Eiwiranmmtal Protscin
Economic Impact Analysis of the Industrial
Boilers and Process Heaters NESHAP
-------
EPA-452/R-04-001
February 2004
Economic Impact Analysis of the Industrial Boilers and Process Heaters NESHAP
By:
RTI International*
Health, Social, and Economics Research
Research Triangle Park, NC 27709
Prepared for:
John L. Sorrels
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Innovative Strategies and Economics Group
C339-01
Research Triangle Park, NC 27711
Contract No. 68-D-99-024
*RTI International is a trade name of Research Triangle Institute.
-------
CONTENTS
Section Page
1 Introduction 1-1
1.1 Agency Requirements for an EIA 1-1
1.2 Scope and Purpose 1-2
1.3 Organization of the Report 1-3
2 Boiler and Process Heater Technologies 2-1
2.1 Characteristics of Steam 2-2
2.2 Fossil-Fuel Boiler Characterization 2-4
2.2.1 Industrial, Commercial, and Institutional Boilers 2-5
2.2.2 Heat Transfer Configurations 2-5
2.2.3 Major Design Types 2-6
2.2.3.1 Stoker-Fired Boilers (Coal) 2-6
2.2.3.2 Pulverized Coal Boilers (Coal) 2-6
2.2.3.3 Fluidized Bed Combustion (FBC) Boilers (Coal) ... 2-7
2.2.3.4 Tangentially Fired Boilers (Coal, Oil, Natural Gas) . . 2-7
2.2.3.5 Wall-fired Boilers (Coal, Oil, Natural Gas) 2-8
2.3 Process Heater Characterization 2-8
2.3.1 Classes of Process Heaters 2-8
2.3.2 Major Design Types 2-9
2.3.2.1 Combustion Chamber Set-Ups 2-10
2.3.2.2 Combustion Air Supply 2-10
2.3.2.3 Tube Configurations 2-11
2.3.2.4 Burners 2-12
3 Profile of Affected Units and Facilities and Compliance Costs 3-1
3.1 Regulatory Alternative 3-1
3.1.1 Regulatory Background 3-2
3.1.2 Regulatory Authority 3-3
iii
-------
3.1.3 Regulatory Alternatives and Control Technologies 3-6
3.1.3.1 MACT Floor Development 3-6
3.1.3.2 Consideration of Options Beyond the Floor for
Existing Units 3-14
3.1.3.3 EPA Response to Recent Court Decisions in
Developing the Emission Limitations 3-18
3.1.3.4 How did EPA Determine the Emission
Limitations for New Units? 3-20
3.1.4 Considerations of Possible Risk-Based Alternatives to
Reduce Impacts to Sources 3-30
3.1.4.1 Applicability Cutoffs for Threshold Pollutants
Under Section 112(d)(4) of the CAA 3-31
3.1.4.2 Applicability Cutoffs for Hydrogen Chloride
Controls Under Section 112(d)(4) of the CAA .... 3-32
3.1.4.3 Applicability Cutoffs for Total Selected Metals
Controls Under Section 112(d)(4) of the CAA 3-33
3.2 Profile of Existing Boiler and Process Heaters Units 3-35
3.2.1 Distribution of Existing Boilers and Facilities by Industry ... 3-36
3.2.2 Technical Characteristics of Existing Boilers 3-36
3.2.2.1 Final Rule 3-36
3.3 Methodology for Estimating Cost Impacts 3-39
3.4 Projection of New Boilers and Process Heaters 3-48
3.5 National Engineering Population, Cost Estimates, and
Cost-Effectiveness Estimates 3-49
Profiles of Affected Industries 4-1
4.1 Textile Mill Products (SIC 22/NAICS 313) 4-1
4.2 Lumber and Wood Products (SIC 24/NAICS 321) 4-1
4.2.1 Supply Side of the Industry 4-2
4.2.1.1 Production Processes 4-2
4.2.1.2 Types of Output 4-4
4.2.1.3 Major By-Products and Co-Products 4-4
4.2.1.4 Costs of Production 4-4
4.2.1.5 Capacity Utilization 4-5
4.2.2 Demand Side of the Industry 4-5
4.2.3 Product Characteristics 4-6
4.2.4 Uses and Consumers of Outputs 4-6
4.2.5 Organization of the Industry 4-6
4.2.6 Markets and Trends 4-9
4.3 Furniture and Related Product Manufacturing (SIC 25/NAICS 337) . . 4-9
iv
-------
4.4 Paper and Allied Products (SIC 26/NAICS 322) 4-10
4.4.1 Supply Side of the Industry 4-11
4.4.1.1 Production Process 4-11
4.4.1.2 Types of Output 4-12
4.4.1.3 Major By-Products and Co-Products 4-12
4.4.1.4 Costs of Production 4-13
4.4.1.5 Capacity Utilization 4-13
4.4.2 Demand Side of the Industry 4-14
4.4.2.1 Product Characteristics 4-14
4.4.2.2 Uses and Consumers of Products 4-14
4.4.3 Organization of the Industry 4-14
4.4.4 Markets and Trends 4-16
4.5 Medicinal Chemicals and Botanical Products and Pharmaceutical
Preparations (SICs 2833, 2834/NAICS 32451) 4-16
4.5.1 Supply Side of the Industry 4-17
4.5.1.1 Production Processes 4-17
4.5.1.2 Types of Output 4-18
4.5.1.3 Major By-Products and Co-Products 4-18
4.5.1.4 Costs of Production 4-18
4.5.1.5 Capacity Utilization 4-20
4.5.2 Demand Side of the Industry 4-20
4.5.3 Organization of the Industry 4-21
4.5.4 Markets and Trends 4-23
4.6 Industrial Organic Chemicals Industry (SIC 2869/NAICS 3251) 4-24
4.6.1 Supply Side of the Industry 4-24
4.6.1.1 Production Processes 4-24
4.6.1.2 Types of Output 4-25
4.6.1.3 Major By-Products and Co-Products 4-26
4.6.1.4 Costs of Production 4-26
4.6.1.5 Capacity Utilization 4-26
4.6.2 Demand Side of the Industry 4-28
4.6.3 Organization of the Industry 4-28
4.6.4 Markets and Trends 4-28
4.7 Electric Services (SIC 4911/NAICS 22111) 4-28
4.7.1 Electricity Production 4-29
4.7.1.1 Generation 4-31
4.7.1.2 Transmission 4-32
4.7.1.3 Distribution 4-32
4.7.2 Cost of Production 4-32
v
-------
4.7.3 Organization of the Industry 4-33
4.7.3.1 Utilities 4-34
4.7.3.2 Nonutilities 4-36
4.7.4 Demand Side of the Industry 4-36
4.7.4.1 Electricity Consumption 4-36
4.7.4.2 Trends in the Electricity Market 4-38
5 Economic Analysis Methodology 5-1
5.1 Background on Economic Modeling Approaches 5-1
5.1.1 Modeling Dimension 1: Scope of Economic
Decisionmaking 5-2
5.1.2 Modeling Dimension 2: Interaction Between
Economic Sectors 5-3
5.2 Selected Modeling Approach for Boilers and Process
Heaters Analysis 5-4
5.2.1 Directly Affected Markets 5-5
5.2.1.1 Electricity Market 5-7
5.2.1.2 Petroleum Market 5-7
5.2.1.3 Goods and Services Markets: Agriculture,
Manufacturing, Mining, Commercial, and
Transportation 5-8
5.2.2 Indirectly Affected Markets 5-11
5.2.2.1 Market for Coal 5-11
5.2.2.2 Natural Gas Market 5-11
5.2.2.3 Goods and Services Markets 5-12
5.2.2.4 Impact on Residential Sector 5-12
5.3 Operationalizing the Economic Impact Model 5-12
5.3.1 Computer Model 5-14
5.3.2 Calculating Changes in Social Welfare 5-17
6 Economic Impact Analysis Results 6-1
6.1 Social Cost Estimates 6-1
6.2 National Market-Level Impacts 6-2
6.3 Executive Order 13211 (Energy Effects) 6-5
6.4 Conclusions 6-6
VI
-------
7 Small Entity Impacts 7-1
7.1 Background on Small Entity Screenings 7-1
7.2 Identifying Small Entities 7-2
7.3 Analysis of Facility-Level and Parent-Level Data 7-3
7.4 Small Entity Impacts 7-7
7.5 Affected Government Entities: Supplemental Analysis 7-7
7.6 Assessment of SBREFA Screening 7-11
References R-l
Appendix A Estimating Economic Impacts in Markets Affected by
the Boilers and Process Heaters MACT A-l
Appendix B Assumptions and Sensitivity Analysis B-l
Appendix C Economic Analysis of Regulatory Alternative: Option 1A C-l
Appendix D Impacts from Application of Risk-Based Alternatives D-l
vn
-------
LIST OF FIGURES
Number Page
2-1 Generating Electricity: Steam Turbines 2-4
3-1 Characteristics of Units Affected 3-39
4-1 Traditional Electric Power Industry Structure 4-30
4-2 Utility and Nonutility Generation and Shares by Class, 1985 and 1995 4-35
4-3 Annual Electricity Sales by Sector 4-38
5-1 Links Between Energy and Goods and Services Markets 5-6
5-2 Market Effects of Regulation-Induced Costs 5-8
5-3 Fuel Market Interactions with Facility-Level Production Decisions 5-10
5-4 Operationalizing the Estimation of Economic Impact 5-13
5-5 Changes in Economic Welfare with Regulation 5-18
7-1 Parent Size by Employment Range 7-6
7-2 Number of Parents by Sales Range 7-7
Vlll
-------
LIST OF TABLES
Number Page
3-1 Emission Limits for Boilers and Process Heaters (Ib/MMBtu) 3-5
3-2 Units and Facilities Affected by the Final Rule by Industry 3-37
3-3 Testing and Monitoring Costs for Units Covered 3-45
3-4 Cost Effectiveness (C/E) of Industrial Boiler and Process Heater MACT
on Existing Units and Subcategories 3-47
3-5 New Unit Projections by Industry, MACT Floor 3-50
3-6 Unit Cost and Population Estimates for the Final Rule by Industry, 2005 . . . 3-52
4-1 Lumber and Wood Products Markets Likely to Be Affected by the
Regulation 4-2
4-2 Value of Shipments for the Lumber and Wood Products Industry
(SIC 24/NAICS 321), 1987-1996 4-3
4-3 Inputs for the Lumber and Wood Products Industry (SIC 24/NAICS 321),
1987-1996 4-5
4-4 Capacity Utilization Ratios for Lumber and Wood Products Industry,
1991-1996 4-6
4-5 Size of Establishments and Value of Shipments for the Lumber and Wood
Products Industry (SIC 24/NAICS 321) 4-7
4-6 Measures of Market Concentration for Lumber and Wood Products
Markets 4-8
4-7 Paper and Allied Products Industry Markets Likely to Be Affected
by Regulation 4-10
4-8 Value of Shipments for the Paper and Allied Products Industry
(SIC 26/NAICS 322), 1987-1996 4-11
4-9 Inputs for the Paper and Allied Products Industry (SIC 26/NAICS 322),
1987-1996 4-13
4-10 Capacity Utilization Ratios for the Paper and Allied Products Industry,
1991-1996 4-14
4-11 Size of Establishments and Value of Shipments for the Paper and Allied
Products Industry (SIC 26/NAICS 322) 4-15
4-12 Measurements of Market Concentration for Paper and Allied Products
Markets 4-16
4-13 Value of Shipments for the Medicinals and Botanicals and Pharmaceutical
Preparations Industries, 1987-1996 4-17
ix
-------
4-14 Inputs for Medicinal Chemicals and Botanical Products Industry (SIC
2833/NAICS 32451), 1987-1996 4-19
4-15 Inputs for the Pharmaceutical Preparations Industry
(SIC 2834/NAICS 32451), 1987-1996 4-20
4-16 Capacity Utilization Ratios for the Medicinal Chemicals and Botanical
Products (SIC 2833/NAICS 32451) and Pharmaceutical Preparations
(SIC 2834/NAICS 32451) Industries, 1991-1996 4-21
4-17 Size of Establishments and Value of Shipments for the Medicinal
Chemicals and Botanical Products (SIC 2833/NAICS 32451) and
Pharmaceutical Preparations (SIC 2834/NAICS 32451) Industries 4-22
4-18 Measures of Market Concentration for the Medicinal Chemicals and
Botanical Products (SIC 2833/NAICS 32451) and Pharmaceutical
Preparations (SIC 2834/NAICS 32451) Industries 4-23
4-19 Value of Shipments for the Industrial Organic Chemicals, N.E.C. Industry
(SIC 2869/NAICS 3251), 1987-1996 4-25
4-20 Inputs for the Industrial Organic Chemicals Industry
(SIC 2869/NAICS 3251), 1987-1996 4-27
4-21 Capacity Utilization Ratios for the Industrial Organic Chemicals
Industry (SIC 2869/NAICS 3251), 1991-1996 4-27
4-22 Size of Establishments and Value of Shipments for the Industrial Organic
Chemicals Industry (SIC 2869/NAICS 3251) 4-29
4-23 Net Generation by Energy Source, 1995 4-31
4-24 Total Expenditures in 1996 ($103) 4-33
4-25 Number of Electricity Suppliers in 1999 4-34
4-26 U.S. Electric Utility Retail Sales of Electricity by Sector, 1989 Through
1998 (106 kWh) 4-37
4-27 Key Parameters in the Cases 4-39
5-1 Comparison of Modeling Approaches 5-2
5-2 Supply and Demand Elasticities 5-15
5-3 Fuel Price Elasticities 5-17
6-1 Social Cost Estimates ($1998 106): Final Rule 6-1
6-2 Distribution of Social Costs by Sector/Market: Final Rule ($1998 106) 6-3
6-3 Market-Level Impacts 6-4
7-1 Summary of Small Entity Impacts 7-1
7-2 Facility-Level and Parent-Level Data by Industry 7-4
7-3 Small Parent Entities by Industry 7-8
7-4 Summary Statistics for SBREFA Screening Analysis 7-10
7-5 Regional Distribution of Municipal Systems 7-11
7-6 Selected Municipal Utilities' Capacity, Usage, and Consumer Types 7-12
x
-------
7-7 Supplemental Screening Analysis for Small Governmental Jurisdictions .... 7-13
7-8 Profit Margins for Industry Sectors with Affected Small Businesses 7-14
XI
-------
SECTION 1
INTRODUCTION
The U.S. Environmental Protection Agency (referred to as EPA or the Agency) is
developing regulations under Section 112 of the Clean Air Act (CAA) or the Act for
industrial, commercial, and institutional (ICI) boilers and process heaters. These combustion
devices are used in the production processes of numerous industries in the U.S. The
hazardous air pollutants (HAPs) are generated by the combustion of fossil fuels and biomass
in boilers and process heaters. The primary HAPs emitted by ICI boilers and process heaters
include arsenic, beryllium, cadmium, lead, hydrochloric acid, mercury, and other HAPs. In
addition, ICI boilers and process heaters also emit non-HAP pollutants such as sulfur dioxide
and particulate matter. To inform this rulemaking, the Innovative Strategies and Economics
Group (ISEG) of EPA's Office of Air Quality Planning and Standards (OAQPS) has
developed an economic impact analysis (EIA) to estimate the potential social costs of the
regulation. This report presents the results of this analysis in which a market model was used
to analyze the impacts of the air pollution rule on society.
1.1 Agency Requirements for an EIA
Congress and the Executive Office have imposed statutory and administrative
requirements for conducting economic analyses to accompany regulatory actions. Section
317 of the CAA specifically requires estimation of the cost and economic impacts for
specific regulations and standards under the authority of the Act. In addition, Executive
Order (EO) 12866 requires a more comprehensive analysis of benefits and costs for
significant regulatory actions.1 Other statutory and administrative requirements include
examination of the composition and distribution of benefits and costs. For example, the
Regulatory Flexibility Act (RFA), as amended by the Small Business Regulatory
Enforcement and Fairness Act of 1996 (SBREFA), requires EPA to consider the economic
impacts of regulatory actions on small entities. The Agency's OAQPS Economic Analysis
'Office of Management and Budget (OMB) guidance under EO 12866 stipulates that a full benefit-cost analysis
is required only when the regulatory action has an annual effect on the economy of $100 million or more.
1-1
-------
Resource Document provides detailed instructions and expectations for economic analyses
that support rulemaking (EPA, 1999).
1.2 Scope and Purpose
The CAA's purpose is to protect and enhance the quality of the nation's air resources
(Section 101(b)). Section 112 of the CAA Amendments of 1990 establishes the authority to
set national emissions standards for HAPs. This report evaluates the economic impacts of
pollution control requirements placed on ICI boilers and process heaters under these
amendments. These control requirements are designed to reduce releases of HAPs into the
atmosphere.
To reduce emissions of HAPs, the Agency establishes maximum achievable control
technology (MACT) standards. The term "MACT floor" refers to the minimum control
technology on which MACT standards can be based. For existing major sources,2 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). For new sources,
the MACT floor must be no less stringent than the emissions control achieved in practice by
the best controlled similar source. The MACT can also be chosen to be more stringent than
the floor, considering the costs and the health and environmental impacts.
The MACT floor will affect approximately 5,600 existing and new units. EPA
developed annual compliance costs for model units in each of 83 different model unit types.
EPA then linked the annualized compliance costs from the model units to the estimated
existing population of boilers and process heaters to obtain national impact estimates. In
addition, the Agency projected entrance of new boilers and process heaters through the year
2005, and linked the annualized compliance costs to these projected new units.
The economic impacts of national compliance costs, including both existing and new
units, on affected markets was then estimated using a computerized market model. EPA
used changes in prices and quantities in energy markets and final product markets to estimate
the firm-, industry-, market-, and societal-level impacts associated with the regulation.
2A 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 considering control, 10 tons or more of
any one HAP or 25 tons or more of any combination of HAPs.
1-2
-------
1.3 Organization of the Report
The remainder of this report is divided into six sections that describe the
methodology and presents the analysis results:
• Section 2 provides background information on ICI boiler and process heater
technologies.
• Section 3 profiles existing ICI boilers and process heaters by capacity, fuel type,
and industry and presents projections of the future population of units in 2005.
National compliance cost estimates are also presented in this section.
• Section 4 profiles the industries with the largest number of affected facilities.
Included are profiles of the lumber and wood products (SIC 24/NAICS 321),
paper and allied products (SIC 26/NAICS 322), and electrical services
(SIC 49/NAICS 221) industries.
• Section 5 describes the methodology for assessing the economic impacts of the
National Emission Standard for Hazardous Air Pollutants (NESHAP).
• Section 6 presents the results of the economic analysis, including market,
industry, and social cost impacts.
• Section 7 provides the Agency's analysis of the regulation's impact on small
entities.
In addition to these sections, Appendix A details the economic model used to predict
the economic impacts of the NESHAP, Appendix B presents the results of sensitivity
analyses on key model assumptions, and Appendix C presents results of analyses for
regulatory alternative Option 1 A.
1-3
-------
SECTION 2
BOILER AND PROCESS HEATER TECHNOLOGIES
The three categories of combustion devices affected under the regulations are
industrial, commercial, and institutional (ICI) boilers and process heaters. Although their
primary function is to transfer heat generated from fuel combustion to materials used in the
production process, the applications for boilers and process heaters are somewhat different.
As a result, the primary industries using boilers may not be the same as those using process
heaters. It is important to note that throughout this report the terms "boilers and process
heaters," and "units" are synonymous with "ICI boilers and process heaters." Utility boilers
primarily engaged in generating electricity are not covered by the NESHAP under analysis
and are therefore excluded from this analysis.
Boilers are combustion devices used to produce steam or heat water. Steam is
produced in boilers by heating water until it vaporizes. The steam is then channeled to
applications within a facility or group of facilities via pipes. Steam is an important power
and heating source for the U.S. economy. It is used in the preparation or manufacturing of
many key products, such as paper, petroleum products, furniture, and chemicals. Steam is
also used to heat buildings and to generate the majority of the electricity consumed in this
country. There are literally thousands of boilers currently being used in the United States
throughout a wide variety of industries.
Process heaters are primarily used as heat transfer units in which heat from fuel
combustion is transferred to process fluids, although they may also be used to transfer heat to
other nonfluid materials or to heat transfer materials for use in a process unit (not including
generation of steam). Process heaters are generally used in heat transfer applications where
boilers are inadequate. Often these are uses in which heat must be transferred at
temperatures in excess of 90° to 204°C (200° to 400°F). Process heaters are used in the
petroleum refining and petrochemical industries, with minor applications in the asphalt
concrete, gypsum, iron and steel, and wood and forest products industries.
Since one of the main uses of boilers is to generate steam, some of the characteristics
of steam are discussed in this section. This section also provides an overview of the various
types of boiler and process heater characteristics and designs.
2-1
-------
2.1 Characteristics of Steam
Steam, an odorless, invisible gas of vaporized water, maybe interspersed with water
droplets, which gives it a cloudy appearance. It is produced naturally when underground
water is heated by volcanic processes and mechanically using boilers and other heating
processes. When water is heated at atmospheric pressure, it remains in liquid form until its
temperature exceeds 212°F, the boiling point of water. Additional heat does not raise the
water's temperature but rather vaporizes the water, converting it into steam. However, if
water is heated under pressure, such as in a boiler, the boiling point is higher than 212°F and
more heat is required to generate steam. Once all the water has been vaporized into steam,
the addition of heat causes the temperature and volume to increase. Steam's heating and
work capabilities increase as it is produced under greater pressure coupled with higher
temperatures. As steam escapes from the boiler, it can be directed through pipes to drive
mechanical processes or to provide heat.
The steam used in most utility, industrial, and commercial applications is referred to
as "clean steam." Clean steam encompasses steam purities ranging from pure, solid-free
steam used in critical processes to filtered steam for less demanding applications. The
various types of clean steam differ in steam purity and steam quality. Steam purity is a
quantitative measure of contamination of steam caused by dissolved particles in the vapor or
by tiny droplets of water that may remain in the steam. Steam quality is a measure of how
much liquid water is mixed in with the dry steam (Fleming, 1992). Firms select the levels of
steam quality and steam purity for their applications based on the sensitivity of their
equipment to impurities, water droplet size, and condensation as well as the requirements for
their production process. Using clean steam minimizes the risk of product contamination
and prolongs equipment life. Although there are infinite possible levels of water purity and
quality, the term "clean steam" generally refers to three basic types of steam:
• filtered steam—produced by filtering plant steam using high-efficiency filters.
Filtered steam is generally of high steam quality because most large water
droplets and other contaminants will be filtered out.
• clean steam—steam that is frequently produced from deionized and distilled
water. Deionized and distilled water is free of dissolved solids and ions, which
may corrode pipework.
• pure steam—similar to clean steam except that it is always produced from
deionized and distilled water.
2-2
-------
Steam applications can be categorized by the amount of pressure required: hot water,
low pressure, and high pressure. Low pressure is 0 to 15 pounds per square inch (psi) and
high pressure steam is above 15 psi (Plant Engineering, 1991). Hot water systems, which
generate little steam, are primarily used for comfort applications, such as hot water for a
building. Low pressure applications include process heat and space heating. High pressure
steam applications are more frequently used in industrial and utility applications. Some high
pressure applications require that the steam be superheated, a process which ensures that the
steam is free of water droplets, to avoid damaging sensitive equipment.
Electric cogenerators, such as large factories and processing facilities, use steam to
drive turbines to generate electricity. A conventional steam electric power plant burns fossil
fuels (coal, gas, or oil) in a boiler, releasing heat that boils water and converts it into high-
pressure steam (see Figure 2-1). The steam enters a turbine where it expands and pushes
against blades to turn the generator shaft and create electric current. In this way, the thermal
energy of steam becomes mechanical energy, which is converted into electricity. Steam used
to drive turbines generates most of the electric power in the United States (TXU, 2000).
Industrial operations use steam to perform work such as powering complex
machinery operations, in the same way that electric utilities use steam to rotate turbines.
Textile mills, pulp and paper mills, and other manufacturing outfits are examples of facilities
that use steam to run machinery. Steam also provides heat and pressure for manufacturing
processes. Industrial establishments use steam to provide heat for drying or to heat and
separate materials. For example, the paper industry uses steam to heat rollers that dry paper
during the final stages of the production process. Petroleum refineries and chemical
producers use steam to heat petroleum, raw materials, and other inputs to separate inputs into
their constituent components or to facilitate chemical interactions. In addition to these
applications, steam is employed in many other industrial processes, including textile
production, wood working, furniture making, metal working, food preparation, and the
manufacture of chemicals. Substitutes for using steam as process heat include electrical
heating equipment, infrared, and other radiant drying techniques. Electricity may be used to
power machinery, as well. However, switching from steam-powered to electricity-powered
machinery would require significant equipment retrofits or replacement.
2-3
-------
Stack
Pulverized
Coal
Combustion Gases
Steam Turbine
Generator
Courtesy of TXU Boiler
Condenser
Electricity
Figure 2-1. Generating Electricity: Steam Turbines
Source: Texas Utilities (TXU). 2000. "Generating Electricity: Steam Turbines." As obtained in September
2000. .
Other steam applications include heating, sanitation, food processing and preparation,
and cleaning. In addition to using boilers to heat water, factories, hospitals, government
buildings, schools and other large buildings use boiler-generated steam to provide space
heating. Substitutes for boilers in heating air and water include electrical water and space
heaters; furnaces; and other heating, ventilation, and air conditioning equipment.
2.2 Fossil-Fuel Boiler Characterization
Section 2.2 discusses the different classes of fossil-fuel boilers, the most common
heat transfer configurations, and the major design types. The discussion indicates the type(s)
of fuel that each design can use to operate.
2-4
-------
2.2.1 Industrial, Commercial, and Institutional Boilers
Industrial, commercial, and institutional boilers are primarily used for process
heating, electrical or mechanical power generation, and/or space heating. Industrial boilers
are used in all major industrial sectors but primarily by the paper products, chemical, food,
and petroleum industries. It is estimated that the heat input capacity for these boilers is
typically between 10 and 250 MMBtu/hr; however, larger industrial boilers do exist and are
similar to utility boilers (EPA, 1997b). Commercial/institutional boilers are generally
smaller than the industrial units, with heat input capacities generally below 10 MMBtu/hr.
These units normally supply the steam and hot water for space heating in a wide range of
locations, including wholesale and retail trade, office buildings, hotels, restaurants, hospitals,
schools, museums, government buildings, and airports. Five hundred ninety-three of the
3,615 units potentially affected by the floor alternative for the regulation are
commercial/institutional units.
A boiler system includes the boiler itself, associated piping and valves, operation and
safety controls, water treatment system, and peripheral equipment such as pollution control
devices, economizers, or superheaters (Plant Engineering, 1991). Most boilers are made of
steel, cast iron, or copper. The primary fuels used by boilers are coal, oil, and natural gas,
but some use electricity, waste gases, or biomass.
Boilers may either be erected onsite (field-erected boilers) or assembled at a factory
(packaged boilers). Packaged boilers are typically lower in initial cost and more simple to
install. However, field-erected boilers may have lower operating costs, less maintenance,
and greater flexibility because the furnace or convection pattern chosen to meet required
steam pressure, capacity, and fuel specifications is tailored to the boiler's potential use (Plant
Engineering, 1991). Applications requiring more than 100,000 pounds of steam per hour are
usually equipped with a field-erected boiler.
2.2.2 Heat Transfer Configurations
The heat transfer configuration of a boiler refers to the method by which heat is
transferred to the water. The four primary boiler configurations are watertube, firetube, cast
iron, and tubeless. Most industrial users tend to rely on either watertube or firetube
configurations.
In a watertube boiler, combustion heat is transferred to water flowing through tubes
lining the furnace walls and boiler passes. The furnace watertubes absorb primarily radiative
heat, while the watertubes in the boiler passes gain heat by convective heat transfer. These
2-5
-------
units have a wide range of heat input capacities (ICI units range from 0.4 to 1,500
MMBtu/hr) and can be either field erected or packaged.1 Watertube boilers with heat input
capacities greater than 200 MMBtu/hr are typically field erected.
Because firetube, cast iron, and tubeless heat transfer configurations typically have
heat input capacities below 10 MMBtu/hr, they will not generally be covered by the
NESHAP. Therefore, this profile focuses on those boiler types that use watertube heat
transfer configurations.
2.2.3 Major Design Types
This section summarizes the five major design types for fossil fuel industrial boilers
that will be covered by the NESHAP. It also discusses, where possible, the fuels used,
capacity, and assembly method of each of these types of boilers.
2.2.3.1 Stoker-Fired Boilers (Coal)
These units use underfeed air to combust the coal char on a stationary grate,
combined with one or more levels of overfire air introduced above the grate. There are three
types of stoker units:
• spreader stokers,
• underfeed stokers, and
• overfeed stokers.
Stokers generally burn all types of coal, with the exception of overfeed stokers, which do not
burn coking bituminous coals. Stokers can also burn other types of solid fuel, such as wood,
wood waste, and bagasse. Spreader stokers are the most common of these boiler types and
have heat input capacities that typically range from 5 to 550 MMBtu/hr. However, some of
these boilers have capacities as high as 1,500 MMBtu/hr. Smaller stoker units (i.e., those
with heat input capacities less than 100 MMBtu/hr) are generally packaged, while larger
units are usually field erected.
2.2.3.2 Pulverized Coal Boilers (Coal)
Combustion in pulverized coal-fired units takes place almost entirely while the coal is
suspended, unlike in stoker units in which the coal burns on a grate. Finely ground coal is
typically mixed with primary combustion air and fed to the burner or burners, where it is
ignited and mixed with secondary combustion air. Depending on the location of the burners
2-6
-------
and the direction of coal injection into the furnace, pulverized coal-fired boilers can be
classified into three different firing types:
• single and opposed wall,
• tangential, and
• cyclone.
Of these types, wall and tangential configurations are the most common. These firing
methods are described further in Sections 2.2.3.4 and 2.2.3.5.
2.2.3.3 FluidizedBed Combustion (FBC) Boilers (Coal)
FBC is an integrated technology for reducing sulfur dioxide (SO2) and NOX emissions
during the combustion of coal. In a typical FBC boiler, crushed coal and inert material
(sand, silica, alumina, or ash) and/or a sorbent (limestone) are maintained in a highly
turbulent suspended state by the upward flow of primary air from the windbox located
directly below the combustion floor. This fiuidized state provides a large amount of surface
contact between the air and solid particles, which promotes uniform and efficient combustion
at lower furnace temperatures than conventional coal-fired boilers. Once the hot gases leave
the combustion chamber, they pass through the convective sections of the boiler, which are
similar or identical to components used in conventional boilers.
For the FBCs currently in use in all sectors, coal is the primary fuel source, followed
in descending order by biomass, coal waste, and municipal waste. The heat input capacities
of all ICI FBC units generally range from 1.4 to 1,075 MMBtu/hr.
2.2.3.4 Tangentially Fired Boilers (Coal, Oil, Natural Gas)
The tangentially fired boiler is based on the concept of a single flame zone within the
furnace. The fuel-air mixture projects from the four corners of the furnace along a line
tangential to an imaginary cylinder located along the furnace centerline. As fuel and air are
fed to the burners and the fuel is combusted, a rotating "fireball" is formed. Primarily
because of their tangential firing pattern, which leads to larger flame volumes and flame
interaction, uncontrolled tangentially fired boilers generally emit relatively lower NOX than
other uncontrolled boiler designs.
Utilities primarily use this type of boiler. Coal is the most common fuel used by
these units. Tangentially fired boilers operated by utilities are typically larger than 400 MW,
2-7
-------
while industrial ones almost always have heat input capacities over 100 MMBtu/hr. In
general, most units with heat input capacities over 100 MMBtu/hr are field erected.
2.2.3.5 Wall-fired Boilers (Coal, Oil, Natural Gas)
Wall-fired boilers are characterized by multiple individual burners located on a single
wall or on opposing walls of the furnace. In contrast to tangentially fired boilers, each of the
burners in a wall-fired boiler has a relatively distinct flame zone, and the burners in wall-
fired boilers do not tilt. Superheated steam temperatures are instead controlled by excess air
levels, heat input, flue gas recirculation, and/or steam attemperation (water spray).
Depending on the design and location of the burners, wall-fired boilers are referred to as
single wall or opposed wall.
Wall-fired boilers are used to burn coal, oil, or natural gas, and some designs feature
multifuel capability. Almost all industrial wall-fired boilers have heat input capacities
greater than 100 MMBtu/hr. Opposed-wall boilers in particular are usually much larger than
250 MMBtu/hr heat input capacity and are much more common in utility rather than in
industrial operations. Because of their size, most wall-fired units are field erected. Field-
erected watertube boilers strictly designed for oil firing are more compact than coal-fired
boilers with the same heat input, because of the more rapid combustion characteristics of fuel
oil. Field-erected watertube boilers fired by natural gas are even more compact because of
the rapid combustion rate of the gaseous fuel, the low flame luminosity, and the ash-free
content of natural gas.
2.3 Process Heater Characterization
Process heaters are heat transfer units in which heat from fuel combustion is
transferred to materials used in a production process. The process fluid stream is heated
primarily for one of two reasons: to raise the temperature for additional processing or to
make chemical reactions occur. This section describes the different classes of process
heaters and major design types.
2.3.1 Classes of Process Heaters
The universe of process heaters is divided into two categories:
• indirect-fired process heater—any process heater in which the combustion gases
do not mix with or exhaust to the atmosphere from the same stack(s) or vent(s)
with any gases emanating from the process or material being processed.
2-8
-------
• direct-fired process heater—any process heater in which the combustion gases
mix with and exhaust to the atmosphere from the same stack(s) or vent(s) with
gases originating from the process or material being processed.
Indirect-fired units are used in situations where direct flame contact with the material
being processed is undesirable because of problems with contamination and ignition of the
process material. Direct-fired units are used where such problems are not an important
factor. Emissions of indirect-fired units consist solely of the products of combustion
(including those of incomplete combustion). On the other hand, direct-fired units will
generate emissions consisting not only of the products of combustion, but also the process
material(s). This means that the emissions from indirect-fired process heaters will be generic
to the fuel in use and are common across industries while emissions from direct-fired process
heaters are unique to a given process and may vary widely depending on the process
material. Only indirect-fired process heaters are considered under this regulation. Many
direct-fired process heaters are being considered under separate MACT-development
projects.
In addition to the distinction between direct- and indirect-fired heaters, process
heaters may also be considered either heated feed or reaction feed. Heated feed process
heaters are used to heat a process fluid stream before additional processing. These types of
process heaters are used as preheaters for various operations in the petroleum refining
industry such as distillation, catalytic cracking, hydroprocessing, and hydroconversion. In
addition, heated feed process heaters are used widely in the chemical manufacturing industry
as fired reactors (e.g., steam-hydrocarbon reformers and olefins pyrolysis furnaces), feed
preheaters for nonfired reactors, reboilers for distillation operations, and heaters for heating
transfer oils. Reaction feed process heaters are used to provide enough heat to cause
chemical reactions to occur inside the tubes being heated. Many chemical reactions do not
occur at room temperature and require the application of heat to the reactants to cause the
reaction to take place. Applications include steam-hydrocarbon reformers used in ammonia
and methanol manufacturing, pyrolysis furnaces used in ethylene manufacturing, and thermal
cracking units used in refining operations.
2.3.2 Major Design Types
Process heaters may be designed and constructed in a number of ways, but most
process heaters include burner(s), combustion chamber(s), and tubes that contain process
fluids. Sections 2.3.2.1 through 2.3.2.4 describe combustion chambers setups, combustion
air supply, tube configurations, and burners, respectively.
2-9
-------
2.3.2.1 Combustion Chamber Set-Ups
Process heaters contain a radiant heat transfer area in the combustion chamber. This
area heats the process fluid stream in the tubes by flame radiation. Equipment found in this
area includes the burners) and the combustion chamber(s). Most heat transfer to the process
fluid stream occurs here, but these tubes do not necessarily constitute a majority of the tubes
in which the process fluid flows.
Most process heaters also use a convective heat transfer section to recover residual
heat from the hot combustion gases by convective heat transfer to the process fluid stream.
This section is located after the radiant heat transfer section and also contains tubes filled
with process fluid. The first few rows of tubes in this section are called shield tubes and are
subject to some radiant heat transfer. Typically, the process fluid flows through the
convective section prior to entering the radiant section to preheat the process fluid stream.
The temperature of the flue gas upon entering the convective section usually ranges from
800°C to 1,000°C (1,500°F to 2,000°F). Preheating in the convective section improves the
efficiency of the process heater, particularly if the tube design includes fins or other extended
surface areas. An extended tube surface area can improve efficiency by 10 percent.
Extended tubes can reduce flue gas temperatures from 800°C to 1,000°C to (1,500°F to
2,000°F) to 120°C to 260°C (250°F to 500°F).
2.3.2.2 Combustion Air Supply
Air for combustion is supplied to the burners via either natural draft (ND) or
mechanical draft (MD) systems. Natural draft heaters use ductwork systems to route air,
usually at ambient conditions, to the burners. MD heaters use fans in the ductwork system to
supply air, usually preheated, to the burners. The combustion air supply must have sufficient
pressure to overcome the burner system pressure drops caused by ducting, burner registers,
and dampers. The pressure inside the firebox is generally a slightly negative draft of
approximately 49.8 to 125 Pascals (Pa) at the radiant-to-convective section transition point.
The negative draft is achieved in ND systems via the stack effect and in MD systems via fans
or blowers.
ND combustion air supply uses the stack effect to induce the flow of combustion air
in the heater. The stack effect, or thermal buoyancy, is caused by the density difference
between the hot flue gas in the stack and the significantly cooler ambient air surrounding the
stack. Approximately 90 percent of all gas-fired heaters and 76 percent of all oil-fired
heaters use ND combustion air supply (EPA, 1993).
2-10
-------
There are three types of MD combustion air supply: forced draft, induced draft, and
balanced draft. The draft types are named according to the position, relative to the
combustion chamber, of the fans used to create the pressure difference in the process heater.
All three types of MD heaters rely on the fans to supply combustion air and remove flue gas.
In forced draft combustion air supply systems, the fan is located upstream from the
combustion chamber, supplying combustion air to the burners. The air pressure supplied to
the burners in a forced draft heater is typically in the range of 0.747 to 2.49 kilopascals
(kPa). Though combustion air is supplied to the burners under positive pressure, the
remainder of the process heater operates under negative pressure caused by the stack effect.
In induced draft combustion air systems, the fan is located downstream of the combustion
chamber, creating negative pressure inside the combustion chamber.
This negative pressure draws, or induces, combustion air into the burner registers.
Balanced draft combustion air systems use fans placed both upstream and downstream
(forced and induced draft) of the combustion chamber.
There are advantages and disadvantages for both ND and MD combustion air supply.
One advantage to natural draft heaters is that they do not require the fans and equipment
associated with MD combustion air supply. However, control over combustion air flow is
not as precise in ND heaters as in MD heaters. MD heaters, unlike ND heaters, provide the
option of using alternate sources of combustion oxygen, such as gas turbine exhaust. They
also allow the use of combustion air preheat. Combustion air preheat has limited application
in ND heaters due to the pressure drops associated with combustion air preheaters.
Combustion air preheaters are often used to increase the efficiency of MD process
heaters. The maximum thermal efficiency obtainable with current air preheat equipment is
92 percent. Preheaters allow heat to be transferred to the combustion air from flue gas,
steam, condensate, hydrocarbon, or other hot streams. The preheater increases the efficiency
of the process heater because some of the thermal energy is reclaimed that would have been
exhausted from the hot streams via cooling towers. If the thermal energy is from a hot
stream other than the flue gas, the entire plant's efficiency is increased. The benefit of
higher thermal efficiency is that less fuel is required to operate the heater.
2.3.2.3 Tube Configurations
The orientation of the tubes through which a process fluid stream flows is also taken
into consideration when designing a process heater. The tubes in the convective section are
oriented horizontally in most process heaters to allow cross-flow convection. However, the
2-11
-------
tubes in the radiant area may be oriented either horizontally or vertically. The orientation is
chosen on a case-by-case basis according to the design specifications of the individual
process heater. For example, the arbor, or wicket, type of heater is a specialty design to
minimize the pressure drop across the tubes.
2.3.2.4 Burners
Many different types of burners are used in process heaters. Burner selection
depends on several factors including process heat flux requirements, fuel type, and draft
type. The burner chosen must provide a radiant heat distribution that is consistent with the
configuration of the tubes carrying process fluid. Also, the number and location of the
burner(s) depend on the process heater application.
Many burner flame shapes are possible, but the most common types are fiat and
conical. Flat flames are generally used in applications that require high temperatures such as
ethylene pyrolysis furnaces, although some ethylene furnaces use conical flames to achieve
uniform heat distribution. Long conical flames are used in cases where a uniform heat
distribution is needed in the radiant section.
Fuel compatibility is also important in burner selection. Burners may be designed for
combustion of oil, gas, or a gas/oil mixture. Gas-fired burners are simpler in operation and
design than oil-fired burners and are classified as either premix or raw gas burners. In
premix burners, 50 to 60 percent of the air necessary for combustion is mixed with the gas
prior to combustion at the burner tip. This air is induced into the gas stream as the gas
expands through orifices in the burner. The remainder of the air necessary for combustion is
provided at the burner tip. Raw gas burners receive fuel gas without any premixed
combustion air. Mixing occurs in the combustion zone at the burner tip.
Oil-fired burners are classified according to the method of fuel atomization used.
Atomization is needed to increase the mixing of fuel and combustion air. Three types of fuel
atomization commonly used are mechanical, air, and steam. Steam is the most widely used
method because it is the most economical, provides the best flame control, and can handle
the largest turndown ratios. Typical steam requirements are 0.07 to 0.16 kilogram (kg)
steam/kg of oil.
Combination burners can burn 100 percent oil, 100 percent gas, or any combination
of oil and gas. A burner with this capability generally has a single oil nozzle in the center of
a group of gas nozzles. The air needed for combustion can be controlled separately in this
type of burner. Another option is to base load the burners with one fuel and to add the other
2-12
-------
fuel to meet increases in load demand. Combination burners add flexibility to the process
heater, especially when the composition of the fuel is variable.
The location and number of burners needed for a process heater are also determined
on an individual basis. Burners can be located on the ceiling, walls, or floor of the
combustion chamber. Floor- and wall-fired units are the most common burner types found in
process heaters because they are both efficient and flexible. In particular, floor-mounted
burners integrate well with the use of combustion air preheat, liquid fuels, and alternate
sources of combustion oxygen such as turbine exhaust.
The number of burners in a heater can range from 1 to over 100. In the refinery
industry, the average number of burners is estimated at 24 in ND heaters with an average
design heat release of 69.4 million Btu per hour (MMBtu/hr). The average number of
burners is estimated at 20 in MD heaters with ambient combustion air and an average design
heat release of 103.6 MMBtu/hr. The average number of burners is estimated at 14 in MD
heaters with combustion air preheat and an average design heat release of 135.4 MMBtu/hr.
In general, the smaller the number of burners, the simpler the heater will be. However,
multiple burners provide a more uniform temperature distribution.
2-13
-------
SECTION 3
PROFILE OF AFFECTED UNITS AND FACILITIES, AND COMPLIANCE COSTS
The floor-level MACT for the regulation will affect existing and new ICI boilers and
process heaters that have input capacity greater than 10 million Btus and are fueled by fossil
and nonfossil fuel solids and liquids. The economic impact estimates presented in Section 6
and the small entity screening analysis presented in Section 7 are based on the estimated
stock of existing units and the projection of new units through the year 2005. They are also
based on the compliance costs associated with applying the final rule to these units. This
section begins with a review of the industry distribution and technical characteristics of
existing boilers and process heaters contained in the Agency's Inventory Database. It also
presents projected growth estimates for boilers and process heaters through the year 2005, a
description of how costs are estimated, and the national engineering cost estimates and cost-
effectiveness (cost/ton) estimates by pollutant controlled.
3.1 Regulatory Alternatives
Section 112 of the CAA requires EPA to promulgate regulations for the control of
HAP emissions from each source category listed under section 112(c). The statute requires
the regulations to reflect the maximum degree of reductions in emissions of HAP that is
achievable taking into consideration the cost of achieving emissions reductions, any nonair
quality health and environmental impacts, and energy requirements. This level of control is
commonly referred to as MACT. The MACT regulation can be based on the emissions
reductions achievable through application of measures, processes, methods, systems, or
techniques including, but not limited to: (1) reducing the volume of, or eliminating emissions
of, such pollutants through process changes, substitutions of materials, or other
modifications; (2) enclosing systems or processes to eliminate emissions; (3) collecting,
capturing, or treating such pollutants when released from a process, stack, storage or fugitive
emission point; (4) design, equipment, work practices, or operational standards as provided
in subsection 112(h); or (5) a combination of the above.
For new sources, MACT standards cannot be less stringent than the emission control
achieved in practice by the best-controlled similar source. The MACT standards for existing
sources can be less stringent than standards for new sources, but they cannot be less stringent
3-1
-------
than the average emission limitation achieved by the best-performing 12 percent of existing
sources for categories and subcategories with 30 or more sources, or the best-performing 5
sources for categories or subcategories with fewer than 30 sources.
In essence, these MACT standards would ensure that all major sources of air toxic
emissions achieve the level of control already being achieved by the better-controlled and
lower-emitting sources in each category. This approach provides assurance to citizens that
each major source of toxic air pollution will be required to effectively control its emissions.
A major source of HAP emissions is any stationary source or group of stationary sources
located within a contiguous area and under common control that emits or has the potential to
emit any single HAP at a rate of 9.07 Mg (10 tons) or more per year or any combination of
HAPs at a rate of 22.68 Mg (25 tons) or more a year. At the same time, this approach
provides a level economic playing field, ensuring that facilities that employ cleaner processes
and good emission controls are not disadvantaged relative to competitors with poorer
controls.
3.1.1 Regulatory Background
In September 1996, EPA chartered the Industrial Combustion Coordinated
Rulemaking (ICCR) advisory committee under the Federal Advisory Committee Act
(FACA). The committee's objective was to develop recommendations for regulations for
several combustion source categories under sections 112 and 129 of the CAA. The ICCR
advisory committee, known as the Coordinating Committee, formed Source Work Groups
for the various combustion types covered under the ICCR. One of the work groups was
formed to research issues related to boilers. Another was formed to research issues related to
process heaters. The Boiler and Process Heater Work Groups submitted recommendations,
information, and data analysis results to the Coordinating Committee, which in turn
considered them and submitted recommendations and information to EPA. The Committee's
recommendations were considered by EPA in developing these proposed standards for
boilers and process heaters. The Committee's 2-year charter expired in September 1998.
Following the expiration of the ICCR FACA charter, EPA decided to combine boilers
with units in the process heater source category covering indirect fired units, and to regulate
both under this NESHAP. This was done because indirect fired process heaters and boilers
are similar devices, burn similar fuel, have similar emission characteristics, and emissions
from each can be controlled using similar control devices or techniques.
3-2
-------
3.1.2 Regulatory Authority
Section 112 of the CAA requires that EPA promulgate regulations requiring the
control of HAP emissions from major sources and certain area sources. The control of HAP
is achieved through promulgation of emission standards under sections 112(d) and (f) and, in
appropriate circumstances, work practice standards under section 112(h) of the CAA.
An initial list of categories of major and area sources of HAP selected for regulation
in accordance with section 112(c) of the CAA was published in the Federal Register on July
16, 1992 (57 FR 31576). Industrial boilers, commercial and institutional boilers, and process
heaters are three of the listed 174 categories of sources. The listing was based on the
Administrator's determination that they may reasonably be anticipated to emit several of the
188 listed HAP in quantities sufficient to designate them as major sources.
This rule affects industrial boilers, institutional and commercial boilers, and process
heaters. In this rule process heaters are defined as units in which the combustion gases do
not directly come into contact with process gases in the combustion chamber (e.g. indirect
fired). Boiler means an enclosed device using controlled flame combustion and having the
primary purpose of recovering thermal energy in the form of steam or hot water. A waste
heat boiler (or heat recovery steam generator) is a device that recovers normally unused
energy and converts it to usable heat. Waste heat boilers are excluded from this rule. A hot
water heater is a closed vessel in which water is heated by combustion of gaseous fuel and is
withdrawn for use external to the vessel at pressures not exceeding 160 psig. Hot water
heaters are excluded from this rule.
Boilers and process heaters emit particulate matter, volatile organic compounds, and
hazardous air pollutants, depending on the material burned. Solid and liquid fuel-fired units
emit metals, halogenated compounds and organic compounds. Gas fuel-fired units emit
mostly organic compounds.
The affected source is each individual industrial, commercial, or institutional boiler
or process heater located at a major facility. The affected source does not include units that
are municipal waste combustors (40 CFR part 60, subparts AAAA, BBBB or Cb), medical
waste incinerators (40 CFR part 60, subpart Ce and EC), fossil fuel fired electric utility steam
generating units, commercial and industrial solid waste incineration units (40 CFR part 60
subparts CCCC or DDDD), recovery boilers or furnaces (40 CFR part 63, subpart MM), or
hazardous waste combustion units required to have a permit under section 3005 of the Solid
Waste Disposal Act or are subject to 40 CFR part 63, subpart EEE.
3-3
-------
The rule applies to an owner or operate a boiler or process heater at a major source
meeting the requirements in section E.G. A major source of HAP emissions is any stationary
source or group of stationary sources located within a contiguous area and under common
control that emits or has the potential to emit any single HAP at a rate of 9.07 Mg (10 tons)
or more per year or any combination of HAP at a rate of 22.68 Mg (25 tons) or more a year.
An affected operator must meet the emission limits for the subcategories in Table 3-1
of this preamble for each of the pollutants listed. Emission limits were developed for new
and existing sources; and for large, small, and limited use solid, liquid, and gas fuel fired
units. Large units are those with heat input capacities greater than 10 MMBtu/hr. Small
units are those with heat input capacities less than or equal to 10 MMBtu/hr. Limited use
units are those with capacity utilizations less than or equal to 10 percent as required in a
federally enforceable permit.
If your new or existing boiler or process heater is permitted to burn a solid fuel, or
any combination of solid fuel with liquid or gaseous fuel, the unit is in one of the solid
subcategories. If your new or reconstructed boiler or process heater burns a liquid fuel, or a
liquid fuel in combination with a gaseous fuel, the unit is in one of the liquid subcategories.
If your new or existing boiler or process heater burns a gaseous fuel only, the unit is in the
gas subcategory and is not required to meet any emission limit.
For solid fuel-fired boilers or process heaters, we are allowing sources to choose one
of two emission limit options: (1) existing and new affected sources may choose to limit PM
emissions to the level listed in Table 3-1 or (2) existing and new affected sources may
choose to limit total selected metals emissions to the level listed in Table 3-1 of this
preamble.
If you do not use an add-on control or use an add-on control other than a wet
scrubber, you must maintain opacity level to less than or equal to the level established during
the compliance test for mercury and PM or total selected metals, and maintain the fuel
chlorine content to less than or equal to the operating level established during the HC1
compliance test.
If you use a wet scrubber, you must maintain the minimum pH, pressure drop and
liquid fiowrate above the operating levels established during the performance tests.
If you use a dry scrubber, you must maintain opacity level and the minimum sorbent
injection rate established during the performance test.
3-4
-------
Table 3-1. Emission Limits for Boilers and Process Heaters (Ib/MMBtu)
Source
New
Boiler or
Process
Heater
Subcategory
Solid Fuel, Large Unit
Solid Fuel, Small Unit
Solid Fuel, Limited
PM
0.04
0.04
0.04
Total
Selected
or Metals
or 0.00007
or 0.00007
or 0.00007
HC1
0.016
0.032
0.032
Mercury
(Hg)
0.0000026
0.0000026
0.0000026
Carbon
Monoxide
(CO-ppm @ 3%
oxygen)
200
200
Use
Liquid Fuel,
Large Unit
Liquid Fuel, Small
Unit
Liquid Fuel, Limited
Use
Gaseous Fuel, Large
Unit
Gaseous Fuel, Small
Unit
Gaseous Fuel, Limited
Use
0.068
0.068
0.068
— 0.00045
— 0.0009
— 0.0009
200
200
200
200
Existing
Boiler or
Process
Heater
Solid Fuel, Large Unit 0.062 or 0.001 0.048 0.000004 —
Solid Fuel, Small Unit — — — — —
Solid Fuel, Limited
Use
Liquid Fuel,
Large Unit
Liquid Fuel, Small
Unit
Liquid Fuel, Limited
Use
Gaseous Fuel
0.21
0.001
3-5
-------
If you use an ESP in combination with a wet scrubber and cannot monitor the
opacity, you must maintain the average secondary current and voltage or total power input
established during the performance test.
There is an alternative compliance procedure and operating limit for meeting the total
selected metals emission limit option. If you have no control or do not want to take credit of
metals reductions with your existing control device, and can show that total metals in the fuel
would be less than the metals emission level, then you can monitor the metals fuel analysis to
meet the metals emissions limitations. Similarly, if you have no control or do not want to
take credit of mercury reduction with your existing control device, and can show that
mercury in the fuel would be less than the mercury emission level, then you can monitor the
mercury fuel analysis to meet the mercury emission limitations.
3.1.3 Regulatory Alternatives and Control Technologies
3.1.3.1 MA CT Floor Development
We considered several approaches to identifying MACT floor for existing industrial,
commercial, and institutional boilers and process heaters. First, we considered using
emissions data on boilers and process heaters to set the MACT floor. However, after review
of the data available, we determined that emissions information was inadequate to set MACT
floors. We then considered using State regulations and permits to set the MACT floors.
However, we found no State regulations or State permits which specifically limit HAP
emissions from these sources.
Consequently, we concluded that the only reasonable approach for determining
MACT floors is to base it on control technology. Information was available on the control
technologies employed by the population of boilers identified by the EPA. We considered
several possible control technologies (i.e., factors that influence emissions), including fuel
substitution, process changes and work practices, and add-on control technologies.
We first considered whether fuel switching would be an appropriate control option
for sources in each subcategory. Both fuel switching to other fuels used in the subcategory
and fuels from other subcategories were considered. This consideration included
determining whether switching fuels would achieve lower HAP emissions. A second
consideration was whether fuel switching could be technically done on boilers and process
heaters in the subcategory considering the existing design of boilers and process heaters. We
also considered the availability of the alternative fuel.
3-6
-------
After considering these factors, we determined that fuel switching was not an
appropriate control technology to be included in determining the MACT floor level of
control for any subcategory. This decision was based on the overall effect of fuel switching
on HAP emissions, technical and design considerations discussed in section IE. A of this
preamble, and concerns about fuel availability.
Based on the data available in the emissions database, we determined that while fuel
switching from solid fuels to gaseous or liquid fuels would decrease PM and some metals
emissions, emissions of some organic HAP would also increase, resulting in uncertain
benefits. We determined that it would be inappropriate in a MACT rulemaking, that is
technology based, to consider a technology that potentially will result in an increase in a
HAP regardless of its potential to reduce other HAP without determining the overall benefit.
Determining the benefits of fuel switching would require an assessment of the risk associated
which each HAP emitted and a determination of which fuel results in the overall lower risk
taking into account the available control technology for each fuel. This assessment will be
performed in a future rulemaking.
A similar determination was made when considering fuel switching to "cleaner" fuels
within a subcategory. For example, the term "clean coal" refers to coal that is lower in sulfur
content and not necessarily lower in HAP content. Data gathered by EPA also indicates that
within specific coal types HAP content can vary significantly. Switching to a "clean coal"
may increase emissions of some HAP. Therefore, fuel switching to a "cleaner" coal would
not be an appropriate option. Fuel switching from coal to biomass would result in similar
impacts on HAP emissions. While metallic HAP emissions would be reduced, emissions of
organics would increase based on information in the emissions database.
Another factor considered was the availability of alternative fuels. Natural gas
pipelines are not available in all regions of the U.S., and natural gas is simply not available
as a fuel for many industrial, commercial, and institutional boilers and process heaters.
Moreover, even where pipelines provide access to natural gas, supplies of natural gas may
not be adequate. For example, it is common practice in cities during winter months (or
periods of peak demand) to prioritize natural gas usage for residential areas before industrial
usage. Requiring EPA regulated combustion units to switch to natural gas would place an
even greater strain on natural gas resources. Consequently, even where pipelines exist some
units would not be able to run at normal of full capacity during these times if shortages were
to occur. Therefore, under any circumstances, there would be some units that could not
comply with a requirement to switch to natural gas.
3-7
-------
Similar problems for fuel switching to biomass could arise. Existing sources burning
biomass generally are combusting a recovered material from the manufacturing or
agriculture process. Industrial, commercial, and institutional facilities that are not associated
with the wood products industry or agriculture may not have access to a sufficient supply of
biomass materials to replace their fossil fuel.
There are many concerns with switching fuels on sources designed and operated to
burn specific fuels. Changes to the fuel type (solid, liquid, or gas) will require extensive
changes to the fuel handling and feeding system (e.g., a stoker using wood as fuel would
need to be redesigned to handle fuel oil or gaseous fuel). Additionally, burners and
combustion chamber designs are generally not capable of handling different fuel types, and
generally cannot accommodate increases or decreases in the fuel volume and shape. Design
changes to allow different fuel use, in some cases, may reduce the capacity and efficiency of
the boiler or process heater. Reduced efficiency may result in a greater degree of incomplete
combustion and, thus, an increase in organic HAP emissions. For the reasons discussed
above, we decided that fuel switching to "cleaner" solid fuels or to liquid or gaseous fuels
would not be appropriate or available as a MACT floor level.
We also determined that using process changes or work practices were not
appropriate in developing MACT floors. HAP emissions from boilers and process heaters
are primarily dependent upon the composition of the fuel. Fuel dependent HAP are metals,
including mercury, and acid gases. Fuel dependent HAP are typically controlled by
removing them from the flue gas after combustion. Therefore, they are not affected by the
operation of the boiler or process heater. Consequently, process changes would be
ineffective in reducing these fuel-related HAP emissions.
On the other hand, organic HAP can be formed from incomplete combustion of the
fuel. Data are not available that definitively show that organic HAP emissions are related to
the operation of the boiler or process heater. Some studies indicate that organic HAP are
greatly influence by time, turbulence and temperature. Other studies indicate that organic
HAP emissions are not affected by the operation of the unit. The measurement of CO is
generally an indicator of incomplete combustion since CO will burn to carbon dioxide if
adequate oxygen is available. Correcting incomplete combustion maybe accomplished
through providing more combustion air. Therefore, we consider monitoring and maintaining
CO emission levels to be associated with minimizing organic HAP emission levels and, thus,
CO monitoring would be a good indicator of combustion efficiency and organic HAP
emissions.
3-8
-------
In summary, we determined that considering process changes and work practices
would not be appropriate in developing MACT floors for existing units. We are requesting
comment, and information on emission reductions, on whether there are other GCP practices
that would be appropriate for minimizing organic HAP emissions from industrial,
commercial, and institutional boilers and process heaters.
Consequently, we concluded that add-on control technology is the only factor that
significantly controls HAP emissions.
In order to determine the MACT floor based on add-on control technologies, we first
examined the population database of existing sources. Units not meeting the definition of an
industrial, commercial, or institutional boiler or process heater, and units located at area
sources were removed from the database. The remaining units were divided first into three
subcategories based on fuel state: gaseous fuel-fired, liquid fuel-fired, and solid fuel-fired
units. Each of these three subcategories was then further divided into subcategories based on
capacity: (1) large boilers and process heaters (units with heat inputs greater than 10
MMBtu/hr); (2) small units (with a maximum rated heat input capacity of 10 MMBtu/hr or
less); and (3) limited use units with capacity utilization less than 10 percent.
We identified the types of air pollution control techniques currently used by existing
boilers and process heaters in each subcategory. We ranked those controls according to their
effectiveness in removing the different categories of pollutants; including metallic HAP and
PM, inorganic HAP such as acid gases, mercury, and organic HAP. The EPA ranked these
existing control technologies by incorporating recommendations made by the ICCR, and by
reviewing emissions test data, previous EPA studies, and other literature, as well as by using
engineering judgement.
Based upon the emissions reduction potential of existing air pollution control
techniques, we listed all the boilers and process heaters in the population database in order of
decreasing control device effectiveness for each subcategory. Then the technology basis of
the existing source MACT floor was determined for each pollutant category by identifying
the best-performing 12 percent of units. We then selected the technology used by the median
unit in the best performing 12 percent of units (i.e., the boiler or process heater unit
representing the 94th percentile) as the technology associated with the MACT floor level of
control for each subcategory. As previously described, emissions data for this category is
insufficient to identify the best-performing units. The most appropriate way to identify the
average emission limitation achieved by the best-performing 12 percent of existing sources is
3-9
-------
to identify the technology used by the unit in the middle of the range of the best performing
12 percent of units, i.e., the median unit).
After establishing the technology basis for the existing source MACT floor for each
subcategory and each type of pollutant, the EPA examined the emissions data available for
boilers and process heaters controlled by these technologies to determine achievable
emission levels. The resulting emission levels associated with the existing source MACT
floors for each pollutant are based on the average of the lowest three run average test data
from units using the technology associated with the MACT floor level of control, and by
incorporating operational variability using results from multiple tests on these best
performing units. This approach reasonably ensures that the emission limit selected as the
MACT floor represents a level of control that can be consistently achieved by a unit in the
subcategory using the control technology associated with the MACT floor. This approach is
reasonable because the most informative way to predict the worst reasonably foreseeable
performance of the best-controlled units, with available data, is to examine the available
long-term performance of the best performing units that had multiple test results. In other
words, the EPA considers all units with the same control technology that is properly
designed and operated to be equally well controlled, even if the emission test results from
such units vary considerably.
The level of control "achieved" by the average of the top performing 12 percent of
units is best represented by the average emissions observed from all units using the same
technology as that employed by the unit representing the median of the top 12 percent.
The EPA's review of emissions data indicates that some boilers and process heaters
within each subcategory may be able to meet the floor emission levels without using the air
pollution control technology that is associated with the MACT floor. This is to be expected,
given the variety of fuel types, fuel input rates, and boiler designs included within each
subcategory and the resulting variability in emission rates. Thus, for instance, boilers or
process heaters within the large unit solid fuel subcategory that burn lower percentages of
solid fuels may be able to achieve the emission levels for the large unit solid fuel
subcategory without the need for additional control devices.
Furthermore, solid fuels, especially coal, are very heterogeneous and can vary in
composition by location. Coal analysis data obtained from the electric utility industry in
another rulemaking contained information on the mercury, chlorine, and ash content of
various coals. A preliminary review of this data indicate that the composition can vary
greatly from location to location, and also within location. Based on the range of variation of
3-10
-------
mercury, chlorine, and ash content in coal, it is possible for a unit with a lower performing
control system to have emission levels lower than a unit considered to be included in the best
performing 12 percent of the units.
This situation is reflected in the emissions information used to set the MACT floor
emission limits. In some instances there are boilers with ESP's or other controls that achieve
similar, or lower, outlet emission levels of non-mercury metallic HAP, PM, or mercury to
fabric filters. In most cases, this is due to concentrations entering these other control devices
being lower, even though the percent reduction achieved is lower than fabric filters.
Additionally, the design of some control devices may have a substantial effect on the
their emission reduction capability. For example, fabric filters are largely insensitive to the
physical characteristics of the inlet gas stream. Thus, their design does not vary widely, and
emissions reductions are expected to be similar (e.g. 99 percent reduction of PM). However,
ESP design can vary significantly.
Consequently, since fuel substitution has been determined not to be an appropriate
MACT floor control technology, EPA still considers the fabric filter to be the
best-performing control for non-mercury metallic HAPs, PM, and mercury and only
emissions information for fabric filters was used to develop emission limits. A detailed
discussion of the MACT floor methodology is presented in the memorandum "MACT Floor
Analysis for New and Existing Sources in the Industrial, Commercial, and Institutional
Boilers and Process Heaters Source Categories" in the docket.
Existing Solid Fuel Boilers and Process Heaters Large Units-Heat Inputs Greater
than 10 MMBtu/hr. The most effective control technologies identified for removing
non-mercury metallic HAP and PM are fabric filters. About 14 percent of solid fuel-fired
boilers and process heater use fabric filters. Because this is the technology used by the 94th
percentile (the median of the best-performing 12 percent), the EPA considers a fabric filter to
be the technology basis for the MACT floor for non-mercury metallic HAP control for
existing boilers and process heaters in this subcategory.
The most effective control technologies identified for removing inorganic HAP that
are acid gases, such as hydrogen chloride, are wet scrubbers and packed bed scrubbers.
These technologies are used by about 12 percent of the boilers and process heaters in the
solid fuel subcategory. About 10 percent of solid-fired boilers and process heaters use wet
scrubbers, and approximately 1 percent use packed bed scrubbers. Because wet scrubbers
are the technology used by the 94th percentile (median of the best-performing 12 percent),
3-11
-------
the EPA considers a wet scrubber to be the technology basis for the MACT floor for acid gas
control for existing boilers and process heaters in the solid fuel subcategory. The MACT
floor emission level based on wet scrubbers and incorporating operational variability is 0.048
Ib HCl/MMBtu.
Based on test information on utility boilers, we have concluded that fabric filters are
most effective in controlling mercury, and units having them would constitute the best
controlled mercury sources. As discussed previously, more than 6 percent of sources in the
subcategory have fabric filters. The MACT floor emission level based on fabric filters and
incorporating operational variability is 0.000004 Ib mercury/MMBtu.
For organic HAP, we assessed whether maintaining and monitoring CO levels would
be part of the MACT floor, and determined that less than 6 percent of the units in this
subcategory do so. Therefore, we concluded the MACT floor for existing sources in this
subcategory is no emissions reductions for organic HAP.
Therefore, the EPA determined that the combination of fabric filter and wet scrubber
control technologies forms the basis for the MACT floor level of control for existing solid
fuel boilers or process heaters in this subcategory. We recognize that some boilers and
process heaters that use technologies other than those used as the basis of the MACT floor
can achieve the MACT floor emission levels. For example, emission test data show that
many boilers with well-designed and operated ESP can meet the MACT floor emission
levels for non-mercury metallic HAP and PM, even though the floor emission level for these
pollutants is based on a fabric filter (however, we would not expect that all units using ESP
would be able to meet the proposed rule).
Small Units—Heat Inputs Less than or Equal to 10 MMBtu/hr. Less than 6 percent of
the units in this subcategory used control techniques that would reduce non-mercury metallic
HAP and PM, mercury, and inorganic HAP, such as HC1. Also, maintaining and monitoring
CO levels was used by less than 6 percent of the units in the subcategory.
Therefore, we determined that the MACT floor emission level for existing units for
any of the pollutant categories in this subcategory is no emissions reductions.
Limited Use Units—Capacity Utilizations Less than or Equal to 10 Percent. The
most effective control technologies identified for removing non-mercury metallic HAP and
PM are ESP and fabric filters. Less than 2 percent of solid fuel-fired boilers and process
heater in this subcategory use fabric filters, and 14 percent use ESP. Because ESP are the
technology used by the 94th percentile (the median of the best-performing 12 percent), the
3-12
-------
EPA considers an ESP to be the technology basis for the MACT floor for non-mercury
metallic HAP control for existing boilers and process heaters in the solid fuel subcategory.
A PM level is set as a surrogate for non-mercury metallic HAP control. The MACT floor
emission level based on ESPs, considering operational variability, is 0.021 Ib PM/MMBtu.
We are also providing an alternative metals limit of 0.001 Ib metals/MMBtu which can be
used to show compliance in cases where metal HAP emissions are low in proportion to PM
emissions.
Similar control technology analyses were done for the boilers and process heaters in
this subcategory for the other pollutant groups of interest, including inorganic HAP, organic
HAP and mercury. Less than 6 percent of the units in this subcategory have controls that
would reduce emissions of organic HAP, mercury, and inorganic HAP, so the existing source
MACT floor for those pollutants is no emissions reductions. Therefore, we determined that
ESP control technology, which achieves non-mercury metallic HAP and PM control forms
the basis for the MACT floor level of control for existing solid fuel boilers and process
heaters in this subcategory.
Existing Liquid Fuel Boilers and Process Heaters. Emissions data for liquid
subcategories was inadequate to identify the best-performing sources for reasons described in
section D of the preamble. We also found no State regulations or permits which specifically
limit HAP emissions from these sources. Therefore, we examined control technology data to
identify a MACT floor. We found that less than 6 percent of the units in each of the liquid
subcategories used control techniques that would reduce non-mercury metallic HAP and PM,
mercury, organic HAP, or inorganic HAP (such as HC1). Therefore, we determined that the
control technique associated with the 94th percentile (the median of the best-performing 12
percent) could not be identified.
Therefore, we are unable to identify the best performing 12 percent of units in the
subcategories. In light of this analysis, we concluded the MACT floor for existing sources in
these liquid subcategory is no emissions reductions for non-mercury metallic HAP, mercury,
inorganic HAP, and organic HAP.
Existing Gaseous Fuel Boilers and Process Heaters. Emissions data for gas
subcategories was inadequate to identify the best-performing sources for reasons described in
section D of the preamble. We also found no State regulations or permits which specifically
limit HAP emissions from these sources. Therefore, we examined control technology data to
identify a MACT floor. We found that no existing units in the gaseous fuel-fired
subcategories were using control technologies that achieve consistently lower emission rates
3-13
-------
than uncontrolled sources for any of the pollutant groups of interest. Therefore, we are
unable to identify the best performing 12 percent of units in the subcategories.
Consequently, the EPA determined that no existing source MACT floor based on control
technologies could be identified for gaseous fuel-fired units. Therefore, we concluded the
MACT floor for existing sources in this subcategory is no emissions reductions for
non-mercury metallic HAP, mercury, inorganic HAP, and organic HAP.
3.1.3.2 Consideration of Options Beyond the Floor for Existing Units
Once the MACT floor determinations were done for each subcategory, the EPA
considered various regulatory options more stringent than the MACT floor level of control
(i.e., technologies or other work practices that could result in lower emissions) for the
different subcategories.
Maintaining and monitoring CO levels was identified as a possible control for
organic HAPs. However, less than 6 percent of the sources in the existing source
subcategories used this control method and it was not considered the MACT floor control
technology. We then looked at it as an above-the-floor option. However, information was
not available to estimate the HAP emissions reductions that would be associated with CO
monitoring and emission limits. This option would also require a high cost to install and
operate CO monitors. Given the cost and the uncertain emissions reductions that might be
achieved, we chose to not require CO monitoring and emission limits as MACT.
The following sections discuss the above-the-floor options analyzed to control
emissions of metallic HAP, mercury, and inorganic HAP. Based on the analysis described in
these sections, the EPA decided to not go beyond the MACT floor level of control for the
proposed rule for any of the subcategories of existing sources.
Existing Solid Fuel Units. Large Units—Heat Inputs Greater than 10 MMBtu/hr.
Besides fuel switching (see section III.D of this preamble), we identified a better designed
and operated fabric filter (the MACT floor for new units) as a control technology that could
achieve greater emissions reductions of metallic HAP and PM emissions than the MACT
floor level of control (i.e., a typical existing fabric filter). Consequently, the EPA analyzed
the emissions reductions and additional cost of adopting an emission limit representative of
the performance of a unit with a better designed and operated fabric filter. The additional
annualized cost to comply with this emission limit was estimated to be approximately 500
million dollars with an additional emission reduction of approximately 100 tons of metallic
HAP. The results indicated that while additional emissions reductions would be realized, the
3-14
-------
costs would be too high to consider it a feasible above the floor option. Non-air quality
health, environmental impacts, and energy effects were not significant factors, because there
would be little difference in the non-air quality health and environmental impacts of
replacing existing fabric filters with improved performance fabric filters. Therefore, we did
not select these controls as MACT. Fuel switching was not considered a feasible
beyond-the-fioor option for the same reasons described in section HIE of the proposal
preamble.
We identified packed bed scrubbers as a control technology that could achieve
greater emissions reductions of inorganic HAP, like HC1, than the MACT floor level of
control (i.e., a wet scrubber). Consequently, the EPA analyzed the emissions reductions and
additional cost of adopting an emission limit representative of the performance of a unit with
a packed bed scrubber. The additional annualized cost to comply with this emission limit
(using a packed bed scrubber) was estimated to be approximately 900 million dollars with
an additional emission reduction of approximately 20,000 tons of HC1. The results indicated
that while additional emissions reductions would be realized, the costs would be too high to
consider it a feasible above the floor option. Non-air quality health, environmental impacts,
and energy effects were not significant factors, because there would be little difference in the
non-air quality health and environmental impacts between packed bed scrubbers and wet
scrubbers. Therefore, we did not select these controls as MACT.
In reviewing potential regulatory options for existing sources, the EPA identified one
existing industrial boiler that was using a technology, carbon injection, used in other
industries to achieve greater control of mercury emissions than the MACT floor level of
control. However, emission data indicated that this unit was not achieving mercury emission
reductions. The EPA does not have information that would show carbon injection is
effective for reducing mercury emissions from industrial, commercial, and institutional
boilers and process heaters. Therefore, carbon injection was not evaluated as a regulatory
option.
However, the EPA requests comments on whether carbon injection should be
considered as a beyond-the-fioor option and whether existing industrial, commercial, or
institutional boilers and process heaters could use carbon injection technology, or other
control techniques to consistently achieve mercury emission levels that are lower than levels
from similar sources with the MACT floor level of control. The EPA is aware that research
continues on ways to improve mercury capture by PM controls, sorbent injection, and the
3-15
-------
development of novel techniques. The EPA requests comment and information on the
effectiveness of such control technologies in reducing mercury emissions.
Small Units—Heat Inputs Less than or Equal to 10 MMBtu/hr. The EPA could not
identify a technology-based level of control for the MACT floor for this subcategory. To
control non-mercury metallic HAP and mercury, we analyzed the above the floor option of a
fabric filter which was identified as the most effective control device for non-mercury
metallic HAP and mercury. To control inorganic HAP such as hydrogen chloride, we
analyzed the above the floor option of a wet scrubber since it was identified as the least cost
option.
The total annualized cost of complying with the fabric filter option was estimated to
be $10 million, with an estimated emission reduction of 1.9 tons per year of non-mercury
metallic HAP and 0.003 tons of mercury. The annualized cost of complying with the wet
scrubber option was estimated to be $11 million, with an emission reduction of 48 per year
of HC1. The results of this analysis indicated that while additional emissions reductions
could be realized, the costs would be too high to consider them feasible options. Therefore,
we did not select these controls as MACT. Non-air quality health, environmental impacts,
and energy effects were not significant factors.
Limited Use Units—Capacity Utilizations Less than or Equal to 10 Percent. The
MACT floor level of control for this subcategory for non-mercury metallic HAP control is an
ESP. Although fabric filters were identified as being more effective, many ESP can achieve
similar levels. Any additional emission reduction from using a fabric filter would be
minimal and costly considering retrofit costs for existing units that already have ESP.
Therefore, an above-the-floor option for metallic HAP was not analyzed in detail, and we did
not select fabric filters as MACT. However, an above the floor option of a fabric filter was
analyzed for mercury control. The total annualized costs of the fabric filter option was
estimated to be an additional $21 million, with an estimated emission reduction of 0.04 tons
of mercury.
The EPA could not identify a technology-based level of control for the MACT floor
for inorganic HAP in this subcategory. To control inorganic HAP, we analyzed the
above-the-floor option of a wet scrubber since it was identified as the least cost option. The
total annualized costs of the wet scrubber option was estimated to be $49 million, with an
estimated emission reduction of 463 tons per year of HC1.
3-16
-------
The results of the above the floor options analyses indicated that while additional
emissions reductions could be realized, the costs would be too high to consider them feasible
options. Therefore, we did not select these controls as MACT. Non-air quality health,
environmental impacts, and energy effects were not significant factors.
Existing Liquid Fuel Units. For the liquid fuel subcategories, the EPA could not
identify a technology-based level of control for the MACT floor. For beyond-the-fioor
options for the liquid subcategory, the EPA identified several PM controls (e.g., fabric filters,
electrostatic precipitators, and venturi scrubbers) that would reduce non-mercury metallic
HAP emissions. For the above-the-fioor analysis, we analyzed the cost and emission
reduction of applying a high efficiency PM control device, such as a fabric filter, since these
would be more likely to be installed for units firing liquid fuel. We identified wet scrubbers
as a technology option beyond the floor for reduction of inorganic HAP, such as HC1. We
identified fabric filters as a technology option beyond the floor for reduction of mercury.
Consequently, the EPA analyzed the emissions reductions and additional cost of applying
high efficiency PM controls and wet scrubbers on liquid fuel-fired units. The additional total
annualized cost of a high efficiency PM control device (such as a fabric filter) was estimated
to be $460 million, with an additional estimated emission reduction of 1,500 tons per year for
non-mercury metallic HAP and 3 tons per year for mercury. The annualized cost of a wet
scrubbers was estimated to be an additional $480 million, with an additional HC1 reduction
of 30 tons per year. The results indicated that while additional emissions reductions would
be realized, the costs would be too high to consider them feasible options. Non-air quality
health, environmental impacts, and energy effects were not significant factors. Therefore,
the EPA chose to not select these controls as MACT for existing liquid units.
Existing Gas-fired Units. For the gaseous fuel subcategories, the EPA could not
identify a technology-based level of control for the MACT floor. The great majority, if not
all, of the emissions from gas-fired units are organic HAP. As discussed in section in.E of
the preamble, CO monitoring and emission limits were considered as an above the floor
option but was not selected as MACT given the costs and uncertain reductions achieved.
Therefore, no above the floor control technique was analyzed for organic HAPs, and MACT
is no emission reduction of non-mercury metallic HAP and mercury, inorganic HAP, and
organic HAP.
Fuel Switching as a Beyond-the-floor Option. For the solid fuel and liquid fuel
subcategories, fuel switching to natural gas is a regulatory option more stringent than the
MACT floor level of control that would reduce mercury, metallic HAP, and inorganic HAP
3-17
-------
emissions. We determined that fuel switching was not an appropriate above-the-fioor option
for the reasons discussed in sections IK A and in.D of this proposal preamble. In some
cases, organic HAP would be increased by fuel switching. Additionally, the estimated
emissions reductions that would be achieved if solid and liquid fuel units switched to natural
gas were compared with the estimated cost of converting existing solid fuel and liquid fuel
units to fire natural gas. The annualized cost of fuel switching was estimated to be $12
billion. The additional emission reduction associated with it was estimated to be 1,500 tons
per year for metallic HAP, 11 tons per year for mercury, and 13,000 tons per year for
inorganic HAP. Additional detail on the calculation procedures is provided in the
memorandum "Development of Fuel Switching Costs and Emissions reductions for
Industrial, Commercial, and Institutional Boilers and Process Heaters" in the docket.
3.1.3.3 EPA Response to Recent Court Decisions in Developing the Emission Limitations
In developing the emission limitations, we tried to be responsive to the recent court
decisions from National Lime Association v. EPA and Cement Kiln Recycling Coalition v.
EPA, regarding the methodology used for determining the MACT floor. In response, we
determined that the most acceptable and appropriate approach for determining the MACT
floor appears to be using only emission data. As discussed and explained in section II.E of
the proposal preamble, we determined that for these source categories and the subcategories
established the use of only the available emission data would be inappropriate for
determining the MACT floor for existing and new units. If only the available emission data
(from a population of units that is deemed unrepresentative) is used, the resulting MACT
floor emission levels would be, in most many cases, unachievable. This is because the
concentration of HAP (metals, HC1, mercury) vary greatly within each fuel type. Some even
have fuel analysis levels below the detection limit. Therefore, some units without any
add-on controls have emission levels below those with add-on controls. Section HIE of the
proposal preamble explains in more detail the approach used to develop the MACT floors for
each subcategory and why the approach is appropriate for the subcategories regulated by this
rule and why the mandating of fuel choice (using low HAP-containing fuel) is also
inappropriate.
In terms of subcategorizing, the main difficulty of establishing a separate subcategory
for each specific fuel type is that many industrial boilers burn a combination of fuels.
Determining which subcategory applies if the mixture varies would be problematic. Would
the applicable emission limits change each time the fuel mixture changes? How would
3-18
-------
compliance be determine and how would continuous compliance be monitored? Because of
these concerns, EPA chose not to further subcategorize sources by each specific fuel type.
However, if we were to further subcategorize solid-fuel units into separate fossil and
non-fossil subcategories, we would first determine if the MACT floor could be developed,
for either subcategory, based on emissions information. If not, then we would look at
developing MACT floors based on control technologies. First we would determine if fuel
switching or work practices could be used. Based on the MACT floor analysis for solid-fuel
fired boilers, it is expected that emissions information and fuel switching would not be
appropriate to develop the MACT floors for a solid fossil or solid non-fossil subcategory.
Similarly, there would be an insufficient number of boilers or process heaters that would be
meeting CO limits to set a level for existing units. However, new units would likely be
subject to a CO limit and monitoring.
In order to determine the MACT floor based on add-on control technologies, we
would follow similar procedures described in section HIE of the preamble. We would
examine the population database of existing sources and subcategorize solid fossil and
non-fossil fuel fired boilers into each of the following three subcategories based on capacity:
(1) large boilers and process heaters (units with heat inputs greater than 10 MMBtu/hr); (2)
small units (with a maximum rated heat input capacity of 10 MMBtu/hr or less); and (3)
limited use units with capacity utilization less than 10 percent.
We would identify the types of air pollution control techniques currently used by
existing boilers and process heaters in each subcategory. Then we would rank those controls
according to their effectiveness in removing the different categories of pollutants; including
metallic HAP and PM, inorganic HAP such as acid gases, mercury, and organic HAP.
Based upon the emissions reduction potential of existing air pollution control
techniques, we would list all the boilers and process heaters in the population database in
order of decreasing control device effectiveness for each subcategory. Then the technology
basis of the existing source MACT floor would be determined for each pollutant category by
identifying the best-performing 12 percent of units. We would then selected the technology
used by the median unit in the best performing 12 percent of units (i.e., the boiler or process
heater unit representing the 94th percentile) as the technology associated with the MACT
floor level of control for each subcategory.
After establishing the technology basis for the existing source MACT floor for each
subcategory and each type of pollutant, we would examine the emissions data available for
3-19
-------
boilers and process heaters controlled by these technologies to determine achievable
emission levels. The resulting emission levels associated with the existing source MACT
floors for each pollutant would be based on the average of the lowest three run average test
data from units using the technology associated with the MACT floor level of control, and by
incorporating operational variability using results from multiple tests on these best
performing units.
The preliminary MACT floor control technology for solid fossil-fuel fired units
would be a combination of a fabric filter and a scrubber. The preliminary MACT floor
control technology for solid non-fossil-fuel fired units would be a combination of an ESP
and a scrubber.
3.1.3.4 How did EPA Determine the Emission Limitations for New Units?
All standards established pursuant to section 112 of the CAA must reflect MACT, the
maximum degree of reduction in emissions of air pollutants that the Administrator, taking
into consideration the cost of achieving such emissions reductions, and any non-air quality
health and environmental impacts and energy requirements, determines is achievable for
each category. The CAA specifies that the degree of reduction in emissions that is deemed
achievable for new boilers and process heaters must be at least as stringent as the emissions
control that is achieved in practice by the best-controlled similar unit. However, the EPA
may not consider costs or other impacts in determining the MACT floor. The EPA may
require a control option that is more stringent than the floor (beyond-the-floor) if the
Administrator considers the cost, environmental, and energy impacts to be reasonable.
Determining the MACT floor for New Units. Similar to the MACT floor process used
for existing units, we considered several approaches to identifying MACT floors for new
industrial, commercial, and institutional boilers and process heaters. First, we considered
using emissions data on boilers and process heaters to set the MACT floor. However, after
review of the data available, we determined that emissions information was inadequate to set
MACT floors. We also reviewed State regulations and permits for these sources, but found
no State regulations or State permits which specifically limit HAP emissions from industrial,
commercial, and institutional boilers and process heaters.
Consequently, we concluded that the only reasonable approach for determining
MACT floors is to base it on control technology. Data were available on the control
technologies employed by the population of boilers identified by the EPA. We considered
3-20
-------
several possible control technologies (i.e., factors that influence emissions), including fuel
substitution, process changes and work practices, and add-on control technologies.
We first considered whether fuel switching would be an appropriate control option
for sources in each subcategory. Both fuel switching to other fuels used in the subcategory
and fuels from other subcategories were considered. This consideration included
determining whether switching fuels would achieve lower HAP emissions. A second
consideration was whether fuel switching could be technically done on boilers and process
heaters in the subcategory considering the existing design of boilers and process heaters. We
also considered the availability of the alternative fuel.
As discussed in section III.D of the proposal preamble, based on the data available in
the emissions database, we determined that while fuel switching would decrease some HAPs,
emissions of some organic HAPs would increase, resulting in uncertain benefits. We
determined that it would be inappropriate in a MACT rulemaking, that is technology based,
to consider a technology that potentially will result in an increase in a HAP regardless of its
potential to reduce other HAP without determining the overall benefit. A detailed discussion
of the consideration of fuel switching is discussed in proposal preamble section in.D.
We also determined that using process changes or work practices were not
appropriate in most cases for developing MACT floors. HAP emissions from boilers and
process heaters are primarily dependent upon the composition of the fuel. Fuel dependent
HAP are metals, including mercury, and acid gases. Fuel dependent HAP are typically
controlled by removing them from the flue gas after combustion. Therefore, they are not
affected by the operation of the boiler or process heater. Consequently, process changes
would be ineffective in reducing their emissions. The exception to this conclusion is
monitoring and maintaining CO levels. The measurement of CO is generally an indicator of
incomplete combustion since CO will burn to carbon dioxide if adequate oxygen is available.
Correcting incomplete combustion may be accomplished through providing more
combustion air. Therefore, we consider monitoring and maintaining CO emission levels to
be associated with minimizing organic HAP emission levels and, thus, CO monitoring would
be a good indicator of combustion efficiency and organic HAP emissions. As discussed in
the final preamble, CO is considered a surrogate for organic HAP emissions in this rule.
To determine if CO monitoring would be the basis of the new source MACT floor for
organic emissions control, we examined available information. The population databases did
not contain information on existing units monitoring CO emissions. We reviewed State
regulations applicable to boilers and process heaters that required the use of CO monitoring
3-21
-------
to maintain a specific CO limit. The analysis of the State regulations indicated that at least
one of the boilers and process heaters in the large and limited use subcategories for solid
fuel, liquid fuel, and gaseous fuel were required to monitor CO emissions and meet a CO
limit of 200 parts per million. Therefore, the new source MACT floor level of control
includes a CO emission limit of 200 parts per million for large and limited use units.
We concluded that, except for CO monitoring for organic HAP, add-on control
technology is the only factor that significantly controls emissions. To determine the MACT
floor for new sources, the EPA reviewed the population database of existing major sources.
Based upon the emission reduction potential of existing air pollution control devices,
the EPA listed all the boilers and process heaters in the population database in order of
decreasing control device effectiveness for each subcategory and each type of pollutant.
Once the ranking of all existing boilers and process heaters was completed for each
subcategory and type of pollutant, the EPA determined the technology basis of the new
source MACT floor by identifying the best-controlled source using the air pollution control
rankings.
After establishing the technology basis for the new source MACT floor for each
subcategory and each type of pollutant, the EPA examined the emissions data available for
boilers and process heaters controlled by these technologies to determine achievable
emission levels for PM (as a surrogate for non-mercury metallic HAP), total selected
non-mercury metallic HAP, mercury, HC1 (as a surrogate for inorganic HAP), and CO (as a
surrogate for organic HAP). This approach is reasonable because the most informative way
to predict the worst reasonably foreseeable performance of the best-controlled unit, with
available data, is to examine the performance of other units that use the same technology. In
other words, the EPA considers all units with the same control technology to be equally well
controlled, and each unit with the best control technology is a "best controlled similar unit"
even if the emission test results from such units vary considerably.
Accordingly, we selected as the floor for new units the level of control that was being
achieved in practice by the best-controlled similar source, that is, the source with emissions
representing the performance of the most effective control technology under the worst
reasonably foreseeable circumstances. A detailed description of the MACT floor
determination is in the memorandum "MACT Floor Analysis for New and Existing Sources
in the Industrial, Commercial, and Institutional Boilers and Process Heaters Source
Categories" in the docket.
3-22
-------
New Solid Fuel-fired Units. Large Units—Heat Inputs Greater than 10 MMBtu/hr.
The most effective control technology identified for removing PM from boilers in this
subcategory is fabric filters. Therefore, the EPA considers a fabric filter to be the technology
basis for the new source MACT floor for non-mercury metallic HAP emissions. The MACT
floor emission level based on fabric filters is 0.04 Ib PM/MMBtu. This PM emission level
was selected from a subset of fabric filters contained in the database. This subset includes
fabric filters assumed to be subject or achieving the NSPS for industrial boilers. The NSPS
(40 CFR 60.40b), which represent best demonstrated technology for criteria pollutants, is
based on the use of a fabric filter for PM and requires the use of a scrubber for sulfur
dioxide. Therefore, fabric filters subjected to the NSPS are assumed to be better designed,
and operated than those built prior to the NSPS.
We are also providing an alternative metals limit of 0.00007 Ib metals/MMBtu which
can be used to show compliance in cases where metal HAP emissions are low in proportion
to PM emissions. The emissions database indicates that some biomass units have low metals
content but high PM emissions. The emission level for metals was selected from metals test
data associated with PM emission tests from fabric filters that met the MACT floor PM
emission level. The most effective control technologies identified for removing inorganic
HAP including acid gases, such as HC1, are wet scrubbers and packed bed scrubbers. Wet
scrubbers is a generic term that is most often used to describe venturi scrubbers, but can
include packed bed scrubbers, impingement scrubbers, etc. One percent of boilers and
process heaters in this subcategory reported using a packed bed scrubber. Emission test data
from other industries suggests that packed bed scrubbers achieve consistently lower emission
levels than wet scrubbers. Therefore, the EPA considers a packed bed scrubber to be the
technology basis for the new source MACT floor for acid gas control for boilers and process
heaters in the solid fuel subcategory. The MACT floor emission level based on packed
scrubbers is 0.016 Ib HCl/MMBtu.
For mercury control, one technology, carbon injection, that has demonstrated
mercury reductions in other source categories (i.e., municipal waste combustors), was
identified as being used on one existing industrial boiler. However, test data on this carbon
injection system indicated that this unit was not achieving mercury emissions reductions.
Therefore, we did not consider carbon injection to be a MACT floor control technology for
industrial, commercial, and institutional boilers and process heaters. Data from electric
utility boilers indicate that fabric filters can achieve mercury emissions reductions.
3-23
-------
Therefore, the EPA considers a fabric filter to be the control technology basis for controlling
mercury in this subcategory. The MACT floor emission level based on fabric filters is
0.0000026 Ib mercury/MMBtu.
Similar control technology analysis was done for the boilers and process heaters in
this subcategory for organic HAP. One control technique, controlling inlet temperature to
the PM control device, that has demonstrated controlling downstream formation of dioxins in
other source categories (e.g., municipal waste combustors) was analyzed for industrial
boilers. Inlet and outlet dioxins test data were available on four boilers controlled with PM
control devices. In all cases, no increase in dioxins emissions were indicated across the PM
control device even at high inlet temperatures. However, we are requesting comment on
controls that would achieve reductions of organic HAP, including any additional data that
might be available. The EPA did find that CO monitoring can reduce organic HAP
emissions, and has included it in the new source MACT floors as described under section
III.F. of this preamble.
In light of this analysis, the EPA determined that the combination of a fabric filter, a
packed bed scrubber, and CO monitoring forms the control technology basis for the new
source MACT floor for boilers and process heaters in this subcategory.
Small Units—Heat Inputs Less than or Equal to 10 MMBtu/hr. The most effective
control technologies identified for removing non-mercury metallic HAP used by units in this
subcategory are fabric filters. Therefore, the EPA considers fabric filters to be the
technology basis for the new source MACT floor for non-mercury metallic HAP control in
this subcategory. The most effective control technology identified for units in this
subcategory for removing acid gases, such as HC1, are wet scrubbers. The most effective
control technologies identified for removing mercury used by units in this subcategory are
fabric filters.
The EPA identified no control technology being used in the existing population of
boilers and process heaters that consistently achieved lower emission rates than uncontrolled
levels, such that a best-controlled similar source for organic HAP could be identified. We
concluded the MACT floor for new sources in this subcategory is no emissions reductions
for organic HAP. Furthermore, CO monitoring is not required for small boilers and process
heaters by any State rules.
3-24
-------
Thus, the EPA determined that the combination of a fabric filter and a wet scrubber
forms the control technology basis for the new source MACT floor for boilers and process
heaters in this subcategory.
The emissions test database did not contain test data for boilers and process heaters
less than 10 MMBtu/hr heat input. In order to develop emission levels for this subcategory,
we decided to use information from units in the large solid subcategory. We considered this
to be an appropriate methodology because although the units in this subcategory are different
enough to warrant their own subcategory (i.e., different designs and emissions), emissions of
the specific HAP for which limits are being proposed (HC1, PM and metals) are expected to
be related more to the type of fuel burned and the type of control used than to the unit design.
Consequently, we determined that emissions information from units greater than 10
MMBtu/hr heat input could be used to establish the MACT floor levels for this subcategory
for HC1, non-mercury metallic HAP (using PM as a surrogate), and mercury because the
fuels and controls are similar.
The MACT floor emission level based on emissions data for fabric filters on solid
fuel-fired boilers is 0.04 Ib PM/MMBtu or 0.00007 Ib selected non-mercury metals/MMBtu,
and 0.0000026 mercury/MMBtu. The MACT floor emission level based on wet scrubbers is
0.032 Ib HCl/MMBtu. We are requesting comment on using emission data from another
subcategory to develop emission levels for this subcategory. We are also requesting any
available emissions information for this subcategory.
Limited Use Units—Capacity Utilizations Less than or Equal to 10 Percent. The
most effective control technologies identified for removing non-mercury metallic HAP and
mercury used by units in this subcategory are fabric filters. Therefore, the EPA considers
fabric filters to be the technology basis for the new source MACT floor for non-mercury
metallic HAP and mercury control in this subcategory. The most effective control
technology identified for units in this subcategory for removing acid gases, such as hydrogen
chloride, are wet scrubbers.
The EPA did find that monitoring CO is used by at least one unit and can reduce
organic HAP emissions, and has included it in the new source MACT floor for this
subcategory as described under section ni.F of this preamble.
Therefore, based on this analysis, the EPA determined that the combination of a
fabric filter, a wet scrubber, and CO monitoring forms the control technology basis for the
new source MACT floor for boilers and process heaters in this subcategory.
3-25
-------
Consequently, we determined that emissions information from units greater than 10
MMBtu/hr heat input could be used to establish MACT floor levels for this subcategory
because the fuels and controls are similar. The MACT floor emission level based on fabric
filters is 0.04 Ib PM/MMBtu or 0.00007 Ib metals/MMBtu, and 0.0000026 mercury/MMBtu.
The MACT floor emission level based on wet scrubbers is 0.032 Ib HCl/MMBtu. We are
requesting comment on using emission data from another subcategory to develop emission
levels for this subcategory. We are also requesting any available emissions information for
this subcategory.
New Liquid Fuel-fired Units. Large Units—Heat Inputs Greater than 10 MMBtu/hr.
The most effective control technologies identified for removing non-mercury metallic HAP
and PM from units in this subcategory are fabric filters. Therefore, the EPA considers a
fabric filter to be the technology basis for the new source MACT floor for non-mercury
metallic HAP. A PM level is set as a surrogate for non-mercury metallic HAP control. The
MACT floor emission level based on emission data for fabric filters on liquid fuel fired
boilers is 0.068 Ib PM/MMBtu. Unlike for solid fuel subcategories, we are not aware of any
liquid fuels that are low in metals but would have high PM emissions. Therefore, we are not
proposing an alternative metals standard for the liquid subcategories.
The most effective control technologies identified for removing inorganic HAP that
are acid gases, such as HC1, are packed bed scrubbers. Therefore, the EPA considers a
packed bed scrubber to be the technology basis for the new source MACT floor for acid gas
control for boilers and process heaters in the liquid fuel subcategory. The MACT floor
emission level based on packed scrubbers is 0.00045 Ib HCl/MMBtu.
Similar control technology analyses were done for the boilers and process heaters in
this subcategory for mercury and organic HAP.
Information in the emissions database or from other source categories does not show
that control technologies, such as fabric filters or wet scrubbers, achieve reductions in
mercury emissions from liquid fuel-fired industrial, commercial, and institutional boilers and
process heaters. Therefore, EPA identified no control technology being used in the existing
population of boilers and process heaters in these subcategories that consistently achieved
lower emission rates than uncontrolled levels, such that a best-controlled similar source for
organic HAP could be identified. However, we did find that monitoring CO is a good
combustion practice that can reduce organic HAP emissions, and has included it in the new
3-26
-------
source MACT floor as described under section III.D of this preamble. We concluded the
MACT floor for new sources in this subcategory is no emissions reductions for mercury.
In light of this analysis, the EPA determined that the combination of a fabric filter, a
packed bed scrubber, and CO monitoring forms the control technology basis for the new
source MACT floor for boilers and process heaters in this subcategory.
Small Units—Heat Inputs Less than or Equal to 10 MMBtu/hr. The most effective
control technologies identified for removing non-mercury metallic HAP used by units in this
subcategory are fabric filters. Therefore, the EPA considers fabric filters to be the
technology basis for the new source MACT floor for non-mercury metallic HAP control in
this subcategory. The most effective control technology identified for units in this
subcategory for removing acid gases, such as hydrogen chloride, are wet scrubbers.
Information in the emissions database or from other source categories does not show
that other control technologies, such as fabric filters or wet scrubbers, achieve reductions in
mercury emissions from liquid fuel-fired industrial, commercial, and institutional boilers and
process heaters. Therefore, EPA could not identify a control technology being used in the
existing population of boilers and process heaters that consistently achieved lower emission
rates than uncontrolled levels, such that a best-controlled similar source for mercury or
organic HAP could be identified. We concluded the MACT floor for new sources in this
subcategory is no emissions reductions for mercury or organic HAP.
Thus, the EPA determined that the combination of a fabric filter and a wet scrubber
forms the control technology basis for the new source MACT floor for boilers and process
heaters in this subcategory.
The emissions test database did not contain test data for boilers and process heaters
less than 10 MMBtu/hr heat input. In order to develop emission levels for this subcategory,
we decided to use information from units in the large liquid subcategory. We considered this
to be an appropriate methodology because although the units in this subcategory are different
enough to warrant their own subcategory (i.e., different designs and emissions), emissions of
the specific types of HAP for which limits are being proposed (HC1 and metals) are expected
to be more related to the type of fuel burned and the type of control than to unit design.
Consequently, we determined that emissions information from units greater than 10
MMBtu/hr heat input could be used to establish MACT floor levels for this subcategory
because the fuels and controls are similar. The MACT floor emission level based on fabric
3-27
-------
filters is 0.068 Ib PM/MMBtu. The MACT floor emission level based on wet scrubbers is
0.0009 Ib HCl/MMBtu.
Limited Use Units—Capacity Utilizations Less than or Equal to 10 Percent. The
most effective control technologies identified for removing non-mercury metallic HAP used
by units in this subcategory are fabric filters. Therefore, the EPA considers fabric filters to
be the technology basis for the new source MACT floor for non-mercury metallic HAP
control in this subcategory. The most effective control technology identified for units in this
subcategory for removing acid gases, such as hydrogen chloride, are wet scrubbers.
Information in the emissions database or from other source categories does not show
that other control technologies, such as fabric filters or wet scrubbers, achieve reductions in
mercury emissions from liquid fuel-fired industrial, commercial, and institutional boilers and
process heaters. The EPA identified no control technology being used in the existing
population of boilers and process heaters that consistently achieved lower emission rates than
uncontrolled levels, such that a best-controlled similar source for mercury could be
identified. We concluded the MACT floor for new sources in this subcategory is no
emissions reductions for mercury.
We did find that monitoring CO can reduce organic HAP emissions and is used by at
least one unit in this subcategory, and have included it in the new source MACT floor as
described under section ELD of this preamble. Therefore, based on this analysis, the EPA
determined that the combination of a fabric filter, a wet scrubber, and CO monitoring forms
the control technology basis for the new source MACT floor for boilers and process heaters
in this subcategory.
The emissions test database did not contain test data for limited use liquid-fired
boilers and process heaters. In order to develop emission levels for this subcategory, we
decided to use information from units in the large liquid subcategory. Consequently, we
determined that emissions information from units greater than 10 MMBtu/hr heat input could
be used to establish MACT floor levels for this subcategory because the fuels and controls
are similar. The MACT floor emission level based on fabric filters is 0.068 Ib PM/MMBtu.
The MACT floor emission level based on wet scrubbers is 0.0009 Ib HCl/MMBtu. We are
requesting comment on using emission data from another subcategory to develop emission
levels for this subcategory. We are also requesting any available emissions information for
this subcategory.
3-28
-------
Gaseous Fuel Subcategories. No existing units were using control technologies that
achieve consistently lower emission rates than uncontrolled sources for any of the pollutant
groups of interest, except organic HAP. At least one unit in the population database in the
large and limited use gaseous fuel subcategories is required to monitor CO. Therefore, the
MACT floor for gaseous fuel-fired units includes a CO monitoring requirement and emission
limit, as described in section in.D of this preamble, but it does not include any emission
limits for PM, metallic HAP, mercury, or inorganic HAP based on the utilization of add-on
control technology.
How EPA Considered Beyond the Floor Options for New Units. The MACT floor
level of control for new units is based on the emission control that is achieved in practice by
the best controlled similar source within each of the subcategories. No technologies were
identified that would achieve non-mercury metals reduction greater than the new source
floors (i.e., fabric filters) for the liquid and solid subcategories or CO monitoring for the
solid, liquid, and gaseous subcategories. For inorganic HAP control, we determined that
packed bed scrubbers achieve higher emissions reductions than MACT floors consisting of a
wet scrubber. Packed bed scrubbers are the technology basis of the MACT floor for the
large unit subcategory, but wet scrubbers were the technology basis of the floors for the
small unit and limited unit subcategories. Therefore, we examined the cost and emission
reductions of applying a packed bed scrubber as a beyond the floor option for new solid and
liquid units within the small and limited use subcategories. We determined that costs were
excessive for the limited emission reduction that would be achieved. Non-air quality health,
environmental impacts, and energy effects were not significant factors, because there would
be little difference in the non-air quality health and environmental impacts between packed
bed scrubbers and wet scrubbers. Therefore, the EPA did not select this beyond-the-floor
option, and the proposed new source MACT level of control for PM, metallic HAP, and
inorganic HAP (HC1) is the same as the MACT floor level of control for all of the
subcategories.
In reviewing potential regulatory options beyond the new source MACT floor level of
control, the EPA identified one existing solid fuel-fired industrial boiler that was using
carbon injection technology for mercury control. However, emission data obtained from this
unit indicated that it was not achieving mercury emission reductions from the uncontrolled
levels. Moreover, we do not have information to otherwise show that carbon injection is
effective for reducing mercury emissions from industrial, commercial, and institutional
boilers and process heaters. Information in the emissions database or from other source
categories does not show that other control technologies, such as fabric filters or wet
3-29
-------
scrubbers, achieve reductions in mercury emissions from liquid fuel-fired industrial,
commercial, and institutional boilers and process heaters. Therefore, carbon injection, for
solid fuel units, and other control techniques, for liquid fuel units, were not evaluated as
regulatory options.
For the solid fuel and liquid fuel subcategories, fuel switching to natural gas is a
potential regulatory option beyond the new source floor level of control that would reduce
mercury and metallic HAP emissions. However, based on current trends within the industry,
the EPA projects that the majority of new boilers and process heaters will be built to fire
natural gas as opposed to solid and liquid fuels such that the overall emissions reductions
associated with this option would be minimal. Furthermore, organic HAP maybe increased
by fuel switching. Limited emissions reductions in combination with the high cost of fuel
switching and considerations about the availability and technical feasibility of fuel switching
makes this an unreasonable regulatory option that was not considered further. Non-air
quality health, environmental impacts, and energy effects were not significant factors. No
beyond-the-fioor options for gas-fired boilers were identified.
Based on the analysis discussed above, the EPA decided to not go beyond the MACT
floor level of control for new sources for MACT in the rule.
3.1.4 Considerations of Possible Risk-Based Alternatives to Reduce Impacts to Sources
The Agency has made every effort in developing this rule to minimize the cost to the
regulated community and allow maximum flexibility in compliance options consistent with
our statutory obligations. However, we recognize that the rule may still require some
facilities to take costly steps to further control emissions even though their emissions may
not result in exposures which could pose an excess individual lifetime cancer risk greater
than one in one million or which exceed thresholds determined to provide an ample margin
of safety for protecting public health and the environment from the effects of hazardous air
pollutants. We therefore solicited comment on whether there are further ways to structure
the rule to focus on the facilities which pose significant risks and avoid the imposition of
high costs on facilities that pose little risk to public health and the environment.
Representatives of the plywood and composite wood products industry provided EPA
with descriptions of three mechanisms that they believed could be used to implement more
cost-effective reductions in risk. The docket for today's rule contains "white papers"
prepared by industry that outline their proposed approaches (see docket number A-98-44,
Item # II-D-525). These approaches could be effective in focusing regulatory controls on
3-30
-------
facilities that pose significant risks and avoiding the imposition of high costs on facilities that
pose little risk to public health or the environment, and we sought public comment on the
utility of each of these approaches with respect to this rule.
One of the approaches, an applicability cutoff for threshold pollutants, would be
implemented under the authority of CAA section 112(d)(4); the second approach,
subcategorization and delisting, would be implemented under the authority of CAA sections
112(c)(l) and 112(c)(9); and, the third approach, would involve the use of a
concentration-based applicability threshold. We sought comments on whether these
approaches are legally justified and asked for information that could be used to support such
approaches.
The approach the Agency has chosen to include in the final rule is the first approach -
an applicability cutoff for threshold pollutants. The threshold pollutants for which an
applicability cutoff is applied are hydrochloric acid (Hcl) and a series of eight metals known
as the total selected metals (TSM).
3.1.4.1 Applicability Cutoffs for Threshold Pollutants Under Section 112(d)(4) of the CAA
This approach is an "applicability cutoff for threshold pollutants that is based on
EPA's authority under CAA section 112(d)(4). A "threshold pollutant" is one for which
there is a concentration or dose below which adverse effects are not expected to occur over a
lifetime of exposure. For such pollutants, section 112(d)(4) allows EPA to consider the
threshold level, with an ample margin of safety, when establishing emissions standards.
Specifically, section 112(d)(4) allows EPA to establish emission standards that are not based
upon the maximum achievable control technology (MACT) specified under section 112(d)(2)
for pollutants for which a health threshold has been established. Such standards may be less
stringent than MACT. Historically, EPA has interpreted 112(d)(4) to allow us to avoid
further regulation of categories of sources that emit only threshold pollutants, if those
emissions result in ambient levels that do not exceed the threshold, with an ample margin of
safety.!
In the past, EPA routinely treated carcinogens as non-threshold pollutants. The EPA
recognizes that advances in risk assessment science and policy may affect the way EPA
differentiates between threshold and non-threshold HAP. The EPA's draft Guidelines for
Carcinogen Risk Assessment (EPA, 1999) suggest that carcinogens be assigned non-linear
'See 63 FR 18754, 18765-66 (April 15, 1998) (Pulp and Paper Combustion Sources Proposed NESHAP).
3-31
-------
dose-response relationships where data warrant. Moreover, it is possible that dose-response
curves for some pollutants may reach zero risk at a dose greater than zero, creating a
threshold for carcinogenic effects. It is possible that future evaluations of the carcinogens
emitted by this source category would determine that one or more of the carcinogens in the
category is a threshold carcinogen or is a carcinogen that exhibits a non-linear dose-response
relationship but does not have a threshold.
The dose-response assessments for formaldehyde and acetaldehyde are currently
undergoing revision by the EPA. As part of this revision effort, EPA is evaluating
formaldehyde and acetaldehyde as potential non-linear carcinogens. The revised
dose-response assessments will be subject to review by the EPA Science Advisory Board,
followed by full consensus review, before adoption into the EPA Integrated Risk Information
System (IRIS). At this time, EPA estimates that the consensus review will be completed
sometime in 2004. The revision of the dose-response assessments could affect the potency
factors of these HAP, as well as their status as threshold or non-threshold pollutants. At this
time, the outcome is not known. In addition to the current reassessment by EPA, there have
been several reassessments of the toxicity of and carcinogenicity of formaldehyde in recent
years, including work by the World Health Organization and the Canadian Ministry of
Health.
3.1.4.2 Applicability Cutoffs for Hydrogen Chloride Controls Under Section 112(d)(4) of the
CAA
HCl Compliance Alternative. As an alternative to the requirement for each large
solid fuel-fired boiler to demonstrate compliance with the HCl emission limit in the final
rule, you may demonstrate compliance with a health-based facility-wide HCl equivalent
allowable emission limit.
The procedures for demonstrating eligibility for the HCl compliance alternative (as
outlined in appendix A of the final rule) are:
(1) You must include in your demonstration every emission point within the facility
that emits a respiratory toxicant included on EPA's list of hazardous air
pollutants.
(2) You must conduct HCl and chlorine emissions tests for every emission point
covered under subpart DDDDD.
(3) You must obtain either through emission testing or through the development and
documentation of best engineering estimates of maximum emissions of
3-32
-------
respiratory toxicants from all emission points at the facility not covered under
subpart DDDDD of part 63 from which a respiratory toxicant might reasonably be
emitted.
(4) You must determine the total maximum hourly mass HCl-equivalent emission
rate for your facility by summing the maximum hourly toxicity-weighted
emission rates of all appropriate respiratory toxicants (calculated using the
maximum rated capacities of the units) for each of the units at your facility.
(5) Use the look-up table in the appendix A of subpart DDDDD to determine if your
facility is in compliance with health-based HCl-equivalent emission limit.
(6) Select the maximum allowable HCl-equivalent emission rate from the look-up
table in appendix A of subpart DDDDD of part 63 for your facility using the
average stack height of your subpart DDDDD emission units as your stack height
and the minimum distance between any respiratory toxicant emission point at the
facility and the closest boundary of the nearest residential (or residentially zoned)
area as your fenceline distance.
(7) Your facility is in compliance if your maximum HCl-equivalent emission rate
does not exceed the value specified in the look-up table in appendix A of subpart
DDDDD.
(8) As an alternative to using the look-up table, you may conduct a site-specific
compliance demonstration (as outlined in appendix A of subpart DDDDD of part
63) which demonstrate that your facility cannot cause an individual chronic
inhalation exposure from respiratory toxicants which can exceed a Hazard Index
(HI) value of 1.0.
3.1.4.3 Applicability Cutoffs for Total Selected Metals Controls Under Section 112(d)(4) of
the CAA
In lieu of complying with the emission standard for TSM in subpart DDDDD of part
63 based on the sum of emissions for the eight selected metals (arsenic, cadmium, chromium,
mercury, manganese, nickel, lead, and ), you may demonstrate eligibility for complying with
the TSM standard based on excluding manganese emissions from the summation of TSM
emissions for the affected source unit.
The procedures for demonstrating eligibility for the TSM compliance alternative (as
outlined in appendix A of the subpart DDDDD) are:
(1) You must include in your demonstration every emission point within the facility
that emits a CNS toxicant included on EPA's list of hazardous air pollutants.
3-33
-------
(2) You must conduct manganese emissions tests for every emission point covered
under subpart DDDDD that emits manganese.
(3) You must obtain either through emission testing or through the development and
documentation of best engineering estimates of maximum emissions of CNS
toxicants from all emission points at the facility not covered under subpart
DDDDD from which a CNS toxicant might reasonably be emitted.
(4) You must determine the total maximum hourly manganese equivalent emission
rate from your facility by summing the maximum hourly toxicity-weighted
emission rates of all appropriate CNS toxicants (calculated using the maximum
rated heat input capacities) for each of the units at your facility.
(5) Use the look-up table in appendix A of subpart DDDDD to determine if your
facility is eligible for complying with the TSM limit based on the sum of
emissions for seven metals (excluding manganese) for the affected source units.
(6) Select the maximum allowable manganese-equivalent emission rate from the
look-up table in appendix A of subpart DDDDD for your facility using the
average stack height of your subpart DDDDD emission units as your stack height
and the minimum distance between any CNS toxicant emission point at the
facility and the closest boundary of the nearest residential (or residentially zoned)
area as your fenceline distance.
(7) Your facility is eligible if your maximum manganese-equivalent emission rate
does not exceed the value specified in the look-up table in appendix A of subpart
DDDDD.
(8) As an alternative to using look-up table to determine if your facility is eligible for
the TSM compliance alternative, you may conduct a site-specific compliance
demonstration (as outlined in appendix A of subpart DDDDD) which
demonstrates that your facility cannot cause an individual chronic inhalation
exposure from CNS toxicants which can exceed a HI value of 1.0.
If you elect to demonstrate eligibility for either of the health-based compliance
alternatives, you must submit certified documentation supporting compliance with the
procedures at least 1 year before the compliance date.
You must submit supporting documentation including documentation of all
maximum capacities, existing control devices used to reduce emissions, stack parameters,
and property boundary distances to each on-site source of HCl-equivalent and/or
manganese-equivalent emissions.
3-34
-------
You must keep records of the information used in developing the eligibility
demonstration for your affected source.
To be eligible for either health-based compliance alternative, the parameters that
defined your affected source as eligible for the health-based compliance alternatives
(including, but not limited to, fuel type, type of control devices, process parameters
documented as worst-case conditions during the emissions testing used for your eligibility
demonstration) must be incorporated as Federally enforceable limits into your title V permit.
If you do not meet these criteria, then your affected source is subject to the applicable
emission limits, operating limits, and work practice standards in Subpart DDDDD.
If you intend to change key parameters (including distance of stack to the property
boundary) that may result in lower allowable health-based emission limits, you must
recalculate the limits under the provisions of this section, and submit documentation
supporting the revised limits prior to initiating the change to the key parameter.
If you intend to install a new solid fuel-fired boiler or process heater or change any
existing emissions controls that may result in increasing HCl-equivalent and/or
manganese-equivalent emissions, you must recalculate the total maximum hourly
HCl-equivalent and/or manganese-equivalent emission rate from your affected source, and
submit certified documentation supporting continued eligibility under the revised
information prior to initiating the new installation or change to the emissions controls.
Facilities that could not demonstrate that they are eligible to be included in the
low-risk subcategory would be subject to MACT and possible future residual risk standards.
3.2 Profile of Existing Boiler and Process Heaters Units
This section profiles existing boilers and process heaters, collectively referred to as
"units," with respect to business applications, industry of parent company, and fuel use. The
unit population database in combination with the model units that helped in preparing that
database were used to determine which types of boilers, fuel, and control devices were in the
existing unit population so that corresponding emission factors could be developed for all
combinations. The development of the population database and the model units are
discussed in the memoranda, "Development of the Population Database for the Industrial,
Commercial, and Institutional Boiler and Process Heater National Emission Standard for
Hazardous Air Pollutants (NESHAP)" and "Development of the Model Units for the
Industrial, Commercial, and Institutional Boiler and Process Heater National Emission
Standard for Hazardous Air Pollutants (NESHAP)." The units contained in the Inventory
3-35
-------
Database are based on information from the Aerometric Information Retrieval System
(AIRS) and Ozone Transport Assessment Group (OTAG) databases, state and local permit
records, and the combustion source Information Collection Request (ICR) conducted by the
Agency in 1997. The list of units contained in the Inventory Database was reviewed and
updated by industry and environmental stakeholders as part of the Industrial Combustion
Coordinated Rulemaking (ICCR), chartered under the Federal Advisory Committee Act
(FACA).
The entire Inventory Database contains more than 58,000 ICI boilers and process
heaters; however, only about 4,000 are estimated to be affected by the final rule. Of these
existing units, a little over half had sufficient information on operating parameters to be
included in the floor-level El A. The number of potentially affected units included in the
profile for the final rule was 2,186.
3.2.1 Distribution of Existing Boilers and Facilities by Industry
Table 3-2 presents the number of existing boilers and process heaters and the number
of facilities owning units by two-digit SIC code and three-digit NAICS code that may be
affected by the final rule. For the final rule, the industries with the largest number of
potentially affected units are the furniture, paper, lumber, and electrical services industries.
These four industries alone account for nearly 60 percent of affected units. Almost all the
process heaters are in the lumber industry. (Section 4 presents industry profiles for the
lumber and wood products, electrical services, and paper industries, among others.) The
remaining units are primarily distributed across the manufacturing sector and service
industries.
3.2.2 Technical Characteristics of Existing Boilers
Figure 3-1 characterizes the population of 2,186 units identified in the Inventory
Database by capacity range, fuel type, and level of preexisting control.
3.2.2.1 Final Rule
• Capacity Range: Unit input capacities in the population are expressed in four
ranges: 0-10, 10-100, 100-250, and >250 MMBtu/hr. Fifty-two percent of the
units affected for this alternative have capacities between 10 and 100 MMBtu/hr.
The two largest capacity ranges each contain approximately one quarter of the
population. Only 1 percent of units have input capacities less than 10 MMBtu/hr.
• Fuel Type: About half of these units consume coal as their primary fuel (1,074
units). After coal, the next most common fuel type is wood (479 units).
3-36
-------
Table 3-2. Units and Facilities Affected by the Final Rule by Industry"
SIC
Code
01
02
07
10
12
13
14
17
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
42
46
NAICS
Code
111
112
115
212
212
211
212
235
311
312
313
315
321
337
322
511
325
324
326
316
327
331
332
333
335
336
334
339
482
484
486
Description
Agriculture — Crop s
Agriculture — Livestock
Agricultural Services
Metal Mining
Coal Mining
Oil and Gas Extraction
Mining/Quarrying — Nonmetallic Minerals
Construction — Special Trade Contractors
Food and Kindred Products
Tobacco Products
Textile Mill Products
Apparel and Other Products from Fabrics
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Printing, Publishing, and Related Industries
Chemicals and Allied Products
Petroleum Refining and Related Industries
Rubber and Miscellaneous Plastics Products
Leather and Leather Products
Stone, Clay, Glass, and Concrete Products
Primary Metal Industries
Fabricated Metal Products
Industrial Machinery and Computer
Equipment
Electronic and Electrical Equipment
Transportation Equipment
Scientific, Optical, and Photographic Equip.
Miscellaneous Manufacturing Industries
Railroad Transportation
Motor Freight and Warehousing
Pipelines, Except Natural Gas
Boilers
3
0
0
9
2
0
8
0
138
11
135
2
335
234
321
0
171
11
17
1
9
41
16
23
5
102
8
2
4
5
0
Heaters
0
0
0
0
0
0
0
0
0
0
0
0
25
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Total
Units
3
0
0
9
2
0
8
0
138
11
135
2
360
234
321
0
174
11
17
1
9
41
16
23
5
102
8
2
4
5
0
Facilities
3
0
0
4
1
0
4
0
60
7
71
2
262
154
194
0
70
8
13
1
7
16
10
12
5
41
4
2
1
1
0
(continued)
3-37
-------
Table 3-2. Units and Facilities Affected by the Final Rule by Industry" (continued)
SIC
Code
49
50
51
55
58
60
59
70
72
76
80
81
82
83
86
87
89
91
92
94
96
97
NA
NAICS
Code
221
421
422
441
722
522
445-454
721
812
811
621
541
611
624
813
541
711/514
921
922
923
926
928
Description
Electric, Gas, and Sanitary Services
Wholesale Trade — Durable Goods
Wholesale Trade — Nondurable Goods
Automotive Dealers and Gasoline Service
Stations
Eating and Drinking Places
Depository Institutions
Miscellaneous Retail
Hotels and Other Lodging Places
Personal Services
Miscellaneous Repair Services
Health Services
Legal Services
Educational Services
Social Services
Membership Organizations
Engineering, Accounting, Research,
Management and Related Services
Services, N.E.C.
Executive, Legislative, and General
Administration
Justice, Public Order, and Safety
Administration of Human Resources
Administration of Economic Programs
National Security and International Affairs
SIC Information Not Available
Boilers
318
3
2
0
0
0
0
1
0
2
37
0
105
2
0
2
2
1
29
1
4
29
7
2,158
Heaters
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
28
Total
Units
318
3
2
0
0
0
0
1
0
2
37
0
105
2
0
2
2
1
29
1
4
29
7
2,186
Facilities
160
2
1
0
0
0
0
1
0
1
18
0
45
1
0
2
1
1
9
1
3
11
4
1,214
a Based on the Inventory Database.
3-38
-------
Floor Alternative (n=2,1S6)
OtolO
1%
100 to 250
25%
Fabric Filter
16%
10to100
52%
Input Capacity
(million Btu)
No Control
13%
Cyclone
^ 36%
Preexisting Control
Technology
Other
Bio mass
14%
Other
4%
Wood
22%
Bagasse
4%
Coal
49%
Coal and
Wood
7%
Fuel Type
Figure 3-1. Characteristics of Units Affected
• Control Level: Eighty-three percent of units have some type of control device
already installed; 289 do not. Typical control devices include fabric filters, wet
scrubbers, and electrostatic precipitators.
3.3 Methodology for Estimating Cost Impacts
The predominant type of control measure that is considered in the analysis of
emission reductions needed for sources to achieve the MACT floor is add-on control
technologies. Add-on control techniques are those technologies that are applied to the vent
gas stream of the boiler or process heater to reduce emissions. The boiler and process
heaters population database includes information on all control techniques that are applied to
industrial, commercial, institutional boilers and process heaters. Generally, they can be
grouped into PM control or acid gas control. The most common technologies, and the ones
analyzed for the impacts analysis, include fabric filters, ESP's, packed scrubbers, venturi
scrubbers, and spray dryers. In addition, when add-on technologies are used, the cost of
ductwork and associated equipment also needed to be considered.
Components of capital cost include
• purchased equipment cost of the primary device and auxiliary equipment,
• instrumentation,
3-39
-------
• sales tax and freight, and
• installation costs. Installation costs include foundations and support, handling
and erection, electrical, piping, insulation, and painting, engineering, construction
and field expenses, contractor fees, start-up, performance tests, and contingencies.
Components of annual cost include
• raw materials,
• utilities (electricity, fuel, steam, air, water),
• waste treatment and disposal,
• labor (operating, supervisory, maintenance),
• maintenance materials,
• replacement parts,
• overhead,
• property taxes,
• insurance,
• administration charges, and
• capital recovery costs.
For this analysis, costs were estimated in 1999 dollars. Capital recovery was calculated
assuming 7 percent interest rate over the life of the equipment. The use of this interest rate is
based on Office of Management and Budget (OMB) guidance (Circular A-94, October 29,
1992).
The algorithms used to estimate these costs were obtained from previous EPA
studies. These cost algorithms are included as appendices to the cost methodology
memorandum in the public docket. Inputs for the algorithms used in the impacts analysis are
also presented in this memorandum.
Fabric Filter
The algorithms used to estimate capital and annual costs of fabric filters were
obtained from EPA's EPA Air Pollution Control Cost Manual. Algorithms were provided
for 4 types of fabric filters: shaker, reversed air, pulse-jet modular, and pulse-jet common.
3-40
-------
The cost algorithms for estimating capital costs reduced to basic equations for each are
provided in Appendix A-l of the cost methodology memorandum (henceforth called the
"cost memo"). Capital costs are based on the gross cloth area of the fabric filter, which is a
function of the gas inlet flow rate. Algorithms for calculating annual costs are provided in
Appendix A-2 of the cost memo. Annual costs include dust disposal, electricity,
maintenance, labor, bag replacement, maintenance labor, compressed air, overhead,
administrative, property taxes, and insurance. Capital recovery is annualized over 20 years
at 7 percent interest. Appendix A-3 of the cost memo presents the values for the inputs used
in this analysis and the reasons for their use.
Electrostatic Precipitator
The algorithms used to estimate capital and annual costs of ESPs were obtained from
EPA's Air Pollution Control Cost Manual. Capital costs are based on the total collection
plate area, which is calculated from the gas inlet flow rate and the required removal
efficiency. The cost algorithms for estimating capital costs of ESPs reduced to basic
equations are provided in Appendix B-l of the cost memo. Algorithms for calculating
annual costs are provided in Appendix B-2 of the cost memo. Annual costs include dust
disposal, electricity, maintenance, labor, maintenance labor, overhead, administrative,
property taxes, and insurance. Capital recovery is annualized at 7 percent interest.
Appendix B-3 of the cost memo presents the values for the inputs used in this analysis and
the reasons for their use.
Venturi Scrubber
The algorithms used to estimate capital and annual costs of venturi scrubbers were
obtained from EPA cost algorithms on EPA's website( http://www.epa.gov/ttn/catc/
products.html#cccinfo.) Capital costs include not only the cost of the venturi scrubber but
also a pump to provide motive force for the solvent. Capital costs are based on the gas flow
rate and saturation temperature of the gas-solvent. The cost algorithms for estimating capital
costs of each piece of equipment were reduced to basic equations in Appendix C-l of the
cost memo. The cost algorithms for estimating annual costs were reduced to basic equations
in Appendix C-2 of the same memorandum. Annual costs include wastewater disposal,
solvent, electricity, maintenance, labor, maintenance labor overhead, administrative, property
taxes, and insurance. Capital recovery is an annualized cost estimated using a 7 percent
interest rate. Appendix C-3 of the cost memo presents the values for the inputs used in this
analysis and the reasons for their use.
3-41
-------
Packed Bed Scrubber
The algorithms used to estimate capital and annual costs of packed bed scrubbers
were obtained from EPA's Air Pollution Control Cost Manual. The capital costs are
comprised of the scrubber tower, packing, pumps, and fans. Capital costs are based
primarily on gas flow rate and removal efficiency. The cost algorithms for estimating capital
costs of packed scrubber equipment reduced to their basic equations for each are provided in
Appendix D-l of the cost memo. The cost algorithms for estimating annual costs of packed
scrubbers are provided in Appendix D-2 of the cost memo. Annual costs include caustic,
wastewater disposal, water, electricity, maintenance, labor, overhead, administrative,
property taxes, and insurance. Capital recovery is an annualized cost estimated using a 7
percent interest rate. Appendix D-3 of the cost memo presents the values for the inputs used
in this analysis and the reasons for their use.
Spray Dryer
The algorithms used to estimate capital and annual costs of spray dryers were
obtained from previous EPA studies. Capital costs include the cost of the spray dryer and
pumps. Capital costs are based on the gas flow rate. The cost algorithms for estimating
capital costs of spray dryer equipment reduced to basic equations are provided in
Appendix E-l of the cost memo. The cost algorithms for estimating annual costs for spray
dryers are provided in Appendix E-2 of the cost memo. Annual costs include lime, water,
electricity, maintenance, labor, maintenance labor, overhead, administrative, property taxes,
and insurance. Capital recovery is an annualized cost estimated using a 7 percent interest
rate. Appendix E-3 of the cost memo presents the values for the inputs used in this analysis
and the reasons for their use.
Ductwork
The algorithms used to estimate capital and annual costs of ductwork were obtained
from EPA's Air Pollution Control Cost Manual. Capital costs include 500 feet of ductwork,
elbows, and fans. The 500 feet of ductwork was based on engineering judgement and
previous experience on the distance between emission points and control devices in chemical
facilities and the availability of space for retrofitting controls. Costs are based on ductwork
diameter, which is calculated from the gas flow rate. The cost algorithms for estimating
capital costs and annual costs reduced to basic equations are provided in Appendix F-l of the
cost memo. Annual costs include electricity, maintenance, maintenance labor, overhead,
administrative, property taxes, and insurance. Capital recovery is an annualized cost
3-42
-------
estimated using a 7 percent interest rate. Required inputs to the ductwork algorithms are
provided in the input tables provided in Appendices A-3, B-3, C-3, D-3, and E-3 of the cost
memo.
Good Combustion Practices
Few sources in the population database specifically reported using good combustion
practices. Boilers and process heaters within each subcategory might use any of a wide
variety of different work practices, depending on the characteristics of the individual unit.
Consequently, any uniform requirements or set of work practices that would
meaningfully reflect the use of good combustion practices, or that could be meaningfully
implemented across any subcategory of boilers and process heaters could not be identified.
Additionally, few of the GCP's have been documented to reduce organic HAP
emissions, and they could not be considered in the MACT analysis. One GCP that may
effect organic HAP emissions is maintaining CO emission levels. CO is generally an
indicator of incomplete combustion because CO will burn to carbon dioxide if adequate
oxygen is available. Controlling CO emissions is a mechanism for ensuring combustion
efficiency, and therefore may be viewed as a kind of GCP.
Capital and annual costs for CO monitoring is presented in Appendix G of the cost
memo. The costing information was obtained from a previous EPA study. Capital costs are
comprised of the initial cost of the equipment. Annual costs include operating and
maintenance costs, annual and quarterly checks, recordkeeping and reporting, taxes,
insurance, and administrative costs. Annualized costs such as capital recovery costs are
calculated assuming an equipment life of 20 years and an interest rate of 7 percent.
Testing and Monitoring Costs
The final rule includes emission limits for HC1, PM, metallic HAP, and mercury.
Additionally, as mentioned in Section 1 of this EIA and the preamble, the rule allows sources
to meet requirements by monitoring fuel content instead of emissions. Consequently, testing
and monitoring costs of meeting the standards were incorporated into the cost estimates.
Capital costs for testing include initial stack tests for PM, HC1, and metals for fossil fuels,
and materials and fuel analysis for biomass. Capital cost components include operation and
maintenance costs and capital recovery assuming the initial capital investment is annualized
over a 5 year period at 7 percent interest. Monitoring costs are included for opacity
3-43
-------
monitoring, HC1 monitoring, and scrubber parametric monitoring.2 Monitoring costs include
the capital cost of monitoring equipment, and the annual costs of capital recovery assuming
the initial capital investment is annualized over a 20 year period at 7 percent interest. Annual
monitoring costs also include operation and maintenance as well as other additional costs.
The testing and monitoring costs are shown in Table 3-3. Appendix G of the cost memo
includes further details on these costs. Information used to estimate testing and monitoring
costs were obtained from previous EPA studies.
Costs to Control Non-Air Effects Related to Rule Implementation
The EPA estimated the additional water usage that would result from the MACT
floor level of control to be 110 million gallons per year for existing sources and 0.6 million
gallons per year for new sources. In addition to the increased water usage, an additional 3.7
million gallons per year of wastewater would be produced for existing sources and 0.6
million gallons per year for new sources. The EPA estimated the additional solid waste that
would result from the MACT floor level of control to be 102,000 tons per year for existing
sources and 1 ton per year for new sources. The costs ($900,000) of handling the additional
solid waste generated from applying MACT floor technology are accounted for in the control
cost estimates for ESP and fabric filter applications. The costs ($20,000) of treating
wastewater from venturi and packed bed scrubber are also accounted for in the control cost
estimates.
Cost Effectiveness
To provide additional information on the magnitude of the cost estimates, Table 3-4
shows the cost-effectiveness (cost/ton reduced estimates) for the HAP and non-HAP
pollutants whose emissions are reduced by this rule.
2The monitoring costs reported for existing units are not the cost of continuous emission monitors
(CEM), but the costs associated with monitoring the process parameters of the control device.
Installation of these process monitors are integral to the control device and would be installed with
or without the monitoring requirements of the MACT. Therefore, even though we present these
monitoring costs separately, they are included in the overall reported control costs and should not
be considered as an additional cost for emission monitoring.
3-44
-------
Table 3-3. Testing and Monitoring Costs for Units Covered
Material or Fuel
Regular Use Units
No, of. . . No. of
Industrial ilu- "
Process
Total Capital
Investment of
Testing and
Monitoring ($)
Total
Annual
Costs of
Testing ($)
Total Annual
Costs of.
Monitoring
($)
Annual Capital
Recovery -Testing
and Monitoring
(1999$)
Total Annual
Costs of Testing
and Monitoring
(1999$)
123,436,995
Liquid/NFF Solid
3,747,240
Biomass/Liquid FF
5,237,834
0
Solid/Gas
Biomass/NFF
Liquid/NFF Solid
6,005,266
0
11,660,854
5,004,638
Units
3,381,728
158,114,555
0
Gas/Wood/Other
-------
Table 3-3. Testing and Monitoring Costs for Units Covered (continued)
.u
Material or Fuel
Limited Use Units
Liquid/NFF Solid
Biomass/Liquid FF
Solid/Gas
Biomass/NFF
Liquid/NFF Solid
Units
Coal/Wood/NFF
Total Capital Total Annual ^^ c Total Annual
Investment of Total Recovery -Testing Costs of Testing
fn°du°sftrial No- of Annual l&JffltS^no , ™ • • and Monitoring
Process Testing and Costs of Monitoring an(j Monitoring
„ ., „ ^ Monitoring ($) Testing ($)
($) (1999$) (19"$)
3,330,416
61,772
169,020
0
$9,696
555,200
97,620
4,263,724
-------
Table 3-4. Cost Effectiveness (C/E) of Industrial Boiler and Process Heater MACT on
Existing Units and Subcategories
Control Costs
($)
PM Emissions
Reduction
(Tons/Year)
C/E ($/tonPM)
Metals
Emissions
Reduction
(Tons/Year)
C/E ($/ton
metals)
HC1 Emissions
Reduction
(Tons/Year)
C/E ($/ton HC1)
HAP Emissions
Reduction
(Tons/Year)
C/E ($/ton
HAP)
Total Large Solid
Annualized Fuel
Costs Subcategory
833,273,781" 810,422,230
565,900 563,060
1,472" 1,439
1,093 1,087
762,373a 745,558a
46,515 46,515
17,914" 17,422a
47,608 47,602
17,502 17,025
Large Solid
Fuel
Subcategory —
Coal Only
669,353,690
359,920
1,860
591
l,132,578a
45,136
14,830a
45,727
14,638
Large Solid
Fuel
Subcategory —
Wood Only
141,068,540
203,140
694
496
284,412"
1,379
102,298"
1,875
75,236
Limited Use
Solid Fuel
Subcategory
22,851,551
2,840
8,046
6
3,808,592"
—
—
6
3,808,500
" The cost-effectiveness value is based on the total annualized cost of the rule and not on the cost for
controlling the specific pollutant, and, thus, overstates the cost/ton for the specific HAP or other pollutant.
b Costs are in 1999 dollars. Emission reductions are calculated for 2005.
3-47
-------
Cost Uncertainties
The primary limitation to the cost estimates developed for the rule is that costs were
calculated for model units rather than each individual boiler or process heater.
Consequently, the costs do not characterize any "real" unit. This was done for practical
reasons. Because there are over 60,000 units in the U.S., it would not be possible to gather
unit-specific information for each unit necessary for estimating costs, such as flue gas
temperatures and flow rates. Additionally, emission information was only available for less
than 1 percent of the units. In order to estimate costs and emission reductions of the rule,
model units were developed to represent the population of boilers and process heaters in the
U.S. While sufficient information was not available for characterizing each unit, sufficient
emissions and process information were available to develop model units. Each unit in the
U.S. was then assigned to a model based on their size and fuel burned. It also should be
noted that the costing methodology is the cost algorithms for the control devices provide a
cost range of+/- 30 percent. This aspect of the costing methodology reflects the degree of
variability typically found in study-level cost estimates. This is also the degree of variability
found in the cost methodology employed in the EPA Air Pollution Control Cost Manual,
which is an important reference for the cost estimates supplied in the RIA. Cost information
available to owners and operators of boilers and process heaters will be more specific and
accurate. Consequently, the cost estimates may overestimate or underestimate costs.
3.4 Projection of New Boilers and Process Heaters
Energy Information Administration fuel consumption forecasts were used in
conjunction with existing model boiler population data to project the number and type of
new boilers to be installed by 2005. EPA used the following steps to calculate new boiler
population estimates:
1. Calculate the percentage change in industrial fuel consumption. Energy
Information Administration data were used to obtain industrial and commercial
fuel use projections. The percentage change in consumption (1998 to 2005) in the
industrial and commercial sectors was calculated for the following fuel categories
using 1998 as the base year (the same year that the model boiler algorithms are
based on): steam coal (2.6%), natural gas (6.3%), residual fuel oil (-7.4%),
distillate fuel oil (12.0%), and biomass (11.5%). It should be noted that 1998 was
a year of below average energy prices, and that current and potential future
energy prices are higher than the historical average. If real fuel prices increase
faster than the EIA's projections, then conservation measures may lead to fewer
projected boilers and process heaters. This trend would lead to an overestimate
(upward bias) of the impact estimates presented in this report.
3-48
-------
2. Estimate the number of new boilers by model number-fuel type. To predict the
number of new boilers in operation by 2005, EPA applied the percentage
difference for each fuel category to the 1998 fuel consumption of boilers
represented by the boiler models to calculate total energy consumed by boilers in
2005 for each model number. The number of new boilers per model was
calculated by dividing the model fuel forecasts by the annual fuel consumption of
one unit and then subtracting the number of units present in 1998, as follows:
Number of _ | Total energy consumed (2005) [MMBtu/yr] j _ Number of
New Units " [ Avg capacity [MMBtu/hr] x 8,760 [hr/yr] J ~ Units
Following these steps, EPA projects that 1,458 boilers and 374 process heaters to be
installed between 1998 and 2005 will be affected by the new source MACT floor. The only
new ICI boilers and process heaters that will be unaffected are those natural gas and distillate
fuel units that have input capacities less than 10 MMBtu/hr. These projections were
developed by model unit type, not by industry. To assess the distribution of the boilers and
process heaters estimated to be operating in 2005 across industries, EPA attached unit-level
weights by model number to each unit in the Inventory Database. These weights allow each
unit in the Inventory Database to represent a number (or fraction) of units that are predicted
to be in use by the end of 2005. The weights were then summed by two-digit SIC code to
estimate the distribution of units by industry.
Table 3-5 presents the projected number of new boilers and process heaters for the
MACT floor. Industries with the estimated greatest concentrations of new units include
chemicals and allied products (295), petroleum refining (198), electric services (134), and
paper and allied products (96).
3.5 National Engineering Population, Cost Estimates, and Cost-Effectiveness
Estimates
The Agency estimates that in 2005, 5,562 units (existing units and new units) may be
affected by the final rule. This population was used to estimate national engineering costs.
The population estimates were determined by unit configuration, not by industry. Thus, the
distribution of units by industry shown in Tables 3-5 and 3-6 was determined by weighting
3-49
-------
Table 3-5. New Unit Projections by Industry, MACT Floor
SIC
Code
01
02
07
10
12
13
14
17
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
42
NAICS
Code
111
112
115
212
212
211
212
235
311
312
313
315
321
337
322
511
325
324
326
316
327
331
332
333
335
336
334
339
482
484
Description
Agriculture — Crop s
Agriculture — Livestock
Agricultural Services
Metal Mining
Coal Mining
Oil and Gas Extraction
M ining/Q uarrying — No nme tallic M inerals
Construction — Special Trade Contractors
Food and Kindred Products
Tobacco Products
Textile Mill Products
Apparel and Other Products from Fabrics
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Printing, Publishing, and Related Industries
Chemicals and Allied Products
Petroleum Refining and Related Industries
Rubber and Miscellaneous Plastics Products
Leather and Leather Products
Stone, Clay, Glass, and Concrete Products
Primary Metal Industries
Fabricated Metal Products
Industrial Machinery and Computer Equipment
Electronic and Electrical Equipment
Transportation Equipment
Scientific, Optical, and Photographic Equipment
Miscellaneous Manufacturing Industries
Railroad Transportation
Motor Freight and Warehousing
New Units
—
—
—
6
1
89
6
—
63
7
73
—
61
47
96
19
295
198
44
5
37
80
53
35
40
80
11
9
—
1
Cost
—
—
—
$47,040
$7,840
$697,760
$87,740
—
$801,836
$54,880
$1,329,391
—
$1,748,655
$1,354,701
$1,526,704
$148,960
$3,793,738
$1,552,320
$385,660
$39,200
$549,975
$2,873,492
$496,920
$396,500
$313,600
$1,133,423
$86,240
$162,323
—
$48,540
(continued)
3-50
-------
Table 3-5. New Unit Projections by Industry, MACT Floor (continued)
SIC
Code
46
49
50
51
55
58
59
60
70
72
76
80
81
82
83
86
87
89
91
92
94
96
97
NA
State
NAICS
Code
486
221
421
422
441
722
445-454
522
721
812
811
621
541
611
624
813
541
711/514
921
922
923
926
928
Description New Units
Pipelines, Except Natural Gas 1
Electric, Gas, and Sanitary Services 134
Wholesale Trade — Durable Goods —
Wholesale Trade — Nondurable Goods —
Automotive Dealers and Gasoline Service Stations —
Eating and Drinking Places —
Miscellaneous Retail —
Depository Institutions —
Hotels and Other Lodging Places —
Personal Services 1
Miscellaneous Repair Services —
Health Services 6
Legal Services —
Educational Services 19
Social Services —
Membership Organizations —
Engineering, Accounting, Research, Management and 2
Related Services
Services, N.E.C. —
Executive, Legislative, and General Administration —
Justice, Public Order, and Safety 4
Administration of Human Resources —
Administration of Economic Programs —
National Security and International Affairs 2
SIC Information Not Available 307
Parent is a State Government —
1,832
Cost
$7,840
$2,094,546
—
—
—
—
—
—
—
$7,840
—
$209,840
—
$815,855
—
—
$388,350
—
—
$153,460
—
—
$97,080
$2,497,327
—
$25,909,574
3-51
-------
Table 3-6. Unit Cost and Population Estimates for the Final Rule by Industry, 2005
SIC
Code
01
02
07
10
12
13
14
17
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
42
NAICS
Code
111
112
115
212
212
211
212
235
311
312
313
315
321
337
322
511
325
324
326
316
327
331
332
333
335
336
334
339
482
484
Description
Agriculture — Crop s
Agriculture — Livestock
Agricultural Services
Metal Mining
Coal Mining
Oil and Gas Extraction
Mining/Quarrying — Nonmetallic Minerals
Construction — Special Trade Contractors
Food and Kindred Products
Tobacco Products
Textile Mill Products
Apparel and Other Products from Fabrics
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Printing, Publishing, and Related Industries
Chemicals and Allied Products
Petroleum Refining and Related Industries
Rubber and Miscellaneous Plastics Products
Leather and Leather Products
Stone, Clay, Glass, and Concrete Products
Primary Metal Industries
Fabricated Metal Products
Industrial Machinery and Computer
Equipment
Electronic and Electrical Equipment
Transportation Equipment
Scientific, Optical, and Photographic
Equipment
Miscellaneous Manufacturing Industries
Railroad Transportation
Motor Freight and Warehousing
Total
Floor
Units
5
—
—
27
6
89
25
—
312
28
360
4
483
311
565
19
644
217
73
7
57
159
87
84
52
300
26
12
9
12
Units
Percent
0.08%
0.00%
0.00%
0.48%
0.10%
1.60%
0.46%
0.00%
5.60%
0.51%
6.47%
0.08%
8.68%
5.59%
10.15%
0.34%
11.58%
3.91%
1.32%
0.13%
1.02%
2.85%
1.56%
1.51%
0.93%
5.39%
0.46%
0.22%
0.16%
0.22%
Total Cost
Floor Costs
(by Unit)
$628,943
—
—
$6,651,678
$683,026
$697,760
$8,253,479
—
$37,774,020
$6,014,216
$74,152,804
$679,510
$48,896,055
$29,632,880
$123,008,263
$148,960
$116,236,183
$4,620,563
$6,356,835
$607,530
$6,253,678
$27,110,619
$10,042,680
$11,208,392
$3,744,828
$55,440,341
$3,511,206
$826,346
$1,251,062
$2,128,148
Percent
0.07%
0.00%
0.00%
0.77%
0.08%
0.08%
0.96%
0.00%
4.38%
0.70%
8.59%
0.08%
5.67%
3.43%
14.25%
0.02%
13.47%
0.54%
0.74%
0.07%
0.72%
3.14%
1.16%
1.30%
0.43%
6.42%
0.41%
0.10%
0.14%
0.25%
(continued)
3-52
-------
Table 3-6. Unit Cost and Population Estimates for the Final Rule by Industry, 2005
(continued)
SIC
Code
46
49
50
51
55
58
59
60
70
72
76
80
81
82
83
86
87
89
91
92
94
96
97
NA
State
NAICS
Code
486
221
421
422
441
722
445-454
522
721
812
811
621
541
611
624
813
541
711/514
921
922
923
926
928
Description
Pipelines, Except Natural Gas
Electric, Gas, and Sanitary Services
Wholesale Trade — Durable Goods
Wholesale Trade — Nondurable Goods
Automotive Dealers and Gasoline Service
Stations
Eating and Drinking Places
Miscellaneous Retail
Depository Institutions
Hotels and Other Lodging Places
Personal Services
Miscellaneous Repair Services
Health Services
Legal Services
Educational Services
Social Services
Membership Organizations
Engineering, Accounting, Research,
Management and Related Services
Services, N.E.C.
Executive, Legislative, and General
Administration
Justice, Public Order, and Safety
Administration of Human Resources
Administration of Economic Programs
National Security and International Affairs
SIC Information Not Available
Parent is a state government
Total
Floor
Units
1
718
6
4
—
—
—
—
2
1
4
86
—
251
5
—
38
2
2
69
2
8
64
326
—
Units
Percent
0.02%
12.91%
0.12%
0.07%
0.00%
0.00%
0.00%
0.00%
0.04%
0.02%
0.08%
1.55%
0.00%
4.52%
0.08%
0.00%
0.68%
0.04%
0.04%
1.23%
0.04%
0.15%
1.16%
5.86%
0.00%
5,562
Total Cost
Floor Costs
(by Unit) Percent
$7,840
$150,341,645
$2,154,760
$1,673,511
—
—
—
—
$567,811
$7,840
$625,531
$15,172,212
—
$60,490,956
$820,191
—
$2,240,544
$918,360
$312,765
$13,707,649
$314,316
$2,300,308
$18,018,010
$6,747,652
—
0.00%
17.42%
0.25%
0.19%
0.00%
0.00%
0.00%
0.00%
0.07%
0.00%
0.07%
1.76%
0.00%
7.01%
0.10%
0.00%
0.26%
0.11%
0.04%
1.59%
0.04%
0.27%
2.09%
0.78%
0.00%
$862,981,906
3-53
-------
existing units by the estimates by unit configuration and tallying weighted units by SIC code.
The average cost of control by unit configuration was multiplied by the weighted number of
units to determine industry-level control cost estimates.
Table 3-6 presents industry-level population and cost estimates for boilers and
process heaters. The distribution of weighted units across industries mirrors that of the
analysis population even though it was determined by weighting units by configuration, not
industry-level growth estimates. The floor cost of control for the estimated 5,562 boilers and
process heaters is $863.0 million, with an average per-unit additional control cost of
$155,157.
3-54
-------
SECTION 4
PROFILES OF AFFECTED INDUSTRIES
This section contains profiles of the major industries affected by the MACT for
boilers and process heaters. Included are profiles of the following industries:
• Textile Mill Products (SIC 22/NAICS 313)
• Lumber and Wood Products (SIC 24/NAICS 321)
• Furniture and Related Product Manufacturing (SIC 25/NAICS 337)
• Paper and Allied Products (SIC 26/NAICS 322)
• Medicinal Chemicals and Botanical Products and Pharmaceutical Preparations
(SICs 2833, 2834/NAICS 32451)
• Industrial Organic Chemicals (SIC 2869/NAICS 3251)
• Electric Services (SIC 4911/NAICS 22111)
4.1 Textile Mill Products (SIC 22/NAICS 313)
The textile industry is one of the few industries found throughout the world, from the
most industrialized countries to the poorest. This industry includes firms producing the
following products: broadwoven fabric; weft, lace, and warp knit fabrics; carpets and rugs;
spun yarn products; and man-made fibers. The United States has typically run a trade deficit
in the textiles sector in recent years, importing about $1.3 billion more than was exported in
1995. Although trade has become an increasingly important part of this industry, trade in
this segment is relatively small compared with trade in the downstream apparel segment. In
1996, the total value of shipments for the textile industry was $80,242 million.
4.2 Lumber and Wood Products (SIC 24/NAICS 321)
The lumber and wood products industry comprises a large number of establishments
engaged in logging; operating sawmills and planing mills; and manufacturing structural
wood panels, wooden containers, and other wood products. Table 4-1 lists the lumber and
wood products markets that are likely to be affected by the regulation on boilers. Most
4-1
-------
Table 4-1. Lumber and Wood Products Markets Likely to Be Affected by the
Regulation
SIC
2421
2434
2449
2491
2493
2499
NAICS
321113
33711
32192
32114
321219
321999
Description
Sawmills and Planing Mills, General
Wood Kitchen Cabinets
Wood Containers, N.E.C.
Wood Preserving
Reconstituted Wood Products
Wood Products, N.E.C.
Source: Industrial Combustion Coordinated Rulemaking (ICCR). 1998. Data/Information Submitted to the
Coordinating Committee at the Final Meeting of the Industrial Combustion Coordinated Rulemaking
Federal Advisory Committee. EPA Docket Numbers A-94-63, II-K-4b2 through -4b5. Research
Triangle Park, North Carolina. September 16-17.
products are produced for the domestic market, but exports increasingly account for a larger
proportion of sales (Haltmaier, 1998). The largest consumers of lumber and wood products
are the remodeling and construction industries.
In 1996, the lumber and wood products industry's total value of shipments was
$85,724.0 million. As seen in Table 4-2, shipment values increased steadily through the late
1980s before declining slightly through the early 1990s as new construction starts and
furniture purchases declined (Haltmaier, 1998). Shipment values recovered, however, as the
economy expanded in the mid-1990s.
4.2.1 Supply Side of the Industry
This section describes the lumber industry's production processes, output, costs of
production, and capacity utilization.
4.2.1.1 Production Processes
Sawn lumber. Sawn lumber is softwood or hardwood trimmed at a sawmill for future
uses in construction, flooring, furniture, or other markets. Softwoods, such as Douglas fir
and spruce, are used for framing in residential or light-commercial construction. Hardwoods,
such as maple and oak, are used in flooring, furniture, crating, and other applications.
4-2
-------
Table 4-2. Value of Shipments for the Lumber and Wood Products Industry
(SIC 24/NAICS 321), 1987-1996
Year Value of Shipments (1992 $106)
1987 85,383.4
1988 85,381.2
1989 85,656.8
1990 86,203.0
1991 81,666.0
1992 81,564.8
1993 74,379.6
1994 79,602.0
1995 87,574.6
1996 85,724.0
Sources: U.S. Department of Commerce, Bureau of the Census. 1996. 1992 Census of Manufactures, Subject
Series: General Summary. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1 990-1998. Annual Survey of Manufactures
[Multiple Years]. Washington, DC: Government Printing Office.
Lumber is prepared at mills using a four-step process. First, logs are debarked and
trimmed into cants, or partially finished lumber. The cants are then cut to specific lengths.
Logs are generally kept wet during storage to prevent cracking and to keep them supple.
However, after being cut, the boards undergo a drying process, either in open air or in a kiln,
to reduce the moisture content. The drying process may take several months and varies
according to the plant's climate and the process used. Finally, the lumber may be treated
with a surface protectant to prevent sap stains and prepare it for export (EPA, 1995a).
Reconstituted wood products. Reconstituted wood products, such as particleboard,
medium density fiberboard, hardboard, and oriented strandboard, are made from raw wood
that is combined with resins and other additives and processed into boards. The size of the
wood particles used varies from sawdust to strands of wood. Once combined, the ingredients
are formed into a mat and then, at high temperatures, pressed into a board. A final finishing
process prepares the boards for delivery.
Wood preserving. Wood is treated with preservative to protect it from mechanical,
physical, and chemical influences (EPA, 1995a). Treatment agents are either water-based
4-3
-------
inorganics, such as copper arsenate (78 percent), or oil-borne organics, such as creosote
(21 percent) (EPA, 1995a). Wood preservatives are usually applied using a pressure
treatment process or a dipping tank. Producers achieve the best results when the lumber's
moisture content is reduced to a point where the preservative can be easily soaked into the
wood. Treated wood is then placed in a kiln or stacked in a low-humidity climate to dry.
4.2.1.2 Types of Output
The lumber and wood products industry produces essential inputs into the
construction, remodeling, and furniture sectors. Lumber and reconstituted wood products are
produced in an array of sizes and can be treated to enhance their value and shelf-life. These
products are intermediate goods; they are purchased by other industries and incorporated into
higher value-added products. In addition to sawmills, the lumber and wood products
industry includes kitchen cabinets, wood containers, and other wooden products used for
fabricating finished goods for immediate consumption.
4.2.1.3 Major By-Products and Co-Products
Shavings, sawdust, and wood chips are the principal co-products of sawn lumber.
Paper mills and makers of reconstituted wood products frequently purchase this material as
an input. By-products are limited to emissions from the drying process and from use of
preservatives.
Very little solid waste is generated by reconstituted wood products manufacturing.
Because the production process incorporates all parts of the sawn log, little is left over as
waste. However, air emissions from dryers are a source of emissions.
Wood preserving results in two types of by-products: air emissions and process
debris. As preservatives dry, either in a kiln or outside, they emit various chemicals into the
air. At plants with dipping processes, wood chips, stones, and other debris build up in the
dipping tank. The debris is routinely collected and disposed of.
4.2.1.4 Costs of Production
The costs of production for the wood products industry fluctuate with the demand for
the industry's products. Most notably, the costs of production steadily declined during the
early 1990s as recession stifled furniture purchases and new housing starts (see Table 4-3).
Overall, employment in the lumber and wood products industry increased approximately 6
percent from 1987 to 1996. During this same period, payroll costs decreased 12 percent,
indicating a decrease in average annual income per employee. New capital investment and
4-4
-------
Table 4-3. Inputs for the Lumber and Wood Products Industry (SIC 24/NAICS 321),
1987-1996
Labor
Year
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Quantity
(103)
698.4
702.4
684.2
677.7
623.6
655.8
685.4
718.5
740.2
738.7
Payroll
(1992 $106)
15,555.5
15,800.0
15,381.3
15,612.9
14,675.8
13,881.8
11,798.9
12,212.5
13,915.4
13,933.7
Materials
(1992 $106)
50,509.2
51,341.0
51,742.2
53,369.0
50,416.3
48,570.0
45,300.3
48,535.6
53,732.9
52,450.1
New Capital
Investment
(1992 $106)
2,234.3
2,099.4
2,329.9
2,315.3
2,006.5
1,760.1
1,538.1
1,956.8
2,553.1
2,659.9
Sources: U.S. Department of Commerce, Bureau of the Census. 1996. 1992 Census of Manufactures, Subject
Series: General Summary. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1 990-1998. Annual Survey of Manufactures
[Multiple Years]. Washington, DC: Government Printing Office.
costs of materials generally moved in tandem over the 10-year period, increasing from 1987
to 1990 and 1994 to 1996 and decreasing from 1991 to 1993.
4.2.1.5 Capacity Utilization
Full production capacity is broadly defined as the maximum level of production an
establishment can obtain under normal operating conditions. The capacity utilization ratio is
the ratio of the actual production level to the full production level. Table 4-4 presents the
historical trends in capacity utilization for the lumber and wood products industry. The
varying capacity utilization ratios reflect adjusting production levels and new production
facilities going on- or off-line. The capacity utilization ratio for the industry in 1996 was 78;
the average over the last 6 years was 79 percent.
4.2.2 Demand Side of the Industry
This section describes the demand side of the market, including product
characteristics, the uses and consumers of the final products, organization of the industry,
and markets and trends.
4-5
-------
Table 4-4. Capacity Utilization Ratios for Lumber and Wood Products Industry,
1991-1996
1991 1992 1993 1994 1995 1996
78 80 81 80 77 78
Note: All values are percentages.
Source: U.S. Department of Commerce, Bureau of the Census. 1998. Survey of Plant Capacity: 1996.
Washington, DC: Government Printing Office.
4.2.3 Product Characteristics
Lumber and wood products are valued both for their physical attributes and their
relative low cost. Wood is available in varying degrees of durability, shades, and sizes and
can be easily shaped. Lumber and wood products have long been the principal raw materials
for the residential and light commercial construction industries, the remodeling industry, and
the furniture industry. Wood is readily available because over one-third of the United States
is forested. The ready supply of wood reduces its costs.
4.2.4 Uses and Consumers of Outputs
Lumber and wood products are used in a wide range of applications, including
residential and noresidential construction; repair/remodeling and home improvement
projects; manufactured housing; millwork and wood products; pulp, paper, and paperboard
mills; toys and sporting goods; kitchen cabinets; crates and other wooden containers; office
and household furniture; and motor homes and recreational vehicles (Haltmaier, 1998).
4.2.5 Organization of the Industry
In 1992, 33,878 companies produced lumber and wood products and operated 35,807
facilities, as shown in Table 4-5. Byway of comparison, in 1987, 32,014 companies
controlled 33,987 facilities. About two-thirds of all establishments have nine or fewer
employees. Between 1987 and 1992, the number of facilities with nine or fewer employees
increased more than 10 percent to 23,590. These facilities' share of the value of shipments
increased about 18.3 percent. Although the number of establishments employing 100 to 249
people decreased during that time, that category's shipment value jumped nearly 40 percent.
The remaining facility categories lost both facilities and value of shipment.
4-6
-------
Table 4-5. Size of Establishments and Value of Shipments for the Lumber and Wood
Products Industry (SIC 24/NAICS 321)
1987
Average Number of Employees
in Establishment
1 to 4 employees
5 to 9 employees
10 to 19 employees
20 to 49 employees
50 to 99 employees
100 to 249 employees
250 to 499 employees
500 to 999 employees
1,000 to 2,499 employees
2,500 or more employees
Total
Number of
Facilities
14,562
6,702
5,353
4,160
1,702
1,190
260
47
4
2
33,987
Value of
Shipments
(1992 $10')
2,769.7
4,264.4
6,982.3
28,551.3
(D)
24,583.3
12,093.4
3,907.9
2,231.3
(D)
85,383.4
1992
Number of
Facilities
15,921
7,669
5,331
3,924
1,615
1,082
219
39
4
3
35,807
Value of
Shipments
(1992 $10')
3,288.9
5,030.4
6,902.8
26,964.9
(D)
34,051.4
(D)
3,331.4
598.6
1,396.4
81,564.8
(D) = undisclosed
Sources: U.S. Department of Commerce, Bureau of the
Series: General Summary. Washington, DC:
U.S. Department of Commerce, Bureau of the
Series: General Summary. Washington, DC:
Census. 1991. 1987 Census of Manufactures, Subject
Government Printing Office.
Census. 1996. 1992 Census of Manufactures, Subject
Government Printing Office.
Market structure can affect the size and distribution of regulatory impacts.
Concentration ratios are often used to evaluate the degree of competition in a market, with
low concentration indicating the presence of a competitive market, and higher concentration
suggesting less-competitive markets. Firms in less-concentrated industries are more likely to
be price takers, while firms in more-concentrated industries are more likely to influence
market prices. Typical measures include four- and eight-firm concentration ratios (CR4 and
CR8) and Herfindahl-Hirschmann indices (HHI). The CR4 for lumber and wood products
subsectors represented in the boilers inventory database ranges between 13 and 50, meaning
that, in each subsector, the top firms' combined sales ranged from 13 to 50 percent of that
respective subsector's total sales. The CR8 ranges from 47 to 66 (U.S. Department of
Commerce, 1995d).
4-7
-------
Although there is no objective criterion for determining market structure based on the
values of concentration ratios, the 1992 Department of Justice's (DOJ's) Horizontal Merger
Guidelines provide criteria for doing so based on HHIs. According to these criteria,
industries with HHIs below 1,000 are considered unconcentrated (i.e., more competitive),
those with HHIs between 1,000 and 1,800 are considered moderately concentrated (i.e.,
moderately competitive), and those with HHIs above 1,800 are considered highly
concentrated (i.e., less competitive) (DOJ, 1992). Firms in less-concentrated industries are
more likely to be price takers, while firms in more-concentrated industries are more likely to
be able to influence market prices. The unconcentrated nature of the markets is also
indicated by HHIs of 1,000 or less (DOJ, 1992). Table 4-6 presents various measures of
market concentration for sectors within the lumber and wood products industry. All lumber
and wood products industries are considered unconcentrated and competitive.
Table 4-6. Measures of Market Concentration for Lumber and Wood Products
Markets
SIC
2421
2434
2449
2491
2493
2499
Description
Saw Mills and Planing
Mills
Wood Kitchen
Cabinets
Wood Containers,
N.E.C.
Wood Preserving
Reconstituted Wood
Products
Wood Products,
N.E.C.
Comparable
NAICS CR4 CR8
321912, 14 20
321113,
321918,
321999
33711 19 25
32192 34 47
321114 17 28
321219 50 66
339999, 13 19
333414,
32192,
321999
Number of
HHI Companies
78 5,302
156 4,303
414 217
152 408
765 193
70 2,656
Number of
Facilities
6004
4323
225
486
288
2754
Sources: U.S. Department of Commerce, Bureau of the Census. 1995d. 1992 Concentration Ratios in
Manufacturing. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1996. 1992 Census of Manufactures, Subject
Series: General Summary. Washington, DC: Government Printing Office.
4-8
-------
4,2,6 Markets and Trends
The U.S. market for lumber and wood products is maturing, and manufacturers are
looking to enter other markets. Although 91 percent of the industry's products are consumed
by the U.S. domestic market, the share of exports increases each year. Exports more than
doubled in value from $3 billion in 1986 to $7.3 billion in 1996 (Haltmaier, 1998). The U.S.
market grew only 2 percent between 1986 and 1996. American manufacturers are focusing
on growing construction markets in Canada, Mexico, and the Pacific Rim, with products
such as durable hardwood veneer products and reconstituted wood boards (EPA, 1995a).
4.3 Furniture and Related Product Manufacturing (SIC 25/NAICS 337)
More than 20,000 establishments in the United States produce furniture and furniture-
related products. These establishments are located across the United States but are
traditionally most concentrated in southern states, such as North Carolina, Mississippi,
Alabama, and Tennessee. According to the "1997 Economic Census," these establishments
employed more than 600,000 people and paid annual wages of nearly $15 billion. The
overall industry-wide value of shipments was $63.9 billion that year (U.S. Department of
Commerce, 2001).
This industry is in a state of change: rapid U.S. economic growth translated into
vigorous sales of household and office furniture, but this trend is unlikely to continue as the
U.S. economy cools after its record run. Adding to industry fluctuation is the merger of two
large firms, Lay-Z-Boy and LADD Furniture. Although the industry includes a multitude of
niche market players, it is really dominated by a few large companies that operate several
subsidiaries, each with its own brand identity. It is unclear whether the merger between two
key players in the market will compel other large manufacturers to pursue mergers and
acquisitions.
What is clear, however, is that large U.S. manufacturers will seek to leverage their
brand identities into wider profit margins by operating direct sales establishments and co-
branding. Manufacturers that are moving into retail and distribution include Bassett
Furniture, Thomasville Furniture, Ethan Allen Interiors, and Drexel. Co-branding efforts are
aimed at capitalizing on the combined power of two identities, such as the Thomas Kinkade
Collection from Lay-Z-Boy and popular artist Thomas Kinkade and the Ernest Hemingway
Collection from Thomasville. The overarching goal is to enhance margins and ward off
invigorated competition from foreign companies that have used this strategy to capture U.S.
market share, such as the Swedish manufacturer Ikea (Lemm, 2000).
4-9
-------
U.S. imports of household furniture totaled nearly $7 billion in 1998. Between 1992
and 1998, furniture imports grew at an annualized rate of nearly 15 percent. Jamie Lemm, an
analyst with the U.S. Department of Commerce's Office of Consumer Goods attributes this
growth to changes in U.S. manufacturing and markets:
A portion of [the] increase can be attributed to the labor-intensive furniture
parts imported by U.S. manufacturers to enhance product lines, but the
increase also signifies the growing importance of the U.S. market to foreign
firms. While some U.S. manufacturers operate showrooms, galleries, and
retail outlets in foreign markets, few sell internationally on a large scale. In
1998, U.S. furniture exports totaled $1.6 billion, accounting for only 6 percent
of all U.S. product shipments.
4.4 Paper and Allied Products (SIC 26/NAICS 322)
The paper and allied products industry is one of the largest manufacturing industries
in the United States. In 1996, the industry shipped nearly $150 billion in paper commodities.
The industry produces a wide range of wood pulp, primary paper products, and paperboard
products such as printing and writing papers, industrial papers, tissues, container board, and
boxboard. The industry also includes manufacturers that "convert" primary paper and
paperboard into finished products like envelopes, packaging, and shipping containers (EPA,
1995b). Paper and allied products industry subsectors that are likely to be affected by the
regulation are listed in Table 4-7.
Table 4-7. Paper and Allied Products Industry Markets Likely to Be Affected by
Regulation
SIC NAICS Industry Description
2611 32211 Pulp Mills
2621 32212 Paper Mills
2676 322291 Sanitary Paper Products
Source: Industrial Combustion Coordinated Rulemaking (ICCR). 1998. Data/Information Submitted to the
Coordinating Committee at the Final Meeting of the Industrial Combustion Coordinated Rulemaking
Federal Advisory Committee. EPA Docket Numbers A-94-63, II-K-4b2 through -4b5. Research
Triangle Park, North Carolina. September 16-17.
4-10
-------
Table 4-8 lists the paper and allied products industry's value of shipments from 1987
to 1996. The industry's performance is tied to raw material prices, labor conditions, and
worldwide inventories and demand (EPA, 1995b). Performance over the 10-year period was
typical of most manufacturing industries. The industry expanded in the late 1980s, then
contracted as demand tapered off as the industry suffered recessionary effects. In the two
years after 1994, the industry's value of shipments increased 9.3 percent to $149.5 billion.
Table 4-8. Value of Shipments for the Paper and Allied Products Industry
(SIC 26/NAICS 322), 1987-1996
Year Value of Shipments (1992 $106)
1987 129,927.8
1988 136,829.4
1989 138,978.3
1990 136,175.7
1991 132,225.0
1992 133,200.7
1993 131,362.2
1994 136,879.9
1995 135,470.3
1996 149,517.1
Sources: U.S. Department of Commerce, Bureau of the Census. 1996. 1992 Census of Manufactures, Subject
Series: General Summary. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1 990-1998. Annual Survey of Manufactures,
[Multiple Years]. Washington, DC: Government Printing Office.
4.4.1 Supply Side of the Industry
4.4.1.1 Production Process
The manufacturing paper and allied products industry is capital- and resource-
intensive, consuming large amounts of pulp wood and water in the manufacturing process.
Approximately half of all paper and allied products establishments are integrated facilities,
meaning that they produce both pulp and paper on-site. The remaining half produce only
paper products; few facilities produce only pulp (EPA, 1995b).
4-11
-------
The paper and paperboard manufacturing process can be divided into three general
steps: pulp making, pulp processing, and paper/paperboard production. Paper and
paperboard are manufactured using what is essentially the same process. The principal
difference between the two products is that paperboard is thicker than paper's 0.3 mm.
Producers manufacture pulp mixtures by using chemicals, machines, or both to
reduce raw material into small fibers. In the case of wood, the most common pulping
material, chemical pulping actions release cellulose fibers by selectively destroying the
chemical bonds that bind the fibers together (EPA, 1995b). Impurities are removed from the
pulp, which then may be bleached to improve brightness. Only about 20 percent of pulp and
paper mills practice bleaching (EPA, 1995b). The pulp may also be further processed to aid
in the paper-making process.
During the paper-making stage, the pulp is strengthened and then converted into
paper. Pulp can be combined with dyes, resins, filler materials, or other additives to better
fulfill specifications for the final product. Next, the water is removed from the pulp, leaving
the pulp on a wire or wire mesh conveyor. The fibers bond together as they are carried
through heated presses and rollers. The paper is stored on large rolls before being shipped
for conversion into another product, such as envelopes and boxes, or cut into paper sheets for
immediate consumption.
4.4.1.2 Types of Output
The paper and allied products industry's output ranges from writing papers to
containers and packaging. Paper products include printing and writing papers; paperboard
boxes; corrugated and solid fiber boxes; fiber cans, drums, and similar products; sanitary
food containers; building paper; packaging; bags; sanitary paper napkins; envelopes;
stationary products; and other converted paper products.
4.4.1.3 Major By-Products and Co-Products
The paper and allied products industry is the largest user of industrial process water
in the United States. In 1988, a typical mill used between 16,000 and 17,000 gallons of
water per ton of paper produced. The equivalent amount of waste water discharged each day
is about 16 million cubic meters (EPA, 1995b). Most facilities operate waste water treatment
facilities on site to remove biological oxygen demand (BOD), total suspended solids (TSS),
and other pollutants before discharging the water into a nearby waterway.
4-12
-------
4.4.1.4 Costs of Production
Historical statistics for the costs of production for the paper and allied products
industry are listed in Table 4-9. From 1987 to 1996, industry payroll generally ranged from
approximately $19 to 20 billion. Employment peaked at 633,200 people in 1989 and
declined slightly to 630,600 people by 1996. Materials costs averaged $74.4 billion a year
and new capital investment averaged $8.3 billion a year.
Table 4-9. Inputs for the Paper and Allied Products Industry (SIC 26/NAICS 322),
1987-1996
Labor
Year
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Quantity (103)
611.1
619.8
633.2
631.2
624.7
626.3
626.3
621.4
629.2
630.6
Payroll
(1992 $106)
20,098.6
19,659.0
19,493.1
19,605.2
19,856.3
20,491.9
20,602.6
20,429.7
18,784.3
19,750.0
Materials
(1992 $106)
70,040.6
73,447.4
75,132.5
74,568.8
72,602.5
73,188.0
73,062.6
76,461.6
79,968.6
75,805.9
New Capital
Investment
(1992 $10')
6,857.5
8,083.8
10,092.9
11,267.2
9,353.9
7,962.4
7,265.2
6,961.7
7,056.8
8,005.9
Sources: U.S. Department of Commerce, Bureau of the Census. 1996. 1992 Census of Manufactures, Subject
Series: General Summery. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1990-1998. Annual Survey of Manufactures
[Multiple Years]. Washington, DC: Government Printing Office.
4.4.1.5 Capacity Utilization
Table 4-10 presents the trend in capacity utilization for the paper and allied products
industry. The varying capacities reflect adjusting production levels and new production
facilities going on- or off-line. The average capacity utilization ratio for the paper and allied
products industry between 1991 and 1996 was approximately 80, with capacity declining
slightly in recent years.
4-13
-------
Table 4-10. Capacity Utilization Ratios for the Paper and Allied Products Industry,
1991-1996
1991 1992 1993 1994 1995 1996
78 80 81 80 77 78
Note: All values are percentages.
Source: U.S. Department of Commerce, Bureau of the Census. 1998. Survey of Plant Capacity: 1996.
Washington, DC: Government Printing Office.
4.4.2 Demand Side of the Industry
4.4.2.1 Product Characteristics
Paper is valued for its diversity in product types, applications, and low cost due to
ready access to raw materials. Manufacturers produce papers of varying durabilities,
textures, and colors. Consumers purchasing large quantities of papers may have papers
tailored to their specification. Papers may be simple writing papers or newsprint for personal
consumption and for the printing and publishing industry or durable for conversion into
shipping cartons, drums, or sanitary boxes. Inputs in the paper production process are
readily available in the United States because one-third of the country is forested, and
facilities generally have ready access to waterways.
4.4.2.2 Uses and Consumers of Products
The paper and allied products industry is an integral part of the U.S. economy; nearly
every industry and service sector relies on paper products for its personal, education, and
business needs. Among a myriad of uses, papers are used for correspondence, printing and
publishing, packing and storage, and sanitary purposes. Common applications are all
manners of reading material, correspondence, sanitary containers, shipping cartons and
drums, and miscellaneous packing materials.
4.4.3 Organization of the Industry
In 1992, 4,264 companies produced paper and allied products and operated 6,416
facilities. Byway of comparison, 4,215 companies controlled 1,732 facilities in 1987.
4-14
-------
Although the number of small firms and facilities increased during those 5 years, the industry
is dominated by high-volume, low-cost producers (Haltmaier, 1998). Even though they
account for only 45 percent of all facilities, those with 50 or more employees contribute
more than 93 percent of the industry's total value of shipments (see Table 4-11). (According
to the Small Business Administration, those companies employing fewer than 500 employees
are "small")
Table 4-11. Size of Establishments and Value of Shipments for the Paper and Allied
Products Industry (SIC 26/NAICS 322)
1987
Number of Employees in
Establishment
1 to 4 employees
4 to 9 employees
10 to 19 employees
20 to 49 employees
50 to 99 employees
100 to 249 employees
250 to 499 employees
500 to 999 employees
1,000 to 2,499 employees
2,500 or more employees
Total
Number of
Facilities
729
531
888
1,433
1,018
1,176
308
145
63
1
1,732
Value of
Shipments
($106)
640.6
(D)
1,563.4
18,328.6
(D)
32,141.7
24,221.1
28,129.1
24,903.1
(D)
129,927.8
1992
Number of
Facilities
786
565
816
1,389
1,088
1,253
298
159
62
6,416
Value of
Shipments
($106)
216
483
1,456.5
6,366.6
12,811.5
35,114.0
22,281.2
31,356.5
23,115.4
133,200.7
(D) = undisclosed
Sources: U.S. Department of Commerce, Bureau of the Census. 1990c
Industry Series: Pulp, Paper, and Board Mills. Washington,
U.S. Department of Commerce, Bureau of the Census. 1995c
Industry Series: Pulp, Paper, and Board Mills. Washington,
. 1987 Census of Manufactures,
DC: Government Printing Office.
. 1992 Census of Manufactures,
DC: Government Printing Office.
For paper and allied products markets likely to be affected by the boilers regulation,
the CR4 ranged between 29 and 68 in 1992 (see Table 4-12). This means that, in each
subsector, the top firms' combined sales ranged from 29 to 68 percent of their respective
industry's total sales. For example, in the sanitary paper products industry, the CR4 ratios
indicate that a few firms control 68 percent of the market. This sector's moderately
concentrated nature is also indicated by its HHI of 1,451 (DOJ, 1992). The remaining two
sectors' HHIs indicate that their respective markets are unconcentrated (i.e., competitive).
4-15
-------
Table 4-12. Measurements of Market Concentration for Paper and Allied Products
Markets
SIC
2611
2621
2676
Description
Pulp Mills
Paper Mills
Sanitary Paper Products
CR4
48
29
68
CR8
75
49
82
HHI
858
392
1,451
Number of
Companies
29
127
80
Number of
Facilities
45
280
150
Sources: U.S. Department of Commerce, Bureau of the Census. 1995d. 1992 Concentration Ratios in
Manufacturing. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1995c. 1992 Census of Manufactures,
Industry Series: Pulp, Paper, and Board Mills. Washington, DC: Government Printing Office.
4,4,4 Markets and Trends
The Department of Commerce projects that shipments of paper and allied products
will increase through 2002 by an annual average of 2.5 percent (Haltmaier, 1998). Because
nearly all of the industry's products are consumer related, shipments will be most affected by
the health of the U.S. and global economy. The United States is a key competitor in the
international market for paper products and, after Canada, is the largest exporter of paper
products. According to Haltmaier (1998), the largest paper and allied products exporters in
the world are Canada (with 23 percent of the market), the United States (10 to 15 percent),
Finland (8 percent), and Sweden (7 percent).
4.5 Medicinal Chemicals and Botanical Products and Pharmaceutical Preparations
(SICs 2833, 2834/NAICS 32451)
The pharmaceutical preparations industry (SIC 2834/NAICS 32451) and the
medicinal chemicals and botanical products industry (SIC 2833/NAICS 32451) are both
primarily engaged in the research, development, manufacture, and/or processing of medicinal
chemicals and pharmaceutical products. Apart from manufacturing drugs for human and
veterinary consumption, the industries grind, grade, and mill botanical products that are
inputs for other industries. Typically, most facilities cross over into both industries (EPA,
1997a). Products include drugs, vitamins, herbal remedies, and production inputs, such as
alkaloids and other active medicinal principals.
Table 4-13 presents both industries' value of shipments from 1987 to 1996.
Medicinals and botanicals' performance during the late 1980s and early 1990s was mixed.
However, shipments increased steadily from 1994 to 1996, increasing 37.7 percent as natural
4-16
-------
Table 4-13. Value of Shipments for the Medicinals and Botanicals and Pharmaceutical
Preparations Industries, 1987-1996
SIC 2833" Medicinals &
Year Botanicals ($106)
1987 4,629.1
1988 5,375.4
1989 5,708.9
1990 5,535.8
1991 6,637.7
1992 6,438.5
1993 5,669.2
1994 5,774.7
1995 6,404.1
1996 7,952.8
SIC 2834" Pharmaceutical
Preparations ($106)
44,345.7
46,399.1
48,083.6
49,718.0
49,866.3
50,417.9
50,973.5
53,144.7
53,225.9
55,103.6
a Comparable NAICS: 325411,325412.
Sources: U.S. Department of Commerce, Bureau of the Census. 1995a. 1992 Census of Manufactures,
Industry Series: Drug Industry. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1 990-1998. Annual Survey of Manufactures
[Multiple Years]. Washington, DC: Government Printing Office.
products such as herbs and vitamins became more popular (EPA, 1997'a). Pharmaceutical
preparations' shipments increased steadily over the 10-year period. From 1987 to 1996, the
industry's shipments increased 24.3 percent to $55.1 billion in 1996.
4.5.1 Supply Side of the Industry
4.5.1.1 Production Processes
The medicinals and botanical products industry and the pharmaceutical preparations
industry share similar production processes. Many products of the former are inputs in the
latter's production process. There are three manufacturing stages: research and
development, preparation of bulk ingredients, and formulation of the final product.
The research and development stage is a long process both to ensure the validity and
benefit of the end product and to satisfy the requirements of stringent federal regulatory
committees. (The pharmaceutical industry operates under strict oversight of the Food and
Drug Administration [FDA].) Therefore, every stage in the development of new drugs is
thoroughly documented and studied. After a new compound is discovered, it is subjected to
4-17
-------
numerous laboratory and animal tests. Results are presented to the FDA via applications that
present and fully disclose all findings to date. As research and development proceeds,
studies are gradually expanded to involve human trials of the new compound. Should FDA
approve the compound, the new product is readied for mass production.
To ensure a uniform product, all ingredients are prepared in bulk using batch
processes. Companies produce enough of each ingredient to satisfy projected sales demand
(EPA, 1997a). Prior to production, all equipment is thoroughly cleaned, prepared, and
validated to prevent any contaminants from entering the production cycle. Most ingredients
are prepared by chemical synthesis, a method whereby primary ingredients undergo a
complex series of processes, including many intermediate stages and chemical reactions in a
step-by-step fashion (EPA, 1997a).
After the bulk materials are prepared, they are converted into a final usable form.
Common forms include tablets, pills, liquids, creams, and ointments. Equipment used in this
final stage is prepared in the same manner as that involved in the bulk preparation process.
Clean and validated machinery is used to process and package the pharmaceuticals for
shipment and consumption.
4.5.1.2 Types of Output
Both industries produce pharmaceutical and botanical products for end consumption
and intermediate products for the industries' own applications. Products include vitamins,
herbal remedies, and alkaloids. Prescription and over-the-counter drugs are produced in
liquid, tablet, cream, and other forms.
4.5.1.3 Major By-Products and Co-Products
Both industries produce many by-products because of the large number of primary
inputs and the extensive chemical processes involved. Wastes and emissions vary by the
process employed, raw materials consumed, and equipment used. In general, emissions
originate during drying and heating stages and during process water discharge. Emissions
controls are in place pursuant to environmental regulations. Other wastes include used
filters, spent raw materials, rejected product, and reaction residues (EPA, 1997a).
4.5.1.4 Costs of Production
Table 4-14 presents SIC 2833 industry's costs of production and employment
statistics from 1987 to 1996. Employment was stable during the late 1980s before steadily
growing in the 1990s. In 1987, medicinals and botanicals employed 11,600 people. By
4-18
-------
Table 4-14. Inputs for Medicinal Chemicals and Botanical Products Industry
(SIC 2833/NAICS 32451), 1987-1996
Labor
Year
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Quantity (103)
11.6
11.3
11.4
10.9
12.5
13.0
13.0
13.9
14.1
16.8
Payroll
($106)
520.2
494.4
504.9
476.4
568.6
587.1
584.3
572.6
625.0
752.1
Materials
($106)
2,229.3
2,658.8
3,118.4
2,902.4
3,368.2
3,245.9
2,638.4
2,755.2
3,006.0
3,793.9
New Capital
Investment ($106)
158.2
194.9
263.4
218.9
512.9
550.5
470.0
480.3
356.2
752.1
Sources: U.S. Department of Commerce, Bureau of the Census. 1995a. 1992 Census of Manufactures,
Industry Series: Drug Industry. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1 990-1998. Annual Survey of Manufactures,
[Multiple Years]. Washington, DC: Government Printing Office.
1996, the industry employed 16,800, an increase of nearly 45 percent. Materials costs
matched the increase in shipments over this same period. Industry growth also fed new
capital investments, which averaged $191.2 million a year in the late 1980s and $515.6
million a year in the early to mid-1990s.
SIC 2834's costs of production and employment for 1987 to 1996 are presented in
Table 4-15. The number of people employed by the industry ranged between 123,000 and
144,000; employment peaked in 1990 before declining by 21,000 jobs by the end of 1992.
During this 10-year period, the cost of materials rose 42.1 percent. The increase is
associated with increased product shipments and the development of new, more expensive
medications (Haltmaier, 1998). New capital investment averaged $2.3 billion a year.
4-19
-------
Table 4-15. Inputs for the Pharmaceutical Preparations Industry
(SIC 2834/NAICS 32451), 1987-1996
Labor
Year
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Quantity
(103)
131.6
133.4
141.8
143.8
129.1
122.8
128.2
134.2
143.0
136.9
Payroll
($106)
5,759.2
5,447.2
6,177.5
6,223.9
5,275.8
4,949.4
5,184.2
5,368.4
5,712.4
5,547.3
Materials
($106)
11,693.7
12,634.8
12,874.2
13,237.6
13,546.6
13,542.5
13,508.7
13,526.1
15,333.6
16,611.1
New Capital
Investment ($106)
2,032.7
2,234.0
2,321.4
2,035.3
1,864.7
2,450.0
2,385.2
2,531.9
2,856.1
2,317.0
Sources: U.S. Department of Commerce, Bureau of the Census. 1995a. 1992 Census of Manufactures,
Industry Series: Drug Industry. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1990-1998. Annual Survey of Manufactures,
[Multiple Years]. Washington, DC: Government Printing Office.
4.5.1.5 Capacity Utilization
Table 4-16 presents the trend in these ratios from 1991 to 1996 for both industries.
The varying capacity ratios reflect adjusting production volumes and new production
facilities and capacity going both on- and off-line. In 1996, the capacity utilization ratios for
SICs 2833 and 2834 were 84 and 67, respectively.
4.5.2 Demand Side of the Industry
New product introductions and improvements on older medications by the drug
industry have greatly improved the health and well-being of the U.S. population (Haltmaier,
1998). Products help alleviate or reduce physical, mental, and emotional ailments or reduce
the severity of symptoms associated with disease, age, and degenerative conditions. Dietary
supplements, such as vitamins and herbal remedies, ensure that consumers receive nutrients
4-20
-------
Table 4-16. Capacity Utilization Ratios for the Medicinal Chemicals and Botanical
Products (SIC 2833/NAICS 32451) and Pharmaceutical Preparations
(SIC 2834/NAICS 32451) Industries, 1991-1996
SIC 2833/NAICS 32451
SIC 2834/NAICS 32451
1991
84
76
1992
86
74
1993
89
70
1994
80
67
1995
90
63
1996
84
67
Note: Capacity utilization ratio is the ratio of the actual production level to the full production level. All
values are percentages.
Source: U.S. Department of Commerce, Bureau of the Census. 1998. Survey of Plant Capacity: 1996.
Washington, DC: Government Printing Office.
of which they may not ordinarily consume enough. Products are available in a range of
dosage types, such as tablets and liquids.
Although prescription medications are increasingly distributed through third parties,
such as hospitals and health maintenance organizations, the general population remains the
end user of pharmaceutical products. As the average age of the U.S. population adjusts to
reflect large numbers of older people, the variety and number of drugs consumed increases.
An older population will generally consume more medications to maintain and improve
quality of life (Haltmaier, 1998).
4.5.3 Organization of the Industry
In 1992, 208 companies produced medicinal chemicals and botanical products and
operated 225 facilities (see Table 4-17). The number of companies and facilities in 1992 was
the same as that of 1987, although shipment values increased almost 40 percent. The
average facility employed more people in 1992 than in 1987. In fact, the number of facilities
employing 50 or more people grew from 37 to 45. These facilities accounted for the lion's
share of the industry's shipments. According to the Small Business Administration,
companies in this SIC code are considered small if they employ fewer than 750 employees.
It is unclear what percentage of the facilities listed in Table 4-17 are small companies.
In 1992, 585 companies manufactured pharmaceutical preparations and operated 691
facilities. By way of comparison, 640 companies operated 732 facilities in 1987. Although
the number of facilities declined by 41, no particular category lost or gained an exceptional
number of facilities. The biggest movement was in the five to nine employees category,
4-21
-------
Table 4-17. Size of Establishments and Value of Shipments for the Medicinal
Chemicals and Botanical Products (SIC 2833/NAICS 32451) and Pharmaceutical
Preparations (SIC 2834/NAICS 32451) Industries
1987
Number of Employees in
Establishment
SIC 2833/NAICS 32451
1 to 4 employees
5 to 9 employees
10 to 19 employees
20 to 49 employees
50 to 99 employees
100 to 249 employees
250 to 499 employees
500 to 999 employees
1,000 to 2,499 employees
Total
SIC 2834/NAICS 32451
1 to 4 employees
5 to 9 employees
10 to 19 employees
20 to 49 employees
50 to 99 employees
100 to 249 employees
250 to 499 employees
500 to 999 employees
1,000 to 2,499 employees
2,500 employees or more
Total
Number of
Facilities
61
34
46
47
15
12
5
4
1
225
158
108
102
117
66
76
50
23
24
8
732
Value of
Shipments
($106)
20.7
38.6
237.0
287.3
273.6
520.6
753.0
2478.2
(D)
4629.1
58.7
178.8
320.3
932.5
1231.0
3596.0
9239.7
4946.9
15,100.9
8740.9
44,345.7
1992
Number of
Facilities
62
42
47
29
25
10
4
3
3
225
152
73
101
110
65
77
56
30
21
6
691
Value of
Shipments
($106)
23.8
58.3
357.1
182.0
653.9
5,163.4
(D)
(D)
(D)
6,438.5
115.6
105.4
284.6
815.7
1,966.8
2,912.4
11,394.6
10,077.7
14,525.7
8,219.4
50,417.9
(D) = undisclosed
Sources: U.S. Department of Commerce, Bureau of the Census. 1990a. 1987 Census of Manufactures,
Industry Series: Drug Industry. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1995a. 1992 Census of Manufactures,
Industry Series: Drug Industry. Washington, DC: Government Printing Office.
4-22
-------
which lost 35 facilities. In both years, facilities with more than 50 employees accounted for
at least 95 percent of the industry's shipments.
Table 4-18 presents the measures of market concentration for both industries. For the
medicinals and botanicals industry, the CR4 was 76 in 1992, and the CR8 was 84 (U.S.
Department of Commerce, 1995b). The highly concentrated nature of the market is further
indicated by an HHI of 2,999 (DOJ, 1992). According to the Department of Justice's
Horizontal Merger Guidelines, industries with HHIs above 1,800 are less competitive.
Table 4-18. Measures of Market Concentration for the Medicinal Chemicals and
Botanical Products (SIC 2833/NAICS 32451) and Pharmaceutical Preparations (SIC
2834/NAICS 32451) Industries
SIC
2833
NAICS
32451
Industry
Medicinal Chemicals and
CR4
76
CR8
84
HHI
2,999
Number of
Companies
208
Number of
Facilities
225
Botanical Products
2834 32451 Pharmaceutical 26 42 341 585 691
Preparations
Sources: U.S. Department of Commerce, Bureau of the Census. 1995d. 1992 Concentration Ratios in
Manufacturing. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1995a. 1992 Census of Manufactures,
Industry Series: Drug Industry. Washington, DC: Government Printing Office.
The pharmaceuticals preparations industry is less concentrated than the medicinal
chemicals and botanical products industry. For SIC 2834, the CR4 and CR8 were 26 and 42,
respectively, in 1992. The industry's HHI was 341, indicating a competitive market.
4.5.4 Markets and Trends
According to the Department of Commerce, global growth in the consumption of
pharmaceuticals is projected to accelerate over the coming decade as populations in
developed countries age and those in developing nations gain wider access to health care.
Currently, the United States remains the largest market for drugs, medicinals, and botanicals
and produces more new products than any other country (Haltmaier, 1998). But, nearly
two-fifths of American producers' sales are generated abroad. Top markets for American
exports are China, Canada, Mexico, Australia, and Japan. Most imports originate in Canada,
Russia, Mexico, Trinidad and Tobago, and Norway.
4-23
-------
4.6 Industrial Organic Chemicals Industry (SIC 2869/NAICS 3251)
The industrial organic chemicals (not elsewhere classified) industry
(SIC 2869/NAICS 3251) produces organic chemicals for end-use applications and for inputs
into numerous other chemical manufacturing industries. In nominal terms, it was the single
largest segment of the $367 billion chemical and allied products industry (SIC 28) in 1996,
accounting for approximately 17 percent of the industry's shipments.
All organic chemicals are, by definition, carbon-based and are divided into two
general categories: commodity and specialty. Commodity chemical manufacturers compete
on price and produce large volumes of staple chemicals using continuous manufacturing
processes. Specialty chemicals cater to custom markets, using batch processes to produce a
diverse range of chemicals. Specialty chemicals generally require more technical expertise
and research and development than the more standardized commodity chemicals industry
(EPA, 1995c). Consequently, specialty chemical manufacturers have a greater value added
to their products. End products for all industrial organic chemical producers are as varied as
synthetic perfumes, flavoring chemicals, glycerin, and plasticizers.
Table 4-19 presents the shipments of industrial organic chemicals from 1987 to 1996.
In real terms, the industry's shipments rose in the late 1980s to a high of $54.9 billion before
declining in the early 1990s as the U.S. economy went into recession. By the mid-1990s, the
industry recovered, as product values reached record highs (Haltmaier, 1998). Between 1993
and 1996, the industry's shipments grew 7.3 percent to $57.7 billion.
4.6.1 Supply Side of the Industry
4.6.1.1 Production Processes
Processes used to manufacture industrial organic chemicals are as varied as the
end-products themselves. There are thousands of possible ingredients and hundreds of
processes. Therefore, the discussion that follows is a general description of the ingredients
and stages involved in a typical manufacturing process.
Essentially a set of ingredients (feedstocks) is combined in a series of reactions to
produce end products and intermediates (EPA, 1995c). The typical chemical synthesis
processes incorporate multiple feedstocks in a series of chemical reactions. Commodity
chemicals are produced in a continuous reactor, and specialty chemicals are produced in
batches. Specialty chemicals may undergo a series of reaction steps, as opposed to
4-24
-------
Table 4-19. Value of Shipments for the Industrial Organic Chemicals, N.E.C. Industry
(SIC 2869/NAICS 3251), 1987-1996
Year Value of Shipments (1992 $106)
1987 48,581.7
1988 53,434.7
1989 54,962.9
1990 53,238.8
1991 51,795.6
1992 54,254.2
1993 53,805.2
1994 57,357.1
1995 59,484.3
1996 57,743.3
Sources: U.S. Department of Commerce, Bureau of the Census. 1995b. 1992 Census of Manufactures,
Industry Series: Industrial Organic Chemicals. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1 990-1998. Annual Survey of Manufactures,
Multiple Years. Washington, DC: Government Printing Office.
commodity chemicals' one continuous reaction because a finite amount of ingredients are
prepared and used in the production process. Reactions usually take place at high
temperatures, with one or two additional components being intermittently added. As the
production advances, by-products are removed using separation, distillation, or refrigeration
techniques. The final product may undergo a drying or pelletizing stage to form a more
manageable substance.
4.6.1.2 Types of Output
Miscellaneous industrial organic chemicals comprise nine general categories of
products:
• aliphitic and other acyclic organic chemicals (ethylene); acetic, chloroaceptic,
adipic, formic, oxalic, and tartaric acids and their metallic salts; chloral,
formaldehyde, and methylamine;
• solvents (ethyl alcohol etc.); methanol; amyl, butyl, and ethyl acetates; ethers;
acetone, carbon disulfide and chlorinated solvents;
• polyhydric alcohols (synthetic glycerin, etc.);
4-25
-------
• synthetic perfume and flavoring materials (citral, methyl, oinone, etc.);
• rubber processing chemicals, both accelerators and antioxidants (cyclic and
acyclic);
• cyclic and acyclic plasticizers (phosphoric acid, etc.);
• synthetic tanning agents;
• chemical warfare gases; and
• esters, amines, etc., of polyhydric alcohols and fatty and other acids.
4.6.1.3 Major By-Products and Co-Products
Co-products, by-products, and emissions vary according to the ingredients, processes,
maintenance practices, and equipment used (EPA, 1997b). Frequently, residuals from the
reaction process that are separated from the end product are resold or possibly reused in the
manufacturing process. A by-product from one process may be another's input. The
industry is strictly regulated because it emits chemicals through many types of media,
including discharges to air, land, and water, and because of the volume and composition of
these emissions.
4.6.1.4 Costs of Production
Of all the factors of production, employment in industrial organic chemicals
fluctuated most often between 1987 and 1996 (see Table 4-20). During that time,
employment fell 8.18 percent to 92,100, after a high of 101,000 in 1991. Most jobs lost were
at the production level (Haltmaier, 1998). Facilities became far more computerized,
incorporating advanced technologies into the production process. Even with the drop in
employment, payroll was $200 million more in 1995 than in 1987. The cost of materials
fluctuated between $29 and $36 billion for these years, and new capital investment averaged
$3,646 million a year.
4.6.1.5 Capacity Utilization
Table 4-21 presents the trend in capacity utilization ratios from 1991 to 1996 for the
industrial organic chemicals industry. The varying capacity utilization ratios reflect changes
in production volumes and new production facilities and capacities going on- and off-line.
The capacity utilization ratio for the industry averaged 85.3 over the 6-year period presented.
4-26
-------
Table 4-20. Inputs for the Industrial Organic Chemicals Industry
(SIC 2869/NAICS 3251), 1987-1996
Year
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Quantity (103)
100.3
97.1
97.9
100.3
101.0
100.1
97.8
89.8
92.1
100.3
Labor
Payroll (1992 $10')
4,295.8
4,045.1
3,977.4
4,144.6
4,297.3
4,504.2
4,540.2
4,476.5
4,510.4
5,144.8
Materials
(1992 $10')
28,147.7
29,492.8
29,676.4
29,579.2
29,335.2
31,860.6
30,920.1
33,267.4
33,163.9
36,068.9
New Capital
Investment
(1992 $10')
2,307.4
2,996.5
3,513.0
4,085.5
4,428.7
4,216.6
3,386.1
2,942.8
3,791.0
4,794.7
Sources: U.S. Department of Commerce, Bureau of the Census. 1995b. 1992 Census of Manufactures.
Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1 990-1998. Annual Survey of Manufactures.
Washington, DC: Government Printing Office.
Table 4-21. Capacity Utilization Ratios for the Industrial Organic Chemicals Industry
(SIC 2869/NAICS 3251), 1991-1996
1991 1992 1993 1994 1995 1996
SIC 2869/NAICS 3251 86 81 91 89 84 84
Note: The capacity utilization ratio is the ratio of the actual production level to the full production level.
All values are percentages.
Source: U.S. Department of Commerce, Bureau of the Census. 1998. Survey of Plant Capacity: 1996.
Washington, DC: Government Printing Office.
4-27
-------
4.6.2 Demand Side of the Industry
Industrial organic chemicals are components of many chemical products. Most of the
chemical sectors (classified under SIC 28) are downstream users of organic chemicals.
These sectors either purchase commodity chemicals or enter into contracts with industrial
organic chemical producers to obtain specialty chemicals. Consumers include inorganic
chemicals (SIC 281), plastics and synthetics (SIC 282), drugs (283), soaps and cleaners (SIC
284), paints and allied products (SIC 286), and miscellaneous chemical products (SIC 289).
4.6.3 Organization of the Industry
Although the industry's value of shipments increased nearly 12 percent between 1987
and 1992, the number of facilities producing industrial organic chemicals only increased by 6
percent. Facilities with 100 or more employees continued to account for the majority of the
industry's shipment values. For example, in 1992, 28 percent of all facilities had 100 or
more employees (see Table 4-22), and these facilities produced 89 percent of the industry's
shipment values. The average number of facilities per firm was 1.4 in both years. According
to the Small Business Administration, an industrial organic chemicals company is considered
small if the total number of employees does not exceed 500. It is unclear what percentage of
facilities are owned by small businesses.
The industrial organic chemicals (not elsewhere classified) industry is unconcentrated
and competitive. The CR4 was 29 and the CR8 43; the industry's HHI was 336.
4.6.4 Markets and Trends
The U.S. industrial organic chemical industry is expected to expand through 2002 at
an annual rate of 1.4 percent (Haltmaier, 1998). U.S. producers face increasing competition
domestically and abroad as chemical industries in developing nations gain market share and
increase exports to the United States. American producers will, however, benefit from
decreasing costs for raw materials and energy and productivity gains.
4.7 Electric Services (SIC 4911/NAICS 22111)
The ongoing process of deregulation of wholesale and retail electric markets is
changing the structure of the electric power industry. Deregulation is leading to the
functional unbundling of generation, transmission, and distribution and to competition in the
generation segment of the industry. This profile provides background information on the
U.S. electric power industry and discusses current industry characteristics and trends that
will influence the future generation and consumption of electricity. It is important to note
4-28
-------
Table 4-22. Size of Establishments and Value of Shipments for the Industrial Organic
Chemicals Industry (SIC 2869/NAICS 3251)
1987
Number of Employees in
Establishment
1 to 4 employees
5 to 9 employees
10 to 19 employees
20 to 49 employees
50 to 99 employees
100 to 249 employees
250 to 499 employees
500 to 999 employees
1,000 to 2,499 employees
2,500 or more employees
Number of
Facilities
97
80
91
137
99
110
41
27
11
6
Value of
Shipments
(1992 $106)
552.8
200.9
484.7
1,749.9
2556.3
10,361.2
17,156.9
9,615.5
9,184.6
7,156.9
1992
Number of
Facilities
100
80
97
125
106
111
41
30
10
5
Value of
Shipments
(1992 $106)
102.6
208.7
533.9
1,701.5
3,460.9
8,855.9
9,971.1
13,755.0
9,051.0
6,613.5
Sources: U.S. Department of Commerce, Bureau of the Census. 1995b. 1992 Census of Manufactures,
Industry Series: Industrial Organic Chemicals. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1990b. 1987 Census of Manufactures,
Industry Series, Paints and Allied Products. Washington, DC: Government Printing Office.
that through out this report the terms "boilers," "process heaters," and "units" are
synonymous with "ICI boilers" and "process heaters." Boilers primarily engaged in the
generation of electricity are not covered by the NESHAP under analysis and are therefore
excluded from this analysis. Utility sources are not affected by this NESHAP except for a
small number of nonfossil fuel units within this industry. Those units in this industry that are
affected may be engaged in activities such as heating and mechanized work.
4.7,1 Electricity Production
Figure 4-1 illustrates the typical structure of the electric utility market. Even with the
technological and regulatory changes in the 1970s and 1980s, at the beginning of the 1990s
the structure of the electric utility industry could still be characterized in terms of generation,
transmission, and distribution. Commercial and retail customers were in essence "captive,"
and rates and service quality were primarily determined by public utility commissions.
4-29
-------
Electricity
Generation
Power Plants
Trans-
mission
High Voltage Lines
Transformer
Distribution
Residential
Customers
Small C/l
Customers
Large C/l
Customers
Figure 4-1. Traditional Electric Power Industry Structure
4-30
-------
The majority of utilities are interconnected and belong to a regional power pool.
Pooling arrangements enable facilities to coordinate the economic dispatch of generation
facilities and manage transmission congestion. In addition, pooling diverse loads can
increase load factors and decrease costs by sharing reserve capacity.
4.7.1.1 Generation
As shown in Table 4-23, coal-fired plants have historically accounted for the bulk of
electricity generation in the United States. With abundant national coal reserves and
advances in pollution abatement technology, such as advanced scrubbers for pulverized coal
and flue gas-desulfurization systems, coal will likely remain the fuel of choice for most
existing generating facilities over the near term.
Table 4-23. Net Generation by Energy Source, 1995
Energy Source
Fossil fuels
Coal
Natural gas
Petroleum
Nuclear
Hydroelectric
Renewable/other
Total
Utility Generators
(MWh)
2,021,064
1,652,914
307,306
60,844
673,402
293,653
6,409
2,994,582
Nonutility Generators
(MWh)
287,696
63,440
213,437
3,957
—
14,515
98,295
400,505
Total (MWh)
2,308,760
673,402
308,168
104,704
3,395,033
Sources: U.S. Department of Energy, Energy Information Administration. 1996. Electric Power Annual,
1995. Vol.1. DOE/EIA-0348(95/1). Washington, DC: U.S. Department of Energy.
U.S. Department of Energy, Energy Information Administration. 1 999b. The Changing Structure of
the Electric Power Industry 1999: Mergers and Other Corporate Combinations. Washington, DC:
U.S. Department of Energy.
Natural gas accounts for approximately 10 percent of current generation capacity but
is expected to grow; advances in natural gas exploration and extraction technologies and new
coal gasification have contributed to the use of natural gas for power generation.
Nuclear plants and renewable energy sources (e.g., hydroelectric, solar, wind)
provide approximately 20 percent and 10 percent of current generating capacity,
respectively. However, there are no plans for new nuclear facilities to be constructed, and
there is little additional growth forecasted in renewable energy.
4-31
-------
4.7.1.2 Transmission
Transmission refers to high voltage lines used to link generators to substations where
power is stepped down for local distribution. Transmission systems have been traditionally
characterized as a collection of independently operated networks or grids interconnected by
bulk transmission interfaces.
Within a well-defined service territory, the regulated utility has historically had
responsibility for all aspects of developing, maintaining, and operating transmissions. These
responsibilities included
• system planning and expanding,
• maintaining power quality and stability, and
• responding to failures.
Isolated systems were connected primarily to increase (and lower the cost of) power
reliability. Most utilities maintained sufficient generating capacity to meet customer needs,
and bulk transactions were initially used only to support extreme demands or equipment
outages.
4.7.1.3 Distribution
Low-voltage distribution systems that deliver electricity to customers comprise
integrated networks of smaller wires and substations that take the higher voltage and step it
down to lower levels to match customers' needs.
The distribution system is the classic example of a natural monopoly because it is not
practical to have more than one set of lines running through neighborhoods or from the curb
to the house.
4.7.2 Cost of Production
Table 4-24 shows total industry expenditures by production activities. Generation
accounts for approximately 75.6 percent of the cost of delivered electric power in 1996.
Transmission and distribution accounted for 2.5 percent and 5.6 percent, respectively.
4-32
-------
Table 4-24. Total Expenditures in 1996 ($103)
Utility
Ownership
Investor-owned
Publicly owned
Federal
Cooperatives
Generation
80,891,644
12,495,324
3,685,719
15,105,404
112,178,091
75.6%
148,370,552
Transmission
2,216,113
840,931
327,443
338,625
3,723,112
2.5%
Distribution
6,124,443
1,017,646
1,435
1,133,984
8,277,508
5.6%
Customer
Accounts
and Sales
6,204,229
486,195
55,536
564,887
7,310,847
4.9%
Administration
and General
Expenses
13,820,059
1,360,111
443,809
1,257,015
16,880,994
11.4%
Sources: U.S. Department of Energy, Energy Information Administration (EIA). 1998b. Financial Statistics of
Major Publicly Owned Electric Utilities, 1997. Washington, DC: U.S. Department of Energy.
U.S. Department of Energy, Energy Information Administration (EIA). 1997. Financial Statistics of
Major U.S. Investor-Owned Electric Utilities, 1996. Washington, DC: U.S. Department of Energy.
Customer accounts and sales and administrative costs accounted for the remaining 16.3
percent of the cost of delivered power.
4.7.3 Organization of the Industry
Because the restructuring plans and time tables are made at the state level, the issues
of asset ownership and control throughout the current supply chain in the electric power
industry vary from state to state. However, the activities conducted throughout the supply
chain are generally the same. This section focuses on the generation segment of the market
because all the boilers affected by the regulation are involved in generation.
As part of deregulation, the transmission and distribution of electricity are being
separated from the business of generating electricity, and a new competitive market in
electricity generation is evolving. As power generators prepare for the competitive market,
the share of electricity generation attributed to nonutilities and utilities is shifting.
More than 7,000 electricity suppliers currently operate in the U.S. market. As shown
in Table 4-25, approximately 42 percent of suppliers are utilities and 58 percent are
nonutilities. Utilities include investor-owned, cooperatives, and municipal systems. Of the
approximately 3,100 utilities operating in the United States, only about 700 generate electric
power. The majority of utilities distribute electricity that they have purchased from power
generators via their own distribution systems.
4-33
-------
Table 4-25. Number of Electricity Suppliers in 1999
Electricity Suppliers Number Percent
Utilities 3,124 42%
Investor-owned utilities 222
Cooperatives 875
Municipal systems 1,885
Public power districts 73
State projects 55
Federal agencies 14
Nonutilities 4,247 58%
Nonutilities (excluding EWGs) 4,103
Exempt wholesale generators 144
Total 7,371 100%
Source: U.S. Department of Energy, Energy Information Administration (EIA). 1999b. The Changing
Structure of the Electric Power Industry 1999: Mergers and Other Corporate Combinations.
Washington, DC: U.S. Department of Energy.
Utility and nonutility generators produced a total of 3,369 billion kWh in 1995.
Although utilities generate the vast majority of electricity produced in the United States,
nonutility generators are quickly eroding utilities' shares of the market. Nonutility
generators include private entities that generate power for their own use or to sell to utilities
or other end users. Between 1985 and 1995, nonutility generation increased from 98 billion
kWh (3.8 percent of total generation) to 374 billion kWh (11.1 percent). Figure 4-2
illustrates this shift in the share of utility and nonutility generation.
4.7.3.1 Utilities
There are four categories of utilities: investor-owned utilities (lOUs), publicly owned
utilities, cooperative utilities, and federal utilities. Of the four, only lOUs always generate
electricity.
lOUs are increasingly selling off generation assets to nonutilities or converting those
assets into nonutilities (Haltmaier, 1998). To prepare for the competitive market, lOUs have
been lowering their operating costs, merging, and diversifying into nonutility businesses.
4-34
-------
Utilities
5%
9%
7%
Shares of Total
Utility Generation
Investor-Owned
Shares of Total
Utility Generation
1988 Generation
Utility 93%
Nonutility 7%
Nonutilities
1998 Generation
Utility 89%
Nonutility 11%
Shares of Total
Nonutility Generation
Shares of Total
Nonutility Generation
a Includes facilities classified in more than one of the following FERC designated categories: cogenerator QF, small power
producer QF, or exempt wholesale generator.
Cogen = Cogenerator.
EWG = Exempt wholesale generator.
Other Non-QF = Nocogenerator Non-QF.
QF = Qualifying facility.
SPP = Small power producer.
Note: Sum of components may not equal total due to independent rounding. Classes for nonutility generation are
determined by the class of each generating unit.
Sources: Utility data: U.S. Department of Energy, Energy Information Administration (EIA). 1996. Electric Power
Annual 1995. Volumes I and II. DOE/EIA-0348(95)/1. Washington, DC: U.S. Department of Energy; Table 8
(and previous issues); 1985 nonutility data: Shares of generation estimated by EIA; total generation from Edison
Electric Institute (EEI). 1998. Statistical Yearbook of the Electric Utility Industry 1998. November.
Washington, DC; 1995 nonutility data: U.S. Department of Energy, Energy Information Administration (EIA).
1996. Electric Power Annual 1995. Volumes I and II. DOE/EIA-0348(95)/1. Washington, DC: U.S.
Department of Energy.
Figure 4-2. Utility and Nonutility Generation and Shares by Class, 1988 and 1998
4-35
-------
In 1995, utilities generated 89 percent of electricity, a decrease from 96 percent in
1985. lOUs generate the majority of the electricity produced in the United States. lOUs are
either individual corporations or a holding company, in which a parent company operates one
or more utilities integrated with one another. lOUs account for approximately three-quarters
of utility generation, a percentage that held constant between 1985 and 1995.
Many states, municipalities, and other government organizations also own and
operate utilities, although the majority do not generate electricity. Those that do generate
electricity operate capacity to supply some or all of their customers' needs. They tend to be
small, localized outfits and can be found in 47 states. These publicly owned utilities
accounted for about one-tenth of utility generation in 1985 and 1995. In a deregulated
market, these generators may be in direct competition with other utilities to service their
market.
Rural electric cooperatives are formed and owned by groups of residents in rural
areas to supply power to those areas. Cooperatives generally purchase from other utilities
the energy that they sell to customers, but some generate their own power. Cooperatives
only produced 5 percent of utility generation in 1985 and only 6 percent in 1995.
Utilities owned by the federal government accounted for about one-tenth of
generation in both 1985 and 1995. The federal government operated a small number of large
utilities in 1995 that supplied power to large industrial consumers or federal installations.
The Tennessee Valley Authority is an example of a federal utility.
4.7.3.2 Nonutilities
Nonutilities are private entities that generate power for their own use or to sell to
utilities or other establishments. Nonutilities are usually operated at mines and
manufacturing facilities, such as chemical plants and paper mills, or are operated by electric
and gas service companies (DOE, EIA, 1998a). More than 4,200 nonutilities operate in the
United States.
4.7.4 Demand Side of the Industry
4.7.4.1 Electricity Consumption
This section analyzes the growth projections for electricity consumption as well as
the price elasticity of demand for electricity. Growth in electricity consumption has
traditionally paralleled gross domestic product growth. However, improved energy
efficiency of electrical equipment, such as high-efficiency motors, has slowed demand
4-36
-------
growth over the past few decades. The magnitude of the relationship has been decreasing
over time, from growth of 7 percent per year in the 1960s down to 1 percent in the 1980s. As
a result, determining what the future growth will be is difficult, although it is expected to be
positive (DOE, EIA, 1999a). Table 4-26 shows consumption by sector of the economy over
the past 10 years. The table shows that since 1989 electricity sales have increased at least 10
percent in all four sectors. The commercial sector has experienced the largest increase,
followed by residential consumption.
Table 4-26. U.S. Electric Utility Retail Sales of Electricity by Sector, 1989 Through
1998 (106 kWh)
Period
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
Percentage change
1989-1998
Residential
905,525
924,019
955,417
935,939
994,781
1,008,482
1,042,501
1,082,491
1,075,767
1,124,004
19%
Commercial
725,861
751,027
765,664
761,271
794,573
820,269
862,685
887,425
928,440
948,904
24%
Industrial
925,659
945,522
946,583
972,714
977,164
1,007,981
1,012,693
1,030,356
1,032,653
1,047,346
12%
Other"
89,765
91,988
94,339
93,442
94,944
97,830
95,407
97,539
102,901
99,868
10%
All Sectors
2,646,809
2,712,555
2,762,003
2,763,365
2,861,462
2,934,563
3,013,287
3,097,810
3,139,761
3,220,121
18%
" Includes public street and highway lighting, other sales to public authorities, sales to railroads and railways,
and interdepartmental sales.
Sources: U.S. Department of Energy, Energy Information Administration (EIA). 1999d. Electric Power
Annual 1998. Volumes I and II. Washington, DC: U.S. Department of Energy.
U.S. Department of Energy, Energy Information Administration (EIA). 1996. Electric Power Annual
1995. Volumes I and II. Washington, DC: U.S. Department of Energy.
In the future, residential demand is expected to be at the forefront of increased
electricity consumption. Between 1997 and 2020, residential demand is expected to increase
at 1.6 percent annually. Commercial growth in demand is expected to be approximately
1.4 percent, while industry is expected to increase demand by 1.1 percent (DOE, EIA,
1999a). Figure 4-3 shows the annual electricity sales by sector from 1970 with projections
through 2020.
4-37
-------
The literature suggests that electricity consumption is relatively price inelastic.
Consumers are generally unable or unwilling to forego a large amount of consumption as the
IS 70
.3020
soso
Figure 4-3. Annual Electricity Sales by Sector
price increases. Numerous studies have investigated the short-run elasticity of demand for
electricity. Overall, the studies suggest that, for a 1 percent increase in the price of
electricity, demand will decrease by 0.15 percent. However, as Table 4-27 shows, elasticities
vary greatly, depending on the demand characteristics of end users and the price structure.
Demand elasticities are estimated to range from a -0.05 percent elasticity of demand for a
"fiat rates" case (i.e., no time-of-use assumption) up to a -0.50 percent demand elasticity for
a "high consumer response" case (DOE, EIA, 1999c).
4.7.4.2 Trends in the Electricity Market
Beginning in the latter part of the 19th century and continuing for about 100 years,
the prevailing view of policymakers and the public was that the government should use its
power to require or prescribe the economic behavior of "natural monopolies" such as electric
utilities. The traditional argument is that it does not make economic sense for there to be
more than one supplier—running two sets of wires from generating facilities to end users is
more costly than one set. However, since monopoly supply is not generally regarded as
likely
4-38
-------
Table 4-27. Key Parameters in the Cases
Case Name
AEO97 Reference Case
No Competition
Flat Rates
(no time-of-use rates)
Moderate Consumer
Response
High Consumer Response
High Efficiency
No Capacity Additions
High Gas Price
Low Gas Price
High Value of Reliability
HalfO&M
Intense Competition
Cost Reduction
and Efficiency
Improvements
AEO97 Reference
Case
No change from
1995
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
Increased cost
savings and
efficiencies
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
Key
Assumptions
Short-Run
Elasticity
of Demand Natural Gas
(Percent) Prices
—
—
-0.05
-0.15
-0.50
-0.15
-0.15
-0.15
-0.15
-0.15
-0.15
-0.15
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
AEO97LowOil
and Gas Supply
Technology Case
AEO97 High Oil
and Gas Supply
Technology Case
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
AEO97 Reference
Case
Capacity
Additions
As needed
to meet demand
As needed
to meet demand
As needed
to meet demand
As needed
to meet demand
As needed
to meet demand
As needed
to meet demand
Not allowed
As needed
to meet demand
As needed
to meet demand
As needed
to meet demand
As needed
to meet demand
As needed to meet
demand
— = not applicable.
Source: U.S. Department of Energy, Energy Information Administration (EIA), Office of Integrated Analysis
and Forecasting. "Competitive Electricity Price Projections."
. As obtained on November 15, 1999c.
4-39
-------
to provide a socially optimal allocation of resources, regulation of rates and other economic
variables was seen as a necessary feature of the system.
Beginning in the 1970s, the public policy view shifted against traditional regulatory
approaches and in favor of deregulation for many important industries including
transportation, communications, finance, and energy. The major drivers for deregulation of
electric power included the following:
• existence of rate differentials across regions offering the promise of benefits from
more efficient use of existing generation resources if the power can be transmitted
across larger geographic areas than was typical in the era of industry regulation;
• the erosion of economies of scale in generation with advances in combustion
turbine technology;
• complexity of providing a regulated industry with the incentives to make socially
efficient investment choices;
• difficulty of providing a responsive regulatory process that can quickly adjust
rates and conditions of service in response to changing technological and market
conditions; and
• complexity of monitoring utilities' cost of service and establishing cost-based
rates for various customer classes that promote economic efficiency while at the
same time addressing equity concerns of regulatory commissions.
Viewed from one perspective, not much changes in the electric industry with
restructuring. The same functions are being performed, essentially the same resources are
being used, and in a broad sense the same reliability criteria are being met. In other ways,
the very nature of restructuring, the harnessing of competitive forces to perform a previously
regulated function, changes almost everything. Each provider and each function become
separate competitive entities that must be judged on their own.
This move to market-based provision of generation services is not matched on the
transmission and distribution side. Network interactions on AC transmission systems have
made it impossible to have separate transmission paths compete. Hence, transmission and
distribution remain regulated. Transmission and generation heavily interact, however, and
transmission congestion can prevent specific generation from getting to market.
Transmission expansion planning becomes an open process with many interested parties.
This open process, coupled with frequent public opposition to transmission expansion, slows
4-40
-------
transmission enhancement. The net result is greatly increased pressure on the transmission
system.
Restructuring of the electric power industry could result in any one of several
possible market structures. In fact, different parts of the country will probably use different
structures, as the current trend indicates. The eventual structure may be dominated by a
power exchange, bilateral contracts, or a combination. A strong Regional Transmission
Organization (RTO) may operate in the area, or a vertically integrated utility may continue to
operate a control area. In any case, several important characteristics will change:
• Commercial provision of generation-based services (e.g., energy, regulation, load
following, voltage control, contingency reserves, backup supply) will replace
regulated service provision. This drastically changes how the service provider is
assessed.
• Individual transactions will replace aggregated supply meeting aggregated
demand. It will be necessary to continuously assess each individual's
performance.
• Transaction sizes will shrink. Instead of dealing only in hundreds and thousands
of MW, it will be necessary to accommodate transactions of a few MW and less.
• Supply flexibility will greatly increase. Instead of services coming from a fixed
fleet of generators, service provision will change dynamically among many
potential suppliers as market conditions change.
4-41
-------
SECTION 5
ECONOMIC ANALYSIS METHODOLOGY
The final rule to control emissions of HAPs from industrial, commercial, and
institutional boilers and process heaters will affect almost all sectors of the U.S. economy.
Several markets will bear the direct compliance costs. In addition, sectors that consume
energy will also bear indirect costs through higher prices for energy. Finally, consumers of
goods and services will experience impacts from higher market prices.
This section presents the methodology for analyzing the economic impacts of the
NESHAP. This economic analysis provides the economic data and supporting information
needed by EPA to support its regulatory determination. The methodology is based on
microeconomic theory and the methods developed for earlier EPA studies. These methods
are tailored to and extended for this analysis, as appropriate, to meet EPA's requirements for
an EIA of controls placed on boilers and process heaters.
This methodology section includes background information on typical economic
modeling approaches, the conceptual approach selected for this EIA, and an overview of the
computerized market model used in the analysis with emphasis on the links between energy
markets and the markets for goods and services. Appendix A includes a description of the
model's baseline data set and specification.
5.1 Background on Economic Modeling Approaches
In general, the EIA methodology needs to allow EPA to consider the effects of the
different regulatory alternatives. Several types of economic impact modeling approaches
have been developed to support regulatory development. These approaches can be viewed as
varying along two modeling dimensions:
• the scope of economic decisionmaking accounted for in the model and
• the scope of interaction between different segments of the economy.
Each of these dimensions was considered in determining the approach for this study. The
advantages and disadvantages of different modeling approaches are discussed below.
5-1
-------
5.1.1 Modeling Dimension 1: Scope of Economic Decisionmaking
Models incorporating different levels of economic decisionmaking can generally be
categorized as with behavior responses and without behavior responses (accounting
approach). Table 5-1 provides a brief comparison of the two approaches. The nonbehavioral
approach essentially holds fixed all interactions between facility production and market
forces. It assumes that firms absorb all control costs and consumers do not face any of the
costs of regulation. Typically, engineering control costs are weighted by the number of
affected units to develop "engineering" estimates of the total annualized costs. These costs
are then compared to company or industry sales to determine the regulation's impact.
Table 5-1. Comparison of Modeling Approaches
EIA With Behavioral Responses
• Incorporates control costs into production function
• Includes change in quantity produced
• Includes change in market price
• Estimates impacts for
-------
regulation, consumers are typically faced with changes in prices that cause them to alter the
quantity that they are willing to purchase. In essence, this approach models the expected
reallocation of society's resources in response to a regulation. The changes in price and
production from the market-level impacts are used to estimate the distribution of social costs
between consumers and producers.
5.1.2 Modeling Dimension 2: Interaction Between Economic Sectors
Because of the large number of markets potentially affected by the regulation on
boilers and process heaters, an issue arises concerning the level of sectoral interaction to
model. In the broadest sense, all markets are directly or indirectly linked in the economy;
thus, the regulation affects all commodities and markets to some extent. For example,
controls on boilers and process heaters may indirectly affect almost all markets for goods and
services to some extent because the cost of fuel (an input in the provision of most goods and
services) is likely to increase with the regulation in effect. However, the impact of rising fuel
prices will differ greatly between different markets depending on how important fuel is as an
input in that market.
The appropriate level of market interactions to be included in the EIA is determined
by the scope of the regulation across industries and the ability of affected firms to pass along
the regulatory costs in the form of higher prices. Alternative approaches for modeling
interactions between economic sectors can generally be divided into three groups:
• Partial equilibrium model: Individual markets are modeled in isolation. The only
factor affecting the market is the cost of the regulation on facilities in the industry
being modeled.
• General equilibrium model: All sectors of the economy are modeled together.
General equilibrium models operationalize neoclassical microeconomic theory by
modeling not only the direct effects of control costs, but also potential input
substitution effects, changes in production levels associated with changes in
market prices across all sectors, and the associated changes in welfare
economywide. A disadvantage of general equilibrium modeling is that substantial
time and resources are required to develop a new model or tailor an existing
model for analyzing regulatory alternatives.
• Multiple-market partial equilibrium model: A subset of related markets are
modeled together, with intersectoral linkages explicitly specified. To account for
the relationships and links between different markets without employing a full
general equilibrium model, analysts can use an integrated partial equilibrium
5-3
-------
model. The multiple-market partial equilibrium approach represents an
intermediate step between a simple, single-market partial equilibrium approach
and a full general equilibrium approach. This approach involves identifying and
modeling the most significant subset of market interactions using an integrated
partial equilibrium framework. In effect, the modeling technique is to link a series
of standard partial equilibrium models by specifying the interactions between
supply functions and then solving for prices and quantities across all markets
simultaneously. In instances where separate markets are closely related and there
are strong interconnections, there are significant advantages to estimating market
adjustments in different markets simultaneously using an integrated market
modeling approach.
5.2 Selected Modeling Approach for Boilers and Process Heaters Analysis
To conduct the analysis for the boilers and process heaters MACT, the Agency used a
market modeling approach that incorporates behavioral responses in a multiple-market partial
equilibrium model as described above. This approach allows for a more realistic assessment
of the distribution of impacts across different groups than the nonbehavioral approach, which
may be especially important in accurately assessing the impacts of a significant rule affecting
numerous industries. Because of the size and complexity of this regulation, it is important to
use a behavioral model to examine the distribution of costs across society. Because the
regulations on boilers and process heaters primarily affect energy costs, an input into many
production processes, complex market interactions need to be captured to provide an accurate
picture of the distribution of regulatory costs. Because of the large number of affected
industries under this MACT, an appropriate model should include multiple markets and the
interactions between them. Multiple-market partial equilibrium analysis provides a
manageable approach to incorporate interactions between energy markets and final product
markets into the EIA to accurately estimate the regulation's impact.
The model used for this analysis includes energy, agriculture, manufacturing, mining,
commercial, and transportation markets affected by the controls placed on boilers and process
heaters.1 The energy markets are divided into natural gas, petroleum products, coal, and
electricity. The residential sector is treated as a single representative demander in the energy
markets.
'These markets are defined at the two- and three-digit NAICS code level. This allows for a fairly disaggregated
examination of the regulation's impact on producers. However, if the costs of the regulation are
concentrated on a particular subset of one of these markets, then treating the cost as if it fell on the entire
NAICS code may still underestimate the impacts on the subset of producers affected by the regulation.
5-4
-------
Figure 5-1 presents an overview of the key market linkages included in the economic
impact model used for analyzing the boilers and process heaters MACT. The analysis'
emphasis is on the energy supply chain and the consumption of energy by producers of goods
and services. The industries most directly affected by the boilers and process heaters MACT
are the electric power industry, chemical industry and pulp and paper industry. However,
changes in the equilibrium prices and quantities of energy and goods and services affect all
sectors of the economy. (See Figure 5-1.) This analysis explicitly models the linkages
between these market segments to capture both the direct costs of compliance and the indirect
costs due to changes in prices. For example, production costs will increase for chemical
companies using boilers and process heaters as a result of the capital investments and
monitoring costs, as well as the resulting increase in the price of electricity used as an energy
input in the production process.
The economic model also captures behavioral changes of producers of goods and
services that feedback into the energy markets. Changes in production levels and fuel
switching in the manufacturing process affect the demand for Btus in fuel markets. The
change in output is determined by the size of the cost increase per Btu (typically variable cost
per output), the facility's production function (slope of supply curve), and the demand
characteristics of the facility's downstream market (other market suppliers and market
demanders). For example, if consumers' demand for a product is not very sensitive to price,
then producers can pass the majority of the cost of the regulation through to consumers and
output may not change appreciably. However, if only a small proportion of market output is
produced by producers affected by the regulation, then competition will prevent the affected
producers from raising their prices significantly.
One possible feedback pathway that this analysis does not model is technical changes
in the manufacturing process. For example, if the cost of Btus increases, a facility may use
measures to increase manufacturing efficiency or capture waste heat. Facilities could also
possibly change the input mix that they use, substituting other inputs for fuel. These facility-
level responses will also act to reduce pollution, but including these responses is beyond the
scope of this analysis.
5.2.1 Directly Affected Markets
Markets where boilers and process heaters are used as an input to production are
considered to be directly affected. As outlined in Section 2, facilities using several types of
boilers or process heaters will be required to add controls. In addition, a larger population of
5-5
-------
•a
&
u
&
u
•a
HI
Figure 5-1. Links Between Energy and Goods and Services Markets
5-6
-------
boilers and process heaters will incur monitoring costs to comply with the regulation.
Therefore, the regulation will increase their production costs and cause these directly affected
firms to reduce the quantity that they are willing to supply at any given price.
5.2.1.1 Electricity Market
Boilers are used to generate power throughout the electric power industry. Even
though utility boilers are not covered under this regulation, the Agency estimates over 300
industrial, commercial, and institutional boilers involved in providing electric services
(SIC 4911/NAICS22111) will be affected. Most of these are owned by municipal electric
service providers.
For this study, the electricity market was modeled as a nationally competitive market.
The direct costs of compliance on affected boilers lead to an upward shift in the total market
supply for electricity. Figure 5-2 illustrates the shifts in the supply curve for a representative
energy market. In addition to the direct costs, the market for electricity will also be indirectly
affected through changes in fuel prices. Electricity generators are extremely large consumers
of coal, natural gas, and petroleum products. For example, some of the impact of control
costs on the petroleum industry will be on the electricity industry in the form of higher prices.
Indirect costs will also lead to an upward shift in the supply curve.
The demand for electricity is derived by aggregating across the goods and services
markets and the residential sector. Because of direct compliance costs on the goods and
services markets, the demand curve for electricity will shift downward. Therefore, it is
ambiguous whether equilibrium quantity will rise or fall. The changes in the price and
quantity are determined by the relative magnitude of the shifts in the price elasticities of the
supply and demand curves.
5.2.1.2 Petroleum Market
Control costs associated with boilers and process heaters will increase the cost of
refining petroleum products. The supply curve for petroleum products will shift upward by
the proportional increase in total production costs caused by the control costs on boilers and
process heaters. For petroleum products, a single composite product was used to model
market adjustment because boilers and process heaters are used throughout the refinement
process, from distillation to reformulation. As a result, assigning costs to specific end
products, such as fuel oil #2 or reformulated gasoline, is difficult. The use of a composite
5-7
-------
sn
Qio
Qn
c
°20
Q20
021
STO
STI
QTO
On
Producers b«* rin|fa
-------
to model the supply and demand. Changes in production levels and fuel switching due to the
regulation's impact on the price of Btus were then linked back into the energy markets.
The impact of the regulation on producers in these sectors was modeled as an increase
in the cost of Btus used in the production process. In this context, Btus refer to the generic
energy requirements used to generate process heat, process steam, or shaft power.
Compliance costs associated with the regulation will increase the cost of Btu production in
the manufacturing sectors. The cost of Btu production for industry increases because of both
direct control costs on boilers and process heaters owned by manufacturers, and increases in
the price of fuels. Because Btus are an input into the production process, these price
increases lead to an upward shift in the facility (and industry) supply curves as shown in
Figure 5-2, leading to a change in the equilibrium market price and quantity.
The changes in equilibrium supply and demand in each market are modeled to
estimate the regulation's impact on each sector. In a perfectly competitive market, the point
where supply equals demand determines the market price and quantity, so market price and
quantity are determined by solving the model for the price where the quantity supplied and
the quantity demanded are equal. The size of the regulation-induced shifts in the supply
curve is a function of the total direct control costs associated with boilers and process heaters
and the indirect fuel costs (determined by the change in fuel price and intensity of use) in
each goods and services market. The proportional shift in the supply curve is determined by
the ratio of total control costs (both direct and indirect) to total revenue.
This impact on the price of Btus facing industrial users feeds back to the fuel market
in two ways (see Figure 5-3). The first is through the company's input decision concerning
the fuel(s) that will be used for its manufacturing process. As the cost of Btus increases,
firms may switch fuels and/or change production processes to increase energy efficiency and
reduce the number of Btus required per unit of output. Fuel switching impacts were modeled
using cross-price elasticities of demand between energy sources. For example, a cross-price
elasticity of demand between natural gas and electricity of 0.5 implies that a 1 percent
increase in the price of electricity will lead to a 0.5 percent increase in the demand for natural
gas. Own-price elasticities of demand are used to estimate the change in the use of fuel by
demanders. For example, a demand elasticity of-0.175 for electricity implies that a 1
percent increase in the price of electricity will lead to a 0.175 percent decrease in the quantity
of electricity demanded.
5-9
-------
Compliance
Costs
Fu
Mar
j
d „
kets ysfu r
L
if
J
Btu
Production
Decision
net Use
A&Btu r
l
Production
Decision
Oirfpuf
Output
* * Market
Figure 5-3. Fuel Market Interactions with Facility-Level Production Decisions
The second feedback pathway to the energy markets is through the facility's change in
output. Because Btus are an input into the production process, energy price increases lead to
an upward shift in the facility supply curves (not modeled individually). This leads to an
upward shift in the industry supply curve when the shifts at the facility level are aggregated
across facilities. A shift in the industry supply curve leads to a change in the equilibrium
market price and quantity. In a perfectly competitive market, the point where supply equals
demand determines the new market price and quantity. The Agency modeled the feedback
into the energy market by assuming that the percentage change in output in the manufacturing
sectors translates into a equivalent percentage change in the demand for energy (Btus). This
implies that there are constant returns to scale from energy inputs in the manufacturing
process over the relevant range of output and time period of analysis. This is an appropriate
assumption for this analysis because the output changes in these sectors being modeled are
relatively small (always less than 1 percent) and reflect short-run production decisions.2
The Agency assumed that the demand curves for goods and services in all sectors are
unchanged by the regulation. However, because the demand function quantifies the change
in quantity demanded in response to a change in price, the baseline demand conditions are
Long-run production decisions of fuel switching and increased energy efficiency are captured by the cross- and
own-price elasticities in the energy markets.
5-10
-------
important in determining the regulation's impact. The key demand parameters are the
elasticities of demand with respect to changes in the price of goods and services. For these
markets, a "reasonable" range of elasticity values is assigned based on estimates from similar
commodities. Because price changes are anticipated to be small, the point elasticities at the
original price and quantity should be applicable throughout the relevant range of prices and
quantities examined in this model.
5.2.2 In directly Affected Markets
In addition to the many markets that are directly affected by the regulation on boilers
and process heaters, some markets feel the regulation's impacts despite having no direct costs
resulting from the regulation. Firms in these markets generally face changes in the price of
energy that affect their production decisions.
5.2.2.1 Market for Coal
The coal market is not directly affected by the regulation, but it has the potential to be
significantly affected through indirect costs. Although boilers and process heaters are not
commonly used in the production or transportation of coal, the supply of coal will be affected
by the price of energy used in coal production. However, the indirect impacts on coal
production costs are relatively small compared to the direct impacts on the production costs
in the electricity and petroleum markets; thus, the "relative" price of coal (per Btu) will
decrease compared with other energy sources.
The demand for coal from the industrial sectors will be affected by differences in
compliance costs by fuel type applied to boilers and process heaters in the industrial sectors.
Because compliance costs are high for coal-fired units, manufacturers will switch away from
coal units toward natural gas units with lower compliance costs. However, the overall impact
on the demand for coal is ambiguous because the relative increase in the cost of producing
Btus by burning coal will be offset by the relative decrease in the price of coal. Similarly, the
demand for coal by utility generators will be affected through changes in the relative price of
alternative (noncoal) energy sources and direct costs on coal boilers.
5.2.2.2 Natural Gas Market
The natural gas market is included in the economic model to complete coverage of
the energy markets. EPA projects that there are no direct and minimal indirect impacts on
the production costs of natural gas. However, the demand for natural gas will increase
5-11
-------
because of the relative decrease in the price of natural gas and the lower relative compliance
costs for gas-fired boilers and process heaters.
5.2.2.3 Goods and Services Markets
Some goods and services markets do not include any boilers or process heaters and
are therefore not directly affected by the regulation. However, these markets will still be
affected indirectly because of the changes in energy prices that they will face following the
regulation. There will be a tendency for these users to shift away from electricity and
petroleum products and towards natural gas and coal.
5.2.2.4 Impact on Residential Sector
The residential sector does not bear any direct costs associated with the regulation
because this sector does not own boilers or process heaters. However, they bear indirect
costs due to price increases. The residential sector is a significant consumer of electricity,
natural gas, and petroleum products used for heating, cooling, and lighting, as well as many
other end uses. The change in the quantity of energy demanded by these consumers in
response to changes in energy prices is modeled as a single demand curve parameterized by
demand elasticities for residential consumers obtained from the literature.
5.3 Operationalizing the Economic Impact Model
Figure 5-4 illustrates the linkages used to operationalize the estimation of economic
impacts associated with the compliance costs. Compliance costs placed on boilers and
process heaters shift the supply curve for electricity and petroleum products. Adjustments in
the electricity and petroleum energy markets determine the share of the cost increases that
producers (electric service providers and petroleum companies) and consumers (product
manufacturers, commercial business, and residential households) bear.
The supply and demand relationships between the energy markets are fully modeled.
For example, changes in electricity production feed back and affect the demand for coal,
natural gas and petroleum products. Similar changes in refinery production affect the
petroleum industry's demand for electricity.
5-12
-------
-a
1
I
B <;
tS
a
O
s
•I
3
tr
B
•u '.0
w
!
•s
1
Figure 5-4. Operationalizing the Estimation of Economic Impact
5-13
-------
Manufacturers experience supply curve shifts due to control costs on affected boilers
and process heaters they operate and changes in prices for natural gas, petroleum, electricity,
and coal. The share of these costs borne by producers and consumers is determined by the
new equilibrium price and quantity in the goods and services markets. Changes in
manufacturers' Btu demands due to fuel switching and changes in production levels feed
back into the energy markets.
Adjustments in price and quantity in all markets occur simultaneously. A computer
model was used to numerically simulate market adjustments by iterating over commodity
prices until equilibrium is reached (i.e., until the quantity supplied equals the quantity
demanded in all markets being modeled). Using the results provided by the model, economic
impacts of the regulation (changes in consumer and producer surplus) were estimated for all
sectors of the economy being modeled.
5.3.1 Computer Model
The computer model comprises a series of computer spreadsheet modules. The
modules integrate the engineering cost inputs and the market-level adjustment parameters to
estimate the regulation's impact on the price and quantity in each market being analyzed. At
the heart of the model is a market-clearing algorithm that compares the total quantity
supplied to the total quantity demanded for each market commodity.
Current prices and production levels are used to calibrate the baseline scenario
(without regulation) for the model. Then, the compliance costs associated with the regulation
are introduced as a "shock" to the system, and the supply and demand for market
commodities are allowed to adjust to account for the increased production costs resulting
from the regulation. Using an iterative process, if the supply does not equal demand in all
markets, a new set of prices is "called out" and sent back to producers and consumers to
"ask" what quantities they would supply and demand based on these new prices. This
technique is referred to as an auctioneer approach because new prices are continually called
out until an equilibrium set of prices is determined (i.e., where supply equals demand for all
markets).
Supply and demand quantities are computed at each price iteration. The market
supply for each market is obtained by using a mathematical specification of the supply
function, and the key parameter is the point elasticity of supply at the baseline condition.
Table 5-2 lists the supply elasticities for the markets used in the model.
5-14
-------
Table 5-2. Supply and Demand Elasticities
Supply Elasticities
Petroleum 0.58b
NaturalGas 0.41b
Electricity 0.7 5C
Coal 1.00b
Demand Elasticities
Industrial Residential" Transportation
Derived -0.28 Derived
Derived -0.26 Derived
Derived -0.23 Derived
Derived -0.26 Derived
NAICS Description Supplyd
311 Food
312 Beverage and Tobacco
313 Textile Mills
3 1 4 Textile Product Mills
315 Apparel
0
Products 0
0
0
0
316 Leather and Allied Products 0
321 Wood Products
322 Paper
0
1
323 Printing and Related Support 0
325 Chemicals
0
326 Plastics and Rubber Products 0
327 Nonmetallic Mineral Products 0
331 Primary Metals
3
332 Fabricated Metal Products 0
333 Machinery
0
334 Computer and Electronic Products 0
335 Electrical Equipment, Appliances, and Components 0
336 Transportation Equipment 0
337 Furniture and Related Products 0
339 Miscellaneous
1 1 Agricultural Sector
0
0
75C
75C
37e
37e
75C
75C
75d
20C
75C
75C
75C
75C
50f
75C
75C
75C
75C
75C
75C
75C
75C
Commercial
Derived
Derived
Derived
Derived
Demandd
-0.30
-1.30
-0.85e
-0.85e
-1.80
-1.20
-0.20
-1.09
-1.80
-1.50
-1.80
-0.90
-0.80
-0.20
-0.50
-0.30
-0.50
-1.00C
-3.40
-0.60
-1.80
(continued)
5-15
-------
Table 5-2. Supply and Demand Elasticities (continued)
NAICS
23
21
48
Commercia
1
Description
Construction Sector
Other Mining Sector
Transportation
Commercial
Supply"
0.75C
0.43
0.75C
0.75C
Demand
-1.00C
-0.30
-0.70
-1.00C
a U.S. Department of Energy, Energy Information Administration (EIA). "Issues in Midterm Analysis and
Forecasting 1999—Table 1." . As obtained on May 8,
2000a.
b Dahl, Carol A., and Thomas E. Duggan. 1996. "U.S. Energy Product Supply Elasticities: A Survey and
Application to the U.S. Oil Market." Resource and Energy Economics\S:243-263.
c Assumed value.
d E.H. Pechan & Associates, Inc. 1997. Qualitative Market Impact Analysis for Implementation of the
Selected Ozone and PM NAAQS. Appendix B. Prepared for the U.S. Environmental Protection Agency.
e Warfield, et al. 2001. "Multifiber Arrangement Phaseout: Implications for the U.S.
Fibers/Textiles/Fabricated Products Complex." www.fibronet.com.tw/mirron/ncs/9312/mar.html> As
obtained September 19, 2001.
f U.S. International Trade Commission (USITC). November 21, 2001. Memorandum to the Commission from
Craig Thomsen, John Giamalua, John Benedetto, Joshua Levy, International Economists. Investigation No.
TA-201 -73: STEEL-Remedy Memorandum.
The demand curves for the energy markets are the sum of demand responses across all
markets. The demand for energy in the manufacturing sectors is a derived demand calculated
using baseline energy usage and changes associated with fuel switching and changes in
output levels. Similarly, the energy demand in residential sectors is obtained through
mathematical specification of a demand function (see Appendix A).
The demand for goods and service in the two- and three-digit NAICS code
manufacturing sectors is obtained by using a mathematical specification of the demand
function. Table 5-2 lists the demand elasticities for the markets used in the model.
EPA modeled fuel switching using secondary data developed by the U.S. Department
of Energy for the National Energy Modeling System (NEMS). Table 5-3 contains fuel price
elasticities of demand for electricity, natural gas, petroleum products, and coal. The diagonal
elements in the table represent own-price elasticities. For example, the table indicates that
for steam coal, a 1 percent change in the price of coal will lead to a 0.499 percent decrease in
the use of coal. The off diagonal elements are cross-price elasticities and indicate fuel
switching propensities. For example, for steam coal, the second column indicates that a
5-16
-------
Table 5-3. Fuel Price Elasticities
Own and Cross Elasticities
Inputs
Electricity
Natural Gas
Steam Coal
Residual
Distillate
Electricity
-0.074
0.496
0.021
0.236
0.247
Natural Gas
0.092
-0.229
0.061
0.036
0.002
Coal
0.605
1.087
-0.499
0.650
0.578
Residual
0.080
0.346
0.151
-0.587
0.044
Distillate
0.017
0.014
0.023
0.012
-0.055
Source: U.S. Department of Energy, Energy Information Administration (EIA). January 2000 b. Model
Documentation Report: Industrial Sector Demand Module of the National Energy Modeling System.
DOE/EIA-M064(2000). Washington, DC: U.S. Department of Energy.
1 percent increase in the price of coal will lead to a 0.061 percent increase in the use of
natural gas.
5.5.2 Calculating Changes in Social Welfare
The boilers and process heaters MACT will impact almost every sector of the
economy, either directly through control costs or indirectly through changes in the price of
energy and final products. For example, a share of control costs that originate in the energy
markets is passed through the goods and services markets and borne by both the producers
and consumers of their products. To estimate the total change in social welfare without
double-counting impacts across the linked partial equilibrium markets being modeled, EPA
quantified social welfare changes for the following categories:
• change in producer surplus in the energy markets;
• change in producer surplus in the goods and services markets;
• change in consumer surplus in the goods and services markets; and
• change in consumer surplus in the residential sector.
Figure 5-5 illustrates the change in producer and consumer surplus in the intermediate energy
market and the goods and services markets. For example, assume a simple world with only
5-17
-------
(a) Change in Consumer Surplus
in the Energy Market
(b) Change in Producer Surplus in
the Energy Market
(c) Change in Consumer Surplus
in Goods and Services Markets
(d) Change in Producer Surplus in
Goods and Services Markets
Figure 5-5. Changes in Economic Welfare with Regulation
one energy market, wholesale electricity, and one product market, pulp and paper. If the
regulation increases the cost of generating wholesale electricity, then part of the cost of the
regulation will be borne by the electricity producers as decreased producer surplus, and part
of the costs will be passed on to the pulp and paper manufacturers. In Figure 5-5(a), the pulp
and paper manufacturers are the consumers of electricity, so the change in consumer surplus
is displayed. This change in consumer surplus in the energy market is captured by the
5-18
-------
product market (because the consumer is the pulp and paper industry in this case), where it is
split between consumer surplus and producer surplus in those markets. Figure 5-5(b) shows
the change in producer surplus in the energy market, where B represents an increase in
producer surplus and C represents a decrease.
As shown in Figures 5-5(c) and 5-5(d), the cost affects the pulp and paper industry by
shifting up the supply curve in the pulp and paper market. These higher electricity prices
therefore lead to costs in the pulp and paper industry that are distributed between producers
and consumers of paper products in the form of lower producer surplus and lower consumer
surplus. Note that the change in consumer surplus in the intermediate energy market must
equal the total change in consumer and producer surplus in the product market. Thus, to
avoid double-counting, the change in consumer surplus in the intermediate energy market
was not quantified; instead the total change in social welfare was calculated as
Change in Social Welfare = £APSE + £APSF + £ACSF + £ACSR (5.1)
where
APSE = change in producer surplus in the energy markets;
APSF = change in producer surplus in the goods and services markets;
ACSF = change in consumer surplus in the goods and services markets; and
ACSR = change in consumer surplus in the commercial, residential, and
transportation energy markets.
Appendix A contains the mathematical algorithms used to calculate the change in producer
and consumer surplus in the appropriate markets. The market analysis is conducted for the
year 2005 and incorporates both growth in supply and demand. As a result, both new and
existing sources are evaluated using the same analysis approach.
The engineering control costs presented in Section 3.3 are inputs (regulatory
"shocks") in the market model approach. The magnitude and distribution of the regulatory
costs' impact on the economy depend on the relative size of the impact on individual markets
(relative shift of the market supply curves) and the behavioral responses of producers and
consumers in each market (measured by the price elasticities of supply and demand).
5-19
-------
SECTION 6
ECONOMIC IMPACT ANALYSIS RESULTS
The underlying objective of the EIA is to evaluate the effect of the regulation on the
welfare of affected stakeholders and society in general. Although the engineering cost
analysis presented in Section 3 does represent an estimate of the resources required to comply
with the rule under baseline economic conditions, the analysis does not account for the fact
that the regulations may cause the economic conditions to change. For instance, producers
may reduce production in the face of higher production costs, thereby reducing market
supply. Moreover, the control costs may be passed along to other parties through various
economic exchanges. Therefore, EPA developed an analytical structure and economic model
to measure and track these effects (described in detail in Section 5 and Appendix A). In this
section, we report quantitative estimates of these welfare impacts and their distribution across
stakeholders.
6.1 Social Cost Estimates
Under the final rule, EPA estimates the total change in social welfare is $862.9
million (see Table 6-1). This estimate is slightly smaller (less than $0.3 million) than the
estimated baseline engineering costs as a result of behavior changes by producers and
consumers. Possible behavior responses include changes in consumption and production
patterns and fuel switching.
Table 6-1. Social Cost Estimates ($1998 106): Final Rule
Change in Social Welfare
Baseline engineering costs $863.0
Social costs with market adjustments $862.9
Difference between engineering and social costs $0.1
6-1
-------
EPA also estimated the distribution of social costs between producers and consumers
and report the distribution of impacts across sectors/markets in Table 6-2. The market
analysis estimates that consumers will bear $414.3 million, or 48 percent of the total social
cost as a result of higher prices and lower consumption levels. Producer surplus is projected
to decrease by $448.7 million, or 52 percent of the total social cost as result of direct control
costs, higher energy costs, and reductions in output.
With exception of the natural gas market, energy producers are expected to
experience producer surplus losses. Under the final rule, electricity, petroleum, and coal
producer surplus is projected to decline by approximately $35 million. In contrast, natural
gas producer surplus is projected to increase by $4 million as they benefit from increased
demand from industries switching from petroleum and electricity.
The majority welfare impacts fall on the agriculture, manufacturing, and mining
industries. EPA estimates total welfare losses of $609.8 million for these sectors.
Manufacturing industries with large number of boilers and process heaters and industries that
consume electricity experience the majority these losses (e.g., chemicals and allied products,
paper, textile mill products, and food). Consumers in these industries experience losses of
$295.2 million and producers bear $314.6 million. The cost of this rule to producers as a
percentage of baseline 2005 shipments is less than 0.5 percent.
EPA also examined the impact on the commercial, transportation and residential
sectors. The total welfare loss for the commercial sector is estimated to be $167.1 million.
Therefore, the regulatory burden associated with the MACT is estimated as 0.001 percent of
total 2005 commercial sector revenues. Consumers in this sector bear approximately $71.6
million and producers bear $95.5 million of these impacts. In contrast, the total welfare loss
for the transportation sector is estimated to be $9.0 million. The regulatory burden associated
with the rule is estimated as 0.003 percent of total 2005 transportation sector revenues.
Transportation consumers bear approximately $4.7 million and producers bear $4.3 million
of these impacts. Finally, the social cost burden to residential consumers of energy, $42.7
million, is 0.037 percent of annual residential energy expenditures in 2005.
6.2 National Market-Level Impacts
Increases in the costs of production in the energy and final product markets due to the
regulation are expected to result in changes in prices, production, and consumption from
6-2
-------
Table 6-2. Distribution of Social Costs by Sector/Market: Final Rule ($1998 106)
Sectors/Markets
Energy Markets
Petroleum
Natural gas
Electricity
Coal
Subtotal
NAICS Code
311
312
313
314
315
316
321
322
323
325
326
327
331
332
333
334
335
336
337
339
11
23
21
48
42; 44-45; 49;
51-56; 61-62;71-
72; 81
Grand Total
SIC Code
20 (pt)
20 (pt); 21
22 (pt)
22 (pt)
23
31
24
26
27
28
30
32
33
34
35
36 (pt)
36 (pt)
37
25
39
01-08
15-17
10; 14
40-47 (pt)
40-48 (pt);
50-99
Description
Food
Beverage and Tobacco Products
Textile Mills
Textile Product Mills
Apparel
Leather and Allied Products
Wood Products
Paper
Printing and Related Support
Chemicals
Plastics and Rubber Products
Nonmetallic Mineral Products
Primary Metals
Fabricated Metal Products
Machinery
Computer and Electronic Products
Electrical Equipment, Appliances,
and Components
Transportation Equipment
Furniture and Related Products
Miscellaneous
Agricultural Sector
Construction Sector
Other Mining Sector
Transportation
Commercial
Residential
Producer
Surplus
-$1.9
$4.1
-$33.7
-$2.7
-$34.2
-$28.2
-$2.4
-$22.7
-$0.1
-$0.4
-$0.3
-$39.1
-$66.1
-$0.2
-$40.9
-$2.2
-$3.4
-$25.2
-$8.5
-$7.3
-$3.6
-$2.5
-$24.6
-$5.4
-$0.8
-$0.6
-$0.8
-$10.1
-$4.7
-$71.6
NA
-$414.3
Change in:
Consumer
Surplus
-$11.3
-$4.1
-$52.0
-$0.1
-$1.1
-$0.4
-$10.4
-$60.0
-$0.4
-$81.8
-$5.4
-$4.0
-$5.7
-$2.3
-$4.9
-$1.4
-$1.6
-$32.8
-$24.6
-$0.7
-$1.3
-$1.1
-$7.0
-$4.3
-$95.5
-$42.7
-$448.7
Social
Welfare
-$39.4
-$6.5
-$74.7
-$0.2
-$1.5
-$0.7
-$49.5
-$126.1
-$0.6
-$122.8
-$7.6
-$7.4
-$30.9
-$10.8
-$12.2
-$5.0
-$4.1
-$57.3
-$30.1
-$1.5
-$1.9
-$1.9
-$17.2
-$9.0
-$167.1
-$42.7
-$862.9
NA = Not applicable.
pt = Part.
6-3
-------
Table 6-3. Market-Level Impacts: Final Rule
Percent Change
Sectors/Markets
Energy Markets
Petroleum
Natural gas
Electricity
Coal
NAICS Code
311
312
313
314
315
316
321
322
323
325
326
327
331
332
333
334
335
336
337
339
11
23
21
48
42; 44-45; 49; 51-56;
61-62; 71-72; 81
SIC Code
20 (pt)
20 (pt); 21
22 (pt)
22 (pt)
23
31
24
26
27
28
30
32
33
34
35
36 (pt)
36 (pt)
37
25
39
01-08
15-17
10; 14
40-47 (pt)
40-48 (pt); 50-
99
Description
Food
Beverage and Tobacco Products
Textile Mills
Textile Product Mills
Apparel
Leather and Allied Products
Wood Products
Paper
Printing and Related Support
Chemicals
Plastics and Rubber Products
Nonmetallic Mineral Products
Primary Metals
Fabricated Metal Products
Machinery
Computer and Electronic Products
Electrical Equipment, Appliances, and
Components
Transportation Equipment
Furniture and Related Products
Miscellaneous
Agricultural Sector
Construction Sector
Other Mining Sector
Transportation
Commercial
Price
0.002%
0.005%
0.050%
-0.007%
0.006%
0.003%
0.025%
0.000%
0.000%
0.002%
0.041%
0.026%
0.000%
0.009%
0.001%
0.003%
0.011%
0.003%
0.002%
0.001%
0.002%
0.004%
0.008%
0.001%
0.000%
0.000%
0.012%
0.001%
0.000%
Quantity
0.000%
0.002%
-0.011%
-0.010%
-0.002%
-0.004%
-0.021%
0.000%
-0.001%
-0.003%
-0.008%
-0.028%
0.000%
-0.013%
-0.002%
-0.003%
-0.009%
-0.001%
-0.001%
0.000%
-0.001%
-0.004%
-0.026%
0.000%
0.000%
0.000%
-0.004%
-0.001%
0.000%
pt = Part.
6-4
-------
baseline levels. As shown in Table 6-3, the electricity market price increases by 0.050
percent, while production/consumption decreases by 0.011 percent as a result of additional
control costs. A significant share of electricity is produced in the United States using coal as
a primary input. Therefore, projected reductions in electricity production also lead to a
decrease in demand for coal. As a result, the price and quantities of coal are projected to fall
by 0.007 percent and 0.010 percent, respectively. In the petroleum market, the model
projects small price and quantity effects (i.e., less than 0.01 percent). In the natural gas
market, the model projects the market price will rise in response to increased demand (0.005
percent). The price increase is the result of additional control costs and increased demand.
Production and consumption quantities also increase in this market (0.002) as a result of
increased demand.
Additional control costs and higher energy costs associated with the regulation lead to
higher goods and services prices in all markets and a decline in output. However, the
changes are generally very small. Under the MACT Floor, three markets have price increases
greater than or equal to 0.02 percent—Wood Products (NAICS 321), Paper (NAICS 322),
and Textile Mills (NAICS 313). The producers in these sectors are expected to face higher
per-unit control costs relative to other industries. In addition, these industries are also
electricity-intensive; therefore, costs of production also increase as a result of higher
electricity prices.
Although the impacts on price and quantity in the goods and services markets are
estimated to be small, one possible effect of modeling market impacts at the two and three
digit NAICS code level is that fuel-intensive industries within the larger NAICS code
definition may be affected more significantly than the average industry for that NAICS code.
Thus, the changes in price and quantity should be interpreted as an average for the whole
NAICS code, not necessarily for each disaggregated industry within that NAICS code.
6.3 Executive Order 13211 (Energy Effects)
Executive Order 13211, "Actions Concerning Regulations That Significantly Affect
Energy Supply, Distribution, or Use" (66 Fed. Reg. 28355 [May 22, 2001]), requires EPA to
prepare and submit a Statement of Energy Effects to the Administrator of the Office of
Information and Regulatory Affairs, Office of Management and Budget, for certain actions
identified as "significant energy actions." Section 4(b) of Executive Order 13211 defines
"significant energy actions" as "any action by an agency (normally published in the Federal
Register) that promulgates or is expected to lead to the promulgation of a final rule or
6-5
-------
regulation, including notices of inquiry, advance notices of rulemaking, and notices of
rulemaking:
• that is a significant regulatory action under Executive Order 12866 or any
successor order, and is likely to have a significant adverse effect on the supply,
distribution, or use of energy; or
• that is designated by the Administrator of the Office of Information and
Regulatory Affairs as a significant energy action."
EPA has provided additional information on the impacts of the final rule on affected
energy markets below.1
Energy Price Effects. As described in the market-level results section, electricity
prices are projected to increase by less than 1 percent. Petroleum and natural gas prices are
all projected to increase by less than 0.1 percent. The price of coal is projected to decrease
slightly.
Impacts on Electricity Supply, Distribution, and Use. We project the increased
compliance costs for the electricity market will result in an annual production decline of
approximately 415 million kWh under the MACT floor.
Impacts on Petroleum, Natural Gas, and Coal Supply, Distribution, and Use. The
model projects decreases in petroleum production/consumption of approximately 68 barrels
per day under the MACT floor. In contrast, natural gas production/consumption is projected
to increase by 1.1 million cubic feet per day under the MACT floor. This is the result of fuel
switching in response to relative price changes. Finally, the model also projects less than a
1,000 tons per day decrease in coal production/consumption in response to reduced output
from the electricity sector (a significant consumer of coal).
6.4 Conclusions
The decrease in social surplus estimated using the market analysis is $862.9 million.
This estimate is slightly smaller than the estimated baseline engineering costs because the
market model accounts for behavioral changes of producers and consumers. Although the
rule affects boilers and process heaters used in energy industries, energy producers only incur
less than 6 percent of the total social cost of the regulation. This burden is spread across
'Conversion factors for heat rates were obtained from AEO 2002, Appendix H. These factors vary by year to
year; 2010 values are reported in this Appendix.
6-6
-------
numerous markets because the price of energy increases slightly as a result of the regulation,
which increases the cost of production for all markets that use energy as part of their
production process.
The remaining share of the social cost is mostly borne by the manufacturing sectors
which operate the majority of the boilers and process heaters affected by the regulation.
Manufacturing industries bearing the largest social costs include percent—Wood Products
(NAICS 321), Paper (NAICS 322), and Textile Mills (NAICS 313). However, the market
model predicts that changes in these industries' price and quantity do not exceed 0.02
percent.
Because of the minimal changes in price and quantity estimated for most of the
affected markets, EPA expects that there would be no discernable impact on international
trade. Although an increase in the price of U.S. products relative to those of foreign
producers is expected to decrease exports and increase imports, the changes in price due to
the industrial boilers and process heaters MACT are generally too small to significantly
influence trade patterns. There may also be a small decrease in employment, but because the
impact of the regulation is spread across so many industries and the decreases in market
quantities are so small, it is unlikely that any particular industry will face a significant
decrease in employment.
6-7
-------
SECTION 7
SMALL ENTITY IMP ACTS
This section investigates the potential impact the regulation will have on small
entities. The Agency has identified 185 small entities that will be affected by the final rule
for the ICI boilers and process heaters NESHAP. For these entities, the average cost-to-sales
ratio (CSR) is 0.78 percent and the average annual control cost is $199,000. Ten entities will
incur annual costs that are greater than or equal to 3 percent of their annual sales (see
Table 7-1).
Table 7-1. Summary of Small Entity Impacts
Final Rule
Number of small entities 185
Total number of entities 576
Average annual control cost per small entity (103) $199
Average control cost/sales ratio 0.78%
Number of small entities with cost-to-sales ratios >1 percent 34
Number of small entities with cost-to-sales ratios >3 percent 10
7.1 Background on Small Entity Screenings
The regulatory costs imposed on domestic producers and government entities to
reduce air emissions from boilers and process heaters will have a direct impact on owners of
the affected facilities. Firms, individuals, or governmental jurisdictions that own the
facilities with boilers and process heaters are typically 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 potentially affect all sizes of businesses, but small businesses may
have special problems relative to large businesses in complying with such regulations.
7-1
-------
The Regulatory Flexibility Act (RFA) generally requires an agency to prepare a
regulatory flexibility analysis of any rule subject to notice and comment rulemaking
requirements under the Administrative Procedure Act or any other statute unless the agency
certifies that the rule will not have a significant economic impact on a substantial number of
small entities. Small entities include small businesses, small organizations, and small
governmental jurisdictions.
For purposes of assessing the impacts of today's rule on small entities, small entity is
defined as: (1) a small business according to Small Business Administration (SBA) size
standards by the North American Industry Classification System (NAICS) category of the
owning entity. The range of small business size standards for the 40 affected industries
ranges from 500 to 1,000 employees, except for petroleum refining and electric utilities. In
these latter two industries, the size standard is 1,500 employees and amass throughput of
75,000 barrels/day or less, and 4 million kilowatt-hours of production or less, respectively.
(2) a small governmental jurisdiction that is a government of a city, county, town, school
district or special district with a population of less than 50,000; and (3) a small organization
that is any not-for-profit enterprise which is independently owned and operated and is not
dominant in its field.
This section investigates characteristics of businesses and government entities that
own existing boilers and process heaters affected by this rule and provides a preliminary
screening-level analysis to assist in determining whether this rule is likely to impose a
significant impact on a substantial number of the small businesses within this industry. The
screening-level analysis employed here is a "sales test," which computes the annualized
compliance costs as a share of sales/revenue for existing companies/government entities.
7.2 Identifying Small Entities
To support the economic impact analysis of the regulation, EPA identified 2,186
boilers and process heaters located at commercial, industrial, and government facilities that
would be affected by the regulation. The population of boilers and process heaters was
developed from the EPA ICCR Inventory Database version 4.1.! The list of boilers and
process heaters contained in these databases was developed from information in the AIRS
and OTAG databases, state and local permit records, and the combustion source ICR
conducted by the Agency. Industry and environmental stakeholders reviewed the units
'The ICCR Inventory Database contains data for boilers, process heaters, incinerators, landfill gas flares,
turbines, and internal combustion engines.
7-2
-------
contained in these databases as part of the ICCR FACA process. In addition, stakeholders
contributed to the databases by identifying and including omitted units. Information was
extracted from the ICCR databases to support the ICI boilers and process heaters NESHAP.
This modified database containing information on only boilers and process heaters is referred
to as the Inventory Database.
The small entities screening analysis for the regulation is based on the evaluation of
existing owners of boilers and process heaters for which information was available. It is
assumed that the size and ownership distribution of units in the Inventory Database is
representative of the entire estimated population of existing boilers and process heaters. In
addition, it is assumed that new sources included in the 2005 population will also be
representative of the Inventory Database. However, because our analysis is based on a
subset of the total population of boilers and process heaters, the number of entities identified
as highly affected in this analysis may not be identical to the actual impact of the regulation
on small entities. The remainder of this section presents cost and sales information on small
companies and government organizations that own existing boilers and process heaters.
7.3 Analysis of Facility-Level and Parent-Level Data
The 2,186 units in the Inventory Database with full information were linked to 1,214
existing facilities. As shown in Table 7-2, these 1,186 facilities are owned by 576 parent
entities. The average number of facilities per entity is approximately 2.0; however, as is also
illustrated in Table 7-2, several large entities in the health services industry and government
sectors own many facilities with boilers and process heaters.
Employment and sales are typically used as measures of business size. Employment,
sales, population, and tax revenue data (when applicable) were collected for the 576 parent
companies and government entities.2 Figure 7-1 shows the distribution of employees by
parent company for the final rule. Employment for parent companies ranges from 5 to
608,000 employees. One hundred seventy-eight or more of the firms have fewer than 500
employees, and 55 companies have more than 25,000 employees.
Sales provide another measure of business size. Figure 7-2 presents the sales or
revenue distribution for affected parent entities. The median sales figure for affected
companies is $300 million ($200 million), and the average sales figure is $4.1 billion
2Total annualized cost is compared to tax revenue to assess the relative impact on local governments.
7-3
-------
Table 7-2. Facility-Level and Parent-Level Data by Industry
SIC
Code
01
02
07
10
12
13
14
17
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
NAICS
Code
111
112
115
212
212
211
212
235
311
312
313
315
321
337
322
511
325
324
326
316
327
331
332
333
335
336
334
339
482
Description
Agriculture — Crop s
Agriculture — Livestock
Agricultural Services
Metal Mining
Coal Mining
Oil and Gas Extraction
Mining/Quarrying — Nonmetallic
Minerals
Construction — Special Trade
Food and Kindred Products
Tobacco Products
Textile Mill Products
Apparel & Other Products from
Fabrics
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Printing, Publishing, and Related
Industries
Chemicals and Allied Products
Petroleum Refining and Related
Industries
Rubber and Misc. Plastics Products
Leather and Leather Products
Stone, Clay, Glass, and Concrete
Products
Primary Metal Industries
Fabricated Metal Products
Industrial Machinery and Computer
Equip.
Electronic and Electrical Equipment
Transportation Equipment
Scientific, Optical, and Photographic
Equipment
Misc. Manufacturing Industries
Railroad Transportation
Number
of
Units
3
—
—
9
2
—
8
—
138
11
135
2
360
234
321
—
174
11
17
1
9
41
16
23
5
102
8
2
4
Number
of
Facilities
3
—
—
4
1
—
4
—
60
7
71
2
262
154
194
—
70
8
13
1
7
16
10
12
5
41
4
2
1
Number
of
Parent
Entities
3
—
—
2
—
—
3
—
32
4
33
1
122
67
68
—
41
9
9
1
4
10
7
9
3
12
3
2
1
Avg.
Number of
Facilities
Per Parent
Entity
1.0
—
—
2.0
—
—
1.3
—
1.9
1.8
2.2
2.0
2.1
2.3
2.9
—
1.7
0.9
1.4
1.0
1.8
1.6
1.4
1.3
1.7
3.4
1.3
1.0
1.0
(continued)
7-4
-------
Table 7-2. Facility-Level and Parent-Level Data by Industry (continued)
SIC
Code
42
46
49
50
51
55
58
59
60
70
72
76
80
81
82
83
86
87
89
91
92
94
96
97
NA
State
NAICS
Code
484
486
221
421
422
441
722
445-454
522
721
812
811
621
541
611
624
813
541
711/514
921
922
923
926
928
Description
Motor Freight and Warehousing
Pipelines, Except Natural Gas
Electric, Gas, and Sanitary Services
Wholesale Trade — Durable Goods
Wholesale Trade — Nondurable Goods
Automotive Dealers and Gasoline
Service Stations
Eating and Drinking Places
Miscellaneous Retail
Depository Institutions
Hotels and Other Lodging Places
Personal Services
Misc. Repair Services
Health Services
Legal Services
Educational Services
Social Services
Membership Organizations
Engineering, Accounting, Research,
Management and Related Services
Services, N.E.C.
Executive, Legislative, and General
Administration
Justice, Public Order, and Safety
Administration of Human Resources
Administration of Economic Programs
National Security and International
Affairs
SIC Information Not Available
Parent is a state government
Total
Number
of
Units
5
—
318
3
2
—
—
—
—
1
—
2
37
—
105
2
—
2
2
1
29
1
4
29
7
—
2,186
Number
of
Facilities
1
—
160
2
1
—
—
—
—
1
—
1
18
—
45
1
—
2
1
1
9
1
3
11
4
—
1,214
Number
of
Parent
Entities
1
—
80
1
1
—
—
—
—
1
—
—
2
—
30
—
—
1
—
—
—
—
1
2
—
10
576
Avg.
Number of
Facilities
Per Parent
Entity
1.0
—
2.0
2.0
1.0
—
—
—
—
1.0
—
—
9.0
—
1.5
—
—
2.0
—
—
—
—
3.0
5.5
—
—
2.0
Source: Industrial Combustion Coordinated Rulemaking (ICCR). 1998. Data/Information Submitted to the
Coordinating Committee at the Final Meeting of the Industrial Combustion Coordinated Rulemaking Federal
Advisory Committee. EPA Docket Numbers A-94-63, II-K-4b2 through -4b5. Research Triangle Park, North
Carolina. September 16-17.
7-5
-------
I
E
to
Q.
200
150
100
50
0
141
10 12
I 1 I I
31
125
52
121
55
<25
25 to 49 50 to 99
100 to 500 to 1 pOO to 5,000 to >25,000
499 999 4,999 24,999
Parent Employment
Figure 7-1. Parent Size by Employment Range
*Excludes 29 parent government entities.
($3.5 billion) (excluding the federal government). As shown in Figure 7-2, revenue and sales
figures vary greatly across the population: 209 firms and governments affected by the final
rule have annual revenues less than $100 million per year. These figures include all sales
associated with the parent company, not just facilities affected by the regulation (i.e.,
facilities with boilers or process heaters).
Based on SBA guidelines, 185 of the entities were identified as small businesses.3
Small businesses by business type are presented in Table 7-3. The lumber and wood
products industry contains the largest number of the small businesses with 84, followed by
furniture and fixtures with 28, electric services with 26, and paper and allied products with
13. The remaining small businesses are distributed across 40 different two-digit SIC code
groupings.
'Small business guidelines typically define small businesses based on employment, and the threshold varies
from industry to industry. For example, in the paints and allied products industry, a business with fewer
than 500 employees is considered a small business; whereas in the industrial gases industry, a business with
fewer than 1,000 employees is considered small. However, for a few industries, usually services, sales are
used as the criterion. For example, in the veterinary hospital industry, companies with less than $5 million
in annual sales are defined as small businesses.
7-6
-------
# 150 -
0)
n
Q_
'S 100-
^
0}
1 50
n .
18
7
128
110
56
51
133
"n 33
2L\ -\ 7
<5 5to9 10to49 50to99 100 to SOOto 1,000 to 5,000to 10,000to >25,000
499 999 4,999 9,999 24,999
Parent Sales ($1Q6)
Figure 7-2. Number of Parents by Sales Range
*Excludes 3 parent entities for which sales or revenue information was unavailable.
7.4 Small Entity Impacts
Table 7-4 presents a summary of the ratio of control costs to sales for affected large
and small entities. The average CSR is 0.14 percent for large entities (excluding the federal
government) and 0.78 percent for small entities. Forty-four small parents had CSRs greater
than 1 percent, assuming add-on control is employed to meet the standard. For these 44
parent companies, the CSRs ranged from 1.00 percent to 7.83 percent. Ten entities out of
these 44 had CSRs ratios greater than 3 percent.
7.5 Affected Government Entities: Supplemental Analysis
Of the 185 small entities identified, 13 were small governmental jurisdictions that
own and operate "public power" producers with affected boilers. The Regulatory Flexibility
Act as amended by the Small Business Regulatory Enforcement Fairness Act provides the
following standard definition of "small governmental jurisdiction": a city, county, town,
township, village, school district, or special district with a population of less than 50,000.
For this analysis, public power producers are defined as nonprofit publicly owned electrical
utilities operated by municipalities, counties, and states or other publicly owned bodies such
as public utility districts. This excludes rural electric cooperatives.
7-7
-------
Table 7-3. Small Parent Entities by Industry
SIC Code
01
02
07
10
12
13
14
17
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
42
46
49
50
NAICS
Code
111
112
115
212
212
211
212
235
311
312
313
315
321
337
322
511
325
324
326
316
327
331
332
333
335
336
334
339
482
484
486
221
421
Description
Agriculture — Crop s
Agriculture — Livestock
Agricultural Services
Metal Mining
Coal Mining
Oil and Gas Extraction
Mining/Quarrying — Nonmetallic Minerals
Construction — Special Trade Contractors
Food and Kindred Products
Tobacco Products
Textile Mill Products
Apparel and Other Products from Fabrics
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Printing, Publishing, and Related Industries
Chemicals and Allied Products
Petroleum Refining and Related Industries
Rubber and Misc. Plastics Products
Leather and Leather Products
Stone, Clay, Glass, and Concrete Products
Primary Metal Industries
Fabricated Metal Products
Industrial Machinery and Computer Equip.
Electronic and Electrical Equipment
Transportation Equipment
Scientific, Optical, and Photographic Equip.
Miscellaneous Manufacturing Industries
Railroad Transportation
Motor Freight and Warehousing
Pipelines, Except Natural Gas
Electric, Gas, and Sanitary Services
Wholesale Trade — Durable Goods
Number of Parent
Entities
3
—
—
2
—
—
3
—
32
4
33
1
122
67
68
—
41
9
9
1
4
10
7
9
3
12
3
2
1
1
—
80
1
Number of Small
Parent Entities
—
—
—
2
—
—
—
—
12
—
5
—
84
28
13
—
4
2
1
1
—
1
3
1
—
1
—
—
—
—
—
26
—
(continued)
7-8
-------
Table 7-3. Small Parent Entities by Industry (continued)
SIC Code
51
55
58
59
60
70
72
76
80
81
82
83
86
87
89
91
92
94
96
97
NA
State
NAICS
Code
422
441
722
445-454
522
721
812
811
621
541
611
624
813
541
711/514
921
922
923
926
928
Description
Wholesale Trade — Nondurable Goods
Automotive Dealers and Gasoline Service
Stations
Eating and Drinking Places
Miscellaneous Retail
Depository Institutions
Hotels and Other Lodging Places
Personal Services
Misc. Repair Services
Health Services
Legal Services
Educational Services
Social Services
Membership Organizations
Engineering, Accounting, Research, Management
and Related Services
Services, N.E.C.
Executive, Legislative, and General
Administration
Justice, Public Order, and Safety
Administration of Human Resources
Administration of Economic Programs
National Security and International Affairs
SIC Information Not Available
Parent is a State Government
Total
Number of Parent
Entities
1
—
—
—
—
1
—
—
2
—
30
—
—
1
—
—
—
—
1
2
—
10
576
Number of Small
Parent Entities
—
—
—
—
—
—
—
—
1
—
—
—
—
—
—
—
—
—
—
—
—
—
185
Source: Industrial Combustion Coordinated Rulemaking (ICCR). 1998. Data/Information Submitted to the Coordinating
Committee at the Final Meeting of the Industrial Combustion Coordinated Rulemaking Federal Advisory
Committee. EPA Docket Numbers A-94-63, II-K-4b2 through -4b5. Research Triangle Park, North Carolina.
September 16-17.
7-9
-------
Table 7-4. Summary Statistics for SBREFA Screening Analysis
Value
Total Number of Small Entities 185
Average Annual Compliance Cost per Small Entity ($103) $199
Entities with Sales/Revenue Data
Compliance costs are <1% of sales 141
Compliance costs are >1 to 3% of sales 34
Compliance costs are >3% of sales 10
Compliance Cost-to-Sales/Revenue Ratios
Average 0.78
Median 0.50
Maximum 7.83
Minimum 0.01
As illustrated in Table 7-5, the vast majority of small municipal systems with affected
boilers are located in the Midwest (11 systems or 85 percent). Four of the 11 municipal
systems are located in Minnesota, with two in Indiana and two in Michigan.
Historically municipal utilities were created to provide residents of a community with
reliable energy. For example, the residential sector accounts for more than two-thirds of
total consumers in all cases (see Table 7-6). However, the residential sector generally
represents the smallest group in terms of total energy consumption. The industrial and
commercial sectors consume approximately 70 percent of total energy supplied. Power not
consumed by the residential, commercial, or industrial sector is sold into the wholesale
energy market.
Public power producers do not pay state or local taxes. However, they typically are
under agreement to make annual contributions to state and local government operating funds.
In addition, they are not guaranteed a rate of return (as regulated public utilities are);
however, their rates are set by agreement with local councils, and these rates are typically
adjusted to reflect changes in operating costs.
7-10
-------
Table 7-5. Regional Distribution of Municipal Systems
Regional Distribution
East
Vermont
Midwest
Indiana
Iowa
Michigan
Minnesota
Ohio
Wisconsin
West
California
Total
Number of Facilities
1
2
1
2
4
1
1
1
13
Municipal utilities can generate capital by issuing tax-exempt municipal bonds.
These municipal bonds are exempt from federal income taxes, which allows the publicly
owned utilities to finance capital projects at a more affordable rate. Additionally, the local
governments investing in municipal utilities generally issue revenue bonds rather than
general obligation bonds. This ensures that the debt can be paid back through revenues from
generating electricity and does not obligate the local government or community tax base.
As shown in Table 7-7, the average total annual compliance costs per entity are
$223,000 under the final rule. The median cost-to-revenue ratio is 0.94 percent, and ratios
range from less than 0.5 percent to 8 percent. Three of the affected small governments have
cost-to-revenue ratios at or above 3 percent.
7.6 Assessment of SBREFA Screening
This analysis indicates that over two-thirds of the entities affected by the industrial
boilers and process heaters standard are large.4 The relatively small proportion of small
entities affected by the regulation at the MACT floor is due in part to the exclusion of ICI
4Based on SB A guidelines for determining small entities.
7-11
-------
Table 7-6. Selected Municipal Utilities' Capacity, Usage, and Consumer Types
Row
ID
1
2
3
4
5
6
7
8
9
10
11
12
Capacity
(MW)
50.5
115
24.3
22.2
34.5
23
35
46
103.1
32
26
34
Energy Usage
(kWh/y)
332,524,000
371,823,000
388,066,000
185,191,000
147,335,000
573,003,000
338,903,000
194,753,000
837,175,000
218,208,000
267,201,000
95,642,000
Distribution of Energy Usage by
Customer Type
Residential
27%
36%
19%
26%
26%
8%
38%
22%
NA
40%
16%
33%
Commercial
NA
28%
10%
14%
27%
NA
8%
NA
NA
3%
NA
67%
Industrial
NA
16%
70%
58%
44%
NA
51%
NA
NA
55%
NA
NA
Total
Consumers
19,313
15,615
9,082
6,235
5,955
7,207
13,247
6,890
NA
10,829
9,471
5,747
Distribution of Customers
Residential
82%
87%
84%
86%
86%
90%
87%
85%
NA
88%
75%
83%
Commercial
15%
11%
14%
13%
14%
7%
11%
13%
NA
3%
24%
17%
Industrial
3.7%
0.3%
1.0%
1.6%
0.3%
1.0%
1.3%
0.1%
NA
8.4%
0.3%
0.3%
to
Source: Giles, Ellen F. 2000. Platt's Directory of Electric Power Producers and Distributors, 109th Edition of the Electrical World Directory.
New York: McGraw Hill.
-------
Table 7-7. Supplemental Screening Analysis for Small Governmental Jurisdictions
Value
Total Number of Small Entities 13
Average Total Annual Compliance Cost (TACC) per Small Entity ($103) $223
Entities with Sales/Revenue Data
Compliance costs are <1% of revenue 7
Compliance costs are >1 to 3% of revenue 3
Compliance costs are >3% of revenue 3
Compliance Cost-to-Sales/Revenue Ratios
Average 1.67
Median 0.94
Maximum 7.83
Minimum 0.02
Source: American Public Power Association (APPA). 2002. Straight Answers to False Charges about
Public Power. Washington, DC: APPA. .
As obtained on November 1 3, 2003.
boilers and process heaters with less than 10 MMBtu input capacity that also use a fossil fuel
liquid or gas as primary fuel. As a result, a large share of small boilers and process heaters,
which are presumably owned disproportionately by smaller entities, will not incur
compliance costs. The Agency estimates that approximately 57 percent of the U.S.
population are less than 10 MMBtus or are emergency units and, hence, are excluded from
the regulation.
Of the small entities affected by the final rule, the majority are in the lumber and
wood products, furniture and fixtures, paper and allied products, and electric, gas and
sanitary services sectors. As shown in Table 7-8, the median profit margin for these four
sectors is approximately 3 percent. Table 7-8 also shows the profit margins for the other
industry sectors with affected small businesses. All profit margins of industry sectors with
affected small businesses are above 2 percent.
7-13
-------
Table 7-8. Profit Margins for Industry Sectors with Affected Small Businesses
SIC
Code
20
22
24
25
26
28
49
NAICS
Code
311
313
321
337
322
325
221
Description
Food and Kindred Products
Textile Mill Products
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Chemicals and Allied Products
Electric, Gas, and Sanitary Services
Median Profit Margin
3.6%
2.1%
3.0%
3.0%
3.3%
2.7%
7.5%
Source: Dun & Bradstreet. 1997. Industry Norms & Key Business Ratios. Desktop Edition 1996-97. Murray
Hill, NJ: Dun & Bradstreet, Inc.
After considering the economic impact of the final rule on small entities, EPA
certifies that this action will not have a significant impact on a substantial number of small
entities. In accordance with the RFA, as amended by the SBREFA, 5 U.S.C. 601, et. seq.,
EPA conducted an assessment of the standard on small businesses within the industries
affected by the rule. Based on SBA size definitions and reported sales and employment data,
the Agency identified 185 entities, or 32 percent. Although small entities represent 32
percent of the SBREFA screening population, they are expected to incur only 8 percent of
the total compliance costs of $445.6 million (1998$). Only ten small entities have
compliance costs equal to or greater than 3 percent of their sales. In addition, only 34 small
entities have CSRs between 1 and 3 percent. Additional analysis of small governmental
jurisdictions shows 3 of 13 have CSRs greater than 3 percent, and 3 have CSRs between 1
and 3 percent.
An EIA was performed to estimate the changes in product price and production
quantities for this rule. As mentioned in the summary of economic impacts earlier in this
report, the estimated changes in prices and output for affected firms (including small firms)
are no more than 0.04 percent.
This rule will not have a significant economic impact on a substantial number of
small entities as a result of several decisions EPA made regarding the development of this
rulemaking which resulted in limiting the impact of this rule on small entities. First, as
7-14
-------
mentioned earlier, EPA identified small units (heat input of 10 MMBtu/hr or less) and
limited-use boilers (operate less than 10 percent of the time) as separate subcategories from
large units. Many small and limited-use units are located at small entities. As also discussed
earlier, the result of the MACT floor analysis for these subcategories of existing sources was
that no MACT floor could be identified except for the limited-use solid fuel subcategory,
which is less stringent than the MACT floor for large units. Furthermore, the results of the
above-the-fioor analysis for these subcategories indicated that the costs would be too high to
be considered feasible. Consequently, this rule contains no emission limitations for any of
the existing small and limited-use subcategories except the existing limited-use solid fuel
subcategory. In addition, the alternative metals emission limit resulted in minimizing the
impacts on small entities because some of the potential entities burning a fuel containing
very little metals are small entities.
7-15
-------
REFERENCES
American Public Power Association (APPA). 2002. Straight Answers to False Charges
about Public Power. Washington, DC: APPA. . As obtained on November 13, 2003.
Dahl, Carol A., and Thomas E. Duggan. 1996. "U.S. Energy Product Supply Elasticities: A
Survey and Application to the U.S. Oil Market." Resource and Energy
Economics\8:243-263.
Dun & Bradstreet. 1997. Industry Norms & Key Business Ratios. Desktop Edition 1996-97.
Murray Hill, NJ: Dun & Bradstreet, Inc.
Eastern Research Group. Memorandum to Jim Eddinger, U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Development of the
Population Database for the Industrial/Commercial/Institutional Boiler and Indirect-
Fired Process Heater National Emission Standard for Hazardous Air Pollutants
(NESHAP). May 18, 2000.
Eastern Research Group. Memorandum to Jim Eddinger, U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Development of Model Units
for the Industrial/Commercial/Institutional Boiler and Indirect-Fired Process Heater
National Emission Standard for Hazardous Air Pollutants (NESHAP). July, 2000.
Edison Electric Institute (EEI). 1998. Statistical Yearbook of the Electric Utility Industry
1997. November. Washington, DC.
E.H. Pechan & Associates, Inc. 1997. Qualitative Market Impact Analysis for
Implementation of the Selected Ozone and PM NAAQS. Appendix B. Prepared for
the U.S. Environmental Protection Agency.
Fleming, Ian. 1992. "Just How Clean is your Steam?: Minimizing Risk of Product
Contamination and Prolonging Filter Life." Manufacturing Chemist 63(10):37-39.
Giles, Ellen F. 2000. Platt's Directory of Electric Power Producers and Distributors, 109th
Edition of the Electrical World Directory. New York: McGraw Hill.
R-l
-------
Haltmaier, Susan. 1998. "Electricity Production and Sales." In U.S. Industry & Trade
Outlook '98, DRI/McGraw-Hill, Standard & Poor's, and U.S. Department of
Commerce/international Trade Administration. New York: McGraw-Hill, pp. 5-1 to
5-9.
Industrial Combustion Coordinated Rulemaking (ICCR). 1998. Data/Information Submitted
to the Coordinating Committee at the Final Meeting of the Industrial Combustion
Coordinated Rulemaking Federal Advisory Committee. EPA Docket Numbers
A-94-63, II-K-4b2 through -4b5. Research Triangle Park, North Carolina.
September 16-17.
Lemm, Jamie. 2000. "Household Furniture." In U.S. Industry & Trade Outlook 2000. New
York: DRI/McGraw-Hill, Standard & Poor's, and U.S. Department of
Commerce/International Trade Administration.
Plant Engineering. 1991. "Boiler Systems." Plant Engineering 45(14):92-94.
Texas Utilities (TXU). 2000. "Generating Electricity: Steam Turbines." As obtained in
September 2000. .
U.S. Bureau of Economic Analysis. 1997. Regional Economic Information System:
1969-1995 on CD-ROM [machine-readable data files]/prepared by the Bureau of
Economic Analysis, U.S. Department of Commerce. Washington, DC: The Bureau
of Economic Analysis [producer and distributor], 1997.
. As obtained February 2002.
U.S. Environmental Protection Agency. July 1999: "Draft Revised Guidelines for
Carcinogen Risk Assessment." NCEA-F-0644. USEPA, Risk Assessment Forum.
Pp 3-9ff. Http://www.epa.gov/ncea/raf/pdfs/cancer_gls.pdf.
U.S. Department of Agriculture, National Agriculture Statistics Service. 1997 Census of
Agriculture, . As
obtained February 2002.
U.S. Department of Commerce, Bureau of the Census. 1990-1998. Annual Survey of
Manufactures [Multiple Years]. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1990a. 1987 Census of
Manufactures, Industry Series: Drug Industry. Washington, DC: Government
Printing Office.
R-2
-------
U.S. Department of Commerce, Bureau of the Census. 1990b. 1987 Census of
Manufactures, Industry Series, Paints and Allied Products. Washington, DC:
Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1990c. 1987 Census of
Manufactures, Industry Series: Pulp, Paper, and Board Mills. Washington, DC:
Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1991. 1987 Census of Manufactures,
Subject Series: General Summary. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1995a. 1992 Census of
Manufactures, Industry Series: Drug Industry. Washington, DC: Government
Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1995b. 1992 Census of
Manufactures, Industry Series: Industrial Organic Chemicals. Washington, DC:
Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1995c. 1992 Census of
Manufactures, Industry Series: Pulp, Paper, and Board Mills. Washington, DC:
Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1995d. 1992 Concentration Ratios in
Manufacturing. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1996. 1992 Census of Manufactures,
Subject Series: General Summary. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. 1998. Survey of Plant Capacity:
1996. Washington, DC: Government Printing Office.
U.S. Department of Commerce, Bureau of the Census. Economy-Wide Key Statistics: 1997.
. As obtained February 2001.
U.S. Department of Commerce, Bureau of the Census. 2001. "1997 Economic
Census—United States." As obtained on March 13, 2001.
.
R-3
-------
U.S. Department of Energy, Energy Information Administration (EIA). 1996. Electric
Power Annual 1995. Volumes I and II. DOE/EIA-0348(95)/1. Washington, DC:
U.S. Department of Energy.
U.S. Department of Energy, Energy Information Administration (EIA). 1997. Financial
Statistics of Major U.S. Investor-Owned Electric Utilities, 1996. Washington, DC:
U.S. Department of Energy.
U.S. Department of Energy, Energy Information Administration (EIA). 1998a. The
Changing Structure of the Electric Power Industry: Selected Issues, 1998.
DOE/EIA-0562(98). Washington, DC: U.S. Department of Energy.
U.S. Department of Energy, Energy Information Administration (EIA). 1998b. Financial
Statistics of Major Publicly Owned Electric Utilities, 1997. Washington, DC: U.S.
Department of Energy.
U.S. Department of Energy, Energy Information Administration (EIA). "Annual Energy
Outlook 1999—Market Trend—Electricity." . As accessed November 15, 1999a.
U.S. Department of Energy, Energy Information Administration (EIA). 1999b. The
Changing Structure of the Electric Power Industry 1999: Mergers and Other
Corporate Combinations. Washington, DC: U.S. Department of Energy.
U.S. Department of Energy, Energy Information Administration (EIA), Office of Integrated
Analysis and Forecasting. "Competitive Electricity Price Projections."
. As obtained on November 15,
1999c.
U.S. Department of Energy, Energy Information Administration (EIA). 1999d. Electric
Power Annual, 1998. Volumes I and n. Washington, DC: U.S. Department of
Energy.
U.S. Department of Energy, Energy Information Administration (EIA). "Issues in Midterm
Analysis and Forecasting 1999—Table 1." . As obtained on May 8, 2000a.
U.S. Department of Energy, Energy Information Administration (EIA). January 2000b.
Model Documentation Report: Industrial Sector Demand Module of the National
R-4
-------
Energy Modeling System. DOE/EIA-M064(2000). Washington, DC: U.S.
Department of Energy.
U.S. Department of Energy, Energy Information Administrations (EIA). 2002. "Annual
Energy Outlook 2002." . As obtained
on February 12, 2002.
U.S. Department of Justice. 1992. Horizontal Merger Guidelines. Washington, DC: U.S.
Department of Justice.
U.S. Environmental Protection Agency (EPA). 1993. Alternative Control Techniques
Document-NO.,. Emissions from Process Heaters (Revised). Washington, DC: U.S.
Environmental Protection Agency.
U.S. Environmental Protection Agency (EPA), Office of Compliance Sector Notebook
Project. 1995a. Profile of the Lumber and Wood Products Industry. Washington,
DC: U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency (EPA), Office of Compliance Sector Notebook
Project. 1995b. Profile of the Pulp and Paper Industry. Washington, DC: U.S.
Environmental Protection Agency.
U.S. Environmental Protection Agency (EPA), Office of Compliance Sector Notebook
Project. 1995c. Profile of the Organic Chemical Industry. Washington, DC: U.S.
Environmental Protection Agency.
U.S. Environmental Protection Agency (EPA). 1997a. EPA Office of Compliance Sector
Notebook Project: Profile of the Pharmaceutical Manufacturing Industry.
Washington, DC: U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency (EPA). 1997b. Regulatory Impact Analysis of Air
Pollution Regulations: Utility and Industrial Boilers. Research Triangle Park, NC:
U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.
Industrial Combustion Coordinated Rulemaking, Inventory Database V4—Process
Heaters. November 13, 1998.
U.S. Environmental Protection Agency (EPA). 1999. OAQPSEconomic Analysis Resource
Document. Durham, NC: Innovative Strategies and Economics Group.
R-5
-------
U.S. Environmental Protection Agency (EPA). July 1999. "Draft Revised Guidelines for
Carcinogen Risk Assessment." NCEA-F-0644. USEPA, Risk Assessment Forum.
pp 3-9ff. http://www.epa.gov/ncea/raf/pdfs/cancer_gls.pdf.
U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.
Industrial Combustion Coordinated Rulemaking, Inventory Database V4.1-Boilers.
February 26, 1999.
U.S. International Trade Commission (USITC). November 21, 2001. Memorandum to the
Commission from Craig Thomsen, John Giamalva, John Benedetto, Joshua Levy,
International Economists. Investigation No. TA-201-73: STEEL-Remedy
Memorandum.
Warfield, et al. 2001. "Multifiber Arrangement Phaseout: Implications for the U.S.
Fibers/Textiles/Fabricated Products Complex."
. As obtained September 19,
2001.
R-6
-------
APPENDIX A:
ECONOMIC MODEL OF MARKETS AFFECTED BY THE BOILERS AND
PROCESS HEATERS MACT
The primary purpose of the EIA for the rule is to describe and quantify the economic
impacts associated with the rule. The Agency used a basic framework that is consistent with
economic theory and the analyses performed for other rules to develop estimates of these
impacts. This approach employs standard microeconomic concepts to model behavioral
responses expected to occur with regulation. This appendix describes the spreadsheet model
in more detail and discusses how the Agency
• collected the baseline data set from the Annual Energy Outlook 2002 (DOE, EIA,
2002), U.S. Census Bureau (U.S. Department of Commerce, 2001), and U.S.
Department of Agriculture (USD A, 2002).
• characterized market supply and demand for each market and specified links
between the energy and agricultural, manufacturing, mining, and commercial
markets.
• introduced a policy "shock" into the model by using control cost-induced shifts in
the supply functions, and
• used a solution algorithm to determine a new with-regulation equilibrium for each
market.
A.I Baseline Data Set
EPA collected the following data to characterize the baseline year, 2005:
• Energy Market Data—The Department of Energy's Supplemental Tables to the
Annual Energy Outlook 2002 report forecasts of price, quantity, and fuel
intensities used to calibrate the model.
• Agriculture, Mining, Manufacturing, Commercial Sectors—EPA obtained
shipment data from the 1997 Economic Census and 1997 Agriculture Census.
We then used annual growth rates reported by the Bureau of Economic Analysis
(BEA, 1997) to estimate baseline shipment data for 2005. The Agency selected
units for output such that the price in each market equals one. We computed
energy demand using fuel intensity data reported in the AEO 2002.
A-l
-------
• Supply and Demand Elasticities—The supply and demand elasticity values used
in the market model are reported in Table 5-2 of this report. Given the
uncertainties regarding these parameters, EPA also conducted several sensitivity
analyses and report these results in Appendix B.
A.2 Multi-Market Model
The model includes four energy markets (coal, electricity, natural gas, and petroleum)
and 24 goods and service markets. The following sections describe model equations the
Agency developed to characterize these markets and estimate welfare changes resulting from
the rule.
A. 1.1 Supply Side Modeling
EPA estimated the change in quantity supplied as follows:
n
Ap-c - Y, OjApj
Aqs = q0s • 8s • ^
Po
where q s is the baseline quantity, Es is the domestic supply elasticity, the term
n
Ap-c - Y a'Ap- is the change in the producer's net price, and p0 is the baseline price.
The change Jn net price is composed of the change in baseline price resulting from the
regulation, the direct shift in the supply function resulting from compliance costs, and the
indirect shift in the supply function resulting from changes in input prices in energy market
(j). The fuel share is allowed to vary using a fuel switching rule relying on cross-price
elasticities of demand between energy sources.
A. 1.1.2 Producer Welfare Measurement
EPA approximated the change in producer surplus with the following equation:
n n
APS = qi-(Ap-c- X ^App - 0.5 -Aq- (Ap-c- £ OjApp (A.2)
Increased control costs, higher energy input costs, and output declines have a
negative effect on domestic producer surplus. However, these losses are mitigated to some
degree as a result of higher market prices.
A-2
-------
A. 1.2 Energy Demand Side Modeling
Market demand in the energy markets is expressed as the sum of the energy,
residential, agriculture, manufacturing, mining, commercial, and transportation sectors:
QDJ = .1 qDji > (A-3)
where j indexes the energy market and i indexes the consuming sector. The change in
residential quantity demanded of energy market j can be approximated as follows:
Di
Di Di Pi
] ' ' ' —
i
' = % ] ' TI ' ' — (A.4)
PJO
where q DJ is baseline consumption, T|Dj is the residential demand elasticity and (Ap) is the
change in the market price.
In contrast, energy demand from energy, agricultural, manufacturing, mining,
commercial, and transportation sectors is modeled as a derived demand resulting from the
production and consumption choices in these industries. Energy demand responds to
changes in sector output and fuel switching that occurs in response to changes in relative
energy prices. For each of these sectors, energy demand is expressed as follows:
BTU..
BTUJU = - * • FSW • qu (A.5)
%
where BTU is demand for energy market j from sector i, q is sector i's output, and FSW is a
factor generated by the fuel switching algorithm. The subscripts 0 and 1 represent baseline
and with regulation conditions, respectively.
A.I. 3 Agriculture, Manufacturing, Mining, Commercial, and Transportation Demand
Side Modeling
The change in quantity demanded in these markets can be approximated as follows:
A DJ Dj DJ ?!
Aq ' = % ' • TI ' • (A.6)
PiO
A-3
-------
where q D' is baseline output, T|D is the demand elasticity of the respective market (i) and (A
p;) is the change in the market price.
The change in consumer surplus in markets is approximated as follows:
ACS = - qj-Ap + 0.5-Aq-Ap (A.7)
As shown, higher market prices and reduced consumption lead to welfare losses for
consumers.
A.2 With-Regulation Market Equilibrium Determination
Market adjustments can be conceptualized as an interactive feedback process.
Supply segments face increased production costs as a result of the rule and are willing to
supply smaller quantities at the baseline price. This reduction in market supply leads to an
increase in the market price that all producers and consumers face, which leads to further
responses by producers and consumers and thus new market prices. The new with-regulation
equilibrium is the result of a series of iterations in which price is adjusted and producers and
consumers respond, until a set of stable market prices arises where total market supply equals
market demand (i.e., Qs = QD) in each market. Market price adjustment takes place based on
a price revision rule that adjusts price upward (downward) by a given percentage in response
to excess demand (excess supply).
The algorithm for determining with-regulation equilibria can be summarized by
seven recursive steps:
1. Impose the control costs on affected supply segments, thereby affecting their
supply decisions.
2. Recalculate the market supply in each market. Excess demand currently exists.
3. Determine the new prices via a price revision rule.
4. Recalculate market supply with new prices, accounting for fuel switching choices
associated with new energy prices.
5. Compute market demand in each market.
A-4
-------
6. Compare supply and demand in each markets. If equilibrium conditions are not
satisfied, go to Step 3, resulting in a new set of market prices. Repeat until
equilibrium conditions are satisfied (i.e., the ratio of supply to demand is
arbitrarily close to one).
A-5
-------
APPENDIX B
ASSUMPTIONS AND SENSITIVITY ANALYSIS
In developing the economic model to estimate the impacts of the industrial/
commercial/institutional boilers and process heaters NESHAP, several assumptions were
necessary to make the model operational. This appendix lists and explains the major model
assumptions and describes their potential impact on the analysis results. Sensitivity analyses
are presented for numeric assumptions.
Assumption: The domestic markets for goods and services are all perfectly competitive.
Explanation: Assuming that these markets are perfectly competitive implies that the
producers of these products are unable to unilaterally affect the prices they receive for their
products. Because the industries used in this analysis are aggregated across a large number of
individual producers, it is a reasonable assumption that the individual producers have a very
small share of industry sales and cannot individually influence the price of output from that
industry.
Possible Impact: If these product markets were in fact imperfectly competitive, implying
that individual producers can exercise market power and thus affect the prices they receive
for their products, then the economic model would understate possible increases in the price
of final products due to the regulation as well as 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, and producer
surplus losses would be smaller in the final product markets.
Assumption: Market Supply and Demand Elasticity Uncertainty
Explanation: The goods and service markets are modeled at the two or three-digit NAICS
code level to operationalize the economic model. Because of the high level of aggregation,
only limited data on elasticities of supply and demand estimates are available. However,
these elasticities strongly influence the distribution of economic impacts between producers
and consumers.
B-l
-------
Sensitivity Analysis: Tables B-la and Table B-lb show how the economic impact estimates
vary as the supply and demand elasticities for goods and services change by 25 percent.
Table B-la. Sensitivity Analysis: Supply and Demand Elasticities in the Goods and
Services Markets
Change Supply
Demand Constant
Change in consumer surplus
Change in producer surplus
Change in social welfare
25% Decrease
-367.8
-495.2
-862.9
Elasticities Reported
in Section 6
-414.3
-448.7
-862.9
25% Increase
-450.5
-412.4
-862.9
Table B-lb. Sensitivity Analysis: Supply and Demand Elasticities in the Goods and
Services Markets
Supply Constant Elasticities Reported
Demand Change 25% Decrease in Section 6 25% Increase
Change in consumer surplus -462.7 -414.3 -364.4
Change in producer surplus -400.2 -448.7 -498.5
Change in social welfare -862.9 -862.9 -862.9
Assumption: Cross-price elasticities of demand for fuels are based on 2015 NEMS
projections.
Explanation: Cross- and own-price elasticities of demand from NEMS were used to capture
fuel switching in the manufacturing sectors in the economic model. As shown in Table 5-2,
allowing manufacturers to switch fuels in response to changes in relative energy prices
decreases the change in social welfare by approximately 10 percent. However, the NEMS
projection reflects aggregate behavioral responses in the year 2015. Because this is a longer
B-2
-------
window of analysis compared to the baseline year 2005, this analysis may overestimate firms'
ability to switch fuels in the short run.
Sensitivity Analysis: Table B-2 shows how the economic impact estimates vary as the own-
and cross-price elasticities used in the EIA are reduced by 50 percent and 75 percent.
Table B-2. Sensitivity Analysis: Own- and Cross-Price Elasticities Used to Model Fuel
Switching
Change in consumer surplus
Change in producer surplus
Change in social welfare
Fuel Price Elasticities
Presented in Table 5-2
-414.3
-448.7
-862.9
Reduced by
50 Percent
-414.6
-448.4
-862.9
Reduced by 75
Percent
-414.9
-448.0
-862.9
Assumption: The domestic markets for energy are perfectly competitive.
Explanation: Assuming that the markets for energy are perfectly competitive implies that
individual producers are not capable of unilaterally affecting the prices they receive for their
products. Under perfect competition, firms that raise their price above the competitive price
are unable to sell at that higher price because they are a small share of the market and
consumers can easily buy from one of a multitude of other firms that are selling at the
competitive price level. Given the relatively homogeneous nature of individual energy
products (petroleum, coal, natural gas, electricity), the assumption of perfect competition at
the national level seems to be appropriate.
Possible Impact: If energy markets were in fact imperfectly competitive, implying that
individual producers can exercise market power and thus affect the prices they receive for
their products, then the economic model would understate possible increases in the price of
energy due to the regulation as well as the social costs of the regulation. Under imperfect
competition, energy producers would be able to pass along more of the costs of the regulation
to consumers; thus, consumer surplus losses would be greater, and producer surplus losses
would be smaller in the energy markets.
B-3
-------
Assumption: The elasticity of supply in the electricity market for existing sources is
approximately 0.75.
Explanation: The price elasticity of supply in the electricity markets represents the
behavioral responses from existing sources to changes in the price of electricity. However,
there is no consensus on estimates of the price elasticity of supply for electricity. This is in
part because, under traditional regulation, the electric utility industry had a mandate to serve
all its customers and utilities were compensated on a rate-based rate of return. As a result,
the market concept of supply elasticity was not the driving force in utilities' capital
investment decisions. This has changed under deregulation. The market price for electricity
has become the determining factor in decisions to retire older units or to make higher cost
units available to the market.
Sensitivity Analysis: Table B-3 shows how the economic impact estimates vary as the
elasticity of supply in the electricity markets varies.
Table B-3. Sensitivity Analysis: Elasticity of Supply in the Electricity Markets
ES = 0.5
ES = 0.75
ES =
Change in consumer surplus
Change in producer surplus
Change in social welfare
-405.0
-457.9
-862.9
-414.3
-448.7
-862.9
-419.6
-443.4
-862.9
B-4
-------
APPENDIX C
ECONOMIC ANALYSIS OF REGULATORY ALTERNATIVE: OPTION 1A
EPA examined one above-the-floor alternative referred to in this appendix as Option
1A. Option 1A broadens the scope of affected units to include those fueled by residual fuel
oil and units of covered fuel types with input capacities less than 10 million Btus. In this
appendix, we describe the engineering compliance costs associated with this option, estimate
the size and distribution of social cost, and report the results of the small entity screening
analysis
C.I Engineering Cost Analysis: Affected Population and Cost Estimates
The entire Inventory Database contains more than 58,000 ICI boilers and process
heaters. The number of units included in the profile was 3,580 for Option 1A. As shown in
Table C-l, the industries with the largest number of potentially affected units are the
furniture, paper, lumber, and electrical services industries. These four industries alone
account for nearly 60 percent of affected units. Almost all the process heaters are in the
lumber industry. The remaining units are primarily distributed across the manufacturing
sector and service industries. The distribution of units affected by the Option 1A alternative
is similar to the final rule, although both the number of units and the number of facilities is
greater for Option 1 A.
We describe the technical characteristics of existing boilers affected under Option 1A
below (see Figure C-l). These characteristics include capacity range, fuel type, and level of
preexisting controls.
• Capacity Range: About half of the 3,580 units affected by this alternative have
input capacities between 10 and 100 MMBtu/hr. Twenty percent have capacities
between 100 and 250, 16 percent have capacities greater than 250, and 13 percent
have capacities less than 10 MMBtu/hr.
• Fuel Type: Coal and residual fuel oil are the primary fuel types each accounting
for slightly less than one-third of the units. The remaining third primarily
consists of units that consume wood or some other type of biomass fuel.
C-l
-------
Table C-l. Units and Facilities Affected by the Option 1A Alternative by Industry"
SIC
Code
01
02
07
10
12
13
14
17
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
42
46
NAICS
Code
111
112
115
212
212
211
212
235
311
312
313
315
321
337
322
511
325
324
326
316
327
331
332
333
335
336
334
339
482
484
486
Description
Agriculture — Crop s
Agriculture — Livestock
Agricultural Services
Metal Mining
Coal Mining
Oil and Gas Extraction
Mining/Quarrying — Nonmetallic Minerals
Construction — Special Trade Contractors
Food and Kindred Products
Tobacco Products
Textile Mill Products
Apparel and Other Products from Fabrics
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Printing, Publishing, and Related Industries
Chemicals and Allied Products
Petroleum Refining and Related Industries
Rubber and Miscellaneous Plastics Products
Leather and Leather Products
Stone, Clay, Glass, and Concrete Products
Primary Metal Industries
Fabricated Metal Products
Industrial Machinery and Computer Equipment
Electronic and Electrical Equipment
Transportation Equipment
Scientific, Optical, and Photographic Equip.
Miscellaneous Manufacturing Industries
Railroad Transportation
Motor Freight and Warehousing
Pipelines, Except Natural Gas
Boilers
6
0
0
10
2
8
10
2
163
22
247
4
434
310
503
8
332
54
56
22
40
83
44
46
45
158
33
14
4
5
3
Heaters
0
0
0
1
0
10
0
0
0
0
3
0
28
0
0
0
101
108
0
0
2
2
0
0
0
0
0
0
0
2
3
Total
Units
6
0
0
11
2
18
10
2
163
22
250
4
462
310
503
8
433
162
56
22
42
85
44
46
45
158
33
14
4
7
6
Facilities
6
0
0
5
1
4
5
1
72
11
134
4
337
209
272
6
163
50
37
12
25
33
28
25
29
61
16
10
1
3
5
(continued)
C-2
-------
Table C-l. Units and Facilities Affected by the Option 1A Alternative by Industry"
(continued)
SIC
Code
49
50
51
55
58
60
59
70
72
76
80
81
82
83
86
87
89
91
92
94
96
97
NA
NAICS
Code
221
421
422
441
722
522
445-454
721
812
811
621
541
611
624
813
541
711/514
921
922
923
926
928
Description
Electric, Gas, and Sanitary Services
Wholesale Trade — Durable Goods
Wholesale Trade — Nondurable Goods
Automotive Dealers and Gasoline Service
Stations
Eating and Drinking Places
Depository Institutions
Miscellaneous Retail
Hotels and Other Lodging Places
Personal Services
Miscellaneous Repair Services
Health Services
Legal Services
Educational Services
Social Services
Membership Organizations
Engineering, Accounting, Research,
Management and Related Services
Services, N.E.C.
Executive, Legislative, and General
Administration
Justice, Public Order, and Safety
Administration of Human Resources
Administration of Economic Programs
National Security and International Affairs
SIC Information Not Available
Boilers
371
3
2
0
0
0
1
1
0
2
40
0
114
3
0
6
2
2
33
1
4
41
24
3,318
Heaters
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
262
Total
Units
372
3
2
1
0
0
1
1
0
2
40
0
114
3
0
6
2
2
33
1
4
41
24
3,580
Facilities
185
2
1
1
0
0
1
1
0
1
19
0
50
2
0
5
1
2
10
1
3
13
18
1,881
a Based on the Inventory Database.
C-3
-------
Option 1A Alternative (n=3,58Q)
>250
16%
0 to 10
13%
Fabric Filter
10%
100 to 250
20%
10 to 100
51% wet
Scrubber
8%
No Control
41%
Residual
Fuel Oil
32%
Cyclone
26%
Bagasse
2%
Input Capacity
(million Btu)
Preexisting Control
Technology
Coal and
Wood
5%
Fuel Type
Figure C-l. Characteristics of Units Affected
Control Level: Forty-one percent have no existing pollution control equipment
installed. Typical control devices include fabric filters, wet scrubbers, and
electrostatic precipitators.
• Fuel Type: This alternative includes those units affected under Option 1 A, as
well as a large number of natural gas units that were not affected under Option
1A. The vast majority of the 78 percent of the total number of potentially
affected units are fueled by natural gas.
• Control Level: Eighty-eight percent of the affected units have no preexisting
control equipment.
The Agency estimates that in 2005, 9,163 units (existing units and new units) may be
affected by the Option 1 A. These populations were used to estimate national engineering
costs. As shown in Table C-2, the cost of controls for Option 1A is $1,995.8 million, with an
average per-unit cost of $218,000.
C-4
-------
Table C-2. Unit Cost and Population Estimates for the Option 1A by Industry, 2005
SIC
Code
01
02
07
10
12
13
14
17
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
42
NAICS
Code
111
112
115
212
212
211
212
235
311
312
313
315
321
337
322
511
325
324
326
316
327
331
332
333
335
336
334
339
482
484
Description
Agriculture — Crop s
Agriculture — Livestock
Agricultural Services
Metal Mining
Coal Mining
Oil and Gas Extraction
Mining/Quarrying — Nonmetallic Minerals
Construction — Special Trade Contractors
Food and Kindred Products
Tobacco Products
Textile Mill Products
Apparel and Other Products from Fabrics
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Printing, Publishing, and Related Industries
Chemicals and Allied Products
Petroleum Refining and Related Industries
Rubber and Miscellaneous Plastics Products
Leather and Leather Products
Stone, Clay, Glass, and Concrete Products
Primary Metal Industries
Fabricated Metal Products
Industrial Machinery and Computer
Equipment
Electronic and Electrical Equipment
Transportation Equipment
Scientific, Optical, and Photographic
Equipment
Miscellaneous Manufacturing Industries
Railroad Transportation
Motor Freight and Warehousing
Total
Option
1A Units
11
—
—
34
6
137
31
2
376
56
673
10
620
421
1,050
37
1,359
677
178
66
154
271
165
151
167
453
104
37
9
19
Units
Percent
0.12%
0.00%
0.00%
0.37%
0.06%
1.50%
0.34%
0.03%
4.10%
0.61%
7.34%
0.11%
6.77%
4.60%
11.46%
0.40%
14.83%
7.38%
1.94%
0.72%
1.68%
2.95%
1.80%
1.65%
1.82%
4.95%
1.13%
0.41%
0.10%
0.21%
Total Cost
Option 1A Costs
(by Unit)
$1,633,841
—
—
$8,952,098
$683,026
$6,070,001
$17,958,177
$230,525
$122,487,346
$13,685,614
$147,094,726
$1,213,586
$89,961,854
$50,045,573
$323,736,302
$1,824,933
$293,027,205
$73,172,001
$18,100,195
$6,924,480
$17,509,996
$65,174,064
$22,066,661
$26,418,385
$18,770,867
$107,402,909
$13,638,983
$4,222,427
$2,240,871
$3,475,610
Percent
0.08%
0.00%
0.00%
0.45%
0.03%
0.30%
0.90%
0.01%
6.14%
0.69%
7.37%
0.06%
4.51%
2.51%
16.22%
0.09%
14.68%
3.67%
0.91%
0.35%
0.88%
3.27%
1.11%
1.32%
0.94%
5.38%
0.68%
0.21%
0.11%
0.17%
(continued)
C-5
-------
Table C-2. Unit Cost and Population Estimates for the Option 1A by Industry, 2005
(continued)
SIC
Code
46
49
50
51
55
58
59
60
70
72
76
80
81
82
83
86
87
89
91
92
94
96
97
NA
State
NAICS
Code
486
221
421
422
441
722
445-454
522
721
812
811
621
541
611
624
813
541
711/514
921
922
923
926
928
Description
Pipelines, Except Natural Gas
Electric, Gas, and Sanitary Services
Wholesale Trade — Durable Goods
Wholesale Trade — Nondurable Goods
Automotive Dealers and Gasoline Service
Stations
Eating and Drinking Places
Miscellaneous Retail
Depository Institutions
Hotels and Other Lodging Places
Personal Services
Miscellaneous Repair Services
Health Services
Legal Services
Educational Services
Social Services
Membership Organizations
Engineering, Accounting, Research,
Management and Related Services
Services, N.E.C.
Executive, Legislative, and General
Administration
Justice, Public Order, and Safety
Administration of Human Resources
Administration of Economic Programs
National Security and International Affairs
SIC Information Not Available
Parent is a state government
Total
Option 1A
Units
19
865
6
4
2
—
3
—
2
1
4
93
—
273
8
—
49
2
5
77
2
8
96
368
—
Units
Percent
0.21%
9.44%
0.07%
0.04%
0.02%
0.00%
0.03%
0.00%
0.02%
0.01%
0.05%
1.01%
0.00%
2.98%
0.08%
0.00%
0.54%
0.02%
0.06%
0.85%
0.02%
0.09%
1.05%
4.01%
0.00%
9,163
Total Cost
Option 1A Costs
(by Unit) Percent
$1,959,589
$331,479,389
$2,675,296
$2,693,380
$195,421
—
$259,585
—
$849,114
$7,840
$1,120,435
$22,545,605
—
$91,770,778
$1,448,405
—
$5,016,627
$1,211,582
$845,423
$21,308,885
$314,316
$4,200,975
$36,080,306
$12,099,975
—
0.10%
16.61%
0.13%
0.13%
0.01%
0.00%
0.01%
0.00%
0.04%
0.00%
0.06%
1.13%
0.00%
4.60%
0.07%
0.00%
0.25%
0.06%
0.04%
1.07%
0.02%
0.21%
1.81%
0.61%
0.00%
$1,995,805,181
C-6
-------
C.2 Economic Impact Analysis Results
As shown in Table C-3, EPA estimates the welfare impacts are over twice as high as
the MACT floor ($1,995.5 million). The market analysis shows that consumers will bear
$955.3 million, or 48 percent of the total social cost as a result of higher prices and lower
consumption levels. Producer surplus is projected to decrease by $1,040.2 million, or 52 of
the total social cost as result of direct control costs, higher energy costs, and reductions in
output.
Table C-3. Social Cost Estimates ($1998 106)
Change in Social Change in Social
Welfare, MACT Floor Welfare, Option 1A
Baseline engineering costs
Social costs with market adjustments
Difference between engineering and social
costs
$863.0
$862.9
$0.1
$1,995.8
$1,995.5
$0.3
With exception of the natural gas market, energy producers are expected to
experience producer surplus losses. Electricity, petroleum, and coal producer surplus is
projected to decline by approximately $113 million under Option 1A. In contrast, natural
gas producer surplus is projected to increase by $2 million as they benefit from increased
demand from industries switching from petroleum and electricity (Table C-4).
The majority welfare impacts fall on the agriculture, manufacturing, and mining
industries and EPA estimates losses of $1,444.3 million for these sectors. Manufacturing
industries with large number of boilers and process heaters and industries that consume
electricity experience the majority these losses (e.g., chemicals and allied products, paper,
textile mill products, and food). Consumers in these industries experience losses of $709.9
million and producers bear $734.4 million.
EPA also examined the impact on the commercial, transportation and residential
sectors. We project the commercial sector has the highest welfare losses among the three
($302 million) with commercial customers bearing approximately 42 percent of these losses,
or $129 million. EPA estimates similar consumer surplus loss ($92 million) for residential
C-7
-------
Table C-4. Distribution of Social Costs by Sector/Market: Option 1A ($1998 106)
Sectors/Markets
Energy Markets
Petroleum
Natural gas
Electricity
Coal
Subtotal
NAICS Code
311
312
313
314
315
316
321
322
323
325
326
327
331
332
333
334
335
336
337
339
11
23
21
48
42; 44-45; 49; 51-
56; 61-62; 71-72;
81
Grand Total
SIC Code
20 (pt)
20 (pt); 21
22 (pt)
22 (pt)
23
31
24
26
27
28
30
32
33
34
35
36 (pt)
36 (pt)
37
25
39
01-08
15-17
10; 14
40-47 (pt)
40-48 (pt);
50-99
Description
Food
Beverage and Tobacco Products
Textile Mills
Textile Product Mills
Apparel
Leather and Allied Products
Wood Products
Paper
Printing and Related Support
Chemicals
Plastics and Rubber Products
Nonmetallic Mineral Products
Primary Metals
Fabricated Metal Products
Machinery
Computer and Electronic Products
Electrical Equipment, Appliances,
and Components
Transportation Equipment
Furniture and Related Products
Miscellaneous
Agricultural Sector
Construction Sector
Other Mining Sector
Transportation
Commercial
Residential
Producer
Surplus
-$27.3
$2.4
-$79.5
-$6.4
-$110.8
-$90.0
-$5.4
-$45.0
-$0.1
-$0.9
-$2.7
-$72.0
-$173.1
-$0.4
-$102.4
-$6.1
-$9.1
-$59.5
-$18.6
-$17.1
-$12.0
-$11.7
-$47.8
-$9.2
-$3.2
-$1.5
-$3.2
-$18.9
-$24.1
-$129.3
NA
-$955.3
Change in:
Consumer
Surplus
-$36.0
-$9.3
-$103.2
-$0.3
-$2.1
-$4.3
-$19.2
-$157.2
-$1.0
-$204.7
-$14.6
-$10.9
-$13.6
-$5.0
-$11.4
-$4.8
-$7.8
-$63.7
-$41.8
-$2.5
-$3.6
-$4.3
-$13.1
-$22.5
-$172.5
-$92.0
-$1,040.2
Social
Welfare
-$126.0
-$14.7
-$148.2
-$0.4
-$3.0
-$7.1
-$91.2
-$330.3
-$1.4
-$307.1
-$20.7
-$20.0
-$73.1
-$23.6
-$28.5
-$16.8
-$19.6
-$111.4
-$51.0
-$5.7
-$5.1
-$7.5
-$32.0
-$46.5
-$301.8
-$92.0
-$1,995.5
NA = Not applicable.
C-8
-------
energy consumers. Finally, the total welfare loss for the transportation sector is estimated to
be $46.5 million, with transportation customers bearing slightly more than half of these
losses (51 percent). However, all of these losses (consumer or producer) for these sectors
represent less than 0.05 percent of baseline value of consumption or shipments.
C. 2.1 Market-Level Impacts
Increases in the costs of production in the energy and final product markets due to the
regulation are expected to result in changes in prices, production, and consumption from
baseline levels. As shown in Table C-5, the electricity market price increases by 0.11
percent, while production/consumption decreases by 0.03 percent as a result of additional
control costs. A significant share of electricity is produced in the United States using coal as
a primary input. Therefore, projected reductions in electricity production also lead to a
decrease in demand for coal. As a result, the price and quantities of coal are projected to fall
by 0.02 percent and 0.02 percent, respectively. In the petroleum market, the model projects
small price and quantity effects (i.e., less than 0.02 percent). In the natural gas market, the
model projects the market price will rise in response to increased demand (0.01 percent).
The price increase is the result of additional control costs and increased demand. Production
and consumption quantities also slightly increase in this market as a result of increased
demand.
Additional control costs and higher energy costs associated with the regulation lead
to higher goods and services prices in all markets and a decline in output. However, the
changes are generally very small. Under Option 1 A, three markets have price increases
greater than or equal to 0.05 percent—Wood Products (NAICS 321), Paper (NAICS 322),
and Textile Mills (NAICS 313). The producers in these sectors are expected to face higher
per-unit control costs relative to other industries. In addition, these industries are also
electricity-intensive; therefore, costs of production also increase as a result of higher
electricity prices.
C.2.2 Executive Order 13211 (Energy Effects)
EPA has provided additional information on the impacts of the rule on affected
energy markets below.1
'Conversion factors for heat rates were obtained from AEO 2002, Appendix H. These factors vary by year to
year; 2010 values are reported in this appendix.
C-9
-------
Table C-5. Market-Level Impacts: Option 1A
Option 1A
Percent Change
Sectors/Markets
Energy Markets
Petroleum
Natural gas
Electricity
Coal
NAICS Code
311
312
313
314
315
316
321
322
323
325
326
327
331
332
333
334
335
336
337
339
11
23
21
48
42; 44-45; 49; 51-
56; 61-62; 71-72;
81
SIC Code
20 (pt)
20 (pt); 21
22 (pt)
22 (pt)
23
31
24
26
27
28
30
32
33
34
35
36 (pt)
36 (pt)
37
25
39
01-08
15-17
10; 14
40-47 (pt)
40-48 (pt);
50-99
Description
Food
Beverage and Tobacco Products
Textile Mills
Textile Product Mills
Apparel
Leather and Allied Products
Wood Products
Paper
Printing and Related Support
Chemicals
Plastics and Rubber Products
Nonmetallic Mineral Products
Primary Metals
Fabricated Metal Products
Machinery
Computer and Electronic Products
Electrical Equipment, Appliances, and Components
Transportation Equipment
Furniture and Related Products
Miscellaneous
Agricultural Sector
Construction Sector
Other Mining Sector
Transportation
Commercial
Price
0.019%
0.005%
0.108%
-0.020%
0.019%
0.007%
0.050%
0.000%
0.001%
0.025%
0.075%
0.068%
0.000%
0.021%
0.003%
0.009%
0.026%
0.007%
0.005%
0.002%
0.009%
0.007%
0.013%
0.003%
0.001%
0.000%
0.023%
0.007%
0.001%
Quantity
-0.005%
0.001%
-0.026%
-0.024%
-0.006%
-0.009%
-0.043%
0.000%
-0.001%
-0.030%
-0.015%
-0.074%
-0.001%
-0.032%
-0.005%
-0.008%
-0.021%
-0.001%
-0.002%
-0.001%
-0.004%
-0.007%
-0.044%
-0.002%
-0.001%
0.000%
-0.007%
-0.005%
-0.001%
pt = Part.
C-10
-------
Energy Price Effects. As described in the market-level results section, electricity
prices are projected to increase by less than 1 percent. Petroleum and natural gas prices are
all projected to increase by less than 0.1 percent. The price of coal is projected to decrease
slightly.
Impacts on Electricity Supply, Distribution, and Use. We project the increased
compliance costs for the electricity market will result in an annual production decline of
approximately 980 million kWh under Option 1 A.
Impacts on Petroleum, Natural Gas, and Coal Supply, Distribution, and Use. The
model projects decreases in petroleum production/consumption of approximately 975 barrels
per day under Option 1A. In contrast, natural gas production/consumption is projected to
increase by 600,000 cubic feet per day under Option 1A. This is the result of fuel switching
in response to relative price changes. Finally, the model also projects less than a 1,000 tons
per day decrease in coal production/consumption in response to reduced output from the
electricity sector (a significant consumer of coal).
C.3 Small Entity Screening
The 3,580 units in the Inventory Database with full information were linked to 1,881
existing facilities. As shown in Table C-6, these are facilities owned by 970 parent
companies. The average number of facilities per company is approximately 2.2; however,
several large entities in the health services industry and government sectors own many
facilities with boilers and process heaters.
Based on SBA guidelines, 369 of the companies were identified as small businesses.2
The lumber and wood products industry contains the largest number of the small businesses
with 134, followed by furniture and fixtures with 55, electric services with 30, and paper and
allied products with 30. The remaining small businesses are distributed across 40 different
two-digit SIC code groupings (Table C-7).
2Small business guidelines typically define small businesses based on employment, and the threshold varies
from industry to industry. For example, in the paints and allied products industry, a business with fewer
than 500 employees is considered a small business; whereas in the industrial gases industry, a business with
fewer than 1,000 employees is considered small. However, for a few industries, usually services, sales are
used as the criterion. For example, in the veterinary hospital industry, companies with less than $5 million
in annual sales are defined as small businesses.
C-ll
-------
Table C-6. Facility-Level and Parent-Level Data by Industry
SIC
Code
01
02
07
10
12
13
14
17
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
NAICS
Code
111
112
115
212
212
211
212
235
311
312
313
315
321
337
322
511
325
324
326
316
327
331
332
333
335
336
334
339
482
Description
Agriculture — Crop s
Agriculture — Livestock
Agricultural Services
Metal Mining
Coal Mining
Oil and Gas Extraction
Mining/Quarrying — Nonmetallic
Minerals
Construction — Special Trade
Food and Kindred Products
Tobacco Products
Textile Mill Products
Apparel & Other Products from
Fabrics
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Printing, Publishing, and Related
Industries
Chemicals and Allied Products
Petroleum Refining and Related
Industries
Rubber and Misc. Plastics Products
Leather and Leather Products
Stone, Clay, Glass, and Concrete
Products
Primary Metal Industries
Fabricated Metal Products
Industrial Machinery and Computer
Equip.
Electronic and Electrical Equipment
Transportation Equipment
Scientific, Optical, and Photographic
Equipment
Misc. Manufacturing Industries
Railroad Transportation
Number
of
Units
6
—
—
11
2
18
10
2
163
22
250
4
462
310
503
8
433
162
56
22
42
85
44
46
45
158
33
14
4
Number
Number of
of Parent
Facilities Companies
6
—
—
5
1
4
5
1
72
11
134
4
337
209
272
6
163
50
37
12
25
33
28
25
29
61
16
10
1
6
—
—
2
—
1
4
1
38
6
73
3
175
100
100
3
91
31
24
8
15
22
18
20
19
26
9
9
1
Avg.
Number of
Facilities
Per Parent
Entity
1.0
—
—
2.5
—
4.0
1.3
1.0
1.9
1.8
1.8
1.3
1.9
2.1
2.7
2.0
1.8
1.6
1.5
1.5
1.7
1.5
1.6
1.3
1.5
2.3
1.8
1.1
1.0
(continued)
C-12
-------
Table C-6. Facility-Level and Parent-Level Data by Industry (continued)
SIC
Code
42
46
49
50
51
55
58
59
60
70
72
76
80
81
82
83
86
87
89
91
92
94
96
97
NA
State
NAICS
Code
484
486
221
421
422
441
722
445-454
522
721
812
811
621
541
611
624
813
541
711/514
921
922
923
926
928
Description
Motor Freight and Warehousing
Pipelines, Except Natural Gas
Electric, Gas, and Sanitary Services
Wholesale Trade — Durable Goods
Wholesale Trade — Nondurable Goods
Automotive Dealers and Gasoline
Service Stations
Eating and Drinking Places
Miscellaneous Retail
Depository Institutions
Hotels and Other Lodging Places
Personal Services
Misc. Repair Services
Health Services
Legal Services
Educational Services
Social Services
Membership Organizations
Engineering, Accounting, Research,
Management and Related Services
Services, N.E.C.
Executive, Legislative, and General
Administration
Justice, Public Order, and Safety
Administration of Human Resources
Administration of Economic Programs
National Security and International
Affairs
SIC Information Not Available
Parent is a state government
Total
Number
of
Units
7
6
372
3
2
1
—
1
—
1
—
2
40
—
114
3
—
6
2
2
33
1
4
41
24
—
3,580
Number
of
Facilities
3
5
185
2
1
1
—
1
—
1
—
1
19
—
50
2
—
5
1
2
10
1
3
13
18
—
1,881
Number
of
Parent
Companies
3
1
98
1
1
1
—
1
—
1
—
—
2
—
35
2
—
2
—
1
—
—
1
2
2
11
970
Avg.
Number of
Facilities
Per Parent
Entity
1.0
5.0
1.9
2.0
1.0
1.0
—
1.0
—
1.0
—
—
9.5
—
1.4
1.0
—
2.5
—
2.0
—
—
3.0
6.5
9.0
—
2.2
Source: Industrial Combustion Coordinated Rulemaking (ICCR). 1998. Data/Information Submitted to the
Coordinating Committee at the Final Meeting of the Industrial Combustion Coordinated Rulemaking Federal
Advisory Committee. EPA Docket Numbers A-94-63, II-K-4b2 through -4b5. Research Triangle Park, North
Carolina. September 16-17.
C-13
-------
Table C-7. Small Parent Companies by Industry
SIC
Code
01
02
07
10
12
13
14
17
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
42
46
49
NAICS
Code
111
112
115
212
212
211
212
235
311
312
313
315
321
337
322
511
325
324
326
316
327
331
332
333
335
336
334
339
482
484
486
221
Description
Agriculture — Crop s
Agriculture — Livestock
Agricultural Services
Metal Mining
Coal Mining
Oil and Gas Extraction
Mining/Quarrying — Nonmetallic Minerals
Construction — Special Trade Contractors
Food and Kindred Products
Tobacco Products
Textile Mill Products
Apparel and Other Products from Fabrics
Lumber and Wood Products
Furniture and Fixtures
Paper and Allied Products
Printing, Publishing, and Related Industries
Chemicals and Allied Products
Petroleum Refining and Related Industries
Rubber and Misc. Plastics Products
Leather and Leather Products
Stone, Clay, Glass, and Concrete Products
Primary Metal Industries
Fabricated Metal Products
Industrial Machinery and Computer Equip.
Electronic and Electrical Equipment
Transportation Equipment
Scientific, Optical, and Photographic Equip.
Miscellaneous Manufacturing Industries
Railroad Transportation
Motor Freight and Warehousing
Pipelines, Except Natural Gas
Electric, Gas, and Sanitary Services
Number of Parent
Companies
6
—
—
2
—
1
4
1
38
6
73
3
175
100
100
3
91
31
24
8
15
22
18
20
19
26
9
9
1
3
1
98
Number of Small
Parent Companies
1
—
—
2
—
1
—
1
15
—
27
2
134
55
30
2
19
9
4
4
3
3
5
5
—
5
1
1
—
1
—
30
(continued)
C-14
-------
Table C-7. Small Parent Companies by Industry (continued)
SIC
Code
50
51
55
58
59
60
70
72
76
80
81
82
83
86
87
89
91
92
94
96
97
NA
State
NAICS
Code
421
422
441
722
445-454
522
721
812
811
621
541
611
624
813
541
711/514
921
922
923
926
928
Description
Wholesale Trade — Durable Goods
Wholesale Trade — Nondurable Goods
Automotive Dealers and Gasoline Service
Stations
Eating and Drinking Places
Miscellaneous Retail
Depository Institutions
Hotels and Other Lodging Places
Personal Services
Misc. Repair Services
Health Services
Legal Services
Educational Services
Social Services
Membership Organizations
Engineering, Accounting, Research,
Management and Related Services
Services, N.E.C.
Executive, Legislative, and General
Administration
Justice, Public Order, and Safety
Administration of Human Resources
Administration of Economic Programs
National Security and International Affairs
SIC Information Not Available
Parent is a State Government
Total
Number of Parent
Companies
1
1
1
—
1
—
1
—
—
2
—
35
2
—
2
—
1
—
—
1
2
2
11
970
Number of Small
Parent Companies
—
—
1
—
1
—
—
—
—
1
—
3
1
—
—
—
—
—
—
—
—
2
—
369
Source: Industrial Combustion Coordinated Rulemaking (ICCR). 1998. Data/Information Submitted to the
Coordinating Committee at the Final Meeting of the Industrial Combustion Coordinated Rulemaking
Federal Advisory Committee. EPA Docket Numbers A-94-63, II-K-4b2 through -4b5. Research
Triangle Park, North Carolina. September 16-17.
C-15
-------
C.3.1 Small Entity Screening Results
As shown in Table C-8, the average cost per entity is $270,000. The median cost-to-
sales ratio is 0.8 percent, and ratios range from 0.01 to 39 percent. Forty-five of the 369
affected small businesses have CSRs at or above 3 percent.
Table C-8. Summary Statistics for SBREFA Screening Analysis
Option 1A
Total Number of Small Entities 369
Average Annual Compliance Cost per Small Entity ($103) $270
Entities with Sales/Revenue Data
Compliance costs are <1% of sales 176
Compliance costs are >1 to 3% of sales 148
Compliance costs are >3% of sales 45
Compliance Cost-to-Sales/Revenue Ratios
Average 1.65
Median 0.77
Maximum 38.83
Minimum 0.009
C. 3.2 Affected Government Entities: Supplemental Analysis
As shown in Table C-9, the average total annual compliance costs per entity are
$548,000 for Option 1A. The median cost-to-revenue ratio is 2.2 percent, and ratios range
from less than 0.5 percent to 16 percent. Five of the 13 affected small governments have
cost-to-revenue ratios at or above 3 percent.
C-16
-------
Table C-9. Supplemental Screening Analysis for Small Governmental Jurisdictions
Option 1A
Total Number of Small Entities 13
Average Total Annual Compliance Cost (TACC) per Small Entity ($) $548
Entities with Sales/Revenue Data
Compliance costs are <1% of revenue 2
Compliance costs are >1 to 3% of revenue 6
Compliance costs are >3% of revenue 5
Compliance Cost-to-Sales/Revenue Ratios
Average 4.18
Median 2.21
Maximum 16.30
Minimum 0.02
Source: American Public Power Association (APPA). 2002. Straight Answers to False Charges about
Public Power. Washington, DC: APPA. .
As obtained on November 1 3, 2003.
C-17
-------
APPENDIX D:
IMPACTS FROM APPLICATION OF RISK-BASED ALTERNATIVES
As an alternative to the requirement for each large solid fuel-fired boiler to
demonstrate compliance with the HC1 emission limit in the final rule, the source may
demonstrate compliance with a health-based facility-wide HC1 equivalent allowable
emission limit. The procedures for complying with this compliance alternative are in
Appendix A of the final rule. Also, in lieu of complying with the emission standard for what
are called total selected metals (TSM) in subpart DDDDD of part 63 based on the sum of
emissions for metals (such as arsenic, cadmium, chromium, mercury, manganese, nickel, and
lead facilities may demonstrate eligibility for complying with the TSM standard based on
excluding manganese emissions from the summation of TSM emissions for the affected
source unit.
The impacts discussed below reflect the effects of compliance with both risk-based
alternatives to the final rule.
Per technical direction received February 5, 2004, RTI performed an economic and
small entity analysis of a modified emission control scenario under the Industrial Boilers and
Process Heaters MACT. The key elements of the sensitivity scenario are as follows:
• Exempt controls for HCL emissions for coal-fired units—(model plants 2d, 3e,
6d, and 7e).
• Exempt controls for manganese emissions for wood-fired units (model plants 30a,
30b, 34a, 34b, and 34d).
• Monitoring, record-keeping, and reporting costs for units linked to these model
plants remain unchanged.
This appendix highlights the key results of these analyses.
• Social Cost Estimates—The social cost estimates fall from $863 billion to $746
billion, or approximately 14 percent (see Table D-l). The sensitivity analysis
does not show significant changes in the distribution of these costs across sectors
and stakeholders (see Table D-2).
D-l
-------
Table D-l. Social Cost Estimates ($1998 106)
Change in Social Welfare
Baseline engineering costs $746.49
Social costs with market adjustments $746.44
Difference between engineering and social cost estimates $.04
Market-Level Impacts—A comparison of the relative price and quantity changes
shows the sensitivity scenario moderates price and quantity effects, but the
differences are very small (see Table D-3).
Energy Effects—Energy impacts are small in both analyses. However, electricity
and natural gas production/consumption impacts in the sensitivity scenario are
approximately half the values projected under the MACT floor. Annual
electricity production declines by 246 million kWh compared to 415 kWh.
Natural gas production/consumption declines by 230,000 cubic feet per day
compared to 1.1 million cubic feet per day. Petroleum and coal sector effects do
not significantly change.
Small Entity Impacts—The average annual control costs fall from $199K to
$142K under the sensitivity scenario (see Table D-4). Similarly, the average
(median) cost-to-sales ratio (CSR) falls from 0.78 percent (0.50 percent) to 0.52
percent (0.16 percent) (see Table D-5). Twenty-two entities have CSRs greater
than 1 percent under the sensitivity scenario compared with 34 entities under the
MACT floor. Eight entities have CSRs that are 3 percent or higher compared to
10 entities under the MACT floor.
Affected Government Entities: Supplemental Analysis—The average annual
control costs fall from $223K to $217K under the sensitivity scenario(see Table
D-6). CSRs remain essentially unchanged (e.g., the average [median] CSR falls
from 1.67 percent [0.94 percent] to 1.66 percent [0.94 percent]). As a result, there
is no change in projections of the number of government entities affected at the 1
or 3 percent level.
D-2
-------
Table D-2. Distribution of Social Costs by Sector/Market: ($1998 106)
Sectors/Markets
Energy Markets
Petroleum
Natural gas
Electricity
Coal
Subtotal
NAICS Code
311
312
313
314
315
316
321
322
323
325
326
327
331
332
333
334
335
336
337
339
11
23
21
48
42; 44-45; 49;
51-56; 61-62;71-
72; 81
Grand Total
SIC Code
20 (pt)
20 (pt); 21
22 (pt)
22 (pt)
23
31
24
26
27
28
30
32
33
34
35
36 (pt)
36 (pt)
37
25
39
01-08
15-17
10; 14
40-47 (pt)
40-48 (pt);
50-99
Description
Food
Beverage and Tobacco Products
Textile Mills
Textile Product Mills
Apparel
Leather and Allied Products
Wood Products
Paper
Printing and Related Support
Chemicals
Plastics and Rubber Products
Nonmetallic Mineral Products
Primary Metals
Fabricated Metal Products
Machinery
Computer and Electronic Products
Electrical Equipment, Appliances,
and Components
Transportation Equipment
Furniture and Related Products
Miscellaneous
Agricultural Sector
Construction Sector
Other Mining Sector
Transportation
Commercial
Residential
Producer
Surplus
-$2.4
$0.9
-$32.3
-$2.6
-$36.3
-$21.4
-$2.1
-$20.8
-$0.1
-$0.3
-$0.3
-$25.2
-$63.3
-$0.2
-$35.5
-$2.1
-$2.6
-$21.6
-$6.3
-$6.3
-$3.2
-$2.4
-$19.9
-$3.3
-$0.8
-$0.5
-$0.7
-$9.9
-$3.5
-$63.7
NA
-$352.3
Change in:
Consumer
Surplus
-$8.6
-$3.7
-$47.6
-$0.1
-$0.8
-$0.4
-$6.7
-$57.5
-$0.4
-$71.0
-$5.0
-$3.1
-$4.9
-$1.7
-$4.2
-$1.3
-$1.6
-$26.6
-$14.8
-$0.6
-$1.2
-$0.9
-$6.9
-$3.2
-$84.9
-$36.3
-$394.2
Social
Welfare
-$30.0
-$5.8
-$68.4
-$0.2
-$1.2
-$0.7
-$31.9
-$120.8
-$0.5
-$106.5
-$7.0
-$5.7
-$26.6
-$8.0
-$10.5
-$4.5
-$4.1
-$46.5
-$18.1
-$1.4
-$1.7
-$1.6
-$16.8
-$6.7
-$148.6
-$36.3
-$746.4
NA = Not applicable.
pt = Part.
D-3
-------
Table D-3. Market-Level Impacts: Final Rule
Sectors/Markets
Energy Markets
Petroleum
Natural gas
Electricity
Coal
NAICS Code
311
312
313
314
315
316
321
322
323
325
326
327
331
332
333
334
335
336
337
339
11
23
21
48
42; 44-45; 49; 51-56;
61-62; 71-72; 81
SIC Code
20 (pt)
20 (pt); 21
22 (pt)
22 (pt)
23
31
24
26
27
28
30
32
33
34
35
36 (pt)
36 (pt)
37
25
39
01-08
15-17
10; 14
40-47 (pt)
40-48 (pt); 50-99
Description
Food
Beverage and Tobacco Products
Textile Mills
Textile Product Mills
Apparel
Leather and Allied Products
Wood Products
Paper
Printing and Related Support
Chemicals
Plastics and Rubber Products
Nonmetallic Mineral Products
Primary Metals
Fabricated Metal Products
Machinery
Computer and Electronic Products
Electrical Equipment, Appliances, and
Components
Transportation Equipment
Furniture and Related Products
Miscellaneous
Agricultural Sector
Construction Sector
Other Mining Sector
Transportation
Commercial
Percent
Price
0.001%
0.002%
0.044%
-0.008%
0.004%
0.003%
0.023%
0.000%
0.000%
0.002%
0.026%
0.025%
0.000%
0.007%
0.001%
0.003%
0.009%
0.002%
0.002%
0.001%
0.002%
0.003%
0.005%
0.001%
0.000%
0.000%
0.012%
0.001%
0.000%
Change
Quantity
0.000%
0.000%
-0.010%
-0.010%
-0.001%
-0.004%
-0.020%
0.000%
-0.001%
-0.003%
-0.005%
-0.027%
0.000%
-0.011%
-0.002%
-0.002%
-0.007%
-0.000%
-0.001%
0.000%
-0.001%
-0.003%
-0.015%
0.000%
0.000%
0.000%
-0.004%
-0.001%
0.000%
pt = Part.
D-4
-------
Table D-4. Summary of Small Entity Impacts
Final Rule
Number of small entities
Total number of entities
Average annual control cost per small entity (103)
Average control cost/sales ratio
Number of small entities with cost-to-sales ratios > 1 percent
Number of small entities with cost-to-sales ratios >3 percent
185
576
$142
0.52%
22
Table D-5. Summary Statistics for SBREFA Screening Analysis
Value
Total Number of Small Entities
Average Annual Compliance Cost per Small Entity ($103)
185
$142
Entities with Sales/Revenue Data
Compliance costs are <1% of sales
Compliance costs are >1 to 3% of sales
Compliance costs are >3% of sales
155
22
Compliance Cost-to-Sales/Revenue Ratios
Average
Median
Maximum
Minimum
0.52
0.16
7.83
0.01
D-5
-------
Table D-6. Supplemental Screening Analysis for Small Governmental Jurisdictions
Value
Total Number of Small Entities 13
Average Total Annual Compliance Cost (TACC) per Small Entity ($103) $217
Entities with Sales/Revenue Data
Compliance costs are <1% of revenue 7
Compliance costs are >1 to 3% of revenue 3
Compliance costs are >3% of revenue 3
Compliance Cost-to-Sales/Revenue Ratios
Average 1.66
Median 0.94
Maximum 7.83
Minimum 0.01
Source: American Public Power Association (APPA). 2002. Straight Answers to False Charges about
Public Power. Washington, DC: APPA. .
As obtained on November 1 3, 2003.
D-6
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO.
EPA-452/R-04-001
4. TITLE AND SUBTITLE
Economic Impact Analysis of the
NESHAP
o
Industrial Boilers and Process Heaters
7. AUTHOR(S)
Mike Gallaher, Alan O'Connor, Brooks Depro
9. PERFORMING ORGANIZATION NAME AND ADDRESS
RTI International
Center for Regulatory Economics and Policy Research, Hobbs Bldg.
Research Triangle Park, NC 27709
12. SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
February 2004
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
RTI Project Number 7647-005-423
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D-99-024
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report evaluates the economic impacts of the Industrial Boilers and Process Heaters
NESHAP. The social costs of the rule are estimated by incorporating the expected costs of compliance in a
partial equilibrium model and projecting the market impacts. The report also provides the screening analysis
for small business impacts.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
economic impacts
small business impacts
social costs
18. DISTRIBUTION STATEMENT
Release Unlimited
b. IDENTIFIERS/OPEN ENDED TERMS
Air Pollution Control
Economic Impact Analysis
Regulatory Flexibility Analysis
19. SECURITY CLASS (Report)
Unclassified
20. SECURITY CLASS (Page)
Unclassified
c. COSATI Field/Group
21. NO. OF PAGES
202
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
------- |