailed breakout of EPA's affordability results
under the proposed BAT Option by model CAFO type, land availability, and type of operation
(broiler, layer, turkey). The results are the same for the two-tier and three-tier structure because
only the numbers of CAFOs represented by each model type changes. The impacts are presented
by model CAFO and indicate the level of impact under each of the economic affordability criteria.
These results reflect a "zero" cost passthrough assumption.
These tables show that show that financial stress impacts for broilers are being driven by
the revenue test in the Large 1 models in both regions (revenue test of greater than 10 percent is
considered an indication of financial stress). Broiler operations, however, primarily operate under
production contracts, regardless of size. Operations with production contracts are often
6-40
-------
Table 6-25. Economic Achievability Results for Broiler CAFOs
Model
Category 1
Sales
DCF
DA
, Category 2
Sales
DCF
DA
Category 3
Sales
DCF
DA
ZeroCPT
SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
10.2%
9.8%
9.4%
11.3%
9.1%
Pass
Pass
Pass
Pass
Pass
0.25
0.25
0.25
0.33
0.33
6.9%
6.3%
5.9%
6.7%
5.2%
Pass
Pass
Pass
"Pass
Pass
0.25
0.25
0.25
0.33
0.33
5.4%
5.1%
4.7%
5.5%
4.6%
Pass
Pass
Pass
Pass
Pass
0.25
0.25
0.25
0.33
0.33
MA Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
9.2%
8.7%
8.4%
10.5%
8.5% '
Pass
Pass
Pass
Pass
Pass
0.27
0.27
0.27
0.37
0.37
7.6%
7.1%
6.6%
8.2%
6.0%
Pass
Pass
Pass
Pass
Pass
0.27
0.27
0.26
0.37
0.37
5.4%
5.0%
4.6%
5.8%
4.7%
Pass
Pass
Pass
Pass
Pass
0.26
0.26
0.26
0.37
0.37
Partial CPT
SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
6.7%
6.4%
6.1%
7.3%
5.9%
Pass
Pass
Pass
Pass
Pass
0.23
0.23
0.23
0.31
0.31
4.5%
4.1%
3.8%
4.4%
3.4%
Pass
Pass
Pass
Pass
Pass
0.23
0.23
0.23
0.31
0.31
3.5%
3.3%
3.1%
3.6%
3.0%
Pass
Pass
Pass
Pass
Pass
0.23
0.23
0.23
0.31
0.31
MA Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
6.0%
5.7%
5.4%
6.9%
5.5%
Pass
Pass
Pass
Pass
Pass
0.25
0.25
0.25
0.35
0.35
5.0%
4.6%
4.3%
5.4%
3.9%
Pass
Pass
Pass
Pass
Pass
0.25
0.25
0.25
0.35
0.35
3.5%
3.2% .
3.0%
3.8%
3.1%
Pass
Pass
Pass
• Pass
Pass
0.25
0.25
0.25
0.34
0.34
Source: USEPA.
6-41
-------
Table 6-26. Economic Achievability Results for Layer and Turkey CAFOs (
Model
Category 1
Sales
DCF
DA
Category 2
Sales
DCF
DA
Category 3
Sales
DCF
DA
. Egg Laying Operations
Wet SO Region
Medium 2
Large 1
2.2%
1.1%
Pass
Pass
0.12
0.11
2.4%
1.1%
Pass
Pass
0.12
0.11
1.6%
. 0.6%
Pass
Pass
0.12
0.11
Dry SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
0.7%
0.6%
0.6%
0.7%
NA
Pass
Pass
Pass
Pass
NA
0.11
0.11
0.11
0.11
NA
0.2%
0.1%
0.1%
0.1%
0.0%
Pass
Pass
Pass
Pass
Pass
0.11
0.1.1
0.11
0.11
0.11
0.1%
0.1%
0.1% •
0.1%
0.0%
Pass
Pass
Pass
Pass
Pass
0.11
0.11
0.11
0.11
0.11
Dry MW Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
1.0%
1.1%
0.9%
NA
NA
Pass
Pass
Pass
NA
NA
0.11
0.11
0.11
NA
NA
0.2%
0.2%
0.2%
0.1%
0.0% '
Pass
Pass
Pass
Pass
Pass
0.11
0.11
0.11
0.11
0.11
0.1%
0.1%
0.1%
0.1%
0.0%
Pass
Pass
Pass
Pass
Pass
0.11
0.11
0.11
0.11
0.11
Turkey Operations
MA Region
Medium la
Medium Ib
Medium 2
Large 1
MW Region
Medium la
Medium Ib
Medium 2
Lar&e I
6.4%
6.2%
5.4%
2.9%
Pass
Pass
Pass
Pass
0.24
0.24
0.24
0.16
2.5%
2.0%
1.6%
0.8%
Pass
Pass
Pass
Pass
0.24
0.24
0.24
0.16
0.8%
0.6%
0.5%
0.2%
7.4%
7.0%
5.8%
3 2%
Pass
Pass
Pass
Pass
0.25
0.25
0.25
0.17
2.5%
2.0%
1.6%
0.7%
Pass
Pass
Pass
Pass
0.25
0.25
0.25
0.17
1.2%
1.1%
0.9%
0.5%
Pass
Pass
Pass
Pass
Pass
Pass
Pass
Pass
0.24 .
0.24
0.24
0.16
„
0.25
0.25
0.25
0.17
Source: USEPA.
6-42
-------
associated with lower revenues and lower costs than operations without a production contract.
Thus, the revenue test may not be a fair assessment of the level of impact on these operations,
particularly if zero CPT is assumed. Under an assumption of partial cost passthrough, all model
CAFOs show revenue tests below 10 percent.
/
EPA believes there may be mitigating circumstances affecting financial impacts in the
poultry sectors other than the potential for cost passthrough. Poultry manure is a valuable soil
amendment. One beneficial effect of complying with the proposed CAFO regulations is that
poultry operations would produce a poultry litter with low moisture content, which is more easily
handled than liquid manures. This manure can, in many cases, be sold to other agricultural
operations for crop amendment purposes. EPA does not include a cost offset for sale of dry
poultry litter in the model cost estimates, as discussed in Section 4.2.7. EPA did, however,
conduct a sensitivity analysis to examine the potential impact such gains could have on analytical
results.
EPA uses an offset of 40 cents per head to estimate manure sales (U.S. EPA, 2000k).
EPA estimates that sales of dry poultry litter in the broiler sector could offset the total costs of
meeting the requirements of the proposed CAFO regulations by approximately one-half. This
would reduce net costs from an estimated $74 million to about $36 million under the co-proposed
two-tier structure. Net costs would be reduced by from an estimated $10 million to about $45
million under the three-tier structure.
Table 6-27 presents the results of EPA's analysis for broiler CAFOs assuming a manure
sale offset for the two-tier and three-tier structures. As shown, if manure is sold, then no broiler
operations are expected to incur stress impacts as a result of the proposed CAFO regulations.
Additionally, only 4 percent of all broiler CAFOs (2 percent of operations with more than 1,000
AU and 6 percent of operations with less than 1,000 AU) are estimated to experience moderate
impacts, m comparison, under a partial CPT assumption of 35 percent for broilers (without
considering manure sales), 5 percent of all broiler operations (4 percent of 1 operations with more
than 1,000 AU and 6 percent of operations with less than 1,000 AU) are estimated to incur
moderate impacts. The analysis shows that sales of poultry litter could offset the cost of CAFO
regulatory requirements even more than the CPT assumption assumed for this analysis.
Furthermore, manure sales could offset these costs sufficiently to eliminate all stress impacts
without the need to assume any cost passthrough.
Past analyses that estimate costs to livestock and poultry producers for environmental
improvements at the farm site often examine the potential for the value of manure (expressed in
terms of nitrogen, phosphorus, and potassium) to offset production costs associated with capital
improvements and annual operation and maintenance costs (see, e.g., Christensen, et al., 1981).
6-43
-------
Table 6-27. Number and Percenta
Size
>1,000 AU
500-1,000 AU
(Two-Tier)
300-1,000 AU
(Three-Tier)
>500 AU
(Two-Tier)
>300 AU
(Three-Tier)
Number
ofCAFOs
3,940
5,840
9,820
9,780
13,720
ge of Affected Broiler CAFOs (Manure Sales Assumption)
Affordable
(number)
3,770
5,530
9,240
9,300
13,020
(percent)
96%
95%
94%
95%
95%
Moderate
(number)
170
310
580
480
740
(percent)
4%
5%
6%
5%
5%
Stress
(number)
0
0
0
0
0
(percent)
0%
0%
0%
0%
0%
Source: USEPA. Proposed BAT Option is Option 5. Results do not include designated CAFOs.
Assume manure sales valued at $.40/head as an offset to estimate compliance costs.
6.4 PROCESSOR ANALYSIS
As discussed in Section 4.3, EPA does not conduct a detailed estimate of the costs and
impacts that would accrue to individual co-permittees due to lack of data and market information.
However, EPA believes that the framework used to estimate costs to CAFOS provides a means to
evaluate the possible upper bound of costs that could accrue to potential co-permittees, based on
the potential share of (pre-tax) costs that may be passed on from the CAFO (described in Section
4.3). EPA is proposing that this amount approximates the magnitude of the costs that may be
incurred by processing firms in those industries that may be affected by the proposed co-
permitting requirements.
Table 6-28 presents the results of EPA's analysis. This analysis focuses on the potential
magnitude of costs to co-permittees in the poultry sector. As presented in Section 2, EPA
estimates that about 227 broiler processors may be subject to the proposed co-permitting
requirements. EPA does not evaluate the potential magnitude of costs to egg and turkey
processors because the compliance costs to CAFOs in these industries are expected to be easily
absorbed by CAFOs (see Table 6-23). Using the framework to estimate costs and impacts to
regulated CAFOs, EPA calculates the estimated upper bound of costs that could accrue to broiler
processors based the estimated pre-tax cost estimated for CAFOs, assuming that either all or a
portion of these costs are absorbed by processors as markets adjust to the proposed CAFO
regulations. EPA's partial cost passthrough scenario assumes that 35 percent of all broiler
compliance costs are passed on to the food processing sectors. (For more information on this
approach, see Section 4.2).
6-44
-------
.Using this approach, EPA estimates that the range of potential annual costs to broiler
processors is $34 million (partial cost passthrbugh, two-tier structure) to $117 million (full cost
passthrough, three-tier structure). These estimates are expressed in 1999 pre-tax dollars.
To assess the magnitude of impacts that could accrue to processors using this approach,
EPA compares the passed through compliance costs to both aggregate processor costs of
production and to revenues (a sales test). The results of this analysis are shown in Table 6-28 and
are presented in terms of the equivalent 1999 pre-tax compliance cost as compared to 1997 data
from the Department of Commerce on the revenue and costs among processors in the broiler ,
industries. As shown, EPA estimates that, even under full cost passthrough, incremental cost
changes are less than two percent and passed through compliance costs as a share of revenue are
less than one percent.
This suggested approach does not assume any addition to the total costs of the rule as a
result of co-permitting. This approach also does not assume that there will be a cost savings to
contract growers as result of a contractual arrangement with a processing firm. This approach
merely attempts to quantify the potential magnitude of costs that could accrue to processors that
may be affected by the co-permitting requirements. Due to lack of data, EPA did not conduct a
detailed analysis of the costs and impacts that would accrue to individual co-permittees. Additional limitations
of this approach as recognized by EPA are discussed in Section 4.3.
Table 6-28. Impact of Passed Through Compliance Costs under Co-proposed Alternatives, Broiler Sector
Sector
Two-Tier
Structure
Three-
Tier
Structure
Passed Through
Pre-tax
Compliance Cost **
Partial
CPT
100%
CPT
1997
Revenues
1997
Delivered
Cost"'
$1999 (million)
$34
$41
$97
$117
$17,700
$9,100
Passed Through
Compliance Cost
as % of Revenues
(Sales Test)
Partial
CPT
100%
CPT
Passed Through
Compliance Cost as %
of Delivered Cost
Partial
CPT
100%
CPT
(percent)
0.2%
0.2%
0.5%
0.6%
0.4%
0.4%
1.0%
1.2%
Source: USEPA. 1997 processor revenues and costs are from the Department of Commerce (USDC, 1999a).
Option/Scenario definitions are provided in Section 3. Proposed BAT Option is Option 5.
"'Pre-tax compliance costs that are estimated to be passed from the CAFO to the processors using a mid-range CPT
of 35% for the broiler sector (see Section 4.2.6.1).
b/Delivered costs include all raw materials put into production during the year.
6-45
-------
6.5 MARKET ANALYSIS
This section presents the results of EPA's market model analysis for the poultry sectors.
The results presented in this section briefly compare the results of the two-tier (500 AU threshold)
and the three-tier (Scenario 3) structures that are being co-proposed by EPA. Additional results
on the alternative regulatory options and scenarios considered by EPA as part of this rulemaking
are provided in Section 5.4. For further explanation of the market model and sources of the
baseline input data, see Section 4.4 and Appendix B.
A summary of the key results of the market model for the broiler, layer, and turkey sectors
is shown in Tables 6-29, 6-30, and 6-31 for both the two-tier and three-tier structures. These
tables indicate the predicted changes in farm and retail prices, quantities, national and regional
employment, and national economic output.
Compared to a baseline producer price of 37 cents per pound (hi 1997 dollars), EPA's
market model predicts that the proposed CAFO regulations will raise broiler producer prices by
0.19 cents per pound to 0.22 cents per pound, or less than 0.06 percent of the baseline producer
price, depending on the co-proposed tier structure (Table 6-29). At the retail level, consumer
prices for broiler products will rise about 0.2 cents per pound.
From a baseline producer price of 69.8 cents per dozen, EPA's market model predicts that
the proposed CAFO regulations will raise egg prices by 0.13 cents per dozen to 0.14 cents per
dozen, or less than 0.03 percent of the baseline producer price, depending on the co-proposed tier
structure (Table 6-30). At the retail level, consumer prices for eggs will rise about 0.14 cents per
dozen.
Compared to a baseline producer price of 40.1 cents per pound, EPA's market model
predicts that the proposed CAFO regulations will raise turkey producer prices by 0.12 cents per
pound to 0.16 cents per pound, or less than 0.4 percent of the baseline producer price, depending
on the co-proposed tier structure (Table 6-31). At the retail level, consumer prices for turkey
products will rise about 0.15 cents per pound".
These price increases are driven by slight changes in the amount of poultry products
produced at the farm level and thus available for consumption. At the commodity level, EPA's
market model predicts that U.S. poultry product imports will not change compared to baseline
imports. U.S. broiler exports will decrease by less than 0.1 percent compared to baseline, while
egg and turkey exports will remain unchanged.
Absorption of compliance costs by the producers and small declines in quantities are
expected to result in fewer jobs in the poultry industry. Tables 6-29 through 6-31 also present
EPA's estimates of both the direct (i.e., farm and processor level) and total (i.e., national level)
reductions in employment for the poultry sector. Overall, changes in national aggregate
employment in the broiler sector are estimated to range from a total reduction of 1,870 to 2,260
6-46
-------
jobs, measured in full-time equivalents (FTEs). Changes in national aggregate employment in the
layer sector are estimated to range from 200 to 220 FTEs. Changes hi national aggregate
employment hi the turkey sector are estimated to range from 370 to 490 FTEs. This analysis does
not adjust for offsetting increases hi other parts of the economy and other sector employment that
may be stimulated as a result of the proposed regulations, such as the construction and farm
services sectors.
. EPA's projected job losses are estimated throughout the entire economy, using available
modeling approaches described in Section 4, and are not attributable to the regulated community
only. As shown hi Tables 6-29 through 6-31, about 80 percent of these estimated job losses are
hi the non-agricultural or farm services support industries (i.e., indirect or induced employment
affects; see Section 4.4).
At the CAFO level, EPA predicts that job losses hi the broiler production sector
associated with the proposed CAFO regulations will range from 340 to 410 jobs under the
proposed BAT Option, depending on tier structure (Table 6-29). EPA estimates that job loss'es hi
the layer sector at the CAFO level will range from 13 to 15 jobs (Table 6-30) and that job losses
hi the turkey sector at the CAFO level will range from and 100 to 130 jobs (Table 6-31). These
estimates include CAFO owner-operators and employed family members, as well as hired farm
labor. These estimated reductions compare to an estimated total farm level employment of
71,800 FTEs hi the poultry sector nationwide (Table 2-17; Abel, Daft, and Barley, 1993, as
updated by EPA). EPA estimates that job losses hi the broiler processing sector will range from
60 to 70 jobs (Table 6-29), while fewer than 10 jobs will be lost hi either the egg or turkey
processing industries (Tables 6-30 and 6-31). These estimated losses compare to the more than
204,000 persons employed hi poultry processing hi 1997 (USDC, 1999a).
Changes hi employment and earnings can affect the vitality of local communities.
Community impacts are usually determined by employment changes at individual facilities. As
facility-specific information and analysis were not within the scope of this study, EPA is not able
to speculate on community impacts. For this analysis, EPA disaggregates the national
employment results to examine the potential regional employment impacts of the proposed CAFO
regulations. The method EPA uses to allocate impacts is based on broiler and egg production at
large operations and turkey production at all farms. This allocation does not take into account
existing environmental practices or other production factors (see Section 4.4). Table 6-29 shows
that the dominant broiler producing regions of the South would be the most affected, followed by
the Mid-Atlantic. Turkey and egg production have their largest impacts hi the Midwest. None of
the impacts represent a significant share of total employment hi these regions. Compared to the
baseline, EPA estimates the loss hi broiler agricultural employment at under 0.01 percent; about
60 percent of the estimated agricultural job losses in the broiler sector are expected hi the South
(Table 6-29). About 40 percent of the egg and turkey industry job losses are expected in the
Midwest (Tables 6-30 and 6-31). Economy-wide employment losses are estimated at under 0.01
percent compared to the baseline.
6-47
-------
Table 6-29. Summary of Market Model Results for the Broiler Sector
Variable
Pre-
Regulatory
Value/Units
Two-Tier Structure
BAT
Option
Range of
Alternative Options
Three-Tier Structure
BAT
Option
Range of
Alternative Options
Farm Products
Price
Quantity
Produced
37.000/lb.
27,551
mil. Ibs.
37.19
27,538
37.17-37.31
27,530 - 27,540
37.22
27,536
37.19 - 37.40
27,524 - 27,538
Retail Products
Price
Quantity
Demanded
Quantity
Exported
Quantity
Imported
151.00^/lb.
27,551
mil. Ibs.
5,048
mil. Ibs.
5
mil. Ibs.
151.19
27,538
5,046
5
151.17-151.31
27,530 - 27,540
5,044 - 5,046
5-5
151.22
27,536
5,045
5
151.19-151.40
- 27,524-27,538
5,043 - 5,046
5-5
Employment Reduction a/b/
Direct Farm
Direct
Processor
Total Economy
71,800
FTEs
204,200
FTEs
129.6 mil.
FTEs-
338
57
1,865
301 - 564
51-96
1,660 - 3,108
411
70
2,262
355 - 727
60 - 123
1,959 - 4,008
Output Reduction
National
$ million
185
165 - 309
225
195-398
Regional Farm and Processing Employment Reduction
•Pacific
Central
Midwest
South
Mid-Atlantic
Total
FTEs
FTEs
FTEs
FTEs
FTEs
FTEs
17
33
18
226
101
396
15-29
29-55
16-31
, 201-377
90 - 169
352 - 660
21
40
22
274
123
480
18-37
.35-71
19-40
237 - 486
106-218
416-851
Source: Post-regulatory changes are estimated by USEPA. Pre-regulatory prices, quantities, and trade volumes, see
Table 4-16 (Section 4). Pre-regulatory employment, see Table 2-17 (Section 2).
"'I FTE = 2,080 hours of labor.
w Estimated employment across all poultry sectors (Table 2-17).
6-48
-------
Table 6-30. Summary of Market Model Results for the Layer Sector
Variable
Pre-
Regulatory
Value/Units
Two-Tier Structure
BAT
Option
Range of
Alternative Options •
Three-Tier Structure
BAT
Option
Range of
Alternative Options
Farm Products
Price
Quantity
Produced
69.80e7doz.
6,473
mil. doz.
69.93
6,472
69.91-70.03
6,472-6,472
69.94
6,472
69.91 - 70.06
6,471 - 6,472
Retail Products
Price
Quantity
Demanded
Quantity
Exported
Quantity
Imported
106.0Qe7doz
5,578
mil. doz.
228
mil. doz.
7
mil. doz.
106.13
5,577
228
7
106.11-106.23
5,577 - 5,577
228 - 228
7-7
106.14
5,577
228
7
106.11 - 106.26
5,576 - 5,577
228 - 228
7-7
Employment Reduction >/b/
Direct Farm
Direct
Processor
Total
Economy
71,800
FTEs
'204,200
FTEs
129.6 mil.
FTEs
13
3
202
11-23
3-6
162 - 348
15
4
218
11-27
3-7
170 - 401
Output Reduction
National
$ million
19
15-33
20
16-38
Regional Farm and Processing Employment Reduction
Pacific
Central
Midwest
South
Mid-Atlantic
Total
FTEs
FTEs
FTEs
FTEs
FTEs
FTEs
2
. 2
6
4
3
17
2-3
1-3
5-11
3-6
3-6
14-29
2
2
7
4
4
18
2-4
1-3
5 -. 13
3-7
3-7
14-33
Source: Post-regulatory changes are estimated by USEPA. Pre-regulatory prices, quantities, and trade volumes, see
Table 4-16 (Section 4). Pre-regulatory employment, see Table 2-17 (Section 2).
"'I FTE = 2,080 hours of labor.
b/ Estimated employment across all poultry sectors (Table 2-17).
6-49
-------
Table 6-31. Summary of Market Model Results for the Turkey Sector
Variable
Pre-
Regulatory
Value/Units
Two-Tier Structure
BAT
Option
Range of
Alternative Options
Three-Tier Structure
BAT
Option
Farm Products
Price
Quantity
Produced
40.10ji/lb.
5,412
mil. Ibs.
40.22
5,409
40.18-40.28
5,407 - 5,410
40.26
5,408
Retail Products
Price
Quantity
Demanded
Quantity
Exported
Direct Farm
Direct Processor
Total Economy
105.100/lb
5,412
mil. Ibs.
598
mil. Ibs.
105.22
5,409
598
105.18 - 105.28
5,407 - 5,410
597 - 598
105.26
5,408
598
Range of
Alternative
Options
40.19-40.35
5,405-5,409
105.19 - 105.35
5,405 - 5,409
597 - 598
Employment Reduction "/b/
71,800
FTEs
204,200
FTEs
129.6 mil.
FTEs
98
9
373
61 - 142
5-13
231 - 543
127
11
486
Output Reduction
National
Pacific
Central
Midwest
South
Mid-Atlantic
$ million
38
24-55
49
Regional Farm and Processing Employment Reduction
FTEs
FTEs
FTEs
FTEs
FTEs
FTEs
10
7
41
15
34
106
6-14
4-10
26-60
9-21
21 - 49
66 - 155
13
9
54
19
44
138
; 75-200
7-18
286 - 765
29-78
7-20
5-14
32-85
11-30
26-69
81-218
Table 4-16 (Section 4). Pre-regulatory employment, see Table 2-17 (Section 2).
"'I FTE = 2,080 hours of labor.
w Estimated employment across all poultry sectors (Table 2-17).
6-50
-------
SECTION SEVEN
SUMMARY OF ECONOMIC IMPACTS:
HOG SUBCATEGORY
This section presents a profile of the hog industry, including farmers (Section 7.1) and
processors (Section 7.2). Following the industry profile, this section provides a detailed summary
of EPA's economic analysis of the proposed CAFO regulations as it affects regulated CAFOs
(Section 7.3), hog processors (Section 7.4), and national markets (Section 7.5).
7.1 PROFILE OF THE HOG PRODUCTION SECTORS
This section presents a profile of hog production operations and provides background
information for analyzing the private sector costs of the proposed CAFO regulations. The
purpose of this profile is to provide a baseline description of the current activities, structure, and
performance of the hog production industries. The following sections describe the types of
operations in this sector and present an overview of the industry, describing the number and size
of operations (including the subset of regulated operations), geographic distribution, supply and
demand conditions, price trends, and the financial conditions that characterize this sector.
7.1.1 Industry Definition
Hog production operations are those that raise or feed hogs and pigs either independently
or on a contract basis. These operations are identified under NAICS 11221, Hog and pig farming
(SIC 0213, Hogs). Hog production operations may be categorized by six facility types based on
the life stage of the animal in which they specialize (NPPC, 1998; USDA/APfflS, 1995b). These
categories include: .
• Farrow-to-wean operations that breed pigs and ship 10- to 15-pound pigs to
nursery operations.
• Fan-owing-nursery operations that breed pigs and ship 40- to 60-pound "feeder"
pigs to growing-finishuig operations.
• Nursery operations that manage weaned pigs (more than 10 to 15 pounds) and
ship 40- to 60-pound "feeder" pigs to growing-finishing operations.
• Grow-finishing or feeder-to-finish operations that handle 40- to 60-pound pigs
and "finish" these to market weights of about 255 pounds.
7-1
-------
• Farrow-to-finish operations that handle all stages of production from breeding
through finishing.
• Wean-to-finish operations that handle all stages of production, except breeding,
from weaning (10- to 15-pound pigs) through finishing.
USDA's Animal and Plant Health Inspection Service (APHIS) reports that in a 1995
survey of hog production operations, 86 percent of respondents produced market hogs
(comprised of 62 percent farrow-to-finish operations and 24 percent grower-to-finish operations),
12 percent produced feeder pigs (nurseries), and 3 percent produced weaned pigs and breeding
stock (USDA/APHIS, 1995b).
Fresh meat cuts from a typical hog carcass constitute nearly 40 percent of carcass weight.
This phase of production is typically handled beyond the production phase by meat packing plants
(NAICS 311611, Animal [except poultry] slaughtering). The remaining carcass weight is further
processed into sausages and other prepared meats (NAICS 311612, Meat processed from
carcasses). Other by-products, such as hides, lard, and offal, have value in the manufacture of
clothing, foodstuffs, fertilizers, and other industrial products. Additional information on the hog
processing sector is presented in Section 7.2.
7.1.2 Overview of the Hog Industry
The hog industry is undergoing rapid and significant change, including changing
technology usage, size of operations, location, and product characteristics (Schrader, 1998).
Structural change is reflected by a tendency toward fewer but larger operations as hog production
shifts from family-based, small-scale, independent production operations to more specialized,
larger production operations that are less dependent on market coordination between production
phases (Schrader, 1998; Warner, 1998). Major geographic shifts are also evident as production
operations relocate from the more traditional Corribelt region to the Southern states. Increased
industrialization and management intensity have accompanied changes in ownership structure and
the increased use of contract production (Lawrence, et al., 1998; Warner, 1998), as discussed
later in Section 7.2. The rise in the use of production contracts has contributed to changes in
several areas, including geographic distribution and prices received by producers, as noted in the
following sections. Meanwhile, market conditions have been unstable in the hog industry and
have been characterized by production expansion in excess of domestic and export demand, which
have pushed prices lower.
7.1.2.1 Trends in the Number and Size
USDA reports that in 1997 there were 109,800 hog farms in the United States, based on
year-end inventories (USDA/NASS, 1999a). See Table 7-1. These data on the number of farms
7-2
-------
include both commercial and non-commercial operations, as well as operations that confine and
graze animals. This estimate includes all facility types, including operations that finish market
hogs for slaughter and operations that raise breeder stock.
The number of hog farms is declining. In 1987 there were 243,400 hog farms in the
United States (USDC, 1994). This compares to 109,800 hog farms in 1997, reflecting a decrease
of nearly 60 percent over the 10-year period (Table 7-1). During the same period, however, total
year-end hog inventory among all operations increased, rising from 52.2 million hogs in 1987 to
61.2 million hogs in 1997—a 17 percent increase (USDA/NASS, 1999a; USDC, 1994).
Increased production is also the result of production efficiency gains and large-scale expansion
among some producers. Average herd size across all operations more than doubled between
1987 and 1997, from a national average of 220 hogs to 560 hogs per operation (USDA/NASS,
1999a; USDC, 1994). Table 7-1 shows these trends based on data on the number of hog
operations and corresponding total number of animals for selected years between 1974 and 1997.
Data on the distribution of animals across the various operation sizes indicate that the
majority of hog operations (93 percent) have fewer than 1,250 head, accounting for about one-
third of overall inventories (USDA/NASS, 1999a). Nearly half the inventories are concentrated
among the 3 percent of operations with more than 2,500 head.
Table 7-1. Number of Hog Operations and Animals, 1974-1997
Year
1974
1978
1982
1987
1992
1997
Operations
470,258
445,117
329,833
243,398
191,347
109,754
Animals
(thousand)
45,504
57,697
55,366
52,217
57,563
61,206
Herd Size
(head)
97
130
168
215
301
558
Percent of
Operations
Percent of
Animals
(>l,000head)
0.9%
1.6% -
2.8%
4.0%
6.2%
11.6%
15.6%
22.7%
30.8%
38.0%
50.3%
75.3%
Source: USDA/NASS, 1999a, and USDC, 1994. Data are based on year-end inventory.
For the purpose of this analysis, EPA estimates the number of confinement operations that
may be subject to the proposed CAFO regulations using 1997 Census data that are aggregated by
USDA's NASS. NASS developed a methodology for identifying farms likely to be CAFOs based
on the Census survey Information and estimated animal units on these operations based on
reported data. A summary of these data are provided in the Development Document, USEPA,
2000a. These summary data reflect average herd size throughout the year, accounting for both
7-3
-------
animals sales and inventories. Where applicable, data are adjusted for the average number of
marketing cycles (USEPA, 2000a). This avoids misrepresentation due to seasonal fluctuations in
inventory and the number and timing of animals sold. From these data, EPA has estimated the
number of confinement operations (referred to here as AFOs) using available data and other
information from the Census as well as other USDA and industry publications (USDA/NASS,
1999a; USDA/APfflS, 1995b; NPPC, 1998). These data may differ from those presented in
Table 7-1.
Expressed on this basis, USDA estimates that there were 117,880 hog operations with
56.7 million market and breeding hogs in 1997 (Table 7-2). Not all of these operations would be
subject to the proposed regulations. As shown in Table 7-2, under the two-tier structure, EPA
estimates that there are 5,860 farrow-frnish feedlots (including breeder and nursery operations)
and 2,690 grow-finish feedlots with more than 1,250 head (i.e., 500 AU equivalent). Under the
three-tier structure, EPA estimates that 5,700 farrow-finish feedlots (including breeder and
nursery operations) and 2,650 grow-finish feedlots with more than 750 head (i.e., 300 AU
equivalent) would meet EPA's proposed "risk-based" conditions and thus require a permit.
(More information on the co-proposed tier structures is provided in Section 3.)
Under the two-tier structure EPA expects that designation of hog operations with fewer
than 1,250 head will be limited to about 20 confinement operations annually, or 200 operations
over a 10-year time period. Under the three-tier structure, EPA expects that about 5 hog
operations with fewer than 750 head would be designated annually, or 50 operations over a 10-
year time period. EPA expects that designated facilities will be located in more traditional
farming regions.
As shown in Table 7-2, a total of 2,690 grow-finish operations and 6,060 farrow-finish
operations are estimated either to be defined (>500 AU) or designated (<500 AU) as CAFOs
under the two-tier structure at the 500 AU threshold. A total of 2,650 grow-finish CAFOs and
5,750 farrow-finish CAFOs are estimated to be defined (>300 AU) or designated (<300 AU)
under the three-tier structure. These estimates adjust for operations with more than a single
animal type.
More information on how EPA estimated the number of affected animal confinement
operations is presented in Section 2 of this report, along with additional estimates on the number
of affected hog operations under other regulatory options considered by EPA.
7.1.2.2 Geographic Distribution
Hog production is concentrated among the top five producing states. In 1997, these five
states supplied 60 percent of the U.S. market for pork, measured in terms of marketed head.
Iowa was the largest hog producing state, representing 20 percent of all hogs marketed in 1997
(Table 7-3). The second largest producing state was North Carolina with 16 percent of sales.
7-4
-------
Table 7-2. EPA's Estimate of the Number of CAFOs Affected Under the Co-Proposed Tier Structures
Sector
Hog-GF
Hog-FF
Total
Total
Number
ofAFOs
53,620
64,260
117,880
Number of CAFOs
>1,000
AIT"
1,670
2,420
4,090
Two-Tier Structure
(500 AU Threshold)
500-1,000
' AU
1,020
3,430
4,450
<500
AU
0
200
200
Total
CAFOs
2,690
6,060
8,750
Three-Tier Structure
(Scenario 3)
300-
1,000 AU
980
3,280
4,260
<300
AU
0
50
50
Total
CAFOs
2,650
5,750
8,400
Source: USEPA, 2000a. See Section 2 for more information. See Table 3-1 for definitions of the
options/scenarios.
"Hogs: FF" are farrow-finish (includes breeder and nursery pigs); "Hogs: GF" are grower-finish only. The number
of operations shown eliminates double counting of operations with mixed animal types.
^As defined for the proposed regulations, one AU is equivalent to 2.5 hogs (over 55 pounds) or 5 nursery pigs.
Other top five producing states included Minnesota with 9 percent, Illinois with 8 percent, and
Missouri with 7 percent of hogs marketed in 1997. Other major producing states in 1997
included Indiana, Nebraska, Oklahoma, Kansas, and Ohio. Combined, the top ten producing hog
states accounted for 82 percent of U.S. production in 1997 (Table 7-3).
Despite North Carolina's large production share, the majority of commercial hog
operations are located in the Midwest: In 1997, Iowa had the most hog operations with 18,400
(see Table 7-3). Other states with large numbers of hog operations included Minnesota (7,700),
Illinois (7,400), Ohio (5,900), Indiana (6,600) and Nebraska (6,300 operations). By comparison,
North Carolina had 2,700 commercial hog operations (USDA/NASS, 1999a).
The Southeast has seen rapid growth in hog production. Consolidation has been one
factor in the region's increasing hog inventories. Other factors include increased vertical
integration (modeled on the integrated poultry operations in the region), proximity to growing
consumer markets, and mild climate (USGAO, 1995; Iowa State University, 1998). The
Southeast offers a number of economic advantages for hog producers, including lower energy
costs to heat facilities and a closer proximity to end markets. The warmer climate of the
Southeast also contributes to improved feed efficiency, which makes proximity to feed grain
sources among Midwestern states less important. Compared with the Southeast, the Midwest
continues to support smaller, traditional, independently owned farms (McBride, 1999; Iowa State
University, 1998; Martinez, 1999).
7-5
-------
Table 7-3. Geographic Distribution of Hos Operations by Major Producing State, 1997
Major Producing
State
owa
'Jorth Carolina
Minnesota
Illinois
vlissouri
Indiana
Nebraska
Oklahoma
Kansas
Ohio
South Dakota
Michigan
Wisconsin
Pennsylvania
Georgia
Texas
Tennessee
Mississippi
California
Florida
All Other
Top 5 states
Top 10 states
Top 20 states
Total U.S.
Marketed Head
(thousands)
21,040
16,735
9,197
8,128
7,443
6,282
5,966
3,846
3,500
3,050
2,305
1,697
1,554
1,536
1;304
854
726
503
378
137
8,373
62,543
85,187
96,181
104,554
(percent)
20%
16%
9%
8%
7%
6%
6%
4%
3%
3%
2%
2%
1%
1%
1%
1%
1%
0%
• 0%
0%
8%
60%
81%
92%
100%
Farms Reporting Sales
(number)
18,370
2,666
7,717
7,447
5,183
6,623
6,296
2,082
2,873
5,938
3,067
2,690
3,591
2,971
1,561
3,659
1,579
425
1,193
971
15,204
41,383
65,195
86,902
102,106
(percent)
18%
3%
8%
7% I!
5%
6%
6%
2%
3%
6%
3%
3%
4%
3%
2%
4%
2%
0%
1%
15%
41%
64%
85% 1
100% ||
Source: USDA/NASS, 1999a and USDA/NASS, 1998e. Data are based on annual sales.
7-6
-------
7.1.2,3 Supply and Demand Conditions
Total U.S. pork production (carcass weight basis) increased slightly between 1992 and
1997, reaching 17.3 billion pounds in 1997 (Table 7-4). During the same period, total domestic
demand for pork products decreased. Expressed on a per-capita basis, adjusted for population
growth, demand dropped by nearly 8 percent from 67.8 pounds per person in 1992 to 62.5
pounds per person per year in 1997 (Table 7-3). Compared to demand levels in the 1970s, when
pork consumption exceeded 70 pounds per person per year, consumption is down by about 10
pounds per person (Table 7-3). Domestic consumption constitutes the bulk (94 percent) of U.S.
annual pork supplies.
As shown in Table 7-4, U.S. pork exports nearly doubled during the 1990s (Putnam and
Allshouse, 1997 and 1999). The U.S. is among the world's lowest cost producer of pork, but still
ranks close to competitors Australia and Argentina in terms of cost of production (Iowa State
University, 1998). .
Table 7-4. Total U.S. Hog Supply and Demand (carcass weight basis), 1992-1997
Year
1992
1993
1994
1995
1996
1997
%92-97
Production
Imports
Exports
Total
Demand
(million pounds red meat carcass weight)
17,234
17,088
17,696
17,849
17,117 .
17,274
0.2%
645
740
743
664
618
633
-1.9%
552
601
549
.787
970
1,044
89.1%
17,330
17,253
17,811
17,768
16,795
16,821
-0.3%
Per Capita
Demand
(lbs./person)
67.8
66.8
68.2
67.4
63.0
62.5
-7.8%
Source: Putnam and Allshouse, 1997 and 1999. Supplemented with information from USDA/ERS, 1998c and
1997f. Excludes beginning and ending stocks and shipments to U.S. territories. Per-capita demand is shown to
depict real demand growth, adjusting for growth in U.S. population (about 1 percent per year).
7.1.2.4 Farm Price Trends
Prices received by farmers tend to vary seasonally according to production cycles
throughout the year and are also prone to wide fluctuations from year-to-year. Prices are often
subject to periods of high instability according to changing market conditions or sharp shifts in
supply in response to changing prices and/or input costs, among other factors. This is especially
7-7
-------
true in the pork sector where prices may vary cyclically and seasonally and are highly dependent
on market demand and current inventory or supply (NPPC, 1998; Schrader, 1998).
Cyclical variations in hog prices occur due to the biological time lags that are inherent hi
pork production. Because hogs take approximately a year to reach market weight, it is impossible
to know if current breeding decisions will accurately reflect demand conditions a year later.
Mismatches between the supply and demand for hogs when those hogs actually reach market will
cause changes in hog prices that will signal farmers to adjust breeding decisions. To expand
future meat supplies in response to expected profits, producers must hold back animals from
market in the near term to build up the breeding herd, which will short the market and increase
prices in the short run. Conversely, when low prices signal a reduction in production, the
resulting herd sell-off will increase supplies and reduce prices in the short run (Kohls and Uhl, ,
1998). The market cycle caused by this biological lag can take three to four years before the
market returns to equilibrium—barring other shocks to the market (Schrader, 1998).
Seasonal changes can also affect prices through extreme weather events that affect supply
and through changes in consumer preferences that affect demand. Based on data for 1985
through 1996, Schrader (1998) shows that the seasonal variation in hog prices indicates a
production-driven rather than market-driven system, based on average trends that show that
production is high when prices are low. These variations in the production cycle and seasonal
pricing patterns indicate that efficiency gains may be attained through closer coordination of
production and marketing (Schrader, 1998). Indeed, the uncertainty and risk caused by these
cycles has been a major factor in driving the trend towards coordination between producers and
processors (Kohls and Uhl, 1998).
Table 7-5 presents actual quarterly and annual prices received by U.S. hog producers from
1992 through 1997. Following a decline in average prices during the mid-1990s when average
hog prices were about $40 per hundredweight (cwt) and about $30 per cwt for sows, prices
surged during the 1996-1997 period to over $50 per cwt for hogs and $40 per cwt for sows
(Table 7-4).' Higher hog prices during 1996 and 1997 followed a contraction in domestic
production in response to lower prices in 1994-1995 and also rapid growth in export demand
(Table 7-4; Table 7-5). Hog prices dropped substantially in 1998, but lower feed prices during
that year somewhat offset the price drop (USDA/ERS, 2000c).
The actual price a farmer receives will depend on whether the operation is an independent
owner-operator or whether the operation grows animals under a production contract with a
processor/integrator.
The contract price that the grower receives is often lower than the market price received
by independent operators, thus contributing to lower gross revenues received by the contract
grower compared to the independent operator. The North Carolina Pork Council report contract
'Expressed in real terms adjusted for producer price inflation, average gains in price are somewhat lower.
7-8
-------
Table 7-5. Actual Average Quarterly and Annual Hog Prices Received by Farmers, Total U.S., 1992-1997
Year
Average
Ql
Average
Q2
Average
03
Average
Q4
Average
Annual
Hogs Monthly Prices Received by Farmers ($/hundredweight or cwt)
1993
1994
1995
1996
1997
43.9
45.3
37.9
45.9
52.0
46.8
42.7 •
38.3
54.3
56.6
46.9
39.8
47.8
57.7
54.9
43.3
30.2
43.0
1 55.2
44.7
45.2
39.5
41.8
53.3 .
52.0
Sows Monthly Prices Received by Farmers ($/hundredweight or cwt)
1993
1994
1995
1996
1997
35.1
36.9
28.3
.33.3
46.3
38.1
35.4
28.8
40.8
46.5
35.7
28.7
32.7
47.5
43.5
34.3
21.0 .
34.0
49.2
35.5
35.8
30.5
31.0
42.7
43.0
Source: USDA/NASS, 1998a.
hog prices in the range of $10 to $11 per hog (Farm Journal, 1998). Schiller (1998) report
contract prices ranging from about $11.60 to $12.80 per hog. At $10 to $13 per hog (assuming a
250 pound finish weight), these prices translate to prices of about $4 to $5 per cwt, or roughly 10
percent of the average USDA-reported producer price (see Table 7-5). Nearly two-thirds of
grow-finish operations that raise market hogs use contracts, whereas fewer than 2 percent of
farrow-finish operations use contracts (USDA/ERS, 2000c).
With a production contract, the contract grower's lower price generally is offset by lower
costs, since the integrator often pays for a substantial portion of the grower's annual variable cash
expenses. Inputs supplied by the integrator may include feeder pigs, feed, veterinary services and
medicines, technical support, and transportation of animals. These variable cash costs comprise a
large component of annual costs, averaging more than 70 percent of total variable and fixed costs
at livestock operations (USDA/ERS, 1999a). In addition, the grower faces reduced risk because
the integrator guarantees the grower a fixed output price. By comparison, the independent
operator must cover all production costs and anticipate market price fluctuations. Therefore, an
independent operator faces relatively higher costs and also assumes greater production and price
risks compared to a contract grower. Yet, because the grower's production costs are lower, the
contract grower's net revenue or profits (gross revenue less costs) may be comparable to that of
an independent operation.
7-9
-------
7.1.3 Financial Characteristics of Hog Operations
7.1.3.1 Overview of Financial Characteristics
In 1997, commercial hog farms in the U.S. generated a total of ,$13.1 billion in annual
revenue (USDA/NASS, 1999a).2 As shown in Table 7-6, nearly 90 percent ($11.6 billion) of
total revenues from commercial hog operations represent the sale of hogs and pigs. Secondary .
livestock revenues, including sales by farms that generate a portion of their total revenue from
other livestock raised on-site totaled $0.5 billion (4 percent of total farm revenues). Crop sales
from hog operations accounted for 8 percent ($1.0 billion) in 1997 (USDA/NASS, 1999a).
Approximately one-fourth of all commercial hog farms generate more than $0.5 million in
revenue annually (USDA/NASS, 1999). See Table 7-6. The remaining three-fourths of farms
generate revenues below $0.5 million. This revenue cut-off corresponds with the definition of a
"small business" in the hog sector established by the Small Business Administration (SBA, 1998;
USGPO, 2000). (Section 9 provides additional information on EPA's small business analysis.)
7.1.3.2 Income Statement and Balance Sheet Information
In recent years, hog farms have faced a number of financial and structural changes. It is
expected that consolidation will continue in the face of negative farm returns to management and
risk. The hog farms most susceptible to closure are small farms, high-cost farms, diversified
single-manager operations, farms with older technology, and farms where the manager is
approaching retirement age with no successor to manage the operation (Boehlje et al., 1997).
Table 7-7 presents average income statement and balance sheet data for commercial hog
farms from 1993 through 1997. The average U.S. hog farm was in a favorable financial position
from 1993 through 1997 with a positive net farm income and a debt-to-asset ratio ranging from
0.22 to 0.30. (USDA's farm performance criteria are described hi Section 4.2.5.) While the
national average income statement shows a positive net income, additional information show that
between 1991 and 1994 about 30 percent of all hog farms experienced negative income
(USDA/ERS, 1997b). Data from Yeske (1996) also show a wide distribution in the financial
performance among hog operations, as indicated by an average breakeven market cost per pig
estimated to range from $37 per pig to $108 per pig across operations. Operations in the poorer
performing category likely are smaller operations that are not affected by the proposed CAFO
regulations.
2USDA defines commercial farms as those with gross sales of $50,000 or more during a given year.
- 7-10
-------
Table 7-6. Farm Revenue at Hog Farms (>$50,000 in Sales), by Revenue Category and Economic Class
Revenue Category/Economic Class
# Hog Farms (1,000's)
Revenues ($1,000)
Sales by Revenue Category (reported and percentage share)
Primary Livestock
Secondary Livestock
Crop Sales
AH Farms
Primary Livestock
Secondary Livestock
Crop Sales
All Farms
23,511
3,130
8,042
34,683
68%
9%
23%
100%
11,594,271
510,494
991,887
13,096,653
89%
4%
8%
100%
Sales by Economic Class (reported and percentage share)
>$1 million in revenue
Between $0.5-$1.0
Between S0.25-S0.50
Between S0.10-S0.25
Between S0.05-S0.10
All Farms
>$1 million in revenue
Between S0.5-S1.0
Between $0.25-50.50
Between S0.10-S0.25
Between $0.05-$0.10
All Farms
3,545
5,641
9,995
12,340
3,162
34,683
10%
16%
29%
36%
9%
100%
7,135,927
2,407,870
1,987,686
1,355,136
210,035
13,096,654
54%
18%
15%
10%
2%
100%
Source: USDA/NASS, 1999a (Table 50 and Table 51). Based on data for commercial farms with more than
$50,000 in annual revenues. Excludes non-commercial farms with revenues below $50,000.
Primary Livestock: Hogs (NAICS 1122) and Poultry (NAICS 1123), respectively.
Secondary Livestock: Beef (beef farming, NAICS 112111, and beef feedlots, NAICS 112112), Dairy (NAICS
11212), miscellaneous categories (NAICS 1122, NAICS 1124, NAICS 1125), along with Hogs (NAICS 1122) and
Poultry (NAICS 1123), respectively.
Crop Sales: Oilseed/Grains (NAICS 1111), Vegetables (NAICS 1112), Fruits/Nuts (NAICS 1113), Greenhouse
(NAICS 1114) and other crops (NAICS 1119).
7-11
-------
Table 7-7. Income Statement and Balance Sheet for Hog Farms (Sales >S50,000), 1993-97
Item
1993
1994
1995
1996
1997
(dollars per farm)
[ncome Statement
Gross cash income
Livestock sales
Crop sales (inch net CCC loans)
Government payments
Other farm-related income ^
Less: Cash expenses
Variable Cash expenses
Fixed Cash expenses
Equals: Net cash farm income
Less: Depreciation
Labor, non-cash benefits
Plus: Value of inventory change
Non-money income w
Equals: Net farm income
$191,566
$144,676
$27,577
$10,094
$9,219
$155,941
$130,181
$25,760
$35,625
$17,092
$318
$3,939
$3,972
$26,126
$195,594
$144,635
$25,593
$6,286
$19,080
$163,433
$138,166
$25,267
$32,160
$16,931
$463
$934
$4,047
$19,748
$227,671
$170,508
$34,850
$6,128
$16,186
$180,698
$147,097
$33,601
$46,973
$20,466
$574
$12,662
$3,866
$42,461
$330,160
$259,556
$47,366
$5,037
$18,201
$245,610
$208,878
$36,733
$84,549
$26,342
$458
$33,366
$4,742
$95,857
$347,371
$272,782
$35,356
$4,635
$34,598
$262,349
$223,247
$39,102
$85,022
$26,937
$264
($2,510)
$3,960
$59,271
Balance Sheet
Farm assets
Current assets
Non-current assets
Farm liabilities
Current liabilities
Non-current liabilities
Farm equity
Debt/asset ratio
$538,454
$110,173
$428,280
$129,150
$43,772
$85,377
$409,304
0.24
$553,871
$124,963
$428,908
$130,321
$40,904
$97,434
$414,231
0.25
$564,979
$126,237
$438,742
$148,480
$53,559
$94,921
$416,499
0.26
$770,265
$183,716
$586,549
$167,792
$55,817
$111,976
$602,472
0.22
$647,007
$123,632
$523,374
$195,555
$48,904
$146,651
$451,452
0.30
Source: USDA/ERS, 1997a.
^ Includes income from machine-hire, custom work, livestock grazing, land rental, contract production fees,
outdoor recreation, and any other farm-related source.
Defined as home consumption and imputed rental value of farm dwellings owned by the farm operation.
Factors affecting the financial differences between operations include the size of
operations, labor input, tenure of the operation, and whether the operation is owner-operated or
under contract (Zulovich, 2000). The costs of raising a hog from farrowing to market weight are
roughly the same whether the function is performed at one facility, e.g., farrow-finish, or at a
series of specialized facilities (Foster, 2000a). Farrow-finish and grow-finish facilities differ
slightly in their cost structure because of the life cycle stages they encompass. Farrow-finish total
costs may tend to be higher than other facilities because they raise the hog through its whole life
cycle and through a period that involves more veterinary care and labor. Alternately, average
7-12
-------
costs at these facilities may be lower than at other operations because they raise the hog through
the early weeks when weight gain is most rapid.
Grow-finish operations, which have on average more than twice the number of animals as
farrow-finish operations, have smaller returns on a cwt gain basis because of higher operating
costs for feeder pigs and marketing at grow-finish operations (USDA/ERS, 2000c; Doane's,
1995). Farrow-finish operations, however, have higher overhead costs. Overhead costs are
higher at farrow-finish operations because they tend to have more buildings and equipment per
hog produced, which results in higher depreciation costs. They also have more hired and unpaid
labor, which results in higher labor overhead and opportunity costs of unpaid labor (USDA/ERS,
2000c; Doane's, 1995).
Data are available data from USDA's ERS cost of production data series that reflect
differences among farrow-to-finish, farrow-to-feeder, and feeder-to-finish operations by select
production regions (USDA/ERS, 2000c). These data are measured in terms of average dollars
per hundredweight (cwt) gain among hog operations. Data for 1999 indicate that the total gross
value of production is higher among grow-finish operations than among farrow-finish operations,
estimated to average $38 per cwt compared to $31 per cwt, respectively (USDA/ERS, 2000c).
This is due mostly to greater revenue per marketed hog at grow-finish operations. Average total
revenue less operating expenses is also higher at grow-finish operations, despite higher total
operating costs at these operation estimated at $42 per cwt in 1999 (compared to $25 per cwt at
farrow-finish operations). Costs for feed and feeder pigs are higher at grow-finish operations,
estimated at $38 per cwt or about 90 percent of total operating costs (USDA/ERS, 2000c).
These broad differences in financial conditions among the types of hog operations are also shown
in Table 7-8 which presents a summary of ERS' cost of production data for the hog sector,
averaged over the 1993-1997 period.
Investments in equipment may also result in differing returns between farrow-finish and
grow-finish operations. A report by the University of Missouri on hog manure technologies
(1999) presents the difference in return on assets (ROA), where assets are the investment in
manure technologies, for different sized farrow-finish and grow-finish operations. The ROA (net
cost) for farrow finish operations with 150 to 1,200 sows ranges from 11 to 19 percent; the ROA
at grow-finish operations with 2,000 to 4,000 hogs ranges from 9 to 11 percent (University of
Missouri, 1999). These reported ROA percentages for relatively small investments in manure
technologies at farrow finish and grow-finish operations are good by most standards. Average
costs are lowest among operations that raise immature animals only (see data for "farrow-feeder"
operations shown in Table 7-8; also, see USDA/ERS, 2000c, and Yeske, 1996).
Despite these differences in the financial conditions among the different types of hog
operations, EPA uses average financial data across all hog operations since these are the data that
are available from ARMS for this analysis. Specifically, EPA assesses impacts at grow-finish and
farrow-finish operations using data for all hog operations. Based on the broad differences
demonstrated by data shown in Table 7-8, EPA believes that these average data more or less
approximate conditions equally across grow-finish and farrow-finish operations, particularly for
7-13 .
-------
Tflhlc 7-8 Costs and Returns for Hoe Farms bv Facility Type, Average 1993-1997
Facility Type
Total
North"
South"'
(average $ per cwt. gain)
411 Hog Farms
Total, gross value of production
Total, variable cash expenses
Total, fixed cash expenses
Total, cash expenses
Gross value of production less cash expenses
Economic (full ownership) costs
Residual returns to management and risk
54.62
41.57
5.75
47.33
7.30
66.14
-11.52
54.40
41.43
6.16
47.59
6.81
66.10
-11.70
Farrow-to-Finish Farms
Total, gross value of production
Total, variable cash expenses
Total, fixed cash expenses
Total, cash expenses
Gross value of production less cash expenses
Economic (full ownership) costs:
Residual returns to management and risk
48.89
36.92
5.15
42.08
6.81
60.43
-11.54
48.77
36.29
5.48
41.77
7.00
59.89
-11.12
Farrow-to-Feeder Farms
Total, gross value of production
Total, variable cash expenses
Total, fixed cash expenses
Total, cash expenses
Gross value of production less cash expenses
Economic (full ownership) costs:
Residual returns to management and risk
82.58
66.76
11.17
77.93
4.65
119.51
-36.93
83.93
65.21
10.53
. . 75.74
8.18
114.88
-30.95
Fee'der-to-Finish Farms
Total, gross value of production
Total, variable cash expenses
Total, fixed cash expenses
Total, cash expenses
Gross value of production less cash expenses
Economic (full ownership) costs:
Residual returns to management and risk
60.48
49.89
5.56
55.45
,5.03
68.83
-8.36
60.64
52.45
6.23
58.67
1.97
72.24
-11.60
55.52
42.14
4.24
46.38
9.14
51.44
4.08
49.33
39.41
3.82
43.23
6.10
62.48
-13.15
83.93
65.21
10.53
75.74
8.18
114.88
-30.95
60.10
44.27
4.01
48.28
11.82
61.24
-1.14
Source: Derived form USDA/ERS, 2000c.
"'North: IL, IN, IA, KS, MI, MN, MO, NE, OH, SD, arid WI.
"•'South: AL, AR, GA, KT, NC, SC, TO, TX, and VA.
7-14
-------
the key financial criteria examined in this analysis (namely, gross revenue and net cash income).
Among grow-finish operations, EPA's use of these average data likely overstate impacts at grow-
finish operations since these operations generally have more favorable financial conditions than
the average hog operation. As shown in Table 7-8, grow-finish operations tend to have higher
average revenue and higher net cash income compared to the average across all operations.
However, EPA's use of these average data may understate impacts at farrow-finish operations
since financial conditions at these operations are generally less favorable than those for the
average hog operation. As shown in Table 7-8, gross revenue and revenue less expenses may
average 10 percent lower at. farrow-finish operations compared to the average. Because farrow-
finish operations account for more than 60 percent of all marketed hogs (USDA/APHIS, 1995b)
and because the impacts may be understated using average data for this subsector, EPA conducts
sensitivity analyses of these financial variables (provided in Appendix D of this report).
7.1.3.3 Baseline Conditions for Hog Operations
Tables 7-9 and 7-10 provide a summary of the financial baseline conditions assumed for
this analysis. These data are aggregated from the 1997 ARMS data set and are obtained by
USDA's ERS, as described in Section 4. These data are separated by select facility size and
production region groupings (see Table 4-4), but do not reflect conditions separately across the
different types of hog operations (i.e., grow-finish and farrow-finish operations, nurseries, farrow-
to-wean, and wean-to-finish operations). Additional information on how these data differ by
region are provided in the record (USDA/ERS, 1999a, see DCN 70063).
According to the 1997 ARMS data the average hog operation demonstrated a favorable
financial position in 1997 with positive net income and a debt-to-asset ratio that ranged from 15
percent to 39 percent, across select operation sizes (USDA/ERS, 1999a). See Table 7-9. These
debt-to-asset ratios indicate that—on average—hog operations are not in a vulnerable financial
position and have a low potential for cash flow problems and a low relative risk of insolvency.
Based on these data, EPA assumes that baseline (prior to regulation) net cash flow for all model
types for the hog sector is positive, and baseline debt-to-asset ratios for all model types are 40
percent or less. All hog operations in this analysis, therefore, are considered financially healthy,
on average, in the regulatory baseline.
Data shown in Table 7-9 are distributed by broad facility size groups. As shown, more
than 90 percent of operations have fewer than 800 hogs and pigs, however, these operations only
account for about 30 percent of all hogs raised annually (Table 7-9). There are fewer larger-sized
operations with more than 2,500 head (2 percent of all farms), but these operations raise over 40
percent of all hogs annually (Table 7-9). Smaller hog operations with less than 800 head are more
diversified than larger ones, with about 50 percent of all farm revenue from crops. This compares
to hog operations with more than 2,500 hogs, where livestock comprises the bulk of all annual
farm sales and only 13 percent of farm revenues are from crops (Table 7-9).
7-15
-------
Table 7-9. Tvoical Financial Characteristics of Hog Operations, By Size of Operation
Item
Number of farms
Percent of farms
Percent of value of production
Livestock value of production
Crop value of production
All Farms
117,552
100.0%
100.0%
62.8%
37.2%
Less than 800
Head
106,463
90.6%
53.8%
50.8%
49.2%
800 to 2,500
Head
8,298
7.1%
21.5%
65.3%
34.7%
More than
2,500 Head
2,79 1
2.4%
24.6%
86.7%
13.3%
..'- ,
Number of hogs and pigs
Distribution of hogs and pigs
Percent of hogs and pigs owned
Percent of hogs and pigs not owned
Number of sample farms with hogs and pigs
40,562,697
100.0%
72.4%
27.6%
986
13,074,799
32.2%
79.1%
20.9%
648
10,896,454
26.9%
79.5%
20.5%
185
16,591,444
40.9%
62.4%
" 37.6%
153
Debt-to-Asset Ratios
AH Regions
Midwest
Mid-Atlantic
0.1837
0.2079
0.1300
0.1509
0.1732
0.0920
0.2534
0.2511
d
EPA Derived Gross Cash Income Per Animal ^
All Regions
Midwest
Mid-Atlantic
$363.00
$377.10
$174.47
$643.63
$606.13
$83.60
$296.66
$303.89
d
0.3927
0.4044
0.3095
$185.43
$228.99
$383.18
EPA Derived Net Cash Income Per Animal ^
All Regions
Midwest
Mid-Atlantic
$69.54
$76.47
$30.64
$110.32
$118.64
$30.98
$63.86
$66.02
d
$41.13
$46.51
$31.16
Source: USEPA and USDA/ERS, 1999a.
''EPA derived gross cash and net cash income by dividing the average gross or net cash income line items by the
average number of animals as for each size group and region.
d = Data insufficient for disclosure.
7-16
-------
Table 7-10. Income Statement and Balance Sheet for Farms with Hogs and Figs, by Size of Operation, 1997
Item
AH Farms
<800 Hogs/ Pigs
800 to 2,500
>2,500 Hogs/ Pigs
Income Statement
Gross cash income
Livestock income
Crop sales (incl. net CCC loans)
Government payments
Other farm-related income "
Total variable expenses
Livestock purchases
Feed
Other variable expenses v
Total fixed expenses
Equals: Net cash farm income
Less: Depreciation and Other 3/
Plus: Value of inventory change
Plus: Nonmoney income "'
Equals: Net farm income
125,259
73,644
35,960
3,808
11,847
80,217
7,561
27,777
44,880
21,046
44,880
11,491
4,203a
5,028
21,736
79,045
40,322
28,122
3,048
7,553
49,939
3,582
13,297
33,061
15,557
33,061
8,011a
4,197a
5,106
14,842
389,557
244,989
106,564
11,203
26,801
249,880
28,405
102,720
118,756
55,819
118,756
35,833a
12,079b
4,364
64,467
1,102,299
835,301
125,004a
10,805
131,189a
730,719
97,357a
357,286
276,077a
127,056
276,077a
71,861
-18,991c
4,010
157,682a
Balance Sheet
Farm assets
Current assets
Non-current assets
Land, buildings, and equipment 5/
Farm liabilities
Current liabilities
Noncurrent liabilities
Farm equity
484,506
65,984
418,522
397,574
89,007
25,885
63,122
395,499
419,939
49,424
370,515
353,413
63,365
18,944
44,421
356,574
944,775
197,938
746,836
702,403
239,367
72,192
167,175
705,408
1,578,920
305,330
1,273,590
1,175,766
620,057
152,953
467,104a
958,863
Source: USDA/ERS, 1999a. Copies of these data are in the rulemaking record (DCN 70063).
"Machine-hire, custom work, livestock grazing, land rental, contract fees, and other farm-related sources.
^Incl. livestock leasing, custom feed processing, bedding, grazing, supply, transportation, storage, general business
expenses, and registration fees. Footnote (a) refers to an RSE on "other livestock-related" portion of the total:
3/Includes labor, non-cash benefits. Footnote (a) refers-to an RSE on "non-cash benefits" portion of the total.
"'The value of home consumption plus an imputed rental value of farm dwellings.
5/The value of the operator's dwelling and associated liabilities are included if the dwelling was located on the farm.
a = Relative standard error (RSE) of the estimate exceeds 25 percent, but no more than 50 percent.
b = RSE of estimate >50%, but <75%. c = RSE of estimate >75%. d = Data insufficient for disclosure.
7-17
-------
Table 7-9 also shows the percentage of hogs and pigs owned by farmers compared to those
not owned by farmers. EPA uses this information on animal ownership as an indication of the
extent of production contract use in these sectors (see Section 2.3). Across all hog operations in
1997, about 30 percent of animals were not owned by farmers (USDA/ERS, 1999a). Percentages
vary across farm sizes, with up to 38 percent of animals not owned by the farming operation for
farms with more than 2,500 hogs and pigs, compared to 28 percent among smaller-sized
operations (Table 7-9). This is consistent with other market information (Hayenga et al., 1996;
Lawrence et al., 1998).
Table 7-10 presents average income statement and balance sheet information for hog
operations in 1997, by size of operation. The financial data used for this analysis do not
distinguish between operations with and without production contracts. Contract operations may
have lower revenues, but lower costs as well. These data also do not distinguish between grow-
finish and farrow-finish operations, as discussed in Section 7.1.3.2. EPA believes that some
of these differences are addressed in its sensitivity analysis that examines varying some of the key
input data used for this analysis, presented in Appendix D.
The data shown in Table 7-10 are differentiated by selected size categories and reveal
differences among operations by size. The income statement data (as well as the data in Table 7-9)
point to increasing specialization as the size of an operation increases. A larger proportion of
animals are not owned at the largest operations than at the smallest operations. However, the
smallest operations have proportionately smaller expenditures on livestock-related expenses than
larger operations. Expenditures on livestock and feed average about one-third of total variable
expenses at an average hog operation with less than 800 hogs; operations with more than 2,500
hogs are associated with expenditures on livestock and feed averaging two-thirds of total variable
expenses (Table 7-10). Explanations for these differences may include differences hi the degree of
specialization and feeding strategies, and other factors.
Smaller and larger operations are also different in terms of government payments. As in
most of the livestock and poultry sectors, the smallest operations receive a greater proportion of
then: gross cash income in the form of government payments-4 percent, compared with 1 percent
for the largest operations. These differences may also reflect the greater diversity of smaller
operations, which could be receiving crop subsidies.
Despite these differences, operating margins (measured in terms of average net cash farm
income as a percentage of average gross cash income) among differently sized operations do not
differ substantially: operations with less than 800 hogs show an operating margin of 17 percent, as
compared to 22 percent at operations with more than 800 hogs (USDA/ERS, 1999a). However,
the smallest operations show the lowest return on assets (measured as average net farm income to
average farm assets): operations with less than 800 hogs show average a return on assets of 3.5
percent, as compared to 6.8 percent and 10 percent at operations with between 800 and 2,500
hogs arid operations with more than 800 hogs, respectively (USDA/ERS, 1999a). See Table 7-10.
The 1997 ARMS data include, among an average farm's assets, the value of the owner's home
when it is located on the farm. Since smaller operations may be more likely to have the owner's
7-18
-------
dwelling located on the farm .than larger operations, if dwelling values were excluded, the returns
on the "business" assets might be higher for these smallest operations.
Section 4 of this report presents key financial data used for this analysis, shown in
Table 7-10, that are calculated onto a per-animal basis. For the hog sector, total gross farm
revenues are estimated to range from $84 to $304 per head (includes revenue from other farm-
related sources). Net cash income ranges from $31 to $66 per head among CAFO models,
depending on facility size and region (see Tables 4-5 and 4-6).
7.2 PROFILE OF THE HOG PROCESSING SECTORS
Hog and poultry farms represent the beginning of the meat and egg products marketing
chain that also includes meat packers, food processors, integrators, and retailers. Farms provide
the raw materials to slaughterers, packers, and processors in the form of live hogs, which then are
converted into cuts of meat and processed foods. These products are eventually sold to consumers
at retail establishments. Because of seasonality of production, perishability, and limited resources
among farmers to handle farm output, farmers are increasingly reliant upon industry middlemen
such as processors, meat packers, and integrators.
Meat packers that slaughter hogs are identified in the 1997 Census of Manufactures under
NAICS 311611, Animal (except Poultry) Slaughtering. Processors that further-process hogs are
under NAICS 311612, Meat Processed from Carcasses. These codes correspond to the SIC codes
of 2011—Meat Packing Plants, and 2013—Sausages and Other Prepared Meats.
Hog farms and packers linked by spot markets are the dominant form of coordination in the
U.S. pork sector (Hayenga et al., 1996); however, vertical coordination, integration, and
specialization through contract farming are rapidly becoming the norm in pork production,
particularly in regions outside the Midwest. Many similarities exist between current changes in the
pork industry and the changes that took place previously in the broiler industry. Martinez (1999)
indicates that, similar to the broiler industry, the pork industry can use vertical integration to
facilitate the adoption of new, cost-saving technologies. These new technologies are facilitated as
a result of reduced transaction costs and increased access to capital, leading to lower processing
costs and higher quality animals (Martinez, i 999).
While contracting is the primary method for marketing hogs in some states, it is not
popular in all states. Kliebenstein and Lawrence (1995) summarize regional production
comparisons in the hog farm sector as follows. Hog production in North Carolina is characterized
by a highly coordinated system of contractual relationships by larger sized operations. Contracts
specify activities and responsibilities such as feed formulations, production facilities, genetics,
internal veterinary care, and management strategies. Coordination efforts in the more traditional
Midwest have been less rapid but more diversified, and involve independent producers entering
into networks that provide many of the same characteristics of the highly integrated systems
(Kliebenstein and Lawrence, 1995). Small sellers generally used the spot market while larger
7-19
-------
sellers used contracts (USDA/GIPSA, 1996a). Packers are expected to be more involved in
influencing hog production .and marketing decisions in the future through expanded use of long-
term marketing contracts or less formal producer-packer relationships based on the quality of hogs
produced or the herd health programs imposed by packers (Hayenga et al., 1996).
Use of marketing contracts between hog producers and meat packers has risen sharply in
recent years, up from 11 percent of all hog marketings in 1993 to 57 percent of 1997 hog
marketings (Lawrence, et al., 1998). Marketings above 50,000 head size class and operations
outside of the Cornbelt had more than 75 percent of their hogs with a packer (Lawrence, et al.,
1998). The emergence of the new "megaproducers" hi North Carolina in the early 1990s and,
more recently, in the western states of Utah, Oklahoma, and Texas, has encouraged the use of
long-term marketing contracts to ensure procurement of hogs for daily slaughter. For large-scale
specialized processing plants, the use of prearranged agreements ensures procurement will meet
capacity on a day-to-day basis. For farms, if allows for risk sharing if prices drop. Based on a
survey of hog producers, the dominant type of agreements are forward contracts that usually
involve formula pricing tied to market prices in the Midwest, plus quality premiums and discounts
(Hayenga et al., 1996). Some contracts allow for risk sharing by Unking prices to production costs
or by setting upper and lower price bounds.
The number of hogs under production contracts has also risen quickly. In 1997, an
estimated 40 percent of the hogs farrowed and 44 percent of the hogs finished were by farms with
production contracts (Lawrence, et al., 1998). This compares to about 30 percent in 1994, with
most of the growth on the larger sized farms (over 500,000 head marketed). Most contract hogs
are produced by large farms, especially in emergent regions. Production contracts provide
producers with a means of procuring feeder pigs, feed, medication, and technical supervision
through contractors (other producers, packers, or feed companies). Growers typically provide
production facilities, labor, utilities, and waste disposal. Compensation is paid per .head or per
pound of grain with discounts and incentives for feed efficiency and death loss (Lawrence, et al.,
1998). .
A common situation in the hog industry is for an operation to establish a contract
relationship to participate in only one of the production stages of raising livestock, such as
livestock contracting for replacement breeding stock (USDA/ERS, 1996c). In the hog sector,
such arrangements include both finishing and farrowing contracts. A farmer could contract with
another to "finish" hogs by having the second operator feed weaner pigs raised by the first farmer
until it was time to sell them to the processor. Use of these types of arrangements allows farm
operators to increase business volume with limited facilities (USDA/ERS, 1996c). Genetic
advances have helped the development of farrowing farms that specialize in breeding feeder pigs.
Several large pork producers have procurement contracts set up with breeding farms. Among the
largest producers, about 40 percent of market hogs are supplied through contracts with farrowing
farms, while the other 60 percent are produced in their own facilities (Hayenga et al., 1996).
In addition to substantially increasing the use of contracts to ensure supply, the U.S.
meatpacking industry has rapidly consolidated over the past 20 years, due in large part to the
7-20
-------
economies of scale that can be achieved through consolidation. Large plants have significant
advantages in slaughter costs, and the largest hog packers can deliver meat to buyers at costs 5
percent less than plants one-quarter as large (MacDonald et al., 2000). As market concentration
and contract production increases, the hog industry may face the same price discovery concerns
mat the beef cattle industry is experiencing. Increased concentration in the processing sectors,
especially among meat packers, has led to concerns about market competition, monopsonistic
control, and noncompetitive pricing practices that fail to adequately compensate farmers for their
production (Hayenga et al., 1996).
7.3 CAFO ANALYSIS
This section presents the results of EPA's CAFO level analysis for the hog sector. As
discussed in Section 4, EPA uses a representative farm approach to estimate the impact of the
proposed CAFO regulations on affected operations. Each model CAFO differs by facility size
groupings and key farm production regions. For the hog sector, the production regions reflected
in this analysis are the Mid-Atlantic (MA) and Midwest (MW) regions, as defined in Table 4-1
(Section 4). Section 4 provides a summary of how EPA developed the various financial models
used for this analysis. The Development Document (USEPA, 2000a) provides additional
information on the cost models developed by EPA.
Results presented in this section focus on the "BAT Option" that refers to EPA's proposed
technology option for the CAFO regulations (described in Section 3). For the purpose of this
discussion, the "two-tier structure " refers to the combination of BAT Option 5 for the swine
subcategory and NPDES Scenario 4a that covers all operations with more than 500 AU. Where
indicated, the two-tier structure may refer to the alternative threshold at 750 AU (Scenario 5).
The "three-tier structure " refers to the combination of ELG Option 5 (swine subcategory) and
NPDES Scenario 3 that covers operations down to 300 AU based on certain conditions. Results
for other technology options and scoping scenarios considered by EPA as part of this rulemaking
are also summarized. Table 3-1 summarizes EPA's proposed and alternative ELG Options and
NPDES Scenarios discussed in this section.
Section 7.3.1 presents a summary of the cost input data EPA uses for this analysis,
including (post-tax) per-animal and per-facility costs for EPA's model CAFOs. Section 7.3.2
presents EPA's estimate of the aggregate, national level costs of the proposed CAFO regulations
for the hog sector. Section 7.3.3 presents EPA's predicted financial impacts to this sector in terms
of the estimated number and percentage of CAFOs that are expected to experience financial stress
as a result of the proposed CAFO regulations. EPA evaluates economic impacts to CAFOs in this
sector two ways—assuming that a portion of the costs may be passed on from the CAFO to the
consumer (Partial CPT) and assuming that no costs passthrough so that all costs are absorbed by
the CAFO (Zero CPT).
7-21
-------
7.3.1 Overview of Cost Input Data
Tables 7-11 and 7-12 presents estimated input costs that EPA uses to assess costs and
impacts to the hog sector. These data include the post-tax annualized compliance costs, estimated
on a per-animal and per-facility. These costs reflect the estimated capital costs, annual operating
and maintenance costs, start-up or first year costs, and also recurring costs estimated by EPA
(discussed in the Development Document, USEPA, 2000a). These facility costs are annualized
using the approach described in Appendix A of this report. Appendix A shows the individual hog
sector costs by model across all technology options.3
Other input data for this analysis include EPA's estimate of the number of affected CAFOs
and baseline financial conditions at model CAFOs. EPA's estimate of the number of animal
confinement operations that would be defined or designated as CAFOs is presented in Section
7.1.2.1 (see Table 7-2). Additional information is provided in Section 2 of this report. The
average baseline financial conditions for model CAFOs that EPA assumes for this analysis are
presented in Section 4. Tables 4-5 through 4-9 in that section present the financial data used in
this analysis and include gross farm revenues, net cash flow, and debt-to-asset ratios for this sector,
as derived by EPA using the 1997 ARMS data.
Table 7-11 presents the estimated post-tax annualized compliance costs per animal (in 1997
dollars) for the hog sector under the proposed BAT Option (Option 5). Table 7-12 presents cost
estimates for an alternative to Option 5 (Option 5 A). Option 5 A is an option that EPA
investigated only for the hog sector that adds additional requirements to Option 5 regarding dry
manure handling systems.4
As shown in Table 7-11, post-tax costs for the BAT Option range from $4.50 per animal to
$9.40 per animal for grow-finish hog operations and from $3.60 per animal to $8.20 per animal for
farrow-finish hog operations. The range of costs for each type is explained by difference in the
assumed availability of land for manure applications (see definition of Category 1,2, and 3 in
Section 4.1), as well as differences across production regions and facility size. Table 7-11 also
presents the range of post-tax annualized compliance costs per CAFO in the grow-finish and
farrow-finish hog sectors. Per CAFO compliance costs for the BAT Option (Option 5) range from
$5,430 to $54,910 per year for grow-finish hog operations and from $4,360 to $80,660 per year
for farrow-finish hog operations. As shown hi Table 7-12, estimated costs for the alternative
Option 5A are much higher than the BAT Option costs; estimated costs for Option 5 A are
estimated to range from $15.40 to .$24.80 per animal.
3The estimated costs are the same across the NPDES Scenarios, i.e., technology option costs do not
change by scenario, although total costs change due to the difference in numbers of CAFOs affected under each
scenario.
4This alternative option is described in Section VIII in the preamble and in the Development Document
(USEPA, 2000a). As described in the preamble, EPA rejected this option on the basis of cost.
7-22
-------
Table 7-11. Per-Anima! and Per-Facility Post-tax Annualized Compliance Costs (Option 5)
Sector
Hog-GF
Hog-FF
Reg.
MW
MA
MW
MA
Size
Ml (a)
Ml(b)
• M2.
LI
L2
Ml(a)
Ml(b)
M2
LI
L2
.Small
Ml(a)
Ml(b)
M2
LI
L2
Ml(a)
Ml(b)
M2
LI
L2
Avg.
Animals
Per
Facility
900
1,422
2,124
3,417
10,029
963
1,521
2,184
3,554
8,893
750
814
1,460
2,152
3,444
13,819
846
1,518
2,165
3,509
17,118
Cakl
Cat. 2
Cat. 3
Per Hog
Cat. 1
Cat. 2
Cat. 3
Per Facility
($1997)
$7.30
$6.82
$5.69
$5.68
$5.02
$7,53
$7.09
$6.90
$5.80
$5.13
$5.80
$7.45
$6.79
$5.68
$5.68
$4.63
$7.69
$7.13
$6.91
$5.80
$4.71
$6.03
$8.72
$6.81
$5.43
$4.45
$9.43
$5.79
$8.55
$7.55
$4.97
NA
$6.64
$6.03
$4.35
$6.19
$5.44
$5.87
$8.18
$8.10
$7.05
$3.63
$7.47
$7.00
$5.90
$6.20
$5.48
$7.34
$7.21
$6.87
'$6.18
$5.52
NA
$6.96
$6.37
$5.35
$5.61
$4.58
$6.92
$6.33
$6.17
$5.61
$4.56
$6,573
$9,698
$12,076
$19,416
$50,301
$7,255
$10,777
$15,068
$20,610
$45,578
$4,354
$6,061
$9,918
$12,233
$19,553
$63,968
$6,503
$10,823
$14,961
$20,339
$80,657
$5,426
$12,395
$14,468
$18,539
. $44,674
$9,084
$8,805
$18,677
$26,841
$44,160
NA
$5,408
$8,801
$9,351
$21,326
$75,202
$4,967
$12,420
$17,543
$24,735
$62,167
$6,723
$9,953
$12,532
$21,174
$54,909
$7,071
$10,972
$15,014
$21,975
$49,066
NA
$5,667
$9,299
$11,512
$19,315
$63,311
$5,851
$9,614
$13,358
$19,677
$77,991
Source: USEPA. See Table 4-1 for definitions of model regions and sizes. Costs reflect the estimated capital costs,
annual operating and maintenance costs, start-up or first year costs, and also recurring costs assumed by EPA (see
the Development Document, USEPA, 2000a) that are annualized using the approach described in Appendix A.
7-23
-------
Table 7-12. Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 5A)
Sector
Hog-
GF
Hog-
FF
Reg.
MW
MA
MW
MA
Size
Ml
M2
LI
L2
Ml
M2
LI
L2
Ml
M2
LI
L2
Ml
M2
LI
L2
Avg
Animals
Per
Facility
1,422
2,124
3,417
10,029
1,521
2,184
3,554
8,893
1,460
2,152
3,444
13,819
1,518
2,165
3,509
17,118
Cat.1
Cat. 2
Cat. 3
Per Hog
Cat. 1
Cat. 2
Cat. 3
Per Facility
($1997)
$19.38
$23.48
$22.96
$22.48
$20.91
$23.64
$24.70
$22.52
'$18.49
$23.49
$23.04
$22.63
$19.94
$23.65
$24.79
$22.68
$17.41
$21.04
$20.71
$20.21
$18.99
$21.49
$22.80
$20.98
$17.08
$21.66
$20.91
$20.38
$18.30
$21.73
$22.95
$21.10
$16.09
$19.59
$19.47
$19.44
$17.16
$19.53
$20.86
$19.44
$15.35
$19.58
$19.47
$19.44.
$16.37
$19.53
$20.86
$19.43
$27,560
$49,880
$78,439
$225,434
$31,800
$51,634
. $87,768
$200,314
$26,998
$50,546
$79,336
$312,790
$30,276
$51,208
$87,002
$388,182
$24,756
$44,684
$70,753
$202,681
$28,877
$46,936
$81,032
$186,532
$24,932
$46,607
$72,002
$281,575
$27,773
$47,045
$80,523
$361,244
$22,882
$41,601.
$66,517
$194,965
$26,096
$42,660
$74,133
$172,897
$22,415
$42,145
$67,042
$268,587
$24,846
$42,291
$73,196
$332,669
Source: USEPA. See Table 4-1 for definitions of model regions and sizes. Costs reflect the estimated capital costs,
annual operating and maintenance costs, start-up or first year costs, and also recurring costs assumed by EPA (see
the Development Document, USEPA, 2000a) that are annualized using the approach described in Appendix A.
For Option 5A, Medium 1 refers to Medium Ib costs. Costs for Medium la and Small are not estimated.
Compared to other regulatory analyses of the hog sector that have been conducted
(NCSU,1999; Fleming et al., 1997; Babcock, et al., 1997; Environmental Defense, 2000), EPA's
estimated costs used for this analysis reflect the upper end of other estimated per-unit costs. For
example, a study by Environmental Defense compiled per-unit costs from a variety of available
research and show estimated costs of alternative manure management technologies that range from
a cost savings to an operation up to about $8 in costs per finished hog (Environmental Defense,
2000). Researchers at Iowa State University estimate an annualized cost for various manure
storage and management practices of under $1 per animal to as much as$12 per sow and $7 per .
market hog (Fleming et al., 1997; Babcock et al. 1997). Researchers at North Carolina State
7-24
-------
University also estimate costs for a range of practices, estimated at up to $10 per finished hog
(NCSU, 1999). In general, these costs arecamortized but do not take into account tax savings.
EPA's equivalent pre-tax costs per animal would be roughly 40 percent greater than those shown
in Table 7-11 and generally exceed other reported upper end values. As documented in the
Development Document (USEPA, 2000a), EPA believes that its estimated costs are conservative.
The costs presented here are those assumed to be incurred by the regulated CAFO and do
not account for the likelihood that some compliance costs will be passed on through the marketing
levels in the industry.
Table 7-13 presents the range of per animal post-tax compliance costs in 1997 dollars for
grow-finish and farrow-finish operations for each option, including the BAT Option and Option
5A. (The proposed and alternative ELG Option and NPDES Scenarios considered by EPA during
this rulemaking are defined in Table 3-1.) As shown, for both farrow- and grow-finish hog
operations, costs for options other than Option 5A range from under $0.10 to about $18 per
animal. The proposed BAT Option costs fall between these ranges.
Table 7-13. Summary of the Range of Post-Tax Annualized Compliance Costs Per Hog, By Option
Option
Option.1 .
Option 2
Option 3
Option 4
Option 5
Option 5A
Option 6
Option 7
Hog-GF
Minimum
Maximum
Hog-FF
Minimum
Maximum
($1997)
$0.04
$1.69
$1.94
$2.21
$4.45
$16.09
$1.36
$1.69
$10.20
$7.85
$18.74
$13.00
$9.43
$24.70
$7.85
$13.00
$0.03
$1.53
$1.76
$1.94
$3.63
$15.35
$1.93
$1.53
$9.01
$7.03
$18.08
$12.72
$8.18
$24.79
$7.15
$10.53
Source: USEPA.
7.3.2 Estimates of National Annual Compliance Costs
Table 7-14 presents EPA's estimate of the aggregate national level compliance costs for the
hog sector under the proposed BAT Option (Option 5) and the co-proposed two-tier structure
(Scenario 4a at 500 AU threshold) and the three-tier structure (Scenario 3). Costs under the two-
tier structure at the 750 AU threshold (Scenario 5) are also briefly discussed, along with other
7-25
-------
regulatory alternatives considered by EPA during this rulemaking. The description of the
proposed BAT Option and the co-proposed NPDES Scenarios is provided in Section 3.
For the hog sector, EPA estimates total incremental cost (post-tax) of the proposed BAT
Option at $199 million per year under the two-tier structure at the 500 AU threshold (Table"7-14).
About three-quarters of this total estimated cost is for operations with more than 1,000 AU. The
cost of the proposed BAT Option under the three-tier structure is estimated at $184 million per
year; about 80 percent of this cost is for operations with more than 1,000 AU (Table 7-14).
Between the two modeled regions (MA and MW), the MW region bears the largest portion (65
percent) of the total costs under both of the co-proposed tier structures.
Table 7-14 also shows other estimated costs for this sector. The proposed BAT Option at
the 750 AU threshold will cost the hog sector $170 million per year (Table 7-14). The costs of
Option 5A are estimated at nearly $930 million annually.
Table 7-14. Total Estimated Post-Tax Compliance Costs
Scenario/Size
>1,000 AU
Total Two-Tier
Structure (>750 AU)
Total Two-Tier
Structure (>500 AU)
Total Two-Tier Structure
(>300AU)
Total Three-Tier
Structure (>300 AU)
Option
Number of CAFOs
BAT Option
Alternative Options
Number of GAFOs
BAT Option
Alternative Options
Number of CAFOs
BAT Option
Alternative Options
Number of CAFOs
BAT Option
Alternative Options
Number of CAFOs
BAT Option
Alternative Ojrtions
Grow-Finish
Farrow-Finish
($1997 millions, except where noted)
1,670
$51.4
$33.2-$51.4
,2,300
$60.0
$36.1 - $60.0
2,690
$63.8
$37.3-$63.8
4,920
$78.5
$42.0-878.5
2,650
$59.7
$35.9-$59.7
2,420
$97.3
$44.1-$97.3
3,460
$110.3
$46.9 -$110.3
6,060
$135.1
$51.5-$135.1
9,590
$156.7
$56.2-$ 156.7
5,750
$124.7
$49.9-$ 124.7
Source: USEP A.
Numbers of CAFOs include defined CAFOs only. Costs include those for designated hog operations.
7-26
-------
7.3.3 Analysis of CAFO Financial Impacts
EPA's impact analysis uses a representative farm approach to estimate the number of
CAFOs that would experience affordable, moderate, or stress impacts as a result of the CAFO
regulations, as described in Section 4. Economic achievability is determined by applying the
proposed criteria, which include a sales test and also analysis of post-compliance cash flow and
debt-to-asset ratio for an average model CAFO. EPA extrapolates impacts to all CAFOs in the
hog sector using the estimated number of operations represented by each model CAFO.
As described in Section 4.2.5, if an average model facility is determined to incur economic
impacts under regulation that are regarded as "Affordable" or "Moderate," then the proposed
regulations are considered economically achievable. ("Moderate" impacts are not expected to
result in closure and are considered to be economically achievable by EPA.) If an average
operation is determined to incur "Stress," then the proposed regulations are not considered to be
economically achievable. "Affordable" and"Moderate" impacts are associated with positive post-
compliance cash flow over a 10-year period and a debt-to-asset ratio not exceeding 40 percent, in
conjunction with a sales test result that shows that compliance costs are less than 5 percent of sales
("Affordable") or between 5 and 10 percent ("Moderate"). "Stress" impacts are associated with
negative cash flow or if the post-compliance debt-to-asset ratio exceeds 40 percent, or sales test
results that show costs equal to or exceeding 10 percent of sales. ,
Using this classification scheme, EPA's analysis indicates that some hog operations would
experience financial stress as a result of the proposed CAFO regulations under the proposed BAT
Option and both co-proposed scenario, assuming compliance costs cannot be passed through the
marketing chain. Tables 7-15 and 7-16 present the results of EPA's analysis. A total of 1,420 hog
operations (17 percent of defined CAFOs) are expected to experience financial stress under both of
the co-proposed tier structures, including the two-tier structure at 750 AU. The hog operations
with these impacts have more than 1,000 AU on site (i.e., no operations with between 300, 500,
750 and 1,000 AU fall in the stress category). No designated CAFOs are expected to experience
financial stress under either co-proposed scenario. Based on these results, EPA proposes that the
proposed CAFO regulations are economically achievable under the co-proposed scenarios.
EPA also evaluates financial impacts with an assumption of cost passthrough. For the
purpose of this analysis, EPA assumes that the hog sector could pass through 46 percent of
compliance costs. EPA derived these estimates from price elasticities of supply and demand for
each sector reported in the academic literature (see Section 4). Assuming this level of cost
passthrough, the magnitude of the estimated impacts decreases to the affordable or moderate
impact category under the proposed BAT Option and the co-proposed scenarios (Table 7-15).
Section 5 provides additional information on how the co-proposed scenarios compare with
the alternative scenarios EPA considered.
7-27
-------
Table 7-15. Impacted CAFOs under ELG Options & NPDES Scenarios, Grow-Finish Hog Operations
Alternative
ELG Options
and
NPDES Scenarios
Total
#
CAFOs
Affordable
Moderate
Stress
Zero Cost Passthrough
Aff.
Mod.
Stress
Partial Cost Passthrough
(Number of Affected Operations)
Two-Tier (>1000)
BAT Option
Alt ELG Options
1,670
680
680-1,230
Two-Tier (>750 AU, Scenario 5)
BAT Option
Alt. ELG Options
2,300
1,310
1,230-1,850
180
180-290
810
180-810
1 £H(\
1,O /U
0
0
180
180-370
810
180-810
2,300
0
0
Two-Tier (>500 AU, Scenario 4a)
BAT Option
Alt. ELG Options
2,690
1,710
1,580-2,250
180
180-410
810
180-810
2,690
0
0
Two-Tier (>300 AU, Scenario 4b)
BAT Option
Alt. ELG Options
4,920
3,900
3,180-4,470
210
210-980
810
180-810
4,920
0
0
Three-Tier (Scenario 3)
BAT Option
Alt ELG Options
2,650
1,660
1,440-2,210
190
190-500
810
180-810
2£C(\
,tou
0
0
0
0
0
0
0
0
0
0
0
0
Source: USEPA. Numbers may not add due to rounding. Option/Scenario definitions provided in I able
' (Section 3). Category definitions ("Affordable," "Moderate" and "Stress") are provided in Table 4-13 (Section 4).
Tables 7-17 and 7-18 present a more detailed breakout of EPA's affordability results under
the proposed BAT Option by model CAFO type, land availability, and type of operation (both
grow-finish and farrow-finish). The results are the same for the two-tier and three-tier structure
because only the numbers of CAFOs represented by each model type changes. The impacts are
presented by model CAFO and indicate the level of impact under each of the economic
affordability criteria. Zero cost passthrough is assumed.
7-28
-------
Table 7-16. Impacted CAFOs under ELG Options & NPDES Scenarios, Farrow-Finish Hog Operations
Alternative
ELG Options
and
NPDES Scenarios
Two-Tier (>1000)
BAT Option
Alt. ELG Options
Total
#
CAFOs
Affordable
Moderate
Stress
Zero Cost Passthrough
Aff.
Mod.
Stress
Partial Cost Passthrough
(Number of Affected Operations)
2,420
1,780
1,650-2,270
30
30-190
610
130-630
Z?4ZU
0
0
0
0
Two-Tier (>750 AU, Scenario 5)
BAT Option
Alt ELG Options
3,460
2,820
2,650-3,300
30
30-220
610
130-630
3,460
0
0
0
0
Two-Tier (>500 AU, Scenario 4a)
BAT Option
Alt. ELG Options
5,860
5,210
5,010-5,700
30
30-260
610
130-630
5,860
0
0
0
0
Two-Tier (>300 AU, Scenario 4b)
BAT Option
Alt. ELG Options
9,450
8,810
7,940-9,290
30
30-890
610
130-630
9,450
0
0
0
0
Three-Tier (Scenario 3)
BAT Option
Alt. ELG Options
5,710
5,070
4,590-5,550
30
30-520
610
130-630
5,710
0
0
0
0
Source: USEPA. Numbers may not add due to rounding. Option/Scenario definitions provided in Table 3-1
(Section 3). Category definitions ("Affordable," "Moderate" and "Stress") are provided in Table 4-13 (Section 4).
These tables show that the financial stress impacts for hogs are being driven by the debt-to-
asset ratios and sales tests (a debt-to-asset ratio greater than 0.40 with a sales test greater than 3
percent is considered an indicator of the potential for financial stress, even if cash flow is positive).
These results are associated with the Large 1 and 2 models in the MA region. Under an
assumption of partial cost passthrough (not shown in the table), these same models show
acceptable sales tests and debt-to-asset results, and all models indicate "affordable" impacts.
7-29
-------
Table 7-17. Economic Affordabilitv Results for Hoe CAFOs, Grow-Finish Operations
Size
Category 1
Sales
DCF
DA
Category 2
Sales
DCF
DA
Category 3
Sales
DCF
DA
Hog-GF-Zero CPT
MW Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
2.4%
2.2%
1.9%
2.5%
2.2%
Pass
Pass
Pass
Pass
Pass
0.29
0.29
0.29
0.46
0.46
2.0%
2.9%
2.2%
2.4%
1.9%
Pass
Pass
Pass
Pass
Pass
0.28
0.28
0.28
0.45
0.45
2.5%
2.3%
1.9%
2.7%
2.4%
Pass
Pass
Pass
Pass
Pass
0.29
0.29
0.29
0.46
0.46
MA Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
4.3%
4.1%
4.0%
6.9%
6.1%
Pass
Pass
Pass
Pass
Pass
0.16
0.16
0.16
0.41
0.41
5.4%
3.3%
4.9%
9.0%
5.9%
Pass
Pass
Pass
Pass
Pass
0.16
0.15
0.15
0.39
0.38
4.2%
4.1%
3.9%
7.4%
6.6%
Pass
Pass
Pass
Pass
Pass
0.16
0.16
0.16
0.41
0.41
Hog-GF-Partial CPT
MW Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
1.3%
1.2%
1.0%
1.3%
1.2%
Pass
Pass .
Pass
Pass
Pass
0.27
0.27
0.27
0.44
0.44
1.1%
1.5%
1.2%
1.3%
1.1%
Pass
Pass
Pass
Pass
Pass
0.27
0.27
0.27
0.43
0.43
1.3%
1.2%
1.0%
1.5%
1.3%
Pass
Pass
Pass
Pass
Pass
0.27
0.27
0.27
0.44
0.44
MA Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
2.3%
2.2%
2.1%
3.7%
3.3%
Pass
Pass
Pass
Pass
Pass
0.15
0.15
0.15
0.37
0.37
2.9%
1.8%
2.6% .
4.9%
3.2%
Pass
Pass
Pass
Pass
Pass
0.14
0.14
0.14
0.35
0.35
2.3%
2.2%
2.1%
4.0%
3.6%
Pass
Pass
Pass
Pass
Pass
0.15
0.15
0.15
0.37
0.37
Source: USEPA.
7-30
-------
Table 7-18. Economic Affordability Results for Hog CAFOs, Farrow-Finish Operations
Size
Category 1
Sales
DCF
DA
Category 2
Sales
DCF
DA
Category 3
Sales
DCF
DA
Hog-FF-Zero CPT
MW Region
Small
Medium la
Medium Ib
Medium 2
Large 1
Large 2
1.0%
2.5%
2.2%
1.9%
2.5%
2.0%
Pass
Pass
Pass
Pass
Pass
Pass
0.29
0.29
0.29
0.29
0.46
0.46
NA
2.2%
2.0%
1.4%
,2.7%
2.4%
NA
Pass
Pass
Pass
Pass
Pass
NA
0.28
0.28
0.28
0.45
0.45
NA •
2.3%
2.1%
1.8%
2.4%
2.0%
NA
Pass
Pass
Pass
Pass
Pass
NA
0.29
0.29
0:29
0.46
0.46
MA Region
Medium la
Medium Ib*
Medium 2
Large 1
Large 2
4.4%,
4.1% '
4.0%
6.9%
5.6%
Pass
Pass
Pass
Pass
Pass
0.16
0.16
0.16
0.41
0.41
3.4%
4.7%
4.6%
8.4%
4.3%
Pass
Pass
Pass
Pass
Pass
0.16
0.15
0.15
0.39
0.38
4.0%
3.6%
3.5%
6.7%
5.5%
Pass
Pass
Pass
Pass
Pass
0.16
0.16
0.16
0.41
0.41
Hog-FF-Partial CPT
MW Region
Small
Medium la
Medium Ib
Medium 2
Large 1
Large 2
0,5%
1.3%
1.2%
1.0%
1.3%
1.1%
Pass
Pass
Pass
Pass
Pass
Pass
0.27
0.27
0.27
0.27
0.44
0.44
. NA
1.2%
1.1%
0.8%
1.5%
1.3%
NA
Pass
Pass
Pass
Pass
Pass
.NA
0.27
0.27
0.27
0.43
0.43
NA
1.2%
1.1%
1.0%
1.3%
1.1%
NA
Pass
Pass
• Pass
Pass
Pass
NA
0.27
0.27
0.27
0.44
0.44
MA Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
2.4%
2.2%
2.1%
3.7%
3.0%
Pass
Pass
Pass
Pass
Pass
0.15
0.15
0.15
0.37
0.37
1.8%
2.5%
2.5%
4.6%
2.3%
Pass
Pass
Pass
Pass
Pass
0.14
0.14
0.14
0.35
0.35
2.1%
2.0%
1.9%
3.6%
2.9%
Pass
Pass
Pass
Pass
Pass
0.15
0.15
0.15
0.37
0.37
Source: USEPA.
7-31
-------
7.4 PROCESSOR ANALYSIS
As discussed in Section 4.3, EPA does not conduct a detailed estimate of the costs and
impacts that would accrue to individual co-permittees due to lack of data and market information.
However, EPA believes that the framework used to estimate costs to CAFO provides a means to
evaluate the possible upper bound of costs that could accrue to potential co-permittees, based on
the potential share of (pre-tax) costs that may be passed on from the CAFO (described in Section
4.3). EPA is proposing that this amount approximates the magnitude of the costs that may be
incurred by processing firms in those industries that may be affected by the proposed co-permitting
requirements.
Table 7-19 presents the results of EPA's analysis. This analysis focuses on the potential
magnitude of costs to co-permittees in the pork sector. As presented in Section 2, EPA estimates
that about 94 hog processors may be subject to the proposed co-permitting requirements. Using
the framework to estimate costs and impacts to regulated CAFOs, EPA calculates the estimated
upper bound of costs that could accrue to hog processors based the estimated pre-tax cost
estimated for CAFOs, assuming that either all or a portion of these costs are absorbed by
processors as markets adjust to the proposed CAFO regulations. EPA's partial cost passthrough
scenario assumes that 46 percent of all hog compliance costs are passed on to the food processing
sectors. (For more information on this approach, see Section 4.2).
Using this approach, EPA estimates that the range of potential annual costs to hog
processors is $135 million (partial cost passthrough, two-tier structure) to $306 million (full cost
passthrough, three-tier structure). These costs, shown in Table 7-19, are expressed in 1999 pre-
tax dollars.
To assess the magnitude of impacts that could accrue to processors using this approach,
EPA compares the passed through compliance costs to both aggregate processor costs of
production and to revenues (a sales test). The results of this analysis are shown in Table 7-19 and
are presented in terms of the equivalent 1999 pre-tax compliance cost as compared to 1997 data
from the Department of Commerce on the revenue and costs among processors in the hog
industries. As shown, EPA estimates that, even under full cost passthrough, incremental cost
changes are less than two percent and passed through compliance costs as a share of revenue are
less than one percent
This suggested approach does not assume any addition to the total costs of the rule as a
result of co-permitting. This approach also does not assume that there will be a cost savings to
contract growers as result of a contractual arrangement with a processing firm. This approach
merely attempts to quantify the potential magnitude of costs that could accrue to processors that
may be affected by the co-permitting requirements. Due to lack of data, EPA does not conduct a
detailed analysis of the costs and impacts that would accrue to individual co-permittees.
Additional limitations of this approach as recognized by EPA are discussed in Section 4.3.
7-32
-------
Table 7-19. Imi
Sector
Two-Tier
Three-Tier
jact of Passed Through Compliance Costs under Co-proposed Alternatives, Hog Sector
Passed Through
Compliance Cost ^
Partial
CPT
100%
CPT
($1999, million)
$135
$141
$294
$306
1997
Revenues
1997
• Delivered
Cost"'
($1997, million)
$38,500
$15,700
Passed through
Cost-to-Revennes
Partial
CPT
100%
CPT
Passed through
Cost-to-Delivered
Cost
Partial
CPT
100%
CPT
(percent, comparing costs in $1997)
0.3%
0.4%
0.7%
0.8%
0.8%
0.9%
1.8%
1.9%
Source: USEPA. 1997 processor revenues and costs are from the Department of Commerce (USDC, 1999a).
Option/Scenario definitions provided in Section 3. Proposed BAT Option is Option 5. Estimated compliance costs
are pre-tax.
"'Pre-tax compliance costs that are estimated to be passed from the production operation to the processors. CPT =
Cost passthrough. Partial CPT assumes 46% CPT for the hog sector (see Section 4.2.6).
''Delivered costs include all raw materials put into production during the year.
7.5 MARKET ANALYSIS
This section presents the results of EPA's market model analysis for the hog sector. The
results presented in this section briefly compare the results of the two-tier (500 AU threshold) and
the three-tier (Scenario 3) structures that are being co-proposed by EPA. These results measure
changes for the pork industry as a whole and do not differentiate between the types of operations
in the sector. Additional results on the alternative regulatory options and scenarios considered by
EPA as part of this rulemaking are provided in Section 5.4. For further explanation of the market
model and sources of the baseline input data, see Section 4.4 and Appendix B.
A summary of the key results of the market model is shown in Table 7-20 for the two-tier
and three-tier structures indicating the predicted changes in farm and retail prices, quantities,
national and regional employment, and national economic output.
Compared to a baseline producer price of $54.30 per hundredweight (cwt), EPA's market
model predicts that the proposed CAFO regulations will raise producer prices by $0.59 per cwt to
$0.64 per cwt, or less than 1.2 percent of the baseline producer price, depending on the
co-proposed tier structure (Table 7-20). At the retail level, consumer prices for pork products will
rise about one cent per pound. These price increases are driven by slight changes in the amount of
pork products produced at the farm level and thus available for consumption (Table 7-20). At the
commodity level, EPA's market model predicts that U.S. pork imports will rise by about 0.2
percent, compared to baseline imports; U.S. pork exports will decrease by about 0.3 percent
compared to baseline.
Absorption of compliance costs by the producers and small declines in quantities are
expected to result in fewer jobs in the hog industry. Table 7-20 also presents EPA's estimates of
7-33
-------
both the direct (i.e., farm and processor level) and total (i.e., national level) reductions in
employment for the hog sector. Overall, EPA estimates changes in national aggregate employment
to range from a total reduction of 6,380 to 6,880 jobs, measured in full-time equivalents (FTEs).
This analysis also does not adjust for offsetting increases in other parts of the economy and other
sector employment that may be stimulated as a result of the proposed regulations, such as the
construction and farm services sectors.
EPA's projected job losses are estimated throughout the entire economy, using available
modeling approaches described in Section 4, and are not attributable to the regulated community
only. As shown in Table 7-20, about 80 percent of these estimated job losses are in the non-
agricultural or farm services support industries (i.e., indirect or induced employment affects; see
Section 4.4).
At the CAFO level, EPA predicts that job losses in the farming sector associated with the
proposed CAFO regulations will range from 930 to 1,010 jobs under the proposed BAT Option,
depending on tier structure (Table 7-20). These estimates include CAFO owner-operators and
employed family members, as well as hired farm labor. This estimated reduction compares to an
estimated total farm level employment of 195,900 FTEs in the hog sector nationwide (Table 2-17;
Abel, Daft, and Barley, 1993, as updated by EPA). EPA estimates that job losses in the hog
processing sectors will range from 250 to 270 (Table 7-20). These estimated losses compare to
the more than 140,000 persons employed in hog processing in 1997 (USDC, 1999a).
Changes in employment and earnings can affect the vitality of local communities.
Community impacts are usually determined by employment changes at individual facilities. As
facility-specific information and analysis were not within the scope of this study, EPA is not able to
speculate on community impacts. However, EPA disaggregates the national employment results to
examine the potential regional employment impacts of the proposed CAFO regulations. The
method EPA uses to allocate impacts is based on hog production and does not take into account
existing environmental practices or other production factors (see Section 4.4). Table 7-20 shows
that the traditional hog growing regions of the Midwest would be the most affected, followed by
the Mid-Atlantic. None of the impacts represent a significant share of total employment in these
regions. Compared to the baseline, EPA estimates the loss in hog agricultural employment at
under 0.01 percent; about 70 percent of the estimated agricultural job losses in the hog sector are .
expected in the more traditional Midwest region (Table 7-20). Economy-wide employment losses
are estimated at under 0.01 percent compared to the baseline.
7-34
-------
Table 7-20. Summary of Market Model Results for the Hoe Sector
Variable
Pre-Regulatory
Value/Units
Two-Tier Structure
BAT
Option
Range of
Alternative
Options
Three-Tier Structure
BAT
Option
Range of
Alternative
Options
Farm Products
Price
Quantity Produced
Quantity Exported
Quantity Imported
$54.30/cwt
23,542 mil. Ibs.
14 mil. Ibs.
814 mil. Ibs.
$54.89
23,472
14
819
$54.56 - 54.89
23,472-23,511
14-14
814-819
$54.94
23,467
14
820
$54.55 - 54.94
23,467-23,512
14-14
816 - 820
Retail Products
Price
Quantity Demanded
Quantity Exported
Quantity Imported
$2.45/lb.
17,274 mil. Ibs.
1,044 mil. Ibs.
633 mil. Ibs.
$2.46
17,229
1,041
634
- $2.45-2.46
17,229 - 17,254
1,041 - 1,042
634 - 634
$2.46
17,225
1,041
634
$2.45 - 2.46
17,225 - 17,256
1,041 - 1,043
634 - 634
Employment Reduction
Direct Farm
Direct Processor
Total Economy
195,900 FTEs "
84,723 FTEs
129.6 mil. FTEs
931
250
6,376
416-931
111-250
2,849-6,376
1,005
269
6,876
402-1,005
108 - 269
2,752 - 6,876
Output Reduction
National
$ million
655
294 - 655
707
283 - 707
Regional Farm and Processing Employment Reduction
Pacific
Central
Midwest
South
Mid-Atlantic
Total
FTEs
FTEs
FTEs
FTEs
FTEs
FTEs
6
88
753
43
292
1,181
3-6
39 - 88
336 - 753
19-43
130 - 292
528-1,181
6
94
812
47
315
1,274
2-6
38 - 94
325 - 812
19-47
.126-315
510 - 1,274
Source: Post-regulatory changes are estimated by USEPA. Pre-regulatory prices, quantities, and trade volumes, see
Table 4-16 (Section 4). Pre-regulatory employment, see Table 2-17 (Section 2).
a/lFTE = 2,080 hours of labor. • .
7-35
-------
-------
SECTION EIGHT
SUMMARY OF ECONOMIC IMPACTS:
BEEF AND DAIRY SUBCATEGORIES
This section presents a profile of the beef and dairy industry, including farmers (Section
8.1) and processors (Section 8.2) in the cattle and dairy industries. Following the industry profile,
this section provides a detailed summary of EPA's economic analysis of the proposed CAFO
regulations as it affects regulated CAFOs (Section 8.3), processors (Section 8.4), and national
markets (Section 8.5). .
8.1 PROFILE OF THE BEEF AND DAIRY PRODUCTION SECTORS
This section presents a profile of cattle and milk production operations and provides
background information for analyzing the costs and benefits of the proposed CAFO regulations.
The purpose of this profile is to provide a baseline description of the current activities, structure,
and performance of the beef and dairy production industries. The following sections describe the
types of operations in this sector and present an overview of the industry, describing the number
and size of operations (including the subset of regulated operations), geographic distribution,
supply and demand conditions, price trends, and the financial conditions that characterize this
sector.
8.1.1 Industry Definition
Beef and Veal Operations
Beef cattle feedlots (includes veal) are identified as NAICS 112112 (SIC 0211). This
sector comprises establishments primarily engaged in feeding cattle for fattening, including: beef
cattle feedlots (except stockyards for transportation) and feed yards (except stockyards for .
transportation). This NAICS code includes operations where beef cattle sales account for the
majority of revenues. USDA defines beef feedlots as those operations that fed any cattle over the
previous year. Unless otherwise noted, EPA uses this broader .USDA definition of beef feedlots
throughout this report.
The beef cattle industry can be divided into four separate producer segments:
~ • Feedlot operations fatten or "finish" feeder cattle prior to slaughter and constitute
the final phase of fed cattle production. Calves usually begin the finishing stage
after 6 months of age or after reaching at least 400 pounds. Cattle are typically
held for 150 to 180 days and weigh between 1,150 to 1,250 pounds (for steers) or
1,050 to 1,150 pounds (for heifers) at slaughter. .
8-1
-------
• Stacker or backgrounding operations coordinate the flow of animals from
breeding operations to feedlots by feeding calves after weaning and before they
enter a feedlot. Calves are kept between 60 days to 6 months or until they reach a
weight of about 400 pounds (Rasby et al., 1994).
• Veal operations raise male dairy calves for slaughter. The majority of calves are
"special fed" or raised on a low-fiber diet until about 16 to 20 weeks of age, when
they weigh about 450 pounds.
» Cow-calf producers typically maintain a herd of mature cows, some replacement
heifers, and a few bulls, and breed and raise calves to prepare them for fattening at
a feedlot. Calves typically reach maturity on pasture and hay and are usually sold
at weaning. Cow-calf operators may also retain the calves and continue to raise
them on pasture until they reach 600 to 800 pounds and are ready for the feedlot.
As cow-calf and stocker/backgrounding operations primarily graze cattle and calves, then-
activities are not expected to be covered under the proposed CAFO regulations.
After slaughter, live cattle are converted into cuts of meat and various processed foods by
meat packers and processors and sold to consumers at retail establishments. Other by-products,
such as hides, lard, and offal, have value in the manufacture of clothing, foodstuffs, fertilizers, and
other industrial products. By NAICS code, the beef processing sector includes animal
slaughterers1 (NAICS 311611), meat (from carcass) processors (NAICS 311612), and rendering
and byproduct processing facilities (NAICS 311613). Additional information on the processing
sector is presented in Section 8.2.
Dairy and Heifer Operations
Production operations that produce cow's milk are classified under NAICS 11212, dairy
cattle and milk production (SIC 0241, dairy farms). Industry coverage between the NAICS and
SIC classifications is not equivalent: NAICS 11212 does not include dairy heifer replacement
farms but SIC 0241 does. Dairy heifer replacement is now classified under NAICS 112111, Beef
Cattle Ranching and Farming. Therefore, the definition under NAICS is more focused on milk
production. A dairy operation may have several types of animal groups present, including:
• Calves (0-5 months);
• Heifers (6-24 months);
JNAICS 311611 covers the slaughter of cattle, calves, steer, heifers, pork, sheep, and lamb.
8-2
-------
Lactating dairy cows (i.e., currently producing milk); and;
Cows close to calving and dry cows (i.e., not currently producing milk); and
Bulls.
Heifer replacement operations raise pre-calving cows themselves or under production
contracts for all or part of the growing period from weaning to calving (USDA/APHIS, 1993).
With increasing specialization, many large dairy operations contract out the raising of replacement
heifers to focus on milk production (Faust, 1995). The use of heifer replacement operations
allows dairies to expand their herd size on the existing facilities and specialize in the production of
milk, thus potentially increasing profitability (USDA/APfflS, 1.993). In 1991-92, an estimated 1.7
percent of dairy farmers had someone else raise their heifers on a contract basis, with larger
operations contracting for replacement heifers more often than small operations (USDA/APHIS,
1993). Cady (2000) estimates that approximately 15 percent of dairy cows have been raised by a
replacement operation at some stage in their rives, with approximately 1 million heifers being
raised by replacement operations at any one time. Heifers can be raised either on pasture in
warmer climates or in confinement in dry feedlots; however, the majority of operations with over
1,000 head are confined feedlot operations (Cady, 2000). Thus these operations may be defined
as CAFOs under the proposed GAFO regulations and would be the only such CAFOs in the beef
and dairy sector where production contracts are somewhat common.
Although heifer operations are .typically characterized as part of the dairy industry, EPA's
analysis groups heifer operations along with the beef cattle and veal operations.
USDA defines dairy farms as those farms that had any dairy cattle on farm during the
previous year. Unless otherwise noted, EPA has used the broader USDA definition of a dairy
farm throughout this report. Beyond the farm level, raw farm milk is converted into processed
fluid milk and dairy products by dairy cooperatives and processing firms. By NAICS code, dairy
processors are classified under dry, condensed and evaporated dairy manufacturing (NAICS
311514); fluid milk manufacturing (NAICS 311511); creamery butter manufacturing (NAICS
311512); cheese manufacturing (NAICS 311513); and ice cream and frozen dessert
manufacturing (NAICS 3115120). Additional information on the processing sector is presented
in Section 8.2.
8.1.2 Overview of the Beef and Dairy Industry
For the purpose of this industry profile, the cattle feeding sector covers operations that
raise beef cattle, veal and heifers. Industry characteristics of these sectors are presented jointly in
this overview. Industry characteristics of the dairy industry are discussed separately.
8-3
-------
Over the past few decades, the total number of cattle feedlots has declined, while size of
operation has increased. Fed cattle production is now dominated by large feedlots, which account
for 2 percent of all feedlots but 85 percent of annual fed cattle sales. Improvements in technology
and industrial-type entrepreneurial skills are cited as largely responsible for the increase in feedlot
size (Krause, 1991). Fed cattle production is concentrated hi the Great Plains states. Unlike
operations in other livestock and poultry sectors, production contracting is not common among
cattle feedlots, and cattle are owned by the beef operation at nearly 95 percent of all such
operations (USDA/ERS, 1999a). Increasingly, the meat packing industry has become integrated.
with livestock feeding operations, either company-owned or through contracts with custom
feedlots (Kohls and Uhl, 1998). Types of feedlots thus can be roughly divided into company-
owned, custom, and independent feedlots. Custom feedlots are a small part of the industry. Less
than 2 percent of beef operations operate under contracts (USDA/ERS, 1996a). About 20 to 25
percent of all cattle are vertically integrated through ownership of the feedlot or contract
arrangements with packers, however (Kohls and Uhl, 1998).
The U.S. dairy industry has undergone significant structural change in recent years. The
dominant trend in dairy operations has been towards increased consolidation and specialization,
resulting in fewer, larger operations. Large-scale expansion among some producers has raised
total U.S. milk production despite continued reductions in the nation's milk cow herd, indicating
higher per-cow productivity and efficiency gains in the sector. The past few decades have also
witnessed a major geographic shift in milk production from the more traditional producing states
of the Midwest and Mid-Atlantic regions to the West and Southwest where operations are
typified by larger herd size and greater use of technology (El-Osta and Johnson, 1998; McBride,
1999; Manchester and Blayney, 1997).
8.1.2.1 Trends in the Number and Size
USDA reports that in 1997 there were a total of 110,620 fed cattle and calf operations in
the United States, based on sales (USDA/NASS, 1999a). The number of dairy operations totaled
116,880 farms based on year-end inventories (USDA/NASS, 1999a). See Table 8-1. These data
on the number of farms include both commercial and non-commercial operations, as well as
operations that confine and graze animals. This estimate includes all facility types, including
operations mat finish cattle for slaughter and operations that raise breeder stock.
Total cattle and calf operations in the U.S. (including grazing operations) totaled over
800,000 in 1997 (USDA/NASS, 1999a). Feedlot operations account for about one-fourth of
total operations. Data on fed cattle operations are reported by USDA as operations with cattle
and calves fattened on grain and concentrates (based on sales). In 1997, USDA data indicate that
between 1982 and 1997, the number of cattle feeding operations dropped by more than one-half,
from 240,015 operations to 110,620 operations (Table 8-1). Total sales at these operations
remained fairly stable at above 27 million head. However, average herd size at these operations
increased from about 120 head per operation to about 250 head per operation (Table 8-1).
8-4
-------
(USDA/NASS, 1999a). Although fewer than 3 percent of cattle feedlots have capacity of more
than 500 head, they dominate the market sales, accounting for nearly 90 percent of annual sales
(Table 8-1).
Cattle feeding has become increasingly concentrated over the last few decades. Feedlots
have decreased in number, but increased in size: in 1972,104,300 feedlots in. 13 states marketed
23.7 million cattle, while in 1995^ 41,400 feedlots in these same states marketed 23.4 million
cattle (Ward and Schroder, no date). Most of this decrease occurred at feedlots with less than
1,000 head. Meanwhile, feedlots with capacity greater than 1,000 head increased by about 30
percent (Krause, 1991). Also, commercial feedlots control a larger share of the number of cattle
on feed, up from 43 percent of all fed cattle in 1980 to over 50 percent in 1990 (Bastian et al.,
1994). The share of fed cattle raised by feedlots is dropping and accounted for under 20 percent
of all fed cattle in 1990 (Bastian et al., 1994).
Table 8-1. Number of Beef and Dairy Operations and Animals, 1974-1997
Year
Operations
Total Animals
(thousand)
Average Herd
(head)
Percent of Farms
Percent of Animals
(>500 head)
Fed Cattle and Calves (fattened on grain and concentrates)
1974
1978
1982
1987
1992
1997
210,725
247,114
240,015
190,008
147,201
110,620
26,070
29,722
27,674
27,818
26,406
27,328
124
120
115
146
179
247
2.3%
2.5%
0.9%
2.4%
2.7%
3.0% ,
69.1%
72.3%
76.8%
78.2%
82.4%
87.2%
Dairy Cows
1974
1978
1982
1987
1992
1997
403,754
312,095
277,762
202,068
155,339
116,874
10,655
10,222
10,850
10,085
9,492
9,095
26
33
39
50
61
78
0.2%
0.3%
0.4%
0.6%
1.1%
1.9%
5.7%
7.1%
8.5%
11.7% -
17.8%
27.6%
Source: USDA/NASS, 1999a. Cattle/calves fattened on grain and concentrates (sales); dairy (year-end inventory).
8-5
-------
In the dairy sector, USD A reports that there were 116,900 dairy farms with a year-end
inventory of 9.1 million milk cows producing 156.1 billion pounds of milk in 1997 (USDA/NASS,
1999a and 1999c). (See Table 8-1.) In comparison, in 1987, there were 202,100 dairy operations
with 10.1 million cows, producing 142.7 billion pounds of milk (USDA/NASS, 1999a; NMPF,
1999). As these data show, although the number of operations dropped by nearly one-half and
the number of milk cows also decreased slightly, production efficiency at U.S. dairy farms is
increasing (Table 8-1). Average herd size at dairy farms is increasing, up from an average of
under 30 cows per operation in 1974 to almost 80 cows per operation in 1997 (Table 8-1). This
indicates that while the overall number of operations is dropping, the remaining operations are
expanding (USDA/NASS, 1999a; NMPF, 1999). In spite of ongoing consolidation in this sector,
the majority of farms are small in size. As shown in Table 8-1, operations with more than 500
cows accounted for under 2 percent of the total number of farms in 1997. These farms accounted
for almost 30 percent of annual inventories, however (Table 8-1). Most of the operations exiting
the dairy industry tend to be small in size: in 1987, there were 196,500 operations with fewer than
200 cows, compared to 109,700 dairy operations with fewer than 200 cows in 1997
(USDA/NASS, 1999a; NMPF, 1999).
For the purpose of this analysis, EPA estimates the number of confinement operations that
may be subject to the proposed CAFO regulations using 1997 Census data that are aggregated by
USDA's NASS. NASS developed a methodology for identifying farms likely to be CAFOs based
on the Census survey information and estimated animal units on these operations based on
reported data. A summary of these data are provided in the Development DocwmewZ, USEPA,
2000a. These summary data reflect average herd size throughout the year, accounting for both
animals sales and inventories. Where applicable, data are adjusted for the average number of
marketing cycles (USEPA, 2000a). This avoids misrepresentation due to seasonal fluctuations in
inventory and the number and timing of animals sold. From these data, EPA has estimated the
number of confinement operations (referred to here as AFOs) using available data and other
information from the Census as well as other USDA and industry publications (USDA/NASS,
1999a, 1999b and 1999c). These data may differ from that presented in Table 8-1.
Expressed on this basis, USDA estimates that there were more than 106,000 beef feedlots
in 1997 (Table 8-2). EPA also estimates that there were 850 veal operations raising 0.3 million
head and 1,250 stand-alone heifer operations raising 0.9 million head in 1997. Only a portion of
these operations would be subject to the proposed regulations. Under the two-tier structure, EPA
estimates that there are 3,080 beef feedlots with more than 500 head (500 AU of beef cattle).
EPA also estimates that there are about 90 veal operations and 800 heifer operations that may be
subject to the proposed regulations. Under the three-tier structure, EPA estimates that 3,210 beef
feedlots, 140 veal and 980 heifer operations with more than 300 head (300 AU) would meet the
"risk-based" conditions and thus require a permit. EPA expects that few operations that confine
fewer than 500 AU of beef, veal, or heifers, would be designated by the permit authority. For the
purpose of estimating costs, EPA assumes that no beef, veal, or heifer operations would be
designated as CAFOs and subject to the proposed regulations under the three-tier structure.
8-6
-------
Table 8-2. EPA's Estimate of the Number of CAFOs Affected under the Tier Structures
Sector
Fed Cattle
Veal
Heifers
Dairy
Sum
Total
Number
ofAFOs
106,080
850
1,250
116,870
225,050
Number of CAFOs
>1,000 AU
2,080
10
300
1,450
3,830
Two-Tier Structure
(500 AU Threshold)
500-1,000
AU
1,000
80
500
2,310
3,890
<500AU
40
0
0
220
260
Total
CAFOs
3,120
90
800
3,980
7,980
Three-Tier Structure
(Scenario 3)
300-1,000
AU
1,140
130
680
5,030
6,970
<300AU
: o
0
0
50
50
Total
CAFOs
3,210
140
980
6,530
10,860
Source: USEPA, 2000a. See Section 2 for more information. See Table 3-1 for definitions of the
options/scenarios.
"Layers: wet" are operations with liquid manure systems; "Layers: dry" are operations with dry systems. The
number of operations shown eliminates double counting of operations with mixed animal types.
"'As defined for the proposed regulations, one AU is equivalent to one slaughter or feeder cattle, calf or heifer; 0.7
mature dairy cattle.
Under the two-tier structure, EPA assumes that about four beef feedlots located in the Midwest
would be designated annually, or 40 beef feedlots projected over a 10-year period (Table 8-2).
In the dairy sector, USDA reports that there were 116,900 dairy operations witfi a year-
end inventory of 9.1 million milk cows (USDA/NASS, 2000a). See, Table 8-2. Only a portion of
these operations would be subject to the proposed regulations. As shown in Table 8-2, under the
two-tier structure, EPA estimates that there are 3,760 dairy operations that confine more than 350
milk cows (i.e., 500 AU equivalent). Under the three-tier structure, EPA estimates that 6,480
dairy operations with more than 200 head (i.e., 300 AU equivalent) would meet the "risk-based"
conditions described in Section VIE of the preamble and thus require a permit. Under the two-tier
structure, EPA expects that designation of dairies with fewer than 350 milk cows would be
limited to about 22 operations annually, or 220 dairies projected over a 10-year time period.
Under the three-tier structure, EPA expects annual designation of dairies with fewer than 200
milk cows would be limited to about 5 operations, or 50 operations over a 10-year period. EPA
expects that designated facilities will be located in more traditional farming regions.
As shown in Table 8-2, EPA estimates that a total of 3,120 beef operations, 90 veal
operations, 800 heifer operations, and 3,980 dairy operations are estimated either to be defined
(>500 AU) or designated (<500 AU) as CAFOs under the two-tier structure. A total of 1,140
beef operations, 140 veal operations, 980 heifer operations, and 6,530 dairy operations are
estimated to be defined (>300 AU, subject to certain risk-based conditions) or designated (<300
8-7
-------
AU) as CAFOs under the three-tier structure. Total GAFOs under either scenario are not
adjusted for operations with more than a single animal type.
More information on how EPA estimated the number of affected animal confinement
operations is presented in Section 2 of this report, along with additional estimates on the number
of affected beef and dairy operations under other regulatory options considered by EPA.
8.1.2.2 Geographic Distribution
The cattle feeding industry is mostly spread across the Great Plains and Midwestern states,
with the exception of a few states on the West Coast. During the 1970s and 1980s, fed cattle
production shifted eastward from California and Arizona, and westward from Iowa and Illinois to
the larger feedlots of the Great Plains states, where cattle feeding operations remain concentrated
(McBride, 1997). The shift in cattle feeding to the Great Plains was in part attributed to
technological development, including increased output of high-energy feed due to irrigation and
crop improvements in these states (Krause, 1991). Enactment of restrictive statutes on corporate
farming in states such as Iowa, Minnesota, South Dakota, Wisconsin, and Nebraska contributed
to declines in cattle feeding in these states (Krause, 1991). Federal income tax laws during the
mid-1970s may have also contributed to construction of large feedlots hi the Plains states
(Krause, 1991). The Plains states also provide a suitable environment for large-scale cattle
feeding, with a mild, dry climate and a low population density (Krause, 1991). Feedlots in these
states are also located close to suppliers of feeder cattle and slaughter plants.
Nearly 75 percent of all beef production is concentrated among the top five producing
states (USDA/NASS, 1999a). In 1997, Texas and Kansas were the largest beef producing states,
each representing about 20 percent of all beef production (Table 8-3). Nebraska accounted for
another 18 percent of production. Colorado and Iowa were also among the top five producing
states, accounting for about 10 percent and 6 percent of U.S. production, respectively. Other top
ten producing states in 1997 included Oklahoma, California, Idaho, South Dakota, and
Washington (Table 8-3). Combined, the top ten producing states accounted for 85 percent of
U.S. beef production in 1997 (Table 8-3).
Over half of all milk production is concentrated among the top five producing states. In
1997, California and Wisconsin were the largest milk producing states, representing about 18 and
14 percent of all milk production, respectively (Table 8-4). New York accounted for another 7
percent of production. Pennsylvania and Minnesota were also among the top five producing
states, accounting for about 7 percent and 6 percent of U.S. production, respectively. Other top
ten producing states in 1997 included Texas, Michigan, Washington, Idaho, and Ohio (Table 8-4).
Combined, the top ten producing states accounted for almost 70 percent of U.S. milk production
in 1997 (Table 8-4).
8-8
-------
Table 8-3. Geographic Distribution of Cattle and Calf Feedlots by Major Producing State, 1997
Major Producing
State
Texas
Kansas
Nebraska
Colorado
Iowa
Oklahoma
California
Idaho
South Dakota
Washington
Arizona
Minnesota
Illinois
Ohio
Indiana
New Mexico
Wisconsin
Michigan
Missouri
North Dakota
AllOther
Top 5 states
Top 10 states
Total U.S.
Marketed Head
(thousand)
5,800
5,210
4,710
2,555
1,544
907
575
554
527
415
398
310
300
300
250
229
220
210
120
80
1,625
19,819
22,797
26,839
(percent)
22%
19%
18%
10%
6%
3%
2%
2%
2%
2%
1%
1%
1%
1%
1%
1%
1%
1%
0%
0%
6%
74%
85%
100%
Beef Feedlots
(number)
147
195
5,100
174
13,310
26
24
60
3,214
16
9
7,500
6,300
7,500
6,500
10
7,500
4,100
4,000
1,500
38,890
42,310
65,024
106,075
(percent)
6%
0%
5%
0%
13%
0%
0%
0%
3%
• 0%
' 0%
7%
6%
7%
6%
0%
7%
4%
4%
1%
37%
40%
61%
100%
Source: USDA/NASS, 1999a and 1999b. Cattle/calves fattened on grain and concentrates (sales).
8-9
-------
Tahle 8-4. Geopranhical Distribution of Dairy Operations bv Major Producing State, 1997
Major Producing
State
California
Wisconsin
New York
Pennsylvania
Minnesota
Texas
Michigan
Washington
Idaho
Ohio
New Mexico
Iowa
Arizona
Vermont
Florida
Missouri
Illinois
Indiana
Virginia
Kentucky
Top 5 States
Top 10 States
Top 20 States
Total U S
Milk Production
(million Ibs.)
27,582
22,368
11,530
10,662
9,210
5,768
5,410
5,305
5,193
4,415
4,011
3,693
2,664
2,600
2,476
2,362
2,203
2,189
1,858
1,815
81,352
107,443
133,314
156,091
(percent)
18%
14%
7%
7%
6%
4%
3%
3%
3%
3%
3%
2%
2%
2%
2%
2%
1%
1%
1%
1%
52%
69%
85%
100%
Dairy Farms
(number)
2,800
25,000
9,000
11,300
10,500
3,500
4,200
1,400
1,400
6,000
600
4,500
250
2,000
650
4,500
2,400
3,400
1,800
3,600
58,600
75,100
98,800
123,700
(percent)
2%
20%
7%
9%
8%
3%
3%
1%
1%
5%
0%
4% .
0%
2%
1%
4%
2%
3%
1%
3%
47%
61%
80%
100%
Source: USDA/NASS, 1999a (number of dairy farms); USDA/NASS, 1999c (milk production).
8-10
-------
Among these top milk producing states, dairy operations can be divided between those in
traditional and nontraditional dairy production states (Kaiser and Morehart, 1994; McBride,
1997). Traditional dairy production states include Wisconsin, Minnesota, Pennsylvania, and New
York. Dairy operations in these states tend to be smaller and are less industrialized. These states
remain competitive, in part, because local feed supplies are plentiful (McBride, 1997).
Nontraditional dairy production states include California, Washington, Texas, New Mexico, and
Idaho. Dairy operations in these states are larger and use newer technologies and production
methods to take advantage of economies of scale, driving down production costs. These
operations are typically more specialized and produce very little if any crops (Outlaw et al., 1996).
Herd size also varies among regions, with larger sized operations being more common in the more
nontraditional milk producing states. For example, in 1992, herd-size in the Pacific region
averaged 240 cows per operation whereas average herd-size in the Midwest and Northeast states
was less than 60 cows (Outlaw et al., 1996). Milk production in these emergent states is among
the fastest-growing in the nation. By contrast, milk production has been contracting in many of
the more traditional states (USDA/NASS, 1999c).
8.1.2.3 Supply and Demand Conditions
Total U.S. beef production (carcass weight basis) increased 10 percent between 1992 and
1997, reaching 25.5 billion pounds in 1997 (Table 8-5). During the same period, total domestic
demand for beef products increased only 6 percent. Expressed on a per-capita basis, adjusted for
population growth, demand rose slightly from 94.7 pounds per person in 1992 to 95.2 pounds per
person per year in 1997 (Table 8-5). Compared to demand levels in the 1970s when beef
consumption was in the range of 114 pounds per person per year, consumption is down by about
20 pounds per person. This change in consumption has been attributed to the loss of market
share to poultry products in response to changes in consumer preferences and increased health
and nutrition concerns, as well as recent food safety concerns (USDA/ERS, 1998a; USDC,
1999b).
Supply and demand for fluid milk and dairy foods is expressed in terms of total milk
equivalent (total solids basis) (NMPF, 1999). From 1992 to 1997, U.S. dairy product supplies
(milk production plus imports) rose 4 percent, reaching 156.1 billion pounds in 1997 (Table 8-5).
During the same period, total utilization (domestic demand plus exports) also rose 6 percent and
totaled 155.6 million pounds in 1997, resulting in a tightened dairy supply-demand situation.
During the 1990s, demand for all fluid milk and dairy products has averaged around 570 pounds
per person annually, expressed in milk equivalents. This represents a substantial recovery from
the early 1970s when U.S. milk and dairy food production had reached a low of under 520
pounds per person per year—a substantial reduction compared to the 1950s. In part, recent gains
in demand are attributable to improved domestic and export promotion (USDA/ERS, 1998a).
8-11
-------
Table 8-5. Total U.S. Beef and Dairy Supply and Demand, 1992-1997
Year
Production
Imports
Exports
Total
Demand
Per Capita
Demand ^
Beef Products "
1992
1993
1994
1995
1996
1997
%92-97
(million pounds ready-to-cook, carcass weight)
23,086
23,049
24,386
25,222
25,525
25,490
10.4%
2,440
2,401
2,369
2,103
2,073
2,343
-4.0%
1,400
1,337
1,611
1,821
1,877
2,136
52.6% '
24,185
23,944
25,125
25,533,
25,863
25,609
5.9%
(lbs./person)
94.7 .
92.7
96.3
96.8
97.1
95.2
0.5%
Milk and Dairy Products b/
1992
1993
1994
1995
1996
1997
%92-97
(million
pounds)
150,847
150,636
153,602
155,292
154,006
156,091
3.5%
(million pounds milk equivalents on a total solids basis)
4,245
4,341
4,837
4,236
4,466
4,383
3.3%
7,032
6,898
5,806
7,088
4,177
5,244
-25.4%
147,176
147,795
152,170
154,792
155,651
155,606
5.7%
(Ibs. ME/person)
570
566
576
574
574
569
-0.2%
Sources:
"'Putnam and Allshouse, 1997 and 1999. Supplemented with information in USDA/ERS 1998c and 1997f.
Excludes beginning and ending stocks, and shipments to U.S. territories.
^National Milk Producers Federation (NMPF, 1999). Utilization (demand and trade) is expressed in terms of the
milk equivalent (M.E.), total solids basis (tsb), of the estimated milk content in all dairy products (e.g., fluid and
dry milk, cheese, butter, ice cream, whey lactose, and other usable by-products).
''Per capita demand is shown to depict real demand growth, adjusting for growth in U.S. population, which has
grown, on average, at about 1% per year.
8-12
-------
8.1.2.4 Farm Price Trends
Output and price cycles are common in the livestock sectors, given long production
periods and the tendency to adjust future production to current prices (Kohls and Uhl, 1998).
Biological time lags in beef production last about 2.5 years between the time a cow-calf operator
decides to breed an animal and when its beef is ready for retail sale; if the operator wants to
expand productipn and add more breeding herd, lags last about 4.5 years (Becker, 1996).
Because it is impossible to know if current breeding decisions will accurately reflect demand
conditions several years later, mismatches between supply and demand may cause changes in
cattle prices that signal farmers to adjust breeding decisions. For example, to expand future meat
supplies in response to expected profits, producers must hold back animals from market in the
near term to build up the breeding herd, which shorts the market and increases prices in the short
run. Conversely, when low prices signal a reduction in production, the resulting herd sell-off will •
increase supplies and reduce prices in the short run (Kohls and Uhl, 1998). The so-called "cattle
cycle" refers to the approximate 10-year period it takes for beef cattle numbers to expand and
contract in response to changes in prices and profitability (Becker, 1996).
Several factors outside the beef market can influence the severity of the cattle cycle
(Matthews, et al., 1999). Weather affects both the quality of pasture and production of feed
crops. High feed prices induce lower calf prices and reductions in herd size as the cost of raising
the calf to market weight will be higher. A 27 percent drop in com production contributed to the
severity of the 1996 cattle sell off (Becker, 1996). Livestock exports and imports are related to
the domestic price and provide alternative sources and markets (Matthews, et al., 1999).
Commodity programs affect the cattle cycle by motivating changes from pasture to cropland
(Matthews, et al., 1999). While there has been a trend toward decreasing per capita beef
• consumption, retail beef prices are also affected by the cattle cycle. Retailers change their shelf
price more slowly than their costs to avoid disturbing consumers with frequent price changes
(Becker, 1996) but Matthews, et al. (1999) found that this practice did not result in abnormally
large price spreads during the 1996-97 low point in the cattle cycle.
Table 8-6 presents actual quarterly and annual fed cattle prices received by U.S. cattle
producers from 1992 through 1998 (NCBA, 2000). Beef prices declined steadily throughout the
1990s from $75.27 per hundred weight (cwt) in 1992 to $66.09 per cwt in 1997, followed by a
period low of $61.73 per cwt in 1998. The worst losses in the recent past were experienced in
1996 and 1997, which resulted in an overall loss in equity for the feedlot industry (Stott, 2000a).
During those years, escalating feed grain prices, dust-bowl conditions, and unfavorable returns
caused a shift in the industry from expansion toward liquidation, pushing the supply of cattle to its
highest point since the mid-1970s—the high point of the 10-year cattle cycle (Becker, 1996). The
subsequent drop in price from the abundant supply of beef caused farmers to further reduce herd
size, resulting in additional liquidation (Uvacek, no date). High prices are expected to peak in
2002 (Stott, 2000a).
8-13
-------
Table 8-6. Actual Average Quarterly and Annual Prices Received by Farmers, Total U.S., 1992-1997
Year
Average Ql
Average Q2
Average Q3
Average Q4
Average Annual
(monthly prices received $/cwt)
Beef
1992
1993
1994
1995
1996
1997
1992
1993
1994
1995
1996
1997
75.75
80.55
73.41
72.39
63.00
66.18
75.64
79.53
68.99
65.20
60.25
66.31
73.78
73.78
66.63
62.62
67.28
65.12
75.89
. 71.70
68.11
66.07
69.48
66.75
75.27
76.40
66.29
66.57
65.00
66.09
Milk
13.0
12.4
13.5
12.6
14.0
13.5
12.9
12.9
13.0
12.3
14.4
12.8
13.5
12.7
12.5
12.5
16.0
12.7
13.2
13.4
13.0
13.8
15.4
14.5
13.2
12.9
13.0
12.8
15.0
13.4
Sources: NCBA, 2000 (Choice Fed Steer) and USDA/NASS, 1998a (Milk sold to plants, eligible tor fluid market.
Includes surplus fluid grade milk diverted to manufacturing).
Fed cattle prices tend to vary seasonally according to production cycles throughout the
year and are often subject to periods of high variability, particularly for feeder calves (Bliss and
Ward, 1999). Calf prices typically average higher than fed cattle prices (Dyuvetter et al., 1998).
Because returns in the feedlot sector are more dependent on buy-sell margins than absolute prices
levels, there is very little correlation between returns and fed-cattle prices (Dyuvetter et al., 1998).
Fed cattle prices are highly variable: between 1980 and 1997, quarterly average fed cattle prices
averaged about $69 per hundredweight (cwt) and ranged from about $54 per cwt to $81 per cwt.
Milk prices also fluctuate throughout the year, reflecting seasonal variation due to annual
production cycles (Table 8-6). For example, prices tend to drop during the summer months
following the spring flush when demand tends to lessen. Prices often recover during the fall and
winter months when milk supplies tighten due to increased demand from school use and holidays.
Moreover, dairy pricing is influenced by storage considerations, given the perishability and flow
characteristics of milk. Inventories of fluid milk and some dairy products cannot be held for long
periods of time. Finally, farm level prices in general are often subject to periods of high instability
8-14
-------
due to changing market conditions and/or sharp shifts in supply in response to changing farm
prices and/or input costs, among other factors.
Table 8-6 presents actual quarterly and annual milk prices received by U.S. dairy farmers
from 1992 through 1997. Averaged over the year, milk prices changed little over the period,
averaging more than $ 13 per cwt. Quarterly fluctuations, however, show wide seasonal
variability as well as cyclical market swings. For example, milk prices jumped sharply in 1996 in
response to high feed grain costs and tightened supply-demand conditions. Moreover, measured
in real terms (i.e., corrected for price inflation), farm level prices have been decreasing despite
strong domestic demand and export growth for dairy foods.
Federal price supports for milk and also marketing contracts with cooperatives have
protected producers from significant fluctuations in milk prices (McBride, 1997). Dairy pricing
has also influenced by the negotiating power of milk cooperatives and other market factors.
Cooperatives market the majority of milk in the U.S. and give farmers the ability to secure a milk
price above federal market order minimum prices through collective bargaining with
processors/manufacturers and/or retail operations (Wolf and Hamm, 1998).
Historically, the price of farm milk and some dairy products has been supported by an
array of government programs and pricing policies. The government's role in dairy pricing has
included Federal and state milk marketing orders and support programs, as well as purchases
through various government programs. The government's role in dairy policies is decreasing,
following deregulation under the 1996 Federal Agricultural Improvement and Reform (FAIR) Act
(also known as the "Farm Bill").
8.1.3 Financial Data Characteristics of Beef and Dairy Operations
8.1.3.1 Overview of Financial Characteristics
USDA reports that sales of all cattle and calves from commercial operations with $50,000
or more annually totaled $32.4 billion in 1997 (USDA/NASS, 1999a).2 (See Table 8-7.) This
includes revenues generated at both feedlot and all other cattle and calf operations. Reported
gross revenue from cattle and calves sold from facilities that "fatten cattle on grains and
concentrates" totaled $20.4 billion in 1997 (USDA/NASS, 1999a). Among all beef farms,
including feedlot operations, nearly 60 percent of total farm revenues are from the sale of cattle.
Secondary livestock revenues, including sales by farms that generate a portion of their total
revenue from the sale of milk and/or beef, dairy, and poultry products accounted for 37 percent of
total farm revenues. Crop sales from these farms accounted for 5 percent in 1997 (USDA/NASS,
1999a). Supplemental information for beef feedlots only on the share of total revenue generated
from beef versus other secondary income sources is not available, although indications are that
feedlot operations tend to specialize in finishing cattle only and generate a smaller share of total
revenue from other sources (Krause 1991; Kohls and Uhl, 1998).
2USDA defines commercial farms as those with gross sales of $50,000 or more during a given year.
8-15
-------
Table 8-7 Farm Revenue at Beef Feedlots and Dairy Farms, by Revenue Category and Economic Class
Revenue Category/
Economic Class
All Beef
Operations
Revenues
($1,000)
Dairy
Farms
Revenues
($1,000)
Sales by Revenue Category (Reported and Percentage Share)
Primary Livestock
Secondary Livestock
Crop Sales
All Farms
Primary Livestock
Secondary Livestock
Crop Sales
• All Farms
9,709
58,677
15,296
83,682
12%
70%
18%
100%
$18,773,077
$11,869,096
$1,716,422
$32,358,595
58%
37%
5%
100%
68,032
3,314
2,009
73,355
93%
4%
3%
100%
$17,624,762
$464,832
$254,645
$18,344,239
96%
3%
1%
100%
Sales by Economic Class (Reported and Percentage Share)
>S1 million in revenue
S0.5-S1.0 million
S0.25-S0.50 million
S0.10-S0.25 million
S0.05-S0.10 million
All Farms
>$1 million in revenue
S0.5-S1.00 million
S0.25-S0.50 million
S0.10-S0.25 million
S0.05-S0.10 million
All Farms
7,074
8,919
16,345
29,505
21,839
83,682
8%
11%
20%
35%
26%
100%
$21,425,530
$3,039,924
$3,099,380
$3,329,680
$1,464,079
$32,358,593
66%
9%
10%
10%
5%
100%
3,359
4,761
12,890
35,845
16,500
73,355
5%
7%
18%
48%
22%
100%
$6,766,331
$2,467,884
$3,368,697
$4,600,585
$1,140,773
$18,344,270
37%
13%
18%
25%
6%
99%
Source: USDA/NASS, 1999a. Based on data for commercial farms with more than $50,000 in annual revenues.
Excludes non-commercial farms with revenues below $50,000.
Primary livestock: cattle (beef feedlots, NAICS 112112) and dairy (NAICS 11212), respectively.
Secondary livestock: beef (beef farming, NAICS 112111, and beef feedlots, NAICS 112112), dairy (NAICS
11212), miscellaneous categories (NAICS 1122, NAICS 1124, NAICS 1125), along with hogs (NAICS 1122) and
poultry (NAICS 1123), respectively.
Crop sales: oilseed/grains (NAICS 1111), vegetables (NAICS 1112), fruits/nuts (NAICS 1113), greenhouse
(NAICS 1114) and other crops (NAICS 1119).
8-16
-------
Across all beef operations, about 7,000 operations (8 percent of all beef farms) generate
more than $1 million in revenue annually (USDA/NASS, 1999a). (See Table 8-7.) This revenue
. cut-off is the nearest approximation to the Small Business Administration's small business size
standard, which defines a "small business" among beef feedlots as an operation that generates less
than $1.5 million annually in total entity revenue (SBA, 1998; USGPO, 2000). (Section 9
provides additional information on EPA's small business analysis.)
At dairy operations, USDA reports that commercial farms in the U.S. generated a total of
$18.3 billion in annual revenue in 1997 (USDA/NASS, 1999a). Virtually all (96 percent) dairy
farm revenues were from the sales of milk. Approximately $0.7 billion of all dairy farm revenue
was generated from the sales of other livestock or crop production (Table 8-7). As shown in
Table 8-7, approximately 11 percent of all commercial dairy farms generate more than $0.5
million in revenue annually (USDA/NASS, 1999a). The remaining farms generate revenues
below $0.5 million. The $0.5 million threshold in annual farm revenues corresponds with SBA's
definition of a "small business" in the dairy sector (SBA, 1998; USGPO, 2000).
8.1.3.2 Income Statement and Balance Sheet Information
Returns to finishing cattle in commercial feedlots are dependent on many factors and vary
considerably from year to year. Returns are related to productivity (average daily gain, feed
efficiency, etc.) and feed costs (Duyvetter et al., 1998). The estimated quarterly profit for feeding
steers in a commercial feedlot has been about $15 per head, ranging from a low of-$86 per head
to a high of $141 per head from 1980 through 1997 (Duyvetter et al., 1998). Returns of about
$15 per head represents a profit margin of about 2 percent (Idaho Cattle Association, 1999).
Other industry information indicate that over the past 20 years, the average loss per head of fed
cattle was $1.56 (NCBA, 1999). Bliss and Ward (1999) also report negative average returns
based on budgeted cattle feeding net margins in the Great Plains states from 1978 through 1987.3
However, although budgeted profits are often negative, Bliss and Ward note that actual profits at
these operations are positive because the feedlot often price cattle and grain using futures market
contracts and options to manage price risk and manage the seasonal variations in profits (Bliss and
Ward, 1999).
Limited financial data are available on cattle feedlot operations. Tables 8-8 and 8-9
present average financial data that characterize conditions at beef feedlot operations. These
sources of publicly available data include RMA and Dun and Bradstreet. Table 8-10 provides
additional financial information based on a survey conducted by National Cattlemen's Beef
Association (NCBA).
3These estimates are prepared by USDA and represent the profits from selling a slaughter steer that
achieves a 500-pound weight gain in 180 days.
8-17
-------
Table 8-8. Composite Income Statement and Balance Sheet in SIC 0211, Feedlots-Beef Cattle, 1997
Size
(in annual revenues)
?et sales
Operating expenses
Operating profit
All other expenses (net)
Profit before taxes
?arm Assets
Current assets
Cash and equivalents
Trade receivables (net)
Inventory
All other current
Non-current assets
Fixed assets
Intangibles
Other non-current assets
Farm Liabilities
Current liabilities
Notes payable in 1 year
Curr. Maturities-L/T/D
Trade payables •
Income taxes payable
Other current liabilities
Non-current liabilities
Long term debt
Deferred taxes
Other non-current liab.
0-S1 mill.
$550,625
$519,239
$31,386
$26,981
$4,405
$1,572,000
$817,440
$69,168
$117,900
$529,764
$99,036
$756,132
$589,500
$0
$166,632
$1,572,000
$699,540
$521,904
$12,576
$26,724
$9,432
$128,904
$103,752
$84,888
' $0
$18,864
$1-3 mill.
$2,015,857
$1,872,731
$143,126
$20,159
$122,967
$1,556,857
$832,919
$84,070
$233,529
$434,363
$80,957
$723,939
$540,229
$7,784
$175,925
$1,556,857
$784,656
$558,912
$42,035
$60,717
$14,012
$110,537
$158,799
$105,866
$23,353
$29,580
$3-5 mill.
$3,832,778
NA
NA
NA
NA
$2,553,333
NA
NA
NA
NA
NA
NA
NA
NA
NA
$2,553,333
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
$5-10 mill.
$7,323,765
$7,147,994
$175,770
$65,914
$109,856
$3,385,353
$2,369,747
$108,331
$809,099
$1,239,039
$216,663
$1,015,606
$886,962
$3,385
$125,258
$3,385,353
$1,845,017
$1,198,415
$37,239
$216,663
$23,697
$369,003
$524,730
$402,857
$33,854
$88,019
$10-25 mill.
$15,884,320
$15,407,790
$476,530
$79,422
$397,108
$7,544,480
$5,794,161
$475,302
$1,576,796
$3,606,261
$143,345
$1,750,319
$1,486,263
$30,178
$233,879
$7,544,480
$3,870,318
$2,814,091
$52,811
$520,569
$15,089
$460,213
$724,270
$467,758
$188,612
$67,900
$25 mill.+
$198,496,267
$193,732,356
$4,763,910
($396,993)
.$5,160,903
$50,917,433
$36,405,965
$2,800,459
$12,983,945
$17,617,432
$3,004,129
$14,511,468
$9,470,643
$1,883,945
$3,156,881
$50,917,433
$23,523,854
$14,155,046
$407,339
$6,008,257
$611,009
$2,393,119
$7,128,441
$4,480,734
$661,927
$1,985,780
NA =- Not available.
Source: RMA, 1998.
8-18
-------
Table 8-9 Financial Information for Establishments in SIC 0211—Beef Cattle Feedlots, 1993-1997
Item
Net Sales
Gross Profit
Net Profit After Tax
Working Capital
Total Assets
Current Assets
Cash
Accounts Receivable
Notes Receivable
Inventory
Other Current
Non-current Assets
Fixed Assets
Other Non-current
Total Liabilities
Current Liabilities
Accounts Payable
Bank Loans
Notes Payable
Other Current
Non-current Liabilities
Other Long Term
Deferred Credits
1993
$7,459,116
$1,737,974
$164,101
$539,698
$3,394,328
$2,104,483
$169,716
$780,695
$105,224
$790,878
$257,969
$1,289,845
$824,822
$465,023
$3,394,328
$1,564,785
$339,433
$20,366
$454,840
$750,146
$1,829,542
$498,966
$3,394
1994
$7,081,199
$1,642,838
. $198,274
$614,119
$3,198,540
,$2,015,080
$102,353
$844,415
$102,353
$758,054
$207,905
$1,183,459
$713,274
$470,185
$3,198,540
$1,400,961
$374,229
$3,199
$351,839
$671,693
$1,797,580
$438,200
$3,199
1995
$7,162,299
$1,618,680
$322,303
$500,041
$2,809,217
$1,702,386
$176,981 .
$640,501
$58,994
$654,548
.$171,362
$1,106,831
$792,199
$314,632
$2,809,217
$1,202,345
$266,876
$14,046
$286,540
$634,883
$1,606,872
$325,869
$5,618
1996
$11,645,125
$2,282,445
$430,870
$752,076
$4,043,420
$2,539,268
$246,649
$938,073
$105,129
$901,683
$347,734
$1,504,153
$1,035,116
$469,037
$4,043,420
$1,787,192
$396,255
$24,261
$566,079
$800,597
$2,256,228
$485,210
$16,174
1997
$12,169,927
$2,178,417
$352,928
$790,504
$4,074,761
$2,632,296
$240,411
$973,868
$130,392
$1,034,989
$252,635
$1,442,465
$1,124,634
$317,831
$4,074,761
$1,841,792
$456,373
$20,374
$631,588
$733,457
$2,232,969
$484,897
$12,224
Source: Dun & Bradstreet, 1993-1997.
8-19
-------
Table 8-10. Financial Characteristics from NCBA Financial Survey (1994-1998)
Item
1994
1995
1996
1997
1998
Average
Gross Cash Income (PHAO)
Total Cash Expenses
Net Cash Income
Depreciation
Current Assets
Current Liabilities
Total Assets
Total Liabilities
Return on Equity
Return on Assets
Current Ratio
Debt to Equity Ratio
$638.81
$557.80
$16.47
$28.22
$275.18
$208.80
$348.36
$256.93
18.0%
4.7%
1.32
2.81
Gross Cash Income (PHAO)
Total Cash Expenses
Net Cash Income
Return on Equity
Return on Assets
Current Ratio
Debt to Equity Ratio
$965.00
$822.00
$61.23
• 58.1%
17.7%
4.81
7.97
$627.93
$566.91
$8.49
$28.03
$261.30
$186.59
$329.98
$225.79
8.2%
2.6%
1.40
2.17
$700.39
$620.39
$19.94
$27.18
$312.87
$235.40
$380.52
$272.62
18.5%
5.2%
1.33
2.53
$666.54
$556.85
$34.39
$29.12
$418.32
$323.94
$479.62
$359.59
28.6%
7.2%
1.29
3.00
$601.32
$562.00
($28.54)
$28.31
$361.03
$278.57
$434.49
$329.99
-27.3%
-6.6%
1.30
3.16
High
$972.00
$793.00
$69.38
114.8%
67.2%
3.29
6.38
$1,085.00
$903.00
$62.48
61.3%
29.5%
3.11
5.06
$1,239.00
$851.00
$136.70
119.2%
51.5%
4.73
206.00"
$1,074.00
$831.00
$56.73
71.8%
29.4%
5.91
6.56
Low
Gross Cash Income (PHAO)
Total Cash Expenses
Net Cash Income
Return on Equity
Return on Assets
Current Ratio
Debt to Equity Ratio
$277.88
$154.00
($24.00)
-15.0%
-7.3%
0.31
0.96
$271.07
$169.00
($87.07)
-81.5%
-15.9%
-0.31
0.71
$207.86
$157.00
($49.79)
-52.6%
-13.1%
-0.39
0.86
$100.00
$102.00
($38.00)
-200.0%
-10.4%
0.51
0.41
$211.56
$186.14
($125.37)
-171.9%
-26.1%
-1.25
• -4.28
Source: NCBA, 1999. Per head average occupancy (PHAO).
"'sic: NCBA table shows 206.00. Using stated total assets and liabilities debt to equity ratio is calculated as 7.25.
8-20
-------
Table 8-8 presents average income statement ancTbalance sheet data for companies in SIC
0211, Feedlots—Beef Cattle, for 1997, as reported by RMA (RMA, various years). RMA
Annual Statement Studies are gathered from financial statements from RMA member bank
customers. More man 150,000 financial statements from primarily small-and medium-sized
companies are gathered annually. The RMA data represent companies' (not individual facilities')
financial statements and also reflect financial statements for companies having less than $250,000
in total assets. Where available, these financial data are presented in six separate size
classifications based on annual company revenues. These RMA data show relative financial health
for beef feedlots for this period, with positive profits before taxes in each size class. These data
indicate that while operations with less than $1 million in sales barely turn a profit and return less
than one percent on assets, operations in the $1 to $3 million size class have a higher net profit
margin before taxes, 6 percent, than any other size operation. Only the largest size class has a
higher return on assets. For 1997, the $5 to $10 million size class has almost the same level of
operating profit, $176,000, as the $1 to $3 million size class, $143,000 even though it has more
than three times the sales for that year.
Table 8-9 presents financial data for establishments in SIC 0211—Feedlots—beef cattle
from 1993-1997 from Dun & Bradstreet (Dun & Bradstreet, various years). The data from Dun
& Bradstreet's "Industry Norms and Key Business Ratios" provide establishment-based financial
data from the Dun & Bradstreet database and provide measures of financial profitability,
efficiency, and solvency for the upper, lower, and median quartiles for each financial ratio within
an SIC code group. These establishments, at the national-level of aggregation, all show positive
net profits before and after tax. The financial data are not gathered by size breakdowns for the
surveyed establishments. Total assets for 1993 through 1997 exceed $2.8 million. These data
indicate that even though average net sales vary from $7.1 million to $12.2 million in the 5-year
period, the gross profit margin varies only 5 percent, from 18 to 23 percent. This shows that
much of the volatility in cattle prices is passed through from changes in feed and other costs of
production. Sales were flat from 1993 to 1995; business turned around in 1995 and 1996 when
net profit margins exceeded four percent and returns on assets were over ten percent (Table 8-9).
By 1997 sales were up 72 percent from their 1994 low.
Table 8-10 presents data from a survey prepared by NCBA (NCBA, 1999). NCBA
provided aggregated summary information on financial conditions at cattle feeding operations
based on responses to a survey questionnaire of its membership. Like the Dun and Bradstreet
data, the NCBA data show improving income from 1994 to 1997. The NCBA data also includes
data for 1998 when cash income per head fell sharply and net income became negative for the
average operation. High income operations continued to be profitable although their net cash
income fell almost 60 percent from 1997 to 1998. Operating margins were less than 2 percent
over the period (Table 8-10).
The least profitable respondents to the NCBA survey receive only about one-fifth as much
per head as the most profitable. The lowest group carries losses every year but they are not as
8-21
-------
highly leveraged as the top producers, i.e., they have lower debt-to-equity ratios, and thus may
remain in business. Additional information on the NCBA data is provided in Section 8.1.3.3.
The reported range Of values in the NCBA data likely also reflect differences across
different types of operations (Stott, 2000a). There is a major distinction in what drives
profitability at custom feedlots and company-owned feedlots. Occupancy rates at custom feedlots
are the critical factor for profitability at these operations, whereas at company-owned feedlots,
profits on cattle govern the overall profitability of the operation (Idaho Cattle Association, 1999).
Presumably, the packer can optimize occupancy rates, whereas custom feedlots must compete for
customers. Limited additional information on the distinctions between these types of operations is
available.
Limited information is also available on veal and heifer operations (Duyvetter et al., 1998).
Foster (2000b) reports that the basic costs of raising a heifer from weaning to 23 months are the
same whether the animal is part of a dairy herd or in a specialized replacement heifer facility. Use
of the specialized facility has advantages for both the dairy producer and the custom grower. The
grower is able to use available crop land and facilities without committing to a complete milk herd
while the dairy producer can focus his management skills on milk production and avoid tying up
facilities with unproductive animals (Endsley, et al., 1996). If costs are estimated accurately, the
custom grower can negotiate a fair contract payment rate.
Table 8-11 presents average income statement and balance sheet data for commercial
dairy farms from 1993 through 1997. According to USD A, the average dairy farm demonstrated
a favorable financial position with positive net income and a debt-to-asset ratio that ranged from
0.20 to 0.22 from 1993 to 1997 (USDA/ERS, 1998d). These debt-to-asset ratios indicate
that—at the national level—average dairy farms are not in a vulnerable financial position because
they have a low potential for cash flow problems and a low relative risk of insolvency. However,
these national average data do not reflect differences by region, size, or type of operation, and
may mask financial differences among dairy farms. Additional information from USDA show that
between 1991 and 1994 about 30 percent of all beef and dairy farms experienced negative income
(USDA/ERS, 1997b).
Regional differences in milk production costs and prices underlie financial differences
among milk producing regions. For example, the upper midwest has lower cash expenses, in part
because farmers there grow their own feed and have plentiful supplies of roughage in the region
(Outlaw et al, 1996). Data from USDA's ERS cost of production data series also reflect the
differences among the major milk production regions, given differences in production practices
and technologies used, degree of specialization, and use of hired labor, among other factors
(USDA/ERS, 1998d). Farm milk prices also play an important role in influencing financial
differences among regions (Outlaw et al, 1996).
8-22
-------
Table 8-11. Income Statement and Balance Sheet for Dairy Farms (Sales >S50,000), 1993-1997
Item
1993
1994
1995
1996
1997
Dollars per Farm
ncome Statement
Gross cash income
Livestock sales
Crop sales (incl. net CCC loans)
Government payments
Other farm-related income2
Less: cash expenses
Variable cash expenses .
Fixed cash expenses
Equals: net cash farm income
Less: Depreciation
Labor, non-cash benefits
Plus: Value of inventory change
Non-money income*
Equals: net farm income
$211,563
$194,214
$6,635
$4,737
$5,976
$162,288
$137,510
$24,778
$49,275
$19,098
$867
$4,757
$5,169
$39,236
$243,568
$223,973
$7,087
$3,411
$9,097
$193,732
$165,350
$28,383
$49,836
$24,358
$1,114
$7,340
$5,404
$37,108
$263,852
$245,286
$8,738
$2,722
$7,107
$205,263
$174,734
$30,530
$58,589
$24,091
$1,440
$1,948
$5,417
$40,424
$277,181
$259,412
$7,862
$2,720
$7,188
$214,980
$185,020
$29,960
$62,202
$26,066
$1,013
$3,374
$6,216
$44,712
$294,723
$273,863
$5,820
$2,645
$12,395
$237,271
$206,238
$31,033
$57,452
$24,139
$1,024
$14,809
$4,725
$51,823
Balance Sheet
Farm assets
Gurrent assets
Non-current assets
Farm liabilities
Current liabilities
Non-current liabilities
Farm equity
Debt/asset ratio
$678,433
$91,749
$586,683
$133,125
$36,500
$96,624
$545,308
0.20
$729,089
$110,975
$618,133
$152,914
$42,720
$110,194
$576,175
0.21
$725,261
$109,833
$615,428
$158,466
$55,374
$103,092
$566,794
0.22
$777,841
$107,798
$670,043
$161,462
$51,650
$109,812
$616,379
0.21
786,155
111,972
674,183
172,625
40,811
131,815
613,530
0.22
Source: USDA/ERS, 1998d.
a Includes income from machine-hire, custom work, livestock grazing, land rental, contract production fees,
outdoor recreation, and any other farm-related source.
b Defined as home consumption and imputed rental value of farm dwellings owned by the farm operation.
8-23
-------
8.1.3.3 Baseline Conditions for Cattle and Dairy Operations
Tables 8-12 through 8-15 provide a summary of the financial baseline conditions assumed
for this analysis. These data are aggregated from the 1997 ARMS data set and are obtained by
USDA's ERS, as described in Section 4. These data are separated by select facility size and
production region groupings for the beef and dairy sectors (see Table 4-4). Additional
information on how these data differ by region are provided in the record (USDA/ERS, 1999a,
see DCN 70063). EPA uses these average data to assess regulatory costs for fed cattle, veal,
heifer, and dairy operations.
The average beef and dairy operation demonstrated a favorable financial position in 1997
with positive net income and debt-to-asset ratios under 40 percent. In the beef sector, estimated
debt-to-asset ratios ranged from 9 percent (small operations) to 13 percent (medium operations),
across select operation sizes; in the dairy sector, estimated debt-to-asset ratios ranged from 17
percent to 26 percent (USDA/ERS, 1999a). See Tables 8-12 and 8-14. These debt-to-asset
ratios indicate that—on average—beef and dairy operations are not in a vulnerable financial
position and have a low potential for cash flow problems and a low relative risk of insolvency.
Based on these data, EPA assumes that baseline (prior to regulation) net cash flow for all model
types for the beef and dairy sectors is positive, and baseline debt-to-asset ratios for all model
types are 40 percent or less. All cattle and dairy operations in this analysis, therefore, are
considered financially healthy, on average, in the regulatory baseline. Tables 8-13 and 8-15
present average income and balance sheet data for commercial beef and dairy operations in 1997,
by size of operation.
Cattle Operations
• Data shown in Table 8-12 are distributed by broad facility size groups. As shown, more
than 96 percent of operations have fewer than 200 head. These operations account for about 67
percent of all beef cattle raised annually (Table 8-12). There are fewer larger-sized operations
with more than 800 head (less than 1 percent of all farms), and these operations raise only 6.8
percent of all beef cattle annually (Table 8-12). Smaller beef operations with less than 200 head
are more diversified than larger ones, with about 50 percent of all farm revenue from crops. This
compares to beef operations with more than 800 head, where livestock comprises the bulk of all
annual farm sales and only about 13 percent of farm revenues are from crops (Table 8-12).
Table 8-12 also shows the percentage of beef cattle owned by farmers compared to those
not owned by farmers. EPA uses this information on animal ownership as an indication of the
extent of production contract use in these sectors (see Section 2.3). Across all beef operations in
1997, only about 5 percent of animals were not owned by farmers (USDA/ERS, 1999a).
Percentages do not vary noticeably across farm sizes (Table 8-12). This is consistent with other
market information (USD A/GBPS A, 1999c).
\
Table 8-13 presents average income statement and balance sheet information for beef
operations in 1997 and reflects the baseline financial conditions assumed by EPA for this analysis.
The financial data used for this analysis do not distinguish between fed cattle operations and cow-
calf operations. These data also do not distinguish between fed cattle, heifer, and veal operations.
8-24
-------
Table 8-12. Typical Financial Characteristics of Fed Beef Operations, By Size of Operation
Item
dumber of farms
Percent of farms
Percent of value of production
Livestock value of production
Crop value of production
All Farms
940,095
100.0%
100.0%
56.0%
44.0%
Less than 200
Beef Cows
907,368
96.5%
77.7%
52.7%
47.3%
200 to 800
Beef Cows
30,816
3.3%
16.4%
60.8%
39.2%
More than 800
Beef Cows
1,912
0.2%
5.8%
86.9%
13.1%b
Number of beef cows
Distribution of beef cows
Percent of beef cows owned
Percent of beef cows not owned
Number of sample farms with beef cows
39,623,169
100.0%
95.3%
4.7%
4,132
26,422,578
66.7%
96.0%
4.0%
3,548
10,511,161
26.5%
93.7%
6.3%a
495
2,689,430
6.8%
94.1%
5.9%a
89
Debt-to- Asset Ratios
All Regions
Midwest
Central
0.0956
0.1447
0.0893
0.0916
0.1448
0.0899
0.1345
0.1700
0.0797
0.0931
d
d
EPA Derived Gross Cash Income Per Animal
All Regions
Midwest
Central
$1,060.05
$1,011.78
$718.32
$1,073.81
$1,096.67
$709.65
$534.86
$501.99
$461.09
$861.83
$854.00
d
EPA Derived Net Cash Income Per Animal
All Regions
Midwest
Central
$143.68
$190.39
$35.42
$127.31
$178.50
$22.80
$79.37
$80.71
$43.70
$256.28
$321,83
d
Source: USEPA and USDA/ERS, 1999a.
a = The relative standard error of the estimate exceeds 25 percent, but no more than
b = The relative standard error of the estimate exceeds 50 percent, but no more than
c = The relative standard error of the estimate exceeds 75 percent.
d = Data insufficient for disclosure.
50 percent.
75 percent.
8-25
-------
Table 8-13. Income Statement and Balance Sheet for Farms with Beef Cows, by Size of Operation, 1997
Item
All Farms
Less than 200
200 to 800
More than 800
Income Statement
Gross cash income
Livestock income
Crop sales (incl. net CCC loans)
Government payments
Other farm-related income "
Total variable expenses
Livestock purchases
Feed
Other variable expenses v
Total fixed expenses
Equals: Net cash farm income
Less: Depreciation and Other B
Plus: Value of inventory change
Plus: Nonmoney income 4/
Equals: Net farm income
Farm assets
Current assets
Non-current assets
Land, buildings, and equipment 5/
Farm liabilities
Current liabilities
Noncurrent liabilities
Farm equity
44,679
21,202
15,171
1,958
6,348
28.818
2,364
4,789
20,325
9,805
6,056
5,380
5,034
4,419
10,130
35,030
15,107
13,162
1,658
5,103
23,006
1,491
3,743
16,540
7,871
4,153
4,573
4,132
4,358
8,069
Balance Sheet
420,235
42,401
377,834
357,202
40,195
13,627
26,568
380,041
358,236
34,751
323,485
308,712
32,131
10,083
22,049
326,105
242,478
148,993
62,080a
d
d
152,124
18,841
23,415
106,125a
54,372
35,982
23,627
44,296
6,206
62,856
1,735,700
211,736
1,523,964
1,370,316
226,235
86,646
139,590
1,509,465
l,436,244a
854,011
212,917b
.d
d
799,682a
151,013b
200,870b
434,771
209,476a
427,086a
94,036a
-199,546c
4,606a
138,110c
8,642,843a
943,771
7,699,072a
7,041,669a
868,438a
518,852a
349,586a
7,774,405a
Source: Based on USDA/ERS, 1999a. Copies of these data are in the rulemaking record (DCN 70063).
"Machine-hire, custom work, livestock grazing, land rental, contract fees, and other farm-related sources.
^ncl. livestock leasing, custom feed processing, bedding, grazing, supply, transportation, storage, general business
expenses, and registration fees. Footnote (a) refers to an RSE on "other livestock-related" portion of the total.
''Includes labor, non-cash benefits. Footnote (a) refers to an RSE on "non-cash benefits" portion of the total.
"'The value of home consumption plus an imputed rental value of farm dwellings.
s'The value of the operator's dwelling and associated liabilities are included if the dwelling was located on the
farm, a = Relative standard error (RSE) of the estimate exceeds 25 percent, but no more than 50 percent.
b = RSE of estimate >50%, but <75%. c = RSE of estimate >75%. d = Data insufficient for disclosure.
8-26
-------
Data shown in Table 8-13 are differentiated by selected size categories and reveal some
differences among operations by size. The income statement data (and data in Table 8-12) point
to increasing specialization as the size of an operation increases. There are no major differences
in the proportion of animals not owned at the largest operations compared with that at the
smallest operations. However, the smallest operations have proportionately lower expenditures
on livestock-related expenses than larger operations. Expenditures on livestock and feed average
about one-quarter of total variable expenses at an average beef operation with less than 200 head;
operations with more than 800 head are associated with expenditures on livestock and feed
averaging over 40 percent of total variable expenses (Table 8-13). These differences may be
explained by differences hi the degree of specialization and feeding strategies, and other factors.
Operating margins (measured in terms of average net cash farm income as a percentage of
average gross cash income) among differently sized operations differ substantially: operations
with less than 200 head show an operating margin of 12 percent, as compared to 30 percent at
operations with more than 800 head (USDA/ERS, 1999a). However, the smallest operations
show a relatively high return on assets (measured as average net farm income to average farm
assets) compared with the largest operations: operations with less than 200 head show an average
return on assets of 2.5 percent, as compared to 3.6 percent and 1.6 percent at operations with
between 200 and 800 head and operations with more than 800 head, respectively (USDA/ERS,
1999a). See Table 8-13. The 1997 ARMS data include, among an average farm's assets, the
value of the owner's home when it is located on the farm. Since smaller operations may be more
likely to have the owner's dwelling located on the farm than larger operations, if dwelling values
were excluded, the returns on the "business" assets might be higher for these smallest operations.
Section 4 of this report presents key financial data used for this analysis, shown in Table 8-
15, that are calculated onto a per-animal basis. For the cattle sectors, total gross farm revenues
are estimated to range from $500 to $860 per head (includes revenue from other farm-related
sources). Net cash income ranges from $80 to $320 per head among CAFO models, depending
on facility size and region (see Tables 4-5 and 4-6).
For the purpose of this analysis and because of lack of other statistically validated survey
data, EPA uses the ARMS data for cow and calf operations to depict conditions at regulated
cattle feeding operations. For the beef sectors, the data used by EPA reflect income and balance
sheet information for farms with beef cows (shown in Tables 8-12 and 8-13). The National
Cattlemen's Beef Association (NCBA) has expressed concern that the ARMS data are more
reflective of cow-calf operations and represent few feedlots and, therefore, might not be
representative of cattle feeding operations. Correspondence between EPA and NCBA on this
issue is documented hi the rulemaking record (Stott, 2000a; USEPA, 1999k and 2000m; NCBA,
1999). NCBA point out that the ARMS data reflect conditions across all cow and calf
operations, both grazing and feedlots, and may not fully represent cattle feeding operations.
Feedlot operations may be characterized by different financial conditions than grazing operations.
8-27
-------
On the one hand, feedlots are more intensive finishing operations and might receive higher
revenues per animal. On the other hand, feedlot operations might have lower non-livestock
related revenues due to the highly specialize nature of these operations. Operating expenses and
other financial indicators might also differ between the different types of operations in this sector.
In addition, Stott (2000b) indicate that a low debt-to-asset ratio for cow-calf operations does not
accurately represent conditions at cattle feedlots operate'that tend to have higher debt due to
cattle financing. Debt-to-asset ratios are higher at cattle feedlots than what is reported by USDA
for cow-calf operations (Stott, 2000b).
As discussed in Section 8.1.3.2, financial data on cattle feeding operations is limited. To
provide EPA with information on this sector, NCBA conducted a survey of its membership and
provided aggregated summary information on financial conditions at cattle feeding operations
(NCBA, 1999). As reported by NCBA, these data "are not intended as a statistical and
conclusive financial analysis of the feedlot industry, but only a summary" (NCBA, 1999). The
data provided to EPA include average, high, and low estimates of gross revenue, net cash income,
assets, and liabilities, expressed on a per-head average occupancy (PHOA) basis and are not
disaggregated by region or size of operation. NCBA also provided certain financial ratios for
certain size categories: return on assets, current ratio, debt-to-equity, return on equity. These
data are shown in Table 8-10 and are provided in the rulemakmg record (NCBA, 1999,
see—DCN 70073).
EPA did not base its analysis on these data since the data are not disaggregated by size
and producing region and are inconsistent with EPA's modeling framework.4 Also, EPA
determined that the NCBA survey data, if used, might lead to difficulties in estimating impacts.
This is particularly true with respect to debt-to-asset ratios derived from the NCBA data
(USEPA, 2000n and ERG, 2000b). EPA calculates these ratios using average debt and average
• assets reported by NCBA (shown in Table 8-10). This results in debt-to-asset ratios that exceed
70 percent, compared to the 40 percent threshold for USDA's "vulnerable" category, as discus'sed
in Section 4.2.5 (USDA/ERS, 2000g, 1977a and 1997e, Sommer et al., 1998). Without
examining the actual underlying survey data provided by NCBA, EPA is not able to assess the
validity of this derived ratio. Use of these data for this analysis, therefore, would typically lead to
cattle feedlots being assumed as "baseline closures" based on the criteria developed for this EA.
As such, these operations would be excluded from analysis since they would assume to close in
the pre-regulatory baseline. Therefore, EPA believes that the USDA data may produce more
usable data to estimate impacts for cattle operations than the NCBA data.
Additional reasons for not using these data in the analysis include: low response rate, lack
of information on the statistical methodology used to compute averages, inability to reproduce
some of the reported data, and various inconsistencies with other reported data by USDA, among
''NCBA reports that there is not a significant difference among Cattle feeding operations by size and
region (NCBA, 1999).
8-28
-------
other factors. This assessment is contained in the rulemaking record (USEPA, 2000n; ERG,
2000b).
Although EPA does not to use these data, the NCBA data is useful, since it allows the
Agency to evaluate how well the ARMS data for cow and calf operations data represent
conditions at cattle feedyards. Compared to the ARMS data, the average NCBA survey data fall
within the range of values for both of the revenue and expense categories reported in the average
ARMS data. For example, average gross revenue reported by NCBA was $670 per head in 1997
(Table 8-10). This compares to a gross revenue range of $500 per head to $860 per head based
on data in the 1997 ARMS (Table 4-5). The ARMS average data, however, do differ from the
NCBA data for net cash income. For example, NCBA reports an average net cash income of $30
PHAQ for 1999; this compares to a net cash income range of $80 to $320 per animal using the
1997 ARMS data (see Table 4-6). However, if the average NCBA data for revenue and expenses
are used to calculate returns, the average net cash income values are about $110 PHAO (see data
in Table 8-10), which is closer to the range of values based on USD A data for some model
CAFOs. Given the lack of information about how these data are computed and the inability to
reproduce the reported NCBA data, EPA is unable to fully evaluate these differences.
Dairy Operations
Data shown in Table 8-14 are distributed by broad facility size groups. As shown, more
than 80 percent of operations have fewer than 100 milk cows. These operations account for
about 40 percent of all milk cows (Table 8-14). There are fewer larger-sized operations with
more than 500 milk cows (2 percent of all farms), and these operations raise only 6.8 percent of
all beef cattle annually (Table 8-14). Smaller dairy operations with less than 100 milk cows are
slightly more diversified than larger ones, with about 20 percent of all farm revenue from crops.
This compares to dairy operations with more than 500 milk cows, where livestock comprises the
bulk of all annual farm sales and only about 5 percent of farm revenues are from crops
(Table 8-12).
Table 8-14 also shows the percentage of dairy cows owned by farmers compared to those
not owned by farmers. EPA uses this information on animal ownership as an indication of the
extent of production contract use in these sectors (see Section 2.3). Across all dairy operations in
1997, less than 2 percent of animals were not owned by farmers (Table 8-14) (USDA/ERS,
1999a).
Table 8-15 presents average income statement and balance sheet information for dairy
operations in 1997, by size of operation. The data shown in Table 8-15 reveal some differences
among operations by size. The income statement data (as well as the data in Table 8-12) point to
some increasing specialization as the size of an operation increases. There are no major
differences in the proportion of animals not owned at the largest operations compared with that at
the smallest operations. Additionally, differences between the smallest operations and all
operations on the basis of expenditures on livestock-related expenses are not large. Expenditures
8-29
-------
Table 8-14 Typical Financial Characteristics of Dairy Operations, By Size of Operation
Item
Number of farms
Percent of farms
Percent of value of production
Livestock value of production
Crop value of production
All Farms
129,034
100.0%
100.0%
85.9%
14.1%
Less than 100
104,863
81.3%
45.3%
82.3%
17.7%
100 to 500
21,678
16:8%
28.7%
83.5%
16.5%
More than 500
2,492
1.9%
26.0%
94.8%
5.2%a
Number of milk cows
Distribution of milk cows
Percent of milk cows owned
Percent of milk cows not owned
Number of sample farms with milk cows
10,503,839
100.0%
98.6%
1.4%
1,334
4,052,239
38.6%
97.8%
2.2%a
707
3,462,016
33.0%
98.3%
1.7%a
487
Debt-to-Asset Ratios
AH Regions
Pacific
Midwest
0.2000
0.2080
0.2115
0.1742
0.1271
0.2014
0.2185
0.2305
0.2260
EPA Derived Gross Cash Income Per Animal
All Regions
Pacific
Midwest
$2,572.65
$2,672.05
$2,583.60
$2,584.26
$6,786; 18
$2,619.74
$2,523.81
$2,342.77
$2,498.11
EPA Derived Net Cash Income Per Animal
All Regions
Pacific
Midwest
$486.92
$371.59
$535.14
$543.20
$564.70
$598.06
$465:64
$224.73
$443.71
2,989,584
28.5%
99.9%
0.1%
140
0.2616
0.2366
d
$2,612.95
$2,567.29
d
$435.17
$401.74
d
Source: USEPA and USDA/ERS, 1999a.
a = The relative standard error of the estimate exceeds 25 percent, but no more than
b - The relative standard error of the estimate exceeds 50 percent, but no more than
c = The relative standard error of the estimate exceeds 75 percent.
d = Data insufficient for disclosure.
50 percent.
75 percent.
8-30
-------
Table 8-15. Income Statement and Balance Sheet for Dairy Operations, by Size of Operation, 1997
Item
AH farms
Less than 100
100 to 500
More than 500
Income Statement
Gross cash income
Livestock income
Crop sales (incl. net CCC loans)
Government payments
Other farm-related income 1;
Total variable expenses
Livestock purchases
Feed
Other variable expenses ^
Total fixed expenses
Equals: Net cash farm income
Less: Depreciation and Other 3/
Plus: Value of inventory change
Plus: Nonmoney income *
Equals: Net farm income
209,423
185,516
10,941
2,547
10,419
145,227
l,300a
70,782
73,143
24,560
39,637
18,643
11,113
4,468
36,574
99,864
83,945
8,161
1,943
5,815
63,732
788
24,536
38,408
15,141
20,991
10,529a
4,848
4,327
19,637
403,057
357,504
d
4,942
d
282,565
d
127,881
145,945
46,128
74,364
36,688
26,254
4,993
68,923
3,134,685
2,962,979
d
7,1 18a
d
2,379,377
d
1,519,860
738,358
233,247
522,061a
203,1 lOa
143,001a
5,828
467,780a
Balance Sheet
Farm assets
Current assets
Non-current assets
Land, buildings, and equipment 5/
Farm liabilities
Current liabilities
Farm equity
657,364
91,278
566,086
474,209
131,456
32,323
525,908
457,331
53,380
403,952
358,787
79,653
19,570
377,679
1,106,763
156,738
. 950,025
603,425
241,821
60,566
864,942
5,164,500
1,116,417
4,048,083
2,168,935
1,351,024
323,253
3,813,476
Source: Based on USDA/ERS, 1999a. Original data are in the rulemaking record (DCN 70063).
"Machine-hire, custom work, livestock grazing, land rental, contract fees, and other farm-related sources.
^Includes livestock leasing, custom feed processing, bedding, grazing, supply, transportation, storage, general
business expenses, and registration fees. .
''Includes labor, non-cash benefits. Footnote a refers to non-cash benefits portion of the total.
"'The value of home consumption plus an imputed rental value of farm dwellings.
s'The value of the operator's dwelling and associated liabilities are incl. if the dwelling was located on the farm.
a = Relative standard error (RSE) of the estimate exceeds. 25 percent, but no more than 50 percent.
b = RSE of estimate >50%, but <75%. c = RSE of estimate >75%. d = Data insufficient for disclosure.
8-31
-------
on livestock and feed average about 40 of total variable expenses at an average dairy operation
with less than 100 milk cows; all operations are associated with expenditures on livestock and
feed averaging about 50 percent of total variable expenses (Table 8-14). These differences may
be explained by the degree of specialization and feeding strategies, and other factors.
Operating margins (measured in terms of average net cash farm income as a percentage of
average gross cash income) among differently sized operations do not differ substantially:
operations with less than 100 milk cows show an operating margin of 21 percent, as compared to
17 percent at operations with more than 500 milk cows (USDA/ERS, 1999a). However, the
smallest operations show a lower return on assets (measured as average net farm income to
average farm assets) compared with the largest operations: operations with less than 100 milk
cows show an average return on assets of 4.3 percent, as compared to 6.2 percent and 9.1 percent
at operations with between 100 and 500 milk cows and operations with more than 500 milk cows,
respectively (USDA/ERS, 1999a). See Table 8-15. The 1997 ARMS data include, among an
average farm's assets, the value of the owner's home when it is located on the farm. Since
smaller operations may be more likely to have the owner's dwelling located on the farm than
larger operations, if dwelling values were excluded, the returns on the "business" assets might be
higher for these smallest operations.
Section 4 of this report presents key financial data used for this analysis, shown in Table
8-15, that are calculated onto a per-animal basis. For the dairy sector, total gross farm revenues
are estimated to range from $2,340 to $2,620 per head (includes revenue from other farm-related
sources). Net cash income ranges from $230 to $600 per head among CAFO models, depending
on facility size and region (see Tables 4-5 and 4-6).
8.2 PROFILE OF BEEF AND DAIRY PROCESSING SECTORS
Beef feedlots and dairy farms represent the beginning of the meat and dairy products
marketing chain, which also includes meat packers, food processors, and retailers. Feedlots and
dairy farms provide me raw materials to slaughterers, packers, and processors in the form of live
cattle and raw farm milk. These are then converted into cuts of meat and various processed foods
and milk and dairy products. These products are eventually sold to consumers at retail
establishments. ,
By NAICS code, beef and dairy processors are classified under animal slaughtering
(NAICS 311611); meat processed from carcasses (NAICS 311612); rendering and meat
byproduct processing (NAICS 311613); dry, condensed and evaporated dairy manufacturing
(NAICS 311514); fluid milk manufacturing (NAICS 311511); creamery butter manufacturing
(NAICS 311512); and cheese manufacturing (NAICS 311513).
Additional information on the processing sectors in these industries is provided in
Section 2, which also shows how EPA estimates the potential number of processors that may be
8-32
-------
affected by the proposed regulations as co-permittees. As discussed in Section 2.4, EPA does not
anticipate that processors within the beef and dairy industries would be subject to the proposed
CAFO regulations as co-permittees. Largely, EPA's determination is based on the fact that
production contracting accounts for only a small share of beef and milk production (USDA/ERS,
1999a and 1996c; Heffeman, et al., 1999). Also, animal ownership on beef and dairy farms is
mostly by the farm operator (USDA/ERS, 1999a).
Contractual agreements in these sectors are typified by marketing contracts and animal
ownership on beef and dairy farms is mostly by the farm operator (USDA/ERS, 1999a and
1996c). Less than two percent of all beef farms produced cattle and calves under contract in
1993, which consisted mostly of marketing contracts where farms or feedlots agree to sell packers
a certain amount of cattle on a predetermined schedule (USDA/ERS, 1996c). In the dairy sector,
milk is produced under marketing orders through verbal agreement with their buyer or
cooperative. Although not technically a "contract" since quantity and a final price are not
specified before the sale, milk production through use of such quasi-marketing contracts leaves
management decision with farmers since ownership is retained while the commodity is produced.
There is limited use of production contracts in these sectors, but these are mostly used to
specialize in one stage of livestock production, such as to raise stocker cattle (e.g., backgrounding
operations) and to raise replacement heifers (USDA/ERS, 1996c). This allows both farm
operators (i.e., contractor and contractee) to increase business volume with limited facilities
through specialization. For example, custom feeding operations provide cattle feeding services
but in some cases do not own the animals. In these cases, a beef feedlot operation may agree to
fatten or "finish" cattle not owned by the operation for a fee based on weight gain prior to
slaughter. Cattle backgrounding or stocker operations grow feeder cattle from any time between
when calves are weaned until they are on a finishing ration in the feedlot (Rasby et al., 1994).
Background cattle can be owned by a finishing feedlot, retained by ranchers or owned by the
backgrounding feedlot (Duncan et al., 1997). The contractuai agreements at these operations can
also vary greatly with ownership retained by the rancher, backgrounding operation, or finishing
feedlot. Stocker operations also typically combine calves from several cow-calf operations into a
larger production unit (Ward, 1997). •
In the dairy sector, farms may use production contracts with other operators to produce a
stream of replacement heifers (USDA/ERS, 1996d). Heifer replacement operations raise pre-
calving bovine cows under contract and may specialize in rasing calves, young heifers, breeding,
or the entire cycle from weaning to calving (USDA/APHIS, 1993). These operations may have
contracts to raise heifers for a period of time or may raise their own heifers to sell.
8.2.1 Structure of the Red Meat (Beef) Industry
One-half of all fresh and frozen meat received by retailers is supplied through own or
group owned meat warehouses, some which have processing facilities on site. Meat packing
8-33
-------
facilities supply the other one-half of all of all fresh and frozen meat received by retailers
(USDA/GIPSA, 1996b).
The meat packing industry is composed of slaughterhouses, where livestock are
slaughtered and further-processed, and specialized meat processors, which do not slaughter but
manufacture sausage, luncheon meats, and other prepared products (Kohls and Uhl, 1998). These
companies may cover all meat product types, including cattle, calves, steer and heifers, pork, and
sheep and lambs.' There are also rendering facilities that are involved in the manufacture of a wide
variety of by-products, including hides, pelts, lard, offal, that have value in the manufacture of
clothing, foodstuffs, fertilizers, and industrial products (Kohls and Uhl, 1998),
Unlike the poultry industry, where a chicken may be hatched, raised, slaughtered and
processed all in the same complex or at least by the same company in a matter of weeks, the
marketing chain for cattle involves ownership that may change hands a number of times, often
across distant locations, e.g.; from a breeding farm in Georgia to foraging in Texas to an Iowa
feedlot before being shipped off to slaughter. Because of obvious biological differences such as
size, feed requirements, reproduction and growth, the same economies of scale that allow for
complete integration in chicken production do not exist in producing a cow. Nevertheless,
economies of scale are significant in the later stages of cattle production, such as at feeder lots
where cattle are fed to market weight and at slaughter and packing facilities.
The marketing of cattle and calves is largely decentralized. Most cattle and calves are
procured through private (non-public) arrangements and spot market agreements whereby farmers
and feedlots sell their cattle directly to packers or use country dealers who buy and sell livestock
for profit. In 1995, 86 percent of all cattle and 75 percent of all calves were procured on the open
or spot market (USDA/GIPSA, 1996b). More traditional methods of procurement, such as public
markets, terminal markets and auctions, comprised little more than one-tenth of all cattle
purchases by meat packers in 1995 (USDA/GIPSA, 1996b). Use of public markets has been
steadily declining, although use is higher in the areas of less concentrated production away from
the Cornbelt Public markets are more significant for exchange between farmers of breeder and
feeder cattle than for trade in fattened cattle.
Although the beef industry is not as integrated as the poultry and hog sectors, the
relationship between cattle producers and processors has become more interconnected, either
through contractual arrangement or through actual ownership (integration) by processors (Bastian
et al., 1994). Less common forms of procurement, such as formula pricing, forward contracting
and marketing agreements,5 are becoming more common in the beef sector, especially among
larger beef packers. Backward integration of meat packers into livestock feeding either through
packer ownership of feedlots or custom feeding by feedlots under contract to meat packers is also
becoming mote prevalent, such as packer feeding of calves.
5Marketing agreements are long-term purchase agreements in which the packer agrees to purchase a
specified number of cattle per specified time period.
8-34
-------
These methods of procurement, however, still comprise a relatively small share of overall.
cattle and calf procurement. A 1992-1993 survey by USDA's GIPSA (Grain Inspection, Packers
and Stockyards Administration) of procurement methods commonly used by steer and heifer
packers indicate that more than 82 percent of procurement transactions (lot purchases) were
conducted through the open (spot) market (USDA/GIPSA, 1996c). Marketing agreements and
forward pricing accounted for 8 percent and 7 percent, respectively, of purchases. Packer-fed or
owned cattle accounted for under 3 percent of procurement transactions.
Two important trends influencing the meat packing industry have been specialization and
decentralization. In the past, plants served centralized markets and consisted of large facilities
that slaughtered most types of animals and also processed various meat products. Today, plants
tend to specialize in slaughtering only one species of. animal; processing is done in a separate
facility (Kohls and Uhl, 1998). This shift is also marked by movement westward closer to fed
cattle supplies, away from terminal markets, and an increase in direct selling to packers (USEPA,
1999h). Meat packers have not integrated forward into retailing operations, although there has
been backward integration of meat packers into livestock feeding. Barriers such as little product
differentiation, geographically dispersed components of production, and small, highly diverse
production units, have prevented further vertical integration (Ward, 1997).
Additionally, rapid consolidation and mergers among packing companies have resulted in
fewer and larger firms and a highly concentrated industry structure. In 1997, the top four largest
meat packing firms slaughtering cattle controlled 68 percent of the market; these four firms
slaughtering steers and heifers accounted for 80 percent of the market (USDA/GIPSA, 1996c and
1998). Consolidation is being driven by the large economies of scale possible in the meat packing
industry. MacDonald (2000) reports that the largest plants have significant advantages in
slaughter costs and the largest cattle packers can deliver beef to buyers at costs 3 percent below
what plants one-quarter the size can.
Pricing issues for fed cattle remain a primary concern in the industry, particularly since
cattle prices have been low in recent years. Rapid consolidation and mergers between packing
companies have resulted in fewer and larger firms and a highly concentrated industry structure. In
1997, the top four largest meat packing firms slaughtering cattle controlled 68 percent of the
market; the four-firm concentration ratio for the slaughter of steers and heifers accounted for 80
percent (USDA/GIPSA, 1996a and 1998). The largest beef packer, IBP Incorporated, alone
controls nearly 40 percent of the market (USDA/GIPSA, 1996a). Such concentration has led to
concerns that producers in most areas may have access to a single buyer, thus reducing the
producers ability to bid up prices and enabling the packer to pay farmers less for their cattle. Of
particular concern among cattle producers is the issue of "captive supplies" of beef held by
packers and allegations that packers are engaging in price discrimination by using their own
supply of fed cattle to undercut prices.
The ability for meat packers to exercise oligopsony or monopsony market power in the
procurement of cattle has been the topic of extensive research and also congressional debate
.8-35
-------
(USDA/GCPSA, 1997). Most research, however, has failed to conclusively show that
concentration in the meat packing industry has negatively affected farm prices paid for cattle
(Nelson and Hahn, 1996). A study by USDA did conclude that higher levels of regional buyer
concentration may have a negative and statistically significant effect on cattle prices; howeyer, the
change in prices paid is most likely small (USDA/GIPSA, 1997b).
8.2.2 Structure of the Milk and Dairy Foods Industry
Most milk enters marketing channels from dairy farms as raw farm milk and is processed
as fluid milk and also a wide range of dairy products, including cheese, butter, ice cream and
frozen desserts, Vsoft" manufactured products (yogurt, cottage cheese, sour cream, etc.),
processed milk products (dry, evaporated, and condensed milk), and also milk by-products
(lactose, whey, casein, etc.). The processing and manufacturing of dairy products from farm milk
is performed primarily by either dairy cooperative or by investor-owned processing firms, such as
Kraft and Dean Foods and also supermarkets like Safeway and Kroger.
Milk and dairy production has become increasingly specialized but has not experienced
vertical integration in the same way as other livestock industries. In part this is attributable to the
large role of farmer-owned, farmer-controlled dairy cooperatives in the industry. Cooperatives
handle about 80 percent of milk delivered to plants and dealers, with the remaining milk produced
by "independent9' or "non-member" dairy farmers (Jacobson and Cropp, 1994). Cooperatives are
farmer-owned, farmer-controlled for-profit corporations which operate at cost by allocating net
margins back to milk producers on a patronage basis. In most cooperatives, milk producers enter
a 1- to 3-year membership agreement committing the producer to marketing all milk through the
cooperative (Jacobson and Cropp, 1994). Because farm milk must be collected daily and requires
refrigeration and other special handling, this places dairy farmers at a marketing disadvantage and
increases the likelihood that they will participate in a cooperative. Through the bargaining power
of cooperatives, dairy farms can often secure prices above the minimum federal or state market
order prices (Wolf and Hamm, 1998).
Cpoperatives may either negotiate milk prices with processors or process milk themselves.
There are two basic types of dairy cooperatives: bargaining-only and manufacturing/processing
(Manchester and Blayney, 1994). Bargaining-only cooperatives accounted for 68 percent of dairy
cooperatives in 1992. Such cooperatives negotiate prices and terms of trade for their members
milk and tend to be smaller milk handlers, supplying approximately 25 percent of cooperative
milk. Farmers are limited to producing raw milk and the role of the cooperative is to negotiate
the best price for that milk. Manufacturing/processing cooperatives, on the other hand, own
facilities to process milk and manufacture dairy products. Dairy cooperatives with processing
facilities represented 32 percent of all cooperatives in 1992, but their share of all cooperative milk
was 75 percent, most of which was produced by the large diversified cooperatives (Manchester
and Blayney, 1994).
8-36
-------
Markets for milk are often regional because of perishability and bulk and high handling
costs. Fluid milk processing is characterized as an oligopolistic industry at the regional level, with
most major cities served by a few processor firms (Kohls and Uhl, 1998). The number of fluid
milk companies continues to decline. From 1982 to 1992, the number of fluid milk processors
dropped from 853 to 525 companies (USDC, 1994). Economies of size in fluid milk processing
and distribution and continued dairy company mergers and plant consolidation has contributed to
farm-gate pricing concerns among dairy farmers. A USDA study of price transmission between
the farm and retail sector for fluid milk from 1983 to 1990 indicate asymmetric pricing adjustment
since retail prices tended to react more quickly and completely to farm price increases than to
farm price decreases (Hansen et al., 1994). Other dairy products, such as cheese and ice cream,
have a wider, more national market in which there is greater product differentiation and
competition.
8.3 CAFO ANALYSIS
This section presents the results of EPA's CAFO level analysis for the dairy and cattle
sectors, including fed cattle, veal and heifer operations. As discussed in Section 4, EPA uses a
representative farm approach to estimate the impact of the proposed CAFO regulations on
affected operations. Each model CAFO differs by facility size groupings and key farm production
regions. For these sectors, the production regions reflected in this analysis are the Midwest
(MW), Central (CE), and Pacific (PA) regions, as defined in Table 4-1 (Section 4). Section 4
provides a summary of how EPA developed the various financial models used for this analysis.
The Development Document (USEPA, 2000a) provides additional information on the cost models
developed by EPA.
Results presented in this section focus on the "BAT Option" that refers to EPA's
proposed technology option for the CAFO regulations that would impose Option 3 requirements
for the beef (including heifers) and the dairy subcategories and Option 5 requirements for veal
subcategory (described in Section 3). For the purpose of this discussion, the "two-tier structure"
refers to BAT Option 3 (beef, heifer, and dairy operations) and BAT Option 5 (veal) in
combination with NPDES Scenario 4a that covers all operations with more than 500 AU. Where
indicated, the two-tier structure may refer to the alternative threshold at 750 AU (Scenario 5).
The "three-tier structure " refers to BAT Option 3 (beef, heifer, and dairy operations) and BAT
Option 5 (veal) in combination with NPDES Scenario 3 that covers operations down to 300 AU
based on certain conditions. Results for other technology options and scoping scenarios
considered by EPA as part of this rulemaking are also summarized. Table 3-1 summarizes EPA's
proposed and alternative ELG Options and NPDES Scenarios discussed in this section.
Section 8.3.1 presents a summary of the cost input data EPA uses for this analysis,
including (post-tax) per-animal and per-facility costs for EPA's model CAFOs. Section 8.3.2
presents EPA's estimate of the aggregate, national level costs of the proposed CAFO regulations
for the beef and dairy sectors. Section 8.3.3 presents EPA's predicted financial impacts to this
8-37
-------
sector in terms of the estimated number and percentage of CAFOs that are expected to experience
financial stress as a result of the proposed CAFO regulations. For some sectors, EPA evaluates
economic impacts to CAFOs in this sector two ways—assuming that a portion of the costs may
be passed on from the CAFO to the consumer (Partial CPT) and assuming mat no costs
passthrough so that all costs are absorbed by the CAFO (Zero CPT).
8.3.1 Overview of Cost Input Data
Tables 8-16 through 8-18 present estimated input costs that EPA uses to assess costs and
impacts to the cattle and dairy sectors. These data include the post-tax annualized compliance
costs, estimated on a per-animal and per-facility. These costs reflect EPA's estimated capital
costs, annual operating and maintenance costs, start-up or first year costs, and also recurring costs
(discussed in the Development Document, USEPA, 2000a). These facility costs are annualized
using the approach described in Appendix A of-this report. Appendix A shows the individual
sector costs by model across all technology options.6
' Other input data for this analysis include EPA's estimate of the number of affected CAFOs
and baseline financial conditions at model CAFOs. EPA's estimate of the number of animal
confinement operations that would be defined or designated as CAFOs is presented in Section
8.1.2.1 (see Table 8-2). Additional information is provided in Section 2 of this report. The
average baseline financial conditions for model CAFOs that EPA assumes for this analysis are
presented in Section 4. Tables 4-5 through 4-9 in that section present the financial data used in
this analysis and include gross farm revenues, net cash flow, and debt-to-asset ratios for this
sector, as derived by EPA using the 1997 ARMS data.
For beef, heifer, and dairy operations, Table 8-16 presents the estimated post-tax
annualized compliance costs (in 1997 dollars) for these sectors under the proposed BAT Option
(Option 3). For veal operations, Table 8-16 presents estimated compliance costs the proposed
BAT Option for that subcategory (Option 5). Table 8-17 presents cost estimates for Option 3A
that reflects the cost of additional requirements, such as liners, groundwater monitoring, and
recordkeeping costs, for those facilities (beef, heifer, and dairy operations only) where there is a
hydrologic link to surface waters. For this analysis, based on available data and information,
EPA's analysis assumes that 24 percent of the affected operations have a hydrologic connection
to surface waters, as described in the Development Document (USEPA, 2000a). Option 3 costs
(Table 8-16) reflect average cost conditions across all operations—including those operations
with and without a hydrologic link.7
6The estimated costs are the same across the NPDES Scenarios, i:e., technology option costs do not
change by scenario, although total costs change due to the difference in numbers of CAFOs affected under each
scenario.
'Alternatively, estimated costs for "Option 3B" reflect representative facility level costs where no
hydrologic link is present. Option 3,3A, and 3B costs are provided in the Development Document.
8-38
-------
As shown in Table 8-16, post-tax costs for the BAT Option (Option 3) range from $1.70
to $41.00 per animal for the beef sector, from $3.70 to $7.20 per animal for the veal sector
(Option 5), from $3.50 to $27.70 per animal for the heifer sector, and from $16.80 to $61.50 per
animal for the dairy sector. The range in costs is explained by differences in the assumed
availability of land for manure applications (see definition of Category 1, 2, and 3 in
Section 4.1.2), region, and size of CAFO. As these tables show, costs on a per-animal basis
generally decrease as model CAFO size increases.
As shown in Table 8-17, the range of costs are much higher under Option 3A. Costs
range from $13.80 to $80.10 per animal for the beef sector, from $18.18 to $60.86 for the heifer
sector, and from $85.60 to $235.20 per animal for the dairy sector. Tables 8-16 and 8-17 also
present compliance costs on a per-CAFO basis. Per CAFO compliance costs range from $3,180
(Option 3) to $725,868 (Option 3A) in the beef sector, from $2,010 to $2,880 (Option 5) in the
veal sector, from $4,250 (Option 3) to $40,710 (Option 3A) in the heifer sector, and from $9,330
(Option 3) to $172,610 (Option 3 A) in the dairy sector. In general, the annualized post-tax
compliance costs per representative CAFO increase with model size.
When Option 3 costs are considered without the 3 A component (the lower end of the
ranges shown in Table 8-16), the costs per animal are generally within ranges that have been
deemed affordable in several studies of regulatory impacts that focused on the dairy sector. In an
analysis of the economic impacts on livestock producers from wastewater and runoff control
requirements in coastal areas, incremental costs were reported on a per-animal basis (Heimlich
and Barnard, 1995; DPRA, 1995). The range of estimated costs—$17 to $49 per dairy cow-was
determined to be affordable for producers (DPRA, 1995). Similarly, researchers at Cornell
University surveyed milk producers in New York who indicated that they would likely stay in
business if they had to pay up-to $50 per cow for environmental improvements (Poe, et al., 1999).
In general, most studies may amortize costs but do not always account for tax savings.
Therefore, EPA's estimates may reflect the upper-end of costs compared to other cost estimates.
.However, as documented in the Development Document (USEPA, 2000a), EPA believes that its
estimated costs are conservative.
The costs presented here are those assumed to be incurred by the regulated CAFO and do
not account for the likelihood that some compliance costs will be passed on through the
marketing levels in the industry.
Table 8-18 presents the range of per animal post-tax compliance costs in 1997 dollars for
the cattle and dairy sectors for each regulatory option, including the BAT Option and Option 3 A.
(The proposed and alternative ELG Option and NPDES Scenarios considered by EPA during this
rulemaking are defined in Table 3-1.) As shown, for beef operations, option costs range from
$0.10 to $80.30 per animal. For veal, costs range from $2.70 to $18.70 per animal (Option 5).
For heifer operations, option costs range from $0.70 to $60.80 per animal. For dairies, the option
costs range from $3.60 to $235.20 per animal. The proposed BAT Option costs (Option 3) has
8-39
-------
Table 8-16. Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 3)
Sector
Beef
Veal
Heifers
Dairy
Reg.
MW
CE
MW
PA
MW
PA
MW
Model
Small
Ml
M2
LI
. L2
Ml
M2
LI
L2
Ml
M2
Ml-
M2
LI
Ml
M2
LI
Ml
M2
LI
Small
Ml
M2
LI
Average
Animals
Per
Facility
112
455
111
1,877
30,003
455
111
1,877
30,003
400
540
400
750
1,500
400
750
1,500
235
'460
1,419
200
235
460
1,419
Cat. 1
Cat. 2
Cat. 3
Per Animal
Cat. 1
Cat. 2
Cat. 3
Per Facility
$1997
$28.37
$29.60
$21.00
$8.26
$4.61
$31.74
$24.12
$9.39
$4.52
$7.20
$3.72
$22.94
$15.41
$5.35
$24.14
$16.11
$5.74
$48.23
$31.66
$20.02
$60.39
$58.74
$39.99
$27.37
" NA
$32.26
$23.29
$15.31
$7.79
$41.03
$31.25
$17.82
$12.06
NA
NA
$27.72
$17.43
. $7.79
$27.49
$18.58
$6.60
$53.78
$47.34
$41.09
NA
$61.51
$50.07
$38.76
NA
$15.87
$11.31
$7.12
$3.83
$10.81
$7.21
$3.37
$1.65
NA
NA
$10.63
$6.86
$3.51
$14.13
$9.37
$5.04
$39.72
$25.49
$16.83
NA
$50.50
$34.05
$24.32
$3,178
$13,470
$16,317
$15,501
$138,195
$14,440
$18,741
$17,623
$135,620
$2,881
$2,009
$9,178
$11,555
$8,032
$9,657
$12,082
$8,603
$11,334
$14,563
$28,406
$12,078
$13,804
$18,394
$38,840
NA
$14,680
$18,096
$28,739
$233,728
$18,669
$24,282
$33,445
$361,844
. NA
NA
$11,088
$13,072
$11,691
$10,997
$13,936
$9,901
$12,637
$21,776
$58,308
NA
$14,454
$23,031
$55,004
NA
$7,221
$8,791
$13,357
$114,866
$4,920
$5,602
$6,318
$49,582
NA
NA
$4,254
$5,142
$5,265
$5,653
$7,031
$7,563
$9,334
$11,724
$23,883
NA
$11,868
$15,665
$34,504
Source: USEPA. Costs are shown for the BAT Option: Option 3 (beef, heifers, and dairy operations) and Option 5
(veal operations). See Table 4-1 for definitions of model regions and sizes. Costs reflect the estimated capital
costs, annual operating and maintenance costs, start-up or first year costs, and also recurring costs assumed by
EPA (see the Development Document, USEPA, 2000a) that are annualized using the approach described in
Appendix A.
8-40
-------
Table 8-17. Per-Animal and Per-Facility Post-tax Annualized Compliance Costs (Option 3A)
Sector
Reg.
Model
Average
Animals
Per
Facility
Cat. 1
Cat. 2
Cat. 3
Per Animal
Cat. 1
Cat. 2
Cat. 3
Per Facility
$1997
Beef
Beef
Veal
Heifers
Dairy
MW
CE
MW
PA
MW
PA
MW
Small
Ml
M2
LI
L2
Ml
M2
LI
L2
Ml
. M2
Ml
M2
LI
Ml
M2
LI
Ml
M2
LI
Small
Ml
M2
LI
112
455
777
. 1,877
30,003
455
111
' 1,877
30,003
NA
NA
400
750
1,500
400
750
1,500
235
460
1,419
200
235
460
1.419
$80.06
$64.02
$47.97
$27.46
$15.88
$70.62
• $54.64
$30.06
$16.65
'NA*
NA
$58.34
$40.34
$23.91
$53.46
$36.78
$18.66
$228.84
$148.83
$100.45
$228.08
$183.75
$126.14
$88.44
NA
$67.24
$50.56
$34.65
$19.08
$80.32
$61.98
$38.59
$24.20
NA
NA
$60.86
$43.96
$27.14
$55.50
$39.57
$20.16
$235.23
$164.88
$121.64
, NA
$187.38
$136.62
$99.96
NA
$51.06
$38.69
$26.50
$15.12
$50.31
$38.06
$24.19
$13.79
NA
NA
$46.71
$32.14
$22.28
$44.33
$30.50
$18.18
$221.62
$143.21
$97.44
NA
$176.79
$120.80
$85.57
$8,967
,$29,127
$37,276
$51,549
$476,586
$32,133
$42,459
$56,426
$499,538
NA
NA
$23,337
$30,259
$35,870
$21,382
$27,587
$27,990
$53,778
$68,462
$142,544
$45,617
$43,181
$58,027
$125.497
NA
$30,592
$39,282
$65,034
$572,348
$36,544
$48,162
$72,441
$725,968
NA
NA
$24,344
$32,973
$40,710
$22,200
$29,680
$30,235
$55,279
$75,843
$172,613
NA
$44,034
$62,845
$141.838
NA
$23,232
$30,065
$49,740
$453,566
$22,893
$29,570
$45,402
$413,786
• NA
NA
$18,686
$24,105
$33,421
$17,733
$22,873
$27,271
$52,080
$65,879
$138,268
NA
$41,547
$55,566
$121.418
Source: USEPA. Costs are shown for the BAT Option: Option 3 (beef, heifers, and dairy operations) and Option 5
(veal operations). See Table 4-1 for definitions of model regions and sizes. Costs reflect the estimated capital
costs, annual operating and maintenance costs, start-up or first year costs, and also recurring costs assumed by
EPA (see the Development Document, USEPA, 2000a) that are annualized using the approach described in
Appendix A.
8-41
-------
Table 8-18. Summary of the Range of Post-Tax Annualized Compliance Costs Per Animal, By Option
Option
Option 1
Option 2
Option 3
Option 4
Option 5
Option 6
Option 7
Beef
Min.
Max.
Veal
Min.
Max.
Heifer
Min.
Max.
Dairy
Min.
Max.
($1997)
$0.08
$0.08
$1.65
$1.77
$26.12
• $0.08
$0.42
$16.69
$35.32
$80.32
$50.20
$68.12
$35.32
$35.32
$2.66
$3.72
• $4.75
$12.86
$3.72
$3.72
$3.72
$3.61
$7.20
$7.78
$18.74
$7.20
$7.20
$7.20
$0.67
$0.67
$3.51
$6.43
$1.42
$0.67
$0.67
$16.85
$20.87
$60.86
$37.32
$23.68
$20.87
$21.92
$3.60
$3.60
$16.83
$19.22
$15.01
$3.98
$10.60
$41.79
$30.66
$235.23
$79.26
$51.41
$51.40
$50.30
Source: USEPA.
the highest maximum costs in the beef, heifer, and dairy sectors, reflecting estimated Option 3A"
costs. The -minimum cost per animal for Option 3 generally ranges between the costs for
alternative options. The maximum cost per animal for Option 3 (without Option 3 A) also ranges
between the maximum costs for alternative options.
8.3.2 Estimates of National Annual Compliance Costs
Table 8-19 presents EPA's estimate of the aggregate national level compliance costs for
the proposed BAT Option (Option 3 for beef, heifer, and dairy operations and Option 5 for veal
operations) and the co-proposed two-tier structure (Scenario 4a at 500 AU threshold) and the
three-tier structure (Scenario 3). Costs under the two-tier structure at the 750 AU threshold
(Scenario 5) are also briefly discussed, along with other regulatory alternatives considered by
EPA during this rulemaking. The description of the proposed BAT Option and the co-proposed
NPDES Scenarios is provided in Section 3.
Across all cattle and dairy, operations, EPA estimates total incremental costs (post-tax) of
the proposed BAT Option at $255 million and $302 million per year under the two-tier structure
(500 AU) and the three-tier structure, respectively (Table 8-19). Under the two-tier structure at
750 AU threshold, estimated costs are $216 million per year. EPA estimates that the largest
portion of total costs would be borne by the beef sector, estimated to incur roughly half of total
costs to these sectors. Among fed cattle operations, EPA estimates that the cost of the BAT
Option is $135 million and $144 million per year under each of the co-proposed structures; more
than 80 percent of these costs are estimated to be borne by operations with more than 1,000 AU.
8-42
-------
Table 8-19. Total Estimated Post-Tax Compliance Costs
Scenario/Size
>1,000 AU
Total Two-
Tier Structure
(>750 AU)
Total Two-
Tier Structure
(>500 AU)
Total Two-
Tier Structure
(>300 AU)
Total Three-
Tier Structure
(>300 AU)
Option
Number of CAFOs
Cost of Proposed
BAT Option
Cost of Alternative
Options
Number of CAFOs
Cost of Proposed
BAT Option
Cost of Alternative
Options
Number of CAFOs •
Cost of Proposed
BAT Option
Cost of Alternative
Options
Number of CAFOs
Cost of Proposed
BAT Option
Cost of Alternative
Options
Number of CAFOs
Cost of Proposed
BAT Option
Cost of Alternative
Options
Fed Cattle
Veal
Heifers
Dairy
. ". ($1997 millions)
2,080
$118.5
$40.2-$500.2
2,480
$125.3
$43.3 - $513.6
3,080
$135.0
$47.8-$532.5
4,070
$148.3
$54.9-$555.2
3,210
$143.5
$48.6-$544.4
10
$0.02
$0.01-$0.07
40
$0.07
$0.05 - $0.24
90
$0.17
$0.12-$0.59
210
$0.50
$0.3-$1.5
140
$0.54
$0.2-$1.2
300
$2.8
$0.5-$4.1
420
$4.2
$1.2 -$6.0
800
$8.6
$3.6412.0
1,050
$10.9
$4.9-$15.4
980
$10.6
$4.5-$ 15.1
1,450
$65.7
$39.2-$74.4
2,260
$86.9
$53.2 -$96.3
3,760
$111.4
$60.7-$125.5
6,970
$152.4
$77.3-$180.4
6,480
$146.9
$72.2-$ 177.8
Source: USEPA. Costs are shown for the'BAT Option: Option 3 (beef, heifers, and dairy operations) and Option 5
(veal operations). See Table 4-1 for definitions of model regions and sizes.
Numbers of CAFOs include defined CAFOs only. Costs include those for designated beef and dairy operations.
8-43
-------
EPA estimates that the dairy sector would incur compliance costs of $111 million per year
under the two-tier structure (500 AU threshold), with about 60 percent of these costs attributable
to CAFOs with more than 1,000 AU. See Table 8-19. Under the three-tier structure, estimated
costs are $147 milh'on per year, with about 45 percent incurred by CAFOs with more than 1,000
AU. Between the two modeled regions (PA and MW), the Pacific region would bear the greatest
proportion of compliance costs, estimated at more than 72 percent under the two-tier structure
and 66 percent under the three-tier structure. Costs are more or less evenly split between the two
modeled regions (CE and MW), but tend to be somewhat higher in the Midwest region.
A total of $9 milh'on per year would be incurred by heifer operations under the two-tier
structure; under the three-tier structure, compliance costs to heifer operations are estimated at
$11 million per year (Table 8-19). The major portion of this cost would be borne by heifer
operations with fewer than 1,000 AU (67 percent to 74 percent) under each of the proposed tier
structures. Between the two modeled regions (CE and MW), the Central region incurs 86 percent
of the costs to this sector under the two-tier structure (500 AU threshold) and 81 percent under
the three-tier structure. At veal operations, EPA estimates that compliance costs would total $0.2
million per year under the two-tier structure and $0.5 million per year under the three-tier
structure (Table 8-19). The major portion of this cost would be borne by veal operations with
fewer than 1,000 AU, estimated at 71 percent to 95 percent of cost depending on scenario.
8.3.3 Analysis of CAFO Financial Impacts
EPA's impact analysis uses a representative farm approach to estimate the number of
CAFOs that would experience affordable, moderate, or stress impacts as a result of the CAFO
regulations, as described in Section 4. Economic achieyability is determined by applying the
proposed criteria, which include a sales test and also analysis of post-compliance cash flow and
debt-to-asset ratio for an average model CAFO. Impacts are extrapolated to all CAFOs in the
beef and dairy sector using the estimated number of operations represented by each model CAFO.
As described in Section 4.2.5, if an average model facility is determined to incur economic
impacts under regulation that are regarded as "Affordable" or "Moderate," then the proposed
regulations are considered economically achievable. ("Moderate" impacts are not expected to
result in closure and are considered to be economically achievable by EPA.) If an average
operation is determined to incur "Stress," then the proposed regulations are not considered to be
economically achievable. "Affordable" and 'Moderate" impacts are associated with positive post-
compliance cash flow over a 10-year period and a debt-to-asset ratio not exceeding 40 percent, in
conjunction with a sales test result that shows mat compliance costs are less than 5 percent of
sales ("Affordable") or between 5 and 10 percent ("Moderate"). "Stress" impacts are associated
with negative cash flow or if the post-compliance debt-to-asset ratio exceeds 40 percent, or sales
test results that show costs equal to or exceeding 10 percent of sales.
8-44
-------
Using this classification scheme, EPA's analysis indicates that some beef and dairy
operations would experience financial stress as a result of the proposed CAFO regulations under,
both the proposed BAT Option and co-proposed scenario, assuming compliance costs cannot be
passed through the marketing chain.
As discussed previously, financial impacts are assessed to cattle and dairy operations
assuming that 24 percent of these operations will incur costs associated with groundwater
controls, such as liners, groundwater monitoring, and recordkeeping, where a hydrologic
connection from the confinement areas to surface water is present (Option 3 A). Tables 8-20 and
8-21 combine the results of the analysis of Option 3 and Option 3A
Table 8-20 presents the results of EPA's analysis for beef, veal, and heifer operations.
Under the two-tier structure, 10 beef operations are expected to experience financial stress under
the two-tier structure; no veal or heifer operations are expected to experience financial stress
under this co-proposed scenario. Under the three-tier structure, 20 beef and 30 heifer operations
are expected to experience financial stress under the three-tier structure; no veal operations are
expected to experience financial stress (Table 8-20). No designated CAFOs in these sectors are
expected to experience financial stress under either co-proposed scenario. All beef CAFOs that
are estimated to experience financial stress are operations that are assumed to have a hydrologic
link to surface water (i.e., assumed to incur estimated Option 3A costs). EPA did not evaluate
economic impacts to cattle operations under a cost passthrough scenario.
Table 8-21 shows results for the dairy sector. Under the two-tier structure, EPA
estimates that 320 dairy operations are expected to experience financial stress among defined
CAFOs. Under the three-tier structure, EPA estimates that 610 dairy operations are expected to
experience financial stress (Table 8-21). All dairies that are estimated to experience financial
stress are operations that are assumed to have a hydrologic link to surface water. EPA also
estimated that an additional 20 designated dairies would experience financial stress under both the
co-proposed structures. The number of designated dairies that are expected to experience stress
in this analysis are operations that are designated due to a groundwater link to surface waters and
are projected over a 10-year period.
These results for the dairy sector assume that no costs are passed through to consumers.
However, EPA expects that long-run market and structural adjustment by milk and meat
producers will diminish the estimated impacts to these sectors as costs are passed through to
consumers. To evaluate economic impacts to dairy operations under a cost passthrough scenario,
EPA assumes a 67 percent cost passthrough assumption. More information on the method and
data that EPA used to estimate this cost passthrough value is provided in Section 4.2.6.8
Assuming this level of cost passthrough, EPA's analysis indicates that no dairy operations would
8However, EPA uses a different estimate of the long-run price elasticity of supply than that shown in
Table 4-14. For this analysis, EPA uses an estimate reported by Buxton (1985) of by 0.501. This has the effect of
lowering the "selected" CPT value from 86 percent (shown in Table 4-14) to 67 percent, assumed for this analysis.
8-45
-------
§
ts
g
O
Jirt
'"SS
w
C3
*«
fcj
O
cu
cy
*,
jjj
a
CO
§
7*
'
V,
c
o
O
H-
a
•s
a
6
|SJ
C)
able 8-20. Impacted
H
1
Hi
0)
*
cu
«
en
en
2
CO
a>
«
cu
t3
^
e»
"3
n
•a
3
en
CO
cu
^
L.
•a
0
^
2
. «*
"a
IN
en
en
cu
CO
"S
a
• c
s
a
2
1
0
•5
^
<§•
*o
•22
o
CM
O
™
w
e
u
!/3
Two-Tier (>300 AU,
o
0^
0
0
oo
o
^
Number of CAFOs
o
v>
0
10
* *
o
oo
o
o
o
01
o
0
oc
oc
of
§
8
H
a
0
ON
o
o
o
o
1— 1
f^
o
CO
o
o
0
01
o
00
1
o
•1
o
oc
o
oc
01
1
t:
C
f5
a
a
a
H
a
j:
o
oo
ON
0
rr
o
o!
CO
Number of CAFOs
o
co
o
^
0
o
oo
o
o
o
*i
0
CO
o
o
o:
c
t
H
PC
o
^
0
§
i
o
o
00
ON
0
O
CO
Q
ct>
o
s
g
t
o
0
3.
0
o
01^
o
oo
o
01
CO
1
o"
1
3
3
ID
o §
y -a
§ "
\j3 O
o 2
^ft. V)
0 S
tS OT
gl
•il
2 to
CO .2
ource: USEPA. Impac
efinitions of model reg
vo.
^*
oo
-------
Table 8-21. Impacted CAFOs Under ELG Options & NPDES Scenarios, Dairy Operations
Alternative
ELG Options
and
NPDES
Scenarios
Total
#
CAFOs
Affordable
Moderate
'Stress
Zero Cost Passthrough
Affordable
Moderate
Stress
Partial Cost Passthrough
(Number of Affected Operations)
Two-Tier (>1000)
BAT Option
Alt. ELG
Options
1,450
1,450
1,450
0
0
0
0
1,450
1,450
0
0
0
0
Two-Tier (>750 AU, Scenario 5)
BAT Option
Alt. ELG
Options
2,260
2,070
1,480-2,260
50
0-220
130
0-560
2,210
0-2,260
50
0-50
0
0
Two-Tier (>500 AU, Scenario 4a)
BAT Option
Alt. ELG
Options
3,760
3,240
1,580-3,760
200
0-850
. 320
0-1,330
3,550
0-3,760
200
0-200
0
0
Two-Tier (>300 AU, Scenario 4b)
BAT Option
Alt. ELG
Options
7,140
5,810
1,580-7,140
640
. 0-2,660
700
0-2,900
6,500
0-7,140
640
0-640
0
0
Three-Tier (Scenario 3)
BAT Option
Alt. ELG
Options
. 6,480
5,300
1,570-6,480
560
0-2,350
610
0-2,560
5,910
0-6,480
560
0-560
0
0
Source: USEPA. Impacts are shown for the BAT Option: Option 3 (beef, heifers, and dairy operations) and Option
5 (veal operations). See Table 4-1 for definitions of model regions and sizes. Results do not show estimated
impacts to designated operations.
experience financial stress as a result of the proposed regulations (Table 8-21). Even without
assumptions of cost passthrough, EPA's analysis shows that stress impacts would not be
experienced by a substantial number of operations, as compared to the total number of affected
confinement operations in these sectors.
8-47
-------
Based on these results, EPA proposes that the proposed CAFO regulations are
economically achievable under the co-proposed scenarios; Section 5 provides additional
information that compares the co-proposed scenarios with other alternative scenarios.
Tables 8-22 through 8-24 present a more detailed breakout of EPA's affordability results
under the proposed BAT Option by model CAFO type, land availability, and type of operation
(beef, veal, heifer, and dairy). The results are the same for the two-tier and three-tier structure
because only the numbers of CAFOs represented by each model type changes. The impacts are
presented by model CAFO and indicate the level of impact under each of the economic
affordability criteria. These results reflect a "zero" cost passthrough assumption.
These tables show that the financial stress impacts at beef operations in the Medium 1 and
Medium 2 models for certain land availability categories in the CE region are being driven by the
revenue test and cash flow criterion (revenue test of greater than 10 percent is considered an
indication of financial stress). Heifer stress impacts are being driven by the revenue test and the
discounted cash flow criterion in the Medium 1 model, MW region, for certain land availability
categories. Dairy stress impacts are being driven by the revenue test and/or the discounted cash
flow criterion in the Mediuml and Medium 2 models (all categories) in the PA region.
8.4 PROCESSOR ANALYSIS
EPA does not evaluate the potential costs to cattle and dairy processors because EPA
does not expect that the proposed co-permitting requirements to affect meat packing and
processing facilities in these industries, for reasons outlined in Section 2. A brief summary of the
basis for EPA's assumption is provided in Section 8.2. As discussed, EPA's determination is
based on the fact that production contracting accounts for only a small share of beef and milk
production (USDA/ERS, 1999a and 1996c; Heffeman, et al., 1999). Also, animal ownership on
beef and dairy farms is mostly by the farm operator (USDA/ERS, 1999a).
8.5 MARKET ANALYSIS
This section presents the results of EPA's market model analysis for the beef and dairy
sectors. The results presented in this section briefly compare the results of the two-tier (500 AU
threshold) and the three-tier (Scenario 3) structures that are being co-proposed by EPA. These
results measure changes for the beef and dairy industries and do not differentiate between the
types of operations within each sector. Additional results on the alternative regulatory options
and scenarios considered by EPA as part of this rulemakmg are provided in Section 5.4. For
further explanation of the market model and sources of the baseline input data, see Section" 4.4
and Appendix B.
8-48
-------
Table 8-22. Economic Achievability Results for Beef/Heifer CAFOs (Option 3) and Veal CAFOs (Option 5)
CAFO
Model
Category 1
Sales
DCF
DA
Category 2
Sales
DCF
DA
Category 3
Sales '
DCF
DA
Beef
MWRegion
Small
Medium 1
Medium 2
Large 1
Large 2
CE Region
Medium 1
Medium 2
Large 1-
Large2
7.5%
5.5%
3.9%
1.0%
0.5%
6.3%
4.8%
1.1%
0.5%
Pass
Pass
Pass
Pass
Pass
Pass
. Pass
Pass
Pass
0.11
0.16
0.16
0.10
0.10
0.19
0.19
0.10
0.10 .
NA
6.0%
4.4%
1.8%
0.9%
8.2%
6.2%
2.1%
1.4%
• NA
Pass
Pass
Pass
Pass
Pass
Pass
Pass
Pass
NA
0.17
0.16-
0.10
, 0.10
0.20
0.19
0.10
0.10
NA
3.0%
2.1%
0.8%
0.4%
2.2%
1.4%
0.4%
0.2%
'NA
Pass
Pass
Pass
Pass
Pass
Pass
Pass
Pass
NA
0.15
0.15
0.10
0.10
•0.18
0.18
0.10
0.10
Veal
MWRegion -
Medium 1
Medium 2
1.3%
0.7%
Pass
Pass
0.13
0.13
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Heifer
MWRegion
Mediuml
Medium 2
Large 1
4.5%
3.0%
0.7%
Pass
Pass
Pass
CE Region
Medium 1
Medium 2
Large 1
4.6%
3.1%
0.6%
Pass
Pass
Pass
0.16
0.15
0.10
5.1%
3.5%
0.8%
Pass
Pass
Pass
0.18
0.19
0.10
5.5%
3.5%
0.9%
Pass
Pass
Pass
0.16
0.15
0.10
0.19
0.19
0.10
2.6%
1.8%
0.6%
. 2.1%
1.4%
0.4%
Pass
Pass
Pass
0.15
0.15
0.10
Pass
Pass
Pass
0.18
0.18
0.10
Source: USEPA.
8-49
-------
Table 8-23. Economic Achievabilifv Results for Dairy CAFOs (Option 3)
CAFO
Model
PA Region
Medium 1
Medium 2
Large 1
MW Region
Small
Medium 1
Medium 2
T sycpt* 1
Category 1
Sales
2.1%
1.4%
0.8%
2.3%
2.4%
1.6%
1 0%
DCF
DA
Category 2
Sales
DCF
Pass
Pass
Pass
Pass
Pass
Pass
Pass
0.27
0.26
0.26
0.27
0.26
0.25
030
2.3%
2.0%
1.6%
NA
2.5%
2.0%
1.5%
Pass
Pass
Pass
NA
Pass
Pass
Pass
DA
0.27
0.26
0.26
NA
0.26
0.25
0.30
Sales
1.7%
1.1%
0.7%
NA
2.0%
1.4%
0.9%
Category 3
DCF
Pass
Pass
Pass
NA
Pass
Pass '
Pass
DA
0.27
0.26
0.26
NA
0.26
0.25
0.30
Source: USEPA.
A summary of the key results of the market model is shown in Tables 8-25 and 8-26 for
the two-tier and three-tier structures indicating the predicted changes in farm and retail prices,
quantities, national and regional employment, and national economic output.
Compared to a baseline producer price of $66.09 per hundredweight (cwt), EPA's market
model predicts that the proposed CAFO regulations will raise producer cattle prices by $0.21 per
cwt to $0.23 per cwt, or less than 0.34 percent of the baseline producer price, depending on the
co-proposed tier structure (Table 8-25). Raw milk prices will rise by under 10 cents per cwt from
the baseline price of $13.38 per cwt. (All prices are in 1997 dollars.) At the retail level,
consumer prices for beef products will rise less than one cent per pound. The retail dairy-product
price index rises by 0.61 to 0.78. These price increases are driven by slight changes in the amount
produced at the farm level and thus available for consumption (Tables 8-25 and 8-26). At the
retail commodity level, EPA's market model predicts that U.S. beef imports will rise by
0.2 percent, compared to baseline imports; U.S. beef exports will decrease by 0.1 percent
compared to baseline. U.S. dairy product imports rise by 0.8 percent compared to baseline; U.S.
dairy product exports decrease by 0.1 percent compared to baseline exports.
8-50
-------
Table 8-24. Economic Achievability Results for Beef, Heifer, and Dairy CAFOs (Option 3A)
CAFO
Model
Category 1
Sales
DCF
• DA
Category 2
Sales
DCF
DA
Sales
Category 3
DCF
DA
Beef
MW Region
Small
ifediuml
vledium2
Large 1
Jarge2
CE Region
Medium 1
Medium 2
Large 1
Large 2
7.5%
9.1%
6.4%
2.3%
1.3%
10.7%
7.8%
2.5%
1.3%
Pass
Pass
Pass
Pass
Pass
Fail
Pass
Pass
Pass
0.13
0.21
0.19
0.12
0.11
NA
9.5%
6.8%
2.9%
1.5%
0.25
0.24
0.12
0.11
12.1%
8.8%
. 3.2%
1.9%
NA
Pass
Pass
Pass
Pass
Fail
Fafl
Pass
Pass
NA
0.21
0.20
. 0.12
0.11
0.25
0.24
0.12
0.11
NA
7.2%
5.2%
2.2%
1.2%
7.6%
5.4%
2.0%
1.1%
NA
Pass '.
' Pass
Pass
Pass
Pass
Pass
Pass
Pass
NA
0.20
0.19
0.12
0.11
0.23
0.22
0.12
0.11
Heifer
MW Region
Medium 1
Medium 2
Large 1
CE Region
Medium 1
Medium 2
Large 1
10.0%
6.9%
2.2%
11.6%
8.0%
2.8%
Pass
Pass
Pass
Fail
Pass
Pass
0.19
0.18
. 0.11
10.4%
7.4%
2.3%
Pass
Pass
Pass
0.24 .
0.22
0.11
12.1%
8.8% .
3.2%
Fail
Pass
Pass
0.19
0.18
0.11
0.24
.0.22
0.11
8.3%
5.7%
2.1%
9.3%
6.4%
2.6%
Pass
Pass
Pass
Pass
Pass
Pass
0.18
0.17
0.11
0.22
0.21
0.11
Dairy
PA Region
Medium 1
Medium 2
Large 1
9.8%
6.4%
3.9%
Fail
Fail
Pass
0.42
0.38
0.36
10.0%
7.0%
,4.7%
Fail
Fail
Pass
0.42
0.38
0.36
9.5%
6.i%
3.8%
Fail
Fail
Pass
0.41
0.38
0.36
MW Region
Small
Medium 1
Medium 2
Large 1
8.7%
7.4%
5.0%
34%
Pass
Pass
Pass
Pass
0.36
0.34
0.31.
0.37
NA
7.5%
5.5%
3.8%
NA
Pass
Pass
Pass
NA
0.34
0.31
0.37
NA
7.1%
4.8% .
3.3%
NA
Pass
Pass
Pass
NA
0.34
0.31
0.37
Source: USEPA.
8-51
-------
Table 8-25. Summary of Market Model Results for the Beef Sector
Variable
Pre-
Regulatory
Value/Units
Two-Tier Structure
BAT
Option
Range of
Alternative Options
Three-Tier Structure
BAT
Option
Range of
Alternative Options
- Farm Products
Price
Quantity ^
Produced
Quantity
Exported
Quantity
Imported
$66.09/cwt
47,967
mil. Ibs.
331
mil. Ibs.
2,400
mil. Ibs.
$66.30
47,902
330
2,408
$66.17 - 66.89
47,724 - 47,944
328-331
2,403 - 2,429
$66.32
47,898
330
2,408
$66.17 - 66.91
47,717 - 47,943
328-331
2,400 - 2,410
Retail Products
Price
Quantity
Demanded
Quantity
Exported
Quantity
Imported
$2.80/lb.
25,824
mil. Ibs.
2,136
mil. Ibs.
2,343
mil. Ibs.
$2.80
25,795
2,134
2,347
$2.80 - 2.82
25,715-25,814
2,129 - 2,135
2,344-2,356
$2.80
25,793
2,134
2,347
$2.80 - 2.82
25,712 - 25,813
2,129-2,135
2,344 - 2,356
Employment Reduction b/
Direct Farm
Direct Processor
Total Economy
336,700 FTEs
145,617 FTEs
129.6 mil.
FTEs
793
109
4,599
284 - 2,969
39-410
1,648-17,218.
850
117
4,929
294 - 3,045
41 - 421
1,704 - 17,661
Output Reduction
National
$ million
458
164-1,715
491
170-1,760
Regional Farm and Processing Employment Reduction
Pacific
Central
Midwest
South
Mid-Atlantic
Total
FTEs
FTEs
FTEs
FTEs
FTEs
FTEs
49
409
439
1
3
902
. 18-184
147 - 1,533
157 - 1,646
0-3
1-13
323 - 3,379
53
439
471
1
4
967
18-189
152-1,573
163 - 1,688
0-3
1-13
334 . 3,466
Source: Post-regulatory changes are estimated by USEPA. Pre-regulatory prices, quantities, and trade volumes,
see Table 4-16 (Section 4). Pre-regulatory employment, see Table 2-17 (Section 2).
07 Includes veal and heifer.
wl FTE = 2,080 hours of labor.
8-52
-------
Table 8-2fi Dairv Summary of Market Model Results for the Dairy Sector
Variable
Price
Quantity
Produced
Price
Quantity
Demanded
Quantity
Exported
Quantity
Imported
Direct
Farm
Direct
Processor
Total
Economy
National
Pacific
Central
Midwest
South
Mid-
Atlantic
Total
Pre-Regulatory
Value/Units
Two-Tier Structure
BAT
Option.
Range of
Alternative Options
Three-Tier Structure
BAT
Option
Range of
Alternative Options
Farm Products
$13.38/cwt
156,100
mil. Ibs.
13.44
155,907
$13.41 - 13.45
155,883-155,994
13.46
155,852
Retail Products
145.50 Index
156,100
mil. Ibs.
5,244
mil. Ibs.
4,383
mil. Ibs.
146.11
155,907
5,239
4,411
145.83 - 146.18
155,883 - 155,994
5,238 - 5,241
4,398 - 4,414
146.28
155,852
5,237
4,419
$13.42 - 13.48
155,792 -"155,975
145.89 - 146.47
155,792 - 155,975
5,235 - 5,241
4,401 - 4,427
Employment Reduction ^
483,800
FTEs
141,400
FTEs
129.6 mil.
FTEs
492
19
3,200
270 - 554
11-22
1,754-3,604
633
25
4,117
319 - 784
13-31
2,075 - 5,099
Output Reduction
$ million
296
162 - 333
381
Regional Farm and Processing Employment Reduction
FTEs
FTEs
FTEs
FTEs
FTEs
FTEs
262
152
30
35
32
512
144 - 295
83 - 171
17-34
19-39
18-37
280 - 576
337
195
39
45
42
658
192 - 472
170 - 418
98 - 241
20 - 49
23 - 56
21-52
332-815
Source: Post-regulatory changes are estimated by USEPA. Pre-regulatory prices, quantities, and trade volumes, see
Table 4-16 (Section 4). Pre-regulatory employment, see Table 2-17 (Section 2).
a/lFTE = 2,080 hours of labor. ,
Absorption of compliance costs by the producers and small declines in quantities are
expected to result in fewer jobs in the beef and dairy industry. Tables 8-25 and 8-26 also present
EPA's estimates of both the direct (i.e., farm and processor level) and total (i.e., national level)
8-53
-------
reductions in employment for the beef and dairy sectors. Overall, changes in national aggregate
employment in the beef sector are estimated to range from a total reduction of 4,600 to 4,930
jobs, measured in full-time equivalents (FTEs). EPA estimates employment losses in the dairy
sector at 3,200 to 4,120 FTEs, depending on tier structure. This analysis also does not adjust for
offsetting increases in other parts of the economy and other sector'employment that may be
stimulated as a result of the proposed regulations, such as the construction and farm services
sectors.
EPA's estimated job losses are estimated throughout the entire economy, using available
modeling approaches described in Section 4, and are not attributable to the regulated community
only. As shown in Tables 8-25 and 8-26, about 80 percent of these estimated job losses are in the
non-agricultural or farm services support industries (i.e., indirect or induced employment affects;
see Section 4.4). , •
At the CAFO level, EPA predicts that job losses in the cattle production sector associated
with the proposed CAFO regulations will range from 790 to 850 jobs under the proposed BAT
Option, depending on tier structure (Table 8-25). Job losses in the dairy farming sector will range
from 490 to 630 jobs. These estimates include CAFO owner-operators and employed family
members, as well as hired farm labor. This estimated reduction compares to an estimated total
farm level employment of 336,700 FTEs in the beef sector and 483,800 FTEs in the dairy sector
nationwide (Table 2-17; Abel, Daft, and Barley, 1993, as updated by EPA). EPA estimates that
job losses in the beef processing sectors will range from 110 to 120 (Table 8-25). In the dairy
processing sector, EPA estimates 20 to 25 jobs (Table 8-26) will be lost. These estimated losses
compare to the more than 145,000 persons employed in beef processing and 141,000 in dairy
processing in 1997 (USDC, 1999a). .
. Changes in employment and earnings can affect the vitality of local communities.
Community impacts are usually determined by employment changes at individual facilities. As
facility-specific information and analysis were not within the scope of this study, EPA is not able
to speculate on community impacts. However, EPA disaggregates the national employment
results to examine the potential regional employment impacts of the proposed CAFO regulations.
The method EPA uses to allocate estimated national level impacts is based on production shares
across states and does not take into account existing environmental practices or other production
factors (see Section 4.4). Table 8-25 shows that the traditional cattle production regions of the
Midwest would be the most affected, followed closely by the Central region. None of the impacts
represent a significant share of total employment in these regions. Compared to the baseline, EPA
estimates the loss in beef agricultural employment at less than 0.02 percent of total regional
employment; about half of the estimated agricultural job losses in the beef sector are expected in
the Midwest region (Table 8-25). Table 8-26 shows that the results of EPA's analysis indicate
that the more recently developed dairy operations in the Pacific region will be most affected,
followed by operations in the Central region. The loss in dairy agricultural employment is
estimated at less than 0.01 percent of total regional employment; about half of the estimated
agricultural job losses in the dairy sector are expected in the Pacific region (Table 8-26).
Economy-wide employment losses are estimated at less than 0.015 percent for both sectors
compared to the baseline.
8-54
-------
SECTION NINE '
INITIAL REGULATORY FLEXIBILITY ANALYSIS
9.1 THE REGULATORY FLEXIBILITY ACT (RFA) AS AMENDED BY THE
SMALL BUSINESS REGULATORY ENFORCEMENT FAIRNESS ACT
(SBREFA)
This section considers the effects that the proposed CAFO regulations may have on small
. livestock and poultry operations as required by the Regulatory Flexibility Act (RFA, 5 U.S.C et
seq., Public Law 96-354) as amended by the Small Business Regulatory Enforcement Fairness Act
of 1996 (SBREFA), The purpose of the RFA is to establish as a principle of regulation that
agencies should tailor regulatory and informational requirements to the size of entities, consistent
with the objectives of a particular regulation and applicable statutes. The RFA generally requires
an agency to prepare a regulatory flexibility analysis of any rule subject to notice and comment
rulemaking requirements under the Administrative Procedure Act or any other statute unless the
agency certifies that the rule will not have a "significant impact on a substantial number of small
entities."1 Small entities include small businesses, small organizations, and governmental
jurisdictions.
For this proposed rulemaking, EPA could not conclude that costs are sufficiently low to
justify "certification." Instead, EPA complied with all RFA provisions and conducted outreach to
small businesses, convened a Small Business Advocacy Review (SBAR) panel, and prepared an
initial regulatory flexibility analysis (IRFA).2 This analysis is detailed in this section and represents
EPA's assessment of the impacts of the proposed CAFO regulations on small businesses in the
livestock and poultry sectors. Section 9.2 outlines EPA's initial assessment of small businesses in
the sectors affected by the proposed regulations. Section 9.3 presents EPA's analysis (IRFA) and
summarizes the steps taken by EPA to comply with the RFA. Section 9.4 presents the data,
methodology, and results of EPA's analysis of impacts to small businesses for this rulemaking.
9.2 INITIAL ASSESSMENT
EPA guidance on implementing RFA requirements suggests the following must be
addressed in an initial assessment (USEPA, 1999i). First, EPA must indicate whether the
proposal is a rule subject to notice-and-comment rulemaking requirements. EPA has determined
1 The preparation of an IRFA for a proposed rule does not legally foreclose certifying no significant
impact for the final rule (USEPA, 1999i).
2This analysis or a summary of the analysis, must be published in the Federal Register at the time of
publication of a proposal.
9-1
-------
that the proposed CAFO regulations are subject to notice-and-comment rulemaking requirements.
Second, EPA should develop a profile of the affected small entities. EPA has developed a profile
of the livestock and poultry sectors, which includes all affected operations as well as small
businesses. This information is provided in Section 2-and also in Sections 6, 7, and 8 of this EA.
Much of the profile information covered in these sections of this report applies to small
businesses. Additional information on small businesses hi the livestock and poultry sectors is
provided in Sections 9.2 and 9.3. Third, EPA's assessment needs to determine whether the rule
would affect small entities and whether the rule would have an adverse economic impact on small
entities.
Section 9.2.1 reviews the SBA definitions of small entities in the livestock and poultry
industry and discusses a rationale for using an alternative definition of small business in one
sector. Section 9.2.2 then uses the definitions of small entities laid out in Section 9.2.1 to
estimate the number of operations that meet this small business definition. Finally, using the
information developed in Sections 9.2.1 and 9.2.2, Section 9.2.3 presents the results of EPA's
initial assessment. This assessment provides a first level screen of potential impacts to small
CAFO businesses and serves as a signal for additional analysis.
9.2.1 Definition of Small CAFO Businesses
The RFA defines a "small entity" as a small not-for-profit organization, small
governmental jurisdiction, or small business. There are no small governmental operations that
operate CAFOs. There may be a few not-for-profit organizations that operate CAFOs, but
complete information is not available to warrant inclusion of not-for-profit organizations in this
analysis. This analysis therefore focuses only on small businesses that are defined or designated as
CAFOs. (Section 3 describes the circumstances under which an AFO is defined or designated as
a CAFO and is subject to the proposed regulations.) ,
The RFA requires, with some exception, that EPA define small businesses according to its
size standards. SBA sets size standards for defining small businesses by number of employees or
amount of revenues for specific industries. These size standards vary by North American Industry
Classification System (NAICS) code. CAFOs are listed under NAICS 112, Animal Production.3
SBA's size standards differ from the revenue cutoff generally recognized by USD A, which
has defined $250,000 hi gross sales as its cutoff between small and large family farms (USDA,
1998).
3 In September, 2000, SBA updated the basis for its size standard to NAICS codes from Standard
Industrial Classification (SIC) codes (USGPO, 2000). By SIC code, these industries are listed under SIC 02,
Livestock and Animal Specialties. The actual size standards for each sector, specified as an annual revenue
threshold, did not change as a result of this update.
9-2
-------
Table 9-1 shows SB A size standards by SIC code for each of the six livestock and poultry
sectors, which are expressed in terms of average "annual receipts" (revenue). With one
exception, current SB A standards define a "small business" within each of the main livestock and
poultry sectors as an operation that generates average revenues ranging from less than $0.5
million per year (for the hog, dairy, broiler, and turkey sectors) to less than $1.5 million per year
(for the beef feedlot sector), averaged over the most recent three fiscal years (USGPO, 1996;
SB A, 1998). The exception is the revenue threshold for a small chicken egg operation (layer
sector), which SBA has defined as a business that generates up to $9 million annually.
Table 9-1. SBA Revenue Size Standards for Small Livestock and Poultry Operations
NAICS Code
(SIC Code)
112112(0211)
11221(0213)
11212(0241)
11232(0251)
11231(0252)
112330253
NAICS
Industry Description
Cattle Feedlots
Hog and pig fanning
Dairy cattle and milk production
Broilers and other meat-type chickens
Chicken egg production
Turkey production
SBA Size
Standard ^
$1.5 million
$0.5 million
$0.5 million
$0.5 million.
$9.0 million
$0.5 million
EPA-Proposed
Revenue Cutoff
same as SBA
same as SBA
same as SBA
same as SBA
$1.5 million
same as SBA
Source: SBA (1998); USGPO (1991a, 1991b and 1996); U.S. Census Bureau (2000).
Size Standards by NAICS code (13 CFR Part 121) correspond to classifications under
SIC classification.
EPA believes that the definition of small business for the egg laying sector (revenues of $9
million per year) might not truly characterize a small business in this sector. Therefore, EPA. is
proposing to use an alternative definition, as allowed by the RFA:
"...an agency, after consultation with the Office of Advocacy of the Small Business
Administration and after the opportunity for public comment, establishes one or more
definitions of such term which are appropriate to the activities of the agency and publishes
such definitions) in the Federal Register." 5 U.S.C. §601(3).
EPA's alternative definition identifies a small business for egg laying operations as any
operation that generates up to $1.5 million in annual revenue (see Table 9-1). Because this
definition of a small business is not the definition established under the RFA, EPA is specifically
seeking comment on the use of this alternative definition. EPA has also consulted with the SBA
Chief Counsel for Advocacy on the use of this alternative definition (USEPA, 1999d). EPA
believes this definition better reflects the agricultural community's sense of what constitutes a
small business and more closely aligns with the small business definitions codified by SBA for
other animal operations. .
9-3
-------
There are four broad reasons why EPA believes that its alternative definition of small egg
laying operations is more appropriate for the purpose of this rulemaking. These include: (1)
EPA's definition is more consistent with size classes used by USDA and industry; (2) EPA's
definition reflects the financial and institutional realities of the egg industry; (3) EPA's definition
reflects similarities among the sectors of the poultry industry; and (4) EPA's definition captures
the relevant segments of the industry (USEPA, 1999d). The four reasons for using the alternative
definition of small egg laying operations are summarized below. Additional supporting data and
analysis are provided in the rulemaking Record (USEPA, 1999d; USEPA, 2000f).
First, EPA's alternative definition is more consistent with size classes used by USDA
(Madison, 1999) and more generally accepted by the regulated community (Gregory, 1999;
Staples, 1998). USDA describes size classes reflective of farm level conditions at egg laying
operations in terms of the number of houses, where a house has approximately 100,000 to
110,000 hens. Based on USDA's size classes, a small farm has a single house; a medium farm has
two to five houses; and a large farm has more tihtan five houses (i.e., more than 500,000 hens).
Using USDA data, EPA estimates that a "small" egg operation by USDA standards generates
approximately $1.5 million in annual revenue (USEPA, 1999d and 2000f).4
In contrast, a definition of $9 million in annual revenue fails to reflect farm level conditions
based on USDA size classes and matching opinions from the farming community. Such an
operation corresponds to an operation with more than six houses (with approximately 600,000
hens). EPA does not believe an operation with six chicken houses should be characterized as
"small" for the proposed CAFO regulations. EPA visited one such facility. The facility resides on
more than 200 acres and has an annual production of over 180 million eggs. The facility's
extensive customer base includes three major supermarket chains and the U.S. military. Its
distribution system spans four states. A facility with such a high production level and extensive
customer base is not a small business.5 EPA's alternative definition would decrease confusion and
facilitate communication with the regulated community (both large and small businesses) and with
other stakeholders.
Second, EPA's alternative definition better reflects the financial and institutional realities
of the egg industry. EPA focuses its regulatory analyses for the proposed CAFO regulations at
the animal production level since it is the operator who directly incurs all costs associated with the
management and disposal of manure generated from animals that are raised or housed onsite.
EPA believes, based on a preliminary review of the background information supporting the SBA
definition (USGPO, 1991a and 1991b) that the $9 million definition applies to entities at a
different level in the marketing chain—e.g., to large cooperatives or integrators, rather than farms.
The alternative definition would allow EPA to better focus on the needs and concerns of those
4EPA estimates are derived using USDA-reported 1997 data: average yield of 255 eggs per layer per year
(USDA/NASS, 1998b) and average annual producer price of 66.7 cents per dozen (USDA/NASS, 1998a).
Information on EPA's farm site visits is in the rulemaking record.
. 9-4 ,
-------
businesses that are most likely to experience economic hardship associated with regulatory
compliance.
Third, EPA's alternative definition better reflects similarities among the sectors of the
poultry industry. EPA's analysis focuses on three sectors: egg laying, broiler, and turkey meat.
The SBA definitions differ substantially between the egg laying sector and the other two sectors.
As shown in Table 9-1, the small-business definition for layer operations is $9 million in annual
revenue; the small-business definition for both broiler and turkey operations is $0.5 million. At
the farm level, however, there are structural similarities among these three sectors, suggesting that
small business definitions should not be so disparate for these operations. The sectors use similar
technologies and similar manure management techniques. They have similar costs of production.
They have similar industrial organization and marketing arrangements. Measured at the animal
production level, the SBA definition of a small broiler or turkey operation is consistent with
USDA's definition of a small- or medium-sized operation (based on the number of animals and
housing structures, as discussed above).
In fact, prior to 1991, the SBA definition for layer operations was much closer to the
definitions for the other two poultry sectors. The earlier SBA definition for layer operations was
$1.0 million. The definition was revised to $7 million in 1991, and then escalated to $9 million to
account for inflationary changes (USGPO, 1991a and 1991b; Ray, 1999). One of the reasons
cited for the 1991 increase was the "limited participation of small egg producers in government
procurement" (USGPO, 1991a). For the regulatory flexibility assessment of the proposed CAFO
regulations, EPA concludes that the alternative definition is more comparable to the definitions
for other livestock sectors and is therefore more appropriate than the existing definition.
Finally, EPA's alternative definition is more appropriate in terms of capturing the relevant
segments of the industry. Under EPA's alternative definition, small layer operations would
account for roughly 60 percent of annual egg production (USEPA, 2000f). In contrast, under
SBA's definition, small operations would account for approximately 90 percent of annual egg
production. If EPA were to use SBA's definition, a very large share of total annual egg output
would be generated from "small" operations. This would be inconsistent with the analysis of the
broiler and turkey sector, where smaller operations represent roughly one-half of each sector's
respective annual production. This would further contradict expectations by SBA in terms of the
percent of sales attributable to small operations. According to SBA, about 99 percent of all farms
in the economy are small and account for approximately 62 percent of sales (Perez, 2000;
USEPA, 2000g). This agrees with the realities of the agricultural sector where the majority of
farms are small, but account for a relatively small share of overall production. The trend in
agriculture towards fewer, larger farms highlights that larger operations—while relatively few in
number—represent a greater share of overall output.
EPA also considered another alternative definition for all six animal sectors based on the
number of animals raised or housed at the CAFO site (USEPA, 2000e, 1999a, 19991, and 1999n).
Following discussions with representatives from both SBA and OMB during the SBAR Panel
9-5
-------
process, EPA decided not to use this alternative definition for each of the animal sectors (USEPA,
2000g). A complete summary of EPA's correspondence with SBA on its proposal and use of an
alternative definition is contained in the rulemaking record (see DCN 70509, DCN 70507, DCN
70473, DCN 70472, DCN 70511, DCN 70797, and DCN 93001).
9.2.2 Number of Small Businesses Affected by the Proposed CAFO Regulations
There are three steps for determining the number of small CAFO businesses that may be
affected by the proposed regulations. First, EPA identifies small businesses in the relevant
livestock and poultry sectors by equating SBA's annual revenue definition with the number of
animals at an operation. Second, EPA estimates the total number of small businesses in these
sectors using farm size distribution data from USD A. Third, based on the regulatory thresholds
being proposed, EPA estimates the number of small businesses that would be subject to the
proposed requirements. These steps are described in the following sections.
9.2.2.1 Equating SBA Size Standards with Animal Inventory
In the absence of entity level revenue data, EPA identifies small businesses in the livestock
and poultry sectors by equating SBA's annual revenue definitions of "small business" to the of
number of animals at these operations (step 1). This step produces a threshold based on the
number of animals that EPA uses to define small livestock and poultry operations and reflects the
average farm inventory (number of animals) that would be expected at an operation with annual
revenues that define a small business. This initial conversion is necessary because USDA data by
farm size are not available by business revenue. With the exception of egg laying operations, EPA
uses SBA's small business definition to equate the revenue threshold with the number of animals
raised on site at an equivalent small business in each sector. For egg laying operations, EPA's
alternative revenue definition of small business is used.
EPA estimates the number of animals at an operation to match SBA's definitions using
SBA's annual revenue size standard (expressed as annual revenue per entity) and USDA-
reported farm revenue data that are scaled on a per-animal basis (expressed as annual revenue per
inventory animal for an average facility). (This calculation is shown below.) Per-animal financial
data are calculated by multiplying the average value of the reported financial data per farm by the
total number of farms and then dividing this by the total number of animals. (More information
on this calculation is presented in Section 4.2.4.2 of this report.) The average per-animal
revenues assumed for this analysis are shown in Table 9-2.
Financial data used by EPA are from the USDA's 1997 ARMS database. These data
include farm financial data and corresponding summary information that match the reported
average revenue to the total number of farms and the total number of animals in the sample set.
9-6
-------
Table 9-2. Number of Small CAFOs That May Be Affected by the Proposed Regulations
Sector
Cattle "
Dairy
Hogs
Broilers
Egg Layers
Turkeys
AH AFOs *
Total
Annual
(Smillion)
Revenue "
(x)
$1.5
$0.5
$0.5
$0.5
$9.0
$1.5
$0.5
NA
Revenue
per
Head b/
(Avg. U.S.)
(y)
$1,060
$2,573
$363
$2
$25
$20
NA
Number
of Animals at
Small CAFO
Businesses
(z=x/y)
1,400
200
' 1,400
260,000
365,000
61,000
25,000
NA
Estimated
Number of
Small
AFOs
106,450
109,740
107,880
34,530
ND
73,710
12,320
355,650
Two-Tier
(500 AU)
"Small" CAFO
Businesses
2,280
50
300
9,470
ND
200
0
10,550
Three-Tier
"Small"
CAFO
Businesses
2,600
50
300
13,410
ND
590
500
14,630
NA=Not Applicable. ND = Not Determined. "AFOs" have confined animals on-site.
"'SBA Size Standards are at 13 CFR Part 121. EPA assumes an alternative definition of $1.5 million in annual
revenues for egg layers.
b/Average revenue per head across all operations for each sector derived from data obtained from USDA's 1997
ARMS data (USDA/ERS, 1999a). See Section 4. •
^ Includes fed cattle, veal and heifers.
^Total adjusts for operations with mixed animal types and includes designated CAFOs (expressed over a 10-year
period). See Section 2 of this document for estimates of the total number of AFOs.
These data were obtained with the assistance of staff at USDA's ERS (as described in Section
4.2.3.2).6 USDA's data report average national revenue for each sector, combining both livestock
and nonlivestock farm revenue (income from crop sales and other farm-related income, including
government payments). Use of total farm revenue corresponds to SBA's size standards that are
expressed in terms of total annual business revenue (SBA, 1998; USGPO, 2QOO).
EPA uses the derived per-animal revenues shown in Table 9-2 to equate SBA's size
standard (in revenues) with farm size based on the number of animals, as follows:
Average # Animals =
Farm
SBA's Small Business Definition ($ per year per farm')
Average Total Revenue per head ($/animal)
6As noted throughout this report, USDA periodically publishes aggregated data from the ARMS and
Census databases and provides customized analyses of the data to members of the public and other government
agencies. In providing such analyses, USDA maintains a sufficient level of aggregation to ensure the
confidentiality of individual facility data.
9-7
-------
The resultant number of animals represents the average animal inventory threshold for a small
business. Estimated "small business" thresholds for each sector are shown in Table 9-2.
For the purpose of conducting its IRFA for this rulemaking, and based on the animal
inventory thresholds discussed above, EPA is evaluating a "small business" for these sectors as an
animal feeding operation that houses or confines less than: 1,400 fed beef cattle; 200 mature dairy
cattle; 1,400 market hogs; 260,000 broilers; 61,000 layers; or 25,000 turkeys. Hereafter, all
references to small CAFO businesses reflect the SBA definitions of "small" and the alternative
definition proposed by EPA for small layer operations, applied on the basis of a calculated number
ofhead.
9.2.2.2 Total Number of Operations that Match SBA Size Standards
Using the threshold sizes identified for small businesses in the livestock and poultry
sectors (Table 9-2), EPA matches these thresholds with the number of operations associated with
those size thresholds, based on available USDA data, to estimate the total number of small animal
confinement operations in these sectors (step 2).
The 1997 Census constitutes the primary data source that EPA uses to match the small
business thresholds to the number of operations by size. Other supplemental data used includes
other published USDA data and information from industry and the state agriculture extension
agencies. In some cases, EPA extrapolated between two size groupings to obtain an estimate of
the number of small livestock and poultry operations. Additional information is also used to
subdivide sector level data into subsectors. For example, the number of hog operations that are
farrow-finish versus grow-finish are distinguished according to market share information
(USDA/APHIS, 1995b). Information that differentiates the number of egg laying operations
according to manure management system (wet versus dry) are approximated based on
conversations with State Extension personnel for selected states, as described in the Development
Document (USEPA, 2000a). The number of breeder and nursery pig operations and veal and
heifer operations are approximated based on information obtained from state extension personnel
and EPA's farm site visits (USEPA, 2000a).
For many of the animal sectors, it is not possible to estimate from available U.S. farm data
what proportion of total livestock and poultry operations have feedlots and what proportion are
grazing operations only. For the beef and hog sectors, the USDA has limited data on the number
of operations that are feedlot operations only (USDA/APHIS, 1995b; USDA/NASS, 1999a and
1999b). For analytical purposes, EPA has assumed that all dairy and poultry operations
potentially are confinement operations. More information on the farm size distribution data that
EPA uses to match the size thresholds to the number of poultry and livestock operations is
documented in the Development Document (USEPA, 2000a).
9-8
-------
Table 9-2 shows EPA's estimates of the total number of small livestock and poultry"
operations using this approach. As shown, an estimated 355,650 animal confinement operations
meet SBA's small business definition. This is 95 percent of the estimated total number of animal
feeding operations (375,700 operations).
EPA recognizes that this approach may not accurately portray actual small businesses in
all cases across all sectors. On the one hand, the resulting small business estimate would suggest
that a 10-house broiler operation with 260,000 birds would be a small business. Information from
industry sources, however, suggest that a two-house broiler operation with roughly 50,000 birds
is small (Madison, 1999; USEPA, 2000e). Therefore, it is likely that some medium- and large-
size broiler operations are being considered small businesses (USEPA, 2000g). '
On the other hand, it is possible that the resulting small business estimate may have failed
to identify some small businesses as "small" in the other sectors. For example, EPA's approach
identifies as a "small business" hog operations with less than 1,400 pigs and turkey operations
with less than 25,000 turkeys, which account for less than 94 percent of all operations and less
than 30 percent of sales in these sectors. These proportions are below SBA's presumed coverage
rates that define as small about 99 percent of all operations that account for approximately 62
percent of sales (Perez, 2000). Therefore, it is likely that there are additional small hog and
turkey businesses that are not captured under the revised methodology (USEPA, 2000g).
9.2.2.3 Total Number of Small CAFOs Subject to the Proposed Regulations
Based on the regulatory thresholds for each co-proposed alternative, EPA estimates the
number of small businesses that will be subject to the proposed requirements (step 3).7 The 1997
Census constitutes the primary data source that EPA uses to match the small business thresholds
(e.g., a small dairy operation has less than 200 milk cows) to the number of facilities that match
mat size group (e.g., the number of dairies with less than 200 cows, as reported by USD A).
Other supplemental data used include other published USDA data and information from industry
and the state extension agencies.
Table 9-2 shows the estimated total number of livestock and poultry operations that meet
the SB A definition of a "small business" in each of the livestock and poultry sectors. Not all of
small confinement operations would be subject to the proposed CAFO regulations, however.
EPA's proposed regulations only apply to those operations that meet the regulatory definition of a
CAFO or those that have been designated as a CAFO by the NPDES permitting authority due to
risks posed to water quality and public health, as discussed in Section 3. The proposed changes
define as a CAFO those operations that confine more than 300 or 500 AU (depending on co-
7In this section, EPA discusses numbers of affected CAFOs and impacts under the two-tier structure at
500 AU threshold (Scenario 4a) and three-tier structure (Scenario 3) only. "Two-tier structure" in this section
refers to the 500 AU,threshold, except where otherwise noted.
9-9
-------
proposed scenario). The proposed requirements may also apply to an operation that confines
fewer than 300 or 500 AU if it is designated as a CAFO by the NPDES permitting authority on a
case-by-case basis, based on an on-site inspection.
Of the estimated 355,650 animal confinement operations that meet SBA's small business
definition, EPA estimates that 10,550 operations that will be subject to the proposed requirements
that are small businesses under the two-tier structure. Under the three-tier structure, an estimated
14,630 affected operations are small businesses. These estimates include expected designated
facilities. The difference in the number of affected small businesses is among poultry producers,
particularly broiler operations. See Table 9-2.
Table 9-3 presents the estimated number of livestock and poultry operations that may be
subject to the proposed requirements under each co-proposed scenario that are also small
businesses ("small CAFO businesses") by facility size category. The number of small CAFO
businesses are shown as follows: (1) operations defined as CAFOs with more than 1,000 AU, (2)
operations defined as CAFOs with between 300 to 1,000 AU or 500 to 1,000 AU, depending on
scenario, and (3) operations that may be designated as CAFOs with fewer than 300 or 500 AU
that may be designated (varies by co-proposed alternative). The number of small CAFO
businesses in each of the three size categories is developed using the same data approach used to
identify the total number of small operations, discussed in Section 9.2.2.2.
Based on estimates shown in Table 9-3, EPA estimates that there are 10,220 operations
with more than 500 AU that may be defined as CAFOs that also meet the "small business"
definition, under the two-tier structure. Under the three-tier structure, there are 14,530
operations with more than 300 AU that may be defined as CAFOs that are small businesses that
meet the proposed risk-based conditions (described briefly in Section 3; more detail is provided in
Section VII of the preamble). By broad facility size group, EPA estimates that about 4,000
operations have more than 1,000 AU, adjusting for operations with more than a single animal
type. EPA estimates that about 6,000 operations have between 500 and 1,000 AU (two-tier
structure) and about 10,000 operations have between 300 and 1,000 AU (three-tier structure),
accounting for mixed operations. EPA's analysis assumes that all small businesses with 300 to
1,000 AU under the three-tier structure obtain a NPDES permit and that none certify out of the
program.
Among operations that are defined as CAFOs, depending on co-proposed scenario, most
small CAFO businesses are in the broiler and cattle sectors. As defined for this analysis, EPA
expects that there are no small CAFO businesses in the dairy sector with more than 300 AU (see
Section 9.2.2.1) and that small dairies will be subject to the regulations only if they are designated
as a CAFO by the Permitting Authority. Also, as defined for this analysis, there are no small
9-10
-------
Table 9-3. Total Number of Small CAFO Businesses Subject to Regulation
Sector
Fed Cattle
Veal
Heifers
Dairy
Hogs
Broilers
Layers
Turkeys
Sum Total
Total
AH "Small" AFOs
All
104,350
850
1,250
109,740
107,800
34,530
73,710
12,320
444,560
355,565
>1,000
AU
350
10
300
0
0
3,610
0
0
4,270
4,060
Two-Tier Structure
500-1,000
" AU
1,000
80
500
0
100
5,840
180
0
7,700
6,160
<500 AU
40
0
0
50
200
20
20
0
330
330
Total
1,390
90
800
50
300
9,470
200
0
12,300
10,550
Three-Tier Structure
300-1,000
AU '
.1,1.40
130
680
0
250
9,800
. 600
500
13,080
10,470
<300AU
0
0
0
50
50
0
0
0
100
100
Total
1,490
140
980
50
300
13,410
590
500
17,300
14,630
Sources: Values presented in the table are EPA estimates, derived from published USDA data, including 1997
Census of Agriculture (USDA/NASS, 1999a) supplemented with other data, as described in the Development
Document (USEPA, 2000a). All numbers are rounded to the nearest ten.
"Total" eliminates double counting of operations with mixed animal types. Based on survey level Census data,
operations with mixed animal types account for roughly 25 .percent of operations less than 1,000 AU; few
operations with more than 1,000 AU have more than a single animal type.
grow-finish hog operations that may be defined as CAFO under either co-proposed scenario; also,
there are no small CAFO businesses in the turkey sector under the two-tier structure (Table 9-3).
The majority (about 90 percent) of small confinement operations have fewer than 300 AU
(Table 9-3). EPA's total estimate of small affected CAFOs includes an additional 330 small
operations with fewer than 500 AU that may be designated as CAFOs under the two-tier structure
over a 10-year period (consistent with the 10-year time frame used for EPA's financial model).
As these facilities are designated, EPA did not adjust this total to reflect possible mixed animal
operations. All of these operations are small businesses. Under the two-tier structure, designated
operations are expected to consist of beef, dairy, hog, egg layer and broiler confinement
operations that are located in more traditional fanning regions and are determined to be significant
9-11
-------
contributors of pollution.8 Under the three-tier structure, EPA expects that 100 dairy and hog
operations will be designated as CAFO and, therefore, subject to the proposed regulations.
These estimates are based on farm data for 1997. Due to continued consolidation and
facility closure since 1997, EPA's estimates may overstate the actual number of small businesses
in these sectors. In addition, ongoing trends are causing some existing small- and medium-size
operations to expand their inventories to achieve scale economies. Some of the CAFOs
considered here as small businesses may no longer be counted as small businesses because they
now have higher revenues.
9.2.3 Results of the Initial Assessment
Early on in the development of this rulemaking, EPA conducted a preliminary assessment
of the potential impacts to small CAFO businesses based on the results of a costs-to-sales test for
operations with more than 500 AU. This screening test indicated the need for additional analysis
to characterize the nature and extent of impacts on small entities. This assessment is conducted
for those CAFOs that are small businesses, as determined by EPA.
Table 9-4 presents the results of this screening test and indicates that about 80 percent
(about 9,700) of the estimated number of small businesses with more than 500 AU that would be
directly subject to the rule as CAFOs (two-tier) may incur costs in excess of three percent of
sales. Compared to the total number of all small animal confinement facilities estimated by EPA
(355,650 facilities), EPA estimates that operations that may incur costs in excess of three percent
of sales comprise less man two percent of all small businesses in these sectors. (The cost and
revenue data EPA uses for this assessment are presented in Section 9.4; more detailed information
on these data is provided in Section 4 of this report.)
Based on the results of this initial assessment, EPA projected that the Agency would likely
not certify that the proposal, if promulgated, would not impose a significant economic impact on a
substantial number of entities. Therefore, EPA convened a Small Business Advocacy Review
Panel and prepared an Initial Regulatory Flexibility Analysis (TRFA) pursuant to Sections 609(b)
and 603 of the RFA, respectively, and prepared an economic analysis (see Sections 9.3 and 9.4).
SEPA expects that USDA will continue to provide voluntary assistance to those additional operations that
are now defined as CAFOs under the current permitting requirements (300 AU to 500 AU) that are not covered by
proposed CAFO revisions under the two-tier structure.
9-12
-------
Table 9-4. EPA's Preliminary Assessment of Small Business Impacts using a Sales Test
Sector
Fed Cattle
Veal
Heifers
Dairy
Hog-FF
Hog-GF
Broilers
Layers-Wet
Layers-Dry
Turkeys
Sum Total
Small AFOs
104,350
850
1,250
109,736
57,800
50,000
34,530
9,010
64,700
12,320
444,560
Small
CAFOs
(>500 AU)
1,350
90
800
0
100
0
9,450
20
160
0
11,970
Costs Exceed 3% of Revenues
#Small
CAFOs
80
10
20
0
20
0
9,450
0
0
0
9,580
%Small
CAFOS
6%
1%
3%
0%
20%
0%
100%
0%
0%
0%
80%
%SmalI AFOs
1%
1%
2%
0%
1%
0%
28%
0%
0%
0%
2%
Source: USEPA. Total does not adjust for operations with mixed animal types, for comparison purposes. Includes
CAFOs with more than 500 AU. Excludes designated operations. Sales test results are shown for the proposed
BAT Option and NPDES Scenario 4a (described in Section 3).-
9.3 EPA COMPLIANCE WITH RFA REQUIREMENTS
9.3.1 Outreach and Small Business Advocacy Review
As required by Section 609(b) of the RFA, as amended by SBREFA, EPA convened a
Small Business Advocacy Review (SBAR) Panel for the proposed rule. The Panel was convened
in December, 1999. Panel participants included representatives from EPA, the Office of
Information and Regulatory Affairs within the Office of Management and Budget (OMB), and the
Office of Advocacy of the Small Business Administration (SB A). "Small Entity Representatives"
(SERs), who advised the Panel, included small livestock and poultry producers as well as
representatives of the major commodity and agricultural trade associations. Throughout the
development of these regulations, EPA conducted outreach to small businesses in the livestock
and poultry sectors. EPA also consulted with SB A on the use of an alternative definition of small
business for the egg laying sector.
Consistent with the RFA/SBREFA requirements, the Panel evaluated the assembled
materials and small entity comments on issues related to the elements of the IRFA. The Panel's
9-13
-------
activities and recommendations are summarized in the Final Report of the Small Business
Advocacy Review Panel on EPA's Planned Proposed Rule on National Pollutant Discharge
Elimination System (NPDES) and Effluent Limitations Guideline (ELG) Regulations for
Concentrated Animal Feeding Operations (USEPA, 2000g), or "Panel Report." This document
is included in the public record (DCN 93001)., Section XII.G of the preamble provides a
summary of the Panel's activities and recommendations and describes the subsequent action taken
by the Agency. Section XTT of the preamble also details various outreach activities conducted by
EPA that include outreach to small businesses in these sectors.
9.3.2 EPA's Initial Regulatory Flexibility Analysis
As required by Section 603 of the RFA, as amended by SBREFA, EPA has conducted a
initial regulatory flexibility analysis. The IRFA must include a discussion of the reason the agency
is considering the proposed rale, as well as the objectives and legal basis for the proposal. It must
also include a description and estimate of the number of small businesses that will be affected. It
must describe the reporting, recordkeeping, and other compliance requirements of the proposed
rule and must identify any federal rules that may duplicate, "overlap, or conflict with the proposed
rule. Finally, the IRFA must describe any significant regulatory alternatives to the rule that would
accomplish the stated objectives of the applicable statutes and which minimize impacts to small
businesses. Sections 9.3.2.1 through 9.3.2.6 below address each of these requirements of the
IRFA that EPA has prepared to support the proposed CAFO regulations.
Section 607 of the RFA further notes that to comply with the IRFA requirements, the
Agency must "provide either a quantifiable or numerical description of the effects of a proposed
rule or alternatives to the proposed rule, or more general descriptive statements if quantification is
not practicable or-reliable." For this rulemaking, EPA has prepared an economic analysis of the
impacts to small CAFO businesses. This analysis is provided in Section 9.4. Based on the results
of this analysis, EPA has determined that the proposed regulations will result in financial stress to
some affected small businesses, but not a substantial number of operations relative to the total
number of affected small businesses in these sectors. Additional information and the detailed
results of this analysis are presented in Section 9.4.2.
9.3.2.1 Reason EPA is Considering the Proposed-Rule
Despite more than twenty years of regulation, there are persistent reports of discharge and
runoff of manure and manure nutrients from livestock and poultry operations. The proposed
revisions to the existing ELG and NPDES regulations for CAFOs are expected to mitigate future
water quality impairment and the associated human health and ecological risks by reducing
pollutant discharges from the animal production industry.
9-14
-------
EPA's proposed revisions also address the changes that have occurred in the animal
production industries in the United States since the development of the existing regulations. The
continued trend toward fewer but larger operations, coupled with greater emphasis on more
intensive production methods and specialization, is concentrating more manure nutrients and
other animal waste constituents within some geographic areas. This trend has coincided with
increased reports of large-scale discharges from these facilities and continued runoff that is
contributing to the significant increase in nutrients and resulting impairment of many U.S.
waterways.
EPA's proposed revisions of the existing regulations will make the regulations more
effective in protecting or restoring water quality. The revisions will also make the regulations
easier to understand and better clarify the conditions under which an AFO is a CAFO and,
therefore, subject to the regulatory requirements.
Additional information on why EPA is revising the existing regulations is provided in
Section IV of the preamble.
9.3.2.2 Objectives and Legal Basis for the Proposed Rule
A detailed discussion of the objectives and legal basis for the proposed CAFO regulations
is presented in Sections I and HI of the preamble.
9.3.2.3 Description and Estimate of Number of Small Entities Affected
As presented in Section 2, EPA estimates that there are about 375,700 livestock and
poultry operations nationwide of which 355,650 (95 percent) are small (Table 9-2). Of these, the
proposed CAFO regulations are expected to affect—and impose compliance costs
on—approximately 10,550 operations or 14,630 operations (Table 9-3), depending on co-
proposed scenario. Most (about 80 percent) of the estimated number of small CAFO businesses
are in the poultry sectors, with the majority in the broiler sector. The cattle sector accounts for
another 15 to 18 percent of small CAFO businesses, depending on tier structure. The remaining
number of affected small CAFO businesses are in the hog and dairy sectors.
Tables 9-5 and 9-6 show the numbers of affected small businesses by EPA's model CAFO
designation, which characterizes each of the small businesses by sector, size, and key production
region. (Values shown in the tables do not adjust for operations with more than a single animal
type.) These estimated CAFO numbers by model type are used to evaluate small business
impacts, presented in Section 9.4 of this report.
9-15
-------
Table 9-5. Numbers of Small CAFO Businesses by Sector, Size, and Region, Two-Tier Structure
Sector
Fed Cattle
Veal
Heifers
Dairy
Hog: FF
Hog: GF
Layer: Wet
Layer: Dry
Broiler
Turkey
Region
CE
MW
MW
MW
MW
PA
MA
MW
MA
MW
SO
MW
SO
MA
SO
MA
MW
Total
CAFOs
OOOAU
-3--."- ~
11 *
^, — -.«.
50
'
— ™"? ~~
50
\
,%-* 4-
-j »>*
- * -- - „
\
^~*- ~« »* *r
—
100
CAFOs
"Medium I"
**
40
t >• **
;•
~C*^C t.-v.'i
* . ~ "~
•» & ^
•v
» 1 1
150
^
* "f "*
~ „!">•' *~
*" *v"
,
- 1&^
740
1,280
^
2,210
CAFOs
"Medium 2"
160
840
80
500
* *""»*< *.
t ™ ^
* -^ sy~ s
^ * ^ j^ a
^ ^- ^^ <
f 4 -^ ^ ~* %
100
^
4^, -s. ^ >*•
i- - xJ_V -'
40
60
100
1,190
2,650
/• ^ •-•
>- "~ *
? ^ * 2
5,720
CAFOs
"Large 1"
70
280
10
300.
sp',. ^ i *
s * rf «^j
V*~~*j * "^^^
'Vj\^ -
980
2,300
*^'^^,
^,r , t ' v'1 *
3,940
CAFOs
"Large 2"
* '/"
* $• ^
~f
ff U^
> ^ .
t-'f^" \
* s '.
- ^, ' ? >
;H'
*-v - ^
» •* ^ s *
* " V~/ ~ -1 -•
r^ ". . » .>"-"
*. ftf t •**.
*•* \ * *^
*-* j ~^ „
^* j
''V . d f
}£ ? *" V .
* j- "- U...
^! /** >*^
{• - *,
<> i' X, ^ »
». »s r*r
f"^r**-X !
3f "-^ i
^ / f ' >-
JprS
, rtl" ' ,,S '
' *' , T "'
rf* ^ y ,
V
Wr *£r V
70
260
> » "* ""
\ '-
330
Source: USEPA. Size and region breakouts are based on 1997 Census data provided in the Development Document (USEPA,
2000a). Facility size and region definitions for model CAFOs are provided in Section 4, Table 4-1. Rounded to nearest ten.
Numbers do not adjust for mixed animal types and include expected designated CAFOs (<500 AU under two-tier and <300
under three-tier structure) are included in the counts and are shown over a 10-year period. Shaded cells indicate that there are
no small CAFO businesses that will be affected by the regulations that meet the SBA definition of a small business.
9-16
-------
Table 9-6. Numbers of Small CAFO Businesses by Sector, Size, and Region, Three-Tier Structure
Sector
Region
CAFOs
OOOAU
CAFOs
"Medium 1"
CAFOs
"Medium 2"
CAFOs
"Large 1"
CAFOs
"Large 2"
Fed Cattle
CE
20
160
70
MW
120
840
280
Veal
MW
50
80
10
Heifers
MW
180
500
300
Dairy
MW
50
PA
Hog: FF
MA
*..**, -;
MW
50
150
100
Hog: GF
MA
MW
•»> ~* v»^ \
Layer: Wet
SO
50
20
Layer: Dry
MW
130
60
SO
230
100
Broiler
MA
3210
1190
980
70
SO
2750
2650
2,300
260
Turkey
MA
320
MW
180
Total
100
7,390
5,700
3,940
330
Source: USEPA. See Table 9-5.
9.3.2.4 Description of the Proposed Reporting, Recordkeeping, and Other
Requirements
The proposed CAFO regulations contain recordkeeping and reporting requirements.
Costs associated with information collection include the recording of animal inventories, manure
generation, findings from visual inspections of feedlot areas and fields, lagoon emptying, and
other activities on a routine basis. Recordkeeping requirements also include collecting
information on field application of manure and other nutrients (including amount, rate, method,
incorporation, and dates), manure and soil analysis compilation, crop yield goals and harvested
yields, crop rotations, tillage practices, rainfall and irrigation, and lime applications. Other
9-17
-------
requirements include manure spreader calibration worksheets, manure application worksheets,
maintenance logs, and soil and manure test results.
EPA has estimated the burden and costs associated with information collection imposed
on CAFOs and states as a result of the proposed CAFO regulations. This analysis is provided in
the Information Collection Request (ICR) document prepared by EPA (USEPA, 2000i). For the
purpose of this analysis, "burden" means the total time, effort, or financial resources expended by
persons to generate, maintain, retain, or disclose or provide information to or for a federal agency.
This includes the time needed to review instructions; develop, acquire, install, and utilize
technology and systems for the purposes of collecting, validating, and verifying information,
processing and maintaining information, and disclosing and providing information; adjust existing
procedures to comply with any previously applicable instructions and requirements; train
personnel to be able to respond to a collection of information request; search data sources;
complete and review the collection of information; and transmit or otherwise disclose the
information.
EPA's labor burden estimates for CAFO and state respondents are the hours of activity
required to comply with changes to the NPDES CAFO program. For each activity, EPA
estimates the burden in terms of the expected effort necessary to carry out these activities under
normal conditions and reasonable labor efficiency. These activities and estimated burden and cost
levels are described in more detail in the ICR (USEPA, 2000i). The ICR also contains a summary
of wage rate information from USD A, state agricultural extension agencies, and the Bureau of
Labor Statistics, compiled by EPA for the purpose of this analysis. Additional information on the
ICR is provided in Section XELF of the preamble to this rulemaking. A summary of the analysis
of impacts to CAFO operators is provided below. Additional information on the estimated
burden and costs to states is provided in the ICR.
EPA identifies five burden activities to CAFO operators, including start-up activities,
permit application, permit nutrient plan development, best available technology requirements, and
ground water monitoring for new facilities. Start-up activities are steps that a CAFO owner or
operator must take in preparation to comply with the information collection requirements of the
proposed rule. Owners or operators that are potentially affected by the rule will need to
familiarize themselves with the changes to me NPDES CAFO program to determine that they
will need to apply for a permit (or certify out of the program, under three-tier structure only),
develop a PNP, and implement the other BAT requirements. PNPs must be reviewed annually
and rewritten every five years. Permit application activities involve completing and submitting
either an NOI under a general permit or an application-for an individual permit. These activities
will be conducted once every five years.
PNP development and implementation will require owners or operators of CAFOs to
apply for a permit and notify their permitting authority when the PNP has been developed or
modified. This notice must include the number of animals covered by the plan, the number of
acres receiving waste, the nutrient content of the manure, the application schedule and rate, and
9-18
-------
the quantity that will be transferred off site. As part of their recordkeeping responsibilities, CAFO
operators will be required to keep the plan on site for inspections and make it available to the
permitting authority on request.
To meet the proposed BAT requirements, CAFO owners or operators will perform
various activities which will need to be recorded, such as visual inspections of the feedlot
facilities, testing or calibration of manure application equipment, collection of soil samples, -
recording of volume of manure and process wastewater produced as well as off-site transfer, and
employee training. Existing beef and dairy sources as well as all NSPS have requirements will
involve documentation of whether ground water is hydrologically linked to surface water at the
CAFO site and, if it is, records of monitoring of ground water quality. Monitoring records must
be maintained to demonstrate that no discharge has occurred.
In addition to recordkeeping costs, EPA estimates the capital and operation and
maintenance (O&M) costs associated with these burden activities. A CAFO will incur capital
costs when it purchases equipment or builds structures that are needed for compliance with the
rule's reporting and recordkeeping requirements that the facility would not use otherwise.
Consistent with Ihe overall cost analysis for the proposed rule, capital costs are anmialized
assuming a 10-year amortization period and a 7 percent interest rate. Capital costs for the
proposed rule include purchasing a soil auger to collect soil samples and a manure sampler.
CAFOs applying manure on site (assumed to be 100 percent, although land application does not
occur at 100 percent of CAFOs) will need to obtain a scale to calibrate the spreader. Some
facilities will also need to install depth markers in their lagoons, and certain sources with ground
water linked to surface water will need to install monitoring wells. EPA's estimates also include
the one time cost for the nutrient management course in this cost category. A facility incurs
O&M costs when it regularly uses services, materials, or supplies needed to comply with the
rule's reporting and recordkeeping requirements that the facility will not use otherwise. Any cost
for the operation and upkeep of capital equipment is considered an O&M cost. O&M costs may
also be incurred on a non-annual basis, such as every three years. O&M costs include laboratory
analysis of soil, manure, and ground water samples, training of person responsible for manure
application, and maintenance of ground water monitoring wells.
EPA estimates that the public burden for this information collection request will require
1.2 to 1.6 million labor hours for all CAFO respondents to comply with the proposed regulations
(USEPA, 2000i). Information collection at a CAFO is associated with permit application, PNP
development, inspection and sampling, and ground water assessment. These estimates include the
time required to review instructions, search existing data sources, gather and maintain all
necessary data, and complete and review the information collection. EPA estimates total costs to
regulated CAFOs associated with reporting and recordkeeping requirements under the proposed
CAFO regulations at $27 million annually (1999 dollars), under the two-tier structure. For the
three-tier structure, EPA estimates costs to regulated CAFOs at $35 million annually (USEPA,
' 2000i). This estimate excludes NPDES burden for CAFOs covered by other ICR estimates, as
well as NPDES burden for co-permittees and off-site manure recipients.
9-19
-------
Under the two-tier structure, EPA estimates that there will be approximately 7,300 CAFO
respondents and an average of 80,700 CAFO responses per year. Under the three-tier structure,
EPA estimates that there will be approximately 9,600 CAFO respondents and an average of .
107,800 CAFO responses. Thus, the average burden per CAFO respondent is 163 to 166 hours
and the average burden per CAFO response is 14 to 15 hours. For this analysis, EPA assumes
that the administrative burden assumptions are generally the same regardless of CAFO size. Only
soil sampling and PNP development burdens would differ by CAFO size. Costs are assessed
using a weighted average acreage for all affected CAFOs and do not contain a breakdown for
CAFOs with more than or less than 1,000 AU. This estimate likely overstates the time
requirements at small CAFO businesses, since it is an average over all operations both large and
small.
More detailed information on the burden and associated costs for each of the activities
described above is provided in the ICR (USEPA, 2000i).
9.3.2.5Identification of Relevant Federal Rules that May Duplicate, Overlap, or
Conflict with the Proposed Regulations
For this analysis, EPA assumes that all CAFOs are already in compliance with existing
federal and state regulations affecting animal production facilities. The Small Business Advocacy
Review Panel did not identify any federal rules that duplicate or interfere with the requirements of
the proposed rule (USEPA, 2000g).
9.3.2.6 Significant Regulatory Alternatives
EPA proposes to focus the regulatory revisions in this proposal on the largest operations,
which present the greatest risk of causing environmental harm, and in so doing, has minimized the
effects of the proposed regulations on small h'vestock and poultry operations. First, EPA is
proposing to establish a two-tier structure with a 500 AU threshold. Unlike the current
regulations, under which some operations with 300 to 500 AU are defined as CAFOs, operations
of this size under the revised regulations would be CAFOs only by designation. Second, EPA is
proposing to raise the size standard for defining egg laying operations as CAFOs. Third, EPA is
proposing to eliminate the "mixed" animal calculation for operations with more than a single
animal type for determining which AFOs are CAFOs.
Under the two-tier structure, EPA is proposing to revise the threshold for being defined as
a CAFO down to 500 AU and eliminate the "middle category" for operations with between 300
and 1000 AU. This proposal would provide relief to small businesses by removing from the
CAFO definition operations with between 300 AU to 500 AU that under the current rules are
defined as CAFOs. EPA estimates that under the co-proposed alternatives, between 64 percent
(two-tier) and 72 percent (three-tier) of all CAFO manure would be covered by the regulation.
9-20
-------
(See Section 2 of this report.) Under the two-tier structure, the inclusion of all operations with
more than 300 AU instead of operations with more than 500 AU, the CAFO definition would
result in 13,800 additional operations being regulated, along with an additional 8 percent of all
manure. An estimated 80 percent of these additional 13,800 CAFOs are small businesses (about
10,870 CAFOs). EPA estimates that by not extending the regulatory definition to operations with
between 300 and 500 AU, these 10,870 small businesses will not be defined as CAFOs and will
therefore not be subject to the proposed regulations. EPA estimates the additional costs of
extending the regulations to these small CAFO businesses at almost $150 million across all
sectors. The difference in costs between the proposed BAT Option/Scenario and the proposed
BAT Option and Scenario 4b combination may be approximated by comparing the estimated
costs for these regulatory options, which are shown in Section 5.
Also, under the two-tier structure, EPA is proposing to raise the size standard for defining
egg laying operations as CAFOs. This alternative would remove from the CAFO definition small
egg laying operations with between 30,000 and 50,000 hens that under the current rules are
defined as CAFOs, if they utilize a liquid manure management system. (The current regulations
affects egg laying operations with more than 30,000 birds that use wet manure management
systems only. Layer operations with dry manure systems are not covered by the regulations.
EPA is proposing to regulate all layer operations of a certain size, regardless of the type of
manure management systems used, as described in Section 3.) To provide relief to smaller
operations, EPA is proposing to raise the size standard to apply to operations with more than
50,000 birds on-site. A higher size standard for egg laying operations is intended to avoid placing
too much burden on small egg laying operations. These operations are virtually all small
businesses (see Table 9-2). Most of these operations are concentrated in the Southern production
regions. Data are not available to determine the number of egg laying operations with 30,000 to
50,000 layers. Therefore, EPA did not estimate the cost savings of raising the size standards for
egg operations.
In addition, under both co-proposed alternatives, EPA is proposing to revise the threshold
for being defined as a CAFO by eliminating the requirements for "mixed" operations (i.e.,
operations with more than a single animal type). Under the existing permit regulation, if a facility
confines more than one animal type, each animal type is assigned a multiplication factor that is
used to calculate the total number of animal units at the facility. Only poultry is excluded from
this mixed animal type calculation under existing regulations. EPA is proposing to exclude mixed
operations with more than a single animal type. The Agency determined that the inclusion of
these operations would disproportionately burden small businesses while resulting in little
additional environmental benefit. Since most mixed operations tend to be smaller in size, this
exclusion represents important accommodations for small businesses. EPA expects that there are
few large operations that confine more than a single animal type. If certain of these smaller
operations are determined to be discharging to.waters of the U.S., States can later designate them
as CAFOs and subject them to the regulations. EPA's decision not to include operations with
more than a single animal type is also expected to simplify compliance and be more
9-21
-------
administratively efficient, since the mixed operation multipliers were confusing to the regulated
community and to enforcement personnel, and did not cover all animal types.
Overall, EPA's decision to mitigate the effects on small CAFO businesses through these
scope considerations is intended to favor smaller—usually more traditional and often more
sustainable—farm production systems where operators grow both livestock and crops and land
apply manure nutrients. This is consistent with EPA's objectives under the USDA-EPA Unified
National Strategy for Animal Feeding Operations, which targets only the largest operations since
these pose the greatest potential risk to water quality and public health given the sheer volume of
manure generated at these operations (USD A and USEPA, 1999). Larger operations that handle
larger herds or flocks often do not have an adequate land base for manure disposal through land
application. As a result, large facilities need to store significant volumes of manure and
wastewater that have the potential, if not properly handled, to cause significant water quality
impacts. In comparison, smaller operations manage fewer animals and tend to concentrate fewer
manure nutrients at a single location. Smaller operations tend to be less specialized and are more
diversified, engaging in both animal and crop production. These operations often have sufficient
cropland and fertilizer needs to land apply manure nutrients generated at a livestock or poultry
business.
9.4 EPA'S ANALYSIS OF SMALL BUSINESS IMPACTS
This section discusses the data and methodology EPA uses to assess economic impacts on
small CAFO businesses (Section 9.4.1) and presents the results of this analysis (Section 9.4.2).
This economic analysis supports the IRFA (Section 9.3) by quantifying the effects of the proposed
CAFO regulations.
9.4.1 Data and Methodology
To examine the economic impacts of the proposed regulations on small CAFO businesses,
EPA uses the same representative farm approach that is used to analyze impacts to all CAFOs
(regardless of size), as described in Section 4 this EA. This approach evaluates impacts to select
model CAFOs and extrapolates these results to the number of operations identified by each
representative model, thus aggregating costs nationally across all sectors. Inputs for this analysis
include the number of CAFOs represented by each model (see Section 9.3.3) and, for each model
CAFO, the costs of the proposed regulations and selected financial characteristics (see Section 4).
EPA's analysis evaluates the economic achievability of the proposed regulatory options at
small CAFO businesses based on changes in representative financial conditions across three
criteria. These criteria are: a comparison of incremental costs to total revenue (sales test),
projected post-compliance cash flow over a 10-year period, and an assessment of an operation's
debt-to-asset ratio under a post-compliance scenario.
9-22
-------
EPA determines economic impacts to small businesses by applying the proposed economic
achievability criteria described in Section 4.2.5, which are used to divide the impacts of the
proposed CAFO regulations into three categories (see Table 4-11). Accordingly, if an average
model facility is determined to incur economic impacts under the proposed CAFO regulations that
are regarded as "Affordable" or "Moderate," then the results are considered to indicate economic
achievability. "Moderate" impacts are not associated with operational change at the CAFO and
are considered by EPA to indicate economic achievability. If an average operation is determined
to incur "Stress," this result is considered to potentially indicate that the proposed regulations
might not be economically achievable, subject to other considerations. "Affordable" and
"Moderate" impacts are associated with positive post-compliance cash flow over a 10-year period
and a debt-to-asset ratio not exceeding 40 percent, in conjunction with a sales test result that
shows that compliance costs are less than 5 percent of sales ("Affordable") or between 5 and 10
percent of sales ("Moderate"). "Stress" impacts are associated with negative cash flow or a post-
compliance debt-to-asset ratio exceeding 40 percent, or sales test results that show costs equal to
or exceeding 10 percent of sales. More detail on this classification scheme, along with a
discussion of the basis for.EPA's determination of these criteria for this analysis, is provided in
Section 4.2.5.
Table 9-7 shows EPA's estimated compliance costs for selected model CAFOs under the
proposed BAT Option. Costs are not presented separately by facility model for each co-proposed
scenario, since the only difference in costs between Hie two scenarios are associated with the
difference in the numbers of regulated CAFOs. All costs shown are expressed on a per-animal
basis and are differentiated by facility size, producing region, facility types, and other factors.
Costs are reported in ranges across three types of land availability for manure application assumed
for this analysis. These land availability types include: Category 1 farms, which have sufficient
cropland for all on-farm nutrients generated; Category 2 farms, which have insufficient cropland;
and Category 3 farms, which have no cropland. Ranges also reflect Option 3 and 3A costs.9
Section 4.2.1. provides additional information on EPA's cost models. Unit costs shown in Table
9-7 are aggregated by the average number of animals assumed for each model CAFO to derive
total entity compliance costs used in this analysis. Information on EPA's model CAFOs used for
this analysis is provided in Section 4.2 of this report.
9Option 3 assesses average costs to operations if there is no direct hydrologic connection to surface waters;
Option 3A reflects costs to operations where there is a determined groundwater hydrologic connection (assumed at
24 percent of all affected operations):
9-23
-------
Table 9-7. Estimated Per-Head Facility Costs (BAT Option/Co-Proposed Scenarios) for Model CAFOs
Sector
Fed Cattle
Veal
Heifers
Dairy
Hog: FF
Hog: GF
Layer: Wet
Layer: Dry •
Broiler
Turkey
Region
CE
MW
MW
MW
PA
MW
PA
MA
MW
MA
MW
SO
MW
SO
MA
SO
MA
MW
Model
CAFOs
<300AU
Model CAFOs
"Medium 1"
Model CAFOs
"Medium 2"
300 - 1,000 AU
Model CAFOs
"Large 1"
Model CAFOs
"Large 2"
>1,000 AU
(incremental compliance costs $ per animal)
- ._-^. •"™'«£,^r-t.*w-^
,,-•; -.—•:,-<•
>-'::£•££-;-
-, /--,:', **v.i.«
l^^vjjgj
-' -i "-.•rre'sir ~~~~
.v.-.'.'-.i-^Si
~ '-'. "~^tt~'*r *•-* — !
.^ii^fS. J
$60.39-
$222.08
", -:*!;^.>"T^?^w»'H
*-• -*.?-""57"*".°J!~-; ;"e ^
$5.80
•«i
.*..;,;*-, •£
$0.07-
$0.13
$0.07-
$0.15
„„ i» — "*
' ^
$10.81-$80.32
$15.87-$67.24
$2.65-$7.78
$14.13-$55.50
$10.63-$60.86
»«*-,.
f » -*. 4 #
1
"y
$6.03-$7.45
**-"'*** v
~v
$0.83
$0.02-$0.27
$0.02-$0.18
$0.07-$0.13
$0.07-$0.15
$0.07-$0.71
$0.12-$0.83
$7.21-$61.98
$11.31-$50.56
$2.54-$4.75
$9.37-$39.57
$6.86-$43.96
"""*"* V *
* "'* x ,
"„/-*•*
T f *
^t ^ js? !
t -^ r»«>» I
1 •-"- - -*»,'' r
$4.35-$5.65
\ 4*
£,""
S~/ vB» -I"
f X
$0.39-$0.60
$0.02-$0.23
$0.02-$0.15
$0.07-$0.12
$0.07-$0.13
'•> ••*>,.
•^ ^- iis x*-
*• ^ (^" - i"*1 >t~-\. ^
$3.37-$38.59
$7.12-34.65
$2.50-$4.75
$5.04-$20.16
$3.51-$27.14
*j» t -**• * »i.
tfff
• "*-.$, ''S
> -^ f »-^f r- *
*f v^l ^ _/X ^
^ , ^ ^^
."-^--ft: **
•i } •* $ \ ^
tirt* «!i.W «t te
, ^4. •„ "^''M't'yJ^'
• .i^lWX^l
"";? ,^5^
- " >;
—^ „», ^.
€-
.^i»m "^ * """
. "" ^ ^
^ tr* ^
^ g&~-
-*.'*„ ~~
$0.07-$0.12
$0.06-$0.13
<• ^^~^ /
, * ' > '
•"T ~""-»^ ~ ? T'T
•^ -, /?•-.
V »> < ' >£
\ ^ ^ ^^»
*•?+.*" "tVf
,^ *& ^ ~" ""^ ^ f
> *'t&* ^
^ff^ L
A-* V~
1 >r*~? ->**» •
"" ** , f
5 > " * , ^ "
j.flw*' i1 ^
$ } $ -S. * -is*
j^.<, *• jji *v*'",.^
f^<*'"^'
,.1 S ^ T,. i
* ' '« ~r ^**
i- if if, s* „' f
& ^ f %
> ,«« 1*^fc%» *
^ >. (, ^ !^ ,4^ "*
A,tr" " ^ «-
J"S*
* V / X
* ** a " '"" 1 f
*#J *. if I •*
y -*
,f
/"
$0.05-$0.10
$0.05-$0.11
" '. f f '
j-, '"V
s^ - '**''• '
Source: USEPA. Annualized costs are shown in Appendix A; actual costs are in the Development Document
(USEPA, 2000a). Facility size and region definitions for model CAFOs are provided in Section 4, Table 4-1.
Large operations roughly correspond to CAFOs with >1,000 AU and Medium operations correspond to CAFOs
with 300-1,000 AU. Shaded cells indicate that there are no CAFOs that will be affected by the proposed
regulations and that meet the SBA definition of a small business.
•9-24
-------
EPA also developed costs to confinement operations with less than 300 or 500 AU that
may be designated as CAFOs by scaling the estimated compliance costs for the available
"medium" and "large" CAFO models. (See Tables 9-5 and 9-6 for expected designated facilities
under each co-proposed alternative.) The resulting costs—derived on a per-head basis—are
adjusted by the average head counts at operations with fewer than 500 AU or 300 AU to derive
the annualized per-facility compliance cost. EPA assumes that CAFOs with fewer than 500 AU
or 300 AU have sufficient cropland for all on-farm nutrients generated (identified in the cost
model as Category 1 costs). More detailed cost information is provided in the Development
Document (USEPA, 2000a).
As explained in Section 4.2 of this report, EPA evaluates the effect of incurred compliance
costs based on the total number of CAFOs in each sector, including mixed operations. This
approach avoids understating costs at operations with more man one animal type that meets the
size threshold for a CAFO or is designated as a CAFO by the Permitting Authority, and thus may
incur costs to comply with the proposed requirements for each type of animal that is raised on
site. Therefore, EPA's compliance costs estimates likely represent the upper bound, since costs at
facilities with more than a single animal type may, in some cases, be lower due to shared
production technologies and practices across all animal types that are produced on site.
The financial data that EPA uses to analyze impacts on small CAFO businesses are from
USDA's ARMS database (see Section 4.2). These data are.available for 1997 by commodity
sector, facility size (animal inventory), and production region. Available 1997 financial data that
are used to characterize average model CAFOs include gross farm revenue, net cash income (used
to project cash flow), and baseline debt-to-asset ratios. Table 9-8 shows the gross revenue that
EPA assumes for this analysis, expressed on a per-animal basis. Unit revenues shown in Table 9-8
are aggregated by the average number of animals assumed for each model CAFO to derive total
entity revenue used in this analysis. Estimated cash flow and debt-to-asset ratios for CAFO
models are provided in Section 4 of this report (Tables 4-5 and 4-7).
As Table 9-8 shows, USDA data indicate that operations with fewer than 300 AU, on
average, have higher gross revenues when expressed on a per-animal basis than operations with
more than 300 AU. This is explained by the fact that smaller farming operations tend to be more
diversified and engage in both livestock and crop production. In general, larger businesses tend to
be more specialized and concentrate on a single enterprise only. Consistent with SBA's size
standards that are expressed in terms of total annual business revenue (SB A, 1998), EPA assesses
financial impacts at model CAFOs based on changes in total farm revenue. Total farm revenue, as
reported in USDA's ARMS database, includes gross cash income from both livestock and crop
sales (including net Commodity Credit Corporation loans), government payments, and other farm-
related income (income from machine-hire, custom work, livestock grazing, land rental, contract
production fees, outdoor recreation, and other farm-related sources) (USDA/ERS, 1999a).
9-25
-------
Table 9-8. Estimated Per-Head Facility Revenues for Model CAFOs
Sector
Fed Cattle
Veal
Heifers
Dairy
Hog: FF
Hog: GF
Layer: Wet
Layer: Dry
Broiler
Turkey
Region
CE
MW
MW
MW
PA
MW
PA
MA
MW
•MA
MW
SO
MW
SO
MA
SO
MA
MW
Model
CAFOs
<300AU
Model CAFOs
"Medium 1"
Model CAFOs
"Medium 2"
300 - 1,000 AU
Model CAFOs
"Large 1"
Model CAFOs
"Large 2"
>1,000 AU
(incremental compliance costs $ per animal)
':'::•'
-ii*^ ,
":€X
riS":
V*' -'
$2,620
•»" •' f ". wT
'.._-> *
$606
_- ~
-
-
-
$502
$535
$535
$535
$502
*• ^
-
* , to " *^
< 3. A. t ^
~\ '"*•* s „.
$304
e m 3 "•*
\' ^ *•*>' ' /
• * ,f
$25
$25
$25
$1.5
. $1.4
$11.2
$11.2
^^, »,r , -
$854
$862
$862
$862
$854
, ^ .''"Y*
>V/*'-^
-^ "^ *'*<"
;j^5^|g
-Vv'^':'4*-
s* V"^i
^ 4 ~^£. eg /
. ^ ^,^>^
% ' « f" ^
: >M^/r'
~ »* j^ «.'-
,>'fitt;l.:
vrf^ "^U.
;^^'^'
^ i» >«<;; x * ^
<;> '/?;
^ ' ,~» 1 X
%Su/ Lr;
$1.1
$1.2.
* "^ * '**
" - 1* ^~
Y'"' X-'T'
., -s.
Source: Derived from USDA/ERS, 1999a (see Section 4.2.4). Facility size and region definitions for model
CAFOs are provided in Section 4, Table 4-1. Large operations roughly correspond to CAFOs with >1,000 AU and
Medium operations correspond to CAFOs with 300-1,000 AU. Shaded cells indicate that there are no CAFOs that
will be affected by the proposed regulations and that meet the SBA definition of a small business.
9-26
-------
Higher total farm, revenues per animal at smaller-sized farms (due to the inclusion of
revenue from all farm-related sources) is demonstrated in the original USDA ARMS data that are
presented in the individual subcategory sections of this report, including Section 6 (poultry),
Section 7 (hogs), and Section 8 (cattle and dairy). Derived on a per animal basis, these data show
that operations with less than 300 AU tend to generate a larger share of total revenue from other
secondary sources, including other secondary livestock revenue as well as revenue from crop
sales. Other sources of farm-related revenue that tend to be greater at operations with less than
300 AU, compared to operations with more than 300 AU, include other farm-related revenue,
such as government payments and nonfarm income. Since EPA's small business analysis
considers a business' total entity revenue, with SBA size standards, the derived per-unit revenues
are relatively lower per-unit for model CAFOs with more than 300 AU compared to model
CAFOs with fewer than 300 AU. EPA's analysis does not consider sources of non-farm revenue
in its analysis, even though data from USDA indicate that nonfarm revenue often constitutes a
significant share of total operating income (USDA/ERS, 2000d, 1996a and 1999a).
The same ARMS financial data, however, consistently indicate that per-unit cash expenses
tend to be greater among smaller producers than among larger operations. This is consistent with
expectations of economies of size in agricultural production. A review of the agricultural
literature suggests that there may be a statistically positive relationship between farm size and per-
unit production costs, such that as farm size (number of animals) increases, per-unit costs are
lower (ERG, 2000d; Lazarus, et al, 1999). This may result in lower per-unit capital costs and
create a competitive advantage among larger-sized operations relative to smaller ones. This
literature review is provided in the rulemaking record (ERG, 2000d—see DCN 70641),
9.4.2 Economic Analysis Results
Using the proposed economic achievability criteria, discussed in Section 9.4.1, EPA's
economic analysis indicates that the proposed regulations will not impose financial stress on a
substantial number of operations, relative to the total number of affected confinement operations
in these sectors. The results of this analysis are presented in Table 9-9 for each of the co-
proposed tier structures. (Results for Scenario 5 (two-tier structure at 750 AU threshold) and
Scenario 6 are not determined, but fall within the range of the results presented.)
Under both the two-tier and three-tier structures, EPA's analysis indicates that the
proposed requirements will not impose stress impacts on any affected small businesses in the veal,
dairy, hog, egg laying, and turkey sectors. Under the two-tier structure, the proposed
requirements will not result in financial stress to affected small operations in the heifer sector. •
Operations in these sectors are expected to be able to absorb the costs associated with the
9-27
-------
Table 9-9. Results of EPA's Small Business Analysis
Sector
Number
of Small
CAFOs
Affordable
Moderate
Stress
Affordable
Moderate
Stress
Zero Cost Passthrough
(Number- of Operations)
(% Affected Operations)
Two-Tier Structure (Proposed BAT Option/Scenario 4a)
Fed Cattle
Veal
Heifer
Dairy
Hogs
Broilers
Layers
Turkeys ,
TOTAL
1,390
90
800
50
300
9,470
200
0
10,550
1,130
90
680
40
300
1,860
200
0
4,300
250
0
120
10
0
7,460
0
0
7,840
10
0
0
0
0
150
0
0
- 160
81%
100%
85%
80%
100%
20%
100%
NA
41%
18%
0%
15%
20%
0%
79%
0%
NA
74%
Three-Tier Structure (Proposed BAT Option/Scenario 3)
Fed Cattle
Veal
Heifer
Dairy
Hogs
Broilers
Layers
Turkeys
TOTAL
1,490
140
980
50
300
13,410
590
500
14,630
1,100
140
800
40
300
1,910
590
460
5,340
380
0
150
10
0
11,220
0
40
11,800
10
0
30
0
0
280
0
0
320
74%
100%
82%
80%
100%
14%
100%
92%
37%
26%
0%
15%
20%
0%
84%
0%
, 8%
81%
1%
0%
0%
0%
0%
2%
0%
NA
2%
1%
0%
3%
0%
0%
2%
0%
0%
2%
Source: USEPA. Impact estimates shown include impacts to designated operations. Option/Scenario definitions
provided in Table 3-1. Category definitions ("Affordable," "Moderate" and "Stress") are provided in Table 4-13.
Numbers may not add due to roundirig. NA = Not Applicable.
Number of operations does not adjust for operations with mixed animal types, for comparison purposes, to avoid
understating costs at operations with more than one animal type that may incur costs to comply with the proposed
requirements for each type of animal that is raised on-site. The number of CAFOs includes designated facilities.
9-28
-------
proposed CAFO regulations without having to rely on cost passthrough. EPA's analysis shows
that operations across most sectors may experience moderate financial impacts (Table 9-9).
Moderate impacts are not associated with operational change at the CAFO (i.e., will not result
in facility or product line closure) and are considered by EPA to be economically achievable.
In the cattle and broiler sectors, however, EPA's analysis indicates that each of the co-
proposed tier structures will result in financial stress on some small businesses in the fed cattle and
broiler sectors, as will the three-tier structure on some small heifer operations. These small
businesses may be vulnerable to closure. Overall, operations that may experience financial stress
comprise about 2 percent of all affected small CAFO businesses. For the two-tier structure, EPA
estimates that 10 small beef operations and 150 small broiler operations will experience financial
stress. For the three-tier structure, EPA estimates that 40 small beef and heifer operations and
280 small broiler operations will experience financial stress. No designated operations under
either co-proposed scenario are estimated to experience financial stress. Small broiler facilities
with stress impacts are larger operations with more than 1,000 AU under both tier structures.
Small cattle and heifer operations with stress impacts are those that have a ground water link to
surface water. This analysis is conducted assuming that no costs are passed through between the
CAFO and processor segments of these industries. Based on the results of this analysis, EPA is
proposing that the proposed regulations are economically achievable to small businesses in these
sectors.
I
EPA believes that the estimated financial impacts shown in Tables 9-9 are worst-case.
These reasons are summarized below.
First, all results are estimated assuming no costs can be passed through between CAFOs
and the processing sectors. As discussed in Section 5 of this report, if modest levels of cost
passthrough are assumed in the broiler sectors, then the proposed regulations are affordable to all
small broiler operations. EPA did not evaluate economic impacts to cattle operations under a cost
passthrough scenario; however, it is expected that long-run market and structural adjustment by
producers in this sector will diminish the estimated impacts. Even without assumptions of cost
passthrough, EPA's analysis shows that adverse impacts will not be experienced by a substantial
number of operations, as compared to the number of affected operations in these sectors. EPA
has conducted an extensive literature review of issues concerning cost passthrough. Based on the
results of the available empirical research on market power and price transmission in these
industries, EPA believes that there is little evidence to support that increased production costs
may not be passed through the market levels. A summary of this literature review is provided in
the rulemaking record (ERG, 2000c — DCN 70640).
Second, as noted in the Panel Report, EPA believes that the number of small broiler
operations is overestimated, m the absence of business level revenue data, EPA estimates the
number of "small businesses" using the approach described in Section 9.2. Using this approach,
virtually all (>99.9 percent) broiler operations are considered "small" businesses. This
categorization may not accurately portray actual small operations in this sector since it classifies a
9-29
-------
10-house broiler operation with 260,000 birds as a small business. Information from industry
sources suggests that a tvvo-house broiler operation with roughly 50,000 birds is more
appropriately characterized as a small business in this sector (Madison, 1999; Staples, 1998).
Therefore, it is likely that the number of small broiler operations may reflect a number of medium
and large size broiler operations being considered as small entities. As discussed in Section 9.2.1,
EPA consulted with SB A on the use of an alternative definition for small businesses in all affected
sectors based on animal inventory at an operation during the development of the rulemaking.
Third, EPA believes that the use of a costs-to-sales comparison is a crude measure of
impacts on small business in sectors where production contracting is commonly used, such as in
the broiler sector (and also in the turkey, egg, and hog sectors, though to a lesser extent). As
discussed in Section 4.2.4.5, lower reported operating revenues in the broiler sector reflect the
predominance of contract growers in this sector. Contract growers receive a pre-negotiated
contract price that is lower than the USDA-reported producer price, thus contributing to lower
gross revenues at these operations (USDA, 1999). Lower producer prices among contract
growers are often offset by lower overall production costs at these operations, since the affiliated
processor firm pays for a substantial portion of the grower's annual variable cash expenses.
Inputs supplied by the integrator may include feeder pigs or chicks, feed, veterinary services and
medicines, technical support, and transportation of animals (USDA, 1996b). These variable cash
costs comprise a large component of annual operating costs, averaging more than 70 percent of
total variable and fixed costs at livestock and poultry operations (USDA, 1999). The contract
grower also faces reduced risk because the integrator guarantees the grower a fixed output price
(see Section 2.3.1 for more details on contracting in animal agriculture). Because production
costs at a contract grower operation are lower than at an independently owned operation, a profit
test (costs-to-profit comparison) is a more accurate measure of impacts at grower operations.
However, financial data are not available that differentiate between contract grower and
independent operations.
Fourth, EPA's initial regulatory flexibility analysis also does not consider a range of
potential cost offsets available to most farms. One source of cost offset is manure sales,
particularly of relatively higher value dry poultry litter. EPA estimates that sales of dry poultry
litter could offset the costs of meeting the regulatory requirements on the order of more than 50
percent. This reduction alone exceeds the level of cost passthrough (42 percent) assumed for the
cost impact analysis of the broiler sector. Details on how EPA calculated these manure sale
offsets and how they would reduce the economic impacts at poultry operations are presented in
Section 6.
Another source of potential cost offset is cost share and technical assistance available to
farmers for on-farm improvements from various state and federal programs, such as the
Envkonmental Quality Incentives Program (EQIP) administered by USD A. The EQIP program
provides cost-share assistance to all livestock and poultry operations, regardless of size, for
terraces, filter strips, and runoff trenches, as well as technical assistance in formulating
conservation plans. More importantly, operations with 1,000 or fewer AU in confinement, which
9-30
-------
make up the majority of small CAFO businesses, are also eligible to receive funding for
construction of animal waste storage and treatment facilities (e.g., lagoons, holding tanks).
Additionally, many poultry operations with more than 1,000 AU are considered small under SB A
definitions, fall below the EQIP size threshold, and are eligible for waste storage and treatment
funding (e.g., poultry operations with less than 455,000 broilers or less than 250,000 layers).
Although funding may be limited, it is expected that the majority of funds are likely to go to
operations eligible for waste storage and treatment funding (ERG, 2000a).
Many other state and federal cost share programs base eligibility not on size thresholds but
on priority watersheds (e.g., USDA's Small Watershed Program; the New York City Watershed
Program), priority contaminants (e.g., Kansas Non-Point Source Pollution Control Fund), or
proposed waste management practices (e.g., Maryland, Minnesota, Missouri, Nebraska, and
North Carolina state programs). However, technical assistance under most programs is available
to all operations, regardless of watershed, contaminants, proposed practices, or size (ERG,
2000a). A review of cost-share and technical assistance programs available to animal feeding
operations is provided in the rulemaking record (ERG, 2000a — DCN 70130).
Finally, this analysis does not take into account certain noneconomic factors that may
influence an operation's decision to weather the boom and bust cycles that are commonplace in
agricultural markets. Farm typology data from USD A indicate that a large share of farming
operations (more than 90 percent) have annual sales of less than $250,000 and are considered
"small family farms" by USDA (USDA/ERS, 2000d and 2000e). Of these, the majority (about 60
percent) are "limited-resource," "retirement," or "residential" operations where farming is not the
primary source of income (USDA/ERS, 2000e and 1999a). In many cases, these operations, have
negative annual income supplemented by sources of off-farm income that subsidize the farming
operation (USDA/ERS, 2000d and 1996a).
USDA's ERS (1996a) reports that about 60 percent of farm operators reporting negative
net income had nonfarm occupations. About 75 to 80 percent of farms rely on some nonfarm
income, and even in the largest operations nonfarm income can be a significant portion of total
household income (USDA/ERS, 1996a). More than 90 percent of farm operators with negative
net income had nonfarm income averaging more than $35,700 per year; even farms with positive
net income rely somewhat on nonfarm income (Heimlich and Barnard, 1995; USDA/ERS, 1996a).
When farm income is negative over a period of time, sales tests can be very difficult to
interpret (Heimlich and Barnard, 1995). One reason that incomes can remain negative over
several years is that operators can supplement farm income with nonfarm income, and these losses
can be used to reduce total income tax liabilities while the real estate value of the farm property
appreciates. Additional noneconomic factors might also include the satisfaction of working for
oneself, the ability to employ family members, a sense of tradition and the ability to pass on that
tradition to future generations, and the fact that the operation is both a home and a livelihood.
These and other noneconomic factors may influence the decision to close a livestock or poultry
operation cannot be adequately addressed in an economic model. To the extent that these factors
9-31
-------
play a role in that decision, EPA's economic model may overstate the possibility of closure among
small businesses.
USDA's farm financial data include operations where farming is part-time and not the
primary occupation, but excludes sources of nonfarm income at these operations. As noted in
Section 4.2, the inclusion of these operations may result in lower average data values than would
be the case if these operations were excluded from the analysis. EPA believes that the inclusion
of these operations may tend to overstate impacts. Previous analyses by USD A and EPA have
also noted the potential effect on average farm data of including these operations and have
regarded these part-time business more as "hobbies or recreational activities" (Heimlich and
Barnard, 1995; DPRA, 1995). Heimlich and Barnard (1995) further indicate that considering
non-farm income in addition to farm income may provide a more appropriate comparison to the
costs of required measures where the motivation for staying in business is not necessarily purely
economic.
Overall, EPA expects that the proposed CAFO regulations will benefit the smallest
businesses in these sectors, since the regulations may create a comparative advantage for smaller
operations (less than 300 or 500 AU), especially those operations that are not subject to the
regulations. Except for the few AFOs that are designated as CAFOs, these smaller operations will
not incur costs associated with the proposed requirements and may benefit from eventual higher
producer prices as these markets adjust to higher production costs in the longer term.
9-32
-------
SECTION TEN
OTHER REGULATORY ANALYSIS REQUIREMENTS
This section addresses the requirements to comply with Executive Order (EO) 12866 and
the Unfunded Mandates Reform Act (UMRA), both which require federal agencies to assess the
costs and benefits of each significant rule they propose or promulgate.
This section is organized as follows. Section 10.1 describes the administrative
requirements of both EO 12866 and UMRA. Section 10.2 identifies the reasons why EPA has
determined that the existing regulations need to be revised. Section 10.3 provides a summary of
the total social costs of the proposed CAPO regulations. Section 10.4 briefly summarizes the
pollutant reductions that are expected under the proposed CAFO regulations. Section 10.5
summarizes the monetized benefits that are expected to accrue under the proposed CAFO
regulations and also provides a comparison of the estimated total social costs and benefits under
the proposed CAFO regulations (Section 10.6).
Much of the information provided in this section is summarized from and extensively
references, other documents that support this rulemaking, as well as other sections of this report,
that present more detailed accounts of EPA's supporting analyses.
10.1 ADDITIONAL ADMINISTRATIVE AND REGULATORY
10.1.1 Requirements of Executive Order 12866
Under Executive Order 12866 (58 FR 51735, October 4, 1993), the Agency must
determine whether a regulatory action is "significant" and therefore subject to OMB review and
the requirements of the Executive Order. The Order defines "significant regulatory action" as one
that is likely to result in a rule that may:
(1) have an annual effect on the economy of,$100 million or more or adversely affect
in a material way the economy, a sector of the economy, productivity, competition,
jobs, the environment, public health or safety, or state, local, or tribal governments
or communities;
(2) create a serious inconsistency or otherwise interfere with an action taken or
planned by another agency;
(3) . materially alter the budgetary impact of entitlements, grants, user fees, or loan
- programs or the rights and obligations of recipients thereof; or
10-1
-------
(4) raise novel legal or policy issues arising out of legal mandates, the President's
priorities, or the principles set forth in the Executive Order."
EPA has determined that the proposed CAFO rulemaMng is a "significant regulatory
action" under the terms of Executive Order 12866 because.the total costs of the rule are
estimated to exceed $100 million annually. As such, this action was submitted to OMB for
review. Changes made in response to OMB suggestions or recommendations will be documented
in the public record.
In addition to submission of the action to OMB, the principal requirements of the
Executive Order are that the Agency perform an analysis comparing the benefits of the regulation
to the costs that the regulation imposes; that the Agency analyze alternative approaches to the
rule; and that the need for the rule be identified. Wherever possible, the costs and benefits of the
rule are to be expressed in monetary terms. To address the analytical requirements, as specified
by the Executive Order, Section 10.2 describes the reasons why EPA is revising the existing
regulations, and Sections 10.3 through 10.6 present the estimated social costs, pollutant
reductions, and monetary benefits of the proposed CAFO regulations. An in-depth profile of
these industry sectors is presented in Section 2 of this report, with additional information for each
affected industry subcategory provided in more detail in Sections 6, 7, and 8. The proposed
revisions to the existing CAFO regulations are discussed in detail in Sections VII and VTH of the
preamble (briefly summarized in Section 3 of this EA).
10.1.2 Requirements of the Unfunded Mandates Reform Act (UMRA)
Title n of the Unfunded Mandates Reform Act of 1995 (UMRA), P.L. 104-4, establishes
requirements for Federal agencies to assess the effects of their regulatory actions on State, local,
and tribal governments and the private sector. Under section 202 of the UMRA, EPA generally
must prepare a written statement, including a cost-benefit analysis, for proposed and final rules
with "Federal mandates" that may result in expenditures to State, local, and tribal governments, in
the aggregate, or to the private sector, of $ 100 million or more in any one year.
Before promulgating an EPA rule for which a written statement is needed, section 205 of
the UMRA generally requires EPA to identify and consider a reasonable number of regulatory
alternatives and adopt the least costly, most cost-effective or least burdensome alternative that
achieves the objectives of the rule. The provisions of section 205 do not apply when they are
inconsistent with applicable law. Moreover, section 205 allows EPA to adopt an alternative other
than the least costly, most cost-effective or least burdensome alternative, if the Administrator
publishes with the final rule an explanation of why that alternative was not adopted.
Before EPA establishes any regulatory requirements that may significantly or uniquely
affect small governments, including tribal governments, it must have developed under section 203
of the UMRA a small government agency plan. The plan must provide for notifying potentially
10-2
-------
affected small governments, thus enabling officials of affected small governments to have
meaningful and timely input in the development of EPA regulatory proposals with significant
Federal intergovernmental mandates, and informing, educating, and advising small governments
on compliance with the regulatory requirements.
EPA has determined that the proposed CAFO regulations contain a federal mandate that
may result in expenditures of $100 million or more for the private sector in any one year (see
Section 10.3). Accordingly, EPA has prepared the written statement required by section 202 of
the UMRA. This and previous sections of the EA constitute this statement: Sections 5 through 8
of the EA identify costs and impacts (burdens) on CAFOs that are subject to the proposed
regulations, as well as impacts on processors in these industries and other market affects.
Appendix E presents information comparing the cost-effectiveness of the proposed regulatory
alternatives. Additionally, EPA's Benefits Analysis (USEPA, 2000d) presents estimated
monetary benefits that may accrue under the proposed regulations, as required under UMRA
when costs of a federal mandate exceed $ 100 million in any one year.
In addition, EPA has determined that the proposed CAFO regulations do not include a
federal mandate that may result in estimated costs of $100 million or more to either state, local, or
tribal governments in the aggregate. Accordingly, the proposed regulations contain no regulatory
requirements that might significantly or uniquely affect small governments and therefore are not
subject to the requirement of section 203 of the UMRA. Costs incurred by state and federal
governments under the regulatory options being considered are presented in Section 10.3. Tribal
governments may also incur compliance costs; however these costs are expected to be modest and
have not been estimated. EPA has determined that the options considered include no regulatory
requirements that might significantly or uniquely affect local governments.
10.2 NEED FOR THE REGULATIONS
Executive Order 12866 requires that the Agency identify the need for the regulation or
regulations being proposed. The specific need for the proposed CAFO regulations are
summarized throughout this report (Sections 1 and 9) and are presented in more detail in Section
IV and V of the preamble of the proposed ralemaking. These reasons are summarized briefly
below:
» In spite of existing regulatory controls, there is continued discharge and runoff of
manure and nutrients from livestock and poultry operations. The proposed
regulations are expected to address the impairment of many U.S. waterways and
the associated human health and ecological risks by reducing nutrient contributions
from animal agriculture.
• Periodic review and revision of existing regulations is envisioned in the CWA. The
existing regulations need to be updated to reflect structural changes in these
10-3
-------
industries over the last few decades. The continued trend toward fewer but larger
operations, coupled with greater emphasis on more intensive production methods
and specialization, is concentrating more manure nutrients and other animal waste
constituents within some geographic areas. This trend has coincided with
increased reports of large-scale discharges from these facilities.
• The existing regulation needs to be more effective at protecting or restoring water
quality. The revisions will make the regulations easier to understand and better
clarify the conditions under which an AFO is a CAFO and, therefore, subject to the
regulatory requirements. Currently, few livestock and poultry operations have
NPDES permits.
Both UMRA and EO 12866 require the statutory authority for the rule to be cited. A
detailed discussion of the objectives and legal basis for the proposed CAFO regulations is
presented in Sections I and n of the preamble.
10.3 TOTAL SOCIAL COSTS
This section provides a summary of the total pre-tax costs of the proposed CAFO
regulations in 1999 dollars. Compliance costs are presented as post-tax costs (i.e., costs that the
CAFOs would face) in Section 5 of this report. The pre-tax costs include the costs to state and
federal government of foregone tax revenues. This section also provides additional details of how
EPA calculated other costs to state and federal governments for administering the NPDES
permitting program, as it applies to CAFOs. A full accounting of these administrative costs is
provided in the Development'Document (USEPA, 2000a). For comparison with benefits
estimates, all cost estimates in this section are reported in 1999 dollars.
Table 10-1 presents a summary of EPA's estimated total costs of the proposed CAFO
regulations. As shown, for the two-tier structure, EPA projects that the total costs of the
proposed regulations are $847 million per year (1999 dollars, pre-tax). For the three-tier
structure, EPA estimates that costs will total $949 million per year. Estimated total costs have
several components, including costs to regulated CAFOs and to offsite recipients of CAFO
manure, and costs to states and federal governments to administer the NPDES permitting
program. These cost components are discussed in more detail 'in the following subsections.
10-4
-------
Table 10-1. Annual Pre-Tax Costs of the Proposed BAT Option under the Co-Proposed Scenarios, S1999
Sector
Two-Tier Structure
(BAT Option/Scenario 4a)
No. of Operations
(number)
Total Cost
($ millions)
Three-Tier Structure
(BAT Option/Scenario 3)
No. of Operations
(number)
Regulated CAFOs "
Beef
Veal
Heifer
Dairy
Hog
Broiler
Layer
Turkey
Subtotal
3,080
90
800
3,760
8,550
9,780
1,640,
1,280
25,540
$216.4
$0.3
$11.6
$177.6
$294.0
$97.1
$14.2
$19.6
$830.7
3,210
140
980
6,480
8,350
13,740
2,010
2,060
31,930
Total Cost
($ millions)
$227.7
$0.8
$14.4
$224.6
$306.1
$116.6
$15.3
$24.9
$930.4
Other Farming Operations *
Offsite
Recipients
17,923
$9.6
21,155
$11.3
. Permitting Authority b/
States
Federal
Subtotal
Total
24,760
1,030
25,590
NA
$5.9
$0.4
$6.2
$846.5
30,650
21,460
31,930
NA
$7.3
$0.4
$7.7
$949.4
Source: USEPA. NA = Not Applicable. Option/Scenario definitions are provided in Table 3-1. Numbers may not
add due to rounding. .
"'Number of affected facilities adjusts for operations with more than a single animal type, includes expected defined
CAFOs only, and excludes designated facilities. Cost estimates include costs to designated CAFOs (see Section 5).
Section 2 provides additional information on the number of affected facilities and offsite manure recipients.
b/Number of permits includes permits for designated facilities on an annualized basis. Table 10-2 provides
additional information on estimated number of permits.
10-5
-------
10.3.1 Costs to Industry (Regulated CAFOs and Offsite Recipients)
The largest component of social cost is the cost to industry of complying with the
regulation. Costs to industry include annualized capital costs, operating and maintenance costs,
start-up and recurring costs, and also recordkeeping costs. Estimated costs cover four broad
categories, including nutrient management planning, facility upgrades, land application, .and
technologies for balancing on-farm nutrients; All capital costs are depreciated over a 10-year
recovery period, based on the Internal Revenue Code's guidance for single purpose agricultural or
horticultural structures.
EPA projects that the total compliance cost to regulated CAFOs is $831 million per year
(pre-tax) and $572 million (post-tax) under the two-tier structure ($1999). By comparison, under
the three-tier structure, EPA estimates that the cost to industry is $930 million per year (pre-tax)
and $658 million (post-tax). In addition, under the two-tier structure, EPA estimates that the
compliance cost to off-site recipients of CAFO manure is $10 million per year. Under the three-
tier structure, the annual compliance cost to off-site recipients of manure is $11 million per year.
See Table 10-1.
With the addition of administrative costs, EPA projects that, in total under the two-tier
structure, private and public sector costs due to compliance with the proposed CAFO regulations
would be $847 million annually, of which $840 million is incurred by CAFO operators (pre-tax)
and offsite recipients of CAFO manure (Table 10-1). For the three-tier structure, EPA estimates
total costs of $949 million annually, of which $936 million is incurred by industry (Table 10-1).
Estimated costs to regulated CAFOs are presented in Section 5, but are incomplete for the
purposes of meeting the requirements of EO 12866 and UMRA. The costs presented in Section 5
are the post-tax costs and represent the costs to industry after compliance costs have been
expensed or depreciated for tax purposes and income taxes have been paid on earnings. These
post-tax costs reflect the tax shield on compliance costs. The tax shield is the cost to the state
and federal governments of subsidizing, in effect, the cost of the proposed CAFO regulations.
Tax shields are also a cost to society and must be included in the estimate of social costs.
For the purpose of this analysis, estimated pre-tax compliance costs can be viewed as an
estimate of the net output loss (not the gross output loss, which is presented in Section 5) to the
economy, plus consumer and producer surplus losses. EPA does not use an estimate of net
output loss because the Agency would then need to compute output gains and consumer and
producer surplus losses associated with the proposed regulations. Because the pre-tax costs
include no cost passthrough assumptions, no consumer surplus is lost. Additionally, the pre-tax
cost will incorporate the loss in producers' surplus. The pre-tax costs of compliance thus serve as
an estimate for the net output loss to the economy plus losses in consumer and producer surplus.
EPA assumes that all confinement operations that are defined or designated as CAFOs will
incur these costs. Cost estimates include costs to facilities that are projected to experience
10-6
-------
financial stress that could lead to facility closure. In some cases, it is possible that a CAFO might
be liquidated instead of incurring these costs. EPA considers the compliance costs assigned to
these model CAFO to be a reasonable upper estimate of the costs to liquidate such operations.1
An operation will choose to liquidate (to the extent that the choice is theirs to make) only when
the costs of liquidating are less than the costs of installing and implementing pollution control.
Estimated costs to offsite recipients of CAFO manure, presented in Section 5, are
expressed on an annualized basis. Additional detail on how these costs are estimated is provided
in the Development Document (USEPA, 2000a).
10.3.2 Costs to the Permitting Authority (States and Federal Governments)
As discussed in Section 10.3.1, the overwhelming majority (about 99 percent) of the
estimated total regulatory costs will accrue to industry and to state and federal governments in the
form of foregone tax revenues. The remaining burden on state and federal governments is
estimated to range from $6 million to $8 million, depending on the co-proposed alternative. For
the two-tier structure, state and federal administrative costs to implement the permit program are
estimated to be $6.2 million per year: $5.9 million for states and $350,000 for EPA (Table 10-1).
For the three-tier structure, state and federal administrative costs to implement the permit
program are estimated by EPA at $7.7 million per year, estimated at $7.3 million for states, and
$416,000 for EPA (Table 10-1). These costs are expressed in 1999 dollars and are annualized
over the 5-year permit life using a 7 percent discount rate.
Regulatory costs will be incurred by the NPDES permitting authority to alter existing state
programs and obtain EPA approval to develop new permits, review new permit applications, and
issue revised permits mat meet the proposed regulatory requirements. Expected administrative
costs will be related to the development, issuance, and tracking of either general or individual
permits, hi most cases, general permits may be issued. Some circumstances may require that an
individual permit be issued. Most of these costs would be incurred by state governments, since
the majority of states are authorized to administer NPDES permits. As shown hi Table 10-1, the
bulk (95 percent) of estimated administrative costs are expected to be incurred by the state
permitting authority.
This section presents EPA's estimate of the number of operations that will be required to
apply for a permit (Section 10.3.2.1) and the expected unit costs that are used to estimate total
administrative costs (Section 10.3.2.2). Unit costs for general permits and individual permits are
discussed separately. Administrative costs are then aggregated using assumptions about the
number of permits and the unit costs, along with assumptions of frequency of occurrence, to
develop total state and federal costs for administering permits (Section 10.3.2.3).
'These liquidation costs include legal fees, broker fees, etc.
10-7
-------
10.3.2.1 Total Number of Permits
Table 10-2 provides a summary of the total number of AFOs that will be required to apply
for a permit (or certify they meet certain requirements, as required under the three-tier structure)
for each of the co-proposed alternatives. Additional information on the estimated number of
permits that would be required under other NPDES Scenarios considered by EPA during the
development of this rulemaking are provided in the Development Document (USEPA, 2000a);
Table 10-2. Summary of the Number of CAFOs Required to Apply for a Permit, by Sector
Sector
Beef
Veal
Heifers
Dairy
Swine
Broilers
Layers
Turkeys
Total
All
Scenarios
>1,000 AU
2,040
10
300
1,420
.4,020
3,880
630
360
12,660
Two-Tier Structure
Total
2,860
60
700
3,380
7,680
8,560
1,440
1,090
25,770
Defined
800
50
400
1,850
3,560
4,670
800
730
12,860
5-year
Designated
20
0
0
110
100
10
10
0
250
Three-Tier Structure
Total
2,950
110
840
5,470
7,450
11,720
1,730
1,710
31,970
Defined
910
100
540
4,020
3,400
7,840
1,100
1,350
19,260
5-year
Designated
0
0
0
30
30
0
0
0
50
Source: USEPA, 2000a. Rounded to nearest ten. Designated facilities (<500 AU and <300 AU) are shown over a
10-year period (however, designated facilities are included in the "total" column projected over a 5-year permit
cycle).
Under the two-tier structure, EPA estimates that 25,770 CAFOs would be permitted.
This estimate consists of 24,760 state permits (17,320 General and 7,420 Individual permits) and
1,030 federal permits (720 General and 310 Individual permits). Under the three-tier structure,
EPA estimates that 31,970 CAFOs would be permitted, consisting of 30,650 state permits
(21,500 General and 9,190 Individual permits) and 1,280 federal permits (900 General and 380
Individual permits). The total number of permits shown includes permits to operations that are
designated as CAFOs (operations with more than 500 AU and more than 300 AU). Designated
facilities are included in the total number of permits column but are shown projected over a 5-year
permit cycle. The total number also adjusts for operations with more than a single animal type
(USEPA, 2000a).
10-8
-------
EPA did not estimate the number of permits and associated administrative costs for
Scenario 5 (two-tier structure at 750 AU threshold) and Scenario 6.
10.3.2.2 Administrative Unit Costs
State and federal administrative costs to issue a general permit include costs for permit
development, public notice and response to comments, and public hearings. States and EPA may
also incur costs each time a facility operator applies for coverage under a general permit due to
the expenses associated with a Notice of Intent (NOI). These per-facility administrative costs
include initial facility inspections and annual recordkeeping expenses associated with tracking
NOIs. Administrative costs for an individual permit include application review by a permit writer,
public notice, and response to comments. An initial facility inspection may also be necessary.
Unit permit costs assumed by EPA for this analysis are obtained from a number of state permitting
employees. The cost assumptions used to develop, review, and approve permits and inspect
facilities, as well as a comparison of administrative costs among the various NPDES Scenarios
EPA considered during this rulemaking, is presented in the Development Document (USEPA,
2000a). These documents also provide detailed discussions on EPA's assumptions of wage rates.
Table 10-3 provides estimates of administrative costs associated with a general permit.
Unit general permit costs for public hearings, public notifications, and response to comments were
provided by a number of state permitting branch employees (USEPA, 2000a). The most pertinent
of these costs came from the State of Maryland, which has recently developed a general permit
Although the State of Washington also provided costs on general permit development, the state
incurred some exceptional expenses that were deemed unrepresentative (the state held 23 public
meetings and took four years to answer all comments).
X
Information regarding costs (for both general and individual permits) is typically specified
in terms of labor hours. Hours were monetized using estimated average wage rates. For states,
the annual average salary is estimated at $42,000, or $20.19 per hour assuming 2,080 work hours
per year. This rate was multiplied by 1.4 to account for benefits to obtain a final- loaded hourly
wage rate of $28.27. Federal wage rates are estimated based on an annual rate of $47,891 (GS
12, Step 1), which was divided by 2,080 and then multiplied by 1.6 to account for benefits,
resulting in a final loaded hourly labor rate of $36.84 (USEPA, 2000a).
To calculate administrative costs, EPA estimates the overall administrative costs
associated with a generalpermit and the per-facility administrative costs. Table 10-3 presents
EPA's estimates of administrative costs associated with a general permit for both state and federal
governments. State costs to issue one general permit and provide for public notification of
applicants are estimated at approximately $35,820 per permit. EPA estimates that Federal
administrative costs are higher at $40,630 per permit. The footnotes to Table 10-3 provide
additional details on how the items associated with a general permit were costed.
10-9
-------
Table 10-3. Administrative Costs Associated With a General Permit, $1999
Item
Range
Low
High
(hours or $)
Representative
Average
State
Cost0
Federal
Cost *
General Permit Development/Administration Costs
Permit development ^
Public notice/response to comments b/
Public hearing(s)c/
Quarterly public notification *
Total
100
90
120
$400
—
300
8,000
360
$8,000
—
200
.120
240
$4,000
—
$5,650
$3,390
$6,780
$20,000
$35,820
$7,370
$4,420
$8,840
$20,000
$40,630
General Permit Costs per Each Facility Covered
Review/approve notice of intent
Facility inspection d'
1
$1,000
1
$1,000
1
$1,000
$30
$1,000
$40
$1,000
Source: USEPA, 2000a.
"'Permit development estimates were made based on the assumption that states would adopt, with minor changes,
the EPA model permit. It is notable that some states have experienced much higher costs, but that is believed to be-
the result of developing a permit without adopting EPA's model. b/Public notice/response to comment estimates
from MD and WA. MD mailed public notice to 10 newspapers (est. 10 hours). Responding to comments required
2 weeks of one FTE (80 hrs). MD total = 10 + 80 = 90 hrs. WA's costs for public notice were nominal.
Responding to comments in WA took four FTE working 25% for 4 yrs (2080 x 4). It is assumed that this cost was
unusually high and the MD experience would be more representative.
Public hearing estimates were based on estimated time/cost per meeting of 60 hours. Assumed states would have
two to six meetings.
•^Cost for each state's permitting authority to publish in a newspaper a list of facilities that have submitted a NOI to
be covered under a general permit. This notification must be made quarterly. Annual cost based on publication
costs of $1,000 per quarter multiplied by 4 to represent annual publication, then multiplied by 5 to represent the
five-year life of the permits. This estimate likely overestimates costs as there may be some quarters in which no
NOIs are received and thus the permitting authority would not need to publish notification. It is also assumed that
in states where EPA administers the program, EPA will also publish quarterly.
''Based on an average cost per inspection of $1,000 (Reg. 6 and TX). Estimate that 10% of facilities will be
inspected.
"Hourly costs monetized using a loaded rate of $28.27 per hour. This is based on $42,000 (1999) per year or
$20.19/hour assuming 2,080 work hours multiplied by 1.4 to account for benefits. Costs rounded to nearest $10.
^Federal costs based on $46,744/year (GS 12, Step 1,1999), divided by 2,080, then multiplied by 1.6 to account for
benefits, resulting in a final loaded hourly labor rate of $36.84 (documented in the USEPA, 2000a and the"
rulemaking record).
10-10
-------
Table 10-4 shows EPA's estimates of the administrative costs associated with individual
permits for both states and the Federal government. Obtaining an individual permit requires a
state or EPA to review the permit application, provide public notice, and possibly respond to
public comments. In a percentage of cases (estimated in this analysis at 12 percent based on
conversations with permitting authorities in Kansas, Indiana, Missouri, Ohio, and Wisconsin), a
public hearing may be necessary. Additionally, an initial facility inspection may be necessary,
estimated to cost the state or EPA approximately $1,000. Unit individual permit costs for permit
review, public hearings, and inspections were provided by several state permitting branch contacts
that issue individual permits (USEPA, 2000a). Additionally, public hearing costs were based on
information obtained from general permit costs. As with the previous table, the footnotes to this
table provide details as to how the individual permit costs were estimated.
Tahlp 10-4 Administrative Costs Associated with an Individual Permit, $1999
Item
Permit review/public notification/response
to comments ^
Public hearing b/
% of applications requiring hearing d
Avg. Public Hearing Cost/Permit
Total
Inspections *
Range
Low
High
(hours or $)
60
100
4
—
—
$1,000
80
300
20
—
—
$1,000
Representative
Average
70
200
12
—
—
$1,000
State
Cost"
$1,980
—
—
$680
$2,660
Federal
Cost"
$2,580
—
—
$880
$3,460
Source: USEPA, 2000a.
"'Response to comments estimates from KS. 2-3 FTEs dedicated to responding to comments - 4,160-6,240 hrs
divided by 50-100 permits per year.b/Based on estimates from WA, which indicated each hearing required
approximately 100-150 hrs.of State employee time. Using best professional judgement, assume 1-2 public
meetings/hearings per permit at 100-150 hours per hearing. ^
"From Kansas (4-8%) and Indiana (15-20%).
'''Based on an average cost per inspection of $1,000 (Reg. 6 and TX). It is estimated that 10% of facilities will be
inspected.
"Hourly costs monetized using a loaded rate of $28.27 per hour. This is based on $42,000 (1999) per year or
$20.19/hour assuming 2,080 work hours multiplied by 1.4 to account for benefits. Costs rounded to nearest $10.
"Federal costs based on $46,744/year (GS 12, Step 1,1999), divided by 2,080, then multiplied by 1.6 to account for
benefits, resulting in a final loaded hourly labor rate of $36.84 (USEPA, 2000a). Costs rounded to nearest $10.
10.3.2.3 Total Administrative Costs
Under the two-tier structure (Scenario 4a), operations with more than 500 AU are CAFOs
and must obtain a permit. Designated facilities with fewer than 500 AU must also obtain a permit.
10-11
-------
In total EPA estimates that 25,770 facilities will be required to apply for a permit (Table 10-2).
Table 10-5 presents EPA's estimate of state and Federal administrative costs to permit CAFOs
under regulatory Scenario 4a.
Under the three-tier structure (Scenario 3), operations with more than 1,000 AU are
CAFOs and must obtain a permit. Operations with between 300 and 1,000 AU are also defined as
CAFOs and must either certify that they do not meet specific conditions or must obtain a permit.
Designated facilities with fewer than 300 AU must also obtain a permit. In total, EPA estimates
that 39,330 facilities will be required to apply for a permit or certify that they do not meet the
criteria specified in the scenario. For purposes of estimating administrative costs, EPA assumes
that 31,930 facilities will actually obtain a permit. (The actual number will be determined by the
permitting authorities on a case-by-case basis.) Table 10-6 presents EPA's estimate of state and
Federal administrative costs to permit CAFOs under regulatory Scenario 3.
The numbers presented in Tables 10-5 and 10-6 reflect the 42 states that are authorized to
administer NPDES programs (and thus are associated with state level costs), and the 8 states
whose programs are administered by EPA (and thus are associated with federal level costs).
Other key assumptions that EPA uses to develop these estimated costs are that 70 percent of all
CAFOs would be covered by general permits and 30 percent would be covered by individual
permits; that inspections would be required for 10 percent of all CAFO applications; and that the
authorized 42 states would account for 96 percent of all permits. See the Development Document
(USEPA, 2000a) for more detailed discussions of the assumptions used in this analysis.
10.3.3 Other Social Costs
An estimate of total social costs of the proposed regulations comprises costs that go
beyond the compliance costs of constructing and implementing pollution control procedures.
Some of these additional costs are monetary, but many are nonmonetary or not easily monetized
(i.e., a dollar value cannot be attributed or is difficult to attribute to the items).
Additional monetary costs include the cost of federal and state subsidies in the form of a
tax shield (or lost tax revenue) and costs of administering a regulation (permitting costs). These
costs are presented in Sections 10.3.1 and 10.3.2. However, other costs might also be incurred
under the proposed regulations and constitute the full range of total social costs. For example,
costs may be incurred as a result of worker dislocations. These costs comprise the value to
workers of avoiding unemployment and the costs of administering unemployment, including the
costs of relocating workers, and the inconvenience, discomfort, and time loss associated with
unemployment (the unemployment benefits themselves are, generally, considered transfer
payments, not costs). Other potential social costs include the cost associated with a slowdown in
the rate of innovation. In theory, there might be some impact on the rate of innovation to the
extent that farms might invest in newer technologies if they did not have to allocate resources to
meeting the requirements of the proposed regulations. Generally, however, unless an industry
10-12
-------
Table 10-5. State and Federal Administrative Costs, Two-Tier Structure (Scenario 4a)
Item
State
Unit
Cost
Number
Req.
Total
Cost
Federal
Unit
Cost
Number
Req.
General Permit
General Permit Development
Costs
$35,820
42
$1,504,440
$40,630
8
Total
Cost
$325,040
General Permit Tracking Costs
Notification of Intent
Inspections
Total Permit Costs
$30
$1,000
—
17,320
1,732
—
$519,600
$1,732,000
$3,756,040
$40
$1,000
—
720
72
—
$28,800
$72,000
$425,840
Individual Permit
Permit Review/Approval
Inspections
Total Permit Costs
Grand Total
Annualized Total
$2,660
$1,000
• — .
—
—
7,420
742
—
—
—
$19,737,200
$742,000
$20,479,200
$24,235,240
$5,910,750
$3,170
$1,000
—
—
—
310
31
—
—
—
$982,700
$31,000
$1,013,700
$1,439,540
$351,090
Source: USEPA, 2000a.
Assumptions:
70% of all CAFOs are assumed to be covered by general permits & 30% by individual permits.
Inspections are estimated to be conducted for 10% of all CAFO permit applications.
The 42 authorized states are estimated to account for 96% of the total permits.
Costs are annualized using a 7% discount rate over a period of five years.
Total CAFOs permitted 25,770
State total 24,740
General (NOI) 17,320 .
Individual 7,420
Federal total 1,030
General (NOI) 720
Individual 310
10-13
-------
Table 10-6. State and Federal Administrative Costs, Three-Tier Structure (Scenario 3)
Item
State
Unit
Cost
Number
Req.
Total Cost
Federal
Unit
Cost
Number
Req.
Total Cost
Certification
Certification requirements
$30
6,300
$189,000
$40
260
$10,400
General Permit
General Permit Development
$35,820
42
$1,504,440
$40,630
8
$325,040
General Permit Tracking
Notification of Intent
Inspections
Total Permit Costs
$30
$1,000
—
21,460
2,146
—
$643,800
$2,146,000
$4,483,240
$40
$1,000
—
900
90
—
$36,000
$90,000
$461,440
Individual Permit
Permit Review/Approval
Inspections
Total Permit Costs
Grand Total
Annualized Total
$2,660
$1,000
—
—
—
9,190
919
—
—
—
$24,445,400
$919,000
$25,364,400
$29,847,640
$7,279,560
$3,170
$1,000
—
—
—
380
38
—
—
—
$1,204,600
$38,000
$1,242,600
$1,704,040
$415,600
Source: USEPA, 2000a.
Assumptions:
70% of all CAFOs are assumed to be covered by general permits & 30% by individual permits.
Inspections are estimated to be conducted for 10% of all CAFO permit applications.
The 42 authorized states are estimated to account for 96% of the total permits.
Costs are annualized using a 7% discount rate over a period of five years.
Total CAFOs certifying and obtaining permits = 39,330
Total CAFOs permitted 31,970
State total 30,690
General (NOT) 21,500
Individual 9,190
Federal total 1,280
General (NOI) 900
Individual 380
10-14
-------
is highly technical, with major investments in research and development, impacts on the rate of
innovation are likely to be minimal.
Monetizing such social costs is difficult. EPA does not evaluate these other potential
social costs and expects that these other costs to society that are not specifically addressed by the
analyses presented in this report will be modest.
10.4 POLLUTANT REDUCTIONS
EPA's estimate of the pollutant reduction expected under the proposed regulations is
presented in Appendix E of this report. For the two-tier structure, EPA estimates that the
proposed regulations will result in pollutant reductions of 8.4 million toxic pounds-equivalent of
priority pollutants, 182 million pounds of nitrogen, 377 million pounds of phosphorus, as well as
removals of other pollutants, such as sediments and pathogens. For the three-tier structure,
expected pollutant reductions are higher, estimated at 9.4 million toxic pounds-equivalent of
priority pollutants, 206 million pounds of nitrogen, and 425 million pounds of phosphorus. These
estimated removals are measured by EPA at the stream level, as described in Appendix E. More
details on the expected pollutant removals is provided in the Benefits Analysis (USEPA, 2000d)
and the Development Document (USEPA, 2000a).
Section 10.5 summarizes the monetized benefits that EPA expects will accrue under the
proposed CAFO regulations, which is based on a more detailed assessment of these benefits in
EPA's Benefits Analysis of the proposed rulemaking (USEPA, 2000d). Finally, Section 10.6
presents a comparison of these cost and benefit estimates.
10.5 BENEFITS ASSESSMENT
EPA estimates that the monetized benefits of the proposed regulations range from $ 146
million to $182 million annually, depending on the co-proposed approach (Table 10-7). Annual
benefits range from $146 million to $165 million under the two-tier structure; under the three-tier
structure, estimated benefits range from $163 million to $182 million annually. EPA is only able
to monetize (i.e., place a dollar value on) a small subset of the range of potential benefits that may
accrue under the proposed regulations. Data and methodological limitations restrict the number
of benefits categories that EPA is able to reasonably quantify and monetize. The proposed
regulations' benefits are primarily in the areas of reduced health risks and improved water quality,
as shown in Table 10-7, which also provides a comparison of the estimated total social costs and
benefits of the proposed CAFO regulations. These benefits categories are described in more
detail as follows:
• Improvements in Water Quality and Suitability for Recreational
Activities: This analysis estimates the economic value of improvements in
inland surface water quality that would increase opportunities for
recreational fishing and swimming.
10-15
-------
Table 10-7. Total Annual Social Costs and Monetized Benefits, $1999
Total Social Costs
"Two-Tier" Structure
(500 AU threshold)
"Three-Tier" Structure
(Scenario 3)
Social Costs
Industry Compliance Costs (pre-tax)
NPDES Permitting Costs
OffsSte Recipients of CAFO Manure
Total Social Costs
$830.7 million
$6.2 million
$9.6 million
$846.5 million
$930.4 million
$7.7 million
$11. 3 million
$949.4 million
Monetized Benefits
Improved surface water quality
Reduced shellfish bed closures
Reduced fish kills
Improved water quality in private wells
Total Monetized Benefits
$108.5 million
$0.2 - 2.4 million
$0.2 - 0.4 million
$36.6-53.9 million
$145.5 - 165.1 million
$127.1 million
$0.2 - 2.7 million
$0.2 - 0.4 million
$35.4 - 52.1 million
$163.0 - 182.3 million
Source: USEPA, 2000d. Estimated costs are presented in this report (Sections 5 and 10). More details on EPA's
offsite recipient costs are in the Development Document (USEPA, 2000a). EPA's benefits estimates are described
in the Benefits Analysis (USEPA, 2000d).
• Reduced Incidence of Fish Kills: This analysis estimates the economic
value of a potential reduction in the number offish kills caused by AFO-
related waste.
• Improved Commercial Shellfishing: This analysis characterizes the impact
of pollution from AFOs on access to commercial shellfish growing waters
and values the potential increase in commercial shellfish harvests that may
result from improved control of that pollution.
• Reduced Contamination of Private Wells: This analysis examines Hie
impact of the revised regulations on groundwater quality and values
predicted improvements in the quality of aquifers that supply private wells.
These individual analyses are discussed in the Benefits Analysis (USEPA, 2000d). These
categories cover a small subset of the broader range of potential benefits that will likely accrue
under the proposed regulation. In addition to these monetized benefits, EPA expects that
additional benefits will accrue under the regulations, including reduced drinking water treatment
costs, reduced odor and air emissions, improved water quality in estuaries, and avoided loss in
property value near CAFOs, among other benefits. These benefits are described in more detail in
the Benefits Analysis (USEPA, 2000d) and other supporting documentation provided in the
10-16
-------
record. These supporting documents also provide estimates of the benefits that would accrue
under alternative regulatory approaches considered by EPA during the development of this
rulemaking.
10.6 COMPARISON OF COST AND BENEFITS ESTIMATES
Table 10-7 compares the costs and benefits of the proposed CAFO regulations under the
two-tier and three-tier structures. Under the two-tier structure, EPA estimates compliance costs
at $847 million, while benefits are estimated at $146 to $165 million. Under the three-tier
structure, EPA estimates costs at $949 million, while benefits are estimated at $163 to $182
million.
10-17
-------
-------
SECTION ELEVEN
REFERENCES
Abel, Daft & Barley. 1993. Meat, Poultry and Dairy Product Exports: A Silent Revolution.
Study commissioned by several agricultural trade associations. Alexandria, VA.
AFPC (Agricultural & Food Policy Center). 2000. Representative Farms Economic Outlook for
the 2000 FAPRI/AFPC Baseline. AFPC Working Paper 00-1. Texas A&M University.
February.
http://www.aQ)c.tamu.edu/pubs/index.php?content=browse&type=0&year=2000.
AFPC (Agricultural & Food Policy Center). 1999. Representative Farms Economic Outlook for
the 1999 FAPRI/AFPC Baseline. AFPC Working Paper 99-1. Texas A&M University.
February, http://www.arpc.tamu.edu/pubs/
Alston, J.M. and J.A. Chalfant. 1993. The Silence of the Lambdas: A Test of the Almost Ideal
and Rotterdam Models. American Journal of Agricultural Economics. May.
American Egg Board. 1998. Egg industry facts sheet. December 2.
http://www.aeb.org/eei/facts/uidustry-facts.html
Aradhyula, S.V. and M.T. Holt. 1989. Risk Behavior and Rational Expectations in the U.S.
Broiler Market. American Journal of Agricultural Economics. November.
Arzac, E.R. and M. Wilkinson. 1979. A Quarterly Econometric Model of United States
Livestock and Feed Grain Markets and Some of Its Policy Implications. American Journal
of Agricultural Economics. May.
AS AE (American Society of Agricultural Engineers). 1993. Manure Production and
Characteristics. ASAED384-1. St. Joseph, Michigan.
Associated Press. 2000. New rules expand liability in livestock environmental damage. Order
includes companies that pay farmers to raise animals; some operations exempt. Published
in the Herald-Ledger (Lexington, KY). February 15.
Aust, P. 1997. An institutional analysis of vertical coordination versus integration: the case of
the U.S. broiler industry. East Lansing, MI: Michigan State University, Department of
Agricultural Economics. Staff Paper 97-24. June.
11-1
-------
Babcock, B A., R. Fleming, and D.S. Bundy. 1997. The Cost of Regulating Hog Manure
Storage Facilities and Land Application Techniques. Iowa State University, The Center
for Agricultural and Rural Development (CARD). Working Paper 97-BP 17. June.
Bailey, K., D. Hardin, J. Spain, J. Garrett, J. Hoehne, R. Randel, R. Ricketts, B. Stevens, and J.
Zulovich. 1997. An Economic Simulation Study of Large-Scale Dairy Units in the
Midwest. Journal of Dairy Science. 80(1). January.
Bastian, C., D. Baily, D. Menkhaus, and T. Glover. 1994. Today's Changing Meat Industry and
Tomorrow's Beef Sector. Utah State University Cooperative Extension Service.
Becker, G.S. 1996. Cattle Prices: Questions and Answers. Congressional Research Service:
Report for Congress. November.
Bennett, M., D. Osburn, and C. Fulhage. 1992. Waste Management Systems for Dairy Herds.
University of Missouri-Columbia, University Extension publication MP666.
Bentley, S.E., P.F. Barlett, F.L., Leistritz, S.H. Murdock, W.E. Saupe, and D.E. Albrecht. 1989.
Involuntary Exits from farming: Evidence from Four Studies. AER-625. November.
Washington DC: U.S. Department of Agriculture, Economic Research Service.
Blayney, D.P. and R.C. Mittelhammer. 1990. Decomposition of Milk Supply Response into
Technology and Price-Induced Effects. American Journal of Agricultural Economics.
72(5):864-872. November.
Bliss, T. andC. Ward. 1999. Seasonal cattle feeding profits. OSU Extension Facts. Oklahoma
Cooperative Extension Service, Division of Agricultural Sciences and Natural Resources.
F-506.
BLS (Bureau of Labor Statistics). 2000. Unemployment in States. U.S. Department of Labor.
http://stats.bls.gov/bls_news/archives/laus_nr.hrm
Boggess, W.B., G. Johns, and C. Meline. 1997. Economic Impacts of Water Quality Programs
in the Lake Okeechobee Watersheds of Florida. Journal of Dairy Science. 80(10): 2682-
2691.
Bonner, L. 2000. Hog-waste accord reached; Smithfield Foods will help pay for efforts to find
alternatives for hog-waste lagoons. July 26.
Brester, G.W. 1996. Estimation of the U.S. Import Demand Elasticity for Beef: The Importance
ofDisaggregation. Review of Agricultural Economics. 18(l):31-42. January.
11-2
-------
Brester, G.W. and T.C. Schroeder. 1995. The Impacts of Brand and Generic Advertising on
Meat Demand. American Journal of Agricultural Economics. 77(4):969-979. November.
Brester, G.W. and M.K. Wohlgenant, 1991. Estimating Interrelated Demands for Meats Using
New Measures for Ground and Table Cut Beef. American Journal of Agricultural
Economics. 73(4): 1182-1194. November.
Brigham, E.F., and L.C.Gapenski. 1997. Financial Management Theory and Practice. Fort
Worth, TX: The Dryden Press.
Brown, D.J. and L.F. Schrader. 1990. Cholesterol Information and Shell Egg Consumption.
American Journal of Agricultural Economics. 72(3):548-555. August.
Buhr, B.L. and H. Kim. 1997. Dynamic Adjustment in Vertically Linked Markets: The Case of
the U.S. Beef Industry. American Journal of Agricultural Economics. 79(1): 126-138.
February.
Buhr, B. 1993. A Quarterly Econometric Simulation Model of the U.S. Livestock and Meat
Sector. .University of Minnesota, Department of Agricultural and Applied Economics.
Staff Paper P93-12. May. http://agecon.lib.umn.edu/mn/p93-12.pdf
Buxton, B. 1985. Factors Affecting U.S. Milk Production. AER 527. Washington, DC: U.S.
Department of Agriculture, Economic Research Service.
Cady, R. 2000. Personal communication between Eastern Research Group, Inc., and Dr. Roger
Cady, of Monsanto Company and founder of the Professional Dairy Heifer Growers
Association, concerning the heifer replacement industry. February 18.
Capps, O., D.E. Farris, PJ. Bryne, J.C. Namken, and C.D. Lambert. 1994. Determinants of
Wholesale Beef-Cut Prices. Journal of Agricultural and Applied Economics. 26(1):183-
199. July. ' '
Capps, O. and J.D. Schmitz. 1991. A Recognition of Health and Nutrition Factors in Food
Demand Analysis. Western Journal of Agricultural Economics. 16(l):21-35. July.
Capps, O. 1989. Utilizing Scanner Data to Estimate Retail Demand Functions for Meat
Products. American Journal of Agricultural Economics. 71(3):750-760. August.
CARD (Center for Agricultural and Rural Development). 1993. The Economic and
Environmental Indicators for Evaluating the National Pilot Project on Livestock and the
Environment. Iowa State University. Staff Report 93-SR 64. October.
CCH (Commerce Clearing House, Inc.). 1999a. 2000 State Tax Handbook. Chicago, IL.
11-3
-------
CCH (Commerce Clearing House, Inc). 1999b. 2000 U.S. Master Tax Guide. Chicago, EL.
CCH (Commerce Clearing House, Inc.). 1995. Personal communication between Eastern
Research Group, Inc., and CCH, Inc., to resolve discrepancies on tax rates for Missouri
and Rhode Island. March 30.
Chavas, J-P. and M.T. Holt. 1993. Market Instability and Nonlinear Dynamics. American
Journal of Agricultural Economics. 75(1): 113-120. February.
Chavas, J-P., A.F. Kraus, and E.V. Jesse. 1990. A Regional Analysis of Milk Supply Response
in the United States. North Central Journal of Agricultural Economics. 12(2):149-176.
July.
Chavas, J-P. and R.M. Klemme. 1986. Aggregate Milk Supply Response and Investment
Behavior on U.S. Dairy Farms. American Journal of Agricultural Economics. 68(1):55-
66. February.
Chavas, J-P. 1983. Structural Change in the Demand for Meat. American Journal of
Agricultural Economics 65(1):148-153. February.
Chavas, J-P. 1982. On the Use of Price Ratio in Aggregate Supply Response: Some Evidence
From the Poultry Industry. Canadian Journal of Agricultural Economics. 64(4):345-358.
November.
Chavas, J-P., and S.R. Johnson. 1982. Supply Dynamics: The Case of U.S. Broilers and
Turkeys. .American Journal of Agricultural Economics. 64(3):558-564. August.
Chavas, J-P. and S.R. Johnson. 1981. An Econometric Model of the U. S. Egg Industry.
Applied Economics. 13:321-335.
Choi, S. and K. Sosin. 1990. Testing for Structural Change: The Demand for Meat. American
Journal of Agricultural Economics. 72(l):227-236. February.
Christensen, L.A., J.R. Trierweiler, T.J. Urlich, M.W. Erickson. 1981. Managing Animal
Wastes: Guidelines for Decisionmaking. Washington, DC: U.S. Department of
Agriculture. Economic Research Service. ERS-671.
Council of Economic Advisors. 2000. Economic Report of the President. Washington, DC:
USGPO. February.
Dhuyvetter, K.C., J. Graff, and G.L. Kuhl. 1998. Kansas Beef Industry: Economic Trends.
Kansas State University Agricultural Extension and Experiment Station and Cooperative
Extension Service. February.
11-4
-------
Doane's. 1995. Livestock Production Costs. Doane's Agricultural Report. Vol.58. Number
32-5. '
DPRA (DPRA Incorporated). 1995. Economic Impact of National Nonpoint Source
Management Measures Affecting Confined Animal Facilities. Unpublished, Manhattan,
Kansas. May.
Dun and Bradstreet. 1998. Industrial Norms and Key Business Ratios. Dun and Bradstreet,
Industry and Financial Consulting Services.
Dun and Bradstreet. 1997. Industrial Norms and Key Business Ratios. Dun and Bradstreet,
Industry and Financial Consulting Services.
Dun and Bradstreet. 1996. Industrial Norms and Key Business Ratios. Dun and Bradstreet,
industry and Financial Consulting Services.
Duncan, M.R., R.D. Taylor, D.M. Saxowsky, and W.W. Koo. 1997. Economic Feasibility of the
Cattle Feeding Industry in the Northern Plains and Western Lakes States. North Dakota
State University, Department of Agricultural Economics. March.
http://agecon.lib.umn.edu/ndsu/aer370.pdf
Bales, J.S., J. Hyde, and L.F. Schrader. 1998. A Note on Dealing with Poultry in Demand
Analysis. Journal of Agricultural and Resource Economics. 23(2):558-567. December.
Bales, J.S. and LJ. Unnevehr. 1994. The Inverse Almost Ideal Demand System. European
Economic Review. 38:101-115.
Bales, J. S. and L. J. Unnevehr. 1993. Simultaneity and Structural Change in U.S. Meat
Demand. American Journal of Agricultural Economics. 75(2):259-268. May.
Bales, J. S. and L. J. Unnevehr. 1988. Demand for Beef and Chicken Products: Separability and
Structural Change. American Journal of Agricultural Economics. 70(3):521-532. August.
Easterling, E.H., and F.A. Lasley. 1985. Estimating Costs and Returns for Poultry and Eggs.
AGES 850703. Washington, DC: U.S. Department of Agriculture, Economic Research
Service. July.
El-Osta, H.S. and J.D. Johnson. 1998. Determinants of Financial Performance Of Commercial
Dairy Farms. Technical Bulletin No. 1859. Washington, DC: U.S. Department of
Agriculture, Economic Research Service.
http://www.ers.usda.gov/epubs/pdf/tbl859/index.htm
11-5
-------
Endsley, J., G.W. Atkeson, and S. Nott. 1996. Income Potential and Guidelines for the Custom
Dairy Heifer Grower. Michigan State University, Department of Agricultural Economics.
Staff Paper 96-89. October.
ENR (Engineering News Record). 2000. Construction Cost Index History (1908-2000).
http://www/enr.com/cost/costcci.asp
Environmental Defense. 2000. Dollars and Sense: An Economic Analysis of Alternative Hog
Waste Management Technologies.
http://www.hogwatch.org/resourcecenter/onlmearticles/econreport/mdex_econreport.html
ERG (Eastern Research Group, Inc.). 2000a. Cost-Share and Technical Assistance Programs
and the CAFO Analysis. Memorandum to R. Johnson, EPA. August 16.
ERG (Eastern Research Group, Inc.). 2000b. NCBA Financial Survey Data Provided to EPA.
Memorandum to R. Johnson, EPA. November.
ERG (Eastern Research Group, Inc.). 2000c. Review of Cost Passthrough and Market Power
Literature-Revised. Memorandum to R. Johnson, EPA. August 24.
ERG (Eastern Research Group, Inc.). 2000d. Review of Productivity Literature—Revised.
Memorandum to R. Johnson, EPA. August 30. .
ERG (Eastern Research Group, Inc.). 2000e. Summary of SEC 1 OK Filings Used in Processor
Analysis. Memorandum to the Rulemaking Record. December.
ERG (Eastern Research Group, Inc.). 2000f. Post-compliance Impacts to New Operations.
Memorandum to R. Johnson, EPA. December 14.
ERG (Eastern Research Group, Inc.). 1999a. Beef and dairy market model development.
Memorandum to R. Johnson, EPA. July 30. •
ERG (Eastern Research Group, Inc.). 1999b. Review of pork and poultry market studies.
Memorandum to R. Johnson, EPA. September 17.
ERG (Eastern Research Group, Inc.). 1999c. Preliminary profile of poultry, hog, beef, and dairy
processors and integrators. Memorandum to R. Johnson, EPA. September 28.
Farm Journal. 1998. Find market niche online. Hogs Today. July/August.
ht1p://www.farmjoumal.com/magazmes/article.cfm?art_ID=3653&ID=3&is
CP=3
11-6
-------
Faust, M. A. 1995. Dairy Replacement Heifer Enterprises. Department of Animal Science, Iowa
State University. DSL-18.
Featherston, J.W. and J. Atwood. 1999. The Role of Economics in Producer's Waste
Management Decisions. Paper presented at an Organized Symposium held in conjunction
with the 1999 American Agricultural Economics Association Meeting, Nashville, TN.
FFSG (Farm Financial Standards Council). 1997. Financial Guidelines for Agricultural
Producers. Naperville, IL: Farm Financial Standards Council.
Fleming, R., B.A. Babcock, and D.S. Bundy. 1997. the Cost of Regulating Hog Manure
Storage Facilities and Land Application Techniques. Iowa State University, The Center
for Agricultural and Rural Development (CARD). Working Paper 97-BP 17. June.
Foster, K. 2000a. Personal communication between Eastern Research Group, Inc., and Ken
Foster of Purdue University. February 29.
Foster, K. 2000b. Personal communication between Eastern Research Group, Inc., and Ken
Foster of Purdue University. October 13.
Foster, K. and O. Burt. 1992. A Dynamic Model of Investment in the U. S. Beef Cattle Industry.
Journal of Business and Economic Statistics. 10:419-426.
Freese, B. 1997. Pork Powerhouses 1997. Successful Farming Online. October.
" http://www.agriculture.com/sfonline/sf/1997/October/pork97/tbll997.html
Gollehon, N. and M. Caswell. 2000. Confined Animal Production Poses Manure Management
Problems. Agricultural Outlook. Washington, DC: U.S. Department of Agriculture,
Economic Research Service. September.
http://www.ers.usda.gov/publications/agoutlook/sep2000/ao274f.pdf
Goodman, P. 1999. Perdue to help farmers dispose of chicken waste; company planning to turns
tons of manure into fertilizer and sell it elsewhere. Published in the Washington Post
(Washington, DC). February 25.
Goodwin, B.K. and M.T. Holt. 1999. Price Transmission and Asymmetric Adjustment in the
U.S. Beef Sector. American Journal of Agricultural Economics. 81(3 August):630-637.
Gray, B., C. Gempesaw, S. Weyerbrock. 1999. Regional Economic Effects of Proposed
Environmental Regulations on the Delmarva Poultry Industry. University of Delaware.
Paper prepared for the-1999 NAREA meeting.
11-7
-------
Gregory, G. 1999. Personal communication between EPA and Gene Gregory, Director of the
United Egg Producers. February.
Hahn, W.F. 1999. Personal communication between Eastern Research Group, Inc., and William
Hahn of U.S. Department of Agriculture, Economic Research Service. July-September.
Hahn, W.F. 1998. Personal communication between Eastern Research Group, Inc., and William
Hahn of U.S. Department of Agriculture, Economic Research Service. July 15.
Hahn, W.F. 1996a. An Annotated Bibliography of Recent Elasticity and Flexibility Estimates for
Meat and Livestock. Washington, DC: U.S. Department of Agriculture, Economic
Research Service. Staff Paper 9611. July.
Hahn, W.F. 1996b. Documentation for the Animal Product Branch's Cost-Benefit Calculation
Model for Red Meat and Poultry. Washington, DC: U.S. Department of Agriculture,
Economic Research Service. Staff Paper 9606. April.
Hahn, W.F. 1988. Effects of Income Distribution on Meat Demand. The Journal of Agricultural
Economics Research. 40(2): 19-24. Spring.
Hansen, B., W. Hahn, and M. Weimar. 1994. Determinants of the Farm-to-Retail Milk Price
Spread. Washington, DC: U.S. Department of Agriculture, Economic Research Service.
AJJB693. March.
Hayenga, M.L., VJ. Rhodes, G.A. Grimes, and J.D. Lawrence. 1996. Vertical coordination in
hog production. Prepared for U.S. Department of Agriculture/Grain Inspection, Packers
and Stockyards Administration. Packers and Stockyards Programs. GIPSA-RR 96-5.
May.
Heffeman, W., M. Hendrickson and R. Gronski. 1999. Consolidation in the Food and
Agriculture System. Report to the National Farmers Union. University of Missouri,
Columbia. February.
Heffernan, B. 1999. Jennie-O is industry's largest turkey firm; cutbacks fall short of
expectations. Turkey World. 75(l):10a-10d. January-February.
Heien,D.M. 1975. An Econometric Model of the U.S. Pork Economy. The Review of
Economics and Statistics. 57(3):370-375. August. .
Heimlich, R. and C.H. Barnard. 1995. Economics of Agricultural Management Measures in the
Coastal Zone. AER 698. Washington, DC: U.S. Department of Agriculture, Economic
Research Service.
11-8
-------
Hennessey, D.A., and J.D. Lawrence. 2000. Contractual Relations, Control, and Quality in the
Hog Sector. Review of Agricultural Economics. 21(l):52-67.
Holt, M.T. and S.V. Aradhyula. 1995. Total Response Measures in Systems of Nonlinear
Equations: An Application to a Model of the U.S. Dairy Sector. Canadian Journal of
Agricultural Economics. Volume 43: 285-304.
Holt, M.T. and S.V. Aradhyula. 1990. Price Risk in Supply Equations: An Application of
GARCH Time-Series Models to the U.S. Broiler Market. Southern Economic Journal.
57(l):230-242. July.
Holt, M.T. and S.R. Johnson. 1988. Supply Dynamics in the U.S. Hog Industry. Canadian
Journal of Agricultural Economics. 36(2):313-335. July.
Hoover's. 1998. Corporate Profiles: An inside look at the billionaires. Meat and Poultry Online
Reports. September.
ht^)://pubs.powerize.hoovers.com/nwsstnd/subscrip.nsf/ByPubNameAC?OpenView
Howard, W.H. and C.R. Shumway. 1988. Dynamic Adjustment in the U.S. Dairy Industry.
American Journal of Agricultural Economics. 70(4):837-847. November.
Huang, K.S. 1994. A Further Look at Flexibilities and Elasticities. American Journal of
Agricultural Economics. 76(2):313-317. May.
Huang, K.S. 1993. A Complete System of U. S. Demand for Food. Technical Bulletin Number
1821. Washington, DC: U.S. Department of Agriculture, Economic Research Service.
Huang, K.S. 1985. U.S. Demand for Food: A Complete System of Price and Income Effects.
Technical Bulletin Number 1714. Washington, DC: U.S. Department of Agriculture,
Economic Research Service.
Huang, K.S. and R.C. Haidacher. 1983. Estimation of a Composite Food Demand System for
the United States. Journal of Business and Economic Statistics. 1(4):285-291. October.
Huslin, A. 2000a. Md. aims to tighten chicken waste rules. Published in the Washington Post
(Washington, DC). August 9.
Huslin, A. 2000b. Chickens, manure found piled at Tyson. Published in the Washington Post
(Washington, DC). August 16.
Huy, B., J.G. Elterich, and C.M. Gempesaw II. 1988. Recent Changes in the Regional Structure
of U.S. Dairy Production. Northeastern Journal of Agricultural and Resource Economics.
April.
11-9
-------
EBP, Inc. 1999. Top 200 Companies: Meat Packers, Processors, and Poultry Companies. Meat
Processing. June.
IBP, Inc. 1997. Top 200 Companies: Meat Packers, Processors, and Poultry Companies. Meat
Processing. June.
Idaho Cattle Association. 1999. Idaho Feedlots Economic Analysis. Presentation by Rick Stott
of Agri-Beef Company. Submitted as industry data to EPA. University Extension. July.
Iowa State University. 1999a. Analyzing a Cash Flow Statement. Iowa State University
Extension. http://www.econ.iastate.edU/ADM/C/c3-16.html.
Iowa State University. 1999b. Financial Troubleshooting. Iowa State University Extension.
http://www.econ.iastate.edU/ADM/C/c3-53.html
Iowa State University. 1999c. Interpreting Financial Performance Measures. Iowa State
University Extension. http://www.econ.iastate.edU/ADM/C/c3-56.html
Iowa State University. 1998. Iowa's Pork Industry: Dollars and Scents. Pm-1746. Iowa State
University of Science and Technology. University Extension. January.
IRS (Internal Revenue Service). 1999a. The Complete Internal Revenue Code. New York, NY:
The Research Institute of America, Inc. July.
IRS (Internal Revenue Service). 1999b. How to Depreciate Property. Washington, DC: Internal
Revenue Service. Publication 946.
IRS (Internal Revenue Service). 1999c. Farmer's Tax Guide. Washington, DC: Internal
Revenue Service. Publication 225.
Jacobson, R., and R. Cropp. 1995. Dairy cooperatives and their role in the United States. In:
Dairy Markets and Policy-Issues and Options, Cornell University. No. M-9. August.
Jamagin,B.D. 1996. Financial Accounting Standards: Explanation and Analysis. 18th Edition.
Chicago: CCH, Inc. :
Johnson, R.S., W.J. Wheeler, and L.A. Christensen. 1999a. EPA's Approach to Controlling
Pollution from Animal Feeding Operations: An Economic Analysis. American Journal of
Agricultural Economics. 81:5(1216-1221). December
Johnson, R.S., W.J. Wheeler, and L.A. Christensen. 1999b. EPA's Approach to Controlling
Pollution from Animal Feeding Operations: An Economic Analysis. Materials presented at
11-10
-------
a principal paper session at the American Agricultural Economics Association of (AAEA)
annual meetings in Nashville, TN, 1999.
Kaiser, H.M., and MJ. Morehart. 1994. A Regional Comparison of Farm Costs and Returns
Among Top Dairy Producers. University of Minnesota, Department of Agricultural and
Applied Economics. Staff Paper P94-11. April.
Katzen, S. 1995. Guidance for implementing Title n of S.I., Memorandum for the Heads of
Executive Departments and Agencies from Sally Katzen, U.S. Environmental Protection
Agency. March 31.-
Kellogg, R.L., C. Lander, D. Moffitt, and N. Gollehon. 2000. Manure Nutrients Relative to the
capacity of cropland and Pastureland to Assimilate Nutrients: Spatial and Temporal
Trends for the U.S.. Washington, DC: USD A, National Resources Conservation Service.
Kinnucan, H.W. and O.D. Forker. 1987. Assymetry in Farm-Retail Price Transmission for Major
Dairy Products. American Journal of Agricultural Economics. 69(2):285-292. May.
Kliebenstein, I, J.D. Lawrence, and M. Duffy. 1998. Economics of the Production Industry. In:
Iowa's Pork Industry—Dollars and Scents. Iowa State University. Department of
Economics.
Kliebenstein, J.B., and J.D. Lawrence. 1995. Contracting and Vertical Coordination in the
United States pork industry. American Journal of Agricultural Economics. 77(5):1213-
1218. December. •
Knoeber, C.R. and W.N. Thurman. 1995. Don't Count Your Chickens...: Risk and Risk Shifting
in the Broiler Industry. American Journal of Agricultural Economics. 77(August):486-
496.
Kohls, R., and J. Uhl. 1998. Marketing of Agricultural Products. Upper Saddle River, NJ:
Prentice-Hall.
Krause,K.R. 1991. Cattle Feeding, 1962-89. AER642. Washington, DC: U.S. Department of
Agriculture, Economic Research Service. January.
Kumbahakar, S.C. 1993. Short-Run Returns to Scale, Farm-Size, and Economic Efficiency.
Review of Economics and Statistics. 75(2). May.
Kumbhakar, S.C., B. Biswas, and D. Bailey. 1989. A Study of Economic Efficiency of Utah
Dairy Farmers: A System Approach. Review of Economics and Statistics. 71(4)
595-604. November.
11-11
-------
Lander, C.H., D. Moffitt, and K. Alt. 1998. Nutrient Available from Livestock Manure Relative
to Crop Growth Requirements. Washington, DC: U.S. Department of Agriculture,
Natural Resources Conservation Service, http://nhq.nrcs.usda.gov/land/pubs/nlweb.html
Lawrence, J., G. Grimes, and M. Hayenga. 1998. Production and Marketing Characteristics of .
U.S. Pork Producers, 1997-1998. Staff Paper 311. Iowa State University, Department of
Economics.
Lazarus, W.F., J. Conlin, K. Edberg, S. Carpenter, D. Johnson, and J. Linn. 1999. Generic
Environmental Impact Statement on Animal Agriculture: A Summary of the Literature
Related to Industry Structure and Competitiveness and Profitability and Economic
Viability. Prepared for the Minnesota Environmental Quality Board. Appendix D and E.
September. http://www.mnplan.state.mn.us/eqb/scoping.htmlSe
. LCBP (Lake Champlain Basin Program). 2000. Preliminary Evaluation of Progress Toward
Lake Champlain Basin Program Phosphorus Reduction Goals. Prepared by the Lake
Champlain Steering Committee. June.
Lemieux, CJVL and M.K. Wohlgenant. 1989. Ex Ante Evaluation of the Economic Impact of
Agricultural Biotechnology: The Case of Porcine Somatotropin. American Journal of
Agricultural Economics. 71(4):903-914. November.
Letson, D. and N. Gollehon. 1996. Confined Animal Production and the Manure Problem.
CHOICES. Third Quarter. A publication of the American Agricultural Economics
Association (AAEA). Ames, IA.
MacDonald, J.M., M.E. Ollinger, K.E. Nelson, and C.R. Handy, 2000. Consolidation in U.S.
Meatpacking. AER 785. Washington, DC: U.S. Department of Agriculture, Economic
Research Service. May 22.
http://www.ers.usda.gov/whatsnew/issues/meatpacking/index.htm
Madison, M. 1999. Personal communication between EPA and Milton Madison, Poultry
Specialist, of U.S. Department of Agriculture, Economic Research Service. January.
Manchester, A.C, and D.P. Blayney. 1997. The Structure of Dairy Markets: Past, Present,
Future. AER 757. Washington, DC: U.S. Department of Agriculture, Economic
Research Service.
Marsh, J.M. 1994. Estimating rntertemporal Supply Response in the Fed Beef Market.
American Journal of Agricultural Economics. 76(3):444-453. August.
11-12
-------
Marsh, J.M. 1992. USDA Data Revisions of Choice Beef Prices and Price Spreads: Implications
for Estimating Demand Responses. Journal of Agricultural and Resource Economics.
17(2):323-334.
Marsh, J.M. 1991. Derived Demand Elasticities: Marketing Margin Methods versus an Inverse
Demand Model for Choice Beef. Western Journal of Agricultural Economics. 16(2):382-
391. December.
Martinez, S.W. 1999. Vertical Coordination in the Pork and Broiler Industries: Implications for
Pork and Chicken Products. AER 777. Washington, DC. U.S. Department of
Agriculture, Economic Research Service. April.
http://www.ers.usda.gov/epubs/pdf/aer777/index.htm
Mathews, Jr., K.H., W.F. Hahn, K.E. Nelson, L.A. Duewer, and R.A. Gustafson. 1999. U.S.
Beef Industry: Cattle Cycles, Price spreads, and Packer Concentration. TB Number 1874.
U.S. Department of Agriculture, Economic Research Service. Washington, DC. April.
http://www.ers.usda.gov/epubs/pdf/tbl874/index.htm
McBride,W.D. 1997. Changes in U.S. Livestock Production, 1969-92. AER 754. Washington,
DC: U.S. Department of Agriculture, Economic Research Service. July.
McElroy,R.G. 1993. Are Small Hog Operations Surviving in Today's Market. Agricultural
Income and Finance. AIS-49. June. Washington DC: U.S. Department of Agriculture,
Economic Research Service.
Mclntosh, C.S., T.A. Park, and C. Karnum. 1997. The Potential Impact of Nutrient
Management Legislation on the U.S. Broiler Industry. Selected paper presented at the
Annual Meeting of the Western Agricultural Economics Association, Reno/Sparks, NV.
Meilke, K.D., A.C. Zwart, and L.J. Martin. 1974. North American Hog Supply: A Comparison
of Geometric and Polynomial Distributed Lag Models. Canadian Journal of Agricultural
Economics.
Milton, B. 1999. Personal communication between Eastern Research Group, Inc., and Bob
Milton of U.S. Department of Agriculture, National Agricultural Statistics Service. May
3.
Milton, B. 1998. Personal communication between Eastern Research Group, Inc., and Bob
Milton of U.S. Department of Agriculture, National Agricultural Statistics Service. April
7. . , ' - •
Montgomery, J. 2000. Poultry firms offer to do more. Published in The News Journal
(Wilmington, DE). December 13.
11-13
-------
Moschini, G., D. Moro, and R.D. Green. 1994. Maintaining and Testing Separability in Demand
Systems. American Journal of Agricultural Economics. 76(l):61-73., February.
Moschini, G. and K.D. Meilke. 1989. Modeling the Pattern of Structural Change in U.S. Meat
Demand. American Journal of Agricultural Economics. 71(2):253-261. May.
NCSA (National Cattlemen's Beef Association). 2000. Cattle and Beef Industry Statistics.
March, http://www.beef.org/library/economics/beefsta1/beefstat.htm, Stat99_l 1 .xls.
NCBA (National Cattlemen's Beef Association). 1999. Executive Summary Report on Financial
Survey. Attachment includes data submitted to EPA. November.
NCSU (North Carolina State University). 1999. Alternative Animal Waste Management
Technologies: A Status Report. North Carolina State University, Animal and Poultry
Waste Management Center (APWMC). June 8.
NCSU (North Carolina State University). 1998. Draft of hog and poultry industry
characterization, waste management practices and model detailed analysis of
predominantly used systems. July 28.
NCSU (North Carolina State University). 1994. Livestock Manure Production and
Characterization in North Carolina. Raleigh, NC: North Carolina Cooperative Extension
Service.
NDB (National Dairy Board). 1995. Putting Differences Among World Producer Milk Prices
into Perspective. Dairy Market Report. 3(7). Prepared by National Milk Producers
Federation. July. ,
Nelson, K. and W. Hahn. 1998. Concentration, in the U.S. Beef Packing. USDA Livestock,
Dairy and Poultry Outlook Report. August.
Nerlove, M. and I. Fornari. 1998. Quasi-rational Expectations, An Alternative to Fully Rational
Expectations:, An Application to U.S. Beef Cattle Supply. Journal of Econometrics.
Volume 83.
NEWWT (Northeast Wisconsin Waters for Tomorrow, Inc.). 1994. Toward a Cost-
Effectiveness Approach to Water Resource Management in the Fox-Wolf River Basin: A
First Cut Analysis. Executive Summary. Green Bay, WI.
http://fwb2k.org/newwttechrept3summary.html
http://fwb2k.org/newwtexecsum.html
http://fwb2k.org/newwttechrept5.htm
11-14
-------
NMPF (National Milk Producers Federation). 1999. 1998 Dairy Producer Highlights.
Supplemented with more recent and updated data for 1992 through 1998, provided
through personal correspondence with Carissa Itle of NMPF in December, 1999.
Arlington, VA.
NMPF (National Milk Producers Federation). 1996. Economic Costs of Environmental
Compliance to Milk Producers. Dairy Farming and the Environment. Spring/Summer.
Arlington, VA.
NPGA (National Poultry Growers Association). 1998. Chicken Feed From a Contract.
http://www.web-span.com/pga/contracts/index.html.
*
NPPC (National Pork Producers Council). 1998. Pork facts 1998/1999. Des Moines, IA:
National Pork Producers Council.
Ogishi, A. and D. Zilberman. 1999. Contract farming in the Livestock Sector and Environmental
Protection. University of California, Berkeley, Department of Agricultural and Resource
Economics.
Ohio State University. 1999. Transferring your farm business to the next generation. Ohio State
University Extension. Bulletin 862. http://ohiouIle.ag.ohio-state.edu/b862^862_5.html.
OMB (Office of Management and Budget). 1992. Guidelines and discount rates for benefit-cost
analysis of federal programs. Appendix A. Revised circular No. A-94. Washington, DC:
Office of Management and Budget. October 29.
OSFIA (Occupational Safety and Health Administration). 1999. Federal Register. 29 CFR Parts
1910,1915, et al. Employer Payment for Personal Protective Equipment; Proposed Rule
March 31.
Ospina, E. and C.R. Shumway. 1979. Disaggregated Analysis of Short-run Beef Supply
Response. Western Journal of Agricultural Economics. 4(2):43-59. December.
Outlaw, J.L, D.P. Anderson, and D.I. Padberg. 1997. Relationships Between Market Price
Signals and Production Management: The Case of Fed Beef. Journal of Agricultural and
Applied Economics. 29(l):37-44. July.
Outlaw, J.L., R.E. Jacobson, R.D. Knutson, and R.B. Schwart. 1996. Structure of the U.S.
Dairy Farm Sector In: Dairy Markets and Policy-Issues and Options. Cornell University.
No. M-4. March. -
11-15
-------
Outlaw, J.L., R.B. Schwart, Jr., R.D. Knutson, A.P. Pagano, J.W. Miller, A.W. Gray. 1993.
Impacts of Dairy Waste Management Regulations. Texas A&M University. AFPC Policy
Working Paper 93-4.
Pearce, D.W. and R.K. Turner. 1990. Economics of Natural Resources and the Environment.
Baltimore, MD: The Johns Hopkins University Press.
Perez, A. 2000. Personal communication between EPA and Austin Perez of the Small Business
Administration. March.
Perry, J., D. Banker, and R. Green. 1999. Broiler Farms' Organization, Management, and
Performance. AIB 748. Washington, DC: U.S. Department of Agriculture, Economic
Research Service. March
Poe, G., N. Bills, B. Bellows, P. Crosscombe, R. Koelsch, M. Kreher, and P. Wright. 1999.
Documenting the Status of Dairy Manure Management in New York: Current Practices an
Willingness to Participate in Voluntary Programs. Cornell University, Department of
Agricultural, Resource, and Managerial Economics. Staff Paper 99-03.
Pritchett, J.G. and DJ. Lui. 1998. Estimating Backward Integration in a Primary Input Market:
The Case of the U.S. Hog Industry. Sixth Joint Conference on Food, Agriculture, and the
Environment. Minneapolis, MN. August 31 - September 2.
Purdy, B.M., M.R. Langemeier, and A.M. Featherstone. 1997. Financial Performance, Risk, and
Specialization. Journal of Agricultural and Applied Economics. 29:1(149-61). July.
Putier, D.S. 1992. Incorporating Reference Price Effects into a Theory of Consumer Choice.
Marketing Science. 11(3):287-309. Summer.
Putnam, J.J., and J.E. Allshouse. 1999. Food Consumption, Prices, and Expenditures, 1970-97.
SB 965. Washington, DC: U.S. Department of Agriculture, Economic Research Service.
http://www.ers.usda.gov/epubs/pdf/sb965/index.htm
Putnam, J.J., and J.E. Allshouse. 1997. Food Consumption, Prices, and Expenditures, 1970-95.
SB 939. Washington, DC: U.S. Department of Agriculture, Economic Research Service.
Randall, C.W. 2000. Personal communication between EPA and Dr. Clifford Randall of Virginia
Tech, Blacksburg, VA. August 6.
Randall, C.W., Z. Kisoglu, D. Sen, P. Mitta, and U. Erdal. 1999. Evaluation of Wastewater
Treatment Plants for BNR Retrofits using Advances in Technology.. Final Report.
Virginia Tech, Blacksburg, VA. Submitted to the Point Source Workgroup, Nutrient
Removal Subcommittee, Implementation Committee, Chesapeake Bay Program. May.
11-16
-------
Rasby, R., I. Rush, and R. Stock. 1994. -Wintering and Backgrounding Beef Calves. University
of Nebraska-Lincoln, Institute of Agriculture and Natural Resources, Cooperative
Extension. October. http://marwww.unl.edu/pubs/beef7gl228Jitm
Ray, R. 1999. Personal communication between EPA and Robert Ray of the Small Business
Administration. January.
Rickman, D.S. and R.K. Schwer. 1995. A Comparison of Multipliers of IMPLAN, REMI, and
RIMS IE: Benchmarking ready-made models for comparison. The Annals of Regional
Science. 29(4):363-374.
RMA (Robert Morris Associates). 1998. Annual Statement Studies 1998. Philadelphia, PA:
RMA.
RMA (Robert Morris Associates). 1997. Annual Statement Studies 1997. Philadelphia, PA:
RMA.
RMA (Robert Morris Associates). 1996. Annual Statement Studies,1996. Philadelphia, PA:
RMA.
Rogers, R.T. and R. J. Sexton. 1994. Assessing the Importance of Oligopsony Power in
Agricultural Markets. American Journal of Agricultural Economics. 76(December):1143-
1150.
Rosen, S , K.M. Murphy, and J.A. Schenkman. 1994. Cattle Cycles. Journal of Political
Economy. Volume 102, Number 3.
Rowland, W.W., R. Langemeier, B.W. Schurle, and A.M. Featherstone. 1998. A Nonparametric
Efficiency Analysis for a Sample of Kansas Swine Operations. Journal of Agricultural and
Applied Economics. 30(1). July.
SBA (Small Business Administration). 1998. Size Standards Used to Define Small Business
Concerns. 13 CFR Part 121. Washington, DC: Small Business Administration.
http://www.sba.gov/regulations/121b.htmK121.201.
Schilling, D. 1998. Benchmarking: Profit Punch. Farm Journal. Hogs Today. July/August.
Http://www.farmjoumal.cx)m/magazrnes/^
ACP=3
Schrader, L.F. 1998. Coordination in the United States Hog/Pork Industry. Purdue University,
Department of Agricultural Economics. Staff Paper #98-19. October.
11-17
-------
Seidl, A. and S. Weiler. 2000. Estimated Economic Impact of Colorado Dairies. Agricultural
and Resource Policy Report. APR-00-01. Department of Agricultural and Resource
Economics. Colorado State University-Fort Collins. January.
http://dare.agsci.colostate.edu/extension/pubs.html
Sims, TJ. 1995. Characteristics of Animal Wastes and Waste-Amended Soils: An Overview of
the Agricultural and Environmental Issues. In: Animal Waste and the Land-Water
Interface. CRC.
Skinner, S.P. 1981. Income Variability among Representative Egg Farms. Bulletin 774.
University of Maine, Life Sciences and Agriculture Experiment Station. January.
Smith, L. 1998. Personal communication between Abt Associates and Lou Smith of USDA's
Agricultural Research Service. June.
Snyder, B. 2000. Personal communication between Eastern Research Group, Inc., and Brett
Snyder of the Office of Policy, USEPA. February 1.
Soliman, M.A. 1971. Econometric Model of the Turkey Industry in the United States. Canadian
Journal of Agricultural Economics. 19:47-60. October.
Sommer, J., R. Hoppe, R. Green, and P. Korb. 1998. Structural and financial characteristics of
U.S. farms, 1995: 20th annual family farm report to Congress. AIB 746. Washington,
DC: U.S. Department of Agriculture, Economic Research Service. July
Sommer, J., D. Banker, R. Green, J. Kabacher, N. Peterson, and T. Sun. 1997. Structural and
financial characteristics of U.S. farms, 1994: 19th annual family farm report to Congress.
AIB 735 Washington, DC: U.S. Department of Agriculture, Economic Research Service.
July.
Southard, L. 1999. Lower Output to Revive Hog Prices in 1999. Agricultural Outlook.
Washington, DC: U.S. Department of Agriculture, Economic Research Service. March.
Sporleder, T.L. and T-R. Liu. 1992. USFOOD, A U.S. Food Industries Input-Output Model.
The Ohio State University, Income Enhancement Study No. 157.
Stam, J.M., D.L. Milkove, and G.B. Wallace. 2000. Indicators of Financial Stress in Agriculture
Reported by Agricultural Banks, 1982-99. Agricultural Income and Finance. AIS-74.
Washington, DC: U.S. Department of Agriculture, Economic Research Service. February.
" http://usda.mannlib/comell.edu/reports/erssor/economics/ais-bb/2000/ais74.pdf
11-18
-------
Stam, J.M., S.R. Koenig, S.E. Bently, H.F. Gale, Jr. 1991. Farm Financial Stress, Farm Exits,
and Public Sector Assistance to the Farm Sector in the 1980's. AER 645. Washington,
DC: U.S. Department of Agriculture, Economic Research Service.
Staples, D. 1998. Personal communication between EPA and David Staples of the United Egg
Producers. February.
Stiegert, K.W., A. Azzam, and B.W. Brorsen. 1993. Markdown Pricing and Cattle Supply in the
Beef Packing Industry. American Journal of Agricultural Economics. August.
Stott, R. 2000a. Personal communication between EPA and Rick Stott, Agri-Beef. August 8.
Followup to correspondence and data on the beef industry from R. Stott to R. Johnson,
EPA (dated July 24).
Stott, R. 2000b. Personal communication between Eastern Research Group, Inc. and R. Stott,
Agri-Beef. July 10.
Tanjuakip, R.V., C.M. Gempesaw, and G.J. Elterich. 1992. An Optimal Control Framework for
Inter-Regional Dairy Policy Analysis. Southern Journal of Agricultural Economics.
December.
Tauer, L.W. 1995. Do New York Dairy Farmers Maximize Profits or Minimize Costs?
American Journal of Agricultural Economics. 77(2). May.
Thornton, G. 1999. Nation's Broiler Industry. Broiler Industry. January.
Thurman, W. and M. Wohlgenant. 1989. Consistent Estimation of General Equilibrium Welfare
Effects. American Journal of Agricultural Economics. 71:1041-1045.
Thurman, W.N., 1987. The Poultry Market: Demand Stability and Industry Structure. American
Journal of Agricultural Economics. 69(l):30-37. February.
Thurman, W.N. 1986. Endogeneity Testing in a Supply and Demand Framework. The Review
of Economics and Statistics 68:638-646.
Tippett, J.P., and R.C. Dodd. 1995. Cost-Effectiveness of Agricultural BMPs for Nutrient
Reduction in the Tar-Pamlico Basin. Prepared by Research Triangle Institute for the
North Carolina Department of the Environment, Health, and Natural Resources. January.
http://www.epa.gov/owowwtrl/watershed/Proceed/mccarthy.html
Tomek, W.G. and K.L. Robinson. 1972. Agricultural Product Prices. Ithaca, NY: Cornell
University Press.
11-19
-------
University of Missouri. 1999. An Economic Analysis of Manure Technologies on Missouri
Swine Operations: A draft report. University of Missouri-Columbia, Commercial
Agriculture Program. April 30.
U.S. Census Bureau. 2000. 1997 NAICS Matched to 1997 SIC Sectors and Subsectors.
http://www.census.gov/epcd/naics/naics3dx.htmSNll
•
U.S. Census Bureau. 1999. Statistical Abstract of the United States: 1999.
http://www.census.gov/prod/www/statistical-abstract-us.html
USDA (U.S. Department of Agriculture). 1999. Correspondence from Dr. Ralph Heimlich,
Deputy Director for Analysis, Economic Research Service, USDA, to Virginia Kibler,
Economist, Office of Wastewater Management, U.S. EPA. October 2.
USDA (U.S. Department of Agriculture). 1998. Report of the USDA National Commission on
SmaU Farms: A Time to Act MP-1545. January.
http://www.reeusda.gov/agsys/smallfarm/report.htm
USDA (U.S. Department of Agriculture). 1992. Agricultural Waste Management Field
Handbook. National Engineering Handbook (NEH). Part 651.
http://www.ncg.nrcs.usda.gov/awrnfh.html.
USDA and USEPA (U.S. Department of Agriculture and U.S. Environmental Protection
Agency). 1999. Unified National Strategy for Animal Feeding Operations. March 9.
http://www.epa.gov/owm/finafost.hmi
USDA/APHIS (U.S. Department of Agriculture, Animal and Plant Health Inspection Service).
1995a. Cattle on Feed Evaluation. Part I: Feedlot Management Practices. January.
h11p://www.aphis.usda.gov/vs/cean/cahm/Beef_Feedlot/cofdesl.pdf
USDA/APHIS (U.S. Department of Agriculture, Animal and Plant Health Inspection Service).
1995b. Swine '95. Reference of 1995 Swine Management Practices: Parts I and n.
October. http://www.aphis.usda.gov/vs/ceah/cahrn/Swine/sw95desl.pdf
USDA/APHIS (U.S. Department of Agriculture, Animal and Plant Health Inspection Service).
1993. Contract Heifer Raising. National Dairy Heifer Evaluation Program (NDHEP).
March. http://www.apMs.usda.gov/vs/ceah/cahrn/Dairy_Cattle/ndhep/ndhepcon.pdf
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 2000a. ERS' New
ARMS Survey. January, http://www.econ.ag.gov/briefing/fbe/sf/sf2.htm
11-20
-------
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 2000b. Farm
Typology. July.
http://www.econ.ag.gov^riefmg/farms1ructiire/Glossary/Text/1^ology.litni
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 2000c. Hog Costs
and Returns, 1998 Costs of Production from the Agricultural Resource Management
Survey. ht1p://www.ers.usda.gov/briefmg/farmmcome/car/hogs2.htm
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 2000d. Off-Farm
Income Supports Many Farm Households. Agricultural Income and Finance. AIS-74.
February. •
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 2000e. USDA's
glossary of farm structure terms.
http://www.ers.usda.gov/briefrng/farm
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 2000f. Statistical
Indicators." Agricultural Outlook. Washington, DC: U.S. Department of Agriculture,
Economic Research Service. March.
http://www.ers.usda.gov/publications/agoutlook/mar2000/ao2069i.pdf
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 2000g. Farm Sector
Debt Anticipated to Stabilize in 2000. Agricultural Income and Finance. AIS-75.
Washington, DC: U.S. Department of Agriculture, Economic Research Service.
September, http://usda.mannlib.comell.edu/reports/erssor/economics/ais-
bb/2000/ais75.pdf
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 2000h. Share of
farms and value of production, by farm typology group, 1997. 1997 Agricultural
Resource Management Study.
http://www.ers.usda.gov/briefmg/fannstructare/qa/Graph/Sharefrni.gif
USDA/ERS (U;S. Department of Agriculture, Economic Research Service). 2000J. Sixty-Eight
Percent of Farm Businesses Entered 2000 hi a Financially Favorable Position. AIS-75.
Washington, DC: U.S. Department of Agriculture, Economic Research Service.
September, http://usda.mannlib.comell.edu/reports/erssor/economics/ais-
bb/2000/ais75.pdf
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1999a. Data from
the Farm Costs and Returns Survey (FCRS) database (11,724 observations). February 17.
• Revised March 25.
11-21
-------
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1999c. Livestock,
Dairy and Poultry Outlook. January 26.
http ://www.ers.usda.gov/publcations/so/view.asp?MivestocMdp-mbb/
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1999d. Livestock,
Dairy and Poultry Outlook. November 23.
http://www.ers.usda.gov/publications/so/view.asp7Mivestock/ldp-mbb/
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1998a. Livestock,
Dairy and Poultry Outlook. May 21.
http://www.ers.usda.gov/publieations/so/view.asp7Mivestock/ldp-mbb/
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1998b. Livestock,
Dairy and Poultry Outlook. December 29.
http://www.ers.usda.gov/publicarions/so/view.asp7Mivestock/ldp-mbb/
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1998c. Livestock,
Dairy and Poultry Outlook. April 21.
http://www.ers.usda.gov/publications/so/view.asp7Mivestock/ldp-mbb/
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1998d. Income
statement for dairy farm businesses, 1993-97.
http://www.econ.ag.gov/briefing/fbe/sf/result97/lotus/incsicl 8.html.
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1997a.
Classification of Overall Financial Performance.
ht1p://www.ers.usda.gov/briefmg/fbe/sf'result96/class.htm
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1997b. Financial
Performance of U.S. Commercial Farms, 1991-94. Agricultural Economic Report No.
751. June.
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1997d. Poultry
Yearbook: 1996. March, http://usda.mannlib.cornell.edu/data-sets/livestock/89007/
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1997e. Staff
Briefing Paper. Financial Performance of U.S. Farm Businesses, 1996.
http://www.econ.ag.gov/briefing/fbe/sf/result96^rief96.htm
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1997f. Livestock,
Dairy and Poultry Outlook. November 19.
http://www.ers.usda.gov/publications/so/view.asp7Mivestock/ldp-mbb/
11-22
-------
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1996a. The
Economic Well-being of Farmer Operator Households, 1991-92. AER-666. Updated
5/96. .
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1996b. Farm'
Business Economic Report, 1995. ECI-1996.
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1996c. Farmers'
use of marketing and production contracts. AER747. December.
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1996d. Economic
Indicators of the Farm Sector: Costs of Production - Major Field Crops and Livestock and
Dairy, 1992. ECIFS 12-3. In Outlaw et al., 1996.
USDA/ERS (U.S. Department of Agriculture, Economic Research Service). 1996e. 1991-95
Farm Costs and Returns Survey; 1996 Agricultural Resource Management Study.
USDA/ERS (U.S. Department'of Agriculture, Economic Research Service). 1993. Farm
Household Income Estimates Provide Additional Perspectives on Farm Families.
Agricultural Income and Finance. AIS-49. Washington, DC: U.S. Department of
Agriculture, Economic Research Service. June.
USDA/GIPSA (U.S. Department of Agriculture/Grain Inspection, Packers and Stockyards
Administration). 1998. Period data on calendar-year Federally inspected slaughter. As
cited in USDA/GIPSA's 1999 Annual Report, http://www.usda.gov/gipsa/pubs/99ar.pdf
USDA/GIPSA (U.S. Department of Agriculture/Grain Inspection, Packers and Stockyards
Administration). 1997. Review of Western Organization of Research Council's (WORC)
Petition for Rulemaking. August. http://www.usda.gov/gipsa/worc__petition/worc.htm
USDA/GIPSA (U.S. Department of Agriculture/Grain Inspection, Packers and Stockyards
Administration). 1996a. Concentration in the Red Meat Packing Industry. Packers and
Stockyards Programs. February.
USDA/GIPSA (U.S. Department of Agriculture, Grain Inspection, Packers and Stockyards
Administration). 1996b. Packers and Stockyards Statistical Report 1995. Packers and
Stockyards Programs. GIPSASR-97-1.
USDA/GIPSA (U.S. Department of Agriculture, Grain Inspection, Packers and Stockyards
Administration). 1996c. Price Determination in Slaughter Cattle Procurement. Texas
A&M Agricultural Market Research Center. Slaughter Cattle Procurement and Pricing
Team. GIPSA-RR 96-2.
11-23
-------
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 2000a.
Farm Production Expenditures. 1999 Summary. SpSyS(OO). July.
http://usda.marmlib.cornell.edu/jeports/nassr/price/2pe-bb/Q)ex0700.pdf
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 2000b.
1997 Census of Agriculture General Information and Frequently Asked Questions.
January, http ://www.nass.usda.gov/census/census97/cenfaqs.htm
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1999a.
1997 Census of Agriculture, Volume 1, Part 51, Chapter 1, United States summary and
state data, http://www.nass.usda.gov/census/
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1999b.
Cattle Final Estimates 1994-1998. January.
http://usda.mannlib.comell.edu/reports/general/sb/b9530199.pdf
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1999c.
Milk Cows and Production: Final Estimates 1993-97. Statistical Bulletin Number 952.
http://usda.mannlib.cornell.edu/reports/general/sb
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1999d.
Poultry Production and Value. Final Estimates: 1994-1997. SB 958. May.
http://usda.mannlib.comell.edu/reports/general/sb/b9580399.pdf
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1998a.
Agricultural Prices 1997 Summary. July.
http://usda.mannlib.comell.edii/reports/nassr/price/zap-bb/agpran98.pdf
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1998b.
Chicken and Eggs. Final Estimates: 1994-1997. SB 944. December.
http://iisda.mannlib.comell.edu/reports/general/sb/b9441298.pdf
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1998c.
Farm Labor. May 22. http://usda.mannlib.comell.edy/reports/nassr/other/pfl-
bb/1998/farm_labor_05.22.98
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1998d.
Livestock Slaughter 1997 Summary. Mt An 1-2-1 (98). March.
http://usda.mannlib.comell.edu/reports/nassr/livestock/pls-
bban/Uvestock_slaughter_annual_summary_03.06.98
11-24
-------
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1998e.
Meat Animals Production, Disposition, and Income: 1997 Summary. April.
http://usda.mannUb.comell.edu/reports/nassr/livestock/zma-bb/meat_animals_04.24.98
USDA/NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 1998f.
Poultry Production and Value: 1997 Summary. April.
http://usda.mannlib.comell.edu/usda/
USDA/NRCS (U.S. Department of Agriculture, Natural Resources Conservation Service). 1999.
Fact Sheet: Environmental Quality Incentives Program.
http://www.nhq.nrcs.usda.gov/OPA/FB96OPA/eqipfact.htm
USDA/WAOB (U.S. Department of Agriculture, World Agricultural Outlook Board). 2000.
USDA Agricultural Baseline Projections to 2009. Staff Report WAOB-00-1. February.
http://www.ers.usda.gov/publicationsAVAOB001/
USDA/WAOB (U.S. Department of Agriculture, World Agricultural Outlook Board). 1999.
USDA Agricultural Baseline Projections to 2008. Staff Report WAOB-99-1.
http://www.ers.usda.gov/publications/WAOB991/
USDC (U.S. Department of Commerce). 1999a. 1997 Census of Manufactures. Includes
industry series reports: Animal (Except Poultry) Slaughtering (EC97M-3116A), Meat
Processed From Carcasses (EC97M-3116B), Poultry Processing (EC97M-3116D), All
Other Miscellaneous Food Manufacturing (EC97M-3119H), Fluid Milk Manufacturing
(EC97M-3115 A), Creamery Butter Manufacturing (EC97M-3115B), Cheese
Manufacturing (EC97M-3115C), Dry, Condensed and Evaporated Dairy Product
Manufacturing (EC97M-3115D), and Ice Cream and Frozen Dessert Manufacturing
(EC97M-3115E). http://www.census.gov/prod/www/abs/97ecmani.html
USDC (U.S. Department of Commerce). 1999b. Processed Food and Beverages. U.S. Industry
and Trade Outlook'99.
USDC (U.S. Department of Commerce). 1997a. 1997 County Business Patterns.
http://www.census.gov.pub/epcd/cbp/view/us97.txt
USDC (US Department of Commerce, Bureau of Economic Analysis). 1997b- Regional
Multipliers, A User Handbook for the Regional Input-Output Modeling System (RIMS
II).
USDC (U.S. Department of Commerce, Bureau of Economic Analysis). 1996. Regional Input-
Output Modeling System (RIMS H) Data Files. June 19.
11-25
-------
USDC (U.S. Department of Commerce). 1994. 1992 Census of Agriculture. Volume 1 (Part
51).
USDC (U.S. Department of Commerce). 1989. 1987 Census of Agriculture. Volume 1 (Part
51).
USDC (U.S. Department of Commerce). 1984. 1982 Census of Agriculture. Volume 1 (Part
51).
USDC (U.S. Department of Commerce). 1980. 1978 Census of Agriculture. Volume 1 (Part
51).
USDC (U.S. Department of Commerce). 1976. 1974 Census of Agriculture. Volume 1 (Part
51).
USEPA (U.S. Environmental Protection Agency). 2000a. Development Document for the
Proposed Revisions to the National Pollutant Discharge Elimination System Regulation
and the Effluent Guidelines for Concentrated Animal Feeding Operations.
USEPA (U.S. Environmental Protection Agency). 2000b. Environmental Assessment of the
Proposed Revisions to the National Pollutant Discharge Elimination System Regulation
and the Effluent Guidelines for Concentrated Animal Feeding Operations.
USEPA (U.S. Environmental Protection Agency). 2000c. Economic Assessment for Final
Action Regarding Pretreatment Standards for the Industrial Laundries Point Source
Category. March.
USEPA (U.S. Environmental Protection Agency). 2000d. Environmental and Economic Benefit
Analysis of the Proposed Revisions to the National Pollutant Discharge Elimination
System Regulation and the Effluent Guidelines for Concentrated Animal Feeding
Operations.
USEPA (U.S. Environmental Protection Agency). 2000e. EPA Alternative Small Business
Definition for Animal Feeding Operations. Memorandum to the Rulemaking Record and
unpublished analysis. R.S. Johnson, U.S. Environmental Protection Agency, Office of
Water. July 7.
USEPA (U.S. Environmental Protection Agency). 2000f. EPA Alternative Small Business
Definition for Chicken Egg Sector. Memorandum to the Rulemaking Record and
unpublished analysis. R.S. Johnson, U.S. Environmental Protection Agency, Office of
Water. July 7.
11-26
-------
USEPA(U S. Environmental Protection Agency). 2000g. Final Report of the SBREFA Small
Business Advocacy Review Panel on National Pollutant Discharge Elimination System
(NPDES) and Effluent Limitations Guidelines for Concentrated Animal Feeding
Operations (CAFOs). March. http:7Avww.epa.gov/sbrefa
USEPA (U.S. Environmental Protection Agency). 2000h. National Water Quality Inventory:
1998 Report to Congress. EPA-841-F-00-006. June.
http://www.epa.gov/305b/98report/index.html
USEPA (U.S. Environmental Protection Agency). 2000L Supporting Statement for the
Information Collection Request for the Proposed Regulatory Revisions to the National
Pollutant Discharge Elimination System Regulations for Concentrated Animal Feeding
Operations and Feedlot Effluent Limitations Guidelines. July 31.
USEPA (U.S. Environmental Protection Agency). 2000J. Toxic and Pollutant Weighting Factors
of the Effluent Limitations Guidelines for Concentrated Animal Feeding Operations.
USEPA (U.S. Environmental Protection Agency). 2000k. Approach for Estimating Value of
Poultry Litter. Memorandum to the Rulemaking Record and unpublished analysis. P.
Shriner, U.S. Environmental Protection Agency, Office of Water. September.
USEPA (U.S. Environmental Protection Agency). 20001. State Compendium: Programs and
Regulatory Activities Related to Animal Feeding Operations. USEPA, Office of
Wastewater Management, http://www.epa.gov/owm/afos/resources.htai
USEPA (U.S. Environmental Protection Agency). 2000m. Memorandum documenting a
meeting between EPA personnel and the National Cattlemen's Beef Association. June 6.
USEPA (U.S. Environmental Protection Agency). 2000n. Memorandum documenting EPA's
use and assessment of NCBA's Financial Survey Data by R.S. Johnson, EPA. December.
USEPA (U.S. Environmental Protection Agency). 2000o. Economic Analysis of Final Effluent
Limitations Guidelines and Standards for the Landfills Point Source Category. EPA-821-
B-99-005.
USEPA (U.S. Environmental Protection Agency). 1999a. Correspondence from Geoffrey H.
Grubbs, Director, Office of Science and Technology, and Michael B. Cook, Director,
Office of Wastewater Management, USEPA, to Mr. Jere Glover, Chief Counsel for
Advocacy, U.S. Small Business Administration. November 19.
USEPA (U.S. Environmental Protection Agency). 1999b. Correspondence from Neil Patel,
Branch Chief, Office of Science and Technology, USEPA, to Dr. Peggy Caswell, Branch
Chief, Economic Research Service, USD A. March 22.
11-27
-------
USEPA (U.S. Environmental Protection Agency). 1999c. Correspondence from Neil Patel,
Branch Chief, Office of Science and Technology, USEPA, to Dr. Peggy Caswell, Branch
Chief, Economic Research Service, USD A. March 25.
USEPA (U.S. Environmental Protection Agency). 1999d. Correspondence from Tudor T.
Davies, Director, Office of Science and Technology, U.S. Environmental Protection
Agency, to Mr. Jere Glover, Chief Counsel for Advocacy, U.S. Small Business
Administration. March 12.
USEPA (U.S. Environmental Protection Agency). 1999e. Correspondence from Virginia Kibler,
Economist, Office of Wastewater Management, U.S. EPA, to Dr. Ralph Heimlich, Deputy
Director for Analysis, Economic Research Service, USD A. December 13.
USEPA (U.S. Environmental Protection Agency). 1999f. Economic Analysis of the Final Phase
n Storm Water Rule. Final Report. October, http://www.epa.gov/cgi-bin/claritgw7op-
Display&documeritr^lser:epa-cirib:466;&rank?=2&template=EPA
USEPA (U.S. Environmental Protection Agency). 1999g. Economic Analysis of Proposed
Effluent Limitations Guidelines and Standards for the Transportation Equipment Cleaning
Industry Point Source Category. May. http://www.epa.gov/ost/guide/teci
USEPA (U.S. Environmental Protection Agency). 1999h. Economic Data Summary.
Preliminary Data Summary! Feedlots Point Source Category Study. Appendix 2.
Prepared by Abt Associates, Inc., for EPA's Office of Water, Engineering "and Analysis
Division. 'EPA-821-R-99-002. January.
http://www.epa.gov/OST/guide/feedlots/pdsappii.pdf
USEPA (U.S. Environmental Protection Agency). 1999L Revised Interim Guidance for EPA
Rulewriters: Regulatory Flexibility Act as amended by the Small Business Regulatory
Enforcement Fairness Act. March 29. http://www.epa.gov/sbrefa/documents/igui99.pdf
USEPA (U.S. Environmental Protection Agency). 1999J. Memorandum documenting meetings
between EPA personnel and National Turkey Federation. December 1998 and February
1999.
USEPA (U.S. Environmental Protection Agency). 1999k. Memoranda and email documenting
correspondence between EPA personnel and the National Cattlemen's Beef Association
(NCBA) regarding NCBA's data survey. July and November.
USEPA (U.S. Environmental Protection Agency). 19991. SBA meeting on alternative definition.
Memorandum to the Rulemaking Record. R.S. Johnson, U.S. Environmental Protection
Agency, Office of Water. December 15.
11-28
-------
USEPA (U.S. Environmental Protection Agency). 1999m. Submitted to rulemaking record.
Electronic files of aggregated ARMS financial data received by USD A, ERS and
spreadsheet versions of these files converted by EPA.
USEPA (U.S. Environmental Protection Agency). 1999n. Correspondence from Sheila Frace,
Director, Engineering and Analysis Division, U.S. Environmental Protection Agency, to
Mr. Austin Perez, Assistant Advocate, U.S. Small Business Administration, December 15.
USEPA (U.S. Environmental Protection Agency). 1999o. Economic Analysis of Final Effluent
Limitations Guidelines and Standards for the Commercial Hazardous Waste Combustors.
EPA-821-B-99-008.
USEPA (U.S. Environmental Protection Agency). 1998a. Correspondence from Renee Johnson,
Economist, Office of Science and Technology, EPA, to Dr. Lee Christensen, Economist,
Economic Research Service , USD A. October 27.
USEPA (U.S. Environmental Protection Agency). 1998b. Economic Analysis of Final Effluent
Limitations Guidelines and Standards for the Pharmaceutical Manufacturing Industry.
July. .
USEPA (U.S. Environmental Protection Agency). 1998c. Economic Analysis of Effluent
Limitations Guidelines and Standards for the Centralized Waste Treatment Industry. EPA
821/R-98-019.
USEPA (U.S. Environmental Protection Agency). 1997a. Economic Analysis for the National
Emission Standards for Hazardous Air Pollutants for Source Category: Pulp and Paper
Production; Effluent Limitations Guidelines, Pretreatment Standards, and New Source
Performance Standards: Pulp, Paper, and Paperboard Category-Phase I. EPA 68-C3-
0302. October 27.
USEPA (U.S. Environmental Protection Agency). 1997b. Economic Assessment for Proposed
Pretreatment Standards for Existing and New Sources for the Industrial Laundries Point
Source Category.
USEPA (U.S. Environmental Protection Agency). 1996. Economic Analysis of Final Effluent
Limitations Guidelines and Standards for the Pesticide Formulating, Packaging, and
Repackaging Industry. EPA 821-R-96-017. September.
USEPA (U.S. Environmental Protection Agency). 1995a. Economic Impact Analysis and
Regulatory Flexibility Analysis of Proposed Effluent Guidelines for the Pharmaceutical
Manufacturing Industry. EPA 821/R-95-018. September.
11-29
-------
USEPA (U.S. Environmental Protection Agency). 1995b. Interim Economic Guidance for Water
Quality Standards: Workbook. EPA-823-B-25-002. March.
USEPA (U.S. Environmental Protection Agency). 1995c. Economic Impact Analysis of
Proposed Effluent Guidelines and Standards for the Metal Products and Machinery
Industry (Phase 1). EPA 821-R-95^022. April.
USEPA (U.S. Environmental Protection Agency). 1994. Medical Waste Incinerators -
Background Information for Proposed Standards and Guidelines: Analysis of Economic
Impacts for New Sources. EPA 453/R-94-047a. July.
USEPA (U.S. Environmental Protection Agency). 1993. Economic Impact and Regulatory
Flexibility Analysis of the Proposed Effluent Guidelines and NESHAP for the Pulp, Paper,
and Paperboard Industry. EPA 821-R-93-021. November.
USEPA (U.S. Environmental Protection Agency). 1987. Economic Impact Analysis of Effluent
Limitations Guidelines and Standards for the Organic Chemicals, Plastics and Synthetic
Fibers Industry. EPA 440/2/87-007. September.
USGAO (U.S. Government Accounting Office). 1997. Drinking Water: Information on the
Quality of Water Found at Community Water Systems and Private Wells.
USGAO (General Accounting Office). 1995. Animal Agriculture: Information on Waste
Management and Water Quality Issues. GAO/RCED-95-200BR. June 28.
USGPO (U.S. Government Printing Office). 2000. Small Business Size Regulations; Size
Standards and the North American Industry Classification System; Correction. 13 CFR
Part 121. Washington, DC: Small Business Administration. Federal Register. 65(172):
53533-53558. September 5. http://www.sba.gov/library/lawroom.html
USGPO (U.S. Government Printing Office). 1999. Standard for Pesticide Containers and
Containment. Federal Register. 64(203): 56917-56944. October 21. Additional
supporting information of EPA's economic analysis for the proposed rulemaking is in
"Regulatory Impact Analysis: Standards for Pesticide Containment Structures Under the
Federal Insecticide, Fungicide, and Rodenticide Act as Amended, 1988" (1993).
USGPO (U.S. Government Printing Office). 1998. Notice of the Proposed Effluent Guidelines
Plan. Federal Register. 63(102): 29203-29213. May 28.
http://frwebgate2access.gpo.gov/waisgate.cgi?WAISdocID=889397817+0+0+0&
WAISaction=retriev
11-30
-------
USGPO (U.S. Government Printing Office). 1996. Several minor corrections were published on
February 23,1996 (61 FR 6421). Federal Register. Vol. 61, No. 3175. January 31,
1996. http://www.sbaonline.sba.gov/gopher/Financial-Assistance/Size-Standards.
USGPO (U.S. Government Printing Office). 1991a. Small Business Size Standards; Egg
Chicken Industry. Vol. 56, No. 83, 19821-19825. April. .
USGPO (U.S. Government Printing Office). 1991b. Small Business Size Standards; Egg
Chicken Industry. Vol. 56, No. 209, 55617. October.
USGPO (U.S. Government Printing Office). 1986. Best Conventional Pollutant Control
Technology. Effluent Limitations Guidelines. Final Rule. Federal Register.
51(103):24974-25002. July 9.
USITC (U.S. International Trade Commission). 1998a. Industry & Trade Summary: Dairy
Products. USITC Publication 3080. Washington DC. December.
http://www.usitc.gov/wais/reports/arc/w3080.htm
USITC (U.S. International Trade Commission). 1998b. Industry & Trade Summary: Poultry.
USITC Publication 3148. Washington DC. :
http://www.usitc.gov/wais/reports/arc/w3148.htm
. Uvacek, E. 1999. High 70's, Low 80's by Spring. Ed's Market Comments. Home Farm.
January, http://www.homefarrn.com/beef7ernc/9901.htm.
Van Arsdall, R.N. and K.E. Nelson. 1985. Economies of Size hi Hog Production. TB 1712.
Washington, DC: U.S. Department of Agriculture, Economic Research Service.
December.
Van Home, James C. 1986. Financial Management and Policy (7th ed.). Englewood-Cliffs, NJ:
Prentice Hall/
Vukina, T. 2000. Personal communication between Eastern Research Group, Inc., and Tomislav
Vukina of North Carolina State University. March 3.
Ward, C.E. 1997.. Vertical Integration Comparison: Beef, Pork, and Poultry. Selected paper
submitted to the Western Agricultural Economics Association. February.
Ward, C.E. and T.C. Schroeder. No Date. Structural Changes in Cattle Feeding and Meat .
Packing.
http://www.agecon.uwyo.edu/AGECON/ProgramsMarketmg/MngTCMkt^efault.htm.
11-31
-------
Warner, M. 1998. Changes in Animal Agricultural Production. In: Animal Agriculture in South
Carolina: A Fact Book. EER 172. Clemson University. November.
Weersink, A. and L.W. Tauer. 1991. Causality Between Dairy Farm Size and Productivity.
American Journal of Agricultural Economics. 73(4). November.
Weersink, A. and W. Howard. 1990. Regional Adjustment Response in the U.S. Dairy Sector to
Changes in Milk Support Price. Western Journal of Agricultural Economics. 15(1):13-21.
July.
Westat. 2000. Methodology for imputing missing or omitted USD A Census data. Memorandum
from Westat to M. Smith, U.S. Environmental Protection Agency, Office of Water.
Westerbarger, D.A. and D. Letson. 1995. Livestock and Poultry Waste-Control Costs.
CHOICES. Second Quarter. A publication of the American Agricultural Economics
Association (AAEA). Ames, IA.
Wiedeman, A. 1998. Correspondence from Allison Wiedeman, Point Source Coordinator,
Chesapeake Bay Program, U.S. Environmental Protection Agency, to Mr. John Adsit,
Stewards of Jackson River. May 3.
Wohlgenant, M.K. 1989. Demand for Farm Output in a Complete System of Demand Functions.
American Journal of Agricultural Economics. 71(2):241-252. May.
Wolf, C.A. and L.G. Hamm. 1998. The Role of Cooperatives in Milk Marketing. 1998
American Agricultural Economics Association, Annual Meeting Selected Paper.
Yeske,P. 1996. Efficient Pork Production. Presented at the 1996 Annual Meeting of the
American Association of Swine Practitioners. Perry, Iowa.
Zulovich, J. 2000. Personal communication between Eastern Research Group, Inc., and Joseph
Zulovich of University of Missouri. August 21.
11-32
-------
APPENDIXA
COST AJNimJALIZATION MODE
This appendix provides an overview of the cost annualization model used by EPA. A
discussed in Section 4 of this report, the annualization-model calculates four types of compliance
costs for a model CAFO:
• Present value of expenditures—before-tax basis
• Present value of expenditures—after-tax basis
• Annualized cost—before-tax basis
• Annualized cost—after-tax basis
The following sections present the input data and assumptions (Section A.1) and provide details
about .the workings of the annualization model (Section A.2). All tables in this Appendix are
provided at the end of Appendix H.
A.1 INPUT DATA SOURCES
There are four key data inputs to the cost annualization model, including:
• Capital and O&M costs (including startup, recurring, and annual O&M costs)
» Depreciable life of the asset
• Discount rate
• Marginal tax rate
The capital and O&M costs that EPA uses in the cost annualization model are developed
by EPA. The capital cost is the initial investment needed to purchase and install the structure; it is
a one-time cost. The O&M cost is the annual cost of operating and maintaining the structure.
O&M costs can be incurred in the first year (startup O&M costs), in periodic intervals (recurring
O&M costs), or every year of the structure's operation (annual O&M costs).
The depreciable life of the asset refers to EPA's assumption on the time period used to
depreciate capital improvements that are made due to the proposed CAFO regulations.
A-l
-------
EPA's annualization model uses a real discount rate of 7 percent, as recommended by the
Office of Management and Budget (OMB) (OMB, 1992). EPA assumes this input to be a real
interest rate and therefore it is not adjusted for inflation.
The marginal tax rate (used to compute tax shield) depends on the amount of taxable
earnings (estimated as net cash income minus depreciation plus value of inventory) at the model
CAFO. Inputs to the cost annualization model to calculate an average operation's tax shield
include both federal and state tax rates.
Additional information about compliance cost estimates and development of the model
CAFOs is provided in Section 4 of this report. Detailed information about the costs used as
inputs to the annualization model is provided in the Development Document (USEPA, 2000a).
Section A. 1.1 below discusses the tax rate and Section A. 1.2 discusses the depreciation
method of the annualization model schedule in more detail.
A.1.1 Marginal Tax Rate
EPA conducts its financial analysis at the CAFO level using representative average
models. The cost annualization model uses as inputs both federal and state tax rates to calculate
an average operation's tax shield (see Table A-3 for sample worksheet). For this analysis, EPA
uses federal and state corporate income tax rates because it is not possible to definitively identify
whether CAFOs represented by each model pay taxes at the corporate or the individual rate.
Table A-3 lists the range of federal tax rates that EPA assumes for this analysis that are
attributed to model CAFOs based on estimated taxable earnings. As shown, federal tax rates
range from 15 percent to 34 percent, depending on the amount of taxable income at a facility
(CCH, 1999b). As an example, using these rateSj model CAFOs with earnings greater than or
equal to $335,000 would be assigned the federal tax rate of 34 percent; model CAFOs with
earnings greater than or equal to $100,000 but less than $335,000 would be assigned a tax rate of
28.3 percent. Examples of taxable income levels at EPA's model CAFOs are presented in
Sections 6 through 8 that show average income statements for each sector.
Table A-l lists each state's top corporate tax rates, as well as rates on individual income
(CCH, 1999a and CCH, 1995). The cost annualization model refers to reported average state tax
rates, however, because of the uncertainty over which state tax rate to apply to a given model
CAFO, EPA uses the national average across all states. Table A-l lists the national average value
that EPA assumes for this analysis (CCH, 1999a and CCH, 1995). As shown, the average
national rates are 6.6 percent (corporate income) and 5.8 percent (personal income). As
discussed previously, EPA uses the higher corporate income tax rate for this analysis.
A-2
-------
Table A-l. State Income Tax Rates
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
Corporate Income
Tax Rate
5.00%
9.40%
9.00%
6.50%
9.30%
5.00%
11.50%
8.70%
5.50%
6.00%
6.40%
8.00%
4.80%
3.40%
12.00%
4.00%
8.25%
8.00%
8.93%
7.00%
9.50%
2.30%
9.80%
5.00%
6.25%
6.75%
7.81%
0.00%
7.00%
Basis for States
With Graduated
Tax Tables
$90,000+
$100,000+
$100,000+
$250,000+
$50,000+
$250,000+
$200,000+
$250,000+
,-
$10,000+
$50,000+
Personal Income
Tax Upper Rate
5.00%
0.00%
6.90%
7.00%
11.00%
5.00%
4.50%
7.70%
0.00%
6.00%
10.00%
8.20%
3.00%
3.40% .
9.98%
7.75%
6.00%
6.00%
8.50%
6.00%
5.95%
4.40%
8.50%
5.00%
6.00%
11.00%
6.99%
0.00%
0.00%
Basis for States
With Graduated
Tax Tables
$3,000+
$150,000+
$25,000+
$215,000+
$40,000+
$7,000+
$21,000+
$20,000+
$47,000+
$30,000+
$8,000+
$50,000+
$33,000+
$100,000+
$50,000+
$10,000+
$9,000+
$63,000+
$27,000+
A-3
-------
Table A-l. State Income Tax Rates (continued)
State
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Average:
Corporate Income
Tax Rate
7:25%
7.60%
9.00%
7.75%
10.50%
8.90%
6.00%
6.60%
9.90%
9.00%
, 5.00%
0.00%
6.00%
0.00%
5.00%
8.25%
6.00%
0.00%
9.00%
7.90%
0.00%
6.61%
Basis for States
With Graduated
Tax Tables
$1 Million+
$50,000+
Based on'Stock
Value
1997 and thereafter
$250,000+
Personal Income
Tax Upper Rate
6.65%
8.50%
7.88%
7.75%
12.00%
7.50%
7.00%
9.00%
2.80%
10.40%
7.00%
0.00%
0.00%
0.00%
7.20% •
9.45%
5.75%
0.00%
6.50%
6.93%
0.00%
5.84%
Basis for States
With Graduated
Tax Tables
$75,000+
$42,000+
-$13,000+
$60,000+
$50,000+
$200,000+
$10,000+
$5,000+
$250,000+
$11,000+
$4,000+
$250,000+
$17,000+
$60,000+
$20,000+
Source: CCH, 1999a and 1995.
Basis for rates is reported to nearest $1,000. Personal income tax rates for Rhode Island and Vermont based on
federal tax (not taxable income). Tax rates given here are equivalents for highest personal federal tax rate.
The cost annualization model can incorporate variable tax rates according to the level of
income to address differences between small and large model CAFOs. For example, a large
model CAFO might have a combined.tax rate of 40.6 percent (34 percent federal rate plus 6.6
percent state rate). After tax shields, this model CAFO would pay 59.4 cents for every dollar of
A-4
-------
incremental animal waste management costs. A small model CAFO might be in the 21.6 percent
tax rate (15 percent federal rate plus 6.6 percent state rate). After tax shields, the small model
CAFO would pay 78.4 cents for every dollar of incremental animal waste management costs. The
net present value of after-tax cost is used in the CAFO level impact analysis because it reflects the
impact the business would actually see in its net income.1
A.1.2 Depreciation Method
EPA uses the Modified Accelerated Cost Recovery System (MACRS) to depreciate
capital investments after examining three alternatives, including MACRS, straight-line
depreciation, and Section 179 of the Internal Revenue Code. MACRS allows businesses to
depreciate a higher percentage of an investment in the early years, and a lower percentage in the
later years. In contrast, straight-line depreciation writes off a constant percentage of the
investment each year. MACRS offers companies a financial advantage over the straight-line
method because a model CAFO's taxable income may be reduced under MACRS by a greater
amount in the early years when the time value of money is greater. EPA also considered using the
Internal Revenue Code Section 179 provision to elect to expense up to $17,500 in the year the
investment is placed in service, assuming that the investment costs do not exceed $200,000 (IRS,
1999a). However, EPA assumes that this provision is already applied to other investments at the
CAFO.
To determine the recovery period of depreciable property, IRS identifies asset classes
based on the activity in which the property is being used. If there is not an activity that matches
the use then IRS provides classes for specific depreciable assets that are used across multiple
business activities such as office furniture, information systems, and automobiles. Under
MACRS, the cost of property is recovered over a set period. The recovery period is based on the
property class to which your property is assigned. If thfe property of interest is not identified by
the IRS then it generally has a recovery period of 7 years (IRS, 1999b).
The capital costs required by this regulation fall across three IRS asset classes including:
land improvements (15 year recovery period), agriculture (7 year recovery period), and single
purpose agricultural or horticultural structures (10 year recovery period). Table A-2 presents
these IRS asset classes as well as the capital costs associated with them. EPA has identified the
appropriate class for each type of cost and has judged that a 10-year time frame is appropriate for
this analysis for the following reasons:
i
The cost annualization model does not consider tax shields on interest paid to finance animal waste
management investments. The cost annualization model assumes a cost to the operation to use the money (the
discount/interest rate), whether the money is paid as interest or is the opportunity cost of internal funding. Tax
shields on interest payments are not included in the cost annualization model because it is"not known what mix of
debt and capital an operation will use to finance the cost of incremental animal waste management investments
and to maintain a conservative estimate of the after-tax annualized cost.
A-5
-------
• A 10-year depreciation time frame is consistent with the 10-year property
classification of a single purpose livestock structure which is defined under
Section(i)(13)(B) as any enclosure or structure specifically designed, constructed
and used for housing raising and feeding a particular kind of livestock including
their produce or for housing the equipment necessary for the housing rasing and
feeding of livestock (IRS, 1999a).
• A 10 year depreciation time frame is a fairly conservative assumption considering
that some assets such as land improvements have a longer 157year time frame and
others such as agricultural equipment have a shorter 7-year time frame.
• This assumption provides a uniform time frame for use in the annualization model
and prevents the use of separate annualization calculations for individual capital
costs.
• A 10-year time frame is consistent with the practice of cost-share programs which
typically organize contracts over 5- to 10-year periods (USDA, 1999).
EPA conducted initial sensitivity analyses of the annualization model using initial cost estimates
and determined that the differences between using a 7-, 10-, or 15-year time frame for
depreciation did not result in large changes in annualized costs.
A.2 SAMPLE COST ANNUALIZATION SPREADSHEET
Table A-3 shows a sample cost annualization worksheet. The top of the spreadsheet
shows the data inputs described in Section A.l. The spreadsheet contains numbered columns that
calculate the before- and after-tax annualized cost of the investment to the .CAFO. Column 1 of
Table A-3 lists each year of the investment's life span, from its installation through its 10-year
depreciable lifetime (shown over years 1 through 11, since a mid-year convention is used).
Column 2 of Table A-3 represents the percentage of the capital costs that can be written <
offer depreciated each year. These rates are based on the MACRS and are taken from CCH,
1999b. Multiplying these depreciation rates by the capital cost gives the annual amount the
model CAFO may depreciate, which is listed in Column 3. EPA uses depreciation expense to
offset annual income for tax purposes; Column 4 shows the tax shield provided from the
depreciation expense—the overall tax rate times the depreciation amount for the year.
Column 5 of Table A-3 is the annual O&M expense. Due to the mid-year convention
assumption for depreciation, Year 1 and Year 11 show only six months of annual O&M costs.
Year 1 O&M also includes the startup O&M cost. Years 2 through 10 Include annual O&M plus
recurring O&M costs for every third and fifth year. Column 6 is the tax shield or benefit provided
from expensing the O&M costs.
A-6
-------
a
o
3
'S
IS
o
(*•<
O
gj
•*•»
.S
1
e
TRS Asset Class Lives and Recovery Periods Relevant for the An
-
--2 <§"<
0 ^,
"•O "
0) i
"3 '
* "S '
" --! »
.1 "
. £• " " "
.
.2 ffi *« "
1 ^ o
60
_g
*^H
o
" *S
> "§ 2
•5 3 » -
o S g u
3*1 » " §
^ 1 i s o i •
i co co i i !>
- -1" '
it! t|?l
|| ^1 1||
|J|||^if
_J"* ^ ^ ^ Q^ "(J ^Q O J-4
"t^ EM ^ O W) ^ "^"* tf 3
03 O S n-t *^ 'o -fc* O • tf
rT*l (^J ^^ *2 ^- C W ^,) ^
• • • « • ^^ • «
t-^
o
• .S
& eg M.g .=!
*"* W £3 C *•"' £*
f- S I -g S'il "3
§ "B, § -JJ12 -S^g g.
•ill a i •* i fi s
£ i -» !-< G S J- . — «. -S o ^
£P
>. c3 «o
en --n -5 g &,
.5 m > * S
to el. . G d. *2 ^ •
600 H^ ^^/ j -i
•S I 60° 0^-§"^ O-o-Sl"
I'll 1 >fl l^'ll'iu
,U,£, , « I" T3 l tS 0
'
O
Agriculture: Includes machinery and equipment, grain bins, and
fences but no other land improvements, that are used in the
production of crops or plants, vines, and trees; livestock; the
operation of farm dairies, nurseries, greenhouses, sod farms,
mushroom cellars, cranberry bogs, apiaries, and fur farms; the
performance of agriculture, animal husbandry, and horticultural
services.
o
11
00
^e
CO
O
2
8
|s>
is
o —
sl
1 t*H
o
10
Single purpose agricultural or horticultural structures (within
the meaning of section 168(1)(13) of the code)
1
a, o,?
-------
000 = OsPvPsP
yj ^^ *— ^ S^ O*^ &*• O^
o H^uS'OOod-^-
>jS X«-«'-'CS(Sc»l
< § S
•g Js t t t t t
O 43
J i
1
•"*
III!
V V V V
1111
o c> co o o
&9 C3 C3 O O
0 ,, O O O O
O /\ <3 vT O VI
j2 S 6? 2 S
[2 3 2 A A A A
* C3 ^3
S* X cfl
6 ra ep
1 1
S -S?
cj [U
1
^
Cj>
!0.
§
2-
i 2 s | s | a ^ 1
2s 8 s
=
D^^^^orrto^&rtOC-
3 j^ ^ ^ oo ^i* r"- c*" C"-» t"-' co
| "- 5 S 5 5 5 S §" S 5
A vS» sS-
OOOOOOOOOOOI
ooooooooooo
V~l V} O VI O O V} 10 ^ O Vy
l.ggggsimsgsl
OOOOOOOOOOOI
Z) O O- O O O O O O O O
Booooooooo
SJ^^a^e^SJSJ^SJSJ'
§iclSsii§llsj
8" S S S S S S S S S d
CD^OOOOOOOOOI
3OOCDOOCDOOOCS
^222zi2S22«&^
WWWW^OT^WW
§21||i§||§SJ
Tj-r— tn^cooicNC^tsoi^
„„««„„„ ww«w«
Illllllllll
69 69 69 <» .
0 0,P 0 OvPspNOxP-^xP
8S°5ncNm!nlolnlScN
O OO ^" ^-t C\ f*» ^O ^O VO ^O CO
1— < ^-< *— « ^-
-2-
vb co jj co o
vo" ^n rt tn1 oT
^ ON c_i ON '—•
^ft 0
d
O CO -J CO »— '
o o ?7} o o\
O ^" ^" *rt
O* ^^ m ^"* C?
2 2 H 2 S
a
§ON
^r
in" co"
O ^f
g s
2" if
69
I 2
cs* oC '
O 00
O .— H
2 **"
2 a
\o oo
o" oT .
s . s
cT ro*
o r~
69
SP
o
o a
o «
2 S
s
O 0
a> o> O
^ "S "c3
1 § § i
CO CU &-i •<
CO
1
•3'
1-
II
g 3
s o
!l
'O ^.
2 §
»
So
PH
-fij o
overy sys
10-year pi
CJ .
ct> i«
«2
If
^5 x
-§ t
o 8
o 4—
03 co
^i «i
oi o
15 s
II
M G3
.T^ S
"S •£
(D 2
li
3/3
CO
-------
Columns 7 and 8 represent negative costs that should be evaluated when considering
compliance costs for model CAFOs, including payments from federal, state, or local cost-share
programs and offsets from the change in manure use.
Column 9 lists a model CAFO's annual cash outflow, or total expenses, associated with
the additional animal waste management. Total expenses include capital costs, assumed to be
incurred during the first year when the structure is installed, plus each year's O&M expense.
Column 10 lists the annual cash outflow less the tax shields from the O&M expenses and
depreciation; a model CAFO will recover these costs in the form of reduced income taxes. The
sum of the 11 years of after-tax expenses is $106,546 (1997 dollars). The equation EPA uses to
calculate the present value of cash flow is :
NPV =
n
I
VT
Where:
r
n
i
= series of cash flows
= interest rate
= number of cash flow periods
= current iteration
EPA uses the present value of the after-tax cash outflow in the CAFO level impact analysis to
calculate the post-regulatory present value of future earnings for a model CAFO.
EPA transforms the present value of the cash outflow into a constant annual payment for
use as the annualized model CAFO's compliance cost. Column 9 calculates the annualized cost as
a 10-year annuity that has the same present value as the total cash outflow. The annualized cost
represents the annual payment required to finance the cash outflow after tax shields. In essence,
paying the annualized cost each year and paying the amounts listed in Column 8 for each year are
equivalent. EPA calculates the annualized cost as:
Annualized Cosf = Present value of cash outflows *
real discount rate
1 - (real discount rate + 1)"
where n is the number of payment periods.
In the example used in Table A-3, based on the capital investment of $100,000, startup
O&M costs of $1,000, 3-year recurring O&M costs of $500, 5-year recurring O&M costs of
$1,500, annual O&M costs of $10,000 per year, a tax rate of 40.6 percent, and a real discount rate
A-9 :
-------
of 7 percent, the model CAFO's annualized cost is $20,591 on a pre-tax basis and $12,729 on a
post-tax basis.2
EPA uses the pre-tax annualized cost to calculate the total social cost of the regulation
(presented in Section 10). This approach incorporates the cost to industry for the purchase,
installation, and operation of additional animal waste management structures, and also federal and
state government from lost tax revenues. (Every tax dollar that a business does not pay due to a tax
shield is a tax dollar lost to the government.)
EPA uses the post-tax annualized cost to reflect what a business actually pays to comply
with incremental animal waste management requirements (presented in Section 5). The post-tax
present value of incremental animal waste management expenditures is used in the CAFO level
impact analysis.
A.3 ANNUALIZED COMPLIANCE COSTS
Tables A-4 through A-ll show EPA's estimated post-tax annualized costs to regulated
CAFOs to comply with the proposed revisions to the CAFO regulations. Annualized costs are
shown in 1997 dollars and are expressed on a per-animal (inventory) basis. EPA estimates per-
animal costs for operations that raise fed cattle, veal, heifers, milk cows, hogs (both farrow-finish
and grow-finish systems), broilers, egg laying hens (both liquid and dry manure systems), and
turkeys. The tables show these costs for each of the seven ELG technology options considered by
EPA during the development of this rulemaking. Costs to confinement operations with less than
300 or 500 AU that may be designated as CAFOs are developed by scaling the estimated
compliance costs for the available "medium" and "large" CAFO models. (See Section 2 for
information on expected designated facilities under each co-proposed alternative.) The resulting
costs-derived on a per-head basis-are adjusted by the average head counts at operations with fewer
than 500 AU or 300 AU to derive the annualized per-facility compliance cost. Costs for CAFOs
with fewer than 500 AU or 300 AU assume that these operations have sufficient cropland for all on-
farm nutrients generated (identified in the cost model as Category 1 costs).
2 There are two ways to calculate post-tax annualized cost. One is to calculate the annualized cost as the
difference between the annuity value of the cash flows (Column 9) and the tax shields (Columns 4 and 6). The
second is to calculate the annuity value of the cash flows after tax shields (Column 10). Both methods yield the
same result.
A-10
-------
Table A-4
Total Annualizcd Compliance Costs per Head for Option 1
Sector/Model
Beef
MW Region
Small
Medium 1
Medium 2
Large 1
Large2
CE Region
Medium 1
Medium 2
Large 1
Large2
Dairy
PA Region
Medium 1
Medium 2
Large 1
MW Region
Small
Medium 1
Medium 2
Large 1
Veal
MW Region
Medium 1
Medium 2
Heifers
PA Region
Medium 1
Medium 2
Large 1
MW Region
Medium 1
Medium 2
Large 1
Head Category 1 Category 2 ' Category 3
112
455
777
1,877
30,003
455
777
1,877
30,003
235
460
1,419
200
235 .
460
1,419
400
540
400
750
1,500
400
750
1,500
$16.08
$16.69
$10.41
$1.36
$0.85
$15.55
$9.58
$1.09
$0.57
' $16.84
$10.77
S5.77
$16.42
$14.77
$9.39
$4.50
$3.61
$2.66 •
$13.76
$8.50
$1.06
$13.72
$8.65
. $0.95
NA
$16.49
$10.52
$9.40
$3.94
.$15.53
$10.51
$15.32
$4.71
$23.97,
$41.79
$39.19
- NA
$16.65
$2430
$22.35
$2.76
$2.03
$16.85
$8.63
$1.29
$16.84
$9.44
$1.04
NA
$5.17
$2.56
$0.60
$0.16
$5.11
$2.51,
$0.52
$0.08
$13,76
SS.Sl
$5.00
NA
$10.60
$6.66
$3.60
$2.50
$1.84
$5.67
$328
$0.75
$5.60
$3.23
$0.67
A-ll
-------
Table A-4 (continued)
Total Annualized Compliance Costs per Head for Option 1
Sector/Model
Hogs (a)
MW Region (GF)
Medium la
Medium Ib
Medium 2
Largel
. Large2
MA Region (GF)
Medium la
Medium Ib
Medium 2
Largel
Large 2
MW Region (FF)
Small
Medium la
Medium Ib
Medium 2
Largel
Large 2
MA Region (FF)
Medium la
Medium Ib
Medium 2
Largel
Large 2
Broilers
SO Region
Medium la
Medium Ib
Medium 2
Largel
Large 2
MA Region
Medium la
Medium Ib
Medium 2
Largel
Large 2
Layers
Wet SO Region
Small
Medium 2
Largel
Dry SO Region
Medium la
Medium Ib
Medium 2
Largel
Large 2
Dry MW Region
Medium la
Medium Ib
Medium 2
Largel
Large 2
Turkeys
MA Region
Medium la
Medium Ib
Medium 2
Largel
MW Region
Medium la
Medium Ib
Medium 2
Largel
Head C
900
1,422
2,124
3,417
10,029
963
1,521
2,184
3,554
8,893
750
814
1,460
2,152
3,444
13,819
846
1,518
2,165
• 3,509
17,118
36,634
51^62
73,776
117,581
281,453
36,796
51,590
73,590
115,281
303,155
1,000
3,654
86,898
32,375
44,909
97,413
293,512
884,291
37,906
52,582
97,484
279,202
1,229,095
18,539
31,267
45,193 •
97,111
18,092
30,514
45,469
158,365
ategoryl Category 2 Category 3
$0.70
$0.47
$0.29
$0.09
$0.04
$0.56
$0.40
$0.31
$0.09
$0.04
$0.48
$0.77
$0.46
$0.29
$0.09
$0.03
$0.62
$0.40
$031
$0.09
$0.03
$0.09
$0.09
$0.08
$0.08
$0.07
$0.09
$0.09
$0.08
$0.08
$0.06
$0.14
$0.07 .
$0.01
$0.03
$0.03
$0.02
$0.02
$0.02
$0.03
$0.03
$0.02
$0.02
$0.02
$0.13
$0.11
$0.09
$0.08
$0.17
$0.15
$0.13
$0.12
$1.52
$5.47
$1.46
$3.43
$3.86
$6.80
$3.50
$6.63
$7.36
$6.18 .
NA
$1.70
$1.07
$0.71
$0.48
$4.02
$1.51
$5.60
$6.04
$5.65
$2.23
$0.09
$0.08
$0.08
$0.07
$0.06
$0.09
$0.08
$0.08
$0.07
$0.06
NA
$0.55
$0.27
, $0.03
$0.02
$0.02
$0.02
$0.02
$0.03
$0.03
$0.02
$0.02
$0.02
$0.13
$0.11
$0.08
. $0.07
$0.17
$0.15
$0.13
$0.12
$10.20
$9.75
$8.20
$8.81
$7.81
$10.17
$9.80
$9.60
$8.80
$7.82
NA
$9.01
$8.41
$7.08
$7.56
$6.16
$8.97
$8.46
' $8.26
$7.56
$6.15
$0.08
$0.07
$0.07
$0.06
$0.05
$0.08
$0.07
$0.07
$0.07
$0.05
NA
$0.40
$0.16
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.09
$0.07
$0.06
$0.05
$0.13
$0.12
$0.10
$0.10
(a) Two categories of hog farms are included: grower-finish (GF) and farrow-finish (FF).
A-12
-------
Table A-5
Total Annualized Compliance Costs per Head for Option 2
Sector/Model
Beef
MW Region
Small
Medium 1
Medium 2
Large 1
Large 2
. CE Region
Medium 1
Medium 2
Large 1
Large 2
Dairy
PA Region
Medium 1
Medium 2
, Large 1
MW Region
Small
Medium 1
Medium 2
Large 1
Veal
MW Region
Medium 1
Medium 2
Heifers
PA Region
Medium 1
Medium 2
Large 1
MW Region
Medium 1
• Medium 2
Large 1
Head Category 1 Category 2 Category 3
112
455
777
1,877
30,003
455
777
1,877
30,003
235
460
1,419
200
235
460
1,419
400
540
400
750
1,500
400
750
1,500
$17.86
$18.34
$11.96
$1.61
$0.93
$25.63
$19.21
$6.44
$2.94
$21.42
$14.62
$8.07
$19.39
$17.97
$12.20
$6.53
$7.20
$3.72
$17.55
' $11.61
$2.45
$14.98
$9.64
$1.21
NA'
$21.56
$14.54
$8.80
$4.12
$35.32
$26.55
$14.97
$10.48
$27.81
$30.66
$29.26
NA
$21.60
$22.68
$18.05
$2.76
$2.03
$20.87
$14.38
$5.03
$18.78
$11.13
$2.23
NA
$5.17
$2.56
$0.60
$0.16
$5.11
$2.51
$0.52
$0.08
$13.76
$8.81
$5.00
NA
$10.60
$6.66
$3.60
$2.50
$1.84
$5.67
$3.28
• $0.75
$5.60
$323
$0.67
A-13
-------
Table A-5 (continued)
Total Annualized Compliance Costs per Head for Option 2
Sector/Model
Hogs (a)
MW Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA.Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MW Region (FF)
Small
Medium la
Medium Ib
Medium 2
Largel
Large 2
MA Region (FF)
Medium la
Medium Ib
Medium 2
Largel
Large 2
Broilers
SO Region
Medium la
Medium Ib
.Medium 2
Largel
Large 2
MA Region
Medium la
Medium Ib
Medium 2
Largel
Large 2
Layers
Wet SO Region
Small
Medium 2
Largel
Dry SO Region
Medium la
Medium Ib
Medium 2
Largel
Large 2
Dry MW Region
• Medium la
Medium Ib
Medium 2
Largel
Large2
Turkeys
MA Region
Medium la
Medium Ib
Medium 2
Largel
MW Region
Medium la
Medium Ib
Medium 2
Largel
Head Category 1 Category 2 Category 3
900
' 1,422
2,124
3,417
10,029
963
1,521
2,184
3,554
8,893
750
814
1,460
2,152
3,444
13,819
846
1,518
2,165
3,509
17,118
36,634
51,362
73,776
117,581
281,453
36,796
51,590
73,590
115,281
303,155
1,000
3,654
86,898
32,375
44,909'
97,413
293,512 ,
884,291
37,906
52,582
97,484
279,202
1,229,095
18,539
31,267
45,193
97,111
18,092
30,514
45,469
158,365
$2.89
$2.67
$2.15
$1.94
$1.69
$3.09
$2.89
$2.83
$2.06
$1.79
$2.28
$2.95
$2.65
$2.15
$1.93
$1.53
$3.15
$2.93
$2.84
$2.06
$1.62
$0.15
$0.14
$0.13
$0.13
$0.11
$0.13
$0.13
S0.12
$0.12
$0.10
$0.32
$024
$0.15
S0.18
$0.14
$0.15
$0.18
$0.00
$0.25
$027
$0.23 •
$0.19
$0.00
$0.71 •
$0.70
$0.61
$0.57
$0.83
$0.79
$0.65
' $0.65
$3.52
$6.78
$4.48
$5.33
$4.60
$7.85
$5.86
$7.07
$6.78
$5.61
NA
$3.87
$4.74
$2.18
$2.65
$4.44
$3.30
$6.70
$7.03
S6.13
$3.65
$0.10
$0.09
$0.08
$0.08
$0.06
$0.11
$0.10
$0.10
$0.09
$0.07
NA
$0.60
$0.27
$0.04
$0.04
$0.03
$0.02
$0.02
$0.05
$0.04
$0.04
$0.03
$0.02
$029
$0.23
$0.18
$0.16
$0.28
$022
$0.18
$0.15
$7.47
$7.00
, $5.90
$6.20
$5.48 -
. $7.34
$7.21
$6.87
$6.18
$5.52
NA
$6.96
$6.37
$535.
$5.61
$4.58
$6.92
$6.33
$6.17
$5.61
$4.56
$0.08
$0.07
$0.07
$0.06
$0.05
$0.08
$0.07
$0.07
$0.07
$0.05
NA
$0.39
$0.15
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.09
$0.07
$0.06
$0.05
$0.13
$0.12
$0.10
$0.10
(a) Two categories of hog farms are included: grower-finish (GF) and farrow-finish (FF).
A-14
-------
Table A-6
Total Annualized Compliance Costs per Head for Option 3
Sector/Model
Beef
MW Region
Small
Medium 1
Medium!
Large 1
Large 2
CE Region
Medium 1
Medium 2
Large 1
Large 2
Dairy
PA Region
Medium 1
Medium 2
Largel
MW Region
Small
Medium 1
Medium 2
Largel
Veal
MW Region
Medium 1
Medium 2
Heifers
PA Region
Medium 1
Medium 2
Largel
MW Region
Medium 1
Medium 2
Largel
Head Category 1 Category 2 Category 3
112
455
' 777
1,877
30,003
455
777
1,877
30,003
235
460
1,419
200
235
460
1,419
400
540
400
750
1,500
400
. 750
1,500
$28.37
$29.60
$21.00
$8.26
$4.61
$31.74
$24.12
'$9.39
$4.52
$48.23 .
$31.66
$20.02
$60.39
$58.74
$39.99
$27.37 .
$7.78
$4.75
$22.94
$15.41
' $5.35
$24.14
$16.11
$5.74
NA
$32i6
$23.29
$15.31
$7.79
$41.03
$3125
$17.82
$12.06
$53.78
$47.34
$41.09
NA
$61.51
$50.07
$38.76
$2.91 .
$2.73
$27.72
$17.43
$7.79
$27.49
$18.58
$6.60
NA*
$15.87
$1131.
$7.12 '
$3.83
$10.81
$721
$3.37
$1.65
$39.72
$25.49
$16.83
NA
$50.50
$34.05
$2432
$2.65
$2.54
$10.63
$6.86
$3.51
$14.13
$937
$5.04
A-15
-------
Table A-6 (continued)
Total Annnalized Compliance Costs per Head for Option 3
Sector/Model
Hogs (a)
Head Category 1 Category 2 Category 3
.MW Region (GF)
Broilers
Layers
Turkeys
Medium la
Medium Ib
Medium 2
Large 1
Large2 .
MA Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MW Region (FF)
Small
Medium la
Medium Ib
Medium 2
Large 1
Large2
MA Region (FF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large2 .
MA Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Wet SO Region
Small
Medium 2
Large 1
Dry SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Dry MW Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region
Medium la
Medium Ib
Medium 2
Large 1
MW Region
Medium la
Medium Ib
Medium 2
Large 1
900
1,422
2,124
3,417
10,029
963
1,521
2,184
3,554
8,893
750
, 814
1,460
2,152
3,444
13,819
846
1,518
2,165
3,509
17,118
' 36,634
51,362
73,776
117,581
281,453
36,796
51,590
73,590
115281
303,155
1,000
3,654
86,898
32,375
44,909
97,413
293,512
884291
37,906
52,582
97,484
279202
1229,095
18,539
31267
45,193
97,111
18,092
30,514
45,469
158,365
$3.92
$3.41
$2.67
$2.36
$1.94
$3.92
$3.54 •
$3.34
$2.39
$2.01
$2.99
$4.07
$3.39
. $2.66
$2.35
$1.76
$4.07
$3.53 '
$3.32
$2.40
$1.81
$0.16
$0.15
$0.14
$0.14
$0.11
$0.15
$0.14
$0.13
$0.13
$0:10
$0.57
$0.36
$0.15
$0.19
$0.18
$0.15
$0.10
$0.00
$026
$0.25
$023
$0.00
$0.00
$0.75
$0.72
$0.62
$0.58
' $0.83
$0.81
$0.67
$0.66
$4.44
$7.44
$4.91
$5.67
$4.80
$8.61
$6.35
$7.48
$7.06
$5.78
NA
$4.90
$5.42
$2.60
$2.99
$4.61
^
$4:14
$7.31
$7.44
$6.40,
$3.79
$0.11
$0.10
$0.09.
$0.08
$0.06
$0.13
$0.12
$0.11
$0.10
$0.07
NA
$0.73
$028
$0.05
$0.04
$0.03
$0.02
$0.02
$0.06
$0.05
$0.04
$0.03
$0.02
$0.32
$0.25
$0.19
$0.17
$0.32
$024
$0.19
$0.16
$8.49
$7.77
$6.38
$6.62
$5.74
$826
$7.79'
$7.36
$6.56
$5.73
NA
$8.08
$7.11
$5.83
$6.03
$4.82
$7.73
$6.90
$6.67
$5.94
$4.74
$0.09
$0.08
$0.08
$0.07
$0.05
$0.10
$0.09
$0.08
$0.07
$0.06
NA
$0.51
$0.16
$0.03
$0.03
$0.02
$0.02
$0.02
$0.04
$0.03
$0.02 •
$0.02
$0.02
$0.12
$0.09
$0.07
$0.05
$0.17
$0.15
$0.12
$0.10
(a) Two categories of hog farms are included: grower-finish (GF) and farrow-finish (FF).
A-16
-------
Table A-7
Total Annualized Compliance Costs per Head for Option 3A
Sector/Model
Beef
MW Region
Small
Medium 1
Medium 2
Large 1
Large2
CE Region
Medium 1
Medium 2
Large 1
Large 2 ,
Dairy
PA Region
Medium 1
Medium 2
Large 1
MW Region
Small
Medium 1
Medium 2
Large!
Veal
MW Region
Medium 1
Medium 2
Heifers
CE Region
Medium 1
Medium 2
Large 1
MW Region
Medium 1
Medium 2
Large 1
Head C
112
455
777
1,877
30,003
455
777
1,877
30,003
235
460
1,419
200
235
460
1,419
NA
NA
400
750
1,500 .
400
750
1,500
ategoryl Category 2 Category 3
$80.06
$64.02
$47.97
$27.46
$15.88
$70.62
$54.64 •
$30.06
$16.65
$228.84
$148.83
$100.45
$228.08
$183.75
$126.14
$88.44
NA
NA
$58.34
$40.34
$23.91
$53.46
$36.78
$18.66
NA
$67.24
$50.56
$34.65
$19.08
$80.32
$61.98
$38.59
$24.20
$235.23
$164.88
$121.64
NA
$187.38
$136.62
$99.96
NA
NA
$60.86
$43.96
$27.14
$55.50
$39.57
$20.16
NA
$51.06
$38.69
$26.50
$15.12
$50.31
$38.06
$24.19
$13.79
$221.62
$143.21
$97.44
NA
$176.79
$120.80
$85.57
-. NA
NA
$46.71
$32.14
$2228
$44.33
$30.50
$18.18
A-17
-------
Table A-7 (continued)
Total Annualized Compliance Costs per Head for Option 3A
Sector/Model
Hogs (a)
MW Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large2
MW Region (FF)
Small
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region (FF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Broilers
SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large2
MA Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Layers
Wet SO Region
Small
Medium 2
Large 1
Dry SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Dry MW Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Turkeys
MA Region
Medium la
Medium Ib
Medium 2
Large 1
MW Region
Medium la
Medium Ib
Medium 2
Large 1
Head Category 1 Category 2 Category 3
900
1,422
2,124
3,417
10,029
963
1,521
2,184
3,554
8,893
' 750
814
1,460
2,152
3,444
13,819
846
1,518
2,165
3,509
17,118
36,634
51,362
73,776
117,581
281,453
36,796 .
51,590
73,590
115,281
303,155
1,0.00
3,654
86,898
32,375
44,909
97,413
293,512
884,291
37,906
52,582
97,484
279,202
1,229,095
18,539
31,267
45,193
97,111
18,092
30,514
45,469
158^65
$9.25 '
$7.18
$5.19
$4.63
$330
$8.92
$7.00
$5.96
$4.58
$3.37
$2.28
$9.84
$7.08
$5.16
$4.62
$2.94
$9.63
$7.00
$5.98
$4.60
$2.87
$0.18
$0.16
$0.15
• $0.14
$0.11
$0.17
$0.15
$0.14
$0.14
$0.10'
$0.46
$0.57
$0.18
'$022
$0.22
$0.15
$0.02
$0.00
$0.30
$0.23
$0.24
$0.00
$0.00
$0.76
$0.75
$0.53
$0.92
$0.87
$0.85.
$0.62
$0.60
$8.89
$11.04
$539
$6.94
$6.23
$13.96
$8.95
$11.16
$10.83
$8.61
NA
$9.58
$6.55
$4.61
$3.97
$6.11
$9.31
$11.06
$10.59
$9.13
$426
$0.13
$0.12
$0.10
$0.09
$0.07
$0.15
$0.13
$0.12
$0.11
$0.07
NA
$0.93
$0.30
$0.07
$0.06
$0.04
$0.03
$0.02
$0.08
$0.06
$0.05
$0.03
$0.02
$0.36
$0:28
$0.21
$0.18
$0.36
$0.28
$021
$0.17
$18.74
$16.45
$13.09
$13.35
$11.07
$18.53
$16.40 '
$1525
$13.30
$11.15
NA
$18.08
$15.03
$11.94
$12.09
$9.08
$17.98
$15.06
$13.93
$12.07
$9.00
. $0.11
$0.10
$0.09
$0.08
$0.06
$0.12
$0.10
$0.09
$0.08
$0.06
NA
$0.72
$0.18
$0.05
• $0.04
$0.03
$0.02
$0.02
$0.05
$0.04
$0.03
$0.02
$0.02
$0.16
$0.12
$0.09
$0.07
$0.22
• $0.18
$0.14
$0.11
(a) Two categories of hog farms are included: grower-finish (GF) and farrow-finish (FF).
A-18
-------
Table A-8
Total Annualized Compliance Costs per Head for Option 4
Sector/Model
Beef
MW Region
Small
Medium 1
Medium!
Large 1
Large 2
CE Region
Medium 1
Medium 2
Large 1
Large 2
Dairy
PA Region
Medium 1
Medium 2
Large 1
MW Region
Small
Medium 1
Medium 2
Large 1
Veal
MW Region
Medium 1
Medium 2
Heifers
. PA Region
Medium 1
Medium 2
Large 1
MW Region
Medium 1
Medium 2
Large 1
Head Category! Category 2 Category 3
112
455
777
1,877
30,003
455
777
1,877
30,003
235
460
1,419
200
235
460
1,419
400
540
400
750
1,500
400
750
1,500
$37.00
$38.77 -
$25.79
$10.24
$4.72
$40.91
$28.91
$11.37
$4.63
$66.89
$39.75
$22.41,
$80.53
$76.50
$48.08
$29.77
$18.74
$12.86
$33.90
$20.97
$8.28
$35.10
$21.67
$8.22
NA
$41.44
$28.08
$17.29
$7.90
$50.20
$36.04
$19.80
$12.17
$72.43
$55.43
$43.49
NA
$79.26
$58.16
$41.16
$13.87
$10.85
$37.00
$22.99
$10.72
$37.32
$24.14
$9.08
NA
$25.04
$16.10
$9.10
$3.94
$19.98
$12.00
$535
$1.77
$58.38
$33.58
. $19.22
NA
$6826
$42.15
$26.71
$13.61
$10.66
$21.59
$12.42
.$6.43
$25.09
$14.94
$7.52
A-19
-------
Table A-8 (continued)
Total Annualized Compliance Costs per Head for Option 4
Sector/Model
Hogs (a)
MW Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region (GF)
Medium la
Medium Ib
Medium 2
Large!
Large 2
MW Region (FF)
Small
Medium la
Medium Ib
Medium 2
Large 1
Large 2,
MA Region (FF)
Medium la
Medium Ib
Medium 2
Large!
Large 2
Broilers
SO Region
Medium la
Medium Ib
Medium 2
Large!
Large2
MA Region
Medium la
Medium Ib
Medium 2
Large!
Large 2
Layers
' Wet SO Region
Small
Medium 2
Large 1
Dry SO Region
Medium la
Medium Ib
: Medium 2
Large!
Large 2
Dry MW Region
Medium la
Medium Ib
Medium 2
Large!
Large 2
Turkeys
MA Region
Medium la
Medium Ib
Medium 2
Large 1
MW Region
Medium la
Medium Ib
Medium 2
Large!
Head Category 1 Category 2 Category 3
900
1,422
2,124
3,417
10,029
963
1,521
2,184
3,554
8,893
750
814
1,460
2,152
3,444
13,819
846
. 1,518
2,165
3,509
17,118
36,634
51,362
73,776
117,581
281,453
36,796
51,590
73,590
115,281
303,155
1,000
3,654
86,898
32,375
44,909
97,413
293,512
884,291
37,906
52,582
97,484
279,202
1,229,095
18,539
31,267
' 45,193
97,111
18,092
30,514
45,469
158,365
$6.54
' $5.07
$3.61
$3.24
$221
$8.31
$6.32
. $527
$3.36
$2.36
$4.57
$6.96
$5.01
$3.58
$323
$1.94
$9.07
$632
$5.28
$338
$1.97
$028
$0.24
$0.20
$0.17
$0.12
$0.27
$0.23
$0.19
$0.16
$0.11
$2.98
$1.59
$0.19
$0.30
$026
$0.18
$0.11
$0.00
$036
$032
$027
$020
$0.00
$0.98
$0.86
$0.70
$0.62
$1.07
$0.95
$0.75
$0.68
$7.06
$9.09
$5.85
$6.56
$5.06
$13.00
$9.13
$9.42
$8.03
$6.12
NA
$7.80
$7.03
$3.53
$3.86
$4.79
$9.13
$10.10
$9.39
$7.38
$3.96
$024
$0.19
$0.15
$0.12
$0.08
$0.25
$0.20
$0.17
$0.14
$0.08
NA
$1.95
$0.32
$0.17
$0.13
$0.07
$0.04
$0.02
$0.16
$0.12
$0.08
$0.04
$0.02
$0.56
$0.39
$027
$0.20
$0.56
S0.39
$0.27
$0.18
$11.11
$9.43
$7.32
$7.50
$6.01
$12.65
$10.57
$9.29
$7.53
$6.08
NA
$10.97
$8.73
$6.76
$6.90
$4.99
$12.72
$9.69
$8.62
$6.92
$4.91
$022
$0.17
$0.14
$0.11
$0.07
$0.22
$0.17
$0.14
$0.11
$0.07
NA
$1.74
$0.20
$0.15
$0.11
$0.05
$0.03
$0.02
$0.13
$0.10
$0.06
$0.03
$0.02
$0.36
$0.23
$0.15
$0.09
$0.41
$0.29
$020
$0.13
(a) Two categories of hog farms are included: grower-finish (GF)and farrow-finish (FF).
A-20
-------
Table A-9
Total Annualized Compliance Costs per Head for Option 5
Sector/Model
Beef
MW Region
Small
Medium 1
Medium 2
Large 1
Laige2
CE Region
Medium 1
Medium 2
Large 1
Large 2
Dairy
PA Region
. Medium 1
Medium 2
Large 1
MW Region
Small
Medium 1
Medium 2
Large 1
Veal
MW Region
Medium 1
Medium 2
Heifers
PA Region
Medium 1
Medium 2
Large 1
MW Region
Medium 1
Medium 2
Large 1
Head Category 1 Category 2 Category 3 -
112
455
777
1,877
30,003
455
777
1,877
30,003
235
460
1,419
200
235
460
1,419
400
540
400
750
1,500
400
750
1,500
$42.32
$49.02
$40.19
$29.30
$26.89
$58.44
$49.47
$36.77
$31.59
- $45.02
$33.07
$18.08
$48.75
$47.64
$37.04
$22.52
$7.20
$3.72
$20.40
$13.30
$3.30
$17.83
$11.33
$1.95
NA
$52.23
$42.74
$36.45
$30.37
. $68.12
$56.76
$45.28
$39.08
$51.41
$49.32
$38.17
NA
$50.75
$48.77
$33.19
$2.76
$2.03
$23.68
$15.50
$5.86
$22.75
$12.79
$2.97
NA
$35.85
$30.80
$28.28
$26.12
$37.92
$32.76
$30.85
$28.73
$37.36
. $2736
$15.01
NA
$4026
$31.50
$19.59.
$2.50
$1.84
$8.52
$4.97
$1.60
$8.44
-$4.92
$1.42
A-21
-------
Table A-9 (continued)
Total Annualizcd Compliance Costs per Head for Option 5
Sector/Model
Hogs (a)
MW Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MW Region (FF)
Small
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region (FF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Broilers
SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Layers
Wet SO Region
Small
Medium 2
Large 1
Dry SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large2
Dry MW Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Turkeys
MA Region
Medium la
Medium Ib
Medium 2
Large 1
MW Region
Medium la
Medium Ib
Medium 2
Large 1
Head Category 1 Category 2 Category 3
900
1,422
2,124
3,417
10,029
963
1,521
2,184
3,554
8;893 '
750
814
1,460
2,152
3,444
13,819
846
1,518
2,165
3,509
17,118
36,634
51,362
73,776
117,581
281,453
36,796 '
51,590
73,590
115,281
303,155
1,000
3,654
86,898
32,375
44,909
97,413
293,512
884,291
37,906
52,582
97,484
279,202
1,229,095
18,539
31,267
45,193
97,111
18,092
30,514
45,469
158,365
$7.30
$6.82
$5.69
$5.68
$5.02
$7.53
$7.09
$6.90
$5.80
$5.13
$5.80
$7.45
$6.79
$5.68
$5.68
$4.63
$7.69
$7.13
$6.91
$5.80
$4.71
$0.15
$0.14
$0.13
$0.13
$0.11
$0.13
$0.13
$0.12
$0.12
$0.10
$0.83
$0.55
$0.27
$0.18
$0.14
$0.15
$0.18
$0.00
$0.25
$0.27
$0.23
$0.19
$0.00
$0.71
$0.70
$0.61
$0.57
$0.83
$0.79
$0.65
$0.65
$6.03 .
$8.72
$6.81
$5.43
$4.45
$9.43
$5.79
$8.55
$7.55
$4.97
NA
$6.64
$6.03
$4.35
$6.19
$5.44
$5.87
$8.18
$8.10
$7.05
$3.63
$0.10
$0.09
$0.08
$0.08
$0.06
$0.11
$0.10
$0.10
$0.09
$0.07
NA
$0.60
$0.27
$0.04
$0.04
$0.03
$0.02
$0.02
$0.05
$0.04
$0.04
$0.03
$0.02
$0.29
$0.23
$0.18
$0.16
$028
$0.22
$0.18
$0.15
$7.47
$7.00
$5.90
$6.20
$5.48
$7.34
$7.21
$6.87
$6.18
$5.52
NA
$6.96
. $6.37
$5.35
$5.61
$4.58
$6.92
$6.33
$6.17
$5.61
$4.56
$0.08
$0.07
$0.07
$0.06
$0.05
$0.08
$0.07
$0.07
$0.07
$0.05
NA
$0.39
$0.15 .
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02 •
$0.09
$0.07
$0.06
$0.05
$0.13
$0.12
$0.10
$0.10
(a) Two categories of hog farms are included: grower-finish (GF) and farrow-finish (FF).
A-22
-------
Table A-10
Total Annualized Compliance Costs per Head for Option 6
Sector/Model
Beef
MW Region
Small
Medium 1
Medium 2
Large 1
Large2
CE Region
Medium 1
Medium 2
' Laigel
Large 2
Dairy
PA Region
Medium 1
Medium 2
Large 1
MW Region
Small
Medium 1
* Medium 2
Large 1
Veal
MW Region
Medium 1
Medium 2
Heifers
PA Region
Medium 1
Medium 2
Large 1
MW Region
Medium 1
Medium 2
Large 1
Head Category 1 Category 2 Category 3
112
455
777 .
1,877
30,003
455
777
1,877
30,003
235
460
i;419
200
235
460
1,419
400
540
--
400
750
1,500
400
750
1,500
$17.86
$18.34
$11.96
$1.61
$0.93
$25.63
$19.21
$6.44
$2.94
$21.42
$32.21
$7.05
$24.58
$17.97
$30.74
$7.62
$7.20
$3.72
$17.55
$11.61
$2.45
$14.98
$9.64
$1.21
NA
$21.56
$14.54
' $8.80
$4.12
$35.32
$26.55
$14.97
$10.48
$3127
$51.40
$30.15
NA
$24.85
. $4437
$21.08
$2.76
$2.03
$20.87
$13.85
$5.03
$17.83
$11.13
$2.23
NA
$5.17
$2.56
$0.60
$0.16
$5.11
$2.51
$0.52
$0.08
$13.76
$26.41
$3.98
NA -
$10.60
$2520
$4.68
$2.50
$1.84
$5.67
$3.28
$0.75
$5.60
$3.23
$0.67
A-23 '
-------
Table A-10 (continued)
Total Annualized Compliance Costs per Head for Option 6
Sector/Model
Hogs (a)
MW Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MW Region (FF)
Small
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region (FF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Broilers
SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region
Medium la
Medium Ib
Medium 2
Large 1
Large2
Layers
Wet SO Region
Small
Medium 2
Large 1
Dry SO Region
Medium la
Medium Ib
Medium 2
Large 1
Large 2
Dry MW Region
Medium la
Medium Ib
Medium 2
Large 1
Large2
Turkeys
MA Region
Medium la
Medium Ib
Medium 2
Large 1
MW Region
Medium la
Medium Ib
Medium 2
Large 1
Head Category 1 Category 2 Category 3
900
1,422
2,124
3,417
10,029
963
1,521
2,184
3,554
8,893
750
814
1,460
2,152
3,444
13,819
846
1,518
2,165
3,509
17,118
36,634
51,362
73,776
117,581
281,453
36,796
51,590
73,590
115,281
303,155
1,000
3,654
86,898
32,375
44,909
97,413
293,512
884,291
37,906
52,582
97,484
279,202
1,229,095
18,539
31,267
45,193
97,111
18,092
30,514
45,469
158,365
$2.89
$2.67 ..
$2.15
$1.94
$1.36
$3.09
$2.89
$2.83
$2.06
$2.68
$0.99
$2.95
$2.65
$2.15
$1.93
$4.10
$3.15
$2.93
$2.84
$2.06
. $3.55
$0.15
$0.14
$0.13
$0.13
$0.11
$0.13
$0.13
$0.12
$0.12
$0.10
$0.32
$024
$0.15
$0.18
$0.14
$0.15
•$0.18
$0.00
$0.25
$0.27
$0.23 •
$0.19
$0.00
$0.71
$0.70
$0.61
$0.57
$0.83
$0.79
$0.65
$0.65
$3.52
$6.78
$4.48
$5.33
$4.27
$7.85
$5.86
$7.07
$6.78
$6.50
NA
$3.87
$4.74
$2.18
$2.65
$7.00
$3.30
$6.70
$7.03
$6.13
$5.58
$0.10
$0.09
$0.08
$0.08
$0.06
$0.11 '
$0.10
$0.10
$0.09
$0.07
NA.
$0.60
$027
$0.04
$0.04
$0.03
$0.02
$0.02
$0.05
$0.04
$0.04
$0.03
$0.02
$0.29
$023
$0.18
$0.16
$028
$022
$0.18
$0.15
$7.47
$7.00
$5.90
$6.20
$5.15
$7.34
$7.21
$6.87
$6.18
$6.41
NA
$6.96
$6.37
$5.35
$5.61
$7.15
$6.92
$6.33
$6.17
$5.61
$6.48
$0.08
$OX>7
$0.07
$0.06
$0.05
$0.08
$0.07
$0.07
. $0.07
$0,05
NA
$0.39
$0.15
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
. $0.09
$0.07
$0.06
$0.05
$0.13
$0.12
$0.10
$0.10
(a) Two categories of hog farms are included: grower-finish (GF) and farrow-finish (FF).
A-24
-------
Table A-ll
Total Annualized Compliance Costs per Head for Option 7
Sector/Model
Beef
MW Region
Small
Medium 1
Medium 2
Large 1
Large 2
CE Region
Medium 1
Medium 2
Large 1
Large 2
Dairy
PA Region
Medium 1
Medium 2
Large 1
MW Region
Small
Medium 1
Medium 2
Large 1
Veal
MW Region
Medium 1
Medium 2
Heifers
PA Region
Medium 1
Medium 2
Large 1
MW Region
Medium 1
Medium 2
Large 1
Head Category 1 Category 2 Category 3
112
455
777
1,877
30,003
455
777
1,877
30,003
235
460
1,419
200
235
460
1,419
400
540
400
750
1,500
400
750
1,500
$18.21
$18.34
$13.18
$2.42
$1.19
$25.63
$21.39
$7.87
$3.43
$21.42
$34.26
$24.69
$13.14
$17.97
$37.31 .
$27.62
$7.20
$3.72
$17.55
$11.61
' $2.45
$14.98
$9.64
$1.21
NA
$21.56
$15.76
$9.61
$4.38
$35.32
$28.73
$16.40
$10.98
$27.81
$50.30
$45.88
NA
$21.60
$47.79
$39.13
$2.76
' $2.03
$21.92
$14.38
$5.03
$17.83
$11.13
$223
NA
$5.17
$3.79
$1.41
$0.42
$5.11
$4.69
$1.95
$0.57
$13.76
$28.45
$21.62
NA
$10.60
$31.77
$24.68
$2.50
$1.84
$5.67
$3.28
$0.75
$5.60
$3.23
$0.67
A-25
-------
Table A-ll (continued)
Total Annualized Compliance Costs per Head for Option 7
Sector/Model
Hogs (a)
MW Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MA Region (GF)
Medium la
Medium Ib
Medium 2
Large 1
Large 2
MW Region (FF)
Small
Medium la
Medium Ib
Medium 2
Largel
Large 2
MA Region (FF)
Medium la
Medium Ib
Medium 2
Largel
Large 2
Broilers
SO Region
Medium la
Medium Ib
Medium 2
Largel
Large 2
MA Region
Medium la
Medium Ib
Medium 2
Largel
Large 2
Layers
Wet SO Region
Small
Medium 2
Largel
Dry SO Region
Medium la
Medium Ib
Medium 2
Largel
Large 2
Dry MW Region
Medium la
Medium Ib
Medium!
Largel
Large2
Turkeys
MA Region
Medium la
Medium Ib
Medium 2
Largel
MW Region
Medium la
Medium Ib
Medium 2
Largel
Head ^ Category 1 . Category 2 Category 3
900
1,422
2,124
3,417
10,029
963
1,521
2,184
3,554
8,893
750
814
1,460
2,152
3,444
13,819
846
1,518
2,165
3,509
' 17,118
36,634
51,362
73,776
117,581
281,453
36,796
51,590
73,590
115,281
303,155
1,000
3,654
86,898
32,375
44,909
97,413
293,512
884,291
37,906
52,582
97,484 .
279,202
1,229,095
18,539
31,267
45,193
97,111
18,092
30,514
45,469
,158,365
$2.89
$2.67
$2.15
$1.94
$1.69
$3.09
$2.89
$2.83
$2.06
$1.79
$2.28
$2.95
$2.65
$2.15
$1.93
$1.53
$3.15
$2.93
$2.84 ..
$2.06 ..
$1.62
$0.15
$0.14
$0.13
$0.13
$0.11
$0.13
$0.13
$0.12
$0.12
$0.10
$0.32
$0.24
$0.15
$0.18
$0.14
$0.15
$0.18
$0.00
$0.25
$0.27
•$0.23
$0.19'
$0.00
$0.71
. $0.70
$0.61
$0.57
$0.83
$0.79
$0.65
$0.65
$8.79
$10.68
$7.08
$7.52
$5.80
$13.00
$9.68
$10.15
$8.92
$6.87
NA
$9.52
$8.58
$4.77
$4.83
$5.41
$8.93
$10.53
$10.12
$8.29
$4.56
$0.10
$0.09
$0.08
$0.08
$0.06
$0.11
$0.10 ,
$0.10
$0.09
$0.07
NA
$0.60
$0.27
$0.04
$0.04
$0.03
$0.02
$0.02
$0.05
$0.04
$0.04
$0.03
$0.02
$0.29
$023
$0.18
$0.16
$0.28
$0.22
$0.18
$0.15
$7.47
$7.00
$5.90
$6.20
$5.48
$7.34
$7.21
$6.87
$6.18
$5.52
NA
$6.96
$6.37
$5.35
$5.61
$4.58
$6.92
$6.33
$6.17
$5.61
$4.56
$0.08
$0.07
$0.07
$0.06
$0.05
$0.08
$0.07
$0.07
$0.07
$0.05
NA .
$0.39
$0.15
$0.02 -
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.02
$0.09
$0.07
$0.06
$0.05
$0.13
$0.12
$0.10
$0.10
(a) Two categories of hog farms are included: grower-finish (GF) and farrow-finish (FF).
A-26
-------
APPENDIX B
MARKET MODEL DESCRIPTION
This appendix describes the market model that EPA uses to estimate changes in market
prices and quantities attributable to the proposed CAFO regulations. Section B.I presents a non-
technical overview of the market model. Section B.2 describes the selection of baseline
parameters and elasticities for the model. Section B.3 presents the model in more technical detail
and is intended for readers with a background in microeconomics. Section B.4 discusses how the
model is used to estimate the impacts of regulatory costs. Section B.5 contains a glossary of
notation for the COSTBEN and EPA market models.
B.I INTRODUCTION AND OVERVIEW
The market model for this analysis is adapted from the COSTBEN model developed by
the U.S. Department of Agriculture, Economic Research Service (ERS) to analyze the effects of
policy changes on livestock and poultry markets (Hahn, 1996). COSTBEN uses a linear, partial
equilibrium model of supply and demand for a product with two stages of production: a market
for the final product, e.g., retail beef products, and a market for the intermediate product, e.g., fed
cattle. The model is typical of textbook supply and demand analysis (Tomek and Robinson, 1972;
Kohls and Uhl, 1998). The model assumes perfect competition in all markets. Supply and
demand "curves" are assumed to be straight lines (i.e., linear model).
The EPA market model differs from COSTBEN in several ways. COSTBEN forecasts
the short-run dynamics of the market as it evolves to a new long-run equilibrium using an iterative
process, in which the model is recalibrated to a forecast baseline at each iteration. A long-run
static analysis is more appropriate for analysis of the ultimate market effects of the proposed
CAFO regulations. EPA's market model is simpler than COSTBEN because it extracts only
those elements of the COSTBEN model essential to finding a long-run equilibrium. COSTBEN
includes models for more general categories of poultry and red meat, and sector-specific models
for cattle, hogs, chicken, and turkeys sold for slaughter. EPA expands on the COSTBEN data
sets to include separate models for each sector, including cattle, hogs, broilers, and turkeys, as
well as the non-meat sectors (egg layers and milk cows). For the egg and dairy sectors, the
intermediate product/final product model structure reflects the market conditions for eggs and
milk.
The EPA market model, like COSTBEN, treats each animal sector separately. Each
sector has two markets: the market for the intermediate products of farm production, cattle, hogs,
broilers, turkeys, raw milk, and eggs (henceforth termed collectively as "farm products") and the
retail market for their related finished products (poultry and red meats, fluid milk and dairy
products, and shell eggs). The markets are related since the supply of farm products affects the
B-l
-------
supply of retail products and the demand for finished goods generates a derived demand for farm
products. The model, basically, puts the supply of each farm product into the same units as
demand for the related retail product and finds the price and quantity where supply equals
demand. This is the long-run equih'brium point. Regulatory costs are modeled as shifts in the
demand or supply functions. The proposed CAFO regulations increase producers' costs so they
are modeled as an upward shift in the supply function. EPA measures these impacts as the change
from the baseline equih'brium to the new post-regulatory equilibrium.
The EPA market model can be thought of as starting with the domestic farm product
supply function. For any price, this function tells how much product will be raised for sale by
U.S. farmers. Additional supplies of imported farm products are added to domestic supply, and
exports are subtracted to yield a trade-adjusted supply function for the farm product. The trade-
adjusted supply function is a modeling convenience which summarizes in one equation all of the
farm product available to U.S. processors.1 In the poultry and dairy markets, where there is no
significant international trade ,in the raw farm products, the trade-adjusted supply function is
identical to the domestic supply function. While international trade in cattle and hogs is not a
large factor in the U.S. market, free trade agreements with Canada and Mexico have resulted hi
importation of fed cattle and hogs for slaughter by U.S. packers.
Raw farm products are processed into finished products. The EPA market model
describes the processing step with a conversion ratio and marginal processing cost. The
conversion ratio expresses the amount of farm output required to make the finished product. For
example, in the beef model it converts number of cattle slaughtered to pounds of beef produced.
The marginal processing cost per animal shows the processing industry's costs of production. In
the EPA market model, marginal costs of processing are just the difference between the retail
price and the farm level price adjusted so they are in the same units. The marginal processing cost
includes transportation to market and retail packaging, as well as those activities typically .
associated with packing and processing animal products. The trade-adjusted supply function for
farm output can be converted to a supply function for finished, retail products using the
conversion ratio and marginal processing cost.
To find the long-run equih'brium, supply must equal demand. The model has a linear
domestic demand function for the retail product. In addition to domestic production, retail
products are imported and exported. Supply and demand conditions in these markets are
discussed in detail in Section 2.2.3. The domestic retail demand is adjusted by retail product
imports and exports to yield a trade-adjusted demand function for the retail product. As the
trade-adjusted supply function is a convenient summary of inputs available to domestic
processors, so trade-adjusted demand is a convenient summary of distribution of their products.
The trade-adjusted retail supply function is equated to the trade-adjusted retail demand function
to find the long-run equilibrium price and quantity in the retail product market.
'Trade-adjusted supply differs from "Total Supply," which is domestic production plus imports, in that it
includes exports. Trade-adjusted supply also differs from "Net Trade" in that it includes domestic production.
B-2
-------
Once the long-run equilibrium retail price is determined, the conversion ratio and the
marginal processing cost relationships allow EPA to translate the retail price to a price for the
farm product. The farm product price applied to the farm product supply function determines the
quantity that will be produced domestically at the market equilibrium. The retail and farm product
prices also determine the quantities of imports and exports of the farm and retail products when
applied to the import and export equations.
The EPA market model deals with each of six animal sectors individually. The model does
not account .for possible multi-market effects and interactions between one sector and the other
livestock and poultry sectors or other sectors hi the economy. (The general equilibrium price
effects are discussed in more detail in Section B.3.) The model also does not account for the
possibility that consumers may substitute red meat for chicken (when chicken prices rise, for
example). Studies have shown that interventions hi one market may have effects in other markets
(Thurman and Wohlgenant, 1989).
Like COSTBEN, the EPA market model is used to evaluate alternatives by changing some
assumptions or parameters of the model. For analysis of the proposed CAFO regulations, the
compliance costs increase farmers' marginal costs of production. EPA estimates the impact of the
regulation by how that shift affects the overall market. Changes in prices and welfare are
measured as differences between the baseline, preregulatory equilibrium and the shocked,
postregulatory equilibrium. Figure B-l summarizes the baseline and postregulatory states of the
model (subscripted as b and 1 respectively).2 Higher farm costs shift the domestic farm product
supply function to the left, which similarly shifts the finished product supply function and the
equilibrium price at the intersection of retail supply and demand rises. There is considerable
evidence in the literature that shocks to farm level prices are quickly transmitted to retail markets
(Mclntosh, Park, and Karnum, 1997; Goodwin and Holt, 1999). Higher retail prices imply
higher farm product prices and lower farm product sales. A summary of the empirical research on
the ability of producers to pass on compliance costs is provided in the rulemaking record (ERG,
2000c).
These changes hi prices and quantities directly affect the revenue, costs, production, and
employment of firms throughout the marketing chain. (These relationships "are also discussed in
Section 4.4.1 of this report). Farms must pay for improvements'to comply with the new
regulations as well as sell a smaller quantity because of the increase in price. The shaded area hi
the Domestic Farm Product Market panel of Figure B-2 illustrates the direct impact on farms.
The processing industry also feels the effects of smaller production. Consumers absorb some of
the compliance costs through the cost passthrough process. These direct impacts are shown by
the shaded areas hi the Retail Product Market panel of Figure B-2. These direct effects have
2Figures are included in this section for illustration and will be discussed in more depth in Sections B.3
and B.4. Variables are defined in Section B.3 and appear in a glossary in Section B.5.
B-3
-------
Figure B-1 Livestock and Poultry Products Market Model
Retail Product Market
Retail
Price
PMO
Trade Adjusted
Supply-)
Trade Adjusted
/ Supplyo
Shift in Farm Product Supply
I shifts Retail Product Supply
Trade Adjusted
Demand
Retail Product Quantity
Farm Level
Price
Pu,
PLI
Domestic Farm Product Market
Supply-i
Supplyo
Compliance costs shift
Farm Product Supply
Farm Product Quantity
B-4
-------
Figure B-2. Direct Impacts
Retail
Price
Retail Product Market
Trade Adjusted
Supply.,
. Costs Passed
Through to
Consumers
Trade Adjusted
Supply0
Direct Impact
on Processors
Trade Adjusted
Demand
Retail Product Quantity
Domestic Farm Product Market
Farm Level
Price
Supply.,
Supply0
Direct Impact
on Farms
QUI QU,
Farm Product Quantity
B-5
-------
ripple effects throughout the economy. The EPA market model calculates the changes in
quantities supplied and demanded for each commodity at the farm, processor, and consumer
levels. Using the estimated market value of these quantity changes, EPA is. able to evaluate other
direct and indirect effects, including changes in national employment and changes in national
economic flows throughout the economy.
B.2 MODEL PARAMETERS AND DATA
Section 4.4.2 of .this report discusses the sources used for data and parameters in the EPA
market model. This section discusses the selection of elasticities and baseline values for the
model. COSTBEN and the EPA market model require specification of six long-run elasticity
estimates: farm product supply, farm product import supply, farm product export demand, retail
import supply, retail export demand, and domestic retail product demand. These elasticities
determine the slopes of the model functions and thus show how much quantities change when
prices change. To establish the starting conditions for the model, a base year is selected. All of
the coefficients of the model are derived so that the undisturbed model yields the baseline results.
EPA has conducted an extensive literature review to find elasticity values for relevant
markets from existing agricultural economic studies. This literature review is summarized in
Appendix C of this report. Although there are many studies, there is little consensus on the
correct demand and supply elasticities for the relevant markets. Table B-l summarizes the results
of the literature search with the maximums and minimums observed. Differences among estimates
depend on many factors. Many different econometric methods are used in these studies.
Additionally, the studies encompass differing production time frames and sample years.
EPA uses the price elasticity of demand values reported by USD A from an integrated
analysis of retail demand for food (Huang, 1993). The use of results from a single study ensures
consistency of methods and sample. Table B-l shows that the values Huang found are
comparable to other demand elasticities in the literature. In general, food demand tends to be
inelastic because food is a necessity. For some products, such as beef, sales are more sensitive to
price because many beef products are considered luxuries and have many substitutes. Milk and
eggs are staples in many peoples' diets and have few substitutes, so demand is very inelastic, i.e.,
quantities do not change much in response to changes in price. Retail demand elasticities are
appropriate for the COSTBEN and EPA market models which convert farm product supply into
retail market terms.
The available estimates of price elasticities of supply are not estimated in a similar unified
way because each sector has different characteristics. For this analysis, EPA uses supply
elasticities for the model that originate from various different sources. In some cases, both short-
and long-run price elasticity of supply estimates are available. Short-run price elasticities of
B-6
-------
Table B-l. Elasticity Estimates in the Agricultural Economics Literature
Variable
Beef'
Dairy
Hog
Broiler "
Layer
Price Elasticity of Demand
Maximum
Minimum
Selected (Huang, 1993)
-0.45
- -1.27
-0.621
-0.05
-0.65
-0.247
-0.07
-1.23
-0.728
Price Elasticity of Supply
Maximum
Minimum
Selected *
3.24
0.06
1.020
6.69
0.07
1.527
1.80
0.01
0.628
-0.10
-1.25
-0.372
-0.02
-0.78
-0.110
0.59
0.06
0.20
0.94
0.03
0.942
Turkey
-0.37
-0.68
-0.535
0.52
0.21
0.20
"'See Appendix C for maximum and minimum citations.
b/Excludes ground beef.
^Includes various forms of chicken meat.
""Selected elasticity of supply sources: Beef, including heifers and veal (Foster and Burt, 1992, adjusted by Foster,
2000a); Dairy (Chavas, Kraus, and Jesse, 1990); Hog (Holt and Johnson, 1988); Broiler and Turkey (Vukina,
2000); Layer (Chavas and Johnson, 1981).
supply tend to be inelastic because the flow of animal products to market is determined by
decisions made several months, or years, before the products will reach the market. In contrast,
some livestock economic models assume that long-run elasticities of supply are perfectly elastic
because producers have time to adjust their production decisions and adapt to changing
conditions (Hahn, 1998). The biology of the animal production process limits the adaptability of
producers to changing market conditions. The "long run" is longer for animals with slow
production cycles than for those with relatively fast cycles. Chicken flock size, for example, can
be adjusted within a few months, while cattle herds require years to adjust. As this is a long-run
analysis, EPA has generally selected more elastic price elasticity of supply values from the range
in the literature.
EPA did not identify estimates of the elasticity of imports and exports for these sectors
from the available literature. Many factors affect imports and exports including exchange rates,
foreign economic conditions, weather in competing production areas, and foreign government
policies. For simplicity, COSTBEN and the EPA market model summarize all of these factors
into linear supply and demand functions that represent the rest of the world. Although there is a
general consensus that foreign markets are more sensitive to prices than domestic markets
(Foster, 2000a), rest-of-the-world response to changes in U.S. prices for animal products has not
been widely studied. EPA assumes that elasticities for imports are the same as domestic supply
elasticities, and export demand elasticities are the same as domestic demand elasticities.
B-7
-------
To assess how the elasticity choices affect the market model results, EPA conducted
sensitivity analyses using the minimum and maximum elasticities as well as changing the trade
assumptions. These sensitivity analysis results appear in Appendix D of this report. These
sensitivity analyses conclude that the results presented in Section 5 of this report are stable across
a range of possible modeling assumptions.
To scale the model to observed values, a base year must be selected. The model
coefficients are calculated so that the baseline values are an equilibrium point in both the farm
product and retail markets. Any year with adequate data could serve as a base year. EPA
selected 1997 to correspond with the year of the engineering cost and financial data. Sensitivity
analyses in Appendix D indicate that the selection of baseline prices has little effect on model
outcomes given the size of shocks being evaluated. The baseline values and sources for all
variables, i.e., prices and quantities, and parameters, i.e., elasticities and multipliers, are
summarized in Section 4.4.2 and Table 4-14 of this report.
B.3 MODEL IN DETAIL
The framework of EPA's market model is based on USDA's COSTBEN model. This
section discusses the economic assumptions underlying the two models and highlights the
differences between them. COSTBEN is a highly simplified, partial equilibrium model intended
for "quick turn-around analysis of policy changes and other shocks to a single species' livestock
and meat market" (Hahn, ,1996). To maintain its simple structure, much of the detail of animal
products markets is assumed away. For example, all imported, exported, and produced livestock
are assumed to be the same. Imports, exports, and domestic production all have the same price
because of this assumption. Each retail product, e.g. beef, is also assumed to be uniform and
therefore have the same price. The problems of making pounds of filet mignon exported
equivalent to pounds of ground beef imported is avoided by assuming all beef is the same. Each
sector has similar equivalency issues which are handled by calibrating the retail quantities to
USDA's product disappearance tables (Putnam and Allshouse, 1999).
A shortcoming of COSTBEN and the EPA market model is their failure to account for
general equilibrium effects of incremental compliance costs. All of the animal sectors interact in
both supply and demand. Animal sectors compete for feed, for example, such that increased
demand in one sector may drive up feed prices in other sectors. Milk and egg producers that
principally raise animals for non-meat production may at times cull their herd or flock and sell
animals to meat markets. Many of the retail products are substitutes for each other, so price
changes in one sector will affect demand in others. This partial market abstraction can only be
justified if the expected shocks are so small that the general equilibrium effects would be
negligible.
COSTBEN and the EPA market model are designed to analyze competitive markets hi
which an intermediate product makes up a large proportion of the value of the finished product.
B-8
-------
This may be a fair characterization of the cattle industry and much of hog production. Poultry,
however, is produced by large integrators who sell to a small number of distributors and grocery
chains, i.e., a bilateral oligopoly with fully integrated production. Dairy markets are dominated by
farmer cooperatives and retail chains, i.e., a bilateral oligopoly. COSTBEN and the EPA market
model are not sophisticated enough to capture any of the game theoretic workings of oligopolistic
markets. A model including these aspects of the market would require assumptions about many
more parameters. The values of these parameters have not been established in the empirical
literature. While these are significant deficiencies, the general directions of changes in prices and
quantities do not change because of these market structure issues.
In highly integrated production processes, such as broilers and turkeys, the farm price
reported by USDA is an internal transfer price imputed from the retail price and other
information. The amount farmers receive per pound for contract production is considerably less.
As the actual contract payments are not public information, they could not be used in the EPA
market model. Thus, poultry impacts and production changes are in the context of USDA prices
which may be considerably different from the situation for contract growers.
The model for each animal sector has eleven equations.3 Table B-2 summarizes the model
equations. Up to three equations model the farm product supply sector. One equation accounts
for domestic production in each animal sector. In the cattle and hog sectors, two equations also
model animal imports and exports. A fourth equation combines these together into the trade-
adjusted retail product supply function. Similarly, three equations model the retail product
market, accounting for domestic retail product demand, foreign imports, and exports. A fourth
equation combines these together into the trade-adjusted retail product demand function. A single
equation models the processing sector. A final equation closes the model by finding the long-run
market equilibrium price of the retail product. All of the relationships within the model are linear
of the form Q=A+BP; thus all of the equilibria are unique and stable. The notation for the model
is summarized in a glossary at the end of this appendix. The glossary shows each symbol, its
definition, and the numbers of the equations in which it is used.
B.3.1 Farm Production Sector
This section describes the four equations that define the supply of farm product available
for processing, which include: the farm product imports equation (Section B.3.1.1); the farm
product exports equation (Section B.3.1.2); the domestic farm product supply equation
(Section B.3.1.3); and the trade-adjusted farm product supply equation (Section B.3.1.4).
3Of these equations: (1) two equations are summary equations that combine equations to determine trade-
adjusted supply and demand; and (2) two equations are identities that close the model by rinding the long-run
market equilibrium price of the retail product.
B-9
-------
Table B-2. General Structure of the Model
Equation
Name
Farm Production Sector
1
2
3
4 (summary)
Farm Product Imports Equation
Farm Product Exports Equation
Domestic Farm Product Supply Equation
Trade-Adjusted Farm Product Supply Equation, combines 1-3
Processing Services Sector
5
6
Definition of Conversion Ratio
Marginal Costs of Processing
Retail Product Sector
7
8
9
10 (summary)
13
Retail Product Import Equation
Retail Product Export Equation
Domestic Retail Product Demand Equation
Trade-Adjusted Retail Product Demand Equation, combines
Price of Retail Product at Market Equilibrium, equates 4 and
7-9
10
S.3.1.1 Farm Product Imports Equation
Cattle and hogs are imported into the U.S. for slaughter. This equation and the analogous
export equation are not used in the other sector models. Imports of these farm products are
considered a function of the U.S. price, PL4:
QLr = a, .+ pt PL
(1)
where the intercept, al} and the price coefficient, pl5 are calculated from the specified elasticities
and baseline market conditions. The model is of constant slope form. The relationships between
4A11 farm product variables are subscripted L; retail market variables are subscripted, M. This is a
holdover from the red meat-oriented COSTBEN model which dealt solely in livestock and meat. The subscript b
indicates the value of the variable in the baseline equilibrium. Superscript I indicates imports, X indicates exports,
and S indicates a net total of domestic products, imports, and exports.
B-10
-------
prices and quantities are specified as elasticities, i.e., r\= (5Q/3P)(P/Q). Elasticities must be
converted to slopes, P= 3Q/8P, to define the model's functions. At the baseline equilibrium price
and quantity, P= ^(Q/P). The intercept, a, is found by inserting P and the baseline values for Q
and P into equation 1 and solving for a. A similar process is followed to calibrate all of the
coefficients:
QL = Quantity of farm product imported
PL = Price of farm product
ctj = Intercept of farm product import supply
j = Coefficient on PL of farm product import supply
baseline imported product quantity
- - -
baseline farm price
LMSE = Import supply elasticity of farm product
B.3.1.2 Farm Product Exports Equation
Cattle and hog exports are also a function of current U.S. price:
x
P2?L
(2)
where QLX is the quantity of farm products exported and the coefficients are calibrated as in
equation 1.
B-ll
-------
B.3.1.3 Domestic Farm Product Supply Equation
In the short-run COSTBEN model, domestic farm product supply converges to the long-
run equilibrium by means of a lagged output coefficient. In repeated solutions of the model, the
lagged prices and quantities move the solution gradually toward the long-run equilibrium as prices
and quantities adapt. In the EPA long-run model, the lagged variables are irrelevant because the
markets have reached a stable equilibrium such that lagged variables equal current variables, i.e.,
P, = Pt.j and Qt = QM. The long-run domestic farm product supply is simply:
(3)
where QL is domestic production of the farm'product and the coefficients are calculated from
baseline data and parameters in the same way as in equation 1. ,
B3.1.4 Trade-Adjusted Farm Product Supply Equation
For convenience, it is useful to combine the farm product export, import, and domestic
supply equations into a single, long-run, trade-adjusted farm product supply equation by summing
the farm production and import equations (equations 3 and 1), and subtracting the farm product
exports equation, (equation 2). The result is equation 4:
(4)
where:
Qjf = Quantity of farm product available for processing
«0 = Intercept of trade-adjusted farm product supply equation
= «I - «2 + «3
P0 = Coefficient on PL of trade-adjusted farm product supply equation
= P, - P2 + P3
B-12
-------
B.3.2 The Processing Sector
The processing industry uses labor, capital, and other inputs to convert farm products into
a finished retail product. COSTBEN and the EPA market model similarly convert farm product
prices and quantities into retail product prices and quantities. The conversion ratio (Conv) is the
baseline quantity of retail product produced divided by the baseline amount of farm product
introduced to the process:
Conv =
(5)
The conversion ratio expresses the amount of farm product needed to make a unit of
finished product and changes the units of measure from farm level units, such as number of
animals or hundredweight of milk, in the farm product market to pounds of retail product.
The processor also adds value to the product using labor and capital. Thus, the price of
the retail product is higher than the price of the farm product by the processor's marginal costs.
The EPA market model differs slightly from COSTBEN in its treatment of processors' costs.
COSTBEN assumes marginal costs increase linearly by a factor of St as the quantity of retail
product processed increases. This assumption avoids the need to include the costs of other
processing inputs in the model and may be defensible, since farm products are a large proportion
of processors' costs. In COSTBEN, the price of the farm level output is equal to the price of the
retail product less the marginal cost of processing times the conversion ratio:
PT = Conv*(PM - (! + S^QM)) (6a)
where:
PM = Price of retail product
= Quantity of net retail product demand
„_,'.. baseline retail product quantity
Conv = Conversion ratio = > — —
baseline farm product quantity
B-13
-------
(j)j = Intercept of marginal cost of processing services equation
= p;
M,b
Conv
= Coefficient on Q^ of marginal cost of processing services equation
Measurement of the rate of change of processors' costs is difficult, so reliable estimates of
8j are not available. The EPA market model carries COSTBEN's simplification further by
assuming the marginal costs of processing are constant in the long run and estimating them from
the baseline data as 4>t. With the assumption of constant processors' marginal costs in the EPA
market model, the processing services equation becomes:
PL = Conv*(PM -
(6)
B.3.3 Retail Product Sector
This section describes the four model equations that characterize the retail product
market, which include: the retail product 'import equation (Section B.3.3.1); the retail product
export equation (Section B.3.3.2); the domestic retail product demand equation (Section B.3.3.3);
and the trade-adjusted retail product demand equation (Section B.3.3.4).
B.3.3.1 Retail Product Import Equation
The retail product import equation is:
QM = °i
•M
(7)
B-14
-------
where:
M = Quantity of retail products imported
j = Intercept of retail product import supply equation
= Coefficient on PM of retail product import supply equation
MMSE = Elasticity of retail product import supply
B.3.3.2 Retail Product Export Equation
The retail product export demand equation is:
QM = °2
(8)
where QMX is the quantity of retail product exported and the coefficients are calibrated similarly to
equation 7.
B.3.3.3 Domestic Retail Product Demand Equation
The domestic retail product demand equation is:
QM = °
:M ~ "3 '3XM
(9)
where QM is the quantity of domestic retail product demand and the coefficients are calibrated
similarly to equation 7.
B-15
-------
B.3.3.4 Trade-adjusted Retail Product Demand Equation
As in the production sector, it is convenient to combine the retail product import, export,
and domestic demand equations into a single trade-adjusted retail product demand equation by
adding retail product exports and domestic demand (equations 8 and 9), and subtracting retail
product imports'(equation 7):
QM = v*
M
(10)
where:
QM = Quantity of net retail product demand
o"0 = Intercept of trade-adjusted retail product demand
= a3 + a2 - a,
Y0 = Coefficient on PM of trade-adjusted retail product demand
= ,Y3 + Y2 - Yi
B.3.4 The Long-Run Market Equilibrium
Algebraically, the first step to solving for the market equilibrium is to put the trade-
adjusted farm product supply in terms of the retail product market. To accomplish this, the
processing services equation (equation 6) is substituted for PL in the trade-adjusted farm output
supply equation (equation 4). Equation 5 converts the QLS on the left-hand side to QMS. The
result is equation 11, which is the farm product supply function converted into retail market
terms:
QM
Conv
P0(Conv(PM -
(11)
Solving the supply equation 11 for QMS and setting it equal to demand (equation 10) yields:
°o + YOPM = Conv(a0 + p0Conv(PM - 4),)) (12)
B-16
-------
Solving for PM gives the equilibrium retail price in terms of the coefficients of the model:
o0 - Conv(cc0 + Conv
Conv2p0-Y0
(13)
The analogous retail product price equation in COSTBEN is more complex because it
includes the lagged farm output price and quantities and the variable marginal processing costs of
the retail product. The lagged variables were part of the short-run adjustment process in the
model and are unnecessary in EPA's long-run equilibrium model. As discussed above, the
variable marginal costs contribute to the reality of the model by indicating the additional costs of
other inputs, but also introduce an additional parameter which is poorly measured. For the small
changes considered in this model, the assumption of constant marginal processing costs is of little
consequence to the results.
After the retail product price is calculated, the quantity of retail product imported,
exported, and consumed domestically can be calculated from their respective equations. The
price of the farm product is calculated from the wholesale price of the retail product using the
marginal cost of processing services equation, and the quantity of farm output imported,
exported, and produced domestically can be calculated from their respective equations.
B.4 USING THE MARKET MODEL
B.4.1 Measuring Changes in Prices and Quantities
In the COSTBEN model, all of the coefficients can be adjusted to reflect changes in
policy. The COSTBEN model contains the flexibility to analyze export promotion and import
restriction programs as well as shifts in demand and supply. EPA expects that the proposed
CAFO regulations will primarily affect domestic farm product supply. The market model
evaluates impacts in terms of market responses to a shift in the domestic farm product supply
function. Knowing this in advance, the EPA market model can be simplified to accommodate
only a single parameter change.
There are several conceivable ways to shift the domestic supply of farm product function
in response to the proposed CAFO regulations. The EPA market model takes a constant slope
approach while COSTBEN takes a constant elasticity approach. In the COSTBEN model, shocks
in domestic supply enter the model as proportional changes in the coefficients of the initial supply
function. (The COSTBEN model user must calculate the proportional change in supply
attributable to the policy question of interest through some process outside of the model itself.)
The COSTBEN postregulatory domestic farm product supply function is:
B-17
-------
QL = (a3 + p3PL )*(!+% supply shift)
(14)
Both the intercept, a3, and the slope, P3, of the domestic supply equation change from their
preregulatory values by the same proportion. Postregulatory equilibrium prices and quantities are
then calculated using this postregulatory domestic farm product supply equation in place of the
preregulatory farm product supply equation. The COSTBEN approach to modeling the supply
shift ensures that the postregulatory supply function has the same elasticity as the preregulatory
supply function, but it will have a different slope.
The EPA market model's focus is a shift in supply caused by regulatory compliance costs.
The shift is considered to be the change in price required to supply the initial quantity of farm
products and cover the new costs of the proposed regulations, i.e., a parallel upward shift in the
supply curve. CAFO regulatory compliance involves both capital improvements and changes in
operations. The EPA market model is a long-run, national analysis; all factors of production can
be changed, and all costs may be considered continuously variable. Thus CAFO regulatory
compliance costs may reasonably be spread over all of the units of production without regard to
fixed or variable costs or the "lumpiness" of compliance investments. The supply curve must rise
by the average compliance cost per unit of farm product (Shock) to cover the farm costs of the
proposed regulation. The shocked intercept, a3Shocked, is calculated as if the price at the
preregulatory quantity includes the shock:
..Shocked
= QL,b - PsO^b + shock>
(15)
where:
..Shocked
-------
B.4.2 Industry Direct Impacts
One measure of direct impacts is the change in the total value of the products produced by
the industry, i.e., industry output. The shift in the supply curve discussed above causes four
changes in industry output. At the farm level, compliance costs are absorbed and the quantity
sold is reduced. At the retail level, processors' sales are reduced and compliance costs are
absorbed by consumers. Each of these impacts is distinct, so they may be added together to show
the total direct impact of the proposed regulation.
The shaded area of the Domestic Farm Product Market panel in Figure B-2 illustrates the
direct impact of additional compliance costs on farm output. Farm impacts have two
components: absorbed compliance costs and lost sales volume due to the price increase. The
shocked supply function, Supply^ is higher than the baseline supply function by the shock,
Shock=PL1-PL', where PL' is the price corresponding to the new quantity on the baseline supply
function, Supply0. The proportion of compliance costs per unit that farms cannot pass on through
the marketing chain is PLb-PL'. Thus total costs to be absorbed at the farm level are (PLb-PL')QLi-
The lost sales volume is simply the change in quantity multiplied by the baseline price, PLb(QLb-
Qn). Summing these two quantities gives the farm level direct impact, PLhOurPi/Qu-
The EPA assumption of constant marginal costs precludes processors from absorbing
passed through costs. For the more vertically integrated animal sectors, there is strong evidence
that integrators do not absorb changes in input costs but quickly pass them on to the retail level
(Mclntosh et al., 1997). The direct impact on the processing sector is the change in quantity
produced times its baseline price, PMb(QMbS-QMiS)3 shown in the Retail Product Market panel of
Figure B-2.
There is also an impact on consumers who must devote more resources to buy a similar
amount of animal products. This direct impact on consumers' spending is the change in price
multiplied by the postregulation quantity, (PMrpMb)QMiS-
B.4.3 Input/Output Analysis
EPA applies Regional Input-Output Modeling System, version 2 (RIMS IT)(USDC,
1997b) multipliers to estimate the effect of the proposed regulations on national employment
(measured hi terms of full-time equivalents) and economic output (measured in terms of changes
hi Gross Domestic Product). These estimated changes are based on the estimated direct impacts,
described in the preceding section, which are in terms of dollars of industry output per year.
The published RIMS II multipliers apply to changes in industry final demand rather than
changes in output. To apply to industry output, the row of published multipliers for the affected
industry must be divided by the multiplier for that industry. The new row sum will be the output-
B-19
-------
driven multiplier for the industry (USDC, 1997b). All of the multipliers used in this analysis are
output-driven. Table 4-14 presents the multipliers used in this analysis.
B.4.3.1 Employment
A HEMS n own-industry multiplier5 expresses for each industry the number of full-time
equivalent jobs per million dollars of industry output (USDC, 1997b). The employment
multipliers are driven by a dollar measure, output, and so must be adjusted for inflation. The
employment multipliers are derived from 1992 data. They are adjusted to the 1997 price levels of
the baseline data using the Consumer Price Index for all urban consumers (Council of Economic
Advisors, 2000). The direct impact on farm employment is:
CPI
Direct' Employment Impact = (PL)bQL>b -
E (
'"2
)
(16)
where prices and quantities are as previously defined and:
E = Own-industry RIMS n employment multiplier for production sector
CPI1992 = Consumer Price Index (CPI-U) for 1992
CPI1997 = Consumer Price Index (CPI-U) for 1997
Total effects multipliers, minus the own-industry portion, estimate the ripple effect of farm
level changes on other industries throughout the national economy. Thus, the indirect
employment effects are:
/
Indirect Employment Impact = (PLbQLb-PLQL,i) (V~E)
)
(17)
where V is the RIMS n total employment multiplier for the production sector, e.g. poultry and
egg production. Processing industry and household indirect employment impacts are calculated
similarly from the direct impacts on those industries. The agricultural component is subtracted
from the processing industry multiplier to avoid counting the impact on farms twice.
5See Section 4.4.1.2 for a discussion of input-output analysis, multipliers, and RIMS II.
Households are treated as an industry.
B-20
-------
B.4.3.2 National Output
EPA estimates the total reduction in national output (Gross Domestic Product) by
multiplying the direct impacts by a RIMS n total output multiplier for the industry, adjusted to
avoid double counting the agricultural component. The inflation adjustment used to calculate the
indirect employment effects is unnecessary for output multipliers.
Farm, processing, and household impacts are added together to show total reductions.
The netting out of own-industry and agricultural effects from the multipliers ensures that impacts
are not double counted when sectors are added together.
Spending to build treatment lagoons and buy manure handling equipment is a stimulus to
the national economy. Although many economists argue that spending for pollution control
equipment should be considered a cost to "right past wrongs," such spending does stimulate the
economy and add jobs and income. None of the results presented in Section 5 include any
stimulus effects. In most cases, a stimulus of this sort offsets the impacts of the regulatory action
(see, for example, USEPA, 1998b, or USEPA, 2000c).
B.5 GLOSSARY OF NOTATION
Tables B-3, B-4, and B-5 indicate, respectively, the COSTBEN/EPA model variables,
coefficients, and parameters used.
B-21
-------
Table R-3. COSTBEN/EPA Model Variables
Symbol "
QL'
PL
PL'
QLX
QL
QLS
QMS
PM
QM'
n x
VM
Q
Description
Quantity of farm product imported
Price of farm product
Price of farm product , postregulatory quantity
on preregulatory supply function
Quantity of farm product exported
Quantity of farm product produced domestically
Quantity of farm product available for
processing
Quantity of net retail product demand
Price of retail product
Quantity of retail product imported
Quantity of retail product exported
Quantity of domestic retail product demand
Units b/
head:BH; lbs:RTD; doz:L
$/cwt:BHD; ji/lb:RT; 0/doz:L
$/cwt:BHD; £/lb:RT; ji/doz:L
head:BH; lbs:RTD; doz:L
head:BH; lbs:RTD; doz:L
head:BH; lbs:RTD; doz:L
lbs:BDHRT; doz:L
$/lb:BH; indexrD; jzi/lb:RT;
£/doz:L
lbs:BDHRT; doz:L
lbs:BDHRT; doz:L
lbs:BDHRT; doz:L
Equations
1
1,2,3A6
16, 17
2
3,5
4
5,10
6,7,8,9,10,13
7
8
6,9
Data variables and source for prices and quantities (production, utilization, and trade) used for this analysis are as
follows (data are shown in Table 4-14):
Prices:
Beef: NCBA, 2000. Stat99_ll.xls, Table 4.1, Farm-level=Choice fed steers, Cattle-fax average, carcass weight;
Retail-level=Choice retail beef.
Dairy: USDA/ERS, 1998b. December 28,1998, U.S. Dairy Situation at a glance. Farm-level=Milk eligible for
fluid use; Retail-leveI=Consumer Price Index for all dairy products (1982-84=100).
Hog: Pork price spread tables at http://www.econ.ag.gov/Briefing/meatbrif/, see USDA/ERS, 1999c. January
26,1999. Farm-level=51-52% lean, Hog, carcass price; Retail-leveHPork composite retail.
Poultry: All Prices from USDA/WAOB, 1999. p. 74 and 75.
Chicken: Farm-level=Broilers, Average price received by farmers, cents per liveweigtit pound; Retail-level=Young
chicken, composite retail.
Turkey: Farm-level=Turkey, Average price received by farmers, cents per liveweight pound; Retail-level=Whole
frozen birds.
Eggs: Farm-level=Table eggs, prices received by farmers (excludes hatching eggs); Retail-level=Grade A, Large
B-22
-------
Table B-3. COSTBEN/EPA Model Variables (continued)
Quantities:
Beef: Farm-Level Domestic Production: USDA/ERS, 1998b. December 28, 1998, Meat Statistics, Commercial
Slaughter, head (includes calves). Where necessary, head is converted to live weight based on 1997 annual
average live weight of 1,173 Ibs/head for cattle, USDA/NASS, 1998d, Livestock Slaughter, 1997 Annual
Summary, Livestock Slaughter: Number of Head Slaughtered and Average Live Weight by Species and
Month, United States, 1997;
Retail-Level Domestic Production: Beef & Veal: Putnam and Allshouse, 1999. Total Production.
Imports and exports: USDA/ERS, 1999d. November 23, 1999, Cumulative U.S. Livestock and Meat
Imports and Exports, Cattle Imports and Exports, head; Beef &Veal Imports and Exports.
Dairy: Domestic Production: USDA/ERS, 1998b. December 28, 1998, U.S. Dairy Situation at a Glance, Milk
Production, U.S. est.
Imports and exports: NMPF, 1999. Milk equivalent, total solids basis.
Hogs: Domestic Production: USDA/ERS, 1998b. December 28, 1998, Meat Statistics Commercial Slaughter,
head; Imports and exports: USDA/ERS, 1999d. November 23,1999, Cumulative U.S. Livestock and
Meat Imports and Exports, head. Where necessary head is converted to live weight based on 1997 annual
average live weight of 256 Ibs/head, USDA/NASS, 1998d, Livestock Slaughter, 1997 Annual Summary,
Livestock Slaughter: Number of Head Slaughtered and Average Live Weight by Species and Month,
United States, 1997;
Pork Quantities: USDA/WAOB, 1999. p. 73. Total production, imports, and exports.
Poultry: All quantities from USDA/ERS, 1998a. May 21,1998, .
Poultry supply and utilization tables and Egg supply and utilization table.
Chicken: Carcass weight, Total of broilers and other chicken for Net RBC production,
imports, and exports.
Turkey: Carcass weight, Net RBC production, imports, and exports.
Eggs: Egg Supply, million dozen, Total eggs, imports, and exports.
^Q designates quantities and P designates prices. All farm product variables are subscripted L; retail market
variables-are subscripted M. This is a holdover from the red-meat-oriented COSTBEN model which dealt solely in
livestock and meat. When necessary, an additional subscript is added to any variable to denote preregulatory and
postregulatory values. Subscript b indicates baseline, preregulatory value; subscript 1 denotes shocked,
postregulatory value. Superscript I indicates imports,. X indicates exports, and S indicates a net total of domestic
products, imports, and exports.
b/B=Beef, D=Dairy, H=Hogs, RHBroilers, L=Layers, T=Turkeys. Cattle and Hog number of head is converted to
pounds at the rate of 1,173 Ib/head for cattle and 256 Ib/head for hogs (USDA/NASS, 1998d).
B-23
-------
Table B-4. COSTBEN/EPA Model Coefficients
Symbol
a,
P,
«2
P*
«3
P3
«0
Po
6,
*•
Ol
Yi
02
Y2
03
Y3
a0
Yo
... Shocked
OCj
Description
Intercept of farm product import supply
Coefficient on PL of farm product import supply
Intercept of farm product export demand
Coefficient on PL of farm product export demand
Intercept of domestic farm product supply
Coefficient on PL of domestic farm product supply
Intercept of trade-adjusted farm product supply,
= ara2+a3
Coefficient on PL of trade-adjusted farm product
supply, = P,-P2+P3
Coefficient on QMS of marginal cost of processing
services equation (assumed to be zero)
Intercept of marginal cost of processing services
equation
Intercept of retail product import supply
Coefficient on PM of retail product import supply
Intercept of retail product export demand
Coefficient on PM of retail product export demand
Intercept of domestic retail product demand
Coefficient on PM of domestic retail product
demand
Intercept of trade-adjusted retail product demand,
= 03+a2-a,
Coefficient on'PM of trade-adjusted retail product
demand, = Ys+YrYi '
Shocked Intercept of domestic farm product
supply
Units *
head:BH; lbs:RTD; doz:L
head/$/cwt:BH; lbs/$/lb:D;
lbs/0/lb:RT; doz/0/doz:L
head:BH; lbs:RTD; doz:L
head/$/cwt:BH; Ibs/$/lb:D;
Ibs/jzi/lb:RT; doz/0/doz:L
headrBH; IbsrRTD; dozrL
head/$/cwt:BH; lbs/$/lb:D;
lbs/0/lbrRT; doz/0/dozrL
headrBH; lbs:RTD; doz:L
head/$/cwt:BH; lbs/$/lb:D; .
lbs/0/lb:RT; doz/0/dozrL
$/lb/lb:BH; index/lb:D;
0/lb/lb:RT; 0/doz/doz:L
$/lb:BH; index:D; 0/lb:RT;
ji/doz:L
lbs:BDHRT; doz:L
lbs/$/lb:BD; lbs/0/lb:HRT;
doz/0/doz:L
lbs:BDHRT; doz:L
lbs/$/lb:BD; lbs/0/lb:HRT;
doz/0/doz:L
lbs:BDHRT; doz:L
lbs/$/lb:BD; lbs/)i/lb:HRT;
doz/^/doz:L
IbsrBDHRT; doz:L
lbs/$/lb:BD; lbs/0/lb:HRT;
doz/0/doz:L
lbs:BDHRT; doz:L
Equations
1
1
2
2
3 •
3
4,13
4,13
6a
6a,6,13
7
7
8
8
9
9
10,13
10,13
15
f, D=Dairy, H=Hog, R=Broiler, L=Layer, T=Turkey. Beef and Hog number of head is converted to pounds
at the rate of 1,173 Ib/head for beef and 256 Ib/head for hogs (USDA/NASS, 1998d).
B-24
-------
Table B-5. COSTBEN/EPA Model Parameters
Symbol
Conv
LRSE
LMDE
LMSE
LXDE
MMSE
MXDE
Shock
E
V
CPI1992
CPI1997
Description
Conversion Ratio from farm to retail
Long-run farm product supply elasticity
Long-run retail product demand elasticity
Import supply elasticity of farm product
Export demand elasticity of farm product
Import supply elasticity of retail product
Export demand elasticity of retail product
Annualized compliance costs per unit of
farm product
Own-industry employment multiplier
Industry total employment multiplier
Consumer Price Index for all urban
consumers (1982-84=100) for 1992,
= 140.3
Consumer Price Index for all urban
consumers (1982-84=100) for 1997,
= 160.5
Units "
lbs/head:BH; Ib retail/lb farm:DRT
doz retail/doz farm:L
unitless
unitless
unitless
Unitless
unitless
unitless
$/cwt:BHD; ji/ftcRT; j£/doz:L
jobs/$ output .
jobs/$ output
unitless
unitless
Related
Coefficient
none
Ps
Ys
P.
P,
Yi
Yz
• Shocked
CC3
none
none
none
none
=Beef, D=Dairy, H=Hog, R=Broiler, L=Layer, T=Turkey.
B-25
-------
-------
APPENDIXC
SUMMARY OF DEMAND AND SUPPLY
ELASTICITY LITERATURE
This appendix presents the results of EPA's literature review and the magnitudes of
published demand and supply elasticities for the beef, dairy, pork, and poultry sectors.
V . •
EPA has reviewed the available literature on the demand and supply characteristics of the
beef, dairy, pork, and poultry markets (ERG, 1999a and ERG, 1999b). These expanded reviews
include an annotated summary of each study and are contained in the record (DCN 70642 and
DCN 70362). The majority of the models in the literature are based on econometric estimations
of various demand and supply system specifications, such as the Almost Ideal Demand System
(AIDS) and the Rotterdam model. However, given the prevalence of non-theoretical approaches
to estimating demand and supply responses in the literature using such techniques as vector
autoregression (VAR), EPA also includes those studies in the tables where applicable.
EPA summarizes the estimates of demand and supply elasticities across a wide range of
poultry meat products including beef, milk, broilers, chicken1, turkey, and poultry meat as a
composite commodity. EPA groups the market studies in a general poultry meat category
because many market studies include either both turkey and chicken elasticities or a general
poultry category.
market studies identified demand elasticities for "chickens" instead of "broilers." EPA has
identified these elasticity estimates as chicken to be consistent with the author's language.
C-l
-------
Table C-l. Demand Elasticities for Beef Products Ranked from the Lowest Estimate to the Highest Estimate
Source
Bales and Unnevehr (1988)
Capps (1989)
Brester and Wohlgenant (1991)
Heien and Pompelli (1988) *
Moschini and Meiike (1989)
Huang and Hahn (1995) ^
Gao and Shonkwiler (1993) ^
Kesavanetal.(1993)a/ .
Brester and Wohlgenant (1991)
Ospina and SHumway (1979)
Alston and Chalfant (1993)
Choi and Sosin (1990)
Brester (1996)
Chavas (1983)
Hahn (1 994)*
Bales and .Unnevehr (1993)
Heien and Pompelli (1988) ^
Moschini, Moro, and Green (1994)
Ospina and Shumway (1979) '
Brester and Wohlgenant (1991)
Brester (1996)
Wohlgenant (1989)
Marsh (1992)
Heien and Pompelli (1988) ^
Capps (1989)
Brester (1996)
Bales and Unnevehr (1988)
Elasticity Estimate
-2.59 (hamburger)
-1.27 (roast beef)
-1.155 (fed beef)
-1.11 (roast)
-1.05 (beef)
-1.036 (high quality beef)
-1.03 (beef)
-1.02 (long-run, beef)
-1.015 (ground beef)
-0.98 (fed beef; Langemeier and Thompson, 1967)
-0.98 (beef)
-0.971 (red meat)
-0.96 (ground beef)
-0.916 (beef)
-0.869 (beef)
-0.850 (beef)
-0.85 (ground beef)
-0.84 (beef) .
-0.83 (fed beef; Freebairn and Rausser, 1975)
-0.811 (table-cut beef)
-0.80 (table-cut beef)
-0.76 (beef and veal)
-0.742 (retail beef)
-0.73 (steaks)
-0.72 (steak)
-0.70 (beef)
-0.68 (table-cut beef)
C-2
-------
Table C-l. Demand Elasticities for Beef Products Ranked from the Lowest Estimate to the Highest Estimate
(continued from Drevious
Source
Marsh (1991)
Huang (1993)
Huang (1986)
Hahn(1988)
Bales and Unnevehr (1988)
Ospina and Shumway (1979)
Marsh (1992)
Marsh (1992)
Arzac and Wilkinson (1979)
Brester and Wohlgenant (1993) ^
Huang and Hahn (1995) *
Capps (1989)
Elasticity Estimate
-0.66 (choice slaughter beef)
-0.6212 (beef and veal)
-0.6166 (beef and veal)
-0.58 (beef)
-0.570 (beef)
-0.57 (wholesale beef) .
-0.536 (farm beef)
-0.495 (wholesale beef)
-0.49 (fed beef)
-0.45 (beef)
-0.401 (manufacturing grade beef)
-0.15 (ground beef)
As cited in Hahn (1996a).
Table C-2. Sunolv Elasticities for Beef Products Ranked from the Lowest Estimate to the Highest Estimate
Source
Marsh (1994)
Ospina and Shumway (1979)
Ospina and Shumway (1979)
Marsh (1994)
Marsh (1994)'
Marsh (1994)
Marsh (1994)
Marsh (1994)
Marsh (1994)
Marsh (1994)
Buhr(1993)
Elasticity Estimate
-0.17 (short-run, fed cattle)
0.06 (steer-heifer fed beef; Folwell and Shapouri, 1977)
0.14 (slaughter beef)
0.14 (all beef; Freebairn and Rausser, 1975)
0.14 (fed beef; Shuib and Menkhaus, 1977)
0.200 (wholesale fed beef; Bedinger and Bobst, 1988)
0.23 (fed beef; Langemeier and Thompson, 1967)
0.606 (intermediate run, fed cattle)
0.993 (beef; Tvedt, etal., 1991)
3.24 (long-run, fed cattle) •
9.505 (beef, long-run - 5 years) ^
37The estimate does not reflect GET because it is not comparable to the other elasticity estimates. The reported
figure is the impact of a 10 percent change in farm price rather than the standard 1 percent. Given the nonlinear
nature of the system, the figure cannot be translated into a standard elasticity estimate via division by 10.
C-3
-------
Table C-3. Demand Elasticities for Milk Ranked from the Lowest Estimate to the Highest Estimate
Source
Holt and Aradhyula (1995)
Tanjuakio, Gempesaw, and Elterich
(1992)
Tanjuakio, Gempesaw, and Elterich
(1992)
Tanjuakio, Gempesaw, and Elterich
(1992)
Tanjuakio, Gempesaw, and Elterich
(1992)
Huang (1986)
Huang (1993)
Holt and Aradhyula (1995)
Tanjuakio, Gempesaw, and Elterich
(1992)
Tanjuakio, Gempesaw, and Elterich
(1992)
Elasticity Estimate
-0.65 (long-run, fluid milk)
-0.48 (fluid milk; Bailey, et al., 1990)
-0.47 (manufactured milk)
-0.43 (manufactured milk; Kaiser and Tauer, 1988)
-0.37 (manufactured milk; Bailey et al, 1990)
-0.2580 (fluid milk)
-0.2472
-0.23 (short-run, fluid milk)
-0.19 (fluid milk)
-0.05 (fluid milk; Kaiser and Tauer, 1988)
Table C-4. Supply Elasticities for Milk Ranked from the Lowest Estimate to the Highest Estimate
Source
Elasticity Estimates
Short-Run
Howard and Shumway (1988)
Chavas and Klemme (1986)
Tanjuakio, Gempesaw, and Elterich (1992)
Weersink and Howard (1990)
-0.075 (milk, short-run)
0.07 to 0.16 (milk, short-run; Hammond et al., 1974,
Chen et al., 1972, and Hutton and Helmberger, 1982) '
0.11 to 0.43 (short-run, milk; Weersink and Tauer, 1990)
0.1 18 to 0.639 (short-run, milk)
Long-Run
Huyetal. (1988)
Blayney and Mittelhammer (1990)
Chavas and Klemme (1986)
Chavas, Kraus, and Jesse (1990)
Chavas and Klemme (1986)
Howard and Shumway (1988)
-0.322, to .884 (milk)
0.12 (milk, at 1 year; Dahlgran, 1985)
0.12 (milk, at 1 year)
0.139 (milk, at 1 year)
0.14 (milk, long-run; Hammond, 1974)
0.144 (milk, long-run)
C-4
-------
Table C-4. Supply Elasticities for Milk Ranked from the Lowest Estimate to the Highest Estimate
(continued from previous page)
Source
Buxton(1985)
Tanjuakio, Gempesaw, and Elterich (1992)
Tanjuakio, Gempesaw, and Elterich (1992)
Buxton(1985)
Weersink and Howard (1990)
Blayney and Mittelhammer (1990)
Chavas, Kraus, and Jesse (1990)
Chavas and Klemme (1986)
Chavas and Klemme (1986)
Chavas and Klemme (1986)
Chavas, Kraus, and Jesse (1990)
Chavas and Klemme (1 986)
Elasticity Estimates
0.175 (milk, at 1 year)
0.016 to 0.29 (milk)
0.25 to 0.46 (long-run, milk; Weersink and Tauer, 1990),
0.5 10 (milk, at 4 years)
0.145 to 0.664 (long-run, milk)
0.8932 (milk)
1.527 (milk, at 10 years)
2.20 (milk, long-run; Dahlgran, 1985)
2.46 (milk, at 10 years)
2.53 (milk, long-run; Chen et al., 1972)
4.787 (milk, at 29 years)
6.69 (milk, at 30 years)
Table C-5. Demand Elasticities for Pork Ranked from the Lowest Estimate to the Highest Estimate
Source
Bales and Unnevehr (1993)
Kesavanetal(1993)a/
Gao and Shonkwiler (1993) *
Arzac and Wilkinson (1979)
Moschini and Meilke (1989)
Huang and Hahn (1995) ^
Huang (1994)
Capps (1989)
Bales and Unnevehr (1993)
Lemieux and Wohlgenant (1989)
Hahn(1988)
Brester and Wohlgenant (1991)
Brester and Wohlgenant (1991)
Elasticity Estimate
-1.234 (pork - AIDS with SI)
-0.99 (pork - long-run)
-0.95 (pork)
-0.87 (pork)
-0.839 (pork)
-0.838 (pork)
-0.8379 (pork)
-0.8279 (pork loin)
-0.801 (pork -AIDS without SI) ' - • . •
-0.80 (pork)
-0.784 (pork)
-0.779 (pork - ground beef model)
-0.775 (pork - nonfed model)
C-5
-------
Table C-5. Demand Elasticities for Pork Ranked from the Lowest Estimate to the Highest Estimate
(continued from previous page)
Source
Bales and Unnevehr (1988)
Huang (1986)
Huang (1993)
Chavas (1983)
Chavas (1983)
Capps (1989)
Moschini, Moro, and Green (1994)
Hahn (1994) *
Brester and Schroeder (1995)
Bales and Unnevehr (1988)
Bales etal (1998)
Wohlgenant(1989)
Capps and Schmitz (1991)
Wohlgenant (1989)
Capps (1989)
Capps (1989)
Alston and Chalfant (1 993)
Alston and Chalfant (1993)
Elasticity Estimate
-0.762 (pork - aggregate system)
-0.7297 (pork)
-0.7281 (pork)
-0.723 (pork - SC)
-0.714 (pork - WSC)
-0.7005 (pork chops)
-0.68 to -0.72 (pork)
-0.699 (pork)
-0.69 (pork)
-0.565 (pork - disaggregated system)
-0.52 (pork)
<-0.51 (pork - unrestricted)
-0.4510 (pork)
-0.36 (pork - restricted)
-0.3596 (ham) ' .
-0.2639 (composite pork commodity)
-0.17 (pork - Rotterdam)
-0.07 (pork - AIDS)
As cited in Hahn (1996a).
C-6
-------
Table C-6. Supply Elasticities for Pork Ranked from the Lowest Estimate to the Highest Estimate
Source
Elasticity Estimate
Short-Run
Holt and Johnson (1988)
Heien(1975)
Meilke et al. (1974)
Meilke et al. (1974)
Lemieux and Wohlgenant (1 989)
Buhr (1993)
0.007 (pork, short-run - 3 quarters)
0.09 (pork) ^
0.16 (hog, short-run - GDL)
0.17 (hog, short-run - PDL)
0.4 (pork, short-run)
2.63 (pork, short-run - 1 quarter) b/
Intermediate-Run
Meilke etal. (1974)
Holt and Johnson (1988)
Lemieux and Wohlgenant (1989)
0.24 (hog, intermediate-run - PDL)
0.338 (pork, intermediate-run - 10 quarters)
1.8 (pork, intermediate-run
Long-Run
Meilke et al. (1974)
Meilke et al. (1974)
Holt and Johnson (1988)
Buhr (1993)
0.43 (hog, long-run - GDL)
0.48 (hog, long-run - PDL)
0.628 (pork, long-run - 40 quarters)
7.35 (pork, long-run - 5 years) b/
37 The reported figure is the elasticity of total number of pigs slaughtered with respect to the ratio of farm to retail
price of pork.
b/The estimate does not reflect GPT because it is not comparable to the other elasticity estimates. The reported
figure is the impact of a 10 percent change in farm price rather than the standard 1 percent. Given the nonlinear
nature of the system, the figure cannot be translated into a standard elasticity estimate via division by 10.
C-7
-------
Table C-7. Demand Elasticities for Broilers/Chickens Ranked from the Lowest to the Highest Estimate
Source
Kesavanetal(1993)a/
Arzac and Wilkinson (1979)
Alston and Chalfant (1993) .
Bales and Unnevehr (1988)
Capps (1989)
Bales and Unnevehr (1988)
Huang (1986)
Gao and Shonkwiler (1993) ^
Huang (1993)
Hahn (1994) *
Bales and Unnevehr (1988)
Bales and Unnevehr (1993)
Huang and Hahn (1995) ^
Huang (1994)
Bales and Unnevehr (1.993)
Bales etal (1998)
Bales etal (1998)
Hahn (1988)
Bales etal (1998)
Moschini and Meilke (1989)
Elasticity Estimate
-1.25 (chicken - long-run)
-0.98 (chicken)
-0.94 (chicken - AIDS and Rotterdam)
-0.677 (chicken - whole bird)
-0.6557 (chicken)
-0.610 (chicken - parts/processed)
-0.5308 (chicken)
-0.47 (chicken)
-0.3723 (chicken)
-0.299 (chicken)
-0.276 (chicken)
-0.233 (chicken - AIDS with SI)
-0.197 (broiler)
-0.1969 (broiler)
-0.162 (chicken - AIDS without SI)
-0.15 (chicken - Model 3)
-0.14 (chicken - Model 1)
-0.140 (chicken)
-0.13 (chicken - Model 2)
-0.104 (chicken)
" As cited in Hahn (1996a).
C-8
-------
Table C-8. Supply Elasticities for Broilers/Chickens Ranked from the Lowest to the Highest Estimate
Source
Elasticity Estimate
Short-Run
Ghavas and Johnson (1982)
Chavas (1982)
Holt and Aradhyula (1990)
Holt and Aradhyula (1990)
Aradhyula and Holt (1989)
Holt and Aradhyula (1990)
Buhr (1993)
0.064 (broiler, short-run)
0.072 (broiler, short-run) d
0.216 (broiler, short-run-adaptive expectations) ^
0.232 (broiler, short-run - GARCH) ^
0.305 (broiler, short-run) ^
0.399 (broiler, long-run - adaptive expectations) ^
0.49 (chicken, short-run - 1 quarter) b/
Long-Run
Holt and Aradhyula (1990)
Holt and Aradhyula (1990)
Buhr (1993)
0.399 (broiler, long-run - adaptive expectations) ^
0.587 (broiler, long-run - GARCH) *
0.68 (chicken, long-run - 5 years) w
7The reported elasticity figure is based on the expected rather than the actual mean price of broilers.
w The estimate does not reflect CPT because it is not comparable to the other elasticity estimates. The reported
figure is the impact of a 10 percent change in farm price rather than the standard 1 percent. Given the nonlinear
nature of the system, the figure cannot be translated into a standard elasticity estimate via division by 10.
d The reported figure is the elasticity of supply with respect to the one-quarter lagged product price.
C-9
-------
Table C-9. Demand Elasticities for Eggs Ranked from the Lowest Estimate to the Highest Estimate
Source
Fuller (1992)
Gerra (1959) "
Chavas and Johnson (1981)
Fuller (1992)
Chavas and Johnson (198 1)
George and King (1971)"
Brown and Schrader (1990)
Brown and Schrader (1990)
Wohlgenant(1989)
Huang (1986)
Huang and Haidacher (1983)
Brown and Schrader (1990)
Huang (1993)
Brown and Schrader (1990)
Brown and Schrader (1990)
Brown and Schrader (1990)
Wohlgenanl (1989)
Brown and Schrader (1990)
Brown and Schrader (1990)
Elasticity Estimate
-0.78 (for price increase)
-0.40
-0.34 (short-run)
-0.33 (for price decrease)
-0.3298 (long-run)
-0.318
-0.17 (with TIME, WW, CHOL)
-0.164 (with TJJME,.CHOL)
-0.15 (unrestricted)
-0.1452
-0.14
-0.129 (with TIME, WW)
-0.1103
-0.109 (with WW, CHOL)
-0.094 (with TIME)
-0.089 (with CHOL)
-0.05 (restricted)
-0.043 (with WW)
-0.022 (no structural change)
31 As cited in Chavas and Johnson (1981).
C-10
-------
Table C-10. SuDmlv Elasticities for Eggs Ranked from the Lowest Estimate to the Highest Estimate
Source
Elasticity Estimate
Short-Run
Chavas and Johnson (1981)
0.031
Long-Run
Chavas and Johnson (1981)
0.9415 (long-ran)
Table C-ll. Demand Elasticities for Turkey Ranked from the Lowest Estimate to the Highest Estimate
Source
Huang (1986) .
Bales et al (1998)
Huang (1993)
Hahn (1994) *
Soliman (1971)
Soliman (1971)
Soliman (1971)
Soliman (1971)
Elasticity Estimate
-0.6797 (turkey)
-0.63 (turkey - Model 1)
-0.5345 (turkey)
-0.459 (turkey)
-0.412 (turkey - 3SLS)
-0.411 (turkey -LISE)
-0.394 (turkey - 2SLS)
-0.372 (turkey - OLS)
" As cited in Hahn (1996a).
Table C-12. Supply Elasticities for Turkey Ranked from the Lowest Estimate to the Highest Estimate
Source
Elasticity Estimate
Short-Run
Chavas and Johnson (1982)
0.21.0 (turkey, short-ran)
Chavas (1982)
0.222 (turkey, short-run)2
Soliman (1971)
0.353 (turkey, short-ran) b/
Long-Run
Soliman (1971)
0.518 (turkey, long-ran)b/
1 The reported figure is the elasticity of supply with respect to the one-quarter lagged product price.
b The reported figure is the elasticity of turkey production with respect to the lagged turkey-feed price ratio.
C-ll
-------
-------
APPENDIX D
SENSITIVITY ANALYSES
This appendix presents EPA's analyses to test the sensitivity of the CAFO model and the
national market model to the model input assumptions. For the CAFO models, EPA conducts a
sensitivity analysis of the financial operating conditions assumed in the baseline by varying key
variables used in the analysis and also by using differing the methodological approach used to
assess impacts (Section D.I). EPA also tests the results of the national market model for
livestock and poultry products to varying commodity prices and price response assumptions
(Section D.2). The results of these sensitivity analyses show that EPA's CAFO model and market
model are reasonably stable over a realistic range of potential variability in impact and market
measures.
For the purpose of this discussion, the "Main Analysis Value" corresponds to data used by
EPA based on the 1997 ARMS data that are assumed to depict baseline financial conditions at
model GAFOs for EPA's main analysis (results presented .in Section 5.2). The "Sensitivity
Analysis Value" corresponds to data that have been modified by EPA to examine variability of the
key input assumptions used for the sensitivity analysis presented here. In each case where the
baseline data have been varied (or the approach varied), all other key input variables and
assumptions are assumed to be the same as they are in the main analysis.
During the development of EPA's analytical model to examine the regulatory costs and
impacts that might accrue to CAFOs as a result of the proposed regulations, EPA also conducted
extensive sensitivity analyses with respect to the estimated compliance costs for model CAFOs
developed by EPA. These results are presented in the Development Document that supports this
rulemaking (USEPA, 2000a). This appendix describes the results of EPA's sensitivity analysis to
EPA's financial models only.
D.I CAFO MODEL SENSITIVITY ANALYSIS
To examine the stability of the results of EPA's CAFO level analysis, EPA conducts
various sensitivity analyses that examine the results of EPA's analysis under both differing
baseline value assumptions, as well as differing methodological approaches. EPA examines the
results of its analysis under differing baseline value assumptions for three key variables, including:
modified gross farm revenue values to reflect lower farm revenues from an operation's livestock
business for use in a sales test (Section D. 1.1), modified net cash income to reflect higher
operating expenses for use in a discounted cash flow analysis (Section D.I.4), and modified debt-
to-asset ratios to reflect greater financial risk for use in a financial ratio analysis (Section D.I.5).
EPA also investigates two alternative methods for undertaking the sales test. These include: use
of pre-tax versus post-tax gross farm revenues for use in a sales test (Section D.1.3) and use of
D-l
-------
livestock income variables only versus entity farm revenue also for use in a sales test
(Section D.I.2).
EPA conducts these sensitivity analyses by estimating the number of adversely affected
operations using the approach for the main analysis (described in Section 4.2.5). Accordingly,
EPA investigates three key financial criteria used in the impact analysis (sales tests, discounted
cash flow, and debt-to-asset ratios) and classifies CAFO level economic impacts into three
categories—affordable, moderate, and stress. Facilities that fall within the "stress" impact
category are considered by EPA to be vulnerable to closure and may indicate that an option is not
economically achievable, subject to other considerations. As in the main analysis, for some
sectors EPA evaluates impacts under both a Zero and a Partial cost passthrough scenario (CPT),
using the approach described in Section 4.2.6 of this report.
EPA re-examines the results of its economic analysis by comparing these results to the
results of these sensitivity analyses. Differences are measured in terms of the number of affected
operations under the stress impact category. The ELG Option and NPDES Scenario combination
examined for this analysis are the proposed BAT Option and Scenario 4b. This scenario would
apply to all CAFOs with more than 300 AU. Although EPA is not proposing this scenario for the
CAFO regulations, EPA evaluates results under this scenario for these sensitivity analyses since
this scenario results in the greatest number of facilities affected and thus allows EPA to more fully
evaluate the results of its financial models used in the main analysis.
EPA decided-to conduct some of these sensitivity analyses because of perceived
limitations with the ARMS financial data that are used for EPA's CAFO analysis. Among the
limitations of these data are that they reflect average financial conditions across an entire sector
•and may not reflect conditions at different subsectors. Variance analysis of these data also are not
available for use in defining a distribution of financial characteristics within a model CAFO. The
ARMS data are also limited in that they represent conditions at a single point in a single year
* (1997), and therefore do not embody the expected variability of farm financial conditions year-to-
year.1 These and other limitations of the ARMS data for use in this analysis are discussed in
Section 4 and in other sections throughout this report, including Sections 6, 7, and 8.
The concern is that use of these limitations may result in understating EPA's economic
impact analysis by overstating a model CAFO's baseline financial conditions and its ability to pay
for on-farm improvements under the proposed regulation (i.e.,'concluding that a model CAFO is
able to afford the estimated regulatory costs when, in fact, it cannot). The'possibility that EPA
may be understating financial conditions at a model CAFO is less problematic since this would
mean that EPA's analysis is overly conservative (i.e., overstates impacts by not fully accounting
for an average facility's ability to afford the estimated regulatory costs when, in fact, it can could
afford costs that are even higher).
'Attachment 1 at the end of this appendix-shows selected annual" market data for each sector from 1992 to
2000, where available. For most sectors, the data reviewed for 1997 are within the range of the period average.
D-2
-------
EPA recognizes the limitations of the ARMS financial data for use in this analysis;
however, EPA also believes that use of these data may produce overly conservative impact
results. The ARMS data are representative of the population, which contains more smaller-sized
operations and fewer larger-sized operations, such as those that would be subject to the proposed
regulations. The ARMS data also include all types of operations—noncommercial, higher cost
producers, as well as smaller scale production units—which may also result in the average
financial statistics being lower than might be expected for the regulated population. Financial
conditions at the larger-sized facilities that would be defined as CAFOs are likely more favorable
than at smaller operations that are more representative of the ARMS data. Since the ARMS data
are likely less representative of the types of larger scale operations that would be affected by the
proposed CAFO regulations, use of these data may overstate estimated impacts to model
facilities. In addition, on the cost side, EPA believes that its estimated costs are conservative, as
documented in the Development Document (USEPA, 2000a).
D.I.I Sales Test Analysis (Lower Livestock Revenue)
The sensitivity analysis described in this appendix section assesses EPA assumptions of the
baseline conditions as they affect the results of EPA's sales test. EPA's main analysis evaluates
impacts using gross farm revenue from the 1997 ARMS data provided by USDA and derived by
EPA on a per-animal basis using available data from ARMS that correspond to-the financial data
(described in Section 4). These data are average and may not reflect conditions at some
operations that may receive lower than average national revenues (e.g., due to differences in
prices received by producers).
To evaluate the baseline assumptions on revenues and to account for revenue differences
among producers, EPA modifies the baseline data used in the main analysis by subtracting 10
percent off the livestock portion of total gross farm revenue and recalculating total revenues,
assuming that the non-livestock revenue portion of revenue remains constant. Table D-l presents
a comparison of the revenues used in the main analysis and the modified revenues used in this
sensitivity analysis (expressed on a per-animal basis).
Table D-2 compares the results of EPA's sensitivity analysis to the results of the main
analysis. Under a zero CPT assumption, the percentage of CAFOs experiencing financial stress
increases from 5.6 percent in the main analysis to 6.0 percent of all CAFOs in the sensitivity ,
analysis. More CAFOs are estimated to experience financial stress in the fed cattle, heifer, and
broiler sectors in the sensitivity analysis. Under a partial CPT assumption, an additional 200
CAFOs (990 CAFOs versus 790 CAFOs) are estimated to experience financial stress under the
alternative revenue assumption compared to the main analysis. As in the main analysis, no hog or
broiler operations are expected to experience financial stress under a partial CPT assumption.
The percentage of operations experiencing financial stress assuming partial CPT increases from
1.7 percent (main analysis) to 2.2 percent (sensitivity analysis) of all CAFOs. Based on the results
of this sensitivity analysis, EPA would likely not change the proposal made in the main analysis
that the proposed CAFO regulations would be economically achievable.
D-3
-------
Table D-l. Baseline Revenues in Main Analysis and Sensitivity Analysis (Lower Livestock Revenues)
Sector
Beef
Dairy
Hog
Broiler
Layer
Turkey
Region
MW
CE
MW
PA
MA
MW
MA
SO
MW/MA
MW/MA
Facility Size
Grouping
Large
Medium
Large
Medium
Large
Medium
Small
Large
Medium
Large
Medium
Large
Medium
Small
Large
Medium •
Large
Medium
All
All
Medium
Main Analysis
Revenue/Head
$861.83
$534.68
$854.00
$501.99
$2,612.95
$2,498.11
$2,619.74
$2,567.29
$2,342.77
$83.60
$296.66
$228.99
$303.89
$606.13
$1.13
$1.47
$1.16
$1.42
$24.63
$20.06
$11.24
Sensitivity
Analysis
Revenue/Head
$810.58
$502.00
$806.09
$468.00
$2,365.97
$2,273.14
$2,404.92
$2,326112
$2,126.16
$78.90
$278.00
$210.38
$28,4.59
$575.40
$1.11
$1.45
$1.14
$1.41
$22.93
$18.42
$10.60
Source: Data used for main analysis are USDA/ERS (1999a), derive.d on a per-animal basis by EPA. Data used for
sensitivity analysis are derived from the data in the main analysis, with USEPA assumptions (adjusts the USDA-
reported livestock portion of total farm revenue is 10 percent lower; farm revenue from other sources remain
constant). Table 4-1 defines regions and facility size groups.
D-4
-------
Table D-2. Number of CAFOs Affected Assuming Alternative Assumption (Lower Livestock Revenues)
Sector
Fed Cattle
Veal
Heifer
Dairy
Hogs
(Zero CPT)
Hogs
(Partial CPT)
Broilers
(Zero CPT)
Broilers
(Partial CPT)
Layers - Wet
Layers - Dry
Turkeys
Total
(Zero CPT)
Total
(Partial CPT)
No. of
CAFOs
4,070
210
1,050
7,140
14,370
14,370
14,140
14,140
360
1,700
2,100
45,140
45,140
Scenario 4b-Sensitivity Analysis
Aff.
Mod.
Stress
(number)
2,780
210
690
5,590
12,660
14,250
0
9,940
360
1,700
1,990
25,980
37,500
1,070
0
300
850
290
120
13,500
4,200
0
0
110
16,120
6,650
230
0
60
700
' 1,420
0
640
0
0
0
0
3,050
990
Scenario 4b-Main Analysis
Aff.
Mod.
Stress
(number)
2,880
210
850
6,010
12,710
14,370
1,960
12,690
360
1,700
1,990
28,680
41,070
1,150
0
150
440
240
0
11,860
1,450
0
0
110
13,950
3,300
40
0
50
700
1,420
0
320
0
0
0
0
•2,520
790
Source: USEPA. Numbers may not add due to rounding.
Evaluated for the BAT Option and Scenario 4b (Options/Scenarios are defined in Table 3-1). Category definitions
("Affordable," "Moderate" and "Stress") are provided in Table 4-43.
D-5
-------
D.1.2 Sales Test Analysis (Livestock Revenue Only)
The sensitivity analysis described in this appendix section assesses a change in
methodology, not a change in baseline assumptions. As described in Section 4, EPA's main
analysis evaluates cost-to-sales ratios (among other criteria) using financial data for the farm
operation as a whole and does not differentiate between an operation's livestock and other
business enterprises (discussed in Section 4). EPA conducts this sensitivity analysis to examine
whether the results of the analysis would differ substantially if only the livestock portion of a
model facility's farm revenue were used to evaluate cost-to-sales ratios in the analysis (i.e., to
examine product line closures). These revenue data are from the 1997 ARMS data. However,
the ARMS data do not differentiate between an operation's cost of production by enterprise; as a
result, the cash flow portion of this analysis uses net cash flow based on an operating costs at the
entity level.
Table D-3 presents a comparison of the revenues used in the main analysis and the
modified revenues used in this sensitivity analysis (expressed on a per-animal basis). As shown,
the reported ARMS data show that livestock revenues comprise roughly one-half of an entity's
operating revenue for most sectors. In the broiler sector, however, livestock revenues are up to
ten times lower than those reported for the entire operation (Table D-3).2 As discussed in Section
4 of this report, EPA believes that an analysis that looks only at an operation's livestock revenue
and not that of an entity may be unrealistic and overly conservative.
Table D-4 compares Hie results of EPA's sensitivity analysis to the results of the main
analysis. Under a zero CPT assumption, an additional number of CAFOs are estimated to
experience financial stress in the fed cattle, heifer, hog, and turkey sectors. All broiler operations
are estimated to experience financial stress even under assumptions of partial cost passthrough.
As in the main analysis, no hog operations are expected to experience financial stress under a
partial CPT assumption. Under a partial CPT assumption—excluding estimated impacts to broiler
operations—an additional 660 CAFOs (1,450 CAFOs versus 790 CAFOs) are estimated to
experience financial stress under the alternative revenue assumption compared to the main
analysis. Disregarding the broiler sector impacts, the percentage of CAFOs considered to
experience financial stress increases from 2.5 percent (main analysis) to 4.7 percent (sensitivity
analysis) of all CAFOs under a partial cost passthrough assumption.
Despite assumptions of cost passthrough, the results for broilers continue to show a high
number of CAFOs in the stress category. EPA considers these impacts to be overstated because
the overall approach overstates costs (i.e., costs are entity level and not by enterprise) and does
not reflect production cost savings to contract growers in some sectors (see Section 2). Given
considerations about the appropriateness of this approach and the use of these data for this
analysis, EPA would likely not change its general approach that assesses impacts using entity level
^he derived revenue per broiler estimate approximates the per-unit revenues received by a contract
grower in the broiler sector, as reported by Perry (1999).
D-6
-------
financial data. Aside from the estimated impacts to the broiler sector, EPA would likely not
change its proposal of economic achievabiliry if livestock revenues had been used to judge
impacts. For the broiler sector, EPA would likely argue that this approach does not appropriately
measure impacts and thus cannot be used to assess economic achievabiliry.
Table D-3. Baseline Revenues in Main Analysis and Sensitivity Analysis (Livestock Revenues Only)
. Sector
Beef
Dairy
Hog
Broiler
Layer
Turkey
Region
MW .
CE
MW
PA
MA
MW
MA
SO
MW/MA ;
MW/MA
Facility Size
Grouping
Large
Medium
Large
Medium
Large
Medium
Small
Large
Medium
Large
Medium
Large
Medium
Small
Large
Medium
Large
Medium
All
All
Medium
Main Analysis
Entity Revenue
per Head
$861.83
$534.68
$854.00
$501.99
$2,612.95
$2,498.11
$2,619.74
$2,567,29
$2,342.77
$83.60
$174.47
$228.99
$303.89
$606.13
$1.13
$1.47
$1.16
$1.42
$24.63
$20.08
$11.24
Sensitivity Analysis
Entity Revenue
per Head
$512.46
$328.65
$479.09
$339.85
$2,649.82
$2,249.62
$2,148.21:
$2,411.70
$2,166.11
$46.94
$101.69
$186.19
$192.93
• . $307.26
$0.16
$0.18
$0.17
$0.13
$16.98
$16.61
$6.47
Source: Data used for main analysis are USDA/ERS (1999a), derived on a per-animal basis by EPA. Data used for
sensitivity analysis are derived from the data in the main analysis, with USEPA assumptions (uses the USDA-
reported livestock portion of total farm revenue only and disregards revenue from other farm-related sources,
including crops). Table 4-1 defines regions and facility size.
D-7
-------
Table D-4. Number of CAFOs Affected Assuming Alternative Assumption (Livestock Revenues Oniy)
Sector
Fed Cattle
Veal
Heifer
Dairy
Hogs
(Zero CPT)
Hogs
(Partial CPT)
Broilers
(Zero CPT)
Broilers
(Partial CPT)
Layers - Wet
Layers - Dry
Turkeys
Total
(Zero CPT)
Total
(Partial CPT)
Total
(excl. broilers)
(Partial CPT)
No. of
CAFOs
4,070
210
1,050
7,140
14,370
14,370
14,140
14,140
'360
1,700
2,100
45,140
45,140
45,140
Scenario 4b— Sensitivity Analysis
Aff.
Mod.
Stress
(number)
2,140
210
630
5,770
9,950
12,960
0
0
360
1,700
1,990
22,750
25,760
25,760
1,450
0
250
670
1,400
1,410
0
0
0
0
10
3,780
3,790
3,790
490
0
170
700
3,020
0
14,140
14,140
0
0
100
18,610
15,600
1,450
Scenario 4b-Main Analysis
Aff.
Mod.
Stress
(number)
2,880
210
850
6,010
12,710
14,370
1,960
12,690
360
1,700
1,990
28,680
41,070
28,380
1,150
0
150
440
240
0
11,860
1,450
0
0
110
13,950
3,300
1,850
40
0
50
700
1,420
0
320
0
0
0
0
2,520
790
790
Source: USBPA. Numbers may not add due to rounding.
Evaluated for the BAT Option and Scenario 4b (Options/Scenarios are defined in Table 3-1). Category definitions
("Affordable," "Moderate" and "Stress") are provided in Table 4-13.
D-8
-------
D.1.3 Sales Test Sensitivity (Pre-tax Compliance Cost Assumption)
The sensitivity analysis described in this appendix section assesses a change in
methodology, not a change in baseline assumptions. As described in Section 4, EPA's main
analysis evaluates cost-to-sales ratios (among other criteria) using EPA's estimated compliance
costs for model facilities adjusted for expected tax savings (i.e., evaluating impacts using post-tax
costs). An alternative approach would be to evaluate cost impacts before taxes (i.e., using pre-tax
costs). Although EPA believes that pre-tax cqsts fail to reflect actual bottom-line impacts at a
facility, EPA conducts this sensitivity analysis to examine whether the results of the analysis
would differ substantially if pre-tax costs were used in the analysis.
Data inputs to this sensitivity analysis on an annual, pre-tax, per-head basis are not
reproduced here, but can be found in the rulemaking record (DCN 70601). For comparison, EPA
presents the post-tax costs in Appendix A. The equivalent after-tax cost is approximately 60 to
70 percent of the pre-tax cost.
Table D-5 compares the results of EPA's sensitivity analysis to the results of the main
analysis. Under a zero CPT assumption, the percentage of CAFOs expected to experience
financial stress increases from 5.5 percent to 13.9 percent of all CAFOs in the sensitivity analysis.
More CAFOs are estimated to experience financial stress in the fed cattle, heifer, hog, and broiler
sectors in the sensitivity analysis. Under a partial CPT assumption, an additional 270 CAFOs
(1,060 CAFOs versus 790 CAFOs) are estimated to experience financial stress compared to the
main analysis. As in the main analysis, no hog or broiler operations are expected to experience
financial stress under a partial CPT assumption. The percentage of operations experiencing
financial stress assuming partial CPT increases from 1.7 percent (main analysis) to 2.3 percent
(sensitivity analysis) of all CAFOs. Based on the results of this sensitivity analysis, EPA would
likely not change the proposal made in the main analysis that the proposed CAFO regulations
would be economically achievable.
D.1.4 Discounted Cash Flow Analysis (Lower Net Cash Income)
The sensitivity analysis described in this appendix section assesses EPA assumptions of the
baseline conditions as they affect the results of EPA's discounted cash flow analysis. EPA's main
analysis evaluates impacts using net cash farm income from the 1997 ARMS data provided by
USD A and derived by EPA on aper-animal basis using available data from ARMS that
correspond to the financial data (described in Section 4). These data are average and may not
reflect conditions at some operations that may generate net cash income below the national
average (e.g., due to higher-cost, less efficient production).
To evaluate the baseline assumptions on net cash income and to account for differences in
net returns among producers, EPA modifies the baseline data used in the main analysis by adding
5 percent to the USDA-reported variable costs and then re-calculating net cash income, assuming
D-9
-------
that gross farm revenues and fixed costs remain the same. A similar approach was used by DPRA
(1995), wherein published average data from USDA were used estimate net cash income for
animal feeding operations by characterizing vulnerable operations from the average by increasing
the estimated average totals cost per facility by 5 percent and favorable operations by decreasing
total cost per facility by 5 percent. Table D-6 presents a comparison of net cash income used in
the main analysis and the modified returns used in this sensitivity analysis (expressed on a per-
animal basis).
Table D-5. Number of CAFOs Affected Assuming Alternative Assumption (Pre-Tax Costs)
Sector
Fed Cattle
Veal
Heifer
Dairy '
Hogs
(Zero CPT)
Hogs
(Partial CPT)
Broilers
(Zero CPT)
Broilers
(Partial CPT)
Layers - Wet
Layers - Dry
Turkeys
Total
(Zero CPT)
Total
(Partial CPT)
No. of
CAFOs
4,070
210
1,050
7,140
14,370
14,370
14,140
14,140
360
1,700
2,100
45,140
45,140
Scenario 4b-Sensitivity Analysis
Aff.
Mod.
Stress
(number)
2,180
210
690
5,430
9,430
14,070
0
7,000
360
1,700
1,990
22,000
33,640
1,610
0
270
1,010
1,190
300
12,690
7,140
0
0
110
16,890
. 10,450
280
0
80
700
3,750
0
1,450
0
0
0
0
6,260
1,060
Scenario 4b-Main Analysis
Aff.
Mod.
Stress
(number)
2,880
210
850
6,010
12,710
14,370
1,960
12,690
360
1,700
1,990
28,680
41,070
1,150
0
150
440
240
0
11,860
1,450
0
0
110
13,950
3,300
40
0
50
700
1,420
0
320
0
0
0
0
2,520
790
Source: USEPA. Numbers may not add due to rounding.
Evaluated for the BAT Option and Scenario 4b (Options/Scenarios are defined in Table 3-1). Category definitions
("Affordable," "Moderate" and "Stress") are provided in Table 4-13.
D-10
-------
Table D-6. Baseline Net Cash Income in Main Analysis and Sensitivity Analysis (Lower Net Cash Income)
Sector
Beef
Dairy
Hog
Broiler
Layer
Turkey
Region
MW
CE
MW
PA
MA
MW
MA
SO
MW/MA
MW/MA
Facility Size
Grouping
Large
Medium
Large
Medium
Large
Medium
Small
Large
Medium
Large
Medium
Large
Medium
Small
Large
Medium
Large
Medium
All
All
Medium
Main Analysis Cash
Income/Head
$256.28
$79.37
$321.83
$80.71
$435.17
$443.71
$598.06
$401.74
$224.73
$31.16
$30.64
$46.51
$66.02
$118.64
$0.48
$0.57
$0.49
$0.57
$4.06
$1.77
$2.55
Sensitivity Analysis
Cash Income/Head
$232.28
$62.59
$299.88
$66.21
$336.00
$358.22
$518.48
$302.36
$129.55
$29.06
$24.79
$38.68
$56.33
$100.38
$0.46
$0.54
$0.47
$0.54
$3.22
$0.90
$2.19
Source: Data used for main analysis are USDA/ERS (1999a), derived on a per-animal basis by EPA. Data used for
sensitivity analysis are derived from the data in the main analysis, with USEPA assumptions (adjusts the USDA-
reported variable costs upward by 5 percent and re-calculates net cash income, assuming all other costs remain
constant). Table 4-1 defines regions and facility size.
D-ll
-------
Table D-7 compares the results of EPA's sensitivity analysis to the results of the main
analysis. Under a zero CPT assumption, the percentage of CAFOs experiencing financial stress
increases from 5.6 percent in the main analysis to 7.3 percent of all CAFOs in the sensitivity
analysis. More CAFOs are estimated to experience financial stress in the fed cattle and dairy
sectors in the sensitivity analysis. Under a partial CPT assumption, an additional 770 CAFOs
(1,560 CAFOs versus 790 CAFOs) are estimated to experience financial stress compared to the
main analysis. As in the main analysis, no hog or broiler operations are expected to experience
financial stress under a partial CPT assumption. The percentage of operations experiencing
financial stress assuming partial CPT increases from 1.7 percent (main analysis) to 3.4 percent
(sensitivity analysis) of all CAFOs. Based on the results of this sensitivity analysis, EPA would
likely not change the proposal made in the main analysis that the proposed CAFO regulations
would be economically achievable.
D.1.5 Debt-to-Asset Analysis (Higher Debt-to-Asset Levels)
The sensitivity analysis described in this appendix section assesses EPA assumptions of the
baseline conditions as they affect the results of EPA's financial ratio analysis. EPA's main
analysis evaluates impacts using debt-to-asset ratios from the 1997 ARMS data provided by
USDA. These data are average and may not reflect conditions at some operations that face
higher financial risk (e.g., due to higher debt levels relative to assets at some farms).
To evaluate the baseline assumptions and to account for differences hi debt-to-asset levels
among producers, EPA modifies the baseline data used in the main analysis by adjusting the
USDA-reported baseline values upward. For this analysis, EPA re-calculates these ratios by
reducing asset levels by a fixed amount without changing the reported debt levels. (Although the
adjustment is made on the basis of assets, the same result could have been achieved using
alternative debt assumptions to reflect higher debt levels at some operations.3)
Analyses of USDA's Census data by Letson and Gollehon (1996) and Gollehon and
Caswell (2000) show that larger livestock and poultry farms have less land devoted to non-
livestock purposes than smaller farms. To conduct a simple sensitivity test, EPA assumes that
CAFOs with less land may have higher debt due to lower asset levels.4 For this sensitivity
analysis, EPA adjusts the USDA-reported debt-to-asset ratios at Category 2 and Category 3
CAFOs.5 At Category 3 CAFOs among the larger-sized operations with more than 1,000 AU,
3For example, Stott (2000b) has noted that feedlots generally maintain'a higher level of debt than cow-calf
operations that are represented by the ARMS data used for this analysis to examine impacts to the beef sector.
4However, it is also possible that operations with less land have less debt on average (e.g., the additional
land at CAFOs with sufficient land for land application might be mortgaged).
5EPA's model CAFOs are divided into three land availability types: CAFOs with sufficient land for land
application of manures (Category 1), CAFOs with insufficient land for land application of manures (Category 2),
and CAFOs with no land for land application of manures (Category 3).
D-12
-------
Table D-7. Number of CAFOs Affected Assuming Alternative Assumption (Lower Net Cash Income)
Sector
Fed Cattle
Veal
Heifer
Dairy
Hogs
(Zero CPT)
Hogs
(Partial CPT)
Broilers
(Zero CPT)
Broilers
(Partial CPT)
Layers - Wet
-Layers - Dry
Turkeys
Total
(Zero CPT)
Total
(Partial CPT)
No. of
CAFOs "
4,070
210
1,050
7,140
14,370
14,370
14,140
14,140
360
1,700
. 2,100
45,140
45,140
Scenario 4b-Sensitivity Analysis
Aff.
Mod.
Stress
(number)
2,870
210
850
5,620
12,710
14,370
1,960
12,690
360
1,700
1,990
28,270
40,660
950
0
150
260
240
0
11,860
1,450
0
0
110
13,580
2,930
260
0
50
1,250
1,420
0
320
0
0
0
0
3,290
1,560
Scenario 4b-Main Analysis
Aff.
Mod.
Stress
(number)
2,880
210
850
6,010
12,710
14,370
1,960
12,690
360
1,700
1,990
28,680
41,070
1,150
0
150
440
240
0
11,860
1,450
0
0
110
13,950
3,300
40
0
50
700
1,420
0
320
0
0
0
0
2,520
790
Source: USEPA. Numbers may not add due to rounding.
Evaluated for the BAT Option and Scenario 4b (Options/Scenarios are defined in Table 3-1). Category definitions
("Affordable," "Moderate" and "Stress") are provided in Table 4-13.
D-13
-------
EPA adjusts asset levels downward by 15 percent at poultry operations and by 25 percent at other
livestock operations. No adjustments are made on the reported data for larger Category 2
CAFOs, which are assumed to represent the average land availability category for larger facilities.
Among medium-sized operations with between 300 and 1,000 AU, EPA adjusts asset levels
downward by 5 percent at poultry operations and by 10 percent at other livestock operations at
Category 2 CAFOs. At Category 3 medium-sized operations, EPA adjusts asset levels downward
by 10 percent at poultry operations and by 20 percent at other livestock operations. Variability of
assets is expected to be greater among the livestock sectors than among the poultry sectors
because of the greater land requirements in general for livestock production compared to poultry
production. No adjustments are made to data assumed for the smaller-sized operations with less
thanSOOAU. . ,
Table D-8 presents a comparison of the debt-to-asset levels used in the main analysis and
the modified revenues used in this sensitivity analysis. As shown, these higher debt-to asset ratios
result in all large hog model CAFOs (Category 3) for both the Mid-Atlantic and Midwest regions
to become identified as financially vulnerable in the baseline. These models would typically not be
analyzed as part of a regulatory impact analysis since these operations are already assumed to
close in the pre-regulatory baseline. For this sensitivity analysis, however, EPA continues to
include these operations in the impact analysis. The remaining CAFOs represented by all other
models are considered to be financially healthy in the baseline.
Table D-9 compares the results of EPA's sensitivity analysis to the results of the main
analysis. Under a zero CPT assumption, the percentage of operations experiencing financial stress
increases from 5.6 percent (main analysis) to 6.4 percent (sensitivity analysis) of all CAFOs.
More CAFOs are estimated to experience financial stress in the broiler sector only, resulting in an
additional 380 operations estimated to be vulnerable to closure as a result of compliance. Under a
partial cost passthrough assumption, no broilers or hogs are expected to experience financial
stress in either the main analysis or the sensitivity analysis. Under partial cost passthrough, the
results of this sensitivity analysis and the main analysis are identical. Based on the results of this
sensitivity analysis, EPA would likely not change the proposal made in the main analysis that the
proposed CAFO regulations would be economically achievable. (For comparison purposes, EPA
includes expected baseline closures in the hog sector (Category 3 large hog CAFOs). Excluding
these operations would result in fewer operations considered potential closures, as compared to
the main analysis, estimated at 810 fewer operations, as shown in Table D-9.)
D.2 MARKET MODEL SENSITIVITY ANALYSIS
To examine the stability of the results of EPA's market model results, EPA conducts
various sensitivity analyses that examine the results of EPA's analysis under differing baseline
value assumptions only. For this analysis, EPA examines alternative values for the baseline price
elasticity and also market price.
D-14
-------
Table D-8. Baseline Debt-to-Asset Ratios in Main Analysis and Sensitivity Analysis (Higher Debt-to-Assets)
Sector
Beef
Dairy
Hog
Broiler
Layer
Turkey
Region
MW
CE
MW
PA
MA
MW
MA
SO
MW/SO
MW/MA
Size (Proxy)
Large
Medium
Large
Medium
Large
Medium
Small
Large
Medium
Large
Medium
Large
Medium
Small
Large
Medium
Large
Medium'
All
All
Medium
Main Analysis
Debt-to- Assets/Farm
0.09
0.13
0.09
.0.17
0.26
0.23
0.20
0.24
0.23
0.31
0.13
0.40
0.25
0.17
0.30
0.21
0.26
0.19
0.11
0.15
0.23.
Sensitivity Analysis
Debt-to-Assets/Farm
Cat. 2
0.09
0.15
0.09
0.19
0.26
0.25
' • NA
0.24
0.26
0.31
0.14
0.40
0.28
NA
0.30
0.22
0.26
0.20
0.11
• • 0.16
0.23
Cat. 3
0.12
0.17
0.12
0.21
0.35
0.28
NA
0.32
0.29
0.41
0.16
0.54
031
NA
0.34
0.23
0.29
0.21
0.12
0.17
0.26
Source: Data used for main analysis are USDA/ERS \>.sssu.j. ^^.^. i»^~v» *«. ~~*M.~,*,J _.™.-/
USDA/ERS, 1999a, with USEPA assumptions (adjusts the USDA-reported debt-to-asset ratios upward for selected
model CAFOs). Table 4-1 defines regions and facility size. NA= Not Applicable.
D-15
-------
Table D-9. Number of CAFOs Affected Assuming Alternative Assumption (Higher Debt-to-Asset Ratios)
Sector
Fed Cattle
Veal
Heifer
Dairy
Hogs
(Zero CPT)
Hogs
(Partial CPT)
Broilers
(Zero CPT)
Broilers
(Partial CPT)
Layers - Wet
Layers - Dry
Turkeys
Total
(Zero CPT)
Total (excl. hog
pre-reg closures)
(Zero CPT)
Total
(Partial CPT)
No. of
CAFOs
4,070
210
1,050
7,140
14,370
14,370
14,140
14,149
360
1,700
2,100
45,140
45,140
45,140
Scenario 4b— Sensitivity Analysis
Aff.
Mod.
Stress
(number)
2,870
210
850
5,810
12,710
14,370
1,910
12,690
360
1,700
1,990
28,400
28,400
40,840
1,160
0
150
640
240
• 0
11,530
1,450
0
0
110
13,840
13,840
3,520
40
0
50
700
1,420
0
700
0
0
0
0
2,910
2,100
790
Scenario 4b-Main Analysis
Aff.
Mod.
Stress
(number)
2,880
210
850
6,010
12,710
14,370
1,960
12,690
360
1,700
1,990
28,680
28,680
41,070
1,150
0
150
440
240
0
11,860
1,450
0
0
110
13,950
13,950
3,300
40
0
50
700
1,420
0
320
0
0
0
0
2,520
1,710
790
Source: USEPA. Numbers may not add due to rounding.
Evaluated for the BAT Option and Scenario 4b (Options/Scenarios are defined in Table 3-1). Category definitions
("Affordable," "Moderate" and "Stress") are provided in Table 4-13.
D-16
-------
EPA has reviewed a wide range of price elasticity values for livestock and poultry
products that are reported in the academic literature (see Appendix C). EPA realizes that the
elasticity values selected for the market model, while carefully considered, are not the only
possible set of realistic values. Therefore, EPA conducts a sensitivity analysis with several
alternative sets of elasticities to assess whether the outcome would have been substantially
different had different input elasticities been used. The analysis indicates that although the results
would have differed, they would have been within the same order of magnitude as those generated
iri the main analysis.
EPA also investigates the effect that the baseline prices have on the analysis. The baseline
prices that EPA uses for this analysis are 1997 averages, which may not reflect regional
differences or different stages in the animal marketing cycle. (The attachment at the end of this
appendix provides a summary of the key market data, including prices, for these sectors from
1992 to 2000 where available.) To test the impact of alternate price assumptions, different prices
are substituted into the model. The results indicate that the price assumptions in the main analysis
have little effect on predicted model outcomes.
D.2.1 Price Elasticities
For the main analysis EPA refers to elasticities reported in the academic literature and uses
a "selected value" that generally reflects the approximate mid-range value or a value chosen based
on best professional judgement (Table 4-14). For this sensitivity analysis, EPA investigates eight
different alternative combinations of elasticities to test the sensitivity of the results to different
model input. The eight combinations include the extremes of supply and demand elasticities
found in the literature. Four combinations show the effect of changing each supply and demand
elasticity to its most elastic or inelastic value while all other elasticities remain at their selected
values. EPA also constructs minimum and maximum partial CPT estimates by applying an
approach described in Section 4.2.6. This approach allows EPA to evaluate the results of the
main analysis using combinations of supply and demand elasticities that minimize or maximize
GPT within the market model.
EPA also investigates the effect of varying trade elasticity assumptions. The trade
combinations test the influence of import and export assumptions on the results. The eight
combinations of elasticities are shown in Table D-10. The actual elasticities used can be found in
Table 4-14 (selected values) and Table B-l (minimum and maximum values).
Table D-l 1 shows the range of farm product price results derived from changing the
elasticity assumptions to evaluate the compliance costs of the proposed BAT Option in the model.
The main analysis value shows the predicted change in farm product price under a post-regulatory
scenario using the selected elasticity values presented in Table 4-14. The range of results is quite
small in all sectors. The sensitivity analysis results are expressed as a range: the "low" value
shown is the lowest farm product price and the "high" value shown is the highest farm product
D-17
-------
price derived from any of the sets of alternative elasticity assumptions. Lower predicted prices
result from the rninimum CPT assumptions, while most of the high-end predicted prices result
from the maximum CPT assumptions. Since the predicted changes in farm product prices are
small, the corresponding changes in retail product prices are even less and are not presented here.
Table D-12 shows the changes in total employment that result from alternative elasticity
assumptions. The main analysis value corresponds to the market model results presented in
Section 5.4. Although the ranges shown for the sensitivity analysis may appear sizable, they are
Table D-10. Elasticity Sensitivity Test Sets
Set Name
Inelastic Supply
Elastic Supply
Inelastic Trade
Elastic Trade
Inelastic Demand
Elastic Demand
Max CPT
Min CPT
Elasticity of Supply
Minimum from Table B-l
4
Selected
Selected
Selected
Selected
4
Minimum from Table B-l
Elasticity of Demand
Selected
Selected
Selected
Selected
Maximum from Table B-l
Minimum from Table B-l
Maximum from Table B-l
Minimum from Table B-l
Trade Elasticities
Selected
Selected
0
4
Selected
Selected
Selected
Selected
Source: USEPA. CPT= Cost passthrough. "Selected" indicates the selected elasticity value shown in Table 4-14.
Alternative values (minimums and maximums) can be seen in Table B-l .
Table D-ll. Range of Postregulatory Farm Product Price Results with Different Elasticity Assumptions
Sector
Beef
Dairy
Hog
Broiler
Layer
Turkey
Units
S/cwt
S/cwt
$/cwt
cents/lb.
cents/doz.
cents/lb.
Farm Product Price
Main Analysis Value
66.30
13.44
54.89
37.19
69.93
40.22
Low Value "
66.11
13.39
54.32
37.05
69.81
40.21
High Value "'
66.37
13.45
55.13
37.57
69.95
40.34
Source: USEPA.
31 All low values are generated by assuming minimum CPT.
'"'All high values shown are generated assuming maximum CPT except for broilers and layers, in which the elastic
import/export assumption generated the high value shown.
D-18
-------
not large in terms of the percentages of sector employment they represent. Also, the width of the
ranges shown varies with the magnitude of compliance costs estimated for each sector. EPA
estimates that costs to the hog sector account for the majority of the total incremental costs of the
rule and therefore show the widest range of predicted employment loss. EPA estimates that the
poultry sector will incur lower costs and therefore fewer employment losses. The losses
estimated in the main analysis are more similar to the high-end value of the estimated ranges
shown for most sectors (Table D-12).
Based on the results of this sensitivity analysis, EPA would likely not change the proposal
made in the main analysis that the proposed CAFO regulations would be economically achievable.
Table D-12. Range of Total Employment Change Results with Different Elasticity Assumptions
Sector
Beef
Dairy
Hog
Broiler
Layer
Turkey
Main Analysis Value
Sensitivity Analysis Value
Low Value "
(FTE losses)
4,599
3,200
6,376
1,865
202
373
. 1,876
2,753
1,388
1,404
184
357
High Value b/
5,397
3,816
7,332
2,265
292
415
Source: USEPA.
31 Low values for beef and dairy are generated by assuming inelastic supply. Low values for layer and turkey are
generated by assuming inelastic demand. Minimum CPT and elastic imports and exports generate hog and broiler
low values, respectively.
b/All high values shown are generated by assuming elastic demand except for hog and turkey in.which the elastic
supply assumption generated the high value shown.
D.2.2 Prices
In addition to the price elasticities, the market model results also depend on baseline price
and quantities as input values to the model. Table D-13 shows the effects of raising or lowering
the baseline farm and retail prices used in the main analysis by 50 percent. The main analysis
value shows the predicted change in farm and retail prices under a post-regulatory scenario when
the baseline values assume the selected levels presented in Table 4-16 (baseline prices). The farm
and retail price range shown in Table D-13 indicate the range of predicted prices in the sensitivity
analysis when the baseline prices are raised or lowered by 50 percent.
D-19
-------
As shown in Table D-13, hog farm prices show the widest range of change in farm
product prices, but even the ranges for this sector are fairly narrow. This analysis demonstrates
that even if baseline prices are assumed to vary +/-50 percent, this would not substantially change
the results of EPA's market model.
Table D-13. Range of Price Changes with Different Baseline Price Assumptions
Sector
Baseline Price
Main Analysis
Units
Main Analysis
Change
Sensitivity Analysis Value
Farm Price
Range
Retail Price
Range
Change in Farm Product Prices
Beef
Dairy
Hog
Broiler
Layer
Turkey
66.09
13.38
54.30
37
40
70
$/cwt
$/cwt
$/cwt
cents/lb.
cents/dozen
cents/lb.
0.21
0.06
0.59
0.19
0.13
0.12
0.19-0.24
0.06-0.07
0.52-0.68
0.16-0.22
0.13-0.14
0.10-0.16
0.17-0.23
0.05-0.06
0.46-0.65
0.14-6.21
0.12-0.13
0.08-0.15
Change in Retail Prices
Beef
Dairy
Hog
Broiler
Layer
Turkey
2.80
145.5
2.45
151
106
105
$/lb.
Index
$/lb.
cents/lb.
cents/dozen
cents/lb.
0.00
0.61
0.01
0.19
0.13
0.12
0.00-0.00
0.56-0.66
0.01-0.01
0.16-0.22
0.13-0.14
0.10-0.16
0.00-0.00
0.53-0.64
0.01-0.01
0.14-0.21
0.12-0.13
0.08-0.15
Source: .USEPA. Low and high values are created assuming that baseline prices are 50 percent lower or higher
than the baseline prices used in the main analysis.
D-20
-------
o
o
OS
OS
T— 1
k.
o
1
£"
n
y.
"S
»
"«
B
Attachment 1. Key An
1
2
3
6?
s
o\
1.
oo;
£
T-1
*
r-,
VO"
c\
Os
T-tA
in
o\
Os
•<*•
OS
r «
f*}
T"(
a
„ ^*
<
'
-
OS
en
CN
60
g
.
60
OS
&O
^
^*%
a
r-
c-i
60
s
en
oo
cs
60
O
.
O
si
g
3
E
Beef Feedldt Sector
2
vri
t/3
|
O)
S
0
SO
>r>
O
^r^
so
O
r-
06
60
o
oo
SO
o •
r-
SO
O
so
CNI
t-.
o
en
IBeef cattle farm-level price
($/cwt, all beef) b
oo
6O
7,
7
^
S
6O
eM
en
so
en
cs
6O
u-i
„
e>l
OS
60
IBeef cattle feed price
($/ton, 32-36% protein) c
en
en
.^
cs
0
o
r-__
CN
o
en
«
en
cQ
oo
en
a
"V
e»
OI
cN
so
00
en
S
5
O^
en
cs
SO
oo
en"
IBeef cattle production
(million Ibs. carcass) d
so
*r\
en
|
O
SO
^^
SO
en
SO
oo
• en'
so
•en
r—
r-
en
OS
en
so^
en
«
en
SO
en^
en
en
en
r-
• oo
C"f
en
in
1 Commercial cattle slaughh
(1,000 head).
S
W
£
,
•~r
f't
oo
OS
eN
o
60
55
^*
S.
oo
oo
en
en
en
OS
oo
en
oo
O
en
6O
SO
en
6O
0
3.
1 Cow-calf operation total vi
production ($/cow) f
^
oo
evl
"-^
z
,
17-
f-*
-
OS
6O
SO-
^.
o
e
fn
E
en.
t—
oo
r~
^—s
so
00
es
oo
V)
t
en
60
o
S
o
o
c
. J3
1 Cow-calf operation net cas
($/cow) g
Z
^
c—
*s~*
'--'
OS
en
en
•^
en
i
en
en
OS
OJ
oo
3
T3
2
CA
8
1 Choice beef farm-retail pri
(cents, 1982-84=100) h
|
-
£
en
SO-
O
u-i
OS
OS
oo
ts
0
C5
o
os
o
o
SO
8
0\
< — r
^
OS
o
• _
1 Producer price index
(beef and veal, 1982=100)
Z
o
5?
es
OS
en
so
en
1 '
oq
so
en
en
o\
en
• p
so
en
-:
en
en
c-1
en
—
o
o
Consumer price index
(beef and veal, 1982-84=1
Q
-------
|
1
"8
0>
CO
>>
??
c
Attachment 1. Key Annual Si
«
2
*
bs
C9
OS
CO
T-*
VO
to
T-*
*•<
to
o\
0,
s
(Veal Sector
o
00
69
$
VO
OO
69
O
CO
CO
69
O
OS,
CO
69
CO
69
O
69
O
oo
69
S
69
C\
CO
69
Farm-level price ($/cwt calves) b
CO
01
oi
a
04
vo
CN
CO
CO
CO
CO
OS
CO
CO
Os
CO
CN
o
CO
•a
J Veal production (million Ibs. veal)
5
2
••3-
CO
Tt-
CO
VI
VI
5
o
CO
oo
VO
25
g
1 Calf slaughter (1,000 head) e
z
I
CO
VO
o
VI
cs
- 'co •-
••'• e4 '•
o •
o
o
c5
o
vq
CO
O
OS
i
(Producer price index
(beef and veal, 1982=100) ,
z
0
CN
OS
CO
Vl
VO
CO
oo
vo
CO
Vi
CO
CO
o
vd
CO
CO
CO
oi
CO
To
(Consumer price index (beef and ve
1982-84=100) j
1 Dairy Farm Sector
VO
VI
CO
69
Vi
Vi
oi
69
OO
CO
69
vi
69
CO
CO
69
s
69
oi
69
r-
01
69
O
CO
oi
69
O
CO
69
•a
c
IA11 milk price (delivered to plants
dealers, $/cwt)b
vo
0
13
2
2
g
69
CN
vo
CO
69
CO
CO
CO
69
£
69
s
CO
VI
CO
CN
69
O\
B
Dairy feed price
($/ton, 32% protein cone) c
01
VO
VI
V)
vi
vo
C-
oi
vo
5
»— t
vo
v>
o
•<*
VI
CO
VI
VI
vq
CO
VI
vq
VI
o
VI
IMilk production
(billion Ibs., milkfat basis)b
OS
0
69
2
•2
CO
vd
69
r-
oo
69
f-
69
S
69
O
69
s
69
CO
69
C
Dairy farm total value of productic
($/cwt) f
oi
69
%•
2
3.
69
VO
&
69-.
OS
oi
69
^
69
O
vo
69
$
69
69
CO
1 Dairy farm net cash income ($/cwt
*
2
K
01
vq
vi
oo
CO
oo
CO
r-
ro
t-
vo
vo
oi
vo
vq
CO
v»
•e
s
Dairy products farm-retail price spi
(cents, 1982-84=100) h
2
$ .
V)
CO
CO
CO
CO:
OI
o
CO
C-;
V)
OS
CO
5
Producer price index
(dairy products, 1982=100),
z
0
Vl
vo
OS .
Vi
CO
o
VI
v>
v>
oi
oo
oi
CO
C-;
CO
OS
OI
v>
CO
01
1
•g
Consumer price index (dairy produ
incl. butter, 1982-84=100) j
CO
^
Q
-------
o
0
o
P*J
OS
OS
1— 1
e
cj
V
^
A
V
•*-*
•*-»
CO
«
Attachment 1. KeyAnm
&A
a
1
H
0
o
OS
•••fi^
^
i
t
C"*
"t.
;
OS
OS
1-4
1/1-
Os
r"
3
OS
,.-
O\
r
? r
o\
~ s
^
x
Hog Sector ; ; v
\o
CN)
^j.
W
<
o
CO
C2
en
6O
O
o
cn
6O
o
cn
O
01
01
6O
O
SO
o
0
OS
vi
6O
O
oi
6O
Barrow & gilt price ($/cwt,) t
OS
r-
01
en
6O
<
so
en
en
01
6O
en
.OS
CS
6O
CS
SO
01
r-
os
o
60
vi
0
cn
CO
. vi
cn
60
f-
r-
oi
en
6O
Sow price ($/cwt) b
VO
g
2
r-
cn
60
cn
60
O
OS
OJ
60
0
£
o
VI
o
60
O
CS
OS
en
o
01
•*!•
60
o
SO
6O
•o
01
O
"5
|"
a:
C
C
.e
V.
01
60
'|
|
r-
en
6O
os
oo
en
_
en
01
r-
cn
s
cn
o
a
IHog feed price
($/ton, 38-42% protein cone.]
ol
r~"
o
cn
so
,00
oo
0
en
Os
o
os"
£
*x
£
£
oo.
OS
r:
oo
oo
o
r-
' ^
en
*— '
Ipork production
(million Ibs. pork carcass) d
SO
u-r
<
K
T-^
O
OS
CN
O
g
^N
OS
g;
of
§
OS
OS
CS
Cs
SO
0
en
OS
00
oo
-*"
CS
-2
i^.
C3
ti
C
•=!
MO
X 0
VI
Vj
6O
|
|
en
OS
en
VI
OJ
cn -
£
o
SO
60
r-
cn
06
6O
o
CO
60
cn
€O
Cs
06
6O
C
o
o •
3
Hog farm total value of prod
($/cwt gain) f
cn
m
&^
^
^
V)
oo
v>
60,
O
r-
T— I
6O
cn
o
6O
Vt
en
oo
en
cn
6O
S
0
€^
V,
6O
'3
on
§
IHog farm net cash income ($
e
*jr
£
00
T— «
VI
oo
t-;
OS
''^
cn
SO
cn
06
Ol
vi
en
Ol
01
0
cn
IFork farm-retail price spread
(cents, 1982-84=100) h
z
£
F-
vi
Cs
vq
cs
en
ol
OS
1—1 .
o
en
o
s
cs
00
Cs
0s
o
II
01
oo
Cs
Producer price index (pork, ;
^
o
OS
cs
*-^
Vi
CO
OS
V)
01
CO
*"—
oo
en
Cs
cn
en
t-;
en
oo
B
Consumer price index
(pork, 1982-84=100) j
-------
<=>
o
o
1
0\
OS
o
a
CJ
00
>-.
CO
~
in
CO
Attachment 1. Key Annun
1*
451
fch
0
§5
Os
oo
2*
i^
c^
vo
»-1
ir>
§i
r-(
o\
CO
T-l
i
TK
1,
Broiler Sector
oo
oo
vi
CO
^
2
*
in
oo
c\
CO
<0
OS
VO
CO
VO
oo
CO
VO
vq
CO
cs
«3-
CO
CO
CO
>0
00
CO
JS
1 Broiler price received (cents/lb.
s
CO
69
^
Z
g
69
O
OO
CM
69
0
P
^_«
69
O
£
•o
CO
OS
CO
o
vo
CO
69
o
CO
69
o
-*
u
1 Broiler grower feed price ($/ton
O
CO
vo
CM
u->
5?_
CO
00
VO
Cs
a
p-
co
oo
t
•",
p-
cs
vo
vo
cs
8
CO
cs1"
{C
~^
o
CO
cs
CM
CO
cs
CM
c
.2
1 Broiler and chicken meat produ
(million Ibs.) d
0 CS
i—i
r-
<
&
oo
CM
oo
oo o
°X
vo"
0
oo
CO CS
co"
, 0
CS
v> oo
oo
OS
CS
V0_
oo
o"
CO
"T,
OS U->
Os
CO
o
Os
vo
oo
VO
^
CO
CO
g
VO
VO
o
*s*
Chicken slaughter (thousand he
CO
cs
^
"2
CM
CO
VO
CS
VO
CO
vo
CM
CS
cs
•*
VO
cs
p-
Broiler farm costs (cents/lb) k
cs
CO
^
£
*
vo
CO
^.
r-
oo
vd
CO
06
00
o
CO
es
^^
1 Broiler farm net returns (cents/1
Z
^
^
p-
cs
o
cs
OS
CS
OS
(
%
i-H
-
VO
cs
oo
p-
p-^
p-
VO
CS
p-
cs
vd
vo
o
CO
VO
1 Poultry farm-retail price spread
(cents, 1982-84=100) h
£
^
^
0
p-
I— 1
„.
p~ -
oq
Os -
CO
oo
^^
p~
^^
o
cs
o
1 Producer price index
(processed poultry, 1982=100),
|-.
0
oo
VI
Os
ln_(
p~
1 — 1
VO
vo
cs
•n
i
-
T^-
OS
vd
CO
^
CO
1 Consumer price index
(poultry, 1982-84=100) j
. . .
Turkey Sector ;;f ; :
OS
OS
OS
CO
^
z
*
oo
oo
CO
•o
o
TT
oo
CO
oo
o
5
^
o
cs
Cs
06
CO
CO
vq
CO
^^
1 Turkey prices received (cents/I
cs
CS
vd
cs
ff,
^
Z
vo
oo
CO
OS
o
vi
69
0
^^
o
OO
wi
VD
69
O
O
CO
o
VI
69
0
CO
CO
^^
&9
0
CO
u
1 Turkey grower feed price ($/ton
o
CO
>o
cs
CO
co^
0
CO
o
CS
5,
0
"O
OS
vo
o_
p-
^v
oo
OS
p-
^i"
^
p-^
Turkey production (million Ibs.
carcass) d
1^
Os
p-
CS
^
Z
o
s
cs
CO
o
CO
«S
O
cs
o
Cs
CM
O
CO
OS
CS
CS
00
cs
cs
^is
cs
cs
VO
CS
^.
oo
CS
1 Turkey slaughter (-1,000 head) c
OS
CO
^
Z
p-
cs
CO
p-
VO
CO
9
Os:
5
CO
*r\
vq
CO
^
CO
vd
CO
CS
CO
Turkey farm costs (cents/lb) k
2
^
z
*
cs
—
0
00
•7
00
1
s
CO
Os
. cs
oo
o
1 Turkey farm returns (cents/lb) |
Q
-------
o
1/3
i»
^™
Attachment 1. Key Annual S
1
g
JtTn
o
e
oC
Ov
^?
£ *"
I
-1-1
1
\0
s
v>
^
^.
s
to'
a
,
s
•rt
-
•z
$
r--*
i
CM
o\
CM
o\
06
o\
vq
oo
g
vo
f2
CM
vo
VO
0
VO
1 Poultry farm-retail price spread
(cents, 1982-84=100) h
~z.
<
p..
^
r~-
o
CM
^
oo
CO
00
^
*""'
p
o
1 Producer price index
(processed poultry, 1982=100) ,
Z .
o
00
»o
5
^-H
^ ,
r^
AO
^-i
CS
\n
CO
2..
o\
VO
CO
^^
en
Consumer price index
(poultry, 1982-84=100).
-
Layer; Sector; :'.;;:'•;.•- '-.•-, ," :
CO
^
&
£
VO
10
CO
f-, .
oo
CO
CO
vo
VO
co
o
o •
.CM
' Os
CO
o
p
u-i
J3 j=
!Egg price' (cents/dozen eggs, ta
eggs excluding hatching eggs)
CM
CM
^
^
g
^
CM
CM
CM
OO
CM
10
OS
•O
1—4
CM
O
CM
CO
0
<=/*
Laying feed price ($/ton) c
o
CM
VO
o
00
r-
o\
VD
oo
vo
vo"
CO
vo"
»— 4
>o
co
VO
VD
T— <
CM
vo
00
r-
1—4
vo
vo
o
o
VO
o
Egg production (million dozen
00
^
' *
2
CM
•^-
vq
1—4
10
t-;
c-^
IO
p
£
o
vd
CM
vd
JaJ
Layer farm costs (cents/dozen)
5
;< '
f-i
%
^
r-
r— 4
VO
vd
CO
CM
s
CM
!— *
t
1 Layer farm net returns
(cents/dozen) i
g
<.
f't
CO
CM
CM
•O
CM
CM
CM
p
co
CM
OS
CM
OS
VO
oo
.1—4
CO
VO
1 — 4
'
Eggs farm-retail price spread
(cents, 1982-84=100) „
g
<
r^ i
£
o\
oo
VO
2
p
1— <4
1—4
r-~
0
CO
r-4
O
oo
0
s
o?
00
60
a?
Producer price index (chicken
1982=100) ,
g.
p
vd
CM
oo
CM
1—4
^.
IO
CO
1—4
0
*— «
S
T— 4
1O
o
CM
co
ir-4
1—4
I""*
CO
00
1—4
t
CM
oo
1 Consumer price index (eggs, 1
84=100) ,
-S
t-t
PH
II
AH
CO
NA = Not available. F = Foreca
Sources:
CN
-------
\o
os
cv
S
o\
52
.0
•a
to
V
_o
a.
CO
2
Mi
ro
CO
3
Ov
C\
cs by Sector, 1
Key Annual Statist!
•
»H
«*••
CJ
ui
•S u
U p
S 3
•< co
ious years.
)_«
d
&
CO
^^
ca
g
<
. Agricultural Prices
CO
CO
$
<^c
^0
«a
»
|
&
t--
o\
c\
V— 1
CO
U
u
'C
. 1998. Agricultural P
CO
CO
g
Q
s
i U
s
*— t
•+-*
CO
1
o\
o\
1
.1
T3
S
t
•3
CO
.0
b4
PL,
d1
'a,
CO
C
O
W
£•
P
•a
li
co -g
So
O\ 13
1 veal slaughte
ance- Situatior
rbook. Beef, hog, am
Farm Sector Perform
000
Sa
eu _^r
^
T3 t/5
0 g
u £
tistical Indicators. Farm-i
C3
00
5>
fe4
various years.
Agricultural Outlook,
eg
P^
«
£
CO
! .=
CO
-OJ
•3
o
0
U
1
)ducer Price In
or Statistics Data. Pr<
•S
_i
CM
0
•1
3
1 Urban Consumers.
Index-Al
ansumer Price
ior Statistics Data. C(
"CO
hJ
•g
CO
2
«
. Poultry Yearbook
CO
CO
S
O
B
, ji
racting Farm Costs from ]
•s
CO
>>
^i
13
turns calculate
and Turkey Farm Re
I
j§
2
_
S
S
X",
ections t(
I Baseline Proj
:s: USDA Agricultura
forecast
H-t
O
o
cs
E
vo
CS
-------
APPENDIX E
COST-EFFECTIVENESS ANALYSIS
As part of the process of setting effluent limitations guidelines and. developing standards,
EPA uses cost-effectiveness calculations to compare the efficiencies of regulatory options for
removing priority and non-conventional pollutants.1 Although not required by the Clean Water
Act, a cost-effectiveness (C-E) analysis offers a useful metric to compare the efficiency of
alternative regulatory options in removing pollutants and to compare the proposed technology
option to other regulatory alternatives that were considered by EPA.
For the purpose of this regulatory analysis, EPA defines cost-effectiveness as the
incremental annualized cost of a technology option per incremental pound of pollutant removed
annually by that option.2 The analyses presented in this section include a standard cost-
effectiveness (C-E) analysis, based on the approach EPA has historically used for developing an
effluent guideline for toxic pollutants, but also expands on this approach to include an analysis of
the cost-effectiveness of removing nutrients and sediments. This expanded approach is necessary
to evaluate the broad range of pollutants in animal manure and wastewater, which include metallic
compounds, nutrients, total suspended solids, and pathogens. Of these pollutants, EPA's
standard C-E analysis is suitable for analyzing only the removal of toxic pollutants. EPA's
standard C-E analysis does not adequately address removals of nutrients, total suspended solids,
and pathogens. To account for the estimated removals of nutrients and sediments under the
proposed CAFO regulations in the analysis, the Agency has developed an alternative approach to
evaluate the pollutant removal effectiveness relative to cost. At this time, EPA has not developed
an approach that would allow a similar assessment of pathogen removals. Estimates of pathogen
reductions in manure and wastewater discharged from CAFOs under the proposed regulations are
provided in the Development Document (USEPA, 2000a).
The organization of this section is as follows. Section E.I provides an introduction and
describes the types and concentrations of pollutants found in animal manure and wastewater.
Section E.2 summarizes EPA's estimated baseline loadings and removals of metals, nutrients,
total suspended solids and pathogens from feedlot and land application areas, for selected
!A list of priority ("toxic") and conventional pollutants are defined at 40 CFR Part 401. There are more than
120 priority pollutants, including metals, pesticides, and organic and inorganic compounds. Conventional
pollutants include biological oxygen demand (BOD), total suspended solids (TSS), pH, fecal coliform, and oil and
grease. Non-conventional pollutants comprise all other pollutants, including nutrients (i.e., they do not include
conventional and priority pollutants).
2EPA defined cost-effectiveness similarly for Phase II of the Storm Water rule (USEPA, 1999f) and examined
the incremental annualized cost of each pollution control option to the incremental pound of TSS removed
annually.
E-l
-------
regulatory alternatives. Section E.3 presents a standard C-E analysis and focuses on a subset of
metallic compounds that are found in animal manure and wastewater. Section E.4 then presents
an analysis of the-cost-effectiveness of loadings reductions of nutrients and total suspended solids.
For this rule, EPA estimates the expected percentage reductions in-pathogens from agricultural .
runoff but does not compare these removals to costs of the regulatory controls. Section E.5
concludes the section by presenting EPA's assessment of the overall effectiveness of the proposed
combination of options to remove the pollutants of concern relative to the effectiveness of other
options considered in these analyses.
More detail on the environmental damages associated with livestock and poultry
operations and the pollutants in animal manure is provided in the Environmental Assessment
(USEPA, 2000b) and the Benefits Analysis (USEPA^ 2000d). Additional information on EPA's
estimated loadings and removals under post-compliance conditions is provided hi the
Development Document (USEPA, 2000a) and also in the Benefits Analysis (USEPA, 2000d).
E.1 POLLUTANTS OF CONCERN
E.I.I Introduction
Manure and wastewater from animal feeding operations have the potential to contribute
pollutants such as nutrients (e.g., nitrogen and phosphorus), organic matter, sediments,
pathogens, metals and metallic compounds, hormones, antibiotics, and ammonia to the
environment (USEPA, 2000b; USDA and USEPA, 1999). Additional information on the
pollutants ha animal manure and on water quality impairment and risks associated with manure
discharge and runoff is provided in the Section V of the preamble and in the Environmental
Assessment (USEPA, 2000b). .
National and local studies confirm the presence of manure pollutants in U.S. waters.
EPA's 1998 National Water Quality Inventory (USEPA, 2000h), prepared under section 305(b)
' of the Clean Water Act, presents information on impaired water bodies based on reports from the
states. Agricultural operations—including animal feeding operations—are considered a
significant source of water pollution in the United States (USEPA, 2000h). As shown in
Table E-l, agriculture3 is the leading contributor to identified water quality impairments in the
nation's rivers and lakes and the fifth leading contributor to identified water quality impairments in
the nation's estuaries (USEPA, 2000h).
Table E-l also lists the leading pollutants that impair surface water quality hi the United
States, as identified hi the National Water Quality Inventory. Livestock and poultry operations
are a potential source of all of these, but are most commonly associated with nutrients, pathogens,
Includes crop production, pastures, rangeland, feedlots, animal holding areas, other animal feeding
operations, pasture and range grazing, concentrated and confined animal feeding operations, and aquaculture.
E-2
-------
Table E-l. Leading Sources and Pollutants of Water Quality Impairment in the United States, 1998
Rank
Rivers & Streams
Lakes, Ponds, & Reservoirs
Estuaries
Sources ^
1
2
3
4
5
Agriculture (59%)
Hydromodification (20%)
Urban Runoff/Storm Sewers
(11%)
Municipal Point Sources (10%)
Resource Extraction (9%)
Agriculture (31%)
Hydromodification (15%)
Urban Runoff/Storm Sewers
(12%)
Municipal Point Sources (1 1%)
Atmospheric Deposition (8%)
Municipal Point Sources
(28%)
Urban RunofPStorm Sewers
(28%)
Atmospheric Deposition
(23%)
Industrial Discharges (15%)
Agriculture (15%)
Pollutants b/
1
2
3
4
5
Siltation (38%)
Pathogens (36%)
Nutrients (29%)
Oxygen-Depleting Substances
(23%)
Metals (21%)
Nutrients (44%)
Metals (27%)
Siltation (15%)
Oxygen-Depleting Substances
(14%)
Suspended Solids (1.0%)
Pathogens (47%)
Oxygen-Depleting Substances
(42%)
Metals (27%)
Nutrients (23%)
Thermal Modifications (18%)
Source: USEPA, 2000h. Figure totals exceed 100 percent because water bodies may be impaired by more than one
source or pollutant.
"'Fraction of impairment attributed to each source is shown in parentheses. For example, agriculture is listed as a
source of impairment in 59 percent of impaired river miles. The portion of "agricultural" impairment attributable
to animal waste (as compared to crop production, pasture grazing, range grazing, and aquaculture) is not specified.
b/Percent impairment attributed to each pollutant is shown in parentheses. For example, siltation is listed as a
cause of impairment in 51 percent of impaired river miles.
oxygen-depleting substances, and solids (siltation). Animal operations are also a potential source
of other leading causes of water quality impairment, such as metals and pesticides, and can
contribute to the growth of noxious aquatic plants due to the discharge of excess nutrients.
Animal operations may also contribute loadings of priority toxic organic chemicals and oil and
grease, but most likely to a lesser extent than they contribute to loadings of other leading
pollutants.
Table E-2 presents additional summary statistics from the 1998 National Water Quality
Inventory. These figures indicate that agriculture contributes to the impairment of at least
170,000 river miles, 2.4 million lake acres, and almost 2,000 estuarine square miles. The portion
of impairment attributable to animal agriculture nationwide is unknown, since not all states and
E-3
-------
Table E-2. Summary of Statistics from the National U.S. Water Quality Impairment Survey, 1998
Total Quantity in U.S.
Rivers
3,662,255 miles
Lakes, Ponds, and
Reservoirs
41. 6 million acres
Estuaries
90,465 square miles
Waters Assessed
23% of total
(840,402 miles)
42% of total
(17.4 million
acres)
32% of total
(28,687 sq.
miles)
Quantity Impaired by All
Sources
35% of assessed
(291,263 miles)
45% of assessed
(7.9 million acres)
44% of assessed
(12,482 square miles)
Quantity Impaired by
Agriculture ^
59% of impaired
(170,750 miles)
31% of impaired
(2,417,801 acres)
15% of impaired
(1,827 square miles)
Source: USEPA, 2000h.
"'AFOs are a subset of the agriculture category. Summaries of impairment by non-agricultural sources are not
presented here.
tribes identified sources at the more specific levels. However, 28 state and tribes did identify
specific agricultural activities contributing to water quality impacts on rivers and streams, and 16
states and tribes identified specific agricultural activities contributing to water quality impacts on
lakes, ponds, and reservoirs. (In the 1996 Inventory, the following states reported impairment
due to intensive animal operations: Hawaii, Illinois, Kansas, Louisiana, Michigan, Minnesota,
Mississippi, Montana, Nebraska, Ohio, Oklahoma, Rhode Island, South Carolina, Tennessee,
Virginia, West Virginia, Wisconsin, and Wyoming. Other states reporting agricultural sources of
impairment include: Arizona, Nevada, New Hampshire, New Mexico, and North Dakota.)
For rivers and streams, estimates from 38 states indicate that 16 percent of the total
reported agricultural impairment is from animal feeding operations (including feedlots, animal
holding areas, and other animal operations), and 17 percent of the agricultural impairment is from
both range and pasture grazing. For lakes, ponds, and reservoirs, estimates from 16 states
indicate that 4 percent of the total reported agricultural impairment is from animal feeding
operations (including feedlots, animal holding areas, and other animal operations), and 39 percent
of the agricultural impairment is from both range and pasture grazing. Impairment specifically
due to land application of manure was not reported.
E.1.2 Pollutant Concentrations in Animal Manure and Wastewater
Table E-3 lists the reported amounts of macro- and micro-nutrients in livestock arid
poultry waste, along with documented levels of other inorganic and metallic constituents. As
shown, the American Society of Agricultural Engineers (ASAE) reports that the constituents
present in livestock and poultry manure include: boron (B), cadmium (Cd), calcium (Ca), chlorine
E-4
-------
Table E-3. Nutrients, Metals, and Pathogens in Livestock and Poultry Manures
Constituent
Mass of animal
Manure (wet basis)
Urine
Density (Ibs/ft3)
Total Solids .
Volatile Solids
BOD (5-day)
COD (5-day)
pH
Nitrogen (Total Kjeldahl)
Nitrogen (Ammonia)
Phosphorus (Total)
Orthophosphorus
Potassium
Calcium
Magnesium
Sulfur
Sodium
Chloride
Iron
Manganese
Boron
Molybdenum
Zinc
Copper
Cadmium
Nickel
Sector
Beef
Dairy
Veal
Hog
Broiler
Layer
Turkey
(average pounds per 1,000 pounds live animal weight per day)
793.7
58
18
1000
8.5
7.2
1.6
7.8
7.0
3.40e-01
8.60e-02
9.2e-02
3.0e-02
2.1e-01
1.4e-01
4.9e-02
4.5e-02
3.0e-02
7.8e-03
1.2e-03
8.8e-04
4.2e-05
l.le-03
3.1e-04
1410.9
86
26
990
12.0
10.0
1-6
11.0
7.0
4.50e-01
7.90e-02
9.4e-02
6.1e-02
2.9e-01
, 1.6e-01
7.1e-02
5.1e-02
5.2e-02
1.3e-01
1.2e-02
1.9e-03
7.1e-04
7.4e-05
1.8e-03
4.5e-04
3.0e-06
2.8e-04
200.6
62
1000
5.2
2.3
1.7
5.3
8.1
2.70e-01
1.20e-01
6.6e-02
2.8e-01
5.9e-02
3.3e-02
8.9e-02
3.3e-04
1.3e-02
4.8e-05
134.5
84
39
990
11.0
8.5
3.1
.8.4
7.5
5.20e-01
2.90e-01
1.8e-01
1.2e-01
. 2.9e-01
3.3e-01
7.0e-02
7.6e-02
6.7e-02
2.6e-01
1.6e-02.
1.9e-03
3.1e-03
2.8e-05
5.0e-03
1.2e-03
2.7e-05
992.1
85
1000
22.0
17.0
16.0
1.10e+00
3.0e-01
4.0e-01
4.1e-01
1.5e-01
8.1e-02
l.SerOl
3.6e-03
9.8e-04
4.0
64
970
16.0
12.0
3.3
11.0
6.9
8.40e-01
2.10e-01
3.0e-01
9.0e-02
3.0e-01
1.3e+00
1.4e-01
1.4e-01
l.Oe-01
5.6e-01
6.0e-02
l.Oe-03
1.8e-03
3.0e-04
1.9e-02
8.3e-04
3.8e-05
2.5e-04
4.0
47
1000
12.0
9-1.
2.1
9.3
6.206-01
S.OOe-02
2.3e-01
2.4e-01
6.3e-01
7.3e-02
6.6e-02
7.5e-02
2.4e-03
1.5e-02
7.1e-04
E-5
-------
Table E-3. Nutrients, Metals, and Pathogens in Livestock and Poultry Manures (continued)
Constituent
Lead
Arsenic
Total coliform bacteria
Fecal coliform bacteria
Fecal streptococcus
bacteria
Sector
Beef
Dairy
Veal
Hog Broiler
Layer
(average pounds per 1,000 pounds live animal weight per day)
na
29
13
14
na
500
7.2
42
na
8.4e-05
6.9e-04
21
8
240
4.8e-04
7.4e-04
5.5e-06
50
3.4
7.4
Turkey
na
0.62
Source: ASAE, 1993. Arsenic values are from NCSU, 1994. Converted from reported (lb./yr./l,000 Ib. animal
weight) values of 0.002 (layers), 0.176 (broilers), and 0.252 (hogs). All values are in pounds unless otherwise
noted.
(Cl), copper (Cu), iron (Fe), lead (Pb), magnesium (Mg), manganese (Ma), molybdenum (Mo),
nickel (Ni), potassium (K), sodium (Na), sulfur (S), zinc (Zn), nitrogen (N) and phosphorus (P)
species, total suspended solids (TSS), and pathogens (ASAE, 1993). Other research conducted
by various land grant universities reports that arsenic, selenium, and other constituents are also
present in some animal manures (NCSU, 1994; Sims, 1995).
Concentrations shown in Table E-3 are reported in pounds per 1,000 pounds of live
animal weight per day and vary by animal species. As shown in Table E-3, poultry manure
generally has a higher nutrient content than most other farm animal manure and wastewater.
Actual nutrient values of manure depend on many factors, including: animal size, maturity, and
species; health and diet of the animals; and the feed composition and the protein content of the
ration fed (USDA, 1992). Additional details on the constituents hi animal manure are provided in
the Environmental Assessment (USEPA, 2000b) and the Development Document (USEPA,
2000a).
E.2 ESTIMATED POLLUTANT REMOVALS
For this analysis, EPA estimates the expected reduction of selected pollutants for each of
the regulatory options considered. These estimates measure the amount of nutrients, sediments,
and metals originating from animal production areas that are removed under a post-regulation
scenario (as compared to a baseline scenario) and do not reach U.S. waters. Additional
information on EPA's estimated loadings and removals under post-compliance conditions is
provided in the Development Document (USEPA, 2000a) and the Benefits Analysis (USEPA,
2000d) that support the rulemaking.
E-6
-------
USDA estimates that manure nutrients available for land application from all confined
livestock and poultry operations totaled 2.6 billion pounds of nitrogen and 1.4 billion pounds of
phosphorus in 1997 (Kellogg, et al., 2000).4 EPA equates these estimates to the total amount of
manure and nutrients generated at animal feeding operations. Of this amount generated, EPA
estimates that between 64 percent (two-tier) and 72 percent (three-tier) of all CAFO manure
would be covered by the proposed regulation. More information on these estimates are presented
in Section 2 of this report. Table E-4 shows these estimates broken out by major subcategory.
F.-4 Nutrients Generated from CAFOs and Loadings that Discharge to U.S. Waters (Baseline)
Sector
Cattle
Dairy
Hogs
All Poultry
Sum Total
"Available for Land Application"
(>500AU) *
Nitrogen
Phosphorus
"Edge-of-Field"
(>500AU) b/
Nitrogen
Phosphorus
(million pounds)
450.9
254:3
187.5
967.6
1 860 3
313.8
97.6
189.4
464.8
1,065.5
122.7
56.9
68.8
186.2
434.6
55.9
138.7
129.9
405.6
730.0
NA=Not Available.
Source(s): .
"Edge-of-field" loadings measure cropland runoff from land application of manure and wastewater. Estimates of
the amount discharged to U.S. waters are derived based on percentage difference, in estimated "At-Stream"
loadings. Baseline reflects pre-regulatory loadings. •
"'Manure nutrients generated at animal feeding operations with more than 500 AU are derived by EPA from
estimated loadings from USDA/NRCS staff (Kellogg, et al., 2000 forthcoming) using 1997 Census of Agriculture
(USDA/NASS, 1999a) and procedures documented in Kellogg et al. (2000) and Lander et al. (1998). Additional
information is presented in Section 2 of this report.
^USEPA, 2000a. "Edge-of-field" loadings measure cropland runoff from land applied manure and wastewater.
Table E-4 also shows EPA's aggregated estimates of loadings that reach the "edge-of-
field" from land applied manure by AFOs with more than 500 AU. EPA estimates these loadings
using a simulation modeling approach based on representative model CAFOs (similar to that used
to estimate compliance costs of the proposed regulations, as described in Section 4). This model
uses estimates of manure generation and information on cropping systems specific to animal
operations under various pre- and post-regulation model simulation conditions. Model CAFOs
take into account differing conditions at CAFOs. These conditions include animal type,
production region, facility size, current management systems and practices, and regionally based
estimates do not include manure generated from other animal agricultural operations, such as sheep
and lamb^ goats, horses, and other miscellaneous animal species.
E-7
-------
physiographic conditions regarding soil, rainfall, hydrology, crop rotation, and other factors.
More details on these models and a summary of the estimated loadings and post-compliance
reductions are provided in the Development Document (USEPA, 2000a).
EPA evaluates post-regulatory conditions in terms of expected reductions from estimated
baseline (pre-regulatory) loadings. (Baseline loadings are shown in Table E-4 and E-5, and
estimated post-compliance reductions are discussed later in Sections E-3 and E-4.) EPA's model
simulates loadings and reductions for total nitrogen (TN), total phosphorus (TP), and total
suspended solids (TSS). Under this approach, EPA also simulates loadings and reductions for six
selected metallic compounds that are present in animal manure (Cd, Cu, Ni, Pb, Zn, As) and also
pathogens (expressed as fecal coliform and fecal streptococcus). Estimated baseline metallic
compound loadings from CAFOs are shown in Table E-5. Additional information on this
approach is provided in the Development Document (USEPA, 2000a).
Table E-5. Estimated Metals Generated and "Edge-of-Field" Loadings from CAFOs
Sector
Cattle
Daiiy
Hogs
All Poultry
Sum Total
"Edge-of-Field" Baseline Loadings
Cadmium
Copper
Lead
Nickel
Zinc
Arsenic
(million pounds)
0.06
0.00
0.01
0.10
0.17
1.65
0.57
0.07
3.07
5.36
1.54
0.11
0.04
2.85
4.54
0.38
0.35
0.13
0.80
1.66
34.90
2.26
2.70
66.06
105.91
0.20
0.00
0.33
0.40
0.75
Source: USEPA, 2000a. "Edge-of-field" loadings measure cropland runoff from land application of manure and
wastewater. Estimates of the amount discharged to U.S. waters are derived based on percentage difference in
estimated "At-Stream" loadings.
EPA uses the estimated "edge-of-field" loadings shown in Table E-4 as inputs to its water
quality modeling framework that determines the pollutant loadings reaching U.S. waterways.
Information on EPA's fate and transport model and a comparison of "edge-of-field" loadings and
pollutants that reach U.S. waters (measured as rivers and streams), referred to here as "at-stream"
loadings, is provided in the Benefits Analysis (USEPA, 2000d). This analysis indicates that about
70 percent of all land applied manure phosphorous and sediments in applied manure reach U.S.
waters; about 90 percent of all land applied manure nitrogen is estimated to reach U.S. waters
(USEPA, 2000d). This level of nutrients reaching U.S. waters is explained by differing
assumptions on a variety of levels, including manure generation by animal species, the share of
animals in confinement, and losses due to volatization and management practice, as well as other
factors, including rounding and truncation error and assumptions regarding background levels.
More information is available in the Development Document (USEPA, 2000a) and the Benefits
Analysis (USEPA, 2000d).
E-8
-------
E.3 COST-EFFECTIVENESS ANALYSIS: TOXIC POLLUTANTS
This section reviews EPA's C-E analysis for selected toxic pollutant removals found in
animal manure. This analysis follows the guidelines of a standard C-E analysis commonly used by
EPA to compare the efficiency of regulatory options for effluent guidelines in removing priority
and non-conventional pollutants.
This analysis evaluates cost-effectiveness as the incremental and average annualized cost
of a pollution control option in an industry (or industry subcategory) per incremental and total
pound equivalent of pollutant (i.e., pound of pollutant .adjusted for toxicity) removed by that
control option. Section E.3.1 outlines this approach. The C-E analysis primarily enables EPA to
compare the removal efficiencies of regulatory options under consideration for a rule. Section
E.3.2 presents these results. A secondary use is to compare the cost-effectiveness of the
proposed option for the effluent guidelines to that of effluent guidelines for other industries, as
shown in Section E.3.3.
E.3.1 Methodology
Performing a standard EPA C-E analysis involves the following seven steps:
1. Determine the pollutants of concern (priority or other pollutants).
2. Estimate relative toxic weights for these pollutants.
3. Define the regulatory pollution control options.
4. Calculate pollutant removals for each control option.
5. Determine the total annualized cost for each control option.
6. Calculate cost-effectiveness values (and adjust to 1981 dollars).
7. Compare cost-effectiveness values.
For this C-E analysis, EPA has identified six toxic pollutants of concern (step 1), including
arsenic, zinc, copper cadmium, nickel, and lead. Factors considered in selecting these pollutants
included toxicity, frequency of occurrence in wastestream effluent, and amount of pollutant in the
waste stream (USEPA, 2000a). As shown in Table E-3, this is a subset of all the toxic
compounds reported to be present in farm animal manures (which varies by animal type).
Therefore, the cost-effectiveness results presented here are conservative (i.e., a higher per-pound
removal value is estimated than would be the case if all metals Were considered in this analysis).
Table E-6 presents the toxic weighting factors (TWFs) for the regulated pollutants used in
this C-E analysis (step 2). In C-E analyses, EPA measures pollutant removals in toxicity
normalized units called "pounds-equivalent," where the pounds-equivalent removed for a
particular pollutant is determined by multiplying the number of pounds of a pollutant removed by
each option by a TWF for each pollutant. EPA uses these factors to account for differences in
toxicity among the pollutants and to adjust the estimated pollutant loading values to account for
E-9
-------
Table E-6. Total Metal Removals "At Stream" by Regulatory Option Considered
Metal
Zinc
Copper
Cadmium
Nickel
Lead
Arsenic
Total
TWF
0.047
0.63
2.6
0.11
2.2
4
NA
Proposed BAT Option
Lbs.
Lbs.-Eq.
Option 1
Lbs.
Lbs.-Eq.
Option 3+5
Lbs.
Lbs.-Eq.
(millions of pounds or pounds-equivalent)
59.0
2.9
0.1
2.4
0.9
0.4
657
2.8
1.8
0.3
0.3
1.9
1.4
84
51.2
2.4
0.1
2.1
0.7
0.2
56.7
2.4
1.5
0.3
0.2
1.5
1.0
6.8
59.0
2.9
0.1
2.4
0.9
0.4
65.7
2.8
1.8
0.3
0.3
1.9
1.4
8.4
Source: Toxic weighting factors (TWF) are calculated by EPA and are based on freshwater chronic criteria lor
copper (USEPA, 2000J). Pound removals are estimated by EPA (USEPA, 2000a). Pounds-equivalent are
calculated by multiplying the estimated pounds of removals by the TWFs shown here.
the fact that different pollutants have different potential effects on human and aquatic life (e.g., a
pound of zinc in an effluent stream has a different effect than a pound of arsenic).
EPA derives TWFs for pollutants using ambient water quality criteria and toxicity values.5
The factors are standardized by relating them to a "benchmark" toxicity value that was based on
the toxicity of copper when the methodology was developed.6 For most pollutants, EPA uses
established chronic freshwater aquatic criteria to derive toxic weighting factors. In cases where a
human health criterion has also been established for the consumption of fish, the sum of both the
human and aquatic criteria are used. The toxic weighting factor is the sum of two criteria-
weighted ratios: the "benchmark/old" copper criterion divided by the human health criterion for
the particular pollutant and the "benchmark/old" copper criterion divided by the aquatic chronic
criterion. This is shown in the following equation:
.
J HHC- AqHC
J J
where:
5Human health and aquatic chronic criteria are maximum contamination thresholds. Units for criteria are
micrograms of pollutant per liter of water. Most values are those reported in the toxicology literature.
6Although the water quality criterion has been revised (to 12.0 ug/1), all cost-effectiveness analyses for effluent
guideline regulations continue to use the "old" criterion of 5.6 ug/1 as a benchmark so that cost-effectiveness values
can continue to be compared to those for other effluent guidelines. Where copper is present in the effluent, the
revised higher criterion for copper results in a toxic weighting factor for copper of 0.467 rather than 1.0.
E-10
-------
HHCj
AqHCj
Constant (5.6) =
Human Health Criteria for pollutant j (if available)
Aquatic Health Criteria for pollutant j
"Benchmark/old" criterion for copper
Given the established values for these criteria, the TWF for each pollutant can be calculated
(USEPA, 2000f). For example, for cadmium, the human health criteria is 84 and the aquatic
health criteria is 1.1. This translates to a TWF for cadmium of 5.16 (Table E-6). EPA uses
TWFs for the selected priority pollutants (Cd, Cu, Ni, Pb, Zn, As) to adjust estimated removals
to a "pounds-equivalent" basis (as shown in. Table E-6) by multiplying the estimated pounds of
removals by the TWFs.
This C-E analysis investigates three pollution control options (step 3). As discussed in
Section 3 of this report, the proposed BAT Option being proposed for this rulemaking is a
combination of Option 3 for beef and dairy (except veal) and Option 5 for pork, veal, and poultry.
In addition, the analysis presents results for the two co-proposed regulatory scenarios (Scenario
4a, the two-tier structure, and Scenario 3, the three-tier structure). Results for Scenario 5 (two-
tier structure at 750 AU threshold) and Scenario 6 are not determined, but fall within the range of
results shown over all scenarios discussed. The two-tier structure includes all operations with
more than 500 AU, while the three-tier structure may include operations with more than 300 AU
that meet the "risk-based" conditions (see Section 3). Two alternative regulatory options
considered for comparison in this cost-effectiveness analysis are Option 1 and a"worst-case"
option that combines the full cost of Option 3 and with the full cost of Option 5 for all
subcategories.7 This latter option is called Option 3+5. These alternative options were chosen for
comparison because they generally represent the lower and upper bound of costs and removals:
Option 1 represents the lowest costs and least removals and Option 3+5 represents the highest
costs and removals.8 In addition, two alternative NPDES Scenarios are shown: CAFOs with
more than 1,000 AU and with more than 300 AU. The differences in CAFO coverage provide an
upper and lower bound of the analysis to roughly depict the alternative NPDES scenarios,
including Scenario 4b (operations >300 AU) and Scenario 1 (approximated using estimates for
operations >1,000 AU). These options/scenarios are described in detail in Section 3 of this
report.
First, EPA calculates pollutant loadings for each animal sector under each regulatory
option (step 4). EPA then calculates pollutant removals as the difference between baseline and
post-treatment discharges. Tables E-4 and E-5 present the baseline (pre-regulatory) loadings and
•7,
Calculated for this analysis as the reported cost of Option 3 (Tables 5-1 and 5-2) plus the cost of Option 5,
minus the cost of Option 2 (see Tables 5-1 and 5-2), since both Option 3 and 5 are incremental to Option 2.
8
'See Section 10 for pre-tax annualized costs and the Development Document (USEPA, 2000a) for estimated
removals.
E-ll
-------
removals for the proposed BAT option, as described in Section E.2.9 The estimated removals are
weighted using the toxic weighting factors (Table E-6) and are reported in pounds-equivalent (lb.-
eq.) across all pollutants. Finally, EPA calculates total weighted removals for each regulatory
option by summing the removals for each of the identified priority pollutants for each option (see
Table E-7). For more information, see the Development Document (USEPA, 2000a).
Pre-tax annualized costs of compliance for each regulatory option (step 5) have been
developed by EPA and are presented in Section 10 of this report. Estimated costs include private
sector costs of purchasing, installing, and operating pollution control systems,10 as well as
administrative costs to state and federal governments for authorizing NPDES permits
(Section 10) and costs to offsite recipients of CAFO manure (Section 5 and, also, USEPA,
2000a). Table E-7 presents the aggregate annual pre-tax costs by option. For the purpose of
comparing C-E values of options under review to those of other promulgated rules, estimated
compliance costs (which are reported in 1999 dollars in Section 10, but are originally developed
in 1997 dollars) are adjusted to 1981 dollars using Engineering News Record's Construction Cost
Index (CCI) (ENR, 2000). Table E-7 shows compliance costs for the proposed CAFO
regulations in 1981 dollars. For this analysis, this adjustment factor is calculated as follows:
Adjustment factor = (1981 CCI)/(1997 CCI) = 3,535/5,825 = 0.6069
EPA calculates cost-effectiveness values separately for each regulatory option (step 6) and
reports these in units of dollars per pound-equivalent of pollutant removed. Generally, options
first are ranked in ascending order of pounds-equivalent of pollutants removed. The incremental
cost-effectiveness value for a particular control option is calculated as the ratio of the incremental
annual cost to the incremental pounds-equivalent removed. The average cost-effectiveness value
for each option is calculated as total dollars for the option divided by total pounds-equivalent
removed by the option. The incremental cost-effectiveness values are viewed in comparison to
the baseline (zero costs/zero removals) for the first option and to the preceding regulatory option.
Cost-effectiveness values are reported in units of dollars per pound-equivalent of pollutant
removed. In this report, EPA presents average cost-effectiveness values, as discussed below.
Cost-effectiveness values can be compared across regulatory alternatives according to
either the average or the incremental cost-effectiveness (step 7). Average cost-effectiveness
values for each option are calculated as total dollars for the option divided by total pounds-
equivalent removed by the option. Average cost-effectiveness reflects the "increment" between
no regulation and the regulatory options shown. Incremental cost-effectiveness is the appropriate
'Pollutant loadings and removals for Options 1 and 3+5 are included in the Development Document (USEPA,
2000a) and the Environmental Assessment (USEPA, 2000b).
IOEvery dollar spent on compliance can be applied against a firm's taxable income. Due to various tax
mechanisms such as accelerated depreciation, this reduction means that firms face only about 70 percent of
compliance costs after taxes.
E-12
-------
measure for comparing one regulatory option to an alternative, less stringent regulatory option for
the same subcategory. The incremental cost-effectiveness value for a particular control option is
calculated as the ratio of the incremental annual cost (e.g., the difference in cost between the .
option removing the least pollutants to the next better option) to the incremental pounds-
equivalent removed' (calculated between the same options). The equation EPA uses to calculate
incremental cost-effectiveness is:
where:
Cost-effectiveness of Option k
Total annualized treatment cost under Option k
Pounds-equivalent removed by Option k
ATCk =
PEk =
The numerator of the equation, ATCk minus ATCk.!, is me incremental annualized
treatment cost in going from Option k-1 (an option that removes fewer pounds-equivalent of
pollutants) to Option k (an option that removes more pounds-equivalent of pollutants). The >
denominator is similarly the incremental removals achieved in going from Option k-1 to k. Thus,
cost-effectiveness measures the incremental unit cost of pollutant removal of Option k (in pound
equivalents) in comparison to Option k-1. Average cost-effectiveness values can also be derived
using this equation by setting ATCk.j to zero and.by setting the pollutant loadings (PE^) to the
baseline loadings. . ,
Because the options are ranked in ascending order of pounds-equivalent of pollutants
removed, any option that has higher costs but lower removals than another option immediately
can be identified (the cost-effectiveness value for the next option becomes negative). When
negative values are computed for Option k, Option k-1 will be noted as "dominated" (having a
higher cost and lower removals than Option k). Option k-1 is then removed from the cost-
effectiveness calculations, and all cost-effectiveness values within a regulatory grouping are then
recalculated without the "dominated" option. This process continues until all "dominated"
options are eliminated. The remaining options can then be presented in terms of their incremental
cost-effectiveness values and are considered viable options for regulatory consideration.
These values can be used, with caution, to compare an option to previously promulgated
effluent limitations guidelines. Because Option 3+5 (all subcategories) removes no more
pollutants at a higher cost than the proposed BAT Option, Option 4 is dominated. Because
Option 4 is dominated, incremental cost-effectiveness is not meaningful. Thus EPA presents, in
Section E.3.2, the average cost-effectiveness results for the proposed BAT Option, Option 1, and
Option 3+5 at each size cutoff.
E-13
-------
EPA generally ranks options in order of increasing pounds-equivalent removed to identity
the point at which increased removal of pollutants is no longer cost-effective. EPA typically
determines this point be where costs (per pound-equivalent removed) increase sharply, that is,
where relatively few incremental pounds are removed for steady increases in cost. The
accompanying figure (Figure E-l) shows this point as Point A, where the cost-effectiveness curve
Rangeof
noncost-
effective
removals
Rage of
cost-cflccthrc
removals
I
100
Percentage of pounds-equivalent removed
Figure E-l. Cost-Effectiveness
becomes nearly vertical. Increases in removals beyond this point come only at relatively high unit
costs, which in many cases EPA will determine exceed the benefit of the increased removals to
society. .
E.3.2 Cost-Effectiveness Results
Table E-7 presents cost-effectiveness results for the proposed BAT Option under both co-
proposed scenarios, Scenario 4a (two-tier structure) and Scenario 3 (three-tier structure) with
regard to priority pollutants. Also shown are the results for alternative regulatory options that
were considered by EPA. Results shown in Table E-7 for select technology options, including the
proposed BAT Option (Option 3 for beef and dairy subcategories, except veal, and Option 5 for
E-14'
-------
Table E-7. Cost-Effectiveness Results by Select Regulatory Option/Scenario ($1981)
Option
Total Annual
Pounds-equivalent
Removed ^
(million pounds)
Total Cost b/
($ millions)
Average
Cost-
Effectiveness
Incremental
Cost-
Effectiveness
($/lbs.-eq.)
ELG Option 3 (Beef/Dairy) and 5 (Swine/Veal/Poultry)
>1000 AU
•>500 AU "Two-Tier"
>300 AU "Three-tier"
>300 AU (Scenario 4b)
5.3
8.4
9.4
9.4
$402
$491
$518
$579
$76
$58
.$55
$62
$76
$29
$27
undefined
ELG Option 1 (All Subcategories)
>1000 AU
>500 AU "Two-Tier"
>300 AU "Three-tier"
>300 AU (Scenario 4b)
4.2
6.8
7.5
7.5
$211
$258
$277
$297
$50
$38
$37
$40
ELG Option 3+5 (All Subcategories)
>1000 AU
>500 AU "Two-Tier"
>300 AU "Three-tier"
>300 AU (Scenario 4b)
5.3
.8.4
9.4
9.4
$1,047
$1,212
$1,254
$1,354
$197
$144
$134
$144
$50
$18
$27
undefined
$197 , .
$53
$43
undefined
Options/Scenarios are described in Table 3-1.
^Pounds-equivalent removals are calculated from removals estimated by EPA's loadings analysis, described in the
Benefits Analysis (USEPA, 2000d) and the Development Document (USEPA, 2000a), adjusting for each
pollutant's toxic weight (as described Section E.3.1).
w Costs are pre-tax and indexed to 1981 dollars using the Construction Cost Index (ENR, 2000).
the swine, veal, and poultry Subcategories) and Option 3+5 (both Option 3 and 5 for all
Subcategories). Options are shown for four CAFO coverage scenarios, including CAFOs with
more than 1,000 AU and CAFOs with more than 500 AU (two-tier structure), and operations
with more than 300 AU, both under Scenario 4b and as defined under Scenario 3 (three-tier
structure). The differences in CAFO coverage provide "an upper and lower bound of the analysis
to roughly depict the alternative NPDES scenarios. Both incremental and average C-E values are
shown.
Incremental cost-effectiveness is the appropriate measure for comparing one regulatory
alternative to another for the same subcategory: In general, the lower the incremental C-E value,
the more cost-efficient the regulatory option is in removing pollutants, taking into account their
E-15
-------
toxicity. For this rulemaMng, EPA compares the cost-effectiveness across alternative NPDES
Scenarios to assess the Agency's decision to define as CAFO operations with more than 500 AU
(two-tier structure) and, alternatively, operations with more than 300 AU, as defined by Scenario
3 (three-tier structure).
As shown in Table E-7, the BAT Option is the most cost-efficient under each of the co-
proposed alternatives. Under both the two-tier (500 AU) and three-tier structures, EPA estimates
an incremental cost-effectiveness value of about $30 per pounds-equivalent (Ibs.-eq.) removed. In
comparison, an alternative scenario covering operations with more than 1,000 AU has a higher
estimated incremental cost-effectiveness ($76/lbs.-eq., as shown in Table E-7). (Since the change
in removals between Scenario 3 and Scenario 4b is zero, the incremental C-E value is
"undefined.") The BAT Option is also more efficient than requiring Option 3+5 for all
subcategories, which has higher costs but results in no additional pollutant removals compared to
the BAT Option. This is because the ELG options differ mostly in terms of their monitoring and
sampling requirements but establish no additional pollutant controls. (Since the change in
removals between the BAT Option and Option 3+5 is zero, the incremental C-E value is
undefined.)
The average cost-effectiveness reflects the "increment" between no regulation and
regulatory options shown. For the BAT Option, EPA estimates an average value at $55 per Ibs.-
eq. to $58 per Ibs.-eq., depending on the proposed tier structure (Table E-7). These estimated
average values are low compared to the alternative NPDES scenarios since the average cost-
effectiveness value is higher ($76/lbs.-eq., if all CAFOs with more than 1,000 AU are regulated;
$62/lbs.-eq. for all CAFOs with more than 300 AU). This average cost is also low compared to
previous ELG rulemakings, where estimated costs have, in some cases, exceeded $100/lbs.-eq.
removed. This information is provided in Section E.3.3. In addition, as shown in Table E-7,
average cost-effectiveness is more than twice as high under the more stringent Option 3+5 for all
subcategories (estimated at more than $100 per Ibs.-eq. removed). Costs, but also removals, are
lower under the-less stringent Option 1 (also referred to as the "nitrogen-based" option)
compared to other technology options. The average cost-effectiveness of Option 1 is $38/lbs.-eq.
under the two-tier structure and $37/lb.-eq. under the three-tier structure. As described in
Section VHI of the preamble, EPA determined that this option would not represent the best
available technology and so chose not to propose it.
Tables E-8 and E-9 show the results of EPA's analysis of average cost-effectiveness by
major subcategory groupings (cattle, dairy, hogs, and poultry) for the BAT Option under the two-
tier and three-tier structures, respectively. As shown in Tables E-8 and E-9, under both co-
proposed scenarios, there is a wide discrepancy in cost-effectiveness among the commodity
sectors. In particular, the cost to remove a pound of toxic pound-equivalent is greatest in the hog
and dairy sectors. The average cost-effectiveness of the proposed BAT Option is more than $100
per pound-equivalent in these two sectors under either co-proposed scenario, whereas removal
costs are about $55/lbs.-eq. for cattle (includes beef, veal, and heifer) and about $17/lbs.-eq. for
poultry under either tier (Tables E-8 and E-9).
E-16
-------
Table E-8. Cost-Effectiveness Results by Sector under the Two-Tier Structure (Scenario 4a)(S1981)
Sector
Total Annual
Pounds-equivalent
Removed "
(million pounds) .
Total Cost b/
($ millions)
Average
Cost-Effectiveness
($/lbs.-eq.)
ELG Option 3 (Beef/Dairy) and 5 (Swine/Veal/Poultry)
Cattle
Dairy
Hogs
Poultry
2.4
0.7
0.7
4.6
$135.2
$104.8
$171.9
$79.0
$55.2
$150.8
$260.9
$17.1
ELG Option 1 (AH Subcategories)
Cattle
Dairy
Hogs
Poultry
2.3
0.6
0.2
• 3.7
$50.0
$61.1
$80.7
$66.2
. , $21.5
$109.0
$403.1
$17.8
ELG Option 3+5 (AH Subcategories)
Cattle
Dairy
Hogs
Poultry
2.4
0.7
0.7
4.6
$670.7
$206.7
$288.4
$161.6
$273.9
$297.3 .
$437.7
$35.0
Options/Scenarios are described in Table 3-1. "Cattle" include beef, heifer, and veal operations. "Poultry"
includes broiler, egg, and turkey operations.
"'Pounds-equivalent removals are calculated from removals estimated by EPA's loadings analysis, described in the
Benefits Analysis (USEPA, 2000d) and the Development Document (USEPA, 2000a), adjusting for each
pollutant's toxic weight (as described Section E.3.1).
b/Costs are pre-tax and indexed to 1981 dollars using the Construction Cost Index (ENR, 2000).
E-17
-------
Table E-9. Cost-Effectiveness Results by Sector under the Three-Tier Structure (Scenario 3)(S1981)
Sector
Total Annual
Pounds-
equivalent
Removed ^
(million pounds)
Total Cost b/
($ millions)
Average
Cost-Effectiveness
($/lb.-eq.)
ELG Option 3 (Beef/Dairy) and 5 (Swine/Veal/Poultry)
Cattle
Dairy
Hogs
Poultry
2.5
1.0
0.8
5.2
$136.7
$128.0
$160.3
$92.5
$55.7
$132.3
$212.4
$17.7
ELG Option 1 (All Subcategories)
Cattle
Dairy
Hogs
Poultry
2.3
0.8
0.2
4.2
$51.1
$70.3
$78.5
$77.2
$21.9
$90.9
$355.2
$18.4
ELG Option 3+5 (AH Subcategories)
Cattle
Dairy
Hogs
Poultry
2.5
1.0
0.8
5.2
$671.8
$250.4
$270.4
$190.7
$273.7
$258.8
$358.3
$36.6
Options/Scenarios are described in Table 3-1. "Cattle" include beef, heifer, and veal operations. "Poultry"
includes broiler, egg, and turkey operations.
"'Pound-equivalent removals are calculated from removals estimated by EPA's loadings analysis, described in the
Benefits Analysis (USEPA, 2000d) and the Development Document (USEPA, 2000a), adjusting for each
pollutant's toxic weight (as described Section E.3.1).
w Costs are pre-tax and indexed to 1981 dollars using the Construction Cost Index (ENR, 2000).
E.3.3 Comparison of Cost-Effectiveness Values with Promulgated Rules
Incremental cost-effectiveness is the appropriate measure for comparing one regulatory
option to an alternative, less stringent regulatory option for the same rule. For comparing the
overall cost-effectiveness of one rule to another, average cost-effectiveness may be a more
appropriate measure, but must be considered in context with caution. (Average cost-effectiveness
can be thought of as the "increment" between no regulation and the selected option for any given
rule.)
E-18
-------
Table E-10 presents the incremental cost-effectiveness values for the proposed regulatory
option under the two co-proposed regulatory scenarios and the values for effluent guidelines
issued for other industries. The numbers presented here for this rulemaking are pre-tax costs.11
As Table E-10 shows, the incremental cost-effectiveness of the proposed BAT Option is $29 or.
$27 per pound-equivalent removed, for the two-tier and three-tier structures, respectively (also
see Table E-7). Compared with other effluent guidelines, EPA considers this rule to be cost-
effective under either co-proposed scenario.
Table E-10. Industry Comparison of BAT Cost-Effectiveness for Direct Dischargers
a/
Industry
Aluminum Forming
Battery Manufacturing
CAFOs (two-tier)
CAFOs (three-tier)
Canmaking
Centralized Waste Treatmentb/
Coal Mining
Coil Coating
Copper Forming
Electronics I
Electronics n
Foundries
Inorganic Chemicals I
Inorganic Chemicals II
Iron and Steel
Leather Tanning
Metal Finishing
Metal Products and Machinery11
Nonferrous Metals Forming
Pounds-Equivalent
Currently
Discharged
Pounds-Equivalent
Remaining at
Selected Option(s)
(1,000's)
1,340
4,126
12,273
13,700
12
3,372
BAT=BPT
2,289
70
9
NA
2,308
32,503
605
40,746
259
3,305
140
34
90
5
3,857
4,305
0
1,261 - 1,267
BAT=BPT
9
8
3
NA
39
1,290
27
1,040
112
3,268
70 '
2
Cost-Effectiveness of
Selected Option(s)
(S/PE removed)
($/lb-equiv. removed)
121
2
29
27
10
5-7
BAT=BPT
49
27
404
NA
84
<1
6
2
BAT=BPT
1.2
50
69
"Average C-E values presented for other effluent guidelines (Table E-10) may be expressed in terms of after-
tax costs—that is, the costs actually faced by the firms, not the total cost of the regulation to industry (which is
subsidized by reductions in taxable income). Because of these factors, direct comparisons between this rulemaking
and others cannot be made easily. The equivalent after-tax cost is approximately 60-70 percent of pre-tax costs.
E-19
-------
Table E-10. Industry Comparison of BAT Cost-Effectiveness for Direct Dischargers (continued)
Industry
Nonferrous Metals Manufacturing I
Nonfenous Metals Manufacturing n
Oil and Gas: Offshore
Coastal Produced Water/TWC
Drilling Waste
Organic Chemicals'1
Pesticides
A/C
B/D
Plastics Molding and Forming
Porcelain Enameling
Petroleum Refining
Pulp and Paper
Textile Mills
Transportation Equipment Cleaning0
Pounds-Equivalent
Currently
Discharged
Pounds-Equivalent
Remaining at
Selected Option(s)
(1,000's)
6,653
1,004
3,809
951
BAT=Current
Practice
54,225
2,461
560
0.1
44.0
1,086
BAT=BPT
15,524
BAT=BPT
15
313
12
2,328
239
BAT=Current Practice
9,735
371
550
— "
41
63
BAT=BPT
4,069
BAT=BPT
0.8
Cost-Effectiveness of
Selected Option(s)
(S/PE removed)
($/lb-equiv. removed)
4
6
33
35
B AT=Current Practice
5
14
224-
BAT=BPT
BAT=BPT
6
BAT=BPT
14
BAT=BPT
108
Toxic and Nonconventional Pollutants Only; Copper-Based Weights; $1981.
"'Although toxic weighting factors for priority pollutants varied across these rules, this table reflects the cost-
effectiveness at the time of regulation.
b/Reflects costs and removals of both air and water pollutants.
E.4 COST-EFFECTIVENESS ANALYSIS: NUTRIENTS AND SEDIMENTS
In addition to conducting a standard C-E analysis for selected toxic pollutants
(Section E.3), EPA also evaluates the cost-effectiveness of removing selected non-conventional
and conventional pollutants, including nitrogen (N), phosphorus (P), and sediments. For this
analysis, sediments are used as a proxy for TSS.
This analysis does not follow the methodological approach of a standard C-E analysis.
Instead, this analysis compares the estimated compliance cost per pound of pollutant removed to a
recognized benchmark, such as EPA's benchmark for conventional pollutants or other criteria for
existing treatment, as reported in available cost-effectiveness studies. A review of this literature is
provided in Section E.4.1. EPA uses these estimates to evaluate the efficiency of regulatory
options in removing a range of pollutants and to compare the proposed BAT Option under the
co-proposed scenarios to other regulatory alternatives (Section E.4.2). This approach also allows
E-20
-------
EPA to assess the types, of management practices that would be implemented to comply with the
proposed regulations.
E.4.1 Review of Literature
EPA has reviewed the available information on pollutant removal costs for nutrients and
sediments. This research can be broadly grouped according to estimates derived for industrial
point sources (PS) and various nonpoint sources (NPS), including agricultural operations. In
general, the PS research provides information on technology and retrofitting costs—and in some
cases, cost per pound of pollutant removed—at municipal facilities, including publicly owned
treatment works (POTWs) and wastewater treatment plants (WWTPs). This research differs
from other cost-effectiveness estimates, since it utilizes actual cost data collected at a particular
facility undergoing an upgrade. Other cost-effectiveness information is available based on the
effectiveness of various nonpoint source controls and Best Management Practices (BMPs) and
other pollutant control technologies that are commonly used to control runoff from agricultural
lands. Typically, this information uses a modeling approach and simulates costs for a
representative facility.
Table E-l 1 summarizes the cost-effectiveness values reported in the studies that were
reviewed for this analysis. A wide range of costs per pound of pollutant removed is estimated by
those studies where estimates span both point source and nonpoint sources, as well as those
where estimates span a range of municipal, urban, and agricultural practices. Annualized costs
also vary widely depending on a variety of factors, including the type of treatment system or
practice evaluated, and whether the costs are evaluated as a retrofit to an existing operation or as
construction of a new facility. .
A series of case studies were compiled by researchers at Virginia Tech, who evaluated
total costs for biological nutrient removal (BNR) retrofits at WWTPs throughout the Chesapeake
Bay Watershed (Randall et al., 1999). These case studies were compiled to estimate a range of
costs per pound of nitrogen removed at these facilities. This research was commissioned by
EPA's Chesapeake Bay Program and was conducted with the assistance of the Maryland
Department of the Environment and the Public Utilities Division of Anne Arundel County. As
part of this work, BNR retrofit costs were estimated for 51 WWTPs located in Maryland,
Pennsylvania, Virginia, and New York. The final report in this series compares these costs to the
projected change in effluent total nitrogen concentrations, assuming that the influent flow meets
the design or projected flow after 20 years (Randall, et al., 1999). As shown in Table E-l 1, this
study concludes that the costs of nitrogen removal are very plant-specific and the costs per pound
of addition nitrogen removal ranged from a projected savings of $0.79 per pound to a cost of
E-21
-------
Type of
Pollutant
Total
Nitrogen
(IN)
Total
Phosphorus
(TP)
Total
Suspended
Solids
Low
Estimate
High
Estimate
(S per pound removed)
$0.79
~
$0.91
$0.09
$9.64
$270.34
$439.99
$2.72
$0.36
$0.01
—
$5.92
$3.64
$9.53
$2.18
$165.00
$1,179.35
$544.32
$135.17
$34.27
$4.61
$0.25 a
$0.18
Treatment
Type
WWTPs
WWTPs
Ag. Lagoon
Ag. Land Appl.
Ag.(low) to municipal (high)
Large PS facility (0.2 mg/L)
Agricultural BMPs
Ag. Lagoon
Ag. Land Appl.
Ag.(low) to municipal (high)
POTWs (BOD and TSS)
Urban Stormwater Controls
Literature
Sources
Randall etal (1999)
Wiedeman (2000)
Tippett and Dodd (1995)
Tippett and Dodd (1995)
NEWWT 1994
LCBP (1995)
LCBP (1995)
Tippett and Dodd (1995)
Tippett and Dodd (1995)
NEWWT (1994)
USGPO (1986)
USEPA (1999f)
\V VY IJtr S — Wttblc W aLCl ncauiicui, JL. 101110, JL w A »* & * \*vit.-mj VTI***.« «.»«.«.—— -.
"'TSS and BOD removals (1976 dollars). Indexed to $1999, estimated costs are $70 per pound removed.
Full citations are provided in references. Timeframe of dollar values shown vary by source (shown below).
Notes summarize timeframe of analysis, study assumptions (where available), and range of sources/treatment.
Randall etal. (1999): 1995-1998; 6% interest and 20-year capital renewal; BNR retrofits at WWTP only.
NEWWT (1994): 5% interest and 20-year capital renewal; low bound is agricultural BMPs and higher bound is
municipal treatment facilities.
Tippett and Dodd (1995): No discount rate was applied and annual cost equals total lifetime costs adjusted by
design life (varies by practice); "Ag. Lagoon" signifies aerobic lagoon and "Ag. Land" is land application (both
with varying increasing over-application of land applied manure under pre-existing conditions). Cost-effectiveness
values that assume direct discharge of animal wastes are not shown.
LCBP (20001:1995: No discount rate was applied and annual cost equals total lifetime costs adjusted by design life
(varies by practice); low bound is agricultural BMPs and high bound is larger industrial point source.
S5.92 per pound (Randall et al., 1999).12 The range of these estimates is comparatively narrow
given that the study examines a single retrofit category across similar facilities. The time frame
for this analysis ranges from 1995 to 1998 according to the available case study data for each
WWTP. A 20-year capital renewal period is assumed; interest and inflation rates of 6 and 3
percent, respectively, are used (Randall, 2000). The primary emphasis in this study is nitrogen,
12The costs per pound of additional nitrogen removed were flow-weighted to determine the average for each
state and for all plants evaluated.
E-22
-------
since the cost to upgrade for phosphorus removal is both configuration- and site-specific (Randall,
2000).13
Based on this analysis and other data from the Maryland Department of the Environment,
EPA's Chesapeake Bay Program Office has derived a cost-effectiveness value for BNR of $3.64
per pound of nitrogen removed, as shown in Table E-8 (Wiedeman, 1998). Based on this
information and the results of the Randall study, EPA's cost-effectiveness analysis assumes that
an estimated cost to remove nitrogen of less tibtan $4 per pound demonstrates cost-effectiveness.
A number of other studies have been conducted to assess the cost-effectiveness of various
state-level programs to reduce nutrients in Wisconsin (NEWWT, 1994), Vermont (LCBP, 2000),
and North Carolina (Tippett and Dodd, 1995). In Wisconsin, a series of studies were conducted
to compare the cost-effectiveness of point and nonpoint source controls across 41 subwatersheds
in the Fox-Wolf watershed in Wisconsin (NEWWT, 1994). These studies estimated the cost of
reducing phosphorus and suspended solids (TSS) loads from municipal treatment facilities and
agricultural sources. Baseline projections are compared to necessary reductions to meet future
water quality objectives, as mandated by that State's current regulations. The base year for the
analysis is 1990. Phosphorus removal costs'for rural sources are estimated at $9.64 per pound,
whereas municipal treatment facilities are associated with an average annual cost of $165 per
pound of phosphorus removed (NEWWT, 1994).
The Lake Champlain Basin Program (LCBP) conducted a similar study to evaluate costs
to meet Vermont's water quality goals. This study estimated phosphorus removal costs ranging
from $270 to more than $1,000 per pound at a large municipal facility, compared to $440 to $544
per pound of phosphorus removed using agricultural BMPs (LCBP, 2000). The base year for this
analysis is 1995. Another study by the Research Triangle Institute (RTI) assessed the cost-
effectiveness of agricultural BMPs for North Carolina's nutrient trading program. Estimated
costs ranged from $2.72 to $135.17 per pound of phosphorus removed using anaerobic lagoons,
and $0.36 to $34.27 per pound of phosphorus removed for land application practices (Tippett and
Dodd, 1995). Estimated costs reflect the wide range of costs associated with land application,
given pre-existing practices at different types of operations. Costs summarized for this analysis
span 1985 to 1994. Estimated values are shown in Table E-10.
For this analysis, EPA assumes LCBP's estimate of less than $10 per pound to remove
phosphorus from discharge sources. This estimate is a conservative estimate given the range of
estimates in the literature (Table E-10). In addition, observations by researchers at Virginia Tech
who estimated removal costs for nitrogen at WWTPs conclude that it will cost about the same to
remove a pound of phosphorus as it costs to remove a pound of nitrogen, if removing only one
13For conventional plug-flow activated sludge configurations, all that is required for phosphorous removal is
the installation of relatively low-cost baffles and mixers; for oxidation ditches, the addition of an anaerobic reactor
separate from the ditch is needed (Randall, 2000).
: • '• E-23 . -
-------
nutrient. If the facility is upgraded to remove both nitrogen and phosphorus, the cost typically
will be only slightly more than the cost to remove nitrogen alone (Randall, 2000).
EPA's benchmark to compare the potential costs per pound of sediments removed is
EPA's cost reasonableness test established by EPA in developing technology-based effluent limits
for conventional pollutants (see 51 FR 24982). This benchmark measures the cost per pound of
TSS and BOD (biological oxygen demand) removed for an "average" POTW with a flow of 2.26
million gallons per day (USGPO, 1986).14 Indexed to 1999 dollars, these costs are about $0.70
per pound of TSS and BOD removed. EPA used this benchmark to evaluate the estimated cost
per pound of TSS removed by municipalities in a recent EPA rulemaking, which estimated the
range of costs for stormwater controls to be between $0.04 to $0.18 per pound of TSS removed
(USEPA, 1999J). The NEWWT studies also estimated the average cost to reduce TSS, reported
at $0.008 per pound removed from rural land and $4.61 per pound removed at municipal
treatment facilities (NEWWT, 1994).
E.4.2 Cost-Effectiveness Results
Table E-7 presents cost-effectiveness results for the proposed BAT Option under both co-
proposed scenarios, Scenario 4a (two-tier structure) and Scenario 3 (three-tier structure) with
regard to nutrients and sediments. Also shown are the results for alternative regulatory options
that were considered by EPA. Results shown in Table E-12 for select technology options,
including the proposed BAT Option (Option 3 for beef and dairy subcategories, except veal, and
Option 5 for the swine, veal, and poultry subcategories) and Option 3+5 (both Option 3 and 5 for
all subcategories). Options are shown for four CAFO coverage scenarios, including CAFOs with
more than 1,000 AU and CAFOs with more than 500 AU (two-tier structure), and operations
with more than 300 AU, both under Scenario 4b and as defined under Scenario 3 (three-tier
structure). The differences in CAFO coverage provide an upper and lower bound of the analysis
to roughly depict the alternative NPDES scenarios.
The values in Table E-12 are average cost-effectiveness values that reflect the increment
between no regulation and the considered regulatory options. All costs are expressed in pre-tax
1999 dollars. Estimated compliance costs used to calculate the cost-effectiveness of the proposed
regulations includes total estimated costs to CAFOs and offsite recipients of CAFO manure
(Section 5) and costs to the permitting authority (Section 10).
14The technologies used for secondary treatment at POTWs removes both TSS and BOD at the same time.
Estimating only the tons of TSS removed from secondary treatment is not possible.
E-24
-------
Table E-12. Cost-Effectiveness Results bv Select Regulatory Option/Scenario, Nutrients (31999)
Option/
Scenario
>1000 AU
>500AU
"Two-Tier"
>300 AU
"Three-tier"
>300 AU
.(Scenario 4b)
Total
Cost37
(Smill.)
Nitrogen
Removals
(mill. Ibs.)
Avg. C-E
Increm. C-E
($/lb. removed)
Phosphorus
Removals
(mill. Ibs.)
Avg. C-E
Increm. C-E
($/lb. removed)
ELG Option 3 (Beef/Dairy) and 5 (Swine/Veal/Poultry)
$688
$840
$886
$992
136
182
206
206
$5.10
$4.60
$4.30
$4.80
$5.10
$3.30
$1.90
undefined
280
377
425
425
$2.50
$2.20
$2.10
$2.30
ELG Option 1 (AH Subcategories)
>1000 AU
>500 AU
"Two-Tier"
>300 AU
"Three-tier"
>300 AU
(Scenario 4b)
>1000 AU
>500 AU
"Two-Tier"
>300 AU
"Three-tier"
>300 AU
(Scenario 4b)
$362 •
$442
' $474
$509
64
90
106
106
$5.60
$4.90
$4.50
$4.80
$5.60
$3.00
$2.10
undefined
226
303
339
339
$1.60
$1.50
$1.40
$1.50
$2.50
$1.60
$0.90
undefined
$1.60
$1.00
$0.90
undefined
ELG Option 3+5 (All Subcategories)
$1,793
$2,074
$2,147
$2,318
136
182
206
206
$13.20
$11.40
$10.40
$11.20
$13.20
$6.10
$3.00
undefined
280
377
425
425
$6.40
$5.50
$5.10
$5.50
$6.40
$2.90
$1.50
undefined
Options and Scenarios are described in Table 3-1.
''Costs are pre-tax ($ 1999).
Under the proposed BAT Option, EPA estimates an average cost-effectiveness of nutrient
removal at $4.60 per pound (two-tier) or $4.30 per pound (three-tier) of nitrogen removed. For
phosphorus removal, removal costs are estimated at $2.20 or $2.10 per pound of phosphorus
removed under each co-proposed alternative (Table E-12). EPA's estimated cost-effectiveness to
remove nitrogen exceeds EPA's benchmark for nitrogen removal of $4 per pound, but falls within
the estimated range of removal costs for this benchmark (as discussed Section E.4.1). EPA's
E-25
-------
estimated cost-effectiveness to remove phosphorus is lower than the benchmark established for
phosphorus of roughly $10 per pound (discussed in Section E.4.1). Based on these results, EPA
concludes that these values are cost-effective.
As shown in Table E-12, the average cost-effectiveness of each of the co-proposed
alternatives is roughly equivalent or within the range of estimated removal costs compared to the
alternative NPDES scenario 4b that would cover all CAFOs with more than 300 AU. However,
the BAT Option is incrementally the most cost-efficient under the three-tier structure, estimated
at $1.90 per pound of nitrogen removed and $0.90 per pound of phosphorus removed.
Incremental cost-effectiveness values are higher under both the two-tier, estimated at $3.30 and
$2.20 per incremental pound of nitrogen and phosphorus removed, respectively. Compared to the
three-tier structure (Scenario 3), Scenario 4b removes no more pollutants, but has a higher cost
and, therefore, an undefined cost-effectiveness value (Table E-12).
A comparison of the cost-effectiveness of the BAT Option to alternative ELG options
show that estimated average cost-effectiveness of the BAT Option is slightly higher at Ihe 1,000
AU threshold compared to the 500 AU cutoff (Table E-12). Costs and removals are more than
twice as high under the more stringent Option 3+5 for all subcategories (Table E-12). Costs and
removals are lower than under Option 1, but EPA chose not to propose Option 1 because it does
not represent the best available technology (described in Section Vm of the preamble).
Table E-13 shows that the cost to remove sediments under the BAT Option is estimated at
$0.003 per pound of sediment removal under both tier structures. This estimated per-pound
removal cost is low compared to EPA's POTW benchmark for conventional pollutants of about
$0.70 per pound of TSS and BOD removed ($1999). Option 1 results across the range of
NPDES Scenarios are estimated at about $0.05 per-pound removal of sediments (Table E-13).
Tables E-14 and E-15 show the results of EPA's C-E analysis by major subcategory
groupings (cattle, dairy, hogs, and poultry) for the BAT Option under the two-tier and three-tier
structures, respectively. As these tables show, there is.a wide discrepancy in cost-effectiveness
among the sectors. In particular, the cost to remove a pound of nutrients is greatest in the hog
and cattle (includes beef, veal, and heifer) sectors. The average cost-effectiveness of the BAT
Option is more than $20 per pound of nitrogen removed in the hog sector and about $10 per
pound of phosphorus removed in cattle sectors. Removal costs are less than $5 or less per pound
of nitrogen/phosphorus respectively, in Ihe dairy and poultry sectors (Tables E-14 and E-15).
E-26
-------
Table E-13. Cost-Effectiveness Results by Select Regulatory Option/Scenario, Sediments (S1999)
Option/
Scenario
Total
Cost "
(Sm 1999)
Sediments
Removed
(million Ibs.)
Average
Cost-Effectiveness
Incremental
Cost-Effectiveness
(average $ per pound removed)
ELG Option 3 (Beef/Dairy) and 5 (Swine/Veal/Poultry)
>1000 AU
>500 AU "Two-Tier"
>300 AU "Three-tier"
>300 AU (Scenario 4b)
$688
$840
$886
$992
209,050
299,708
335,456
335,456
.$0.003
$0.003
$0.003
$0.003 -
$0.003
$0.002
$0.001
undefined
ELG Option 1 (All Subcategories)
>1000 AU
>500 AU "Two-Tier"
>300 AU "Three-tier"
>30Q AU (Scenario 4b)
$362
$442
$474
$509
7,867
11,856
11,886
11,886
$0.046
$0.037
$0.040
$0.043
$0.046
$0.020
$1.093
undefined
ELG Option 3+5 (All Snbcategories)
>1000AU
>500 AU "Two-Tier"
>300 AU "Three-tier"
>300 AU (Scenario 4b)
$1,793
$2,074 .
$2,147
$2,318
209,050
299,708
335,456
335,456
$0.009
$0.007
$0.006
$0.007
$0.009
$0.003
$0.002
undefined
Options and Scenarios are described in Table 3-1.
^Costs are pre-tax.
•E-27
-------
Table E-14. Cost-Effectiveness Results by Sector under the Two-Tier Structure (Scenario 4a)(S1999)
Sector
Total Cost b/
($ millions)
Estimated "At-Stream" Removals
Sediments
Nitrogen
Phosphorus
(million pounds) **
Average Cost Effectiveness
Sediments
Nitrogen
Phosphorus
($/lb. removed)
ELG Option 3 (Beef/Dairy) and 5 (Swine/Veal/Poultry)
Cattle
Dairy
Hogs
Poultry
$231
$179
$294
$135
14,427
5,050
4,965
275,266
42
38
11
90
23
81
42
231
$0.016
$0.036
$0.059
$0.000
$5.50
$4.80
$26.70
$1.50
$10.00
$2.20
$7.00
$0.60
ELG Option 1 (AH Subcategories)
Cattle
Dairy
Hogs
Poultry
$86
$105
$138
$113
6,775
4,745
70
• 266
12
28
1
50
14
68
' 13
208
$0.013
$0.022
$1.977
$0.427
$7.40
$3.80
$120.70
$2.30
$6.00
$1.50
$10.90
' $0.50
ELG Option 3+5: (All Subcategories)
Cattle
Dairy
Hogs
Poultry
$1,148
$354
$494
$277
14,427
5,050
4,965
275,266
42 .
38
11
90
23
81
42
231
$0.080
$0.070
$0.099
$0.001
$27.10
$9.40
$44.70
$3.10
$49.50
$4.40
$11.80
$1.20
Options and Scenarios are described in Table 3-1. "Cattle" include beef, heifer, and veal operations. Poultry
includes broiler, egg, and turkey operations.
"'Pound-equivalent removals are calculated from removals estimated by EPA's loadings analysis, described in the
Benefits Analysis (USEPA, 2000d) and the Development Document (USEPA, 2000a), adjusting for each
pollutant's toxic weight (as described Section E.3.1).
w Costs are pre-tax and indexed to 1999 dollars using the Construction Cost Index (ENR, 2000).
E-28
-------
Table E-15. Cost-Effectiveness Results by Sector under the Three-Tier Structure (Scenario 3) ($1999)
Sector
Total Cost b/
($ millions)
Estimated "At-Stream" Removals
Sediments
Nitrogen
Phosphorus
(million pounds) **
Average Cost Effectiveness
Sediments
Nitrogen
Phosphorus
($/lb. removed)
ELG Option 3 (Beef/Dairy) and 5 (Swine/Veal/Poultry)
Cattle
. Dairy
Hogs
Poultry
$234
$211
$274
$158
14,434
5,159
5,822
310,041
44
50
12
101
24
98
48
255
$0.016
$0.041
$0.047
$0.001
$5.30
$4.20
$23.60
$1.60
$9.80
$2.20
$5.70
$0.60
ELG Option 1 (All Subcategories)
Cattle
Dairy
Hogs
Poultry
$87
$120
$134
$132
6,778
4,752
79
277
12
37
1
55
15
82
14
228
$0.013
$0.025
$1.699
$0.477
$7.00
$3.30
$108.40
$2.40
$6.00
$1.50
$9.70
$0.60
ELG Option 3+5 (All Subcategories)
Cattle
Dairy
Hogs
Poultry
$1,150
$429
$463
$326
14,434
5,159
5,822
310,041
•44
50
12
101.
24
98
48
255
$0.080
$0.083
$0.080
$0.001
$26.20
$8.60
$39,80
$3.20
$48.20
$4.40
$9.60
$1.30
Options and Scenarips are described in Table 3-1. "Cattle" include beef, heifer, and veal operations. "Poultry"
includes broiler, egg, and turkey operations.
^Pound-equivalent removals are calculated from removals estimated by EPA's loadings analysis, described in the
Benefits Analysis (USEPA, 2000d) and the Development Document (USEPA, 2000a), adjusting for each
pollutant's toxic weight (as described Section E.3.1).
b/Costs are pre-tax and indexed to 1999 dollars using the Construction Cost Index (ENR, 2000).
E-29
-------
------- |