United States
Environmental Protection
Agency
Office Of Air Quality
Planning And Standards
Research Triangle Park, NC 27711
EPA-452/R-01-005
January 2001
FINAL REPORT
Air
Economic Impact Analysis for the Final
Vegetable Oil Processing NESHAP
Final Report
-------
Economic Impact Analysis
for the Proposed Vegetable Oil
Processing NESHAP
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Innovative Strategies and Economics Group, MD-15
Research Triangle Park, NC 27711
Prepared Under Contract By:
Research Triangle Institute
Center for Economics Research
Research Triangle Park, NC 27711
January 2001
-------
This report has been reviewed by the Emission Standards Division of the Office of Air Quality
Planning and Standards of the United States Environmental Protection Agency and approved for
publication. Mention of trade names or commercial products is not intended to constitute
endorsement or recommendation for use. Copies of this report are available through the Library
Services (MD-35), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711, or
from the National Technical Information Services 5285 Port Royal Road, Springfield, VA 22161.
-------
CONTENTS
Section Page
Executive Summary ES-1
1 Introduction 1-1
1.1 Scope and Purpose 1-1
1.2 Organization of the Report 1-2
2 Industry Profile 2-1
2.1 Production 2-2
2.2 Market Data 2-4
2.2.1 Quantity Data 2-4
2.2.2 Baseline and Historical Price Data 2-6
2.3 Affected Producers 2-10
2.3.1 Manufacturing Facilities 2-10
2.3.2 Companies 2-21
2.4 Consumption and Uses of Vegetable Oils and Meals 2-24
2.4.1 Vegetable Oil Consumption and Uses 2-24
2.4.2 Oilseed Meal Consumption and Uses 2-26
3 Engineering Cost Analysis 3-1
3.1 Control Options and Costs 3-2
3.1.1 Control Options 3-2
3.1.2 Control Costs 3-5
3.2 National Emissions Reductions and Compliance Costs 3-8
in
-------
4 Economic Impact Analysis 4-1
4.1 Economic Analysis Inputs 4-1
4.1.1 Producer Characterization 4-1
4.1.2 Vegetable Oil and Meal Markets 4-2
4.1.3 Regulatory Control Costs 4-2
4.2 Economic Impact Methodology 4-5
4.3 Economic Model Assumptions 4-8
4.3.1 Operational Assumptions 4-8
4.3.2 Numerical Assumptions 4-9
4.3.2.1 Elasticity Assumptions 4-9
4.3.2.2 Allocation of Compliance Costs to Oils and Meals . 4-11
4.3.2.3 Allocation of Seed Costs to Oils and Meals 4-11
4.4 Economic Impact Results 4-12
4.4.1 Market-Level Results 4-12
4.4.2 Industry-Level Results 4-12
4.4.3 Social Costs of the Regulation 4-16
5 Small Business Impacts 5-1
6 Assumptions and Limitations of the Economic Model 6-1
References R-l
Appendix A Economic Model of The U.S. Vegetable Oil and Meal Markets A-l
Appendix B Estimates of the Demand and Supply Elasticities for Vegetable Oils
and Meals B-l
Appendix C Economic Model Sensitivity Analysis C-l
IV
-------
LIST OF FIGURES
Number Page
2-1 Simplified Solvent Extraction Process for Vegetable Oils 2-2
2-2 U.S. Consumption of Fats and Oils by Use, 1995 2-26
2-3 U.S. Processed Feeds by Type, 1995 2-28
4-1 Supply Curve for a Representative Affected Facility 4-6
4-2 Market Equilibrium Without and With Regulation 4-7
v
-------
LIST OF TABLES
Number Page
2-1 Summary of Output Shares Relative to Input Volumes (short tons) 2-3
2-2 U.S. Inventories, Production, Foreign Trade, and Apparent Consumption of
Vegetable Oils by Market: 1995 (short tons) 2-5
2-3 Prices of Vegetable Oils, 1990-1996 (cents/lb) 2-7
2-4 Prices of Meal and Similar Products, 1990-1996 (cents/lb) 2-8
2-5 Prices of Oilseeds: 1990-1996 (cents/lb) 2-9
2-6(a) Solvent Extraction Corn Oil Manufacturing Facilities and Locations 2-11
2-6(b) Solvent Extraction Cottonseed Oil Manufacturing Facilities
and Locations 2-12
2-6(c) Solvent Extraction Soybean Oil Manufacturing Facilities
and Locations 2-14
2-6(d) Solvent Extraction Other Vegetable Oil Manufacturing
Facilities and Locations 2-18
2-7 Baseline Vegetable Oil Volumes and Shares by Market and Extraction
Method: 1995 (short tons) 2-20
2-8 Sales and Employment Data for Solvent Extraction Vegetable
Oil Companies 2-22
2-9 Per Capita Consumption of Vegetable Oils: 1990-1996 (Ibs) 2-25
2-10 Per Capita Consumption of Meal Products: 1990-1996 (Ibs) 2-27
3-1 Summary of Control Technologies Assigned to Model Plants 3-4
3-2 Cost Estimates for MACT Floor Control Technologies for
Each Model Plant (1995 dollars) 3-6
3-3 Incremental Cost Estimates for Above-the-MACT-Floor Control
Technologies for Each Model Plant (1995 dollars) 3-9
3-4 Summary of National Emissions and Costs for the MACT Floor and
Above-the- MACT-Floor Control Scenarios (1995 dollars) 3-10
4-1 Summary of Baseline Vegetable Oil and Meal Values: 1995 4-3
4-2 Distribution of Facility-Level Compliance Cost-to-Sales Ratios: 1995 4-4
4-3 Elasticity Estimates Used in the Economic Impact Analysis 4-10
VI
-------
4-4 Summary of Market-Level Impacts of Vegetable Oil
Production NESHAP: 1995 4-13
4-5 Summary of Industry-Level Impacts of Vegetable Oil Production
NESHAP: 1995 4-15
4-6 Summary of Distributional Industry Impacts of Vegetable Oil Production
NESHAP: 1995 4-17
4-7 Distribution of the Social Costs Associated with the Vegetable Oil Production
NESHAP: 1995 4-18
5-1 Summary Data for Small Companies: 1995 5-3
5-2 Capacity and Compliance Cost Comparisons for Small and
Large Companies: 1995 5-4
5-3 Summary of Cost-to-Sales Ratios for the Vegetable Oil
NESHAP: 1995 (%) 5-5
5-4 Distribution of Cost-to-Sales Ratios for the Vegetable Oil
NESHAP: 1995 5-6
5-5 Summary of Company Cost-to-Sales Ratios for Companies That
Own Cottonseed Facilities: 1995 (%) 5-8
5-6 Distribution of Cost-to-Sales Ratios for Companies That Own Cottonseed
Facilities: 1995 5-9
5-7 Summary of Facility Cost-to-Sales Ratios for Cottonseed
Facilities: 1995 (%) 5-10
5-8 Summary of Small Business Impacts of Vegetable Oil Production
NESHAP: 1995 5-12
vn
-------
LIST OF ACRONYMS
2SLS
AC
CBI
CSRs
EIA
FTEs
GDP
HAPs
HHIs
ISEG
LP
LDAR
LPC
MACT
MRR
NAICS
NESHAP
O&M
OAQPS
OLS
RFA
SBA
two-stage least squares
annualized capital investment
confidential business information
cost-to-sales ratios
economic impact analysis
full-time equivalents
gross domestic product
hazardous air pollutants
Herfmdahl-Hirschman indexes
Innovative Strategies and Economics Group
lost production
leak detection and repair
lost production costs (operating costs and foregone profits incurred
while the plant is shut down to install capital)
maximum achievable control technology
monitoring, recordkeeping, and reporting costs
North American Industry Classification System
National Emission Standard for Hazardous Air Pollutants
operating and maintenance costs
Office of Air Quality Planning and Standards
ordinary least squares
Regulatory Flexibility Act
Small Business Administration
IX
-------
SBREFA Small Business Regulatory Enforcement Fairness Act of 1996
SIC Standard Industrial Classification
SRC solvent recovery credit
tpd tons per day
tpy tons per year
USDA U.S. Department of Agriculture
VOCs volatile organic compounds
x
-------
EXECUTIVE SUMMARY
This report analyzes the economic impacts of an air pollution regulation to reduce
emissions of hexane, which is a solvent, generated in the production of crude vegetable oils
and meals. Hexane is a hazardous air pollutant (HAP). This analysis presents the economic
impacts of two regulatory alternatives. The first alternative is the MACT floor, and the EPA
is promulgating this regulatory alternative. The economic impact results are also presented
for an above-the-MACT-floor alternative, and these results are presents for comparison
purposes only.
How do emissions of HAPs occur in the production of vegetable oils and meals?
Emissions of HAPs from the production of vegetable oil and meal originate from the
transfer and storage of solvent (hexane); potential leaks of solvent from piping and tanks;
process vents (solvent recovery section, meal dryer, and meal cooler); and solvent retained in
the crude oil or meal after processing.
Which markets are affected by the regulation?
The affected markets are those for crude soybean oil and meal, crude cottonseed oil
and meal, crude corn oil and corn germ meal, and other types of crude vegetable oil and meal.
Other types include safflower, sunflower, flaxseed, canola, and peanut oils and meals. The
markets for refined vegetables oils are not directly affected.
Which producers will be affected?
In 1995, the baseline year of the analysis, the affected producers are the 106 vegetable
oil processing facilities that produce vegetable oil and meal using a solvent extraction
process. Both new and existing producers will be affected. A total of 31 companies are
identified as owners of these vegetable oil and meal plants.
How many small businesses will be affected?
Based on Small Business Administration (SBA) definitions, 15 small companies
owned and operated 21 facilities, or 20 percent of all solvent extraction facilities in 1995.
ES-1
-------
What are the compliance costs associated with the regulation?
The costs that each facility will incur include capital costs; operating and maintenance
costs; monitoring, recordkeeping, and reporting costs; and lost production costs (operating
costs and lost profits incurred while process changes are implemented). On an annualized
basis, the compliance costs for plants operating in 1995 and for three plants that began
operation in 1996 were estimated at $12.3 million with the maximum achievable control
technology (MACT) floor scenario and $204.6 million with the more stringent above-the-
MACT-floor scenario.
What are the expected emissions reductions as a result of the regulation?
The U.S. Environmental Protection Agency (EPA) estimates that a 25 percent
reduction in emissions will be achieved with the MACT floor scenario, and a 43 percent
reduction in emissions will be achieved with the above-the-MACT-floor scenario.
How large are the compliance costs relative to sales for the entire industry?
Cost-to-sales ratios (CSRs) were calculated at the facility level by dividing the
regulatory compliance costs by facility revenue. For the MACT floor scenario, 104 of the
106 facilities have CSRs below 1 percent, two have CSRs from 1 to 2 percent, and no facility
has a CSR above 2 percent. For the above-the-floor scenario, 17 facilities have CSRs below
1 percent, 44 have CSRs from 1 to 2 percent, and 45 have CSRs above 2 percent.
How do the compliance costs relative to sales compare for small businesses?
Under the floor scenario, average CSRs are 0.30 percent for small companies and
0.04 percent for large companies. Under the above-the-floor scenario, average CSRs are 2.97
percent for small companies and 0.45 percent for large companies.
What are the overall expected effects on prices, output, and revenues?
Under the floor scenario, prices for individual vegetable oils and meals are expected
to increase by one-half of 1 percent or less, output is expected to decline by approximately
one-third of 1 percent or less, and revenues are expected to increase by one-tenth of 1
percent. Under the above-the-floor scenario, prices for vegetable oils and meals are expected
to increase by 2 to 13 percent under the above-the-floor scenario, output is expected to
decline by 1 to 6 percent, and revenues are expected to increase by 4 percent.
ES-2
-------
What are the predicted effects of the regulation on employment in the industry?
Employment is expected to decrease by 12 individuals under the floor scenario and by
350 individuals under the above-the-floor scenario.
Are any facilities predicted to close under the regulation?
No product-line or facility closures are predicted with the floor option. Six
product-line closures and three facility closures are predicted with the above-the-floor option.
Will this regulation pose a significant impact on a substantial number of small entities?
No. Under the floor scenario with lost production costs, the screening analysis
(CSRs) and the market impact analysis do not show a significant impact on a substantial
number of small entities. The potential for negative impacts is greater under the above-the-
floor scenario.
How have economic conditions changed in the affected industries since 1995, the
baseline year of the analysis?
The markets for oilseeds, oils, and meals exhibit a great deal of volatility over time.
Since 1995, the prices of the primary oilseeds and similar inputs used in the production of
vegetable oils and meals generally increased, while the prices of crude vegetables and meals
generally decreased. However, the magnitude of the compliance costs relative to sales did
not increase substantially using 1999 data compared to using 1995 data. In 2000, economic
conditions in these industries have generally improved.
ES-3
-------
SECTION 1
INTRODUCTION
The U.S. Environmental Protection Agency (referred to as EPA or the Agency) is
developing an air pollution regulation for reducing emissions of hexane generated in the
production of crude vegetable oils and related products. These products include crude
soybean oil and meal, crude cottonseed oil and meal, crude corn oil and corn germ meal, and
crude specialty vegetable oils and meals. The regulation does not apply to facilities that
refine crude vegetable oil. EPA's Office of Air Quality Planning and Standards (OAQPS) is
developing the National Emission Standard for Hazardous Air Pollutants (NESHAP) under
Section 112 of the Clean Air Act Amendments of 1990 to limit these emissions. The
Innovative Strategies and Economics Group (ISEG) has developed this economic impact
analysis (EIA) to support the evaluation of impacts associated with the regulatory alternatives
considered for this NESHAP. This report presents economic impacts of the maximum
achievable control technology (MACT) floor regulatory alternative promulgated by the EPA
and economic impacts for an above-the-MACT-floor regulatory alternative for comparison
purposes.
1.1 Scope and Purpose
This report evaluates the economic impacts of pollution control requirements in the
production of vegetable oils and related products that are designed to reduce releases of
hazardous air pollutants (HAPs) into the atmosphere. The Clean Air Act's purpose is to
protect and enhance the quality of the nation's air resources (Section 101(b)). Section 112 of
the Clean Air Act Amendments of 1990 establishes the authority to set national emissions
standards for 189 HAPs. Emissions of HAPs from the production of vegetable oil and meal
originate from the transfer and storage of solvent (hexane); potential leaks of solvent from
piping and tanks; process vents (solvent recovery section, meal dryer, and meal cooler); and
solvent retained in the crude oil or meal after processing (Midwest Research Institute, 1995).
1-1
-------
The NESHAP will apply to all existing and new major sources that manufacture
vegetable oil and related products using solvent extraction processes.1 A major source is
defined as a stationary source or group of stationary sources located within a contiguous area
and under common control that emits, or has the potential to emit, 10 tons or more of any one
HAP or 25 tons or more of any combination of HAPs. In 1995, an estimated 106 processing
facilities produce crude vegetable oil and related products in the United States using solvent
extraction processes. Some of these facilities process multiple oilseed types (e.g., soybean,
cottonseed, corn, safflower, sunflower, canola, peanut). Based on 1995 emissions data, EPA
has determined that all of these facilities are major sources of HAPs.
To reduce emissions of HAPs, the Agency establishes MACT standards. The term
"MACT floor" refers to the minimum control technology on which MACT standards can be
based. For existing major sources, the MACT floor is the average emissions limitation
achieved by the best performing 12 percent of sources (if there are 30 or more sources in the
category or subcategory). The MACT can be more stringent than the floor, considering costs,
nonair quality health and environmental impacts, and energy requirements. The estimated
costs for individual plants to comply with the MACT are inputs into the economic impact
analysis presented in this report.
This report analyzes the economic effects of the MACT standard on existing sources.
The MACT standard is the same for both new and existing soybean plants, which contribute
the majority of HAP releases, but slightly more stringent for new plants that process other
oilseed types. However, the economic impacts of the regulation on new sources of all types
are expected to be minimal. Newly installed equipment is expected to be already in
compliance with the MACT standard and no add-on control equipment will be necessary.
Therefore, this report does not explicitly analyze the impact of the regulation on new sources.
However, because the baseline year of the analysis is 1995, this report describes changes in
economic conditions of the affected industries since that time.
1.2 Organization of the Report
The remainder of this report is divided into five sections that describe the
methodology and present results of this analysis:
'Most vegetable oil production processes use solvent extraction. Mechanical extraction accounts for less than 6
percent of production of vegetable oils.
1-2
-------
• Section 2 provides a summary profile of the production of crude vegetable oils
and related products. It presents data on market volumes and prices,
manufacturing plants, and the companies that own and operate these plants.
• Section 3 reviews the regulatory control options and associated costs of
compliance. This section is based on EPA's engineering analysis conducted in
support of the NESHAP.
• Section 4 details the methodology for assessing the economic impacts of the
NESHAP and the results of the analysis, which include market, industry, and
social cost impacts.
• Section 5 provides the Agency's analysis of the regulation's impact on small
businesses.
• Section 6 describes the assumptions used in this analysis.
In addition to these sections, Appendix A describes the economic model used to predict the
economic impacts of the NESHAP, Appendix B provides information on the elasticities of
demand and supply used in the model, and Appendix C provides the results of sensitivity
analyses on the model assumptions.
1-3
-------
SECTION 2
INDUSTRY PROFILE
Most crude vegetable oil and related products are produced using solvent extraction
processes (affected facilities), although a small proportion is still produced using mechanical
or hydraulic extraction processes (unaffected facilities). The affected products produced by
vegetable oil facilities are classified in the following North American Industry Classification
System (NAICS) codes:
• NAICS 311221, Wet Corn Products and NAICS 311211 Flour and Other Grain
Mill Products—corn oil and corn germ meal;
• NAICS 311222, Soybean Products—soybean oil and soybean meal; and
• NAICS 311223, Other Oilseed Products—oils and meals of cottonseed, canola,
flaxseed, rice, safflower, sunflower, and other oilseeds.
In addition to these primary products, other minor products, such as hulls, linters, and
lecithin, are produced as well.1
This section provides a summary profile of the vegetable oil and related products
industries as background information for understanding the technical and economic aspects
of the industries. Section 2.1 presents a brief overview of the production process.
Section 2.2 provides market data on U.S. production, consumption, foreign trade, and prices.
Section 2.3 describes the affected U.S. processing facilities and the companies that own
them. Finally, Section 2.4 provides data on the consumers and uses of vegetable oils and
related products.
2.1 Production
Figure 2-1 shows a simplified process diagram for vegetable oils and related products.
Oilseeds, such as soybeans and cottonseed, or similar inputs, such as peanuts, rice, and corn
'These minor products are not described as part of this summary profile because available data are insufficient
to characterize them.
2-1
-------
Oilseeds or Similar Inputs
Preparation
Hexane
Recycled
Solvent
-> Hulls
Oilseeds from
Preparation
Oil Extraction
Solvent and Extracted Flakes
Hexane
A
Solvent and Crude Oil
Desolventizing/
Toasting/Drying/
Cooling
T
Solvent Recovery
T
Recycled
Solvent
Cake, Meal, and
Other Products
Crude Oil Product
Figure 2-1. Simplified Solvent Extraction Process for Vegetable Oils
germ, are dehulled, cracked and flaked, and prepared for oil extraction. Hexane is added to
dissolve the oil in the prepared oilseed or similar input and then recovered in a desolventizing
(evaporation) process. Recovered hexane is then recycled for reuse in the process.
Crude oil products produced at these facilities are then transferred to a refining
facility where they are prepared for human consumption. Meal products are either further
processed into a variety of products for human consumption or prepared for use in animal
feeds.
Based on the data, it appears that facilities produce relatively fixed proportions of
their outputs to the oilseeds or similar inputs. Table 2-1 shows the average shares of oil
production and meal production volumes relative to oilseed volumes based on U.S.
Department of Agriculture (USDA) data for the years 1975 through 1996 and on the EPA
facility database for 1995. Soybeans generate an average of 18.2 percent oil and 79.2 percent
meal, with 2.6 percent shrink or waste by weight based on USDA data. These numbers are
2-2
-------
Table 2-1. Summary of Output Shares Relative to Input Volumes (short tons)
USDA (1975 -1996)a
Soybean
Mean
Standard deviation
Cottonseed
Mean
Standard deviation
EPA Facility Database
(1995)b
Corn
Cottonseed
Soybean
All other
Total
Crushed
Volume
29,991,290
7,451,987
3,636,076
482,524
2,477,695
3,794,066
41,920,179
2,601,092
Oil Product
Production
Volume
5,472,871
1,423,299
588,727
94,383
1,066,819
622,308
7,968,264
1,026,335
Share
(%)
18.2%
0.45%
16.2%
0.96%
43.1%
16.4%
19.0%
39.5%
Meal
Production
Volume
23,740,258
5,950,354
1,660,773
281,226
1,401,546
1,744,562
31,225,572
1,497,886
Product
Share
(%)
79.2%
1.16%
45.7%
2.65%
56.6%
46.0%
74.5%
57.6%
a The USD A reports total crushed volumes based on a marketing year beginning September 1. These volumes
have been adjusted to reflect a marketing year beginning October 1 to be consistent with reported oil and
meal production.
b Oil and meal product quantities for seven facilities have been adjusted to be consistent with reported total
crushed volumes.
Source: U.S. Department of Agriculture. 1997c. Oil Crops Situation and Outlook Yearbook. Washington,
DC: Government Printing Office.
similar to the figures provided by David Ailor (1998) of the National Oilseed Processors
Association (18.5 percent oil, 79 percent meal, 2.5 percent shrink and waste) and based on
the EPA facility database (19.0 percent oil; 74.5 percent meal; and 6.5 percent shrink, waste,
and hulls). Based on USDA data, cottonseed generated an average of 16.2 percent oil and
45.7 percent meal, with the remainder going to other products, and based on the EPA facility
database, cottonseed generated an average of 16.4 percent oil and 46.0 percent meal. USDA
does not report comparable figures for corn germ or the other oilseed types. However, based
on the EPA facility database, corn germ is on average 43.1 percent oil and 56.6 percent meal.
All other oilseeds are on average 39.5 percent oil and 57.6 percent meal.
2-3
-------
Because the calculated percentages based on USDA data were fairly constant over
time and because the major trade associations verified that these percentages remain constant,
fixed production proportions are assumed in the EIA. Thus, predicted changes in oil and
meal production due to the effects of the regulation were verified against these percentages as
part of the EIA.
2.2 Market Data
This section presents baseline 1995 data for production, exports, imports, and
apparent consumption of each of the three primary oil products and their associated meal
products, as well as other vegetable oils and meals combined. Because the prices for these
products are volatile, both historical and recent price data are included for the major outputs
and the oilseed inputs.
2.2.1 Quantity Data
Table 2-2 provides baseline 1995 data on production, exports, imports, and apparent
consumption of corn oil, cottonseed oil, soybean oil, all other vegetable oils combined, corn
germ meal, cottonseed meal, soybean meal, and all other meals combined, as reported by the
USDA and Department of Commerce.
In 1995, soybean oil accounted for 74 percent of all vegetable oil production.
Approximately 15 percent of soybean oil production was exported. Even greater percentages
of the other oils were exported: 37 percent of corn oil, 23 percent of cottonseed oil, and
62 percent of all other vegetable oils combined. Small quantities of corn, cottonseed, and
soybean oil were imported, but nearly half of all other vegetable oils combined were
imported. Most of this quantity was canola oil, which is currently produced in only small
quantities in the United States.
As with the oil products, soybean meal made up the majority of meal production,
accounting for 92 percent of all meals combined in 1995. Approximately 19 percent of
soybean meal was exported, compared to 5 percent of cottonseed meal and 13 percent of all
other meals. Production and import data were unavailable for corn germ meal. The United
States imports insignificant quantities of soybean and cottonseed meal but imports
approximately half of all other meals (canola, flaxseed, and sunflower).
2-4
-------
Table 2-2. U.S. Inventories, Production, Foreign Trade, and Apparent Consumption of
Vegetable Oils by Market: 1995 (short tons)
Market
Oil products
Corn oil
Cottonseed oil
Soybean oil
All other vegetable
oilsb
Total, oils
Meal products
Corn germ meal
Cottonseed meal
Soybean meal
All other meals0
Total, meals
All other products'1
Beginning
Inventory
45,091
57,370
527,616
135,042
765,118
NA
94,900
241,117
16,512
352,529
NA
Production
1,121,703
645,925
7,818,128
1,015,792
10,601,548
NA
1,826,100
33,340,037
1,091,571
36,257,708
3,781,800
Imports
4,495
106
13,616
529,125
547,341
NA
0
65,405
936,570
1,001,975
NA
Exports
415,270
147,281
1,143,544
632,625
2,338,719
61,950
89,700
6,491,570
138,469
6,781,688
NA
Ending
Inventory
98,203
47,424
704,447
146,792
996,865
NA
21,200
290,100
16,512
327,812
NA
Apparent
Consumption3
657,816
508,697
6,511,369
900,542
8,578,423
NA
1,810,100
26,864,889
1,889,673
30,564,662
NA
a Apparent Consumption = Beginning Inventory + Production + Imports - Exports - Ending Inventory
b Includes canola, flaxseed, peanut, safilower, and sunflower volumes.
0 Includes canola, flaxseed, and sunflower volumes.
d Includes cottonseed hulls, lecithin, cottonseed linters, and soybean hulls.
Sources: U.S. Department of Agriculture, Economic Research Service. Oil Crops Yearbook, [computer file].
Last updated January 1998.
U.S. Department of Commerce. 1996a. 1995 Current Industrial Reports—Fats and Oils: Oilseed
Crushings. M20J. Washington, DC: Government Printing Office.
U.S. Bureau of the Census. 1998a. U.S. Exports History: Historical Summary 1993-1997 on CD-
ROM [machine readable data file]. Washington, DC: Bureau of the Census.
U.S. Bureau of the Census. 1998b. U.S. Imports History: Historical Summary 1993-1997 on CD-
ROM [machine readable data file]. Washington, DC: Bureau of the Census.
2-5
-------
All other vegetable oil products totaled 3.8 million tons in 1995. These include
products such as hulls, lecithin, and cottonseed linters. Because of insufficient data, these
products are not included in the EIA.
2.2.2 Baseline and Historical Price Data
Historical price data for 1990 through 1999 for crude vegetable oils are presented in
Table 2-3, with the 1995 baseline year of analysis in boldface. While the prices for corn,
cottonseed, and soybean enter the model individually, the prices of canola, flaxseed, peanut,
safflower, and sunflower are combined into a weighted average price for all other vegetable
oils (see Table 4-1). Prices of oil tend to fluctuate greatly from year to year and appear to
have peaked in 1994 for many of the oils and then fallen in 1995 and each year since then.
The decrease in prices is attributable primarily to changes in the international markets for oil.
In particular, crushing capacities in South America and Europe have expanded, thus reducing
the demand for vegetable oil exports to these countries (USDA, 1999c).
Prices for oilseed meal and similar products are listed in Table 2-4 for 1990 through
1999, with the baseline 1995 data again in boldface. As with oil prices, these prices fluctuate
greatly from year to year. For these products, prices appear to have peaked in 1993, fallen in
1994 and 1995, peaked again in 1996 and 1997, and then fallen drastically in 1998 and 1999.
As with the prices for vegetable oils, these decreases are most likely attributable to a
reduction in export demand for meals. Thus, for both vegetable oils and meals, prices
received by vegetable oil processors were higher in 1995 than in recent years.
A large percentage of the costs of producing vegetable oils and meals is the cost of
the raw agricultural inputs. In Table 2-5, their prices are presented for 1990 through 1999,
with baseline 1995 data in boldface. Because these are agricultural commodities, acres
planted and weather conditions influence the output in any given year; thus, prices tend to be
volatile. In 1995, prices of oilseeds and similar inputs were substantially higher than in 1999
with the exception of cottonseed. Thus, while output prices for most vegetable oils and
meals have fallen recently for most oilseed types, the cost of the primary input has fallen also.
The situation for cottonseed is different than for the other oilseeds because cottonseed is
being used increasingly as a dairy cow feed.
2-6
-------
Table 2-3. Prices of Vegetable Oils, 1990-1996 (cents/lb)
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
Corn
25.40
28.40
24.00
21.80
27.30
26.60
26.50
24.85
30.33
23.36
Cottonseed
23.90
20.70
21.40
26.00
27.10
26.80
25.90
26.51
31.03
23.95
Soybean
23.40
20.30
19.30
22.70
27.90
26.80
23.80
23.27
25.73
17.60
Canola"
24.40
21.30
20.30
23.70
28.90
27.80
24.80
24.27
26.73
18.60
Flaxseed
40.10
34.50
30.70
31.70
32.50
35.00
37.10
36.25
36.00
36.00
Peanut
45.70
38.06
25.03
34.10
45.91
41.57
40.20
47.20
47.21
38.25
Safflower
55.10
49.20
60.00
70.00
59.00
59.00
59.00
59.00
59.00
59.00
Sunflower
22.10
23.40
22.90
26.80
31.10
28.90
24.66
23.45
24.24
19.00
a USDA does not report a crude canola oil price; thus, it was approximated as one cent per pound over the soybean oil price (Marine, 1999).
Sources: U.S. Department of Agriculture. 2000. Agricultural Statistics 2000. Washington, DC: Government Printing Office.
U.S. Department of Agriculture, Economic Research Service. 1999b. Oil Crops Yearbook, [computer file]. Last updated
November 1999.
U.S. Department of Agriculture, Economic Research Service. 2000. Oil Crops Outlook. September 13, 2000.
U.S. Department of Agriculture. 1997a. Agricultural Statistics 1997. Washington, DC: Government Printing Office.
U.S. Department of Agriculture. 1992. Agricultural Statistics 1992. Washington, DC: Government Printing Office.
U.S. Department of Agriculture, Economic Research Service. 1998. Oil Crops Yearbook, [computer file]. Last updated January
1998.
-------
Table 2-4. Prices of Meal and Similar Products, 1990-1996 (cents/lb)
Corn
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
to NA = Not available.
oo
a /~i _.__._ j _ _i i _ _ _i _i • _ _ i
Germ Meal3
4.76
5.09
5.18
4.39
4.48
4.42
5.82
4.20
3.25
3.11
Cottonseed
7.79
6.74
7.23
8.29
7.53
6.16
10.01
9.42
6.22
5.54
. -t t .
Soybean
9.08
9.21
9.38
9.94
9.13
8.69
12.33
13.32
8.14
7.08
Sunflower
4.57
4.36
3.96
4.51
4.37
3.62
6.33
5.32
3.74
3.33
Flaxseed
6.63
6.37
6.40
7.03
6.00
5.20
8.28
7.71
5.15
4.37
Peanut
NA
NA
NA
NA
8.40
6.84
10.04
10.99
9.36
4.99
Computed by adding $7 per short ton to reported corn gluten feed price (Brenner, 1999).
Sources: U.S. Department of Agriculture, Economic Research Service. 1997. Feed Yearbook, [computer file]. Last updated April 1997.
U.S. Department of Agriculture, Economic Research Service. 2000. Feed Yearbook, [computer file]. Last updated May 2000.
U.S. Department of Agriculture, Economic Research Service. 1998. Oil Crops Yearbook, [computer file]. Last updated January
1998.
U.S. Department of Agriculture, Economic Research Service. 1999b. Oil Crops Yearbook, [computer file]. Last updated November
1999.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook. October 14, 1997.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook. November 13, 1996.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook. March 13, 1996.
U.S. Department of Agriculture, Economic Research Service. 2000. Oil Crops Outlook. September 13, 2000.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook. September 13, 1999.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook. November 12, 1998.
-------
Table 2-5. Prices of Oilseeds: 1990-1996 (cents/lb)
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
Corn Germ3
10.92
12.21
10.32
9.37
11.74
11.44
11.40
10.69
13.04
10.04
Cottonseed
6.01
4.41
4.59
5.72
5.14
4.96
6.03
6.07
5.99
4.99
Soybean
9.70
9.33
9.26
10.07
10.18
9.75
12.13
12.40
10.08
7.61
Peanuts
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Sunflower
11.53
10.19
9.10
11.63
12.97
10.83
12.37
11.39
12.51
9.15
Flaxseed
12.25
7.31
6.71
7.59
8.03
9.02
10.62
10.87
10.47
7.82
Canola
NA
NA
9.84
10.57
11.03
11.10
12.30
11.83
10.63
8.65
Safflower
NA
NA
13.10
14.83
14.80
14.60
16.93
16.30
14.60
13.87
Rice
6.90
7.34
7.03
5.98
8.22
7.62
9.59
9.99
9-10
9-10
to NA = Not available.
a Corn germ price is computed as follows: 0.43 x corn oil price (Brenner, 1999).
Sources: U.S. Department of Agriculture. 1991 a. Agricultural Statistics 1997. Washington, DC: Government Printing Office.
U.S. Department of Agriculture. 1992. Agricultural Statistics 1992. Washington, DC: Government Printing Office.
U.S. Department of Agriculture, Economic Research Service. 1998. Oil Crops Yearbook, [computer file]. Last updated January
1998.
U.S. Department of Agriculture, Economic Research Service. 1999b. Oil Crops Yearbook, [computer file]. Last updated
November 1999.
U.S. Department of Agriculture. 1997b. Agricultural Prices: 1996 Summary. Washington, DC: Government Printing Office.
U.S. Department of Agriculture, National Agriculture Statistics Service. 2000. Agricultural Prices: 1999 Summary. Washington,
DC: Government Printing Office.
U.S. Department of Agriculture, Economic Research Service. 1998. Rice Yearbook, [computer file]. Last updated November 1998.
U.S. Department of Agriculture, Economic Research Service. 2000. Oil Crops Outlook. September 13, 2000.
-------
2.3 Affected Producers
The following section briefly describes vegetable oil processing facilities and the
companies that own them. It also presents the information used to determine the proportion
of products produced by affected solvent extraction facilities versus unaffected mechanical
extraction facilities.
2.3.1 Manufacturing Facilities
Tables 2-6(a) through 2-6(d) provide information on the facilities that produced crude
vegetable oils and meals in the baseline year 1995 and that will be affected by the NESHAP.
In addition, the tables indicate which facilities have closed since 1995 and list new facilities
that have begun operations since 1995. All of these facilities use solvent extraction processes
and are major sources of HAPs. The facilities are organized by the following product
categories:
• corn oil (as represented by NAICS 311221 Wet Corn Products, and NAICS
311211 Flour and Other Grain Mill Products)—As shown in Table 2-6(a), six
companies owned and operated eight facilities producing corn oil in 1995. In
addition, one corn oil facility also produces safflower oil. Since 1995, one facility
has closed.
• cottonseed oil (included in NAICS 311223 Other Oilseed Products)—As shown
in Table 2-6(b), 12 companies owned and operated 25 cottonseed oil facilities in
1995. In addition, one cottonseed oil facility also produces safflower oil, two
facilities also produce peanut oil, and one facility also produces corn oil. Since
1995, ten cottonseed oil facilities have closed or become dormant and three new
facilities have opened.
• soybean oil (as represented by NAICS 311222 Soybean Products)—As shown in
Table 2-6(c), 13 companies own and operate 62 soybean oil facilities. Since
1995, four soybean oil facilities have closed or become dormant and nine new
soybean oil facilities have opened.
• minor vegetable oils (included in NAICS 311223 Other Oilseed Products)—This
classification includes all other producers of vegetable oils, including canola,
flaxseed, peanut, rice, safflower, and sunflower oils. As shown in Table 2-6(d),
six companies owned and operated 11 facilities. Five of these facilities produce
more than one type of vegetable oil product. Since 1995, two facilities have
ceased operations.
2-10
-------
Table 2-6(a). Solvent Extraction Corn Oil Manufacturing Facilities and Locations: 1995
Company Name
Archer Daniels Midland
Bunge Corporation
Cargill Incorporated
CPC International
Mitsubishi Corporation
Tate and Lyle PLC
Facility Name
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Bunge Corp.a
Cargill Inc.
Cargill Inc.
CPC International
California Oils
A.E. Staleyb
Facility Location Other Types Produced
Clinton
Decatur
Danville
Eddyville
Memphis
Bedford Park
Richmond
Loudon
IA
IL
IL
IA
TN
IL
CA Safflower
TN
a Also produces corn oil using a mechanical extraction process.
b Dormant or closed after 1995.
Source: Ailor, David C., National Oilseed Processors Association. July 25, 2000. "Comments of the Vegetable Oil MACT Coalition on the
National Emission Standards for Hazardous Air Pollutants: Solvent Extraction for Vegetable Oil Production, 40 C.F.R. Part 63
Subpart GGGG, Air Docket No. A-97-59." Memorandum.
National Cotton Council of America. July 25, 2000. "Comments of the National Cotton Council and National Cottonseed Products
Association on the Proposed National Emission Standards for Hazardous Air Pollutants: Solvent Extraction for Vegetable Oil
Production (65 FR 34252; May 26, 2000) (Air Docket No. A-97-59). Memorandum.
-------
Table 2-6(b). Solvent Extraction Cottonseed Oil Manufacturing Facilities and Locations: 1995
to
Company Name
Archer Daniels Midland
Chickasha Cotton Oil Mill
Delta Oil Mill
Dunavant Enterprises
Hartsville Oil Mill Incorporated
J.G. Boswell
Osceola Products
Plains Cooperative Oil Mill Incorporated
Planter's Cotton Oil Mill
Producers Cooperative Mill
Facility Name
Southern Cotton Oil Co.
Southern Cotton Oil Co.
Southern Cotton Oil Co.
Southern Cotton Oil Co.
Southern Cotton Oil Co.
Southern Cotton Oil Co.a
Southern Cotton Oil
Chickasha Cotton Oila
Clinton Cotton Oil Milla
Lamesa Cotton Oil Mill
Rio Grande Oil Milla
Delta Oil Mill
Anderson Claytona
Anderson Claytona
Hartsville Oil Mill
J.G. Boswell
Osceola Products Co.a
Osceola Products Co.a
Plains Co-op Oil Mill
Planter's Cotton Oil Mill Inc.
Producers Cooperative Oil Mill
Facility Location Other Types Produced
Memphis
Port Gibson
Lubbock
Levelland
N Little Rock
Quanah
Sweetwater
Casa Grande
Clinton
Lamesa
Harlingen
Jonestown
Phoenix
Chowchilla
Darlington
Corcoran
Kennett
Osceola
Lubbock
Pine Bluff
Oklahoma City
TN
MS
TX
TX
AR
TX
TX
AZ
OK
TX
TX
MS
AZ
CA
SC
CA
MO
AR
TX
AR
OK
Corn
Peanut
Peanut
Safflower
(continued)
-------
Table 2-6(b). Solvent Extraction Cottonseed Oil Manufacturing Facilities and Locations: 1995 (Continued)
Company Name
Valley Cooperative Mills
Yazoo Valley Oil Mill, Incorporated
New Facilities Opened Since 1995
Alimenta
Archer Daniels Midland
Chickasha Cotton Oil Mill
Facility Name
Valley Co-op Oil Mill
Yazoo Valley Oil Mill3
Yazoo Valley Oil Mill
Yazoo Valley Oil Mill"
Alimenta
Southern Cotton Oil
Chickasha Cotton Oil
Facility Location Other Types Produced
Harlingen
Helena
Greenwood
West Monroe
Vienna
Richmond
Tifton
TX
AR
MS
LA
GA Peanut
TX
GA
to a Dormant or closed after 1995.
oo
Source: Ailor, David C., National Oilseed Processors Association. July 25, 2000. "Comments of the Vegetable Oil MACT Coalition on the
National Emission Standards for Hazardous Air Pollutants: Solvent Extraction for Vegetable Oil Production, 40 C.F.R. Part 63
Subpart GGGG, Air Docket No. A-97-59." Memorandum.
National Cotton Council of America. July 25, 2000. "Comments of the National Cotton Council and National Cottonseed Products
Association on the Proposed National Emission Standards for Hazardous Air Pollutants: Solvent Extraction for Vegetable Oil
Production (65 FR 34252; May 26, 2000) (Air Docket No. A-97-59). Memorandum.
-------
Table 2-6(c). Solvent Extraction Soybean Oil Manufacturing Facilities and Locations: 1995
to
Company Name
Ag Processing
Archer Daniels Midland
Facility Name
Ag Processing Inc.
Ag Processing Inc.
Ag Processing Inc.
Ag Processing Inc.
Ag Processing Inc.
Ag Processing Inc.
Ag Processing Inc. Assoc.
Archer Daniels Midland Processing
Archer Daniels Midland Soybean
Processing
Archer Daniels Midland Co.
Archer Daniels Midland Co.a
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Archer Daniels Midland Co.b
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Archer Daniels Midland Co.
Facility
Eagle Grove
Location Other Types Produced
IA
Sergeant Bluff IA
Mason City
St Joseph
Manning
Dawson
Sheldon
Mankato
Kansas City
Des Moines
Decatur
Lincoln
Frankfort
Mexico
Fremont
Kershaw
Clarksdale
Fostoria
Galesburg
Fredonia
Little Rock
Taylorville
Valdosta
IA
MO
IA
MN
IA
MN
MO
IA
IL
NE
IN
MO
NE
SC
MS
OH
IL
KS
AR
IL
GA
(continued)
-------
Table 2-6(c). Solvent Extraction Soybean Oil Manufacturing Facilities and Locations: 1995 (Continued)
Company Name
Facility Name
Facility Location
Other Types
Produced
Bunge Corporation
Cargill Incorporated
to
Bunge Corp.
Bunge Corp.
Bunge Corp.
Bunge Corp.
Bunge Corp.
Bunge Corp. Soybean Processing
Bunge Corp.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc.
Cargill Inc. Protein Products
Decatur AL
Marks MS
Vicksburg MS
Cairo IL
Destrehan LA
Emporia KS
Danville IL
Fayetteville NC
Sidney OH
Sioux City IA
Raleigh NC
Guntersville AL
Des Moines IA
Chesapeake VA
Iowa Falls IA
Bloomington IL
Kansas City MO
Wichita KS
Gainesville GA
Cedar Rapids IA
Lafayette IN
Cedar Rapids IA
(continued)
-------
Table 2-6(c). Solvent Extraction Soybean Oil Manufacturing Facilities and Locations: 1995 (Continued)
to
Company Name
Central Soya Company
Harvest States Cooperative
Moorman Manufacturing
Owensboro Grain Company
Perdue Farms
Riceland Foods Incorporated
Rose Acre Farm Incorporated
Southern Soya Corporation
Townsends
New Facilities Opened Since 1995
Ag Processing
Ag Processing
Bunge Corporation
Central Soya Company
Consolidated Grain and Barge
Facility Name
Central Soya Co.
Central Soya Co.
Central Soya Co.
Central Soya Co.
Central Soya Co.
Honeymead Processing/Refining
Moorman Manufacturing Co.0
Quincy Soybean Co.b
Quincy Soybean Co.0
Owensboro Grain Co.
Perdue Farms Inc.
Perdue Farms Inc.
Riceland Foods Inc.
Rose Acre
Southern Soya Corp.b
Townsendsb
Ag Processing
Ag Processing
Bunge Corporation
Central Soya Co.
Consolidated Grain and Barge
Facility Location Other Types Produced
Decatur
Gibson City
Bellevue
Delphos
Marion
Mankato
Quincy
Helena
Quincy
Owensboro
Cofield
Salisbury
Stuttgart
Seymour
Estill
Millsboro
Hastings
Emmetsburg
Council Bluffs
Morristown
Mt. Vernon
IN
IL
OH
OH
OH
MN
IL
AR
IL
KY
NC
MD
AR
IN
SC
DE
NE
IA
IA
IN
IN
(continued)
-------
Table 2-6(c). Solvent Extraction Soybean Oil Manufacturing Facilities and Locations: 1995 (Continued)
Company Name
New Facilities Opened Since 1995
(continued)
CF Processing
Incobrasa
South Dakota Soybean Processors
Zeeland Farm Soya
Facility Name
CF Processing
Incobrasa
South Dakota Soybean Processors
Zeeland Farm Soya
Facility Location
Creston IA
Oilman IL
Volga SD
Zeeland MI
Other Types Produced
a Two facilities are listed at this location.
b Dormant or closed since 1995.
0 Currently owned by ADM.
Source: Ailor, David C., National Oilseed Processors Association. July 25, 2000. "Comments of the Vegetable Oil MACT Coalition on the
^ National Emission Standards for Hazardous Air Pollutants: Solvent Extraction for Vegetable Oil Production, 40 C.F.R. Part 63
o Subpart GGGG, Air Docket No. A-97-59." Memorandum.
National Cotton Council of America. July 25, 2000. "Comments of the National Cotton Council and National Cottonseed Products
Association on the Proposed National Emission Standards for Hazardous Air Pollutants: Solvent Extraction for Vegetable Oil
Production (65 FR 34252; May 26, 2000) (Air Docket No. A-97-59). Memorandum.
-------
Table 2-6(d). Solvent Extraction Other Vegetable Oil Manufacturing Facilities and Locations: 1995
Company Name
Archer Daniels Midland
Cargill Incorporated
Lubrizol Corporation
Oilseeds International
to Rito Partnership
oo Sessions Company
Facility Name
Archer Daniels Midland Co.
Archer Daniels Midland Co.a
Archer Daniels Midland Co.
Northern Sun
Northern Sun
Cargill Inc.
Stevens Industries
SVO Specialty Products
Oilseeds International8
Rito Partnership15
Sessions Company
Facility
Velva
Augusta
Red Wing
Enderlin
Goodland
West Fargo
Dawson
Culbertson
Grimes
Stuttgart
Enterprise
Location
ND
GA
MN
ND
KS
ND
GA
MT
CA
AR
AL
Types Produced
Canola
Canola and peanut
Flaxseed and sunflower
Sunflower
Sunflower
Flaxseed, sunflower
Canola and peanut
Canola, safflower, and
sunflower
Safflower
Rice
Peanut
a Dormant or closed after 1995.
b The rice oil facility will not be subject to the regulation. However, its production volumes are included in the total for the other vegetable oil
types to protect confidentiality.
Source: Ailor, David C., National Oilseed Processors Association. July 25, 2000. "Comments of the Vegetable Oil MACT Coalition on the
National Emission Standards for Hazardous Air Pollutants: Solvent Extraction for Vegetable Oil Production, 40 C.F.R. Part 63
Subpart GGGG, Air Docket No. A-97-59." Memorandum.
National Cotton Council of America. July 25, 2000. "Comments of the National Cotton Council and National Cottonseed Products
Association on the Proposed National Emission Standards for Hazardous Air Pollutants: Solvent Extraction for Vegetable Oil
Production (65 FR 34252; May 26, 2000) (Air Docket No. A-97-59). Memorandum.
-------
Many cottonseed facilities in particular have ceased operations in the past few years
because of changes in the market for cottonseed. The feed value of cottonseed has risen
relative to the value of oil and meal products processed from cottonseed (USDA, 1997b).
Thus, the price of cottonseed has risen, making cottonseed oil and meal production less
profitable. Facilities owned by small businesses have been particularly affected; of the ten
cottonseed facilities that have closed, seven are owned by small businesses. However, three
new cottonseed facilities have also opened since 1995.
Sales and employment information is not included in Tables 2-6(a) through 2-6(d)
because these data are confidential business information (CBI). For use in the EIA model,
sales at the facility level were calculated by multiplying the quantities produced at each
facility, which is CBI, by the average prices reported by USDA. Facility-level employment
data were available directly as CBI.
In addition to these affected facilities, some facilities in the industry produce
vegetable oil and meal products using mechanical extraction processes. Because of a lack of
data on these unaffected facilities, they were modeled as one aggregate unaffected facility for
each type of vegetable oil.
Table 2-7 presents the 1995 baseline data on affected and unaffected product
volumes. These data are used to determine the production volume of the industry attributable
to the representative unaffected facility. Total 1995 production volumes by type were
obtained from the USDA (see Table 2-2). The volume produced by affected facilities was
obtained by adding the production volumes of the affected facilities in the EPA facility
database. The volume produced by unaffected facilities was obtained by subtracting affected
facility volume from total volume reported by the USDA. For soybean oil and several of the
individual all other oil and meal products, production volume in the EPA facility database
exceeded USDA reported production. In these cases, the facility database volumes were used
as the baseline market values rather than the USDA reported volumes. Because the soybean
oil production volume in the EPA database was assumed to be the market volume, this
analysis assumes there are no unaffected soybean facilities.2
2Because oil and meal are complementary outputs, this analysis also assumes that the soybean meal volume in
the EPA database is the market volume. However, USDA reports a higher volume of soybean meal
production than the EPA database.
2-19
-------
Table 2-7. Baseline Vegetable Oil Volumes and Shares by Market and Extraction Method: 1995 (short tons)
to
to
o
Solvent Extraction
Market
Oil products
Corn oil
Cottonseed oil
Soybean oil
All other vegetable oilsb
Total, oils
Meal products
Corn germ meal
Cottonseed meal
Soybean meal
All other mealsb
Total, meals
Total
Volume
1,121,703
645,925
7,968,264
1,093,411
10,829,303
1,473,651
1,826,100
31,225,572°
1,583,559
38,151,242
Number of
Facilities Volume
9
25
62
15
106
9
25
62
15
106
1,066,819
622,308
7,968,264
1,026,335
10,683,726
1,401,546
1,744,562
31,225,572
1,497,886
35,869,566
Volume
Share (%)
95.1%
96.3%
100.0%
93.9%
98.7%
95.1%
95.5%
100.0%
94.6%
94.0%
Mechanical Extraction
Number of Volume
Facilities3 Volume Share (%)
NA
NA
NA
NA
NA
NA
NA
NA
NA
54,884
23,617
0
67,076
145,577
72,105
81,538
0
85,673
2,281,676
4.9%
3.7%
0.0%
6.1%
1.3%
4.9%
4.5%
0.0%
5.4%
6.0%
a Modeled as a representative plant.
b Includes canola, flaxseed, peanut, rice, safflower, and sunflower volumes.
0 In the economic impacts model, the volume of meal produced by solvent extraction plants was assumed to be the total market volume for
consistency with soybean oil.
NA = not available.
Source: U.S. Department of Agriculture, Economic Research Service. 1998. Oil Crops Yearbook, [computer file]. Lastupdated January 1998.
-------
2.3.2 Companies
A total of 31 companies were identified as owners of vegetable oil manufacturing
plants using the solvent extraction method in 1995. Table 2-8 lists these companies.3 In
addition to the number of facilities owned during 1995, information on sales and employment
at the company level is included as well.4 Archer Daniels Midland (31 facilities) and Cargill
Incorporated (19 facilities) own the largest number of these facilities (47.2 percent of total
solvent extraction facilities).
Firm size is likely to be a factor in the distribution of the impacts of the NESHAP on
companies. Grouping the firms by size facilitates the analysis of small business impacts as
required by the Regulatory Flexibility Act (RFA) of 1982 as amended by the Small Business
Regulatory Enforcement Fairness Act of 1996 (SBREFA). Firms are grouped into small and
large categories using Small Business Administration (SBA) general size standard definitions
for NAICS codes. These size standards are provided either by number of employees or by
annual receipt levels, depending on NAICS code. The SBA defines a small business for
industries affected by this regulation as follows:
• Corn Oil (NAICS 311221)—fewer than 750 total employees and (NAICS
311211)—fewer than 500 employees;
• Cottonseed Oil (NAICS 311223)—fewer than 1,000 total employees;
• Soybean Oil (NAICS 311222)—fewer than 500 total employees; and
• All Other Vegetable Oils (NAICS 311223)—fewer than 1,000 total employees.
Based on these definitions, 15 companies can be classified as small businesses or potentially
small. Two firms do not have employment data available and are included as potentially
small businesses. As of 1995, these 15 companies owned and operated 21 facilities, or
20 percent of all solvent extraction facilities.
3In cases where sales and employment data were not available, the EPA facility information in the CBI file was
used in the EIA based on the assumption that each company owns only the facilities identified therein.
"Sales and employment data were obtained from publicly available sources and reflect the most recently
available information.
2-21
-------
Table 2-8. Sales and Employment Data for Solvent Extraction Vegetable Oil Companies Included in the 1995
Baseline
to
to
to
Company Name
Ag Processing
Archer Daniels Midland Company
Bunge Corporation
Cargill Incorporated
Central Soya Company
Chickasha Cotton Oil Company
CPC International
Delta Oil Mill
Dunavant Enterprises3
Hartsville Oil Mill Incorporated
Harvest States Cooperative15
J.G. Boswell
Lubrizol Corporation0
Mitsubishi Corporationd
Moorman Manufacturing6
Oilseeds International
Osceola Products
Owensboro Grain Company
Perdue Farms
Plains Cooperative Oil Mill,
Incorporated
Planter's Cotton Oil Mill
Producers Cooperative Mill
Riceland Foods Incorporated
Rito Partnership
Organization
Type
Private
Public
Private
Private
Private
Private
Public
Private
Private
Private
Private
Private
Public
Foreign
Private
NA
Private
Private
Private
Private
Private
Private
Private
NA
Number of
Facilities
7
31
8
19
5
4
1
1
2
1
1
1
1
1
o
3
i
2
1
2
1
1
1
1
1
Sales
($ million)
$1,370.0
$13,314.0
$2,570.0
$62,570.0
$1,000.0
$93.0
$9,844.0
$22.5
$720.0
$20.0
$1,000.0
$80.0
$1,600.0
$166,300.0
$800.0
NA
$438.0
$450.0
$2,000.0
$128.0
$35.0
$35.0
$807.6
NA
Total
Employment
3,000
14,811
3,000
73,000
1,200
600
55,300
90
2,000
100
2,400
1,000
4,358
36,000
3,500
NA
189
195
19,000
108
100
100
2,000
NA
Year
Reported
1995
1996
1996
1996
1994
1995
1996
1996
1995
1996
1996
1993
1996
1996
1996
NA
1996
1996
1996
1995
1996
1996
1996
NA
Small
Business
No
No
No
No
No
Yes
No
Yes
No
Yes
No
Yes
No
No
No
Yes
Yes
Yes
No
Yes
Yes
Yes
No
Yes
(continued)
-------
Table 2-8. Sales and Employment Data for Solvent Extraction Vegetable Oil Companies Included in the 1995
Baseline (continued)
to
to
oo
Company Name
Rose Acre Farm Incorporated
Sessions Company
Southern Soya Corporation
TateandLylePLCf
Townsends
Valley Cooperative Mills
Yazoo Valley Oil Mill, Incorporated
Total
Organization
Type
Private
Private
Private
Foreign
Private
Private
Private
Number of
Facilities
1
1
1
1
1
1
o
5
106
Sales
($ million)
$152.0
$30.0
NA
$7,315.4
$270.0
$32.0
$113.7
NA
Total
Employment
900
100
89
17,743
3,000
100
300
NA
Year
Reported
1996
1994
1996
1996
1993
1992
1996
Small
Business
No
Yes
Yes
No
No
Yes
Yes
15
NA = not available
Note: The Small Business Administration (SBA) defines a small business for industries affected by this regulation as follows:
Corn Oil (NAICS 311221) = fewer than 750 total employees and 500 employees for NAICS 311211.
Cottonseed Oil (NAICS 311223) = fewer than 1,000 total employees.
Soybean Oil (NAICS 311222) = fewer than 500 total employees
All Other Vegetable Oils (NAICS 311223) = fewer than 1,000 total employees.
a Owns Anderson Clayton. Queensland Cotton Holdings Limited acquired Anderson Clayton in September 1997.
b Owns Honeymead Processing.
c Owns SVO Specialty Products.
d Owns California Oils.
e Owns Quincy Soybeans. Archer Daniels Midland Company acquired Moorman in late 1997.
f Owns A.E. Staley.
Sources: 1997 Directory of Corporate Affiliations. 1997. Vol.5: International Public and Private Companies. New Providence, RI: National
Register Publishing.
Dun & Bradstreet. 1998. Dun &Bradstreet Market Identifiers [computer file]. New York, NY: Dialog Corporation.
Hoover's Incorporated. 1998. Hoover's Company Profiles. Austin, TX: Hoover's Incorporated, .
Information Access Corporation. 1997. Business Index [computer file]. Foster City, CA: Information Access Corporation.
Mitsubishi Corporation. Annual Report 1996. .
Standard and Poor's Corporations [computer file]. Palo Alto, CA: Dialog Information Service.
-------
2.4 Consumption and Uses of Vegetable Oils and Meals
Vegetable oils are consumed in both edible and inedible products. Oilseed meal
products are consumed both as products for human consumption and as animal feeds. This
section describes consumption and uses of each type of product.
2.4.1 Vegetable Oil Consumption and Uses
In Table 2-9, per capita consumption of corn, cottonseed, soybean, and other
vegetable oils is provided for 1990 through 1999, with the baseline 1995 data in boldface.
Per capita consumption of most vegetable oils has been relatively stable over this time period.
However, soybean oil consumption has been steadily increasing at an average annual growth
rate of 2 percent, and canola oil consumption more than doubled, with an average annual
growth rate of 10 percent. Soybean oil consumption is by far the highest, at nearly 60 pounds
per capita per year in 1999. Corn, cottonseed, and canola oil consumption quantities are each
a few pounds per year, and flaxseed, peanut, safflower, and sunflower consumption quantities
are for the most part each less than 1 pound per year.
In 1995, the baseline year of the analysis, approximately 71.2 percent of all fats and
oils were consumed in edible products, and 28.8 percent were consumed in inedible products.
As Figure 2-2 illustrates, the edible uses include baking and frying fats (28.5 percent), salad
or cooking oil (31.9 percent), margarine (8.7 percent), and other edible products
(2.0 percent). The most significant inedible product uses are animal feed (11.1 percent) and
fatty acids (9.3 percent), but inedible product uses also include soap, paint and varnish, resins
and plastics, lubricants, and other products.
The vegetable oils affected by the regulation make up an estimated 64 percent of the
21 billion pounds of consumption of all fats and oils. In terms of edible uses, these vegetable
oils are often preferred to their substitutes because they have low saturated fat content.
However, functional characteristics of the oils, such as melting behavior, crystal structure,
resistance to oxidation, and flavor, affect preferences as well. Edible substitute products
include coconut oil, palm oil, palm kernel oil, edible tallow, butter, and lard.
2-24
-------
Table 2-9. Per Capita Consumption of Vegetable Oils: 1990-1999 (Ibs)
Calendar Year
1990
1991
1992
1993
1994
1995
1996
1997
1998
to
to 1999
Average Annual
Growth Rate
Corn
4.26
4.64
4.76
4.76
4.76
5.03
5.06
4.70
4.74
4.95
1.8%
Cottonseed
3.17
3.91
4.16
3.82
3.50
3.89
3.81
3.68
3.53
2.94
-0.2%
Soybean
48.88
47.71
49.08
51.30
49.90
49.78
51.72
54.38
57.08
58.14
2.0%
Canola
2.21
2.81
3.37
4.10
4.49
4.69
4.51
4.28
4.59
5.02
10.0%
Flaxseed
0.67
0.67
0.62
0.61
0.64
0.61
0.59
0.57
0.57
0.57
-1.6%
Peanut
0.79
0.77
0.82
0.84
0.75
0.77
0.73
0.76
0.79
0.78
0.1%
Safflower
0.39
0.19
0.10
0.18
0.17
0.18
0.11
0.26
0.28
0.31
10.3%
Sunflower
0.73
0.99
1.36
0.68
0.54
0.65
0.67
0.76
0.75
0.93
7.0%
Note: In cases where monthly data were unavailable, the calendar year data were estimated based on marketing year month shares of the
calendar year. For example, an estimate of a 1996 consumption quantity based on data reported for an October marketing year would be
calculated as follows: (9/12) • 1995 marketing year quantity + (3/12) • 1996 marketing year quantity.
Sources: U.S. Department of Agriculture, Economic Research Service. 1998. Oil Crops Yearbook, [computer file]. Last updated January
1998.
U.S. Department of Agriculture, Economic Research Service. 1999b. Oil Crops Yearbook, [computer file]. Last updated
November 1999.
U.S. Department of Agriculture, Economic Research Service. 2000. Oil Crops Outlook. September 13, 2000.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook. September 13, 1999.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook. November 12, 1998.
U.S. Department of Commerce, Bureau of the Census. National Monthly Population Estimates: 1980-2000.
. Last updated November 2, 2000.
-------
Total Consumption
21,157.4 million pounds
Other edible
2.0%
Other inedible
Fatty acids
9.3%
Feed
11.1%
Margarine
8.7%
Baking and
frying fats
28.5%
Salad or
cooking oils
31.9%
Figure 2-2. U.S. Consumption of Fats and Oils by Use, 1995
Source: U.S. Department of Commerce. 1996b. 1995 Current Industrial Reports—Fats and Oilseed
Crushings, Production, Consumption, and Stocks. M20K. Washington, DC: Government Printing
Office.
In terms of inedible uses, these vegetable oils are used in smaller quantities than some
of their more specialized substitutes. Inedible substitute products include both the edible
substitute products listed above as well as the following oils that are used only in inedible
products: linseed oil, tall oil, caster oil, tung oil, and inedible tallow.
2.4.2 Oilseed Meal Consumption and Uses
In Table 2-10, per capita consumption of corn germ meal, cottonseed meal, soybean
meal, and other meals is provided for 1990 through 1999, with baseline 1995 data in
boldface. Most meal products are consumed in animal feed products, but an estimate of the
proportion of these products used for animal feed versus human consumption is not available.
Soybean products in particular are used in a variety of protein products (i.e., soy flour
concentrates and isolates) in addition to animal feed products. Hence, the approximately
200 pounds per capita consumption
2-26
-------
Table 2-10. Per Capita Consumption of Meal Products 1990-1999 (Ibs)
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
to Average Annual
^ Growth Rate
Corn Germ
Meal
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Cottonseed
10.32
14.73
13.12
11.22
11.10
13.84
12.05
12.39
11.22
8.92
0.1%
Soybean
181.23
182.54
183.89
190.05
200.32
204.58
203.65
207.98
217.99
225.69
2.5%
Canola
2.97
4.28
5.06
7.10
8.26
9.09
9.43
11.28
12.60
13.52
19.0%
Flaxseed
1.05
1.00
0.90
0.86
0.84
0.93
0.84
1.31
1.33
1.28
2.7%
Peanut
1.16
1.47
1.67
1.33
1.32
1.61
1.41
1.01
0.81
NA
-2.5%
Sunflower
2.49
3.01
3.81
3.21
3.06
4.49
3.59
3.60
4.11
4.47
8.6%
Note: In cases where monthly data were unavailable, the calendar year data were estimated based on marketing year month shares of the
calendar year. For example, an estimate of a 1996 consumption quantity based on data reported for an October marketing year would be
calculated as follows: (9/12) • 1995 marketing year quantity + (3/12) • 1996 marketing year quantity.
NA = not available.
Sources: U.S. Department of Agriculture, Economic Research Service. 1998. Oil Crops Yearbook, [computer file]. Last updated January
1998.
U.S. Department of Agriculture, Economic Research Service. 1999b. Oil Crops Yearbook, [computer file]. Last updated
November 1999.
U.S. Department of Agriculture. 1997a. Agricultural Statistics 1997. Washington, DC: Government Printing Office.
U.S. Department of Agriculture. 2000. Agricultural Statistics 2000. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. National Monthly Population Estimates: 1980-2000. . Last updated November 2, 2000.
-------
per year of soybean meal is a combination of animal feed uses and human uses. The other
meals, which account for anywhere from less than one pound per capita to a dozen pounds
per capita, are used in a combination of animal feed and human uses as well. As shown, the
average annual growth rates for meal products over this period are positive with the
exception of peanut meal. Canola meal experienced the largest annual growth rate at 19
percent.
Of the processed feed uses, soybean meal has the largest portion of the market at
57.0 percent of all processed feeds (see Figure 2-3). Cottonseed meal makes up 6.3 percent,
and other oilseed meals (linseed, peanut, sunflower, and canola) make up 4.4 percent. These
products compete with animal proteins (6.9 percent) and other feed products (23.4 percent)
such as millfeeds.
Total Use
42,362 metric tons
Other feeds
23.4%
Animal proteins
6.9%
Gluten feed iiiiii^)^ f m Soybean meal
and meal rZlS^M^^^^M 57.0%
1.9%
Other oilseed
meals
4.4%
Cottonseed meal
6.3%
Figure 2-3. U.S. Processed Feeds by Type, 1995
Source: U.S. Department of Agriculture, Economic Research Service. 1997. Feed Yearbook, [computer
file]. Last updated April 1997.
2-28
-------
SECTION 3
ENGINEERING COST ANALYSIS
This section presents the Agency's estimates of the compliance costs associated with
the NESHAP on the production of vegetable oils and meals. This regulation will affect all
106 facilities (baseline 1995) that use a solvent extraction process to extract oil from oilseeds
or similar agricultural inputs (e.g., soybean, cottonseed, corn germ) because all are major
sources of HAPs.1 These 106 facilities operated 119 product lines during the 1995 to 1996
time period. The primary solvent used in the extraction process is a commercial grade
hexane comprising 60 to 70 percent n-hexane (CAS No. 110-54-3), which is a HAP (Zukor
and Riddle, 1998). The balance of the solvent composition is a blend of hexane isomers,
which are volatile organic compounds (VOCs).
All vegetable oil facilities operate some type of solvent collection and recovery
system. These systems collect solvent-laden process gas streams from a number of key
process units such as the extractor, desolventizer-toaster, process evaporators, and distillation
columns. The solvent collection and recovery system then routes the gathered process gas
streams to a recovery device that is usually a packed-bed mineral oil scrubber. In addition to
the collection and recovery system, source reduction techniques are used as well. By
optimizing the system's performance, process solvent losses are minimized.
Hexane emissions in vegetable oil production facilities occur from the following ten
general sources:
• the main vent (5 to 20 percent of emissions),
• the meal dryer vent and the meal cooler vent (10 to 30 percent of emissions
combined),
• crude meal (10 to 40 percent of emissions),
'Three additional plants using solvent extraction processes come on line in 1996. The cost and emissions data
in this section include data from these plants although they are not included in the 1995 baseline economic
analysis.
3-1
-------
• crude oil (5 to 15 percent of emissions),
• equipment leaks (1 to 25 percent of emissions),
• solvent storage tank leaks (1 to 5 percent of emissions),
• process wastewater collection (1 to 5 percent of emissions),
• facility startups and shutdowns (10 to 20 percent of emissions), and
• operational upsets (1 to 20 percent of emissions) (Zukor and Riddle, 1998).
As described in this section, the Agency estimated the compliance costs for each facility to
install the necessary equipment and process controls that will reduce emissions and bring
each facility into compliance with the NESHAP. The estimation of these costs is currently
applied to existing facilities, although new sources may be considered later. Control options
and costs are described in Section 3.1. National emissions reductions and compliance costs
are described in Section 3.2.
3.1 Control Options and Costs
The NESHAP will limit the gallons of HAP loss per ton of seeds processed rather
than establish regulatory requirements at each emission point. This approach allows industry
the flexibility to implement the most cost-effective method to reduce overall HAP loss and
minimizes the costs associated with monitoring, recordkeeping, reporting, and other
administrative requirements for both industry and the regulatory agencies (Durham, 1998).
The remainder of this section describes the controls based on plant characteristics and
then summarizes their associated costs. In addition to capital costs and operating and
maintenance costs, this section describes monitoring, recordkeeping, and reporting costs and
lost production costs.
3.1.1 Control Options
Solvent losses vary among plants based primarily on the oilseed type, desolventizing
method used, oilseed processing rate, and oilseed prepressing operations. To determine
control options, plants were subcategorized into the following:
• "soybean" plants processing soybeans in both conventional and specialty
desolventizers;
• "corn germ" plants processing corn germ with a wet or dry milling process;
3-2
-------
• "large cottonseed" plants processing 120,000 tons or more of cottonseed per year
as well as plants processing sunflower seed; and
• "small cottonseed" plants processing fewer than 120,000 tons of cottonseed per
year as well as plants processing canola seed, flaxseed, peanuts, and safflower
seed.
To develop model plants, the "soybean" plants were characterized as processing 2,200 tons of
seed per day. "Large cottonseed" and "corn germ" plants were characterized as processing
1,100 tons of seed per day. "Small cottonseed" plants were characterized as processing
400 tons of seed per day. All but specialty soybean plants, which were assumed to operate
300 days per year, were assumed to operate 330 days per year.
Based on their needed emissions reductions to meet the MACT floor, plants were
assigned to one of the following model plants:
• Model MACT Plant: 0 percent emissions reduction;
• Model Plant 1: 30 percent emissions reduction;
• Model Plant 2: 50 percent emissions reduction; and
• Model Plant 3: 70 percent emissions reduction.
One additional model plant was developed to represent soybean facilities operating a
specialty desolventizer. This model plant was used only for calculating impacts for the
"above-the-MACT-floor" option.
Of 119 facility product lines, 42 (35.2 percent) were assigned to the Model
MACT Plant, 50 (42.0 percent) were assigned to Model Plant 1, 17 (14.3 percent) were
assigned to Model Plant 2, and 10 (8.4 percent) were assigned to Model Plant 3 (Zukor and
Snyder, 2000).2 The control technologies assigned to each model plant are described in
Table 3-1. These include, for example, installing additional trays in the desolventizer to
increase residence time (Model Plants 1 and 2), installing an oil dryer in the oil distillation
system (Model Plants 2 and 3), and implementing a leak detection and repair program for
fugitive equipment leaks (Model Plant 3).
2One facility combined reported solvent usage for three oilseed types and thus is treated as a single facility
product line.
3-3
-------
Table 3-1. Summary of Control Technologies Assigned to Model Plants
Model Plant
Required Emissions
Reduction to
Meet MACT
Assigned Control Technologies
MACT Plant
0%
None
Model
Plant 1
30%
Install two additional trays in the desolventizer to
provide increased residence time.
Model
Plant 2
50%
Install two additional trays in the desolventizer to
provide increased residence time.
Install an oil dryer in the oil distillation system to
reduce the residual solvent content in the oil product.
Install a refrigerated condenser on the main vent to
recover solvent vapors in the exhaust.
Model
Plant 3
70%
Install a completely new, counter-current desolventizer.
Install an oil dryer in the oil distillation system to
reduce the residual solvent content in the oil product.
Install a refrigerated condenser on the main vent to
recover solvent vapors in the exhaust.
Vent standing and working losses from fixed-roof
storage tanks to the existing solvent recovery system.
Implement a leak detection and repair (LDAR) program
for fugitive equipment leaks.
Source: Zukor, C. and J. Snyder, Alpha-Gamma Technologies, Inc. November 10, 2000. "Final Summary of
Emission Reductions and Control Costs Associated with Achieving the MACT Floor and a Control
Option Above the MACT Floor." Memorandum submitted to the Vegetable Oil NESHAP Project
File.
In addition to these controls, the Agency also evaluated an above-the-MACT-floor
option, which is more stringent than the MACT floor. For this option, plants would be
required to install a fabric filter and catalytic incinerator to the combined exhaust from the
meal dryer and cooler vents (Zukor and Snyder, 2000). The fabric filter is installed to
remove excess particulate matter in the exhaust stream prior to entering the catalytic
incinerator. The catalytic incinerator is capable of reducing the volume of HAP and VOC
emissions in the exhaust stream by approximately 95 percent. These controls were assumed
to be added on after the plant installed controls to achieve the MACT floor. Unlike the
3-4
-------
control technologies to achieve the MACT floor, no vegetable oil processing plants currently
have a catalytic incinerator in place.
3.1.2 Control Costs
Control costs were estimated for both the MACT floor and the above-the-MACT-
floor options. Compliance costs to achieve the MACT floor include the following
components:
• O&M: operating and maintenance costs;
• MRR: monitoring, recordkeeping, and reporting costs;
• AC: capital investment annualized over 15 years at 7 percent interest;
• SRC: solvent recovery credit (cost savings due to increased reuse of solvent); and
• LPC: lost production costs annualized over 15 years at 7 percent interest
(operating costs and foregone profits incurred while the plant is shut down to
install capital).
The estimated compliance costs for the model plants were scaled to reflect the
processing rate of each affected facility. Table 3-2 presents the MACT floor estimated HAP
emissions reductions, total capital investment costs, annual costs by type, and
cost-effectiveness for each subcategory within each model plant. Soybean model plants
range from 112 to 607 tons per year in HAP emissions reductions at an annual cost of
$227,000 to $575,000 per year. Cost per ton emission reduction measures decrease from
$2,026 per ton of HAP for Model Plant 1 to $948 per ton of HAP for Model Plant 3. Corn
germ model plants range from 112 to 607 tons per year in HAP emissions reductions at an
annual cost of $185,000 to $511,000 per year with associated cost per ton emission reduction
measures of $1,649 to $842 per ton. Large cottonseed model plants range from 139 to 759
tons per year in HAP emissions reductions at an annual cost of $170,000 to $423,000 per year
with associated cost per ton emission reduction measures of $1,222 to $558 per ton. Finally,
small cottonseed model plants range from 71 to 386 tons per year in HAP emissions
reductions at an annual cost of $115,000 to $286,000 per year with associated cost per ton
emission reduction measures of $1,626 to $740 per ton. In all subcategories, plants requiring
larger emissions reductions can achieve them at lower average cost (i.e., $/ton of HAPs) than
plants requiring small emissions reductions.
3-5
-------
Table 3-2. Cost Estimates for MACT Floor Control Technologies for Each Model Plant (1995 dollars)
(Continued)
Model Plant
Model Plant 3
Soybean
Corn Germ
Large Cottonseed
Small Cottonseed
OJ
a, a O&M— Operation
HAP
Emissions
Reduction
(tons/yr)
607
607
759
386
and Maintenanc
Total
Capital
Investment
($103)
$2,771
$2,995
$2,618
$1,371
;e Cost
Annualized Costs ($103/yr)a
O&M
$237
$246
$230
$181
MRR
$40
$40
$40
$40
AC
$305
$329
$287
$150
SRC
$185
$185
$231
$118
MRR — Monitoring, Recordkee
LPC
$178
$81
$97
$32
ping and
Total
$575
$511
$423
$286
Reportin
Cost Per Ton
Emission
Reduction
($/ton HAP)
$948
$842
$558
$740
g Costs
Costs Apply
to Emissions
Reduction
Range
>60
percent
AC—Annualized Capital Costs (15 years at 7%) SRC—Solvent Recovery Credit (cost savings)
LPC—Annualized Lost Production Cost (15 years at 7%) Total Annual Cost = O&M + MRR + AC - ( SRC) + LPC
Source: Zukor, C. and ]. Snyder, Alpha-Gamma Technologies, Inc. November 10, 2000. "Final Summary of Emission Reductions and
Control Costs Associated with Achieving the MACT Floor and a Control Option Above the MACT Floor." Memorandum submitted
to the Vegetable Oil NESHAP Project File.
-------
Table 3-2. Cost Estimates for MACT Floor Control Technologies for Each Model Plant (1995 dollars)
Model Plant
Model Plant 1
Soybean
Corn Germ
Large Cottonseed
Small Cottonseed
Model Plant 2
Soybean
Corn Germ
Large Cottonseed
Small Cottonseed
HAP
Emissions
Reduction
(tons/yr)
112
112
139
71
260
260
325
166
Total
Capital
Investment
($103)
$600
$681
$582
$320
$892
$951
$805
$463
Annualized Costs ($103/yr)a
O&M
$56
$59
$55
$44
$116
$119
$113
$100
MRR
$40
$40
$40
$40
$40
$40
$40
$40
AC
$66
$75
$63
$35
$98
$104
$89
$51
SRC
$34
$34
$42
$22
$79
$79
$99
$51
LPC
$99
$45
$54
$18
$139
$63
$76
$25
Total
$227
$185
$170
$115
$313
$247
$219
$165
Cost Per Ton
Emission
Reduction
($/ton HAP)
$2,026
$1,649
$1,222
$1,626
$1,205
$952
$673
$994
Costs Apply
to Emissions
Reduction
Range
Ito39
percent
40 to 59
percent
(continued)
-------
Table 3-3 presents the above-the-MACT-floor estimated HAP emissions reductions,
total capital investment costs, annual costs by type, and cost per ton emission reduction for
each of the following types of plants: small cottonseed, large cottonseed and corn germ,
conventional soybean, and specialty soybean. HAP emissions reductions range from 27 tons
per year for small cottonseed plants to 566 tons per year for large-size specialty soybean
(2,200 tons per day). Annualized costs for the above-the-MACT-floor scenario include
operating and maintenance costs; annualized capital costs; monitoring, recordkeeping, and
reporting costs; and lost production costs. The costs shown in Table 3-3 are incremental to
the costs for the MACT floor. Lost production costs are assumed to be the same as those
associated with the MACT floor controls. Because the add-on equipment incinerates rather
than recovers solvent, no additional solvent recovery credits are associated with it.
Annualized costs range from $762,000 per year for small cottonseed plants to $19,890,000
per plant for large-size specialty soybean plants (2,200 tons per day). Cost per ton emission
reduction measures range from $15,293 per ton for large cottonseed and corn germ to
$51,462 per ton for small-size specialty soybean plants (220 tons per day).
3.2 National Emissions Reductions and Compliance Costs
Under the NESHAP, the Agency estimates that a 25 percent reduction in emissions
will be achieved with the MACT floor option and a 43 percent reduction in emissions will be
achieved with the above-the-MACT-floor option (Table 3-4). With the MACT floor option,
VOC emissions are expected to decline by 10,600 tons per year, and HAP emissions are
expected to decline by 6,800 tons per year. With the above-the-MACT-floor option, VOC
emissions are expected to decline by 18,300 tons per year, and HAP emissions are expected
to decline by 11,700 tons per year.
Total annual compliance costs for plants operating in 1995 and for three plants that
began operation in 1996 were estimated at $12.3 million with the MACT floor option and
$204.6 million with the above-the-MACT-floor option. With the MACT floor option, cost
per ton emission reduction measures are $1,200 per ton of VOC reductions and $1,800 per
ton of HAP reductions. With the above-the-MACT-floor option, cost per ton emission
reduction measures are $11,200 per ton of VOC reductions and $18,400 per ton of HAP
reductions.
3-8
-------
Table 3-3. Incremental Cost Estimates for Above-the-MACT-Floor Control Technologies for Each Model Plant
(1995 dollars)
Model Plant
Small
Cottonseed
Large Cottonseed and
Corn Germ
Conventional
Soybean
Specialty Soy: 220 tpdc
Specialty Soy: 770 tpdc
Specialty Soy: 2,200 tpdc
HAP
Emission
Reductions
(tons/yr)a
27
82
56
39
185
566
Total
Capital
Investment
($103)
$591
$873
$1,163
$1,295
$4,532
$12,948
Annualized Costs
($103/yr)b
O&M
$686
$1,157
$1,754
$1,880
$6,580
$18,800
MRR
$20
$20
$20
$20
$20
$20
AC
$56
$77
$98
$107
$375
$1,070
SRC
$0
$0
$0
$0
$0
$0
LPC
$0
$0
$0
$0
$0
$0
Total
$762
$1,254
$1,872
$2,007
$6,975
$19,890
Cost Per Ton
Emission
Reduction
($/ton HAP)
$28,222
$15,293
$33,429
$51,462
$37,703
$35,141
a HAP emission reductions achieved by the catalytic incinerator are estimated as 95 percent of the remaining HAP emissions released from the
combined meal dryer and cooler vents after applying MACT floor controls.
b O&M—Operation and Maintenance Cost MRR—Monitoring, Recordkeeping, and Reporting Costs
AC—Annualized Capital Costs (15 years at 7%) SRC—Solvent Recovery Credit (cost savings)
LPC—Annualized Lost Production Cost (15 years at 7%) Total Annual Cost = O&M + MRR + AC - ( SRC) + LPC
0 tpd—tons per day.
Source: Zukor, C. and J. Snyder, Alpha-Gamma Technologies, Inc. November 10, 2000. "Final Summary of Emission Reductions and
Control Costs Associated with Achieving the MACT Floor and a Control Option Above the MACT Floor." Memorandum to the
Vegetable Oil NESHAP Project File.
-------
Table 3-4. Summary of National Emissions and Costs for the MACT Floor and Above-
the-MACT-Floor Control Scenarios (1995 dollars)3
Control Option
MACT Floor
Above MACT Floor
Emissions Reductions
(tons/yr)
VOC HAP
10,600 tons 6,800 tons
18,300 tons 11,700 tons
Overall
Emissions
Reduction
(percent)
25%
43%
Total Annual
Cost
(million $)
$12.3
$204.6
Cost Per Ton Emission
Reduction
($/ton)
VOC
$1,200
$11,200
HAP
$1,800
$18,400
a Totals include all facilities operating in 1995 as well as three facilities that began operation in 1996.
Source: Zukor, C. and J. Snyder, Alpha-Gamma Technologies, Inc. November 10, 2000. "Final Summary of
Emission Reductions and Control Costs Associated with Achieving the MACT Floor and a Control
Option Above the MACT Floor." Memorandum submitted to the Vegetable Oil NESHAP Project
File.
3-10
-------
SECTION 4
ECONOMIC IMPACT ANALYSIS
The rule to control emissions of HAPs from vegetable oil and meal processing
facilities will affect the entire U.S. industry directly (through imposition of compliance costs)
or indirectly (through changes in market prices). Implementation of the rule will increase the
costs of production at solvent extraction plants. As described in Section 3, these costs vary
across facilities depending on their physical characteristics and baseline controls. The
response of producers to these additional costs determines the economic impacts of the
regulation. Specifically, the cost of the regulation may induce some owners to close their
operations (entire facility or individual product lines) or to change their current operating
rates. These choices affect, and in turn are affected by, the market price for vegetable oils
and meals.
This section describes the data and approach used to estimate the economic impacts
of the regulation for baseline year 1995. Section 4.1 presents the inputs for the economic
analysis, including producer characterization, market characterization, and compliance costs
of the regulation. Section 4.2 describes the approach to estimating the economic impacts on
the industry, and Section 4.3 describes the assumptions of the model. Finally, Section 4.4
presents the results of the EIA.
4.1 Economic Analysis Inputs
Inputs to the economic analysis include a baseline characterization of vegetable oil
and meal producers, their markets, and the estimated costs of complying with the regulation.
Each input is described briefly below.
4.1.1 Producer Characterization
The baseline characterization of vegetable oil and meal producers is based primarily
on the facility responses to EPA's Section 114 questionnaires (hereafter called EPA's facility
database). This information includes oilseed purchases, oil production, meal production,
by-product production, capacity, number of employees, facility location, and company
ownership. These facility-specific data on existing major sources were supplemented with
4-1
-------
secondary information on market volumes and market prices from the USDA and on trade
from the U.S. International Trade Commissions. Using these data, the Agency developed
product-specific cost equations for this analysis (described fully in Appendix A).
4.1.2 Vegetable Oil and Meal Markets
Table 4-1 provides 1995 data on the U.S. vegetable oil and meal markets for use in
this analysis. The market prices for each product were obtained from the USDA. Market
output for each product is the sum of U.S. production and foreign imports. The affected
portion of U.S. production of each product is the sum of the individual facility production
levels taken from EPA's facility database, while the unaffected portion is derived as the
difference between the reported U.S. production and affected production volumes (see Table
2-7). Foreign trade data on exports and imports of these products are from the U.S.
International Trade Commission and the USDA.
4.1.3 Regulatory Control Costs
As described in Section 3, the Agency developed compliance cost estimates for each
of the vegetable oil and meal facilities affected by the regulation. Two control scenarios were
considered and thus two sets of costs were developed. The scenarios are as follows:
• the MACT floor including lost production costs and
• the above-the-floor option including lost production costs.
The equipment needed to comply with the MACT floor is an upgrade of existing equipment
at vegetable oil production facilities. In contrast, the equipment needed to comply with the
above-the-floor option is additional equipment added on to the existing process after it is
upgraded to the MACT floor.
Compliance costs are either fixed with regard to the level of production or vary with
the level of production. The costs that are fixed include the following:
• annualized total capital investment (capital recovery)—costs to comply with the
MACT floor as well as above-the-floor equipment costs are annualized over
15 years at 7 percent;
• the catalytic incinerator required for the above-the-floor option must be replaced
every 2 years; thus, its cost is annualized over 2 years at 7 percent;
4-2
-------
Table 4-1. Summary of Baseline Vegetable Oil and Meal Values: 1995
Baseline
Corn
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Cottonseed
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Soybean
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
All Other
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Oil Product
$532.00
1,126,198
1,121,703
1,066,819
54,884
415,270
4,495
$536.00
646,031
645,925
622,308
23,617
147,281
106
$536.00
7,981,880
7,968,264
7,968,264
0
1,143,544
13,616
$658.82
1,623,013
1,093,411
1,026,335
67,076
632,625
529,125
Meal Product
$88.40
1,473,651
1,473,651
1,401,546
72,105
61,950
0
$123.20
1,826,100
1,826,100
1,744,562
81,538
89,700
0
$173.80
31,290,977
31,225,572
31,225,572
0
6,491,570
65,405
$93.22
2,520,129
1,583,559
1,497,886
85,673
138,469
936,570
4-3
-------
• monitoring, recordkeeping, and reporting—fixed cost per facility; and
• lost production costs.
The costs that vary with the level of production include
• operation and maintenance costs for the control equipment, less the cost savings
from reduced solvent purchases.
Because costs were estimated at the facility level but each facility produces oil and
meal jointly, compliance costs were allocated to each product at the facility based on
assumptions described in Section 4.3. Once allocated to each product, compliance costs were
divided by total production of each product at each affected facility to obtain a per-unit
compliance cost or "cost shifter."
As one indicator of the impact of the regulation, cost-to-sales ratios (CSRs) were
estimated at the facility level by dividing the regulatory compliance costs by facility revenue.
Facility revenues were obtained by multiplying market prices as reported by the USDA by
facility-level production. Table 4-2 presents facility-level CSRs for each of the two
regulatory scenarios. For the MACT floor scenario with lost production costs, 104 of the
106 facilities have CSRs below 1 percent, two have CSRs from 1 to 2 percent, and no facility
has a CSR above 2 percent. For the above-the-floor scenario with lost production costs,
17 facilities have CSRs below 1 percent, 44 have CSRs from 1 to 2 percent, and 45 have
CSRs above 2 percent.
Table 4-2. Distribution of Facility-Level Compliance Cost-to-Sales Ratios: 1995
MACT Floor with LPa
0-1%
1-2%
>2%
TOTAL
Number
104
2
0
106
Share
78.1%
1.9%
0.0%
100.0%
Above-the-Floor with LPa
Number
17
44
45
106
Share
16.0%
41.5%
42.5%
100.0%
LP represents lost production costs that may be incurred due to the downtime to install new capital
equipment.
The Agency also evaluated whether these CSRs would change substantially using
1999 data because economic conditions in the industry are different than they were in 1995.
4-4
-------
The compliance costs were projected to 1999 using the producer price index, revenues were
calculated using 1999 market prices, and CSRs were then recomputed. Based on these
calculations, only one facility switched from the 0-1 percent to 1-2 percent category, and all
other facilities had increased CSRs of less than 0.1 percent. Thus, this analysis suggests that
the results of the economic impact analysis would not differ substantially using more recent
data.
The Agency also considered other potential indicator measures including the ratio of
compliance costs to gross margins or profits for the industry. In general, reliable
independently verifiable data are unavailable to compute these ratios for the industry, and the
data that are available are severely limited. In particular, only three of the 31 affected
companies are publicly traded and thus have publicly available financial statements. For
these three companies, the cost-to-profit ratios of the regulation are substantially less than 1
percent.
4.2 Economic Impact Methodology
This section summarizes the Agency's economic approach to modeling the responses
of vegetable oil and meal producers and markets to the imposition of the regulation. In
conducting an economic analysis and determining the economic impacts, the analyst should
recognize the alternatives available to each producer in response to the regulation and the
context of these choices. The Agency evaluated the economic impacts of this NESHAP
using a market-based approach that gives producers the choice of whether to continue
producing these products and, if so, to determine the optimal level consistent with market
signals.
The Agency's approach is soundly based on standard microeconomic theory, employs
a comparative statics approach, and assumes certainty in relevant markets. Prices and
quantities were determined in perfectly competitive markets for each vegetable oil and meal
product. Production decisions involve whether a firm with plant and equipment already in
place purchases inputs to produce outputs. These are called short-run decisions since the
plant and equipment are fixed. A profit-maximizing firm will operate existing capital as long
as the market price for its output exceeds its per-unit variable production costs. As long as
the market price even marginally exceeds the average variable (operating) costs, the facility
will cover not only the cost of its variable inputs but also part of its capital costs. In the short
run, a profit-maximizing firm will not pass up an opportunity to recover part of its fixed
investment in the plant and equipment.
4-5
-------
The Agency developed cost curves for each product at solvent extraction facilities.
Given the capital in place, each vegetable oil and meal product at an affected facility is
characterized by an upward-sloping supply function, as shown in Figure 4-1. In this case, the
supply function is that portion of the marginal cost curve bounded by zero and the technical
capacity at the facility. The facility owner is willing to supply output according to this
schedule as long as the market price is sufficiently high to cover average variable costs. If the
market price falls below the average variable costs, then the firm's best response is to cease
production because total revenue does not cover total variable costs of production.
$/lb
Ibs/year
Figure 4-1. Supply Curve for a Representative Affected Facility
The individual facility-level supply decisions were aggregated to develop the market
supply curve. This economic analysis assumes that prices for vegetable oils and meals are
determined in perfectly competitive markets (i.e., individual facilities have negligible power
over the market price of the products and thus take the prices as "given" by the market). As
shown in Figure 4-2(a), under perfect competition, market prices and quantities are
determined by the intersection of market supply and demand curves. The initial baseline
scenario consists of a market price and quantity (P, Q) that are determined by the downward-
sloping market demand curve (DM) and the upward-sloping market supply curve (SM) that
reflects the sum of the individual supply curves of affected and unaffected facilities.
4-6
-------
+ p
Solvent Extraction
Facilities
= P
I D
Q
Mechanical Extraction
Facilities
Market
,M
a) Baseline Equilibrium
P'
P
Solvent Extraction
Facilities
P'
P
SM7 SM/
Mechanical Extraction
Facilities
Q' Q
Market
b) With Regulation Equilibrium
Figure 4-2. Market Equilibrium Without and With Regulation
Now consider the effect of the regulatory control costs. These costs include the
variable component consisting of the operating and maintenance costs and the nonvariable
component consisting of the compliance capital equipment required for the regulatory option.
Incorporating the regulatory control costs involves shifting upward the supply curve for each
affected facility by the per-unit compliance cost. As a result of the upward shift in the supply
4-7
-------
curve for each affected facility, the market supply curve for each affected vegetable oil will
shift upward to reflect the increased costs of production at solvent-extraction facilities.
The estimated per-unit annual compliance cost of the MACT standard was
incorporated into the baseline market scenario as shown in Figure 4-2(b). In the baseline
scenario without the standards, at the projected price, P, the industry would produce total
output, Q, with solvent-extraction facilities producing the amount qs and mechanical
extraction facilities accounting for Q minus qs, or qm. The regulation raises the average total
cost (annual capital costs plus operating and maintenance) of solvent extraction facilities,
causing their supply curves to shift upward from Ss to Ss' and the market supply curve to shift
upward to SM/. At the new equilibrium with the regulation, the market price increases from P
to P' and market output (as determined from the market demand curve, DM) declines from Q
to Q'. This reduction in market output is the net result of reductions at solvent extraction
facilities and increases at mechanical extraction facilities.
To estimate the economic impacts of the regulation under this scenario, EPA
operationalized the conceptual model described above in a Lotus 1-2-3 multiple spreadsheet
model for each vegetable oil and meal market. Appendix A provides the details of the
operational market model for this economic analysis. In summary, this model characterizes
domestic and foreign producers and consumers of each product and their behavioral
responses to the imposition of the regulatory compliance costs. These costs are expressed per
pound of product for each facility and serve as the input to the market model, or the "cost
shifters" of the baseline supply curves at the facility. Given these costs for directly affected
facilities, the model determines a new equilibrium solution in a comparative static approach
with higher market prices and reductions in output of each product.
4.3 Economic Model Assumptions
In developing and implementing the economic model, several assumptions were
necessary. These assumptions are either numerical or operational. This section describes
each type of assumption.
4.3.1 Operational Assumptions
The operational assumptions of the model influence the structure and coverage of the
model. They are as follows:
4-8
-------
• The domestic markets for crude vegetable oils and meals are perfectly
competitive.
• The U.S. may potentially influence the price of crude vegetable oils and meals in
the world market (i.e., the U.S. is not a price-taker).
• Vegetable oils and meals are produced in fixed proportions relative to the oilseed
inputs.
• The markets for by-products and co-products of the vegetable oil and meal
production process will be unaffected by the regulation.
• The markets for specialty use products, which are primarily produced from crude
meals, will be unaffected by the regulation.
• The baseline year for the analysis, which is 1995, is representative of a typical
year for the industry.
• The markets for the other vegetable oils and meals (canola, flaxseed, peanut,
safflower, and sunflower) are sufficiently similar that they can be combined.
In Section 6, each of these operational assumptions is explained further. In addition,
the impact of each assumption on the results of the model is described. For example, if 1995
was a better than typical year for the industry, then the estimated percentage price effects
would appear smaller for the baseline year than in a typical year. However, based on the
estimates of the CSRs using 1999 data, as described in Section 4.1.3, the results of the model
would likely not differ substantially using more recent data.
4.3.2 Numerical Assumptions
The numerical assumptions of the economic model are the actual values used in
developing the spreadsheet model. These include the demand, supply, and trade elasticities
used in the model as well as the methods for allocating seed costs and compliance costs to the
joint oil and meal products at each facility. These assumptions are described briefly below.
4.3.2.1 Elasticity Assumptions
In determining the elasticity values to use in the economic model, the Agency
reviewed the economics literature and estimated econometric models. Appendix B describes
each paper that cites relevant elasticity estimates and presents the results of the independent
econometric estimates. The elasticities used in the model, which were obtained from both
sources, are listed in Table 4-3 (see Section B.4 for a more complete description).
4-9
-------
Table 4-3. Elasticity Estimates Used in the Economic Impact Analysis
Demand Elasticities
Supply Elasticities
Import Supply Elasticities
Export Demand Elasticities
Product
Corn oil
Cottonseed oil
Soybean oil
All other oils
Corn germ meal
Cottonseed meal
Soybean meal
All other meals
Corn oil
Cottonseed oil
Soybean oil
All other oils
Corn germ meal
Cottonseed meal
Soybean meal
All other meals
All products
All products
Elasticity Used for the EIA
-0.39
-0.65
-0.34
-0.33
-0.46
-1.01
-0.27
-0.64
0.44
0.44
0.44
0.44
0.28
0.28
0.28
0.28
1.00
-1.00
In general, the demand elasticities used in the model were obtained from the
independent econometric estimates. The one exception is that the econometrically estimated
cottonseed oil demand elasticity was outside the range of the econometric estimates; thus, an
estimate from the literature was used. In the case of supply elasticities, the values used are
based on estimates in the literature for soybean products.
To investigate the impacts of alternative elasticity values on the economic model
results, EPA conducted a sensitivity analysis. The economic model was run using, in one
case, the elasticity values expected to generate the greatest effects and, in the other case, the
elasticity values expected to generate the smallest effects. These results are presented in
4-10
-------
Appendix C. In general, the projected effects of the regulation are not particularly sensitive
to changes in the elasticity values.
4.3.2.2 Allocation of Compliance Costs to Oils andMeals
Facility-level compliance costs were allocated to oil and meal products produced at
the facility based on the relative shares of revenue generated by each product. First,
facility-level revenues were calculated for oils and for meals and then expressed as a
proportion relative to total revenue from both products. Then, facility-level compliance costs
were then allocated to each product using these proportions.
This method is preferable to allocating compliance costs based on the proportion of
the weight of the respective products relative to the weight of the oilseed input. This is
because oil generates more revenue relative to its weight than meal. (For example, soybeans
are approximately 20 percent oil and 80 percent meal by weight, but oil generates 45 percent
and meal generates 55 percent of the revenue.) Thus, this method would allocate too much of
the compliance costs to meal.
4.3.2.3 Allocation of Seed Costs to Oils andMeals
Similar to the issue with compliance costs, seed costs were allocated separately to
vegetable oils and oilseed meals to facilitate construction of supply curves for each individual
product. Seed costs were allocated to each product based on its proportion of revenue. First,
facility-level revenues were calculated at each plant using market prices for oils and meals,
and seed costs were calculated using market prices for oilseeds. The industry average
percentage of the oilseed cost relative to total revenue (averaged over oilseed types) was used
to allocate oilseed cost to each individual product. For cottonseed, the average proportion of
cottonseed cost relative to total revenue is approximately 61 percent. Thus, it was assumed
that 61 percent of the revenue generated from cottonseed oil is the cost of cottonseed used in
its production, and 61 percent of the revenue generated from cottonseed meal is the cost of
cottonseed used in its production. For soybeans, the average value is approximately
85 percent and for corn germ, the average value is approximately 77 percent. Since market
prices are missing for most of the other types of oilseeds and oilseed products, 73 percent,
which is the average of the soybean value and cottonseed value, was assumed.
4-11
-------
4.4 Economic Impact Results
This section provides the economic impacts of the MACT regulation. The model
results are summarized below as market-, industry-, and society-level impacts due to the
regulation.
4.4.1 Market-Level Results
Table 4-4 compares 1995 baseline values with the projected effects of the two
regulatory scenarios for each of the eight vegetable oil and meal markets. Compliance costs
are ten to 12 times larger for the above-the-floor option than the MACT floor option and
therefore have much greater market-level effects. In all cases, the following changes occur:
• market price increases,
• production by domestic solvent extraction facilities decreases,
• production by domestic mechanical extraction facilities increases or remains
unchanged,
• exports decrease, and
• imports increase.
Under the floor scenario, the size of the effects in percentage change terms is similar
across all of the markets. Price increases for oils range from 0.16 percent for all other
vegetable oils to 0.48 percent for cottonseed and corn oil. Output decreases range from 0.11
percent for all other vegetable oil to 0.35 percent for cottonseed oil. In the meal markets,
price increases range from 0.10 percent for all other vegetable oils to 0.34 percent for corn
germ meal. Output decreases range from 0.07 percent for all other vegetable meals to 0.27
percent for cottonseed meal.
4.4.2 Industry-Level Results
Table 4-5 compares the 1995 baseline industry measures with the effects of the two
regulatory scenarios for the total domestic industry, domestic solvent extraction facilities, and
domestic mechanical extraction facilities. As with the market-level effects, effects are greater
for the above-the-floor option compared to the floor option. The following general changes
are projected to occur:
• industry revenues increase as a result of higher market prices that offset decreases
in output,
4-12
-------
Table 4-4. Summary of Market-Level Impacts of Vegetable Oil Production
NESHAP: 1995a
Scenario:
Compliance Costs:
Floor with LP"
$12,
151
Above-the-Floor with LP"
$203
,704
Changes from Baseline
Corn Oil
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mech. Extraction
Exports
Imports
Cottonseed Oil
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mech. Extraction
Exports
Imports
Soybean Oil
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mech. Extraction
Exports
Imports
All Other Vegetable Oils
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mech. Extraction
Exports
Imports
Baseline
$532.00
1,126,198
1,121,703
1,066,819
54,884
415,270
4,495
$536.00
646,031
645,925
622,308
23,617
147,281
106
$536.00
7,981,880
7,968,264
7,968,264
0
1,143,544
13,616
$658.82
1,623,013
1,093,411
1,026,335
67,076
632,625
529,125
Absolute
$2.53
-3,280
-3,301
-3,416
115
-1,966
21
$2.55
-2,233
-2,234
-2,283
49
-697
1
$1.84
-11,867
-11,914
-11,914
0
-3,905
47
$1.07
-1,763
2,622
-2,670
48
-1,025
859
Percent
0.48%
-0.29%
-0.29%
-0.32%
0.21%
-0.47%
0.48%
0.48%
-0.35%
-0.35%
-0.37%
0.21%
-0.47%
0.48%
0.34%
-0.15%
-0.15%
-0.15%
NA
-0.34%
0.34%
0.16%
-0.11%
-0.24%
-0.26%
0.07%
-0.16%
0.16%
Absolute
$36.44
-26,786
-27,094
-28,718
1,624
-26,623
308
$30.33
-25,414
-25,420
-25,999
579
-7,887
6
$43.82
-266,695
-267,808
-267,808
0
-86,417
1,113
$15.20
-24,593
-36,798
-37,474
676
-14,263
12,205
Percent
6.85%
-2.38%
-2.42%
-2.69%
2.96%
-6.41%
6.85%
5.66%
-3.93%
-3.94%
-4.18%
2.45%
-5.36%
5.66%
8.17%
-3.34%
-3.36%
-3.36%
NA
-7.56%
8.17%
2.31%
-1.52%
-3.37%
-3.65%
1.01%
-2.25%
2.31%
(continued)
4-13
-------
Table 4-4. Summary of Market-Level Impacts of Vegetable Oil Production
NESHAP: 1995a (Continued)
Scenario:
Compliance Costs:
Corn Germ Meal
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mech. Extraction
Exports
Imports
Cottonseed Meal
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mech. Extraction
Exports
Imports
Soybean Meal
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
All Other Vegetable Oil
Meals
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Floor with LP"
Baseline
$88.40
1,473,651
1,473,651
1,401,546
72,105
61,950
0
$123.20
1,826,100
1,826,100
1,744,562
81,538
89,700
0
$173.80
31,290,977
31,225,572
31,225,572
0
6,491,570
65,405
$93.22
2,520,129
1,583,559
1,497,886
85,673
138,469
936,570
$12,
Absolute
$0.30
-3,304
-3,304
3,373
69
-212
0
$0.34
-5,019
-5,019
-5,081
62
-244
0
$0.47
-35,793
-35,970
-35,970
0
-17,571
178
$0.10
-1,729
-2,704
-2,729
25
-144
975
151
Changes
Percent
0.34%
-0.22%
-0.22%
-0.24%
-0.34%
NA
0.27%
-0.27%
-0.27%
-0.29%
0.08%
-0.27%
NA
0.27%
-0.11%
-0.12%
-0.12%
NA
-0.27%
0.27%
0.10%
-0.07%
-0.17%
-0.18%
0.03%
-0.10%
0.10%
Above-the-Floor with LP"
$203,704
from Baseline
Absolute
$11.33
-84,725
-84,725
-87,201
2,476
-7,036
0
$4.38
-63,320
-63,320
-64,122
802
-3,082
0
$12.49
-892,586
-897,287
-897,287
0
-435,309
4,701
$1.86
-32,548
-51,186
-51,660
474
-2,702
18,638
Percent
12.81%
-5.75%
-5.75%
-6.22%
-11.36%
NA
3.56%
-3.47%
-3.47%
-3.68%
0.98%
-3.44%
NA
7.19%
-2.85%
-2.87%
-2.87%
NA
-6.71%
7.19%
1.99%
-1.29%
-3.23%
-3.45%
0.55%
-1.95%
1.99%
a Facilities that began operation in 1996 or later years are not included in the analysis.
b LP represents lost production costs that may be incurred due to the downtime to install new capital
equipment.
NA = not available.
4-14
-------
Table 4-5. Summary of Industry-Level Impacts of Vegetable Oil Production
NESHAP: 1995a
Floor with LP"
Above-the-Floor with LP"
Changes from Baseline
Total Industry
Revenues ($103)
Costs ($103)
Post-regulatory
Oil and Meal Production
Gross Profits ($103)
Operating Entities
Product-Line Closures
Facility Closures
Employment Loss
Solvent Extraction
Revenues ($103)
Costs ($103)
Post-regulatory
Production
Gross Profits ($103)
Operating Entities
Product Lines
Facilities
Employment
Mechanical Extraction
Revenues ($103)
Costs ($103)
Post-regulatory
Production
Gross Profits ($103)
Operating Entities
Product Lines
Facilities
Employment
Baseline
$11,963,996
$10,567,809
$0
$10,567,809
$1,396,186
NA
NA
NA
$11,853,542
$10,482,567
$0
$10,482,567
$1,370,975
111
106
5,673
$110,454
$85,242
$0
$85,242
$25,212
NA
NA
NA
Absolute
$17,728
-$6,394
$12,076
-$18,470
$24,122
0
0
-12
$17,264
-$6,477
$12,076
-$18,554
$23,741
0
0
-12
$464
$84
$0
$84
$380
NA
NA
NA
Percent
0.1%
-0.1%
NA
-0.2%
1.7%
NA
NA
NA
0.1%
-0.1%
NA
-0.2%
1.7%
0.0%
0.0%
-0.2%
0.4%
0.1%
0.0%
0.1%
1.5%
NA
NA
NA
Absolute
$447,584
-$216,276
$163,977
-$380,253
$663,860
6
3
-350
$440,414
-$217,808
$163,977
-$381,785
$658,222
6
3
-350
$7,170
$1,532
$0
$1,532
$5,638
NA
NA
NA
Percent
3.7%
-2.0%
NA
-3.6%
47.5%
NA
NA
NA
3.7%
-2.1%
NA
-3.6%
48.0%
5.4%
2.8%
-6.2%
6.5%
1.8%
0.0%
1.8%
22.4%
NA
NA
NA
a Facilities that began operation in 1996 or later years are not included in the analysis.
b LP represents lost production costs that may be incurred due to the downtime to install new capital
equipment.
NA = not available.
4-15
-------
• net industry costs decrease,
• production costs decrease for solvent extraction facilities due to decreased
production levels,
• production costs increase for mechanical extraction facilities due to increased
production levels,
• gross industry profits increase (revenue less direct costs of production and
compliance costs), and
• industry employment decreases due to decreased production levels and/or line
closures.
No product-line or facility closures are predicted with the floor option, including lost
production costs. Six product-line closures and three facility closures are predicted with the
above-the-floor option. However, it should be noted that the estimates of facility and
product-line closures are sensitive to the accuracy of the baseline characterization of facilities
and the estimation of their costs to comply with the NESHAP.
Although total gross profits for solvent extraction facilities are projected to increase
1.7 percent under the floor option, some facilities are projected to experience decreased gross
profits. Facilities with profit losses are those with higher than average variable production
costs (labor, energy, and materials) and/or compliance costs relative to the facilities with
profit gains. Facilities with profit gains benefit from increased market price and the ability to
pass some of the regulatory costs on to consumers.
Table 4-6 separates the number of facilities projected to experience profit losses from
those projected to experience profit gains. The factors differentiating facilities with profit
losses and those with profit gains are as follows:
• profit losers have smaller capacity facilities (one-third of the size of profit
gainers),
• profit losers have greater solvent loss ratios per ton of oilseed (two to four times
greater than profit gainers), and
• profit losers have greater compliance costs per ton of oilseed (four times greater
than profit gainers).
4.4.3 Social Costs of the Regulation
The social costs of a regulation are traditionally measured as changes in both
consumer and producer economic welfare (see Appendix A for a more complete discussion).
4-16
-------
Table 4-6. Summary of Distributional Industry Impacts of Vegetable Oil Production
NESHAP: 1995a
Floor with LP"
Above-the-Floor with LP"
Solvent Extraction Facilities
Number
Facility Capacity (tpd)
Total
Average Per Facility
Annual Solvent Loss
Total (103 gallons)
Gallons per Ton of Oilseed
Incremental Compliance Costs
Total ($103/yr)
Per ton of oilseed
Change in Gross Profits ($103/yr)c
Change in Employment (FTEs)d
With
Profit Loss
21
11,362
541
2,419
0.84
$2,502
$0.87
-$1,184
-7
With Profit
Gain
85
161,841
1,904
13,577
0.28
$9,649
$0.20
$24,925
-5
Total
106
173,203
1,634
15,996
0.31
$12,151
$0.24
$23,741
-12
With
Profit
Loss
22
13,610
619
2,682
0.84
$54,191
$16.93
-$16,885
-213
With Profit
Gain Total
84
159,593
1,900
13,314
0.28
$149,513
$3.14
$675,106
-137
106
173,203
1,634
15,996
0.31
$203,704
$4.01
$658,222
-350
a Facilities that began operation in 1996 or later years are not included in the analysis.
b LP represents lost production costs that may be incurred due to the downtime to install new capital
equipment.
0 Gross profits calculated as revenue less costs of production including oilseed costs.
d FTEs = Full-time equivalents.
The vegetable oil NESHAP directly affects consumers and producers of vegetable oils and
meals. Consumers experience reductions in economic welfare (i.e., consumer surplus) due to
increased market prices and decreased market quantity. Producers may experience either
increases or decreases in economic welfare (i.e., producer surplus) as a result of increased
market prices, decreased costs of production, and imposition of the compliance costs.
Table 4-7 compares the welfare effects of the NESHAP on domestic consumers,
domestic producers, and foreign producers. Consumer surplus decreases in all eight
4-17
-------
Table 4-7. Distribution of the Social Costs Associated with the Vegetable Oil
Production NESHAP: 1995a
Floor with LP" Above-the-Floor with LP"
Social Cost Component
Consumer Surplus
Corn Oil
Cottonseed Oil
Soybean Oil
All Other Vegetable Oils
Corn Germ Feed
Cottonseed Meal
Soybean Meal
All Other Vegetable Meals
Producer Surplus
Domestic Producers
Solvent Extraction
Mechanical Extraction
Foreign Producers
Social Costs of Regulation
-$36,929
-$2,845
-$1,644
-$14,649
-$1,734
-$446
-$613
-$14,752
-$244
$24,846
$24,122
$23,741
$380
$725
-$12,083
Change in Value ($103)
-$842,168
-$40,554
-$19,207
-$343,893
-$24,470
-$16,212
-$7,865
-$385,322
-$4,645
$675,388
$663,860
$658,222
$5,638
$11,528
-$166,780
a Facilities that began operation in 1996 or later years are not included in the analysis.
b LP represents lost production costs that may be incurred due to the downtime to install new capital
equipment.
vegetable oil and meal markets under both regulatory scenarios. Consumer surplus losses
range from $36.9 million for the floor option to $842.2 million for the above-the-floor option
including lost production costs. Domestic producer surplus increases under both regulatory
scenarios, although, as noted in Section 4.4.2, some producers gain while others lose.
Foreign producer surplus increases as a result of higher prices and increased imports. In
total, the social costs of the regulation range from $12.1 million to $166.8 million.
4-18
-------
SECTION 5
SMALL BUSINESS IMPACTS
This regulatory action will potentially affect owners of vegetable oil and meal
processing facilities. Firms or individuals that own these facilities are legal business entities
that have the capacity to conduct business transactions and make business decisions that
affect the facility. The legal and financial responsibility for compliance with a regulatory
action ultimately rests with these owners who must bear the financial consequences of their
decisions. Environmental regulations like this rule affect all businesses, large and small, but
small businesses may have special problems in complying with such regulations.
The RFA of 1980 requires that special consideration be given to small entities
affected by federal regulation. The RFA was amended in 1996 by SBREFA to strengthen the
RFA's analytical and procedural requirements.
This section identifies the small businesses that will be affected by this NESHAP and
provides a screening-level analysis to assist in determining whether this rule imposes a
significant impact on a substantial number of small businesses. The screening-level analysis
described in this section is a "sales test," which computes the annualized compliance costs as
a percentage of sales for each company.1 In addition, this section provides information about
the likely impact on small businesses after accounting for producer responses to the
regulation and resulting changes in market prices and output for vegetable oils and meals.
Information on cottonseed facilities is provided separately because they make up a
disproportionate number of the facilities owned by small companies.
The SBA defines a small business involved in vegetable oil and meal processing as
follows:
• Corn Oil (NAICS 311221)—fewer than 750 total employees;
• Corn Oil Dry Milling (NAICS 311211)—fewer than 500 total employees;
'Company-level sales figures used for this analysis were obtained from publicly available sources.
5-1
-------
• Cottonseed Oil (NAICS 311223)—fewer than 1,000 total employees;
• Soybean Oil (NAICS 311222)—fewer than 500 total employees; and
• All Other Vegetable Oils (NAICS 311223)—fewer than 1,000 total employees.
Based on these definitions, 15 companies, which are listed in Table 5-1, can be classified as
small or potentially small businesses. Two firms without available employment data are
included as potentially small businesses. As of 1995, these 15 companies owned and
operated 21 facilities, or 20 percent of all solvent extraction facilities.
For the purposes of assessing the impact of this rule on these small businesses, the
Agency calculated the share of annual compliance cost relative to baseline sales for each
company. For this screening-level analysis, annual compliance costs were defined as the
engineering control costs imposed on these companies; thus, they do not reflect the changes
in production expected to occur in response to imposition of these costs and the resulting
market adjustments. Table 5-2 compares total compliance costs for small and large
companies for each of the two scenarios. Small companies own 20 percent of the facilities
and incur compliance costs of $2.2 million (18 percent) under the floor with lost production
costs scenario and $21.1 million (10 percent) under the above-the-floor with lost production
costs scenario.
Mean, minimum, and maximum CSRs under each scenario are presented in
Table 5-3. Mean CSRs for all companies are 0.16 percent under the floor with lost
production costs scenario and 1.62 percent under the above-the-floor with lost production
costs scenario. Under the floor scenario, the mean CSR for small companies is 0.30 percent,
ranging from a low of 0.03 to 0.61 percent, while the mean ratio is 0.04 percent for large
companies. Under the above-the-floor scenario, the mean CSR for small companies is
2.97 percent, and the mean CSR for large companies is 0.45 percent. Table 5-4 presents the
distribution of CSRs for small, large, and all companies combined for each of the scenarios.
All large and small companies have CSRs less than 1 percent under the floor scenario. Under
the above-the-floor scenario, 13 small companies and three large companies have CSRs
above 1 percent.
5-2
-------
Table 5-1. Summary Data for Small Companies: 1995
Company Name
Chickasha Cotton Oil Company
Delta Oil Mill
Hartsville Oil Mill Incorporated
J.G. Boswell
Oilseeds International
Osceola Products
Owensboro Grain Company
Plains Cooperative Oil Mill, Inc.
Planter's Cotton Oil Mill
Producers Cooperative Mill
Rito Partnership3
Sessions Company
Southern Soya Corporation
Valley Cooperative Mills
Yazoo Valley Oil Mill, Inc.
TOTAL
Number of
Affected Facilities
4
1
1
1
1
2
1
1
1
1
1
1
1
1
3
21
Sales
($ million)
$93.0
$22.5
$20.0
$80.0
NA
$50.0
$450.0
$128.0
$35.0
$35.0
NA
$30.0
NA
$32.0
$113.7
NA
Total
Employment
600
90
100
1,000
NA
170
195
108
100
100
NA
100
89
100
300
NA
Year Reported
1995
1996
1996
1993
NA
1996
1996
1995
1996
1996
NA
1994
1996
1992
1996
NA
NA = Not available.
a Rito Partnership operates a rice oil facility that will not be subject to the regulation. It is included in the aggregate totals to maintain
confidentiality in the tables that present production data.
Note: The Small Business Administration (SBA) defines a small business for industries affected by this regulation as follows:
Corn Oil (NAICS 311221) = fewer than 750 total employees
Corn Oil Dry Milling (NAICS 311211) = fewer than 500 total employees
Cottonseed Oil (NAICS 311223) = fewer than 1,000 total employees
Soybean Oil (NAICS 311222) = fewer than 500 total employees
All Other Vegetable Oils (NAICS 311223) = fewer than 1,000 total employees
Sources: Information Access Corporation. 1997 Business Index [computer file]. Foster City, CA: Information Access Corporation.
The Dialog Corporation. 1997. Dun &Bradstreet Market Identifiers [computer file]. New York, NY: Dun & Bradstreet.
.
-------
Table 5-2. Capacity and Compliance Cost Comparisons for Small and Large Companies: 1995
Capacity
(tons per day)
Company
Size
Small
Large
TOTAL
Number of
Companies
15
16
31
Number of
Facilities
21
85
106
Share of
Facilities
19.8%
80.2%
100.00%
Total
NR
NR
172,803
Share
NR
NR
100.00%
Compliance Costs ($103)
Floor with
LPa
$2,153
$9,998
$12,151
Above-the-Floor
with LP
$21,089
$182,615
$203,704
NR = Not reported to avoid disclosure of CBI.
a LP represents lost production costs that may be incurred due to the downtime to install new capital equipment.
-------
Table 5-3. Summary of Cost-to-Sales Ratios for the Vegetable Oil NESHAP: 1995 (%)
Scenario
Company Size
Small
Large
TOTAL
Mean
0.30
0.04
0.16
Floor with LPa
Min
0.03
<0.01
<0.01
Max
0.61
0.11
0.61
Mean
2.97
0.45
1.62
Above-the-Floor with LP
Min
0.51
<0.01
<0.01
Max
6.11
1.30
6.11
LP represents lost production costs that may be incurred due to the downtime to install new capital equipment.
-------
Table 5-4. Distribution of Cost-to-Sales Ratios for the Vegetable Oil NESHAP: 1995
Scenario
Small Companies
0-1%
1-2%
>2%
TOTAL
Large Companies
0-1%
Y1 1-2%
ON
>2%
TOTAL
All Companies
0-1%
1-2%
>2%
TOTAL
Floor with LPa
Number
15
0
0
15
16
0
0
16
31
0
0
31
Share
100.0%
0.0%
0.0%
100.0%
100.0%
0.0%
0.0%
100.0%
100.0%
0.0%
0.0%
100.0%
Above-the-Floor
Number
2
2
11
15
13
3
0
16
15
5
11
31
with LP
Share
13.3%
13.3%
73.3%
100.0%
81.3%
18.8%
0.0%
100.0%
48.4%
16.1%
35.5%
100.0%
LP represents lost production costs that may be incurred due to the downtime to install new capital equipment.
-------
Because small businesses affected by this regulation own many cottonseed processing
facilities and several of these facilities have closed since the baseline year of analysis,2 the
Agency undertook additional analysis to determine whether companies that operate these
facilities will experience significant economic impacts as a result of the regulation. CSRs
were calculated both for the plants included in the 1995 baseline and then for currently
operating plants using publicly available sales data. Under the floor with lost production
costs scenario, the average CSR for small companies that owned cottonseed processing
facilities in 1995 falls below 1 percent (0.31 percent) with a maximum value of 0.59 percent
(see Tables 5-5 and 5-6). None of these companies have CSRs above 1 percent. However,
under the above-the-floor with lost production costs scenario, the average CSR in 1995 is 2.9
percent with a maximum value of 4.3 percent. All ten companies have CSRs above 1
percent, eight of which are greater than 2 percent. Although ten cottonseed facilities have
closed since 1995, excluding the compliance costs for facilities not currently operating does
not significantly alter the analysis. The one exception is that two companies no longer
operate any cottonseed facilities and therefore are not affected under the above-the-floor
scenario.
CSRs for companies that own cottonseed facilities were also calculated using facility
revenues as the denominator rather than publicly available company sales data. These facility
CSRs exclude revenues that companies may generate from operations other than vegetable oil
production. In this case, for the nine cottonseed processing facilities operating in 1995 and
still currently operating that are owned by small businesses, the average facility-level CSR is
0.28 percent and ranges from 0.05 percent to 0.52 percent (see Table 5-7). Thus, the results
are not substantially different using facility-level revenues instead of company sales. In
addition, the facility CSRs for compliance cost data cottonseed facilities were recomputed
using 1999 price data and projected to 1999 using the producer price index. The purpose of
this analysis was to determine whether cottonseed facility CSRs would change substantially
using more recent data. For the cottonseed processing facilities operating in 1995 and still
operating currently that are owned by small businesses, the average facility-level CSR
increased to only 0.33 percent with a range from 0.06 percent to 0.58 percent.
2In 1995, 25 cottonseed facilities were operating. Since then, ten of these facilities have closed. Closures have
been particularly high in this industry because cottonseed is being used increasingly as a dairy feed. The
value of cottonseed as a feed product has risen relative to its value in producing oil and meal products
(Wedegaertner, 1999). Hence, cottonseed oil and meal processing has become less profitable.
5-7
-------
Table 5-5. Summary of Company Cost-to-Sales Ratios for Companies That Own Cottonseed Facilities: 1995 (%)
Operating 1995"
Floor with LPC
Company Size
Small
Large
TOTAL
Mean
0.31
0.02
0.26
Min
0.03
0.02
0.02
Max
0.59
0.02
0.59
Above-the-Floor with LP
Mean
2.91
0.37
2.49
Min
1.01
0.23
0.23
Max
4.32
0.50
4.32
Floor with
Mean
0.22
0.01
0.18
Min
0.03
<0.01
<0.01
Currently Operating6
LP
Max
0.51
0.02
0.51
Above-the-Floor with LP
Mean
2.35
0.24
1.96
Min
0.86
<0.01
<0.01
Max
4.32
0.47
4.32
a Companies with cottonseed facilities operating in 1995.
b Companies with cottonseed facilities currently operating.
0 LP represents lost production costs that may be incurred due to the downtime to install new capital equipment.
oo
-------
Table 5-6. Distribution of Cost-to-Sales Ratios for Companies That Own Cottonseed Facilities: 1995
Operating 1995a
Floor with LPC
Small Companies
0-1%
1-2%
>2%
TOTAL
Large Companies
0-1%
1-2%
>2%
TOTAL
All Companies
0-1%
1-2%
>2%
TOTAL
Number
10
0
0
10
2
0
0
2
12
0
0
12
Share
100.0%
0.0%
0.0%
100.0%
100.0%
0.0%
0.0%
100.0%
100.0%
0.0%
0.0%
100.0%
Above-the-Floor with LP
Number
0
2
8
10
2
0
0
2
2
2
8
12
Share
0.0%
20.0%
80.0%
100.0%
100.0%
0.0%
0.0%
100.0%
16.7%
16.7%
66.7%
100.0%
Currently Operating13
Floor with LP
Number
9
0
0
9
2
0
0
2
11
0
0
11
Share
100.0%
0.0%
0.0%
100.0%
100.0%
0.0%
0.0%
100.0%
100.0%
0.0%
0.0%
100.0%
Above-the-Floor with LP
Number
1
3
5
9
1
0
0
1
2
3
5
10
Share
11.1%
33.3%
55.6%
100.0%
100.0%
0.0%
0.0%
100.0%
27.3%
27.3%
45.5%
100.0%
a Companies with cottonseed facilities operating in 1995.
b Companies with cottonseed facilities currently operating.
0 LP represents lost production costs that may be incurred due to the downtime to install new capital equipment.
-------
Table 5-7. Summary of Facility Cost-to-Sales Ratios for Cottonseed Facilities: 1995 (%)
Operating 1995"
Currently Operating13
Floor with LPC
Company Size
Small
Large
TOTAL
Mean
0.54
0.25
0.43
Minimum
0.05
0.07
0.05
Maximum
1.57
0.64
1.57
Mean
0.28
0.20
0.25
Minimum
0.05
0.07
0.05
Maximum
0.52
0.52
0.52
a Facilities operating in 1995.
b Facilities currently operating.
0 LP represents lost production costs that may be incurred due to the downtime to install new capital equipment.
-------
In summary, the Agency's screening analysis does not suggest a significant negative
impact on a substantial number of small companies under the MACT floor alternative
promulgated by the EPA (this scenario includes lost production costs). The economic
impacts model verifies that the effects of taking into account market adjustments are small.
However, the potential for negative impacts is greater under the above-the-floor scenario.
Table 5-8 summarizes the economic impacts of the regulation on small companies.
Gross profits are projected to increase by 0.2 percent for small companies under the floor
with lost production costs scenario. The projected effects of the above-the-floor with lost
production costs scenario are greater than the floor scenario, but no closures are expected
under either of the scenarios.
5-11
-------
Table 5-8. Summary of Small Business Impacts of Vegetable Oil Production NESHAP: 1995
to
Floor with LPa
Above-the-Floor with LPa
Changes From Baseline
Revenues ($103)
Costs ($103)
Post-regulatory
Production
Gross Profits ($103)
Operating Entities
Product Lines
Facilities
Employment
Baseline
$780,636
$665,767
$0
$665,767
$114,869
21
20
NR
Absolute
-$793
-$1,010
$2,131
-$3,141
$217
0
0
-6
Percent
-0.1%
-0.2%
NA
-0.5%
0.2%
0.0%
0.0%
NR
Absolute
$13,629
-$7,777
$19,586
-$27,363
$21,405
0
0
-59
Percent
1.7%
-1.2%
NA
-4.1%
18.6%
0.0%
0.0%
NR
NA = Not available.
NR = Not reported to avoid disclosure of CBI.
a LP represents lost production costs that may be incurred due to the downtime to install new capital equipment.
-------
SECTION 6
ASSUMPTIONS AND LIMITATIONS OF THE ECONOMIC MODEL
In developing the economic model of the vegetable oil and meal markets, several
assumptions were necessary to make the model operational. These assumptions are in
addition to the numerical assumptions described in Section 4.3.2 (i.e., elasticity values and
allocation of both seed costs and compliance costs to individual products). In this section,
each operational assumption is listed and explained. Possible impacts and limitations of the
model resulting from each assumption are then described.
Assumption: The domestic markets for crude vegetable oils and meals are perfectly
competitive.
Explanation: Assuming that the markets for crude vegetable oils and meals are perfectly
competitive implies that individual producers cannot individually affect the prices they
receive for their products. The measures available to determine whether a market is perfectly
competitive are the four-firm concentration ratios (CR4) and the Herfindahl-Hirschman
indexes (HHIs). The most recently available measures are based on SIC code-level data for
1992. Based on the 1992 data, the CR4s for vegetable oils range from a low of 62 percent for
SIC 2074 (Cottonseed Oil Mills) to a high of 89 percent for SIC 2076 (Vegetable Oil Mills,
Except Corn, Soybean, and Cottonseed). Similarly, the HHIs range from a low of 1,430 for
SIC 2074 to a high of 2,119 for SIC 2076. These concentration measures considered in
isolation imply that the vegetable oil industry is moderately imperfectly competitive.
However, given the homogeneous nature of crude vegetable oils and meals and that there
appear to be large economies of scale in their production, the assumption of perfect
competition is appropriate.
Possible impact: If the markets for crude vegetable oils and meals were in fact imperfectly
competitive, implying that individual producers can exercise market power and thus affect
the prices they receive for their products, the economic model would understate possible
price increases due to the regulation and the social costs of the regulation. Under imperfect
competition, producers would be able to pass along more of the costs of the regulation to
consumers; thus, consumer surplus losses would be greater.
6-1
-------
Assumption: The U.S. is not a price-taker on the world market for crude vegetable oils
and meals; that is, the U.S. may influence the price of these products on the world
market.
Explanation: Assuming that the U.S. is not a price-taker on the world market for these
products implies that the U.S. is "large" relative to the rest of the world. That is, the U.S.
ships a sufficient quantity of these products that changes in the volume of products imported
or exported may affect prices in the world market. Thus, producers in the U.S. have the
opportunity to pass along some portion of the costs of the regulation to the consumers of
crude vegetable oils and meals (i.e., to the facilities that further process the product for
consumption by humans or animals).
Possible impact: If the U.S. was instead a price-taker on the world market for crude
vegetable oils and meals, producers in the U.S. would not be able to pass along the costs of
the regulation to consumers of these products. If U.S. companies that export crude vegetable
oils and meals attempted to raise prices as a result of the regulation, importing countries
would instead purchase from countries other than the U.S. Likewise, U.S. companies would
be unable to raise prices of these products domestically because consumers of crude
vegetable oils and meals would instead import all these products at the lower world price.
Thus, U.S. producers would have to fully absorb the costs of the regulation, and consumers
would bear none of the costs of the regulation. The potential impact under this scenario on
individual producers is estimated by the CSRs that are documented in Section 4.
Assumption: Crude vegetable oils and meals are produced in fixed proportions relative
to the oilseed inputs.
Explanation: Assuming that crude vegetable oils and meals are produced in fixed
proportions implies that facilities cannot alter the ratio of oil to meal from a given oilseed in
response to the regulation. This assumption is appropriate in light of our calculations using
production data from the USDA and information provided by the trade associations (see
Section 2.1).
Possible impact: If, in fact, facilities could alter the proportion of oil to meal, then facilities
could potentially alter the proportion in response to the regulation and thus could alleviate
some of the burden of the regulation. In particular, a facility may choose to produce more of
the product that generates higher revenues by weight (i.e., oil). However, it appears
technically infeasible for facilities to alter the proportion.
6-2
-------
Assumption: The markets for by-products and co-products of the vegetable oil and
meal production process will be unaffected by the regulation.
Explanation: Products such as cottonseed linters, hulls, and lecithin are produced either as
by-products or co-products of the production process for vegetable oils and meals. These
markets could be potentially affected by the regulation if the regulation changes the
throughput of oilseeds in the production process. They are not modeled in this analysis
because there is insufficient data to characterize their markets (e.g., prices and output at each
facility), and they generate a relatively smaller share of revenue compared to oils and meals.
Possible impact: Because the prices and output of by-products and co-products are assumed
unchanged as a result of the regulation, the analysis may either understate or overstate
changes to revenue and costs for individual facilities. It is expected that the net effects of
these potential revenue and cost changes on individual facilities are minimal and would not
significantly alter the primary conclusions of the EIA.
Assumption: The markets for specialty use products, which are produced primarily
from crude meal, will be unaffected by the regulation.
Explanation: Specialty use products that are produced by vegetable oil mills include
products such as soy protein, tofu, and infant formula ingredients. The production processes
for these products are proprietary to the plant producing them, and only a handful of plants
produce each of these products. Thus, data to characterize these markets are unavailable.
Possible impact: These specialty use products generate higher revenues than crude meal;
thus, revenues are understated in the model for the plants that produce them. Therefore, the
impacts of the regulation on these plants appear larger than they may be in actuality.
Assumption: The baseline year of the analysis, which is 1995, is representative of a
typical year for the industry.
Explanation: The engineering costs of the regulation are estimated for all facilities that
produced crude vegetable oils and meals in 1995. In order for the economic model to be
consistent, all costs, prices, and quantities must be denominated in the same year. In
addition, for consistency between market-level data and facility-level data, both must be
representative of the same year.
Possible impact: If 1995 was a good year relative to typical conditions (i.e., with high output
prices and low input prices), then the impacts of the regulation would appear to be smaller (in
6-3
-------
percentage terms) than they would be for a typical year. As discussed in Section 2, the
markets for vegetable oils and meals and for the oilseeds used in their production exhibit a
great deal of volatility over time. Based on price data for the 1990s, 1995 appears to have
been a relatively good year for vegetable oil producers. To evaluate whether the results of the
economic model would differ substantially using more recent data, facility-level CSRs were
calculated using 1999 data. Based on these calculations, the CSRs increased by less than 0.1
percent except for one facility. Thus, it appears that the results of a model using 1999 data
would not differ substantially from the results using 1995 data. It is likely that the projected
percentage changes in the economic variables would increase slightly, but the overall
economic effects of the MACT floor scenario would still be small. According to a recent
USD A publication, the economic conditions of the vegetable oil and meal industries have
improved in 2000 (USDA, 2000).
Assumption: The markets for specialty oils and meals, which include canola, flaxseed,
peanut, safflower, and sunflower, are sufficiently substitutable that they can be
considered in the same market.
Explanation: Because relatively few facilities produce each of the specialty oils and meals,
and because these products are grouped in the NAICS definitions, they are considered to be
in a single market in the economic analysis. The price in the model for the oilseed input and
the oil and meal outputs are computed as weighted averages of the prices of the individual
products. By grouping these products into a single market, it is assumed that these products
are highly substitutable for one another.
Possible impact: The impacts of the regulation on speciality oils and meals may be
potentially understated or overstated because they are grouped in the analysis. However, the
impacts on each as projected by the economic model are small. There are only a handful of
producers of each of these products, and data on prices of some of the products are not
available; thus, it is not possible to model each market separately.
6-4
-------
REFERENCES
1997 Directory of Corporate Affiliations. 1997. Vol.5: International Public and Private
Companies. New Providence, RI: National Register Publishing.
Agriculture and Agri-Food Canada. September 1998. "An Evaluation of Oilseed Trade
Liberalization." .
Ailor, David C., National Oilseed Processors Association, to Linda Chappell, U.S.
Environmental Protection Agency. May 5, 1998. "NOPA Responses on EPA's
March 1998 Draft 'Vegetable Oil Production: Industry Profile and
Initial/Supplemental to Questions on the Industry'."
Ailor, David C., National Oilseed Processors Association. July 25, 2000. "Comments of the
Vegetable Oil MACT Coalition on the National Emission Standards for Hazardous
Air Pollutants: Solvent Extraction for Vegetable Oil Production, 40 C.F.R. Part 63
SubpartGGGG, Air Docket No. A-97-59." Memorandum.
Baumel, C.P. 1996. "Expanded Soybean Processing Capacity Could Impact Margins."
Feedstuffs 68(May 20):37-41.
Brenner, Kyd, Corn Refiners Association, Inc. E-mail to Mary Muth, Research Triangle
Institute. March 15, 1999.
Chern, Wen S., Edna T. Loehman, and Steven T. Yen. 1995. "Information, Health Risk
Beliefs, and the Demand for Fats and Oils." The Review of Economics and Statistics
77(3):555-64.
Dun & Bradstreet. 1998. Dun & Bradstreet Market Identifiers [computer file]. New York,
NY: Dialog Corporation.
Durham, J.F. June 12, 1998. "Vegetable Oil MACT—Notes for 6/16/98 Teleconference."
Delivered to participants of the Vegetable Oil MACT teleconference.
R-l
-------
Hoover's Incorporated. 1998. Hoover's Company Profiles. Austin, TX: Hoover's
Incorporated, .
Huang, K.S. 1993. U.S. Demand for Food: A Complete System of Price and Income Effects.
U.S. Department of Agriculture, Economic Research Service. Bulletin 1821.
Washington, DC.
Information Access Corporation. 1997. Business Index [computer file]. Foster City, CA:
Information Access Corporation.
Lence, S.H., DJ. Hayes, and W.H. Meyers. 1995. "The Behavior of Forward-Looking Firms
in the Very Short Run." American Journal of Agricultural Economics
77(November): 922-93 4.
Marine, C. Lockwood, Marine Associates. Facsimile to Mary Muth, Research Triangle
Institute. March 10, 1999.
Meyers, W.H., P. Westhoff, D.L. Stephens, B.L. Buhr, M.D. Helmar, and KJ. Stephens.
January 1993. "FAPRIU.S. Agricultural Sector Elasticities Volume I: Crops."
Center for Agricultural and Rural Development, Iowa State University, Ames, Iowa.
Technical Report 92-TR 25.
Midwest Research Institute. November 1995. "Emission Factor Documentation for AP-42,
Section 9.11.1. Vegetable Oil Processing." Prepared for the U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Emission Factor
and Inventory Group. Gary, NC: Midwest Research Institute.
Mitsubishi Corporation. Annual Report 1996. .
National Cotton Council of America. July 25, 2000. "Comments of the National Cotton
Council and National Cottonseed Products Association on the Proposed National
Emission Standards for Hazardous Air Pollutants: Solvent Extraction for Vegetable
Oil Production (65 FR 34252; May 26, 2000) (Air Docket No. A-97-59).
Memorandum.
Standard and Poor's Corporations [computer file]. Palo Alto, CA: Dialog Information
Service.
R-2
-------
Ugarte, D., University of TN at Knoxville. Telecon with Mary Muth, Research Triangle
Institute. March 16, 1999.
U.S. Bureau of Census. 1998a. U.S. Exports History: Historical Summary 1993-1997 on
CD-ROM [machine readable data file]. Washington: Bureau of the Census.
U.S. Bureau of Census. 1998b. U.S. Imports History: Historical Summary 1993-1997 on
CD-ROM [machine readable data file]. Washington, DC: Bureau of the Census.
U.S. Department of Agriculture. 1992. Agricultural Statistics 1992. Washington, DC:
Government Printing Office.
U.S. Department of Agriculture. 1997a. Agricultural Statistics 1997'. Washington, DC:
Government Printing Office.
U.S. Department of Agriculture. 1997b. Agricultural Prices: 1996 Summary. Washington,
DC: Government Printing Office.
U.S. Department of Agriculture. 1997c. Oil Crops Situation and Outlook Yearbook.
Washington, DC: Government Printing Office.
U.S. Department of Agriculture, Economic Research Service. 1997. Feed Yearbook.
[computer file]. Last updated April 1997.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook.
November 13, 1996.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook. March 13,
1996.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook. October
14, 1997.
U.S. Department of Agriculture, Economic Research Service. 1998. Oil Crops Yearbook.
[computer file]. Last updated January 1998.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook.
November 12, 1998.
U.S. Department of Agriculture, Economic Research Service. Rice Yearbook, [computer
file]. Last updated November 1998.
R-3
-------
U.S. Department of Agriculture, Economic Research Service. Sugar and Sweetener
Yearbook, [computer file]. Last updated January 1998.
U.S. Department of Agriculture, Economic Research Service. Oil Crops Outlook.
September 13, 1999a.
U.S. Department of Agriculture, Economic Research Service. 1999b. Oil Crops Yearbook.
[computer file]. Last updated November 1999.
U.S. Department of Agriculture, Economic Research Service. October 1999c. "Oil Crops
Yearbook—Summary." ERS-OCS-1099.
U.S. Department of Agriculture. 2000. Agricultural Statistics 2000. Washington, DC:
Government Printing Office.
U.S. Department of Agriculture, Economic Research Service. 2000. Oil Crops Outlook.
September 13, 2000.
U.S. Department of Agriculture, Economic Research Service. 2000. Feed Yearbook.
[computer file]. Last updated May 2000.
U.S. Department of Agriculture, National Agriculture Statistics Service. 2000. Agricultural
Prices: J999 Summary. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. National Monthly Population
Estimates: 1980-2000. . Last updated November 2, 2000.
U.S. Department of Commerce, Bureau of the Census. 1995. 1992 Census of Manufactures,
Industry Series—Fats and Oils. MC92-1-20D. Washington, DC: Government
Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1995. 1992 Census of Manufactures,
Industry Series—Grain Mill Products. MC92-1-20D. Washington, DC:
Government Printing Office.
U. S. Department of Commerce. 1996a. 1995 Current Industrial Reports—Fats and Oils:
Oilseed Crushings. M20J. Washington, DC: Government Printing Office.
R-4
-------
U. S. Department of Commerce. 1996b. 7995 Current Industrial Reports—Fats and Oilseed
Crushings, Production, Consumption, and Stocks. M20K. Washington, DC:
Government Printing Office.
U.S. Department of Energy, Energy Information Administration. State Energy Price and
Expenditure Report 1994 [computer file].
. Last modified June 25, 1997.
U.S. Department of Labor, Bureau of Labor Statistics. BLS LABSTAT database.
. Data extracted on March 3, April 15, and April 22, 1998.
Wedegaertner, T., Cotton, Inc. Telecon with Mary Muth, Research Triangle Institute.
August 17, 1999.
Yen, S.T., and W.S. Chern. 1992. "Flexible Demand Systems with Serially
Correlated Errors: Fat and Oil Consumption in the United
States." American Journal of Agricultural Economics 74(3):689-697.
Zukor, C. and B. Riddle. June 1, 1998. "Vegetable Oil Production Process Description."
Alpha-Gamma Technologies, Inc. Memorandum to the Vegetable Oil NESHAP
Project File.
Zukor, C. and J. Snyder, Alpha-Gamma Technologies, Inc. November 10, 2000. "Final
Summary of Emission Reductions and Control Costs Associated with Achieving the
MACT Floor and a Control Option Above the MACT Floor." Memorandum
submitted to the Vegetable Oil NESHAP Project File.
R-5
-------
APPENDIX A
ECONOMIC MODEL OF THE U.S. VEGETABLE OIL AND MEAL MARKETS
Implementation of the MACT standards will affect the costs of production in the U.S.
vegetable oil industry for existing solvent extraction plants. Responses at the facility level to
these additional costs collectively determine the market impacts of the regulation.
Specifically, the cost of the regulation may induce some facilities to alter their current level
of production or to close. These choices affect, and in turn are affected by, the market price
for each product. The EIA employs standard concepts in microeconomics to model the
supply of each product and the impacts of the regulation on production costs and the output
decisions of facilities. The main elements are the following:
• characterize production of each product at the individual facility and market
levels,
• characterize demand for each product, and
• develop the solution algorithm to determine the new post-regulatory equilibrium.
A.I Supply of Vegetable Oils and Related Products
Market supply of vegetable oil and related products (Qs) can be expressed as the sum
of domestic and foreign supply, or imports, that is,
Q^q' + q1 (A.I)
where qs is the domestic supply of a particular product, which is the sum of production from
solvent and mechanical extraction facilities, and q1 is the foreign supply, or imports. Each of
these supply components is described below.
A. 1.1 Solvent Extraction Facilities
Individual supply functions were developed for each vegetable oil and related product
at solvent extraction facilities. Producers of vegetable oils and related products have the
ability to vary output in the face of production cost changes. Upward-sloping supply curves
for vegetable oils and related products were developed to allow these facilities to respond in
A-l
-------
this manner to the imposition of regulatory costs. For this analysis, the generalized Leontief
profit function was used to derive the supply curve for vegetable oils and related products at
each facility. This functional form was appropriate given the fixed-proportion material input
(oilseeds) and the variable-proportion inputs of chemicals, labor, electricity, and energy. By
applying Hotelling's lemma to the generalized Leontief profit function, the following general
form of the supply functions for each vegetable oil and related product was obtained:
Q = v- + —
J J
—
2
(A.2)
where p is the net market price for each product after subtracting the cost of the oilseed input,
Ij is the variable production cost variable (described below), YJ and P are model parameters,
and j indexes producers (i.e., individual solvent extraction facilities). The theoretical
restrictions on the model parameters that ensure upward-sloping supply curves are YJ > 0 and
p<0.
Figure A-l illustrates the theoretical supply function represented by Eq. (A.2). As
shown, the upward-sloping supply curve is specified over a productive range with a lower
32
bound of zero that corresponds with a shutdown price equal to -t— -I and an upper bound
4j*
given by the productive capacity of qMj that is approximated by the supply parameter YJ. The
curvature of the supply function is determined by the P parameter.
To obtain the empirical specification of Eq. (A.2), the variable production cost
variable, Ij, was first constructed. It was calculated as a cost-share weighted index of
regional- and state-level average hourly earnings (Wj), average fuel prices (fj), and electricity
prices (ej).1 The Ij variable therefore varies across facilities because of all three variables
(w,f,e). The cost shares used to weight the variable cost components vary by the four
'Chemicals are an input into vegetable oil and meal production but are not included here because they are an
extremely small proportion of total production costs and because we expect little variation in costs across
facilities.
A-2
-------
$/qJ
li
Figure A-l. Theoretical Supply Function for Solvent Extraction Facilities
vegetable oil product classifications—corn oil, cottonseed oil, soybean oil, and all other
vegetable oils. These shares, which were computed from the U.S. Department of
Commerce's 7992 Census of Manufactures, are shown in Table A-l.
Regional- and state-level wage, fuel, and electricity variables were converted into
indexes normalized to the median value of each variable. Table A-2 provides the normalized
indexes used for each product by state. This conversion allows each variable to be measured
in terms of a relative index for use in deriving the cost-share weighted variable production
cost index. The facility-specific index was computed as follows:
Ij = SLwj+Sffj + Seej
where SLis the cost share for labor, Sf is the cost share for fuel, and Se is the cost share for
energy used in the production of vegetable oils (from Table A-l).
Using these constructed values for Ij, the P parameter was computed by substituting
an econometrically estimated or assumed market supply elasticity for a particular vegetable
oil or related product (£), the average annual production level of solvent extraction facilities
(q), the variable production cost index (I), and the market price of the product (p) into the
following equation:
A-3
-------
Table A-l. Cost Shares of Variable Production Factors by Market
SIC
2046
2074
2075
2076
Market
Corn oil
Cottonseed oil
Soybean oil
Minor vegetable oils
Labor
36.4%
50.7%
44.3%
56.9%
Fuels
35.8%
13.9%
33.2%
20.2%
Electricity
27.9%
35.3%
22.5%
22.8%
Sources: U.S. Department of Commerce, Bureau of the Census. 1995. 1992 Census of Manufactures,
Industry Series—Fats and Oils. MC92-1-20D. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1995. 1992 Census of Manufactures,
Industry Series—Grain Mill Products. MC92-1-20D. Washington, DC: Government Printing
Office.
(A.3)
Baseline 1995 market prices and average annual production levels of each product were
provided in Table 4-1. The P parameter for each product was then calculated by
incorporating these values into Eq. (A.3).
Supply Function Intercept. The intercept of the supply function, jp approximates
the productive capacity and varies across products at each facility. This parameter does not
influence the facility's production responsiveness to price changes as does the P parameter.
Thus, the parameter ^ is used to calibrate the model so that each solvent extraction facility's
supply equation is exact using the baseline production data for 1995.
Regulatory Response. The production decisions at solvent extraction facilities are
affected by the total annual compliance costs, Cj, which are expressed per ton of product.2
Each supply equation is directly affected by the regulatory control costs, which enter as a net
2Total annual compliance cost estimates, provided by EPA's engineering analysis, include capital costs, annual
operating and maintenance costs, and applicable monitoring costs. These costs are estimated at the facility
level and allocated to each product using revenue shares at each facility.
A-4
-------
Table A-2. Variable Cost Indexes by Region: 1995
State
AL
AR
AZ
CA
DE
GA
IA
IL
IN
KS
KY
LA
MD
MN
MO
MS
MT
NC
ND
NE
OH
OK
SC
TN
TX
VA
Labor Index
0.79
0.81
0.86
1.20
0.87
0.90
1.13
1.10
1.10
1.03
1.02
0.83
1.15
1.05
1.06
0.72
1.06
0.87
0.95
0.98
1.22
0.79
0.78
1.05
0.88
1.04
Fuel Index
0.77
0.95
1.14
1.03
0.99
1.01
1.17
1.21
1.00
1.01
1.01
0.85
1.04
1.04
1.15
0.92
1.04
0.98
0.92
1.16
1.12
0.81
0.85
0.90
0.99
0.83
Electricity Index
0.93
.04
.27
.60
.04
.03
0.87
1.17
0.89
1.11
0.73
0.95
1.19
0.99
1.04
1.01
0.74
1.11
1.06
0.90
0.93
0.92
0.91
1.02
0.96
0.94
Sources: U.S. Department of Labor, Bureau of Labor Statistics. BLS LABSTAT database.
. Data extracted on March 3, 1998.
U.S. Department of Energy, Energy Information Administration. State Energy Price and Expenditure
Report 1994 [computer file], . Last modified
June 25, 1997.
A-5
-------
price change (i.e., PJ - Cj). Thus, the supply function for each existing facility from Eq. (A.2)
becomes:
P
2
^
(A.4)
The total annual compliance costs per ton were calculated given the annual production per
facility and the regulatory cost estimates for each facility provided by the engineering
analysis.
A. 1.2 Mechanical Extraction Facilities
Mechanical extraction facilities are not directly affected by the regulation and were
modeled as a single representative supplier. Supply of vegetable oils and related products
from these facilities (qm) can be expressed by the following general formula for each product:
qm = Am |pj$ (A 5)
where p is the market price for the product, ^ is the domestic supply elasticity (see Appendix
B), and Am is a multiplicative supply parameter that calibrates the supply equation for each
product given data on price and the supply elasticity to replicate the observed 1995 level of
production from these facilities.
A.I.3 Foreign Supply (Imports)
Similar to mechanical extraction facilities, foreign producers are not directly affected
by the regulation but are included in the model as a single representative supplier. Supply of
vegetable oils and related products from foreign producers (q1) can be expressed by the
following general formula for each product:
q1 = A1 [p]^ (A.6)
where p is the market price for the product, ^ is the import supply elasticity (see Appendix
B), and A1 is a multiplicative supply parameter that calibrates the supply equation for each
A-6
-------
product given data on price and the foreign supply elasticity to replicate the observed 1995
level of imports.
A.2 Demand for Vegetable Oils and Related Products
Market demand for vegetable oil and related products (Qd) can be expressed as the
sum of domestic and foreign demand as follows:
Qd = qd + qx (A.7)
where qd is the domestic demand and qx is the foreign demand, or exports, as described
below. Each of these demand components is described below.
A. 2.1 Domestic Demand
Domestic demand for vegetable oils and related products can be expressed by the
following general formula for each product:
(A.8)
where p is the market price for the product, r|d is the domestic demand elasticity (see
Appendix B), and Bd is a multiplicative demand parameter that calibrates the demand
equation for each product given data on price and the domestic demand elasticity to replicate
the observed 1995 level of domestic consumption.
A.2.2 Foreign Demand (Exports)
Foreign demand, or exports, for vegetable oils and related products can be expressed
by the following general formula for each product:
qx=Bx[p]ix (A.9)
where p is the market price for the product, if is the assumed export demand elasticity (see
Appendix B), and Bx is a multiplicative demand parameter that calibrates the foreign demand
equation for each product given data on price and the foreign demand elasticity to replicate
the observed 1995 level of exports.
A-7
-------
A.3 Post-Regulatory Market Equilibrium Determination
Facility responses and market adjustments can be conceptualized as an interactive
feedback process. Facilities face increased production costs due to compliance, which causes
facility-specific production responses (i.e., output reduction). The cumulative effect of these
responses leads to an increase in the market price that all producers (affected and unaffected)
and consumers face, which leads to further responses by producers (affected and unaffected)
as well as consumers and thus new market prices, and so on. The new equilibrium after
imposing the regulation is the result of a series of iterations between producer and consumer
responses and market adjustments until a stable market price arises where total market supply
equals total market demand (i.e., Qs = Qd).
This process for determining equilibrium price (and output) with the increased
production cost is modeled as a Walrasian auctioneer. The auctioneer calls out a market
price for each product and evaluates the reactions by all participants (producers and
consumers), comparing total quantities supplied and demanded to determine the next price
that will guide the market closer to equilibrium (i.e., where market supply equals market
demand). Decision rules are established to ensure that the process will converge to an
equilibrium, in addition to specifying the conditions for equilibrium. The result of this
approach is a vector of prices with the regulation that equilibrates supply and demand for
each product.
The algorithm for deriving the post-compliance equilibria in all markets can be
generalized to five recursive steps:
1. Impose the control costs on each affected facility, thereby affecting their
supply decisions.
2. Recalculate the market supply of each vegetable oil product.
3. Determine the new prices via the price revision rule for all product markets.
4. Recalculate the supply functions of all facilities with the new prices, resulting
in a new market supply of each product. Evaluate market demand at the new
prices.
5. Go to Step 3, resulting in new prices for each product. Repeat until
equilibrium conditions are satisfied in all markets (i.e., the difference between
supply and demand is arbitrarily small for each and every product).
A-8
-------
A.4 Economic Welfare Impacts
The economic welfare implications of the market price and output changes of
vegetable oils and related products with the regulations can be examined using two slightly
different tactics, each giving a somewhat different insight but the same implications: changes
in the net benefits of consumers and producers based on the price changes and changes in the
total benefits and costs of these products based on the quantity changes. This analysis
focuses on the first measure—the changes in the net benefits of consumers and producers.
Figure A-2 depicts the change in economic welfare by first measuring the change in
consumer surplus and then the change in producer surplus. In essence, the demand and
supply curves previously used as predictive devices are now being used as a valuation tool.
This method of estimating the change in economic welfare with the regulation divides
society into consumers and producers. In a market environment, consumers and producers of
the good or service derive welfare from a market transaction. The difference between the
maximum price consumers are willing to pay for a good and the price they actually pay is
referred to as "consumer surplus." Consumer surplus is measured as the area under the
demand curve and above the price of the product. Similarly, the difference between the
minimum price producers are willing to accept for a good and the price they actually receive
is referred to as "producer surplus." Producer surplus is measured as the area above the
supply curve and below the price of the product. These areas can be thought of as
consumers' net benefits of consumption and producers' net benefits of production,
respectively.
In Figure A-2, baseline equilibrium occurs at the intersection of the demand curve, D,
and supply curve, S. Price is P{ with quantity Q,. The increased cost of production with the
regulations will cause the market supply curve to shift upward to S'. The new equilibrium
price of the product is P2. With a higher price for the product there is less consumer welfare,
all else being unchanged. In Figure A-2(a), area A represents the dollar value of the annual
net loss in consumers' benefits with the increased price. The rectangular portion represents
the loss in consumer surplus on the quantity still consumed, Q2, while the triangular area
represents the foregone surplus resulting from the reduced quantity consumed, Q\-Q2-
In addition to the changes in consumer welfare, there are also changes in producer
welfare with the regulations. With the increase in market price, producers receive higher
revenues on the quantity still purchased, Q2. In Figure A-2(b), area B represents the increase
in revenues due to this increase in price. The difference in the area under the supply curve up
A-9
-------
$/Q
Q2 Q1 Q/t
(a) Change in Consumer Surplus with Regulation
$/Q
P2
p.
Q2 Q1 Q/t
(b) Change in Producer Surplus with Regulation
$/Q
P2
p.
Q2 Q1 Q/t
(c) Net Change in Economic Welfare with Regulation
Figure A-2. Economic Welfare Changes with Regulation: Consumer and Producer
Surplus
A-10
-------
to the original market price, area C, measures the loss in producer surplus, which includes the
loss associated with the quantity no longer produced. The net change in producer welfare is
represented by area B-C.
The change in economic welfare attributable to the compliance costs of the regulation
is the sum of consumer and producer surplus changes, that is, - (A) + (B-C). Figure A-2(c)
shows the net (negative) change in economic welfare associated with the regulation as area
D. However, this analysis does not include the benefits that occur outside the market (i.e.,
the value of the reduced levels of air pollution with the regulation). Including this benefit
may reduce the net cost of the regulation or even make it positive.
A-ll
-------
APPENDIX B
ESTIMATES OF THE DEMAND AND SUPPLY ELASTICITIES FOR VEGETABLE
OILS AND MEALS
This appendix summarizes the demand and supply elasticities used in the economic
impacts model. Because oils and meals for each of corn germ, cottonseed, soybean, and all
other oilseeds combined are modeled individually, estimates of elasticities were required for
each. In cases for which sufficient data were available, EPA estimated the elasticities; in
other cases, the elasticities were obtained directly from the economics literature or assumed
based on similar products' elasticities.
Section B.I reviews previous elasticity estimates obtained from the economics
literature. In Section B.2, the elasticity estimation procedure used by EPA is described.
Section B.3 discusses the results of the estimation procedure. Finally, the elasticities used in
the EIA are summarized in Section B.4.
B.I Review of Previous Elasticity Estimates
The economics literature provides seven sources of estimates of either the demand
elasticities or the supply elasticities for particular vegetable oil or oilseed meal products. The
relevant elasticity estimates from each source are summarized below in reverse chronological
order.
Ugarte (1999)
In developing a national model of the agricultural sector, Ugarte used elasticity
estimates for soybean products as inputs into the analysis. The elasticities used in his model
are based on estimates in the literature. They are as follows:
• soybean meal demand -0.6
• soybean oil demand -0.1
• soybean meal export demand -0.85
• soybean oil export demand -0.75
B-l
-------
Agriculture and Agri-Food Canada (1998)
Under the direction of Professor Karl Meilke at the University of Guelph, Agriculture
and Agri-Food Canada has estimated oil and meal demand elasticities for several countries.
The model aggregates soybeans, canola, and sunflower into a single oilseed commodity for
the purposes of the analysis. All equations were estimated by ordinary least squares. The
years used in the analysis were not indicated in the report. Table B-l provides the estimated
demand elasticities for each commodity aggregate.
Table B-l. Estimates of Oilseed Product Demand Elasticities from Agriculture and
Agri-Food Canada
Product Aggregate Elasticity
Oilseed oil -0.30
Oilseed meal -0.12
Oilseed oil stock
Short-ran -0.80
Long-ran -1.77
Oilseed meal stock
Short-ran -0.50
Long-ran -0.50
Source: Agriculture and Agri-Food Canada. 1998. "An Evaluation of Oilseed Trade Liberalization."
http://aceis.agr.ca/policy/epad/en...pubs/wp-tp/tms/98034tp/toc/toc.htm. September.
Baumel (1996)
Baumel presents historical data on soybean processing margins and describes the
impact of increasing margins on plant expansion by soybean processors. In describing the
effects, he cites estimates of elasticities of demand for soymeal of-0.1 and for soybean oil of
-0.26. However, he does not indicate the source for these estimates.
Chern, Loehman, and Yen (1995)
The study conducted by Chern, Loehman, and Yen used annual observations for the
time period 1950 to 1988 to estimate a system of demand equations for fats and oils that
included variables to measure the effects of health information on consumer demand. The
B-2
-------
components of their system are butter, corn oil, cottonseed oil, lard, peanut oil, and soybean
oil. Table B-2 provides the estimated demand elasticities for each; standard errors are not
included because the authors did not report them.
Table B-2. Estimates of Demand Elasticities for Fats and Oils from Chern, Loehman,
and Yen Using Annual Data, 1950 to 1988
Product Elasticity
Butter -0.816
Corn oil -0.235
Cottonseed oil -0.646
Lard -0.263
Peanut oil -0.242
Soybean oil -0.292
Source: Chern, Wen S., Edna T. Loehman, and Steven T. Yen. 1995. "Information, Health Risk Beliefs, and
the Demand for Fats and Oils." The Review of Economics and Statistics 77(3):555-64.
The elasticity estimates indicate that all fats and oil products are inelastic, and
cottonseed oil is more elastic than corn oil, peanut oil, and soybean oil. While all of these
results appear reasonable, the estimates of the cross-price elasticities and expenditure
elasticities, which are not reported here, are not. For example, the expenditure elasticities for
corn oil and peanut oil are negative, which would suggest that consumers reduce their
purchases of these products as their income increases. Furthermore, we would expect
vegetable oils to be substitutes for one another, yet cross-price elasticities suggest that many
are complements (e.g., peanut oil and cottonseed oil).
Lence, Hayes, and Meyers (1995)
As part of a model to understand forward trading and storage of soybeans, Lence,
Hayes, and Meyers estimated supply elasticities for soybean oil and meal. The data used to
estimate the model are monthly observations for the period September 1965 through
December 1986, and elasticities are evaluated at the sample means. The estimated supply
elasticities from their "very-short-run" and "short-run" models are presented in Table B-3
along with their standard errors. In either case, the supply elasticities are very inelastic. In
B-3
-------
the "very-short-run" model, the elasticities for soybean oil and meal differ, but in the "short-
run" model, they indicate the same supply price responsiveness for each. These differences
likely occur because soybean meal is perishable and more costly to store than soybean oil;
therefore, quantity supplied is less responsive to price changes in the very short run.
Table B-3. Estimates of Supply Elasticities for Soybean Oil and Meal from Lence,
Hayes, and Meyers Using Monthly Data, September 1965 to December 1986
Product
Soybean oil
Soybean meal
Very-Short-Run Supply
Elasticity
0.42
(0.10)
0.11
(0.04)
Short-Run Supply
Elasticity
0.23
(0.09)
0.23
(0.11)
Note: Standard errors are indicated in parentheses.
Source: Lence, S.H., D.J. Hayes, and W.H. Meyers. 1995. "The Behavior of Forward-Looking Firms in the
Very Short Run." American Journal of Agricultural Economics 77(November):922-934.
Huang (1993)
Huang used annual observations for the time period 1953 to 1990 to estimate
elasticities for the fats and oils subgroup within a complete system of demand equations for
food commodities. The components of the fats and oils subgroup were butter, margarine
(which is produced using vegetable oils), and all other fats and oils. Table B-4 presents the
estimated demand elasticities, standard errors, and percentages of each type in the total fats
and oils consumer budget. All the elasticities indicate that demand is very inelastic for these
products.
B-4
-------
Table B-4. Estimates of Demand Elasticities for Fats and Oils from Huang Using
Annual Data, 1953 to 1990
Type of Product
Butter
Margarine
Other fats and oils
Estimated Elasticity
-0.2428
-0.0087
-0.1393
Standard Error
0.1613
0.1470
0.0650
Source: Huang, K.S. 1993. U.S. Demand for Food: A Complete System of Price and Income Effects. U.S.
Department of Agriculture, Economic Research Service. Bulletin 1821.
Meyers et al (1993)
The model developed by Meyers et al. estimates supply and demand elasticities for a
number of agricultural crops using annual data from 1965/1970 (the actual year depends on
the crop) through 1992. They calculated elasticities at the average 1985 through 1989 prices
and quantities. Among the commodities included in their analysis are soybean oil and meal.
Table B-5 presents the estimated supply and demand elasticities; standard errors were not
provided in the report and thus are not included here. The demand elasticities are nearly
perfectly inelastic, indicating almost no change in purchases of soybean oil or meal in
response to price changes. The supply elasticities are inelastic as well but larger in absolute
value than the demand elasticities.
Table B-5. Estimates of Demand and Supply Elasticities for Soybean Oil and Meal
from Meyers et al. Using Annual Data, 1965/1970 to 1992
Product Supply Elasticity Demand Elasticity
Soybean oil 0.660 -0.061
Soybean meal 0.323 -0.097
Source: Meyers, W.H., P. Westhoff, D.L. Stephens, B.L. Buhr, M.D. Helmar, and K.J. Stephens. January
1993. "FAPRI U.S. Agricultural Sector Elasticities Volume I: Crops." Center for Agricultural and
Rural Development, Iowa State University, Ames, Iowa. Technical Report 92-TR 25.
B-5
-------
Yen and Chern (1992)
The study conducted by Yen and Chern used annual observations for the time period
1950 to 1986 to estimate a system of demand equations for fats and oils using a demand
system proposed by Lewbel while correcting for serial correlation of the errors. They
included equations in their model for butter, coconut oil, corn oil, cottonseed oil, peanut oil,
palm oil, lard, soybean oil, and tallow. Table B-6 provides the estimated demand elasticities
for each; standard errors were not reported by the authors.
Table B-6. Estimates of Demand Elasticities for Fats and Oils by Yen and Chern Using
Annual Data, 1950 to 1986
Product
Butter
Coconut oil
Corn oil
Cottonseed oil
Peanut oil
Palm oil
Lard
Soybean oil
Tallow
Elasticity
-0.6711
-0.3959
-0.3063
-1.1185
-1.0145
-1.5168
-0.8620
-0.5523
-1.7380
Source: Yen, S.T., and W.S. Chern. 1992. "Flexible Demand Systems with Serially Correlated Errors: Fat
and Oil Consumption in the United States." American Journal of Agricultural Economics 74(3):689-
697.
B.2 Overview of Elasticity Estimation Procedure
To obtain elasticity estimates, a simultaneous system of equations is required in
which each equation is identified through the inclusions of exogenous variables to control for
shifts in the supply and demand curves over time. A partial equilibrium market
supply/demand model is specified as a system of interdependent equations in which the price
and output of a product are simultaneously determined by the interaction of producers and
consumers in the market. In simultaneous equation models, where variables in one equation
B-6
-------
feed back into variables in another equation, the error terms are correlated with the
endogenous variables (price and output). In comparison, single-equation ordinary least
squares (OLS) estimation of individual equations will lead to biased and inconsistent
parameter estimates because it does not account for the correlation of the error term with the
endogenous variables.
Exogenous variables influencing the demand for vegetable oils and oilseed meals
include measures of general economic activity (such as U.S. gross domestic product [GDP]),
population, and the prices of substitutes. Exogenous variables influencing the level of supply
of vegetable oils and oilseed meals include measures of the change in the costs of production
caused by changes in prices of key inputs like oilseeds, fuel, electricity, and labor.
The supply/demand system for each vegetable oil and oilseed meal at the wholesale
level can be defined as follows:
Qtd = f(Pt,Zt) + ut (B.I)
Qts = g(Pt,Wt) + vt (B.2)
Q,d = Q,s (B.3)
Eq. (B.I) shows per capita quantity demanded as a function of price, Pt; a vector of demand
shifters, Zt(e.g., measures of economic activity and substitute prices); and an error term, ut.
Eq. (B.2) represents quantity supplied as a function of price and a vector of supply shifters,
Wt (e.g., input prices), and an error term, vt, while Eq. (B.3) specifies the equilibrium
condition that quantity supplied equals quantity demanded, creating a system of three
equations with three endogenous variables. The interaction of the specified market forces
solves this system, generating equilibrium values for the variables Pt* and Qt*=Qtd*=Qts*
To generate estimates separately of either the demand elasticities or the supply
elasticities, the two-stage least squares (2SLS) regression procedure can also be used. The
first stage of the 2SLS procedure involves regressing the observed price against the supply
and demand "shifter" variables that are exogenous to the system. This first stage produces
fitted (or predicted) values for the price variable that are, by definition, highly correlated with
the true endogenous variable, the observed price, and uncorrelated with the error term. In the
second stage, these fitted values are then employed as observations of the right-hand side
price variable in the demand function. These fitted values are uncorrelated with the error
term by construction and thus do not incur the endogeneity bias. By converting all variables
B-7
-------
into natural logarithms, each coefficient on the price variable yields an estimate of the
constant elasticity of demand or supply for each vegetable oil or oilseed meal product.
Elasticities estimated in this manner for vegetable oil and oilseed meals were used in the
economic impacts model.
B.3 Results of Elasticity Estimation
The data used in the elasticity estimation for vegetable oils and meals are annual
time-series data covering the period 1984 through 1996 and were obtained primarily from
USDA publications, including Agricultural Statistics, Oil Crops Yearbook, Oil Crops
Situation and Outlook, and Sugar and Sweetener Yearbook.
B. 3.1 Demand Equation Estimation
Demand equations were estimated using a general specification where the per capita
quantity consumed is expressed as a function of own-price, per capita income (for oil
products but not meal products), price of the primary substitute, trend, and trend squared.
Trend and trend squared were included as a general way to model the effects of changes in
tastes and preferences for vegetable oils and oilseed meals. All variables were converted to
natural logs, and all price and income variables were deflated by the implicit GDP deflator.
The endogenous variables in the equations are per capita consumption, own-price, and the
price of the primary substitute. The price of the primary substitute was included as an
endogenous variable because oil products substitute readily for one another as do meal
products; therefore, consumption of each is jointly determined with the price of its
substitutes. The exogenous variables include per capita income, trend, and trend squared.
The list of instruments includes these exogenous variables in addition to supply factors
influencing the price of the product: the price of the oilseed input, wages at the three-digit
SIC level, and the producer price index for fuel.
Using this model, reasonable estimates were obtained for cottonseed oil, cottonseed
meal, soybean oil, and soybean meal. For corn oil, none of the regressors in the model are
significant other than its price and trend. Dropping all of the other regressors in the model
results in a reasonable estimate of the demand elasticity and a reasonable adjusted R2. The
results of these five models are provided in Table B-7.
Sufficient data were not available to estimate a demand equation for corn germ meal.
Sufficient data were available to estimate demand for linseed oil, linseed meal, sunflower oil,
and sunflower meal, but the data series are too volatile to obtain reasonable estimates. Thus,
B-8
-------
Table B-7. Results of Econometric Estimation of Oilseed Product Demand Equations
Dependent
Variable —
Regressor
Intercept
Price
Substitute price3
Income
Trend
Trend squared
Adjusted R2
Durbin-Watson
Observations
Corn Oil
Per Capita
Corn Oil
Consumption
2.640
(4.38)
-0.387
(-2.22)
—
—
0.016
(2.61)
—
0.722
2.753
13
Cottonseed Oil
Per Capita
Cottonseed Oil
Consumption
-14.249
(-1.37)
-0.245
(-0.32)
-3.635
(-1.32)
0.537
(0.77)
0.212
(2.80)
-0.008
(-2.19)
0.7305
2.050
13
Cottonseed
Meal
Per Capita
Cottonseed
Meal
Consumption
-2.682
(-0.43)
-1.011
(-1.60)
1.517
(0.99)
—
0.032
(0.70)
-0.002
(-0.76)
0.643
2.527
13
Soybean Oil
Per Capita
Soybean Oil
Consumption
3.927
(1.46)
-0.340
(-1.92)
0.133
(0.19)
0.412
(2.07)
0.035
(1.80)
-0.001
(-1.27)
0.885
2.001
13
Soybean Meal
Per Capita
Soybean Meal
Consumption
5.287
(33.95)
-0.268
(-0.75)
0.178
(0.51)
—
0.021
(1.63)
0.001
(0.07)
0.895
2.282
13
Notes:
1. Numbers in parentheses are t-ratios (coefficient estimate divided by its standard error).
2. All variables are in natural logs.
a Substitute commodity prices used were as follows:
cottonseed oil equation: soybean oil
cottonseed meal equation: producer price index for animal feed
soybean oil equation: cottonseed oil
soybean meal equation: cottonseed meal
B-9
-------
elasticities for corn germ meal and the other oilseed products were assumed based on the
available estimates.
B. 3.2 Supply Equation Estimation
Supply equations were estimated using a general specification where quantity was
expressed as a function of price, wages at the three-digit SIC level, the price of the oilseed
input, and the price index for fuel. All variables were converted to natural logs, and all price
variables were deflated by the implicit GDP deflator. The endogenous variables in the
equations are price and quantity, and all others are assumed exogenous. The list of
instruments includes these exogenous variables in addition to demand factors influencing the
price of the product: income, price of substitutes, trend, and trend squared.
Reasonable estimates of the supply elasticities were not obtained and thus are not
reported here. Supply of these products is most likely a more complicated process than can
be represented by this simple model. In particular, inventories of products likely influence
the supply response of producers. As a result, supply elasticity estimates used in the EIA
were chosen based on values from the economics literature for soybean oil and meal.
B.4 Results of Elasticity Estimation
Table B-8 summarizes the domestic demand and supply elasticities and the trade
elasticities from the literature and from independent econometric estimates. In addition, the
elasticities used in the EIA are indicated. Corn oil and soybean demand elasticities used in
the EIA were obtained directly from the econometric estimates because they are similar to
those reported in the literature. The cottonseed demand elasticity used in the EIA is the
middle value of the econometrically estimated and reported values because the
econometrically estimated elasticity fell outside the range of reported values. The demand
elasticity for all other oils is a simple average of the three elasticities for the other oils.
Cottonseed meal and soybean meal demand elasticities were obtained directly from the
econometric estimates. The soybean meal estimate from the literature is similar to the
econometric estimate, and no estimates for cottonseed meal were available from the
literature. The demand elasticity used for corn germ meal and all other meals is a simple
average of the cottonseed meal and soybean meal econometric estimates.
The supply elasticities for oils and meals are based on estimates in the literature for
soybean oil and meal. For each of these products, two estimates were available; thus, the
elasticities used in
B-10
-------
Table B-8. Summary of Elasticity Estimates for Vegetable Oils and Meals
Demand
Elasticities
Supply
Elasticities
Import Supply
Elasticities
Export Demand
Elasticities
Product
Corn oil
Cottonseed oil
Soybean oil
All other oils
Corn germ meal
Cottonseed meal
Soybean meal
All other meals
Corn oil
Cottonseed oil
Soybean oil
All other oils
Corn germ meal
Cottonseed meal
Soybean meal
All other meals
All products
All products
Literature Econometric Elasticity Used
Estimates Estimates for the EIA
-0.235, -0.306 -0.387 -0.39
-0.646, -1.118 -0.245 -0.65
-0.292, -0.610, -0.340 -0.34
-0552, -O.la
-0.242 (peanut) -0.33b
-1.014 (peanut)
-0.46
-1.011 -1.01
-0.097, -0.6a -0.268 -0.27
-0.64C
0.44d
0.44d
0.660, 0.230 0.44d
0.44d
0.28e
0.28e
0.323, 0.230 0.28e
0.28e
1.00f
-1.00f
a Value was used in a model developed by Ugarte but was not econometrically estimated.
b Average of corn oil, cottonseed oil, and soybean oil econometric estimates.
0 Average of cottonseed meal and soybean meal econometric estimates.
d Midpoint of soybean oil supply elasticities from the literature.
e Midpoint of soybean meal supply elasticities from the literature.
f Value is assumed.
B-ll
-------
the EIA are an average of the elasticities from each source. For lack of other information, the
soybean elasticities were used for the cottonseed, corn germ, and all other oils and meals as
well.
In general, trade elasticities are expected to be more elastic than domestic supply and
demand elasticities. Except for soybean oil and meal export demand, elasticity values are not
available from either econometric estimates or the literature. For these soybean products,
however, the values reported by Ugarte are assumed rather than econometrically estimated.
Thus, the values of-1 and 1 are assumed for the export demand and import supply
elasticities respectively for all vegetable oil and meal products.
B-12
-------
APPENDIX C
ECONOMIC MODEL SENSITIVITY ANALYSIS
A sensitivity analysis was conducted to determine the effects of different demand and
supply elasticity estimates under the MACT floor regulatory alternative promulgated by the
EPA. This analysis assumes lost production costs. Table C-l presents the sets of elasticities
used in the analysis reported in Section 4 and for two sensitivity analyses. Sensitivity
Analysis A uses the most inelastic demand elasticities and the most elastic supply elasticities
from the literature (see Table B-6). These elasticities are expected to result in the greatest
predicted changes in consumer and producer surplus and the greatest social costs of the
regulation. Sensitivity Analysis B uses the most elastic demand elasticities and the most
inelastic supply elasticities from the literature. Likewise, these elasticities are expected to
result in the smallest predicted changes.
The following results are presented:
• market-level impacts of the NESHAP (Table C-2),
• industry-level impacts of the NESHAP (Table C-3),
• distribution of industry impacts of the NESHAP by profit losers and profit gainers
(Table C-4),
• distribution of the social costs associated with the NESHAP (Table C-5), and
• small business impacts of the NESHAP (Table C-6).
In general, the projected effects of the regulation are not particularly sensitive to changes in
the elasticity values. However, predicted price, output, trade, revenue, and consumer and
producer surplus are greater under Sensitivity Analysis A relative to the base values and less
under Sensitivity Analysis B.
C-l
-------
Table C-l. Summary of Market-Level Impacts of Vegetable Oil Production NESHAP: 1995
Oils
Corn
Cottonseed
Soybean
All Other
Meals
Corn Germ
Cottonseed
C-1 Soybean
to
All Other
Primary
Demand
-0.39
-0.65
-0.34
-0.46
-0.64d
-1.01
-0.27
-0.64d
Supply
0.44
0.44
0.44
0.44
0.28
0.28
0.28
0.28
Sensitivity
Demand
-0.235
-0.245
-2.92
-0.257C
-0.55d
-1.01
-0.097
-0.55d
Analysis Aa
Supply
0.66
0.66
0.66
0.66
0.32
0.32
0.32
0.32
Sensitivity
Demand
-0.39
-1.118
-0.61
-1.044
-0.64d
-1.01
-0.268
-0.64d
Analysis Bb
Supply
0.23
0.23
0.23
0.23
0.23
0.23
0.23
0.23
a Inelastic demand/elastic supply.
b Elastic demand/inelastic supply.
0 Average of corn, cottonseed, soybean.
d Average of cottonseed, soybean.
-------
Table C-2. Summary of Market-Level Impacts of Vegetable Oil Production NESHAP: 1995
p
00
Inelastic Demand/
Elastic Supply
Primary
Elastic Demand/
Inelastic Supply
Changes from Baseline
Scenario
Corn Oil
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Cottonseed Oil
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Soybean Oil
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Baseline
$532.00
1,126,198
1,121,703
1,066,819
54,884
415,270
4,495
$536.00
646,031
645,925
622,308
23,617
147,281
106
$536.00
7,981,880
7,968,264
7,968,264
0
1,143,544
13,616
Absolute
$3.40
-3,705
-3,734
-3,965
232
-2,640
29
$4.14
-2,070
-2,071
-2,191
120
-1,130
1
$2.29
-13,370
-13,428
-13,428
0
-4,864
58
Percent
0.64%
-0.33%
-0.33%
-0.37%
0.42%
-0.64%
0.64%
0.77%
-0.32%
-0.32%
-0.35%
0.51%
-0.77%
0.77%
0.43%
-0.17%
-0.17%
-0.17%
NA
-0.43%
0.43%
Absolute
$2.53
-3,280
-3,301
-3,416
115
-1,966
21
$2.55
-2,233
-2,234
-2,283
49
-697
1
$1.84
-11,867
-11,914
-11,914
0
-3,905
47
Percent
0.48%
-0.29%
-0.29%
-0.32%
0.21%
-0.47%
0.48%
0.48%
-0.35%
-0.35%
-0.37%
0.21%
-0.47%
0.48%
0.34%
-0.15%
-0.15%
-0.15%
NA
-0.34%
0.34%
Absolute
$1.65
-2,141
-2,155
-2,194
39
-1,282
14
$1.18
-1,551
-1,552
-1,564
12
-324
0
$0.94
-9,285
-9,309
-9,309
0
-1,996
24
Percent
0.31%
-0.19%
-0.19%
-0.21%
0.07%
-0.31%
0.31%
0.22%
-0.24%
-0.24%
-0.25%
0.05%
-0.22%
0.22%
0.17%
-0.12%
-0.12%
-0.12%
NA
-0.17%
0.17%
(continued)
-------
Table C-2. Summary of Market-Level Impacts of Vegetable Oil Production NESHAP: 1995 (Continued)
Inelastic Demand/
Elastic Supply
Primary
Elastic Demand/
Inelastic Supply
Changes from Baseline
Scenario
All Other Vegetable Oils
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Corn Germ Meal
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Cottonseed Meal
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Baseline
$658.82
1,623,013
1,093,411
1,026,335
67,076
632,625
529,602
$88.40
1,473,651
1,473,651
1,401,546
72,105
61,950
0
$123.20
1,826,100
1,826,100
1,744,562
81,538
89,700
0
Absolute
$1.57
-2,112
-3,374
-3,479
106
-1,506
1,262
$0.36
-3,447
-3,447
-3,542
95
-254
0
$0.37
-5,561
-5,561
-5,640
79
-271
0
Percent
0.24%
-0.13%
-0.31%
-0.34%
0.16%
-0.24%
0.24%
0.41%
-0.23%
-0.23%
-0.25%
0.00%
-0.41%
NA
0.30%
-0.30%
-0.30%
-0.32%
0.10%
-0.30%
NA
Absolute
$1.07
-1,763
-2,622
-2,670
48
-1,025
859
$0.30
-3,304
-3,304
-3,373
69
-212
0
$0.34
-5,019
-5,019
-5,081
62
-244
0
Percent
0.16%
-0.11%
-0.24%
-0.26%
0.07%
-0.16%
0.16%
0.34%
-0.22%
-0.22%
-0.24%
-0.34%
NA
0.27%
-0.27%
-0.27%
-0.29%
0.08%
-0.27%
NA
Absolute
$0.48
-1,214
-1,600
-1,611
11
-461
386
$0.26
-2,868
-2,868
-2,918
49
-184
0
$0.29
-4,292
-4,292
-4,336
44
-209
0
Percent
0.07%
-0.07%
-0.15%
-0.16%
0.02%
-0.07%
0.07%
0.30%
-0.19%
-0.19%
-0.21%
0.00%
-0.30%
NA
0.23%
-0.24%
-0.24%
-0.25%
0.05%
-0.23%
NA
(continued)
-------
o
Table C-2. Summary of Market-Level Impacts of Vegetable Oil Production NESHAP: 1995 (Continued)
Inelastic Demand/
Elastic Supply
Primary
Elastic Demand/
Inelastic Supply
Changes from Baseline
Scenario
Soybean Meal
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
All Other Vegetable Oil
Meals
Market Price ($/short ton)
Market Output (tpy)
Domestic Production
Solvent Extraction
Mechanical Extraction
Exports
Imports
Baseline
$173.80
31,290,977
31,225,572
31,225,572
0
6,491,570
65,405
$93.22
2,520,129
1,583,559
1,497,886
85,673
138,469
936,570
Absolute
$0.62
-32,124
-32,359
-32,359
0
-23,210
235
$0.12
-1,810
-2,982
-3,017
34
-173
1,172
Percent
0.36%
-0.10%
-0.10%
-0.10%
NA
-0.36%
0.36%
0.13%
-0.07%
-0.19%
-0.20%
0.04%
-0.12%
0.13%
Absolute
$0.47
-35,793
-35,970
-35,970
0
-17,571
178
$0.10
-1,729
-2,704
-2,729
25
-144
975
Percent
0.27%
-0.11%
-0.12%
-0.12%
NA
-0.27%
0.27%
0.10%
-0.07%
-0.17%
-0.18%
0.03%
-0.10%
0.10%
Absolute
$0.42
-31,689
-31,846
-31,846
0
-15,509
157
$0.08
-1,459
-2,282
-2,299
17
-122
823
Percent
0.24%
-0.10%
-0.10%
-0.10%
NA
-0.24%
0.24%
0.09%
-0.06%
-0.14%
-0.15%
0.02%
-0.09%
0.09%
NA = Not available.
-------
Table C-3. Summary of Industry-Level Impacts of Vegetable Oil Production NESHAP: 1995
Inelastic Demand/
Elastic Supply
Scenario
Total Industry
Revenues ($103)
Costs ($103)
Post-regulatory
Oil and Meal Production
Gross Profits ($103)
Operating Entities
Product Line-Closures
Facility Closures
Employment Loss
Solvent Extraction
Revenues ($103)
Costs ($103)
Post-regulatory
Production
Gross Profits ($103)
Operating Entities
Product Lines
Facilities
Employment
Baseline
$11,963,996
$10,698,845
$0
$10,698,845
$1,265,151
NA
NA
NA
$11,853,542
$10,611,376
$0
$10,611,376
$1,242,166
111
106
5,673
Absolute
$27,959
-$6,926
$12,056
-$18,982
$34,885
0
0
-15
$27,222
-$7,089
$12,056
-$19,145
$34,310
0
0
-15
Percent
0.2%
-0.1%
NA
-0.2%
2.8%
NA
NA
NA
0.2%
-0.1%
NA
-0.2%
2.8%
0.0%
0.0%
-0.3%
Primary
Changes from
Absolute
$17,728
-$6,394
$12,076
-$18,470
$24,122
0
0
-12
$17,264
-$6,477
$12,076
-$18,554
$23,741
0
0
-12
Baseline
Percent
0.1%
-0.1%
NA
-0.2%
1.7%
NA
NA
NA
0.1%
-0.1%
NA
-0.2%
1.7%
0.0%
0.0%
-0.2%
Elastic Demand/
Inelastic Supply
Absolute
$10,048
-$2,888
$12,102
-$14,989
$12,936
0
0
-6
$9,802
-$2,923
$12,102
-$15,025
$12,725
0
0
-6
Percent
0.1%
-0.0%
NA
-0.1%
0.8%
NA
NA
NA
0.1%
-0.0%
NA
-0.1%
0.8%
NA
NA
NA
(continued)
-------
Table C-3 Summary of Industry-Level Impacts of Vegetable Oil Production NESHAP: 1995 (Continued)
Inelastic Demand/
Elastic Supply
Primary
Elastic Demand/
Inelastic Supply
Changes from Baseline
Scenario
Mechanical Extraction
Revenues ($103)
Costs ($103)
Post-regulatory
Production
Gross Profits ($103)
Operating Entities
Product Lines
Facilities
Employment
Baseline
$110,454
$87,469
$0
$87,469
$22,985
NA
NA
NA
Absolute
$737
$162
$0
$162
$574
NA
NA
NA
Percent
0.7%
0.2%
0.0%
0.2%
2.5%
NA
NA
NA
Absolute
$464
$84
$0
$84
$380
NA
NA
NA
Percent
0.4%
0.1%
0.0%
0.1%
1.5%
NA
NA
NA
Absolute
$246
$36
$0
$36
$211
NA
NA
NA
Percent
0.2%
0.0%
0.0%
0.0%
0.7%
NA
NA
NA
NA = Not available.
-------
Table C-4. Summary of Distributional Industry Impacts of the Vegetable Oil Production NESHAP: 1995
p
oo
Supply Elasticity
Inelastic Demand/Elastic Supply
Floor
Primary
Elastic Demand/Inelastic Supply
Solvent Extraction Facilities
Number
Facility Capacity (tons
per day)
Total
Per Facility
Annual Solvent Loss
Total (103 gallons)
Gallons per Ton of
Oilseed
Incremental
Compliance Costs
Total ($103/yr)
Per Ton of Oilseed
Change in Gross
Profits ($103/yr)a
Change in
Employment (FTEs)b
With
Profit
Loss
17
8,062
474
1,584
0.86
$1,867
$1.01
-$1,038
-8
With
Profit
Gain
89
165,141
1,856
14,412
0.29
$10,284
$0.21
$35,349
-7
Total
106
173,203
1,634
15,996
0.31
$12,151
$0.24
$34,310
-15
With
Profit
Loss
21
11,362
541
2,419
0.84
$2,502
$0.87
-$1,184
-7
With
Profit
Gain
85
161,841
1,904
13,577
0.28
$9,649
$0.20
$24,925
-5
Total
106
173,203
1,634
15,996
0.31
$12,151
$0.24
$23,741
-12
With
Profit
Loss
32
24,977
781
3,981
0.70
$3,808
$0.67
-$1,473
-3
With
Profit
Gain
74
148,226
2,003
12,015
0.27
$8,343
$0.19
$14,198
-3
Total
106
173,203
1,634
15,996
0.31
$12,151
$0.24
$12,725
-6
a Gross profits calculated as revenue less costs of production including oilseed costs.
b FTEs = Full-time equivalents.
-------
Table C-5. Distribution of the Social Costs Associated with the Vegetable Oil
Production NESHAP: 1995
Floor with Lost Production Costs
Social Cost Component
Consumer Surplus
Corn Oil
Cottonseed Oil
Soybean Oil
All Other Vegetable Oils
Corn Germ Feed
Cottonseed Meal
Soybean Meal
All Other Vegetable Meals
Producer Surplus
Domestic Producers
Solvent Extraction
Mechanical Extraction
Foreign Producers
Social Costs of Regulation
Inelastic
Demand/Elastic
Supply
-$48,321
-$3,827
-$2,673
-$18,259
-$2,548
-$537
-$680
-$19,504
-$294
$35,915
$34,885
$34,310
$574
$1,030
-$12,407
Primary
Change in Value ($103)
-$36,929
-$2,845
-$1,644
-$14,649
-$1,734
-$446
-$613
-$14,752
-$244
$24,846
$23,122
$23,741
$380
$725
-$12,083
Elastic
Demand/Inelastic
Supply
-$25,007
-$1,854
-$763
-$7,476
-$779
-$387
-$524
-$13,017
-$206
$13,314
$12,936
$12,725
$211
$379
-$11,693
C-9
-------
Table C-6. Summary of Small Business Impacts of Vegetable Oil Production NESHAP: 1995
O
o
Floor with Lost Production Costs
Inelastic Demand/Elastic
Supply
Primary
Elastic Demand/Inelastic
Supply
Changes from Baseline
Revenues ($103)
Costs ($103)
Post-regulatory
Production
Gross Profits ($103)
Operating Entities
Product Lines
Facilities
Employment
Baseline
$780,636
$665,767
$0
$665,767
$114,869
21
20
NR
Absolute
-$396
-$1,507
$2,126
-$3,633
$1,111
0
0
-7
Percent
-0.1%
-0.2%
NA
-0.5%
1.1%
0.0%
0.0%
NR
Absolute
-$793
-$1,010
$2,131
-$3,141
$217
0
0
-6
Percent
-0.1%
-0.2%
NA
-0.5%
0.2%
0.0%
0.0%
NR
Absolute
-$739
-$185
$2,139
-$2,324
-$554
0
0
-2
Percent
-0.1%
-0.1%
NA
-0.4%
-0.4%
0.0%
0.0%
NR
NA = Not available.
NR = Not reported to avoid disclosure of CBI.
------- |