EPA-600/5-74-009
February 1974
Socioeconomic Environmental Studies Series
Cost-Effectiveness of A Uniform
National Sulfur Emissions Tax
I
5
LU
CD
Office of Research and Development
Washington, D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
<*. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the SOCIOECONOMIC
ENVIRONMENTAL STUDIES series. This series
describes research on the socioeconomic impact of
environmental problems. This covers recycling and
other recovery operations with emphasis on
monetary incentives. The non-scientific realms of
legal systems, cultural values, and business
systems are also involved. Because of their
interdisciplinary scope, system evaluations and
environmental management reports are included in
this series.
-------
EPA-600/5-74-009
February 1974
COST-EFFECTIVENESS OF A UNIFORM NATIONAL
SULFUR EMISSIONS TAX
By
Tayler H. Bingham
Philip C. Cooley
Mark E. Fogel
Donald R. Johnston
David A. LeSourd
Allen K. Miedema
Richard E. Paddock
Mayrant Simons, Jr.
Macmillan M. Wisler
Contract No. 68-01-0426
Program Element 1HA093
Project Officer
Marshall Rose
Implementation Research Division
Environmental Protection Agency
Washington, D.C. 20460
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
For sale by the Superintendent of Documents, tr.S. Government Printing Office, Washington, D.C. 20402 - Price $2.25
-------
ACKNOWLEDGEMENTS
This project was conducted by the Research Triangle Institute,
Research Triangle Park, North Carolina, pursuant to Contract No.
68-01-0426 with the Environmental Protection Agency. The statements,
findings, conclusions, and recommendations presented in this report
do not necessarily reflect the views of the Environmental Protection
Agency.
This project was under the direct supervision of Tayler H. Bingham
with several individuals contributing to specific aspects of the research
described in this report. Principal among these and their contributions
are:
Philip C. Cooley Model design and programming
Mark E. Fogel Control costs and new
technology options
Donald R. Johnston Analysis of impact of tax
on air quality
David A. LeSourd Area source analysis
Allen K. Miedema New technology analysis
Richard E. Paddock Steam-electric power plant
and sulfuric acid plant
control costs
Mayrant Simons, Jr. Petroleum refinery control
costs
Macmillan M. Wisler Smelter control costs
John R. O'Connor, Frank L. Bunyard, Warren Freas, and Mark Seidel of the
Environmental Protection Agency, Gordon T. C. Taylor of the University of
Maryland and Gerald A. Carlson of North Carolina State University pro-
vided data and/or critical reviews of earlier material and drafts of the
study. Larry Ruff and Marshall Rose, Environmental Protection Agency,
served as project officers. Their interest and guidance are very much
appreciated.
-------
TABLE OF CONTENTS
Chapter Page
1: INTRODUCTION AND SUMMARY 1
1.1 Background 1
1.2 Effects of Sulfur Oxides 2
1.3 The Concept of a Tax on Sulfur Emissions 3
1.4 Approach 3
1.5 Assumptions, Limitations, and Capabilities 6
1.6 Summary of Findings 8
2: THE APPLICATION OF EMISSIONS TAXES FOR POLLUTION CONTROL 11
2.1 Introduction 11
2.2 Air Quality and Market Failure 11
2.3 Emissions Taxes 12
3: ANALYSIS OF THE EFFECTIVENESS AND COSTS OF A TAX ON
SULFUR EMISSIONS 15
3.1 Introduction 15
3.2 The Emissions Response Model 15
3.2.1 Fuel Combustion Sources 16
3.2.1.1 Fuel Supplies and Prices 16
3.2.1.1.1 Fuel Supplies 16
3.2.1.1.2 Fuel Price Adjustment 18
3.2.1.2 Steam-Electric Power Plants 18
3.2.1.3 Area Sources 19
3.2.2 Industrial Process Sources 21
3.2.2.1 Petroleum Refineries 22
3.2.2.2 Sulfuric Acid Plants 23
3.2.2.3 Primary Nonferrous Smelters 23
-------
Chapter Page
3.3 Cost of Control Functions 24
3.4 Effectiveness 25
3.5 Costs 29
3.6 Tax Revenues 30
3.7 Cost-Benefit Analysis 30
3.8 Impacts on Consumer Prices 31
4: SOURCE-BY-SOURCE ANALYSIS OF THE EFFECTIVENESS AND COSTS
OF A TAX ON SULFUR EMISSIONS 37
4.1 Introduction 37
4.2 Steam-Electric Power Plants 37
4.2.1 Background 38
4.2.2 Industry Growth 38
4.2.3 Effectiveness 38
4.2.4 Costs 45
4.2.5 Sensitivity Analysis 47
4.3 Industrial, Commercial, and Residential Space
Heating 50
4.3.1 Background 50
4.3.2 Industry Growth 52
4.3.3 Effectiveness 52
4.3.4 Costs 56
4.4 Petroleum Refineries 56
4.4.1 Background 57
4.4.2 Industry Growth 57
4.4.3 Effectiveness 58
v1
-------
Chapter Page
4.4.4 Costs 61
4.4.5 Sensitivity Analysis 63
4.5 Sulfuric Acid Producers 65
4.5.1 Background 65
4.5.2 Industry Growth 66
4.5.3 Effectiveness 67
4.5.4 Costs 69
4.5.5 Sensitivity Analysis 69
4.6 Primary Nonferrous Smelters 72
4.6.1 Background 73
4.6.2 Industry Growth 73
4.6.3 Effectiveness 73
4.6.4 Costs 79
4.6.5 Sensitivity Analysis 81
5: IMPACT OF SULFUR TAX ON AIR QUALITY IN TWO HYPOTHETICAL
AIR QUALITY CONTROL REGIONS 83
5.1 Introductions 83
5.2 Relationships between SC>2 Concentration and
Sulfur Emissions 83
5.3 Hypothetical AQCR's 84
5.3.1 Hypothetical AQCR A 84
5.3.2 Hypothetical AQCR B 85
5.4 Effect of Sulfur Tax on Emissions and Ambient
S0£ Concentrations 85
5.4.1 Sulfur Emissions 85
5.4.2 Ambient S0£ Concentrations 85
BIBLIOGRAPHY 88
vn
-------
Appendix Page
A: SULFUR EMISSIONS FACTORS AND CONTROL COSTS FOR
FUEL COMBUSTION SOURCES 95
A.I Emission Factors 95
A.2 Sulfur Emissions Control Alternatives and Cost 99
A.2.1 Sulfur Oxide Removal from Stack Gases 99
A.2;l.lr Dry?Limestone Absorption loi
A.2.1.2 Wet Lime Scrubbing 101
A.2.1.3 Magnesia Base Scrubbing 103
A.2.2 Fuel Switching 103
A.2.2.1 Coal 105
A.2.2.2 Oil 110
A.2.2.2.1 Distillate 112
A.2.2.2.2 Residuals 114
A.2.2.3 Gas 117
A.2.2.4 Transportation Costs 117
A.2.3 Emissions Reductions and Costs 118
B: SULFUR EMISSION FACTORS AND CONTROL COSTS FOR
PETROLEUM REFINERIES 121
B.I Emission Factors 121
B.I.I Catalyst Regenerators 121
B.I.2 Claus Plants 123
B.I.3 Fuel Combustion 123
B.2 Sulfur Emissions-Control Alternatives and Costs 124
B.2.1 Catalyst Regenerators 124
B.2.2 Claus Plants 126
B.2.3 Fuel Combustion 127
B.2.4 Emissions Reductions and Costs 127
* • •
vi 11
-------
Ci SULFUR EMISSION FACTORS AND CONTROL COSTS FOR
SULFURIC ACID PRODUCERS 131
C.I Emission Factors 131
C.2 Sulfur Emissions Control Alternatives and Costs 131
C.2.1 Gaseous Emissions Control 132
C.2.2 Mist Control 133
C.2.3 Emissions Reductions and Costs 134
D: SULFUR EMISSION FACTORS AND CONTROL COSTS FOR
PRIMARY NONFERROUS SMELTERS 137
DJ Emission Factors 137
D.I.I Copper Smelters 137
D.I.2 Zinc Smelters 138
D.I.3 Lead Smelters 138
D.2 Sulfur Emissions Control Alternatives and Costs 133
D.2.1 Copper Smelters 139
D.2.2 Zinc Smelters 142
D.2.3 Lead Smelters 145
D.2.4 Emissions Reductions and Costs 145
E: THE PROJECTED MARKET FOR RECOVERED SULFUR 151
E.I Introduction 151
E.2 Major Sources of Sulfur 151
E.3 Consumption of Sulfur 151
E.4 Prices of Sulfur 153
F: THE EFFECT OF THE CORPORATE TAX STRUCTURE ON THE
PROJECTED EMISSION LEVELS 157
1x
-------
G: SOME ASPECTS OF SULFUR EMISSION TAXES IN THE PRESENCE
OF ADVANCING CONTROL TECHNOLOGY FOR POWER PLANTS 161
G.I Introduction . 161
G.2 Some Expected SO? Control Alternatives for
Power Plants
u oup lAJinrui m Ler not i ves lur
, 1980-1985 162
G.2.1 SQ9 Control Alternatives for Existing
Power Plants 162
G.2.1.1 Citrate S02 Removal Process 162
G.2.1.2 Double Alkali S02 Removal Process 162
G.2.1.3 Gasified Coal 163
G.2.2 S02 Control Options for New Power Plants 153
G.2.2.1 Fluidized Bed Combustion Process 163
G.2.2.2 Combined Cycle Power Systems 164
G.2.2.3 Nuclear Power Plants 164
G.2.3 Summary 165
G.3 Theoretical Aspects of Producer Responses to the
Emissions Tax Over Time 165
G.3.1 A Simple Model of Cost Minimization
Without Emissions Control Policies 165
G.3.2 Cost Minimization Over Time in the
Presence of Emissions Control Policies 167
G.4 A Preliminary Empirical Investigation of Sulfur
Emission-Tax-Induced Delays in the Removal of
Sulfur Oxides from Stack Gases 171
H. STATE-BY-STATE PROJECTIONS OF THE EFFECTIVENESS AND
COSTS OF A TAX ON SULFUR EMISSIONS 179
-------
LIST OF FIGURES
Figure Page
1 Emission source behavior in response to a tax 14
2 Socially optimal level of emissions 14
3 Total cost* of reductions in sulfur emissions from
all major sources combined—1978 26
4 Marginal cost* of reductions in sulfur emissions from
all major sources combined--1978 26
5 Effectiveness of a tax on the sulfur emissions from
all major sources combined—1978 27
6 Total costs induced by a tax on the sulfur emissions
from all major sources combined—1978 29
7 Tax revenues from a tax on sulfur emissions from all
major sources combined--1978 30
8 The incidence of a sulfur emissions tax 32
9 Projected impact on consumer prices of a tax on the
sulfur emissions from all major sources—1978 35
10 Steam-electric power plant trends 39
11 Effectiveness of a tax on sulfur emissions: steam-
electric power plants—1978 43
12 Total costs induced by a tax on sulfur emissions:
steam-electric power plants—1978 45
13 Average total incremental costs per kilowatthour induced
by a tax on sulfur emissions: steam-electric power
plants—1978 47
14 Effectiveness of a tax on sulfur emissions: area
sources—1978 55
15 Total cost induced by a tax on sulfur emissions: area
sources—1978 56
16 Petroleum refining trends 58
XI
-------
LIST OF FIGURES
Figure Page
17 Effectiveness of a tax on sulfur emissions:
petroleum refining—1978 61
18 Total costs induced by a tax on sulfur emissions:
petroleum refining—1978 62
19 Average total incremental costs per barrel of refined
oil induced by a tax on sulfur emissions: petroleum
refining—1978 62
20 Sensitivity of the effectiveness and total costs of a
tax on sulfur emissions to the value of recovered sulfur:
petroleum refining—1978 63
21 Sulfuric acid production trends 66
22 Effectiveness of a tax on sulfur emissions: sulfuric
acid producers—1978 69
23 Total costs induced by a tax on sulfur emissions:
sulfuric acid producers—1978 70
24 Average total incremental costs per tone of acid production
induced by a tax on sulfur emissions: sulfuric acid
producers—1978 - 70
25 Sensitivity of the effectiveness and total costs of a
tax on sulfur emissions to the value of recovered
sulfur: sulfuric acid producers—1978 71
26 Primary nonferrous smelting trends 74
27 Effectiveness of a tax on sulfur emissions—1978 79
28 Total costs induced by a tax on sulfur emissions—1978 80
29 Average total incremental costs per ton of product
induced by a tax on sulfur emissions—1978 80
30 Sensitivity of the effectiveness and total costs of
a tax on sulfur emissions to the value of recovered
sulfur—1978 . 81
31 Effectiveness of a tax on sulfur emissions for two
hypothetical Air Quality Control Regions 85
32 Relationship between tax rates and air quality for
two hypothetical Air Quality Control Regions 87
xii
-------
LIST OF FIGURES
Figure Page
A.I Coal-producing districts 108
A.2 Bituminous coal price trends 109
A.3 Sources of supply of petroleum 111
A.4 Distillate consumption 111
A.5 Residual fuel oil consumption 115
A.6 Oil-producing regions 115
A.7 Total cost of reductions in sulfur emissions: steam-
electric utilities 119
A.8 Marginal cost of reductions in sulfur emissions:
steam-electric utilities 119
A.9 Total cost of reductions in sulfur emissions: area
sources 120
V-
A.10 Marginal cost of reductions in sulfur emissions:
area sources 120
B.I Total cost of reductions in sulfur emissions:
petroleum refineries—1978 129
B.2 Marginal cost of reductions in sulfur emissions:
petroleum refineries-1978 129
C.I Total cost of reductions in sulfur emissions:
sulfuric acid producers 135
C.2 Marginal cost of reductions in sulfur emissions:
sulfuric acid producers-1978 136
D.I Total cost of reductions in sulfur emissions:
primary nonferrous smelters-1978 148
D.2 Marginal cost of reductions in sulfur emissions:
primary nonferrous smelters 149
E.I Sulfur consumption trends 152
xiii
-------
LIST OF FIGURES-
Figure Page
E.2 Sulfur price index 154
G.I Loci of anticipated costs of new SO control' processes
(available in T* years beyond 1978) below which a cost
minimizing 1500 MW model power plant would defer S0x
stack gas cleaning 175
G.2 Loci of anticipated costs of new SO control processes
(available in T* years beyond 1978) below which a cost
minimizing 800 MW model power plant would defer SO
stack gas cleaning 176
G.3 Loci of anticipated costs of new SO control processes
(available in T* years beyond 1978) below which a cost
minimizing 350 MW model power plant would defer S0x
stack gas cleaning. 177
H.I Marginal cost of reductions in sulfur emissions from
all major sources combined in selected States 180
H.2 Effectiveness of a tax on the sulfur emissions from
all major sources combined in selected States 180
xiv
-------
LIST OF TABLES
Table Page
1 Major Sources of Sulfur Emissions 4
2 Summary of the Projected Effectiveness and Costs of a
National Tax on the Major Sources of Sulfur Emissions —
1978 8
3 Comparison of Costs of Clean Air and Emissions Tax
Results—1978 9
4 Projected Sulfur Emissions from Major Sources--!978 27
5 Projected Response of all Major Sources Combined
to a National Tax on Sulfur Emissions—1978 28
6 Projected Initial Price Increases Resulting from a
Tax on Sulfur Emissions 34
7 Projected Sulfur Emissions from Steam-Electric
Utilities-1978 41
8 Projected Response of all Steam-Electric Power Plants
National Tax on Sulfur Emissions—1978 42
9 Effects of the Sulfur Emissions Tax on the
Distribution of Flue Gas Desulfurization Choices for
Steam-Electric Power Producing Units 44
10 Distribution of Steam-Electric Utilities' Demand for
Coal and Residual Oil by Sulfur Content--1978 44
11 Percentage Distribution of Control Costs 46
12 Sensitivity of Effectiveness and Total Cost of Tax
on Sulfur Emissions to Residual Fuel Oil Prices:
Steam-Electric Power Plants—1978 48
13 Distribution of Steam-Electric Utilities' Demand for
Coal and Residual Oil by Sulfur Content with Increased
Prices for Residual Oil-1978 49
14 Effects of Residual Oil Price Increases on the Percentage
Distributions of Flue Gas Desulfurization Choices for
Steam-Electric Power Producing Units 51
15 Area Source Fuel Consumption--!970 52
16 Projected Sulfur Emissions from Area Sources—1978 53
xv
-------
LIST OF TABLES
Table Page
17 Projected Response of all Area Sources to a National
Tax on Sulfur Emissions—I978 54
18 Distribution of the Combined Demand of Steam-Electric
Utilities and Area Sources for Coal and Residual Oil —
1978 55
19 Projected Sulfur Emissions from Petroleum Refineries—
1978 59
20 Projected Response of all Petroleum Refineries to a
National Tax on Sulfur Emissions—1978 60
21 Sensitivity of the effectiveness and Total Cost of a
Tax on Sulfur Emissions to the Control Cost Estimates:
Petroleum Refineries--!978 64
22 Projected Emissions from Sulfuric Acid Production—1978 67
23 Projected Response of all Sulfuric Acid Plants to a
National Tax on Sulfur Emissions--!978 68
24 Sensitivity of Effectiveness and Total Costs of Tax
on Sulfur Emissions to the Contro.l Cost Estimates:
Sulfuric Acid Producers—1978 72
25 Projected Sulfur Emissions from Primary Nonferrous
Smelters—1978 75
26 Projected Response of all Primary Nonferrous Smelters
to a National Tax on Sulfur Emissions—1978 76
27 Sensitivity of Effectiveness and Total Cost of Tax on
Sulfur Emissions to Control Cost Estimates: Primary
Nonferrous Smelters—1978 82
28 Sulfur Emissions by Source for two Hypothetical Air
Quality Control Regions 84
29 Annual Sulfur Emissions and Percent Reduction with
Various Tax Rates 86
xvi
-------
LIST OF TABLES
Table Page
A.I Area Source Fuel Consumption by State, 1970 96
A. 2 Steam-Electric Plant Sulfur Emissions Stack Gas
Control Alternatives 99
A. 3 Sulfur Emissions Control Cost Equations for Steam-
Electric Utilities Using Flue Gas Desulfurization 100
A. 4 Geographic Distribution of Known Recoverable
Reserves of Coal 106
A. 5 Production of Commercial Bituminous and Sub-
bituminous Coal— 1970 107
A. 6 Projected Maximum Production of Commercial Bituminous
Coal— 1978 107
A. 7 Coal Prices by District 112
A. 8 Sources of Residual Fuel Oil 113
A. 9 Crude and Residual Oil Imported into the United
States from the Middle East 113
A. 10 Projected Maximum U.S. Production of Residual Oils—
1978 116
A. 11 Oil and Gas Average Btu Contents 116
A. 12 Oil and Gas Average Btu Contents 117
A. 13 Coal Average Btu Contents by Origin
B.I Estimated U.S. Petroleum Refinery Sulfur Balance—
1970 122
B.2 Sulfur Emissions Factors for Petroleum Refineries 123
B.3 Sulfur Emission Control Alternatives for Petroleum
Refineries 125
B.4 Sulfur Emission Control Costs for Petroleum Refineries 126
B.5 Size Distribution of Petroleum Refineries 128
C.I Sulfur Emission Factors for Sulfuric Acid Plants 132
C.2 Sulfur Emission Control Alternatives for Sulfuric
Acid Plants 133
xyn
-------
LIST OF TABLES
Table Page
C.3 Sulfur Emission Control Costs for Sulfuric Acid 135
C.4 Size Distribution of Sulfuric Acid Plants 136
D.I Sulfur Emission Factors for Copper Smelters 137
D.2 Sulfur Emission Factors for Zinc Smelters 138
D.3 Sulfur Emission Factors for Lead Smelters 139
D.4 Sulfur Dioxide Removal Efficiencies of Sulfuric Acid
Plants as a Function of Input Concentrations for Single-
and Double-Absorption Systems 141
D.5 Copper Smelter Plant Types 141
D.6 Sulfur Emission Control Alternatives for Copper
Smelters 142
D.7 Sulfur Emission Control Costs for Copper Smelters 143
D.8 Sulfur Emission Control Alternatives for Zinc Smelters 144
D.9 Sulfur Emission Control Costs for Zinc Smelters 145
D.10 Sulfur Emission Control Alternatives for Lead Smelters 146
D.ll Sulfur Emission Control Costs for Lead Smelters 146
D.12 Size Distribution of Copper Smelters 147
D.13 Size Distribution of Zinc Smelters H8
D.I4 Size Distribution of Lead Smelters 148
E.I Projected Sulfur Emissions—1978 153
E.2 Sulfur Production Costs by Source I56
G.I Projected Costs and Availability of Sulfur Emission
Control Alternatives for Power Production, 1980-1985
6.2 Model Plant Parmeters and Projected SO Removal Costs
n
H.I Projected State-by-State Responses to a National Tax
on Sulfur Emissions
xv1 i i
-------
A PROJECTION OF THE EFFECTIVENESS AND COSTS
OF A NATIONAL TAX ON SULFUR EMISSIONS
Chapter 1: INTRODUCTION AND SUMMARY
1.1 Background
Writing in 1920, the British economist A. C. Pigou observed that
London received only 12 percent of the available sunlight due to the
smoke in the atmosphere from factory chimneys and that the smoke inflicted
"a heavy uncharged loss on the community, in injury to buildings and
vegetables, expenses for washing clothes and cleaning rooms, expenses
for the provision of extra artificial light, and in many other ways."*
Pigou identified the reason for the smoke as the difference between the
private costs and the social costs of production. Industrialists, in
producing consumer goods and services at minimum private costs also
imposed an additional social cost on third parties since the value of
clean air was not included as part of the costs of production. The
assimilative capacity of the atmosphere was free to emitters. Consequently,
industrialists had no incentive to install "smoke-preventing appliances,"
for the costs of such "appliances" would simply raise the costs of
production, thus making the affected products less price-competitive.
Pigou advocated intervention by government to remove the divergence
between social and private costs, specifically citing emissions taxes as
a possible mechanism for removing the divergence. Today, more than half
a century later, economists are reiterating the same basic recommendation.
In particular, a tax on sulfur emissions has been proposed both by members
of government and private citizens on the basis of the effectiveness of
such a tax in reducing sulfur emissions to desirable levels at the smallest
total cost to society. This study provides an initial examination of the
effectiveness and costs of a uniform national tax on the major emitters of
sulfur (or more exactly, sulfur compounds).
Since current legislative and political considerations, coupled with
the still advancing state-of-the-art in sulfur oxide flue gas control techniques,
make the implementation of such a tax unlikely before 1978, this study
*A. C. Pigou, The Economics of Welfare. London: Macmillan and Co.,
Ltd., 1920, pp.
-------
is directed toward the goal of evaluating the potential costs and
implicit reductions in emissions that would occur in the presence
of various tax rates on sulfur emissions during that year. Though
most of the results address the national impact of such a policy tool
on each of five major sulfur emission source categories, some attention
t ••
is /flso given to regional effects and to the intrafuel price effects
of such a tax.
1.2 Effects of Sulfur Oxides
Sulfur is present in polluted atmospheres as a component of both
particulate matter and gases. In particulate matter it may occur as a
sulfate salt or as highly corrosive sulfuric acid. As a gaseous component,
sulfur is present in hydrogen sulfide, mercaptans, and sulfur oxides.
The principal physiological effect of sulfur dioxide (S02) and
sulfuric acid (H2SO*) is bronchoconstriction, leading to an increase
in airway resistance. Epidemiological studies of acute air pollution
episodes have shown a significant association between excess mortality
and morbidity and elevated S02 concentrations with associated particulate
matter. People with preexisting diseases of the heart and lungs are
particularly vulnerable to the effects of SCL.
Nonhealth effects of sulfur compounds include reduction of visibility
by suspended sulfate and sulfuric acid particles, accelerated corrosion
of metals at relative humidities greater than 70 percent, and deterioration
of limestone, marble, roofing slate, and mortar. Textile fibers are
damaged, fabrics fade, leather loses its strength, and paper is embrittled
*
in the presence of S0«. Sulfur dioxide and HpSO. at sufficient concen-
tration for an appropriate length of time also cause injury to ornamental
and economic crops.
Based on an examination of available data on the relationship between
concentration and the occurrence of adverse effects, the Environmental
Protection Agency (EPA) has set National Ambient Air Quality Standards
for S02. By maintaining sulfur oxides concentrations at or below those
specified in the standards, it is hoped'that adverse effects will be
avoided. ,
To achieve the National Ambient Air Quality Standards, several States
t
have developed implementation plans that rely principally on regulation
of sources, specifying such things as the sulfur content of fuels and
allowable S02 emissions. This research evaluates the effectiveness of a
-------
national tax on sulfur emissions either as .an alternative or as a supple-
mental strategy to achieve the National Ambient Air Quality Standards for
sulfur oxides.*
1.3 The Concept of a Tax on Sulfur Emissions
Emissions taxes are government-imposed prices on the discharges of
pollutants to the atmosphere. Their purpose may not be to raise revenue ,v
although they would have that effect; rather their purpose should be to
encourage the equalization of both the marginal costs and the marginal
benefits of using the assimilative capacity of the atmosphere or to induce
the attainment of desired air quality levels at minimum cost. The costs
of emission reductions in the presence of a tax would be internalized to
the firm and typically incorporated to some degree in product prices. This
would force producers and consumers of that product to pay directly the
costs of residuals treatment and disposal. This situation contrasts with
that of present air pollution externalities, in which these costs are
passed along in the form of social costs to all pollution receptor groups.
In general, polluting sources can be expected to control emissions
to the point where the incremental cost of removing the last unit of
effluent from the process off-gases equals the tax rate. Taxes would
then be paid on any uncontrolled discharges of pollutants to the atmos-
phere. The existence of these tax payments, which are costs to the
firm, provide a persistent incentive to seek new, more cost-effective
ways to control waste discharges.
1.4 Approach
This study of a tax on sulfur emissions, applied on a national basis,
uses a comparative statics approach to project emissions in the absence
of air pollution regulations (except the New Source Performance Standards)
and emissions and costs under several alternative tax rates for 1978.
The study is confined to five major sources of sulfur emissions
which account for over 90 percent of all estimated sulfur emissions
(table 1). These sources range from steam-electric plants which contribute
*Note that it is quite reasonable for either Federal or regional
pollution control authorities to consider the addition of a sulfur tax to
present regulations. Such an approach would have two effects: first, it
would provide an additional incentive for plants to meet already existing
regulations; second, it would retain at least some of the cost-effectiveness
properties of a tax while assuming a minimum level of control (under the
regulation) applicable to all sources.
-------
Table 1. Major sources of sulfur emissions
Percent of
Emissions source total
Fuel combustion
Steam-electric 50.6
Area sources* 22.8
Industrial process
Primary nonferrous smelters 11.7
Petroleum refineries 6.3
Sulfuric acid 1.8
Total 93.2
*Space heating and industrial boilers.
Source: Nationwide Inventory of Air Pollutant
Emissions 1968: Raleigh, N. C.: NAPCA, August 1970.
over 50 percent of total sulfur emissions, to sulfuric acid plants which
account for less than 2 percent of the total sulfur emissions.
An inventory of these sources in 1970 was developed which contains
data on plant capacities and the process configurations necessary to
estimate emissions and control costs. This inventory was based on previous
inventories used by RTI and supplemented with current information obtained
from EPA and from trade sources to make it applicable to 1970. In spite
of these efforts, some errors and omissions in the inventory are possible.
It is doubtful, however, that they would be significant enough to affect
the projected effectiveness or costs of the tax on sulfur emissions on
a national level. Because the tax was to be analyzed for 1978, projections
of industry growth were employed. New plants were added to the inventory
on the basis of available projections of industry growth and observed
trends in plant size and process types.'
Sulfur emission control alternatives were identified and their costs
were estimated from data presented in previous studies for EPA, based on
private communications with EPA industry specialists, or developed by RTI.
The control or abatement costs were aggregated from estimates specific
-------
to source location, plant capacity, and process configuration. All control
costs are on an annualized basis.
The number of control alternatives costed for each source depended
on available data. In some cases, only two control alternatives appeared
feasible. In the case of steam-electric utilities, about 1,000 combi-
nations of fuels, fuel origin, sulfur content, and flue gas desulfurization
alternatives were costed for each utility.
A computer model was developed to determine each plant's behavior
under a tax, compute emissions and costs, sum the results over the industry,
and print the results in tabular form. It is assumed that emissions
sources will minimize the sum of the costs of emissions reductions and
tax outlays by selecting the level of emissions reductions where the net
marginal costs of these reductions (MCER), after allowing for the sale
(if any) of byproducts, equals the tax rate (TX). That is, MC™ = TX.
At this level of emissions reductions, the total pollution-related costs
to the source (i.e., annualized control costs plus tax payments) are
minimized. All sulfur values presented in this report are in terms of
sulfur, not sulfur dioxide (S02), which is 50 percent sulfur.
This study has followed the convention used in the Cost of Clean Air
and many other EPA-sponsored studies in using non-tax-adjusted cost estimates,
It is recognized, however, that (because tax payments on sulfur emissions
would be tax deductible expenses) the effective tax rate is overstated
by the marginal percentage rate of corporate income taxes faced by the
firm. This argument, of course, assumes that the firm has enough profits
for the tax payments to be a usable tax deduction. It is also the case,
however, that all of the abatement costs are overestimated by at least
the same factor since all variable costs associated with pollution control
are also fully tax deductible and since the capital costs of pollution
control devices are subject to special accelerated depreciation schemes.
In effect, this study assumes that the tax rate and the pollution control
*Cost of Clean Air, 1973, "Annual Report of the Administrator of the
Environmental Protection Agency to the Congress of the United States."
Subsequent references to Cost of Clean Air mean any issue of the annual
report.
-------
costs are overstated by approximately the same factor. As a result the
projected emissions reductions would be unchanged by including corporate
income tax considerations since both the marginal costs of emissions
reductions (MCER) and the tax rate (TX) can be rescaled by a constant
fraction to derive approximations of the net control costs and net tax
rate. See appendix F for a discussion of the problems and biases implicit
in these assumptions.
Other alterations in the relative price of pollution control hard-
ware—due to the issuance of municipal or State revenue bonds to
subsidize corporate financing of pollution control devices, to preferential
exemptions from property taxes on pollution control gear, and to pollution
control related State income tax preferences—are likely to enhance further
the attractiveness of control hardware over tax payments. These, coupled
with the effects of the Federal corporate tax structure, are, in RTI's
judgment, likely to cause some understatement in the estimated emissions
reductions that would be achieved at the various tax rates projected in
this report. The magnitude of that understatement, though difficult to
evaluate for 1978 in view of continually emerging tax preferences on
pollution control gear at the local level, is not likely to have caused
large errors in predicted emissions reductions.
1.5 Assumptions, Limitations, and Capabilities
Several assumptions and limitations are present in this study
regarding the data inputs and methodology employed.
It has been assumed that the emissions control alternatives identified
in this study will be available in time for installation and operati-o^.
by 1978. If supply conditions delay their applications, the effectiveness
of the tax for 1978 will be less than, and the costs greater than, those
projected.
Because of the lack of data and the scope of this study, only a
limited number of control alternatives have been evaluated for each source.
Most of these alternatives have high control efficiencies (80 to 99 percent),
It is likely, however, that other control alternatives, including process
changes, would be induced with an emissions tax. This would tend to'
increase the projected effectiveness of the tax in mptivating emissions
reductions and, further, it would tend to lower costs from those presented
in this study.
-------
The dynamics of fossil-fuel supply and prices have not been explored
to the extent possible. These factors will play a critical role in
influencing the effectiveness and the costs, not only of the tax but also
of the regulatory approaches to achieving emissions reductions. However,
a preliminary analysis has been conducted of how the projected effective-
ness and costs of the tax may be influenced by future fuels supply.
The most obvious comparison to this report is the annually published
Cost of Clean Air. Any reader making comparisons should be alert to some
critical differences between the underlying assumptions and methodologies
of the two reports. The most obvious difference is that the Cost of Clean
Air shows total costs for reductions in sulfur dioxide emissions while
this report shows costs for sulfur, which constitutes one-half of the
equivalent mass of sulfur dioxide. Secondly, this study accommodates
not just one but several control options for every emissions source. For
fuel combustion sources, these options include: fuel switching among
several sulfur content fuel types, distinguished by location of origin;
three separate flue gas control hardware options; and emissions tax payments,
For other industrial sources, the control options include choices between
various hardware or process changes for each process source and tax pay-
ments. The Cost of Clean Air, on the other hand, generally has taken an
inflexible approach in imposing a control option on specific plants and
totaling up the resultant estimates. Furthermore, where fuel switching
has been considered only, the Cost of Clean Air has incorporated a simple,
low sulfur fuel cost premium to derive the cost of the alternative fuel.
The model of this study was more detailed in that fuel transportation costs
and supply considerations were built into the simulated array of fuel
options available to each plant. A final consideration is that RTI
\'
attempted to incorporate current refinements in the estimates of control
haroware costs. In some cases, these estimates (reported in appendixes
A through D) differed substantially from those used in the Cost of Clean
Air. In summary, the control costs projected here are not fully comparable
with those reported in the Cost of Clean Air owing, mainly, to different
methods of deriving low sulfur fuel prices, to the wider array of control
choices available to plants, and to the incorporation of more recent
estimates of control hardware costs. Despite these differences, however,
-------
the interested reader will note that the cost estimates of the two reports,
after appropriate adjustments for the units difference (sulfur versus
sulfur dioxide), are within the same order of magnitude.
1.6 Summary of Findings
Based on the results of the research presented in this study, it
appears that a national tax on the sulfur emissions of the five major
sources of this pollutant would be an effective means of inducing emissions
reductions. Specifically, table 2 shows the reductions, costs, and tax
payments projected for selected tax rates.
Although no direct comparisons have been developed here between
the costs under a system of emissions standards and those under a system of
emissions taxes, the aggregate cost to the Nation of emissions reductions
with a tax will be no higher than those under an emissions standards
approach to air quality management for a given reduction in emissions.*
In all likelihood, these costs under a tax would be substantially less
Table 2. Summary of the projected effectiveness and costs of a
national tax on the major sources of sulfur emissions--1978
Tax rate Percentage reductions Total Annualized Annual tax
(cents per pound of from unconstrained annual cost control cost payment
sulfur emissions) emission levels* (billions) (billions) (billions)
5
10
15
20
25
30
53
74
78
80
83
85
$1.8
2.7
3.4
3.9
4.4
4.9
$0.9
1.7
2.1
2.4
2.8
3.0
$0.9
1.0
1.3
1.6
1.7
1.9
*The single exception is the assumption that the New Source Performance
Standards are Implemented regardless of the'tax rate.
Source: Research Triangle Institute.
*It is possible, in fact likely, that the sum of tax payments and
control costs would be higher under an emissions tax policy than under
standards. The total cost to society, however, must not include emissions
tax payments since they simply represent income redistributions.
8
-------
than those under an emissions standards approach because of the efficiency
inducing properties of such a tax. A comparison of the emissions and cost
data presented in the Cost of Clean Air with emissions and control cost
data presented in this report under a tax strategy is presented in table 3.
As discussed above, the reader should not conclude that the only basis for
the differences in results between the Cost of Clean Air and this study
is due to the relative efficiencies of taxes and regulations, since
different methodologies were used to develop the cost of control estimates.
Under either a regulatory or tax approach, reductions in the emission
of other pollutants may be achieved when controlling sulfur oxide emissions.
This is due to the technology of the control alternatives. For example,
in applying flue gas desulfurization technologies to sulfur oxide dis-
charges, particulate emissions are usually reduced also. Similarly,
switching from coal to oil may also reduce particulate as well as sulfur
emissions. This added benefit has not been included in this study.
An emissions tax will significantly increase the demand for low sulfur
fuels since they constitute a particularly attractive means of reducing
sulfur emissions caused by fuel combustion. If the long-run supply of these
Table 3. Comparison of Costs of Clean Air and emissions tax results--1978
Source
Steam-electric
utilities
Area sources
Petroleum
refineries
Sulfuric add
plants
Primary non-
ferrous smelters
Cost of Clean Air data*
Uncontrolled
emissions
(thousand
tons of
sulfur)
l
14.075
3,887
2,202
929
2,541
Reductions in Emissions
(thousand
tons of
sulfur)
11,445
3,070
2,153
642
1,990
(percent)
81.3
79.0
97.7
69.1
78.3
Annual 1 zed
control
costs
(million
dollars)
$1,860
1 ,342
34
29
184
Emissions tax datat
Uncontrolled
emissions
(thousand
tons of
sulfur)
11,396
5,678
772
385
1,651
Reductions In
emissions
(thousand
tons of
sulfur)
9,265
4,486
754
266
1.293
Annual ized
control
costs
(million
dollars)
$1 ,600
500
>27
33
50
Required
tax rate
to induce
control
(cents per
pound of
sulfur
emissions
18
20
>30
10
4
*Cost of Clean Air, Environmental Protection Agency, Washington, D.C., 1973.
tOeveloped from data presented in this study for same percent reductions in emissions as implied In the Cost of
Clean Air.
-------
fuels is inelastic, prices may increase fairly substantially. This study
attempts to incorporate (into the determination of those fuel price
projections) supply considerations that, at this juncture, seem reasonable.
The reader is referred to section 3.2 and appendix A for more detailed
discussions of these assumptions.
Though it is difficult to place confidence limits on the estimates
of control costs used in this study, it is perhaps useful to distinguish
the elements of control costs along with RTI's overall confidence in the
estimated costs for those components. In general, the estimates of initial
purchase costs of pollution control gear are quite good; estimation errors
are on the order of+_ 10 percent. Equipment installations costs, on the
other hand, may vary by as much as 100 percent about the mean estimates.
Among annualized costs, the errors in annualized cost are jointly determined
by the above-mentioned capital cost estimates and potential errors in the
discount rate. Operating, maintenance, and replacement costs consist of
labor, power, water, and chemicals costs; all of these are subject to small
variations (on the order of +_ 10 percent) in the short run. A rough
weighting of these estimates of error according to the share of each
component in total annualized costs yields an average range of error in
individual estimates of +_ 16 to 20 percent.
For the industrial sources, use of alternative market values for
recovered sulfur and sulfuric acid of +_ 100 percent from the projected
1978 values ($10 per ton) is estimated to have little impact on the
effectiveness and costs of the tax for most sources. Likewise, control
cost deviations of +_ 20 percent do not significantly alter the results
of the study. It is likely, however, that the results are sensitive to
the number of available control alternatives. If more alternatives were
available to the industrial sources whose current control options manifest
low removal efficiencies, the costs of the tax for these sources would be
less than those projected, and the effectiveness greater at low-to-medium
(5- to 15-cent) tax rates.
10
-------
Chapter 2: THE APPLICATION OF EMISSIONS TAXES
FOR POLLUTION CONTROL
2.1 Introduction
The problem of air pollution, viewed from an economic perspective,
is one of overutilization of a scarce resource. The overutilized
resource is the waste removal capacity of the air; that is, the capacity
of the air to assimilate unwanted byproducts of production and consumption
without imposing damages on such receptors as people and plant life.
Historically, clean air has been a "free good", with more than enough
available to saturate demand. However, the accelerated use of the
atmosphere as a low-cost means of waste (or residuals) disposal has
created health, property, and esthetic damages. Reduction in the
damages inflicted upon society by a polluted atmosphere will require
rationing of the use of the atmosphere for residuals disposal.
Residuals charges are a market-type mechanism for rationing
environmental resources. Emissions taxes, one type of such charges,
have been proposed by an increasing number of people concerned about
the quality of the environment and the efficiency and costs of other,
nonmarket types of strategies. This chapter briefly examines the
rationale behind such charges with reference to a tax on sulfur emissions.
2.2 Air Quality and Market Failure
Most economic goods are rationed through a market process in which
product prices reflect society's tastes and desires and in which costs
reflect productive capabilities. This market process is generally re-
garded as a reasonably efficient means of resource rationing and allo-
cation. Clean air, however, has no effective market. Even though one
may desire clean air, there is no market where this preference can be
registered. The result is the overutilization of the atmosphere for
one service, residuals disposal, thereby transferring to society as a
whole the costs of residuals disposal rather than incorporating these
costs as a part of product costs and price. In this situation, an
external diseconomy is said to exist. Since the costs of residuals
disposal are not "internalized," product prices do not reflect
alternative uses of the atmosphere. As a result, too many private
11
-------
goods whose production generates residuals and too few public goods such '
as clean air are produced. If the value to society of the atmosphere
for purposes other than residuals discharges could be made to bear directly
on discharging activities, polluters would reduce their discharges to the
atmosphere.
Unfortunately, two conditions preclude the existence of an effective
market for clean air. First is the absence of well-defined and enforceable
property rights in the atmosphere. Second is the public-goods nature of
clear air.
Of the two conditions, the public-goods nature of clean air is the
primary reason for the lack of a market for clean air, since the creation
of property rights in the atmosphere has limited practical applicability.*
Public goods are goods that if supplied to one individual (e.g., national
defense) are available to all. In the case of clean air, if air pollution
is reduced in response to a "demand" on the part of some individuals,
pollution will be reduced for all. Thus, appropriate aggregate information
and responses needed to generate a market for public goods are generally
lacking.
Nevertheless, because of the desirability of charging for the
use of scarce resources, such as the assimilative capacity of the air,
other pricing mechanisms are available in the absence of markets. The
most often proposed alternative is the use of emissions taxes.
2.3 Emissions Taxes
Emissions taxes are a form of government intervention; the implicit
rationale for them is that property rights in the atmosphere are vested
in the public with the government acting as agent for the public interest
by rationing the use of the atmosphere for waste disposal.
When prices (taxes) are placed on the use of the atmosphere for
discharging the unwanted byproducts of production (e.g., sulfur oxides),
emitters have an incentive to economize on their use of the environment
just as relative resource prices currently guide decisionmakers to the
most efficient use and combinations of land, labor, and capital. The
abatement analysis burden is placed on corporate management which is,
*A. M. Freeman, III, The Economics of Pollution Control and Environmental
Quality, General Learning Press, 1971, pp. 1-27.~":
12
-------
or will become, more knowledgeable regarding the cost-effectiveness
of various control alternatives than government officials who are likely
to have only incomplete information. Conceptually, at least, emissions
taxes could be adjusted to the time of day, season, weather, or economic
conditions in order to reflect the variable nature of damages.
In the presence of an emissions tax, polluters will adjust their
individual level of control so that the last ton of emission reduction
from all sources imposes the same cost on each source. This is
significant for it means that the reduction in total emissions is
achieved at the least amount of the control cost.
Furthermore, the emissions tax provides a continuing incentive for
firms to seek newer and more efficient means of controlling their
discharge and to avoid judicial delay tactics. These desirable
tendencies are not usually encouraged under a system of emissions
standards.
With the imposition of an emissions tax, polluters can be expected
to pursue the least-cost means of residuals management. These costs can
be conveniently divided into two components: control costs and emissions
tax payments. The sum of these two costs—after adjusting for the effect
of income taxes—will be minimized. The extent to which emissions are
controlled will depend on the relative cost of control and the tax rate.
This decision analysis can also be examined on a per unit basis.
For example, as shown in figure 1, assume that before a tax is levied
on emissions, the plant is emitting E tons per year. A tax (T) per ton
of emissions will induce the plant to effect emissions reductions until
the marginal cost of doing so equals the tax rate; it will produce A units
of pollution control per year or, to put it another way, it will reduce its
emissions from E to E-A. The control cost of emissions reductions will be
the area OAB. The tax bill will be T(E-A); i.e., the tax rate times the
flow of remaining emissions, whose product is equal to the area ABCE.
Ideally, the tax rate should be set at a level sufficient to induce
reductions in emissions to a rate where the incremental damages of pollution
equals the incremental cost of emissions reductions (see fig. 2). However,
the application of emissions taxes to problems of air quality is not
dependent on the availability of reliable information regarding damages
from air pollution. Because air quality goals have been established by
13
-------
EPA, emissions reductions necessary to achieve these air quality standards
can be induced by emissions taxes. The remainder of the study explores
the relationships among alternative emission tax rates, control costs, and
resulting emissions.
MARGINAL COST OF
EMISSION REDUCTIONS
o
o
CONTROL COST OF
EMISSION REDUCTIONS
TAX PAYMENT
REDUCTIONS IN EMISSIONS
(TONS PER YEAR)
Figure 1. Emission source behavior in response to a tax.
MARGINAL DAMAGES
OF EMISSIONS
to
8
MARGINAL COSTS OF
EMISSION REDUCTIONS
Q^£
REDUCTIONS IN EMISSIONS
(TONS PER YEAR)
A E
Figure 2. Socially optimal level of emissions.
14
-------
Chapter 3: ANALYSIS OF THE EFFECTIVENESS AND COSTS
OF A TAX ON SULFUR EMISSIONS
3.1 Introduction
To estimate the probable effectiveness and costs that would result
from implementation of a tax on sulfur released into the atmosphere, an
emissions response model was developed and applied to three industrial
process sources, steam-electric utilities, and area heating sources. When
combined, these sources account for over 90 percent of U.S. sulfur emissions,
The model calculates controlled emissions and costs at specified tax rates
for each plant and aggregates each industry for the Nation as a whole.
Controlled emissions and costs for area sources are also estimated and
added to provide national totals. This analysis was performed for the
year 1978, 5 years hence, on the assumption that the projected control
systems will be availble and could be installed by then. The factors
used and the results of the analysis are given below.
3.2 The Emissions Response Model
The procedure of this study is to approximate, heuristically, the
individual plant's reponse to an emissions tax. This is accomplished by
the use of an emissions response model developed for this study. Through-
out the analysis several simplifying assumptions are made regarding plant •
or source operation regardless of classification. Those assumptions are:
(1) That product output at all existing point sources remains
constant; no plants go out of business or curtail production
in response to emissions control taxes;
(2) That the plant manager chooses the combination of emissions
tax payments and emissions control options in such a way that
his total outlay is minimized;
(3) That expected annualized control costs are sufficiently approxi-
mated by previous studies for EPA cited herein (and discussed
in detail in appendixes A through D);
(4) That the finite number of control options studied here are
representative of those that producers will face in 1978; i.e.,
that no extraordinary technical changes in sulfur oxide control
technology will occur during the next 5 years.
15
-------
The remainder of this section discusses the assumptions and method-
ology for each of the broad emission source categories: fuel combustion
sources and industrial process sources.
3.2.1 Fuel Combustion Sources
The two broad types of fuel combustion sources are steam-electric
power plants and area sources; i.e., commercial, residential, and
industrial space heating. Both of these sources are assumed to have the
option of fuel switching; i.e., changing from one fuel type to another
or from one sulfur content to another. Only power plants are assumed to
have the option of removing entrained sulfur from the carrier gas. The
following paragraphs first discuss the assumptions regarding fuel avail-
ability supplies and prices (sec. 3.2J.I). With that background, the
next sections summarize the additional assumptions and methodology
involved in deriving the cost-minimizing control options for steam-electric
power plants (sec. 3.2.1.2) and area sources (sec. 3.2.1.3). More detailed
discussions are presented in appendix A.
3.2.1.1 Fuel Supplies and Prices
3.2.1.1.1 Fuel Supplies. Because available fuel supplies are not
uniformly distributed with respect to sulfur content, Btu value, and
location, it has been necessary to explicitly incorporate consideration
of these parameters in this study. The approach employed for each fuel
is presented below.
a. Coal. The 1978 maximum production of coal was projected within
each of seven designated coal-producing basins by sulfur contents. The.
method of projection relied on a technique developed by the MITRE Corporation.
The projections involve the use of a growth rate within each of the 7
basins for each of 9 sulfur content groupings. Those projections yield
63 (9 sulfur contents times 7 basins) estimates of maximum commercial
bituminous coal production.
The seven basins were further subdivided according to the 19 coal-
producing districts defined by the U.S. Department of Interior using projected
regional proportions developed by-Battelle. The resulting projections
comprised 171 maximum supply estimates for coal (19 districts times 9 sulfur
contents). The Btu content of these coals were assumed invariant within
each of the 19 districts; the Btu values were averages for each region
16
-------
reported by the MITRE Corporation. Initially prices of the 171 fuels at
their origins were adapted from the Battelle study. Subsequently they were
adjusted in a manner discussed below (sec. 3.2.1.1.2). Because some sulfur
content coals are not available within all regions, there were 21 empty
cells in the fuel price and supply matrices. Consequently, the total number
of coals, distinguished by district of origin and by sulfur content, is
150. To estimate the delivered price of these coals, RTI employed the
transportation cost matrix developed by Battelle. That matrix provides
estimates of shipping costs from the 19 coal regions to 50 destinations.
By adding those transportation costs to the coal prices at their
origin for each of the 50 destinations, tables of delivered coal prices
were developed. Each existing plant or source was then associated with one
of those 50 destinations; this determined which of the 50 delivered-price
vectors was relevant to a particular decision unit. Those vectors of
delivered coal prices then become part of the decision unit's control costs.
b. Residual Oil. Only about one-third of annual U.S. consumption
of residual oil is domestically produced. The maximum expected supplies of
these domestic residual oils were projected for 1978 by Petroleum
Administration for Defense (PAD) districts in a previous study by the
MITRE Corporation. Those projected supplies were then allocated among
the 12 oil-producing districts defined by the U.S. Department of the
Interior, again on the basis of the distributions developed by Battelle.
The result was a matrix of 48 domestically supplied oils (12 origins times
5 sulfur contents minus 12 empty cells). • In addition, imported residual
oil was also assumed available at four ports of entry (east coast, gulf coast,
west coast, Great Lakes). Wellhead or POE prices were adapted from the
Battelle study. A matrix of transportation prices developed by Battelle
was used to develop estimates of delivered prices. These vectors of delivered
oil prices, plus the coal prices determined as described above, are the
fuel cost-sulfur content alternatives.
c. Natural Gas. For utilities, the supply of gas was assumed to be
perfectly elastic to current users up to the quantities currently used and
to be perfectly inelastic above 'those quantities. Therefore, no utilities
were allowed to switch to gas or to increase gas consumption. For
17
-------
area sources, the market share of gas was assumed to remain constant at
1970 proportions among individual States. All gas prices assumed were
those developed by Battelle.
d. Distillate Oil. Distillate oil, used primarily as a source of
energy for area sources, was assumed to be unlimited in supply at prices
projected by Battelle.
3.2.1.1.2 Fuel Price Adjustment. The first iteration of the model
at a zero tax implied that the maximum projected supplies of coal and
residual oil would be exceeded for some of the 198 domestically produced
fuels. The most obvious way to handle this problem was to incorporate
the 198 supply constraints while simultaneously minimizing the control
costs associated with all fuel combustion sources. That approach, however,
would have implied a major linear programming effort beyond the resource
constraints of this project. The alternative chosen here was heuristic.
Whenever the demand for a specific coal exceeded the maximum predicted
supply, its price was arbitrarily increased by a small amount (generally
on the order of 5 percent of base point prices). The entire model was
then iterated again. This prc:ess was continued until the total domestic
demand for coal was smaller than the corresponding total projected domestic
supplies. The implicit assumption here of domestic self-sufficiency of
coal appears warranted in view of the large U.S. reserves of coal, and
historical consumption patterns.
For residual oil, demand exceeded domestic supply. However, addi-
tional supplies were assumed to be available from foreign suppliers at
given prices. The sensitivity of the projected responses of the utilities
to the assumed oil prices was, however, evaluated (see chapter 4). The
general effect of this method of approximating price responses is likely
to induce a downward bias in the costs of switching to low-sulfur fuels.
Consequently, one could expect more control hardware applications than the
model predicts.
3.2.1.2 Steam-Electric Power Plants. Each existing power plant
was identified and its output, size, and operating characteristics re-
corded. Those plants were assumed to face three hardware options for
the removal of sulfur from the combustion gases, each with an assumed
18
-------
potential control efficiency. They are the dry limestone, wet limestone,
and magnesia scrubbing processes. The annualized costs of those options
were intended to approximate the annual outlays whose present value over
the firm's planning horizon equals the expected investment cost plus the
discounted present value of their associated operating costs. Estimates
of those costs were developed as functions of the sulfur content of the
fuel, the megawatt capacities of the boilers at the plant, and the annual
output at the plant. The cost estimating equations, formulated in previous
studies for EPA, were adapted for this study. New power plants were
projected to come on line in accordance with the National Coal Association
listing of conventional steam-electric plants that are planned or under
construction during 1971-1977.* These plants were assumed to adhere to
the New Source Performance Standards. Consequently, new coal- or
oil-fired plants were expected to install either a wet limestone or magnesia
scrubbing system. The costs of those systems were not included in the
total costs of control in this study.
Each plant then was assumed to minimize the total annual outlays of
three components of costs: emissions tax payments, annualized abatement
costs, and delivered fuel costs. For each tax rate considered, the computer
simulation model scanned the sum of these three costs for every fuel type
by sulfur content and for every control hardware option. About 1,000
alternatives exist for each source. The combination of hardware, fuel
type, and emissions tax payments that minimized total outlay was the pre-
dicted response of the plant. The costs of these responses and the corre-
sponding reductions in sulfur emissions at various tax rates per unit of
sulfur emissions were aggregated over all plants.
3.2.1.3 Area Sources. Area sources comprise residential, commercial,
and industrial fuel consumers. Since the available statistics relating
numbers, size, and distribution of these heating units are either scant
or nonexistent, this study attempted to simulate the response of individual
emitters by analyzing each State as an aggregate. Though this approach
'has rather obvious drawbacks, the predicted behavioral response to the
emissions taxes is felt to be a reasonable first approximation.
*National Coal Association, Steam-Electric Plant Factors, 1969.Wash-
ington: National Coal Association, 1969, table 4.
19
-------
The details of RTI's approach for an individual State are as follows.
First, the percentages of total Btu input from coal, residual oil, natural
gas, and distillate oil for each of the three area sources were determined
for each State from published sources. Also, the absolute values of those
fuel consumption rates for 1970 were recorded by State. Then the projected
demands for 1978 were developed by applying separate growth rates to 1970
demands. No hardware control options were allowed for area sources.
Further, among area sources only commercial and industrial users were
allowed fuel switching.
a. Residential Sources. Since residential sources consume only
natural gas and distillate oil, it was assumed that they did not switch
to alternative fuels. It was further assumed that the residential fuel
market shares accounted for by those two fuels remained constant over
time. The growth rates applied to the.1970 State consumption rates were
the projected State population growth rates published in the Survey of
Current Business. Residential sources were then presumed to pay an emissions
tax based on the sulfur content of those respective fuels. It is recognized
that this is a rough first approximation that assumes demand is affected
only by population growth and not by changes in the relative prices of
distillate (because of the sulfur tax). If the demand for nonelectric
home heating is relatively inelastic and if the supplies of natural gas are
also quite limited, these assumptions are not likely to have caused serious
prediction errors.
b. Commercial and Industrial Sources. Commercial and industrial
emission source control options were assumed to be identical. The 1970
shares of total heat input accounted for by natural gas and by distillate
oil among these sources were projected to remain stable through 1978.
Absolute consumption values for each State were developed by applying
the projected growth rate in overall national employment.
The proportion of heat input accounted for by residual oil and coal
among commercial and industrial sources within a State was also assumed
fixed at the 1970 proportion; but the relative shares of those two fuels
within their jointly held market share was not.
The annualized cost whose present value was assumed to approximate
the outlay necessary to install the coal-to-residual-oil-boiler conversion
units was $11 per billion Btu.
20
-------
The total outlays that commercial and industrial boilers are assumed
to minimize, then, are the sum of the emissions tax and fuel costs of
natural gas and distillate oil consumption—over which they have no control
since the market shares accounted for by those two fuels remain constant--
and the emissions tax, the fuel costs, and the boiler conversion costs
of residual oil and coal consumption.
Once again it is clear that many complicated interrelationships
have been rather quickly simplified: relative price sensitivities among
fuels were not built in for natural gas and distillate oils vis-a-vis
the aggregate of residual oil and coal consumption; growth trends were
not developed on a State-by-State basis; no allowance was made for flue
gas cleaning which may be feasible for large industrial sources; and
all commercial-industrial sources in a State were forced to consume their
present residual oil and coal heat inputs as either one or_ the other;
i.e., no variable shares were allowed between these two fuels. Given
sufficient data, these many analytical refinements might have been
justifiable, but it was felt that the paucity of accurate information
about these sources and their operating characteristics simply did not
warrant attempts to build them in for this study. Yet, it was felt that
the analysis did reflect a reasonable first approximation of the national
emissions reductions that a sulfur tax would evoke from those area sources.
3.2.2 Industrial Process Sources
Industrial process sources of sulfur emissions comprise three
industrial groupings: petroleum refineries (appendix B), sulfuric acid
plants (appendix C), and primary nonferrous smelters (appendix D). The
general approach for all of these sources was to determine hardware
alternatives for sulfur oxide control and to annualize the cost of those
installations. The individual plant's response to a tax was assumed to
follow the cost-minimizing hypothesis. Whenever the total annual tax
payments on the portion of emissions that were preventable with one of
the hardware control options exceeded the annual!zed cost of that option,
the plant was assumed to implement the control practice. Some of the
details of this approach are discussed below for each major process source
category.
21
-------
3.2.2.1 Petroleum Refineries. The three major sources of refinery
emissions are catalyst regenerators, Claus sulfur recovery plants, and
fuel combustion sources. Approximately 13 percent of the sulfur content
of crude oil processed at the plant is emitted from these three sources.
The remainder is either recovered, emitted to waterways, or retained in
marketed petroleum products.
The control option that was presumed most feasible for the control
of emissions from catalyst regenerators was hydrodesulfurization of the
catalytic cracker feedstock. This process essentially allows conversion
of some of the sulfur in the feedstock to elemental sulfur. Each existing
refinery in the United States was identified according to the type of
catalyst regenerator in operation at the plant; then annualized costs
of the hydrodesulfurization process were estimated according to several
refinery capacities.
A refinery may or may not have a Claus plant, essentially a process
that is used to convert hydrogen sulfide (HgS) bearing refinery off-gases
to elemental sulfur. If a refinery has no Claus plant, the H2$ stream is
flared to the atmosphere resulting in substantial SOg emissions.
Each existing plant was then identified in terms of the type of
catalyst regeneration it had in place and according to whether it had a
Claus plant. The availability of EPA-estimated sulfur emission factors
then facilitated an estimate of current emissions at each refinery.
The computer model simulated each refinery's response to an emissions
tax in the following way. First, if the specific refinery's capacity did
not correspond exactly to those for which control costs were estimated,
interpolation was used. Then, sulfur emissions estimates were developed
for these refinery sources from the emission factors and from estimates
of refinery throughput. At each tax rate, total annual emissions tax
payments associated with each of the three emissions sources were compared
to the annualized cost of achieving some percentage, reduction in those tax
payments. If those costs were smaller than the associated tax savings, the
refinery was projected to implement the subject control option.
This procedure resulted in aggregated estimates of abatement costs,
of tax payments, and of emissions reductions across all three major
22
-------
emissions sources at any given refinery for each projected tax rate.
Further aggregation across refineries yielded the industry emissions
response curve for different levels of total abatement costs and for
different sulfur tax rates or, identically, for different marginal costs
of emissions reduction.
3.2.2.2 Sulfuric Acid Plants. Sulfuric acid plants emit sulfur in
two forms: (1) as a gas, sulfur dioxide (SCL), and (2) as an acid mist.
The former results from incomplete absorption of SC^ during the produc-
tion process; the latter emerges from the absorption tower in the process
off-gases. The two control techniques that were considered for gaseous
emissions were the dual absorption and the sodium sulfite scrubbing pro-
cesses. Their expected annualized costs are reported in appendix C.
For both gaseous and mist effluents, total potential emissions were
computed using EPA estimates of emission factors and estimates of each
plant's production of acid. By comparing total implied emissions tax
payments under various tax rates to the annualized costs of control for
each type of effluent, RTI projected whether or not abatement practices
would be implemented. The choice criterion was whether or not the tax
savings exceeded the annualized costs of the control option.
Total costs of tax payments, plus abatement costs, less the value
of recovered sulfur were then aggregated for each plant and also across
plants for the industry. The resulting projections of emissions reductions
versus total annualized costs and versus marginal costs (tax rates) are
reported in appendix C.
3.2.2.3 Primary Nonferrous Smelters. Primary nonferrous smelters
include copper, lead, and zinc smelters. Each has a unique process
operation that generates emissions at several points. Copper smelters
emit S02 from the roaster (if the plant has one), reverberatory furnaces,
and converters. Zinc smelters emit sulfur from the roaster or the
roaster-sinterer, depending upon the type of smelting operation.
Emissions from lead smelters derive mostly from the sintering operation,
while small amounts are generated by blast furnace operation.
The major control techniques for smelters are: sulfuric acid plants,
lime and limestone scrubbing, amine absorption, ammonia scrubbing, and
sodium sulfite-bisulfite absorption. The relevant control options and
23
-------
their associated efficiencies for each smelter type were determined and
are reported in appendix D. Similarly, annualized emission control
costs were developed for each relevant option for representative plant
sizes; these are also reported in appendix D.
After identification of each smelter, the annualized control costs
for the specified control options were developed by interpolation from
the previously mentioned cost estimates. By using the same technique
previously applied to other process sources, a control option was
projected to be implemented if its total annual ized costs, less the value
of recovered sulfur, were less than the tax payments implied in the absence
of the device. Also, the results were aggregated for each smelter, across
smelters for each nonferrous metals industry, and finally across all
primary nonferrous sources. The aggregate of projected emissions reductions
versus total annualized control costs and versus marginal costs (tax rates)
is shown in appendix D.
3.3. Cost of Control Functions
A tax on sulfur emissions is expected to induce firms to control
emissions to the level where the sum of the annualized emissions control
costs and tax payments is minimized. Cost minimization is achieved by
selecting the level of emissions reductions at which the incremental
or marginal cost of control equals the tax rate. Projections of the
effectiveness and costs of a tax on sulfur emissions depend, therefore,
on the control alternatives assumed to be available, and their costs
and effectiveness in reducing emissions. Extensive reviews of published
studies and private communications with EPA and other knowledgeable
sources have been conducted in order to identify the sulfur control
alternatives likely to be available by 1978, and their costs and sulfur
removal efficiencies. The alternatives selected and costed are presented
in appendixes A through E. These alternatives have been costed for
controlling sources of various sizes and process configurations
to estimate cost functions for each source. A national listing of the
major source of sulfur emissions (953 steam-electric power plants,
50 area sources (States), 263 petroleum refineries, 183 sulfuric acid plants,
and 28 primary nonferrous smelters) and the process configuration of each
has been used in order to conduct a plant-by-plant analysis.
24
-------
The long run* total and marginal'costs of functions, combined for
all major sources under consideration, are shown in figures 3 and 4,
respectively. These functions are plotted from analytical results of
the emissions tax response model. The total cost function (LTC) increases
at an increasing rate reflecting the higher costs of controlling smaller
plants, the higher cost per ton of sulfur removal at high control
efficiencies, and the price premiums for low sulfur fuels. The marginal
cost function (LMC), being the first derivative of the total cost function,
also increases at an increasing rate throughout the range of emissions
reductions presented.
3.4 Effectiveness
The effectiveness of the tax is defined, for the purposes of this study,
as the reduction in sulfur emissions that would be induced by a tax on
sulfur emissions. The response of the emissions sources to the tax is a
function of the cost of control and the tax rate. This analysis has been
conducted in 5-cent increments for tax rates from 0 to 30 cents per pound
of sulfur emitted.
In the absence of an emissions tax and without the application of
any emissions standards except those applicable to new sources, sulfur
emissions in 1978 are projected to be about 20 million tons anually from
these major sources (see table 4). Eighty-five percent of these emissions
would derive from fuel combustion sources. Since the projected rates of
uncontrolled emissions from these sources require data on fuel demand,
supplies, and prices by sulfur content (see appendix A) (all of which are
difficult to project accurately), these projections of the uncontrolled
emissions of fuel combustion sources should be cautiously interpreted.
The combined responses to the tax of all the major sources are pro-
vided in figure 5 and table 5. For small taxes of 1 to 10 cents per pound
of sulfur, large reductions in sulfur emissions are projected. These pro-
jections hinge on RTI's cost estimates which indicate, for example, that
at a marginal cost of sulfur removal of 10 cents per pound for all plants,
the aggregate of all sulfur emissions sources would reduce emissions by
*The "long run" is a time period long enough for firms to order and
install control equipment or negotiate fuel contracts. It has been assumed
that such adjustments can be made by 1978.
25
-------
10
REDUCTIONS IN SULFUR EMISSIONS
(APPROXIMATE PERCENT)
2O 30 40 SO 60 70
4 68 10 12 14
REDUCTIONS IN SULFUR EMISSIONS
(MILLION TONS PER YEAR)
16
18
Figure 3. Total cost* of reductions in sulfur emissions from all major
sources combined—1978 (*cost does not include emissions tax payments)
(Source: Research Triangle Institute).
600
to
REDUCTIONS IN SULFUR EMISSIONS
(APPROXIMATE PERCENT)
20 3O 40 SO 60 70
90
500
.
o _
300
W Q-
_l £
< E
I 8 200
i
too
ILMC
46 8 K) 12 14
REDUCTIONS IN SULFUR EMISSIONS
(MILLION TONS PER YEAR)
16
18
Figure 4. Marginal cost* of reductions in sulfur emissions from all major
sources combined—1978 (*cost does not include emissions tax payments)
(Source: Research Triangle Institute).
26
-------
Table 4. Projected sulfur emissions from
major sources—1978*
Source
Steam-electric power plants
Area sources
Petroleum refineries
Sulfuric acid plants
Primary nonferrous smelters
Total from all sources
Annual sulfur emissions
(thousand tons of sulfur)
11,396
5,679
772
376
1,650
19,873
Distribution
(percent)
57.3
28.6
3.9
1.9
8.3
100.0
*Assuming only controls required by the New Source Performance Standards.
Source: Research Triangle Institute.
100
5
2>
2 a
8*!
i? to
o
9O
80
SULFUR EMISSIONS REDUCTIONS
(RIGHT SCALE)
SULFUR EMISSIONS
(LEFT SCALE)
o
oc
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
|
o
CO
Figure 5. Effectiveness of a tax on the sulfur emissions from all major
sources combined--!978 (Source: Research Triangle Institute).
27
-------
Table 5. Projected response of all major sources combined
to a national tax on sulfur emissions--1978
Reductions Total Annual i zed
Emissions source Emissions 1n emissions annual control Annual tax
(thousand from zero tax costs costs payment
tons) (thousand tons) (thousands) (thousands) (thousands)
Tax rate
Steam-electric utilities
Area sources
Petroleum refineries
Sulfurlc add plants
: 5 cents per pound of sulfur emitted
5
2
Primary nonferrous smelters
Total from all sources
Tax rate
Steam-electric utilities
Area sources
Petroleum refineries
Sulfurlc add plants
9
:
2
1
Primary nonferrous smelters
Total from all sources
Tax rate
Steam-electric utilities
Area sources
Petroleum refineries
Sulfurlc add plants
5
:
2
1
Primary nonferrous smelters
Total from all sources
Tax rate
Steam-electric utilities
Area sources
Petroleum refineries
Sulfurlc acid plants
4
•
•
1
1
Primary nonferrous smelters
Total from all sources
Tax rate
Steam-electric utilities
Area sources
Petroleum refineries
Sulfurlc add plants
3
:
1
1
Primary nonferrous smelters
Total from all sources
Tax rate
Steam-electric utilities
Area sources
Petroleum refineries
Sulfurlc add plants
3
•
1
1
Primary nonferrous smelters
Total from all sources
3
,159.0
,854.7
608.1
385.4
278.9
,286.1
10 cents
,961.8
,494.3
537.5
96.7
110.6
,200.9
15 cents
,277.7
,383.1
509.6
60.9
94.8
,324.1
20 cents
,948.2
,377.2
499.6
48.7
81.4
,955.1
25 cents
.599.5
,115.2
497.2
47.4
79.6
,338.9
30 cents
,432.1
,030.9
496.4
46.9
79.2
,085.5
6
2
1
10
per
,237.0
,822.8
163.7
0.0
,371.5
,595.0
pound
8,434.2
4
1
•14
per
9
4
1
15
per
,183.2
234.0
288.8
,539.9
,680.1
pound
,118.2
,296.5
261.9
324.4
,555.6
.556.6
pound
9,447.7
4
1
15
per
9
4
1
16
per
9
4
1
16
*
,300.3
271.9
336.7
,569.1
,925.7
pound
,796.5
,562.2
274.4
337.9
,570.9
,541.9
pound
,963.9
,646.6
275.3
338.5
,571 .2
,795.5
$1,171,808 $
466,497
66,403
38,546
85,098
$1,828,352 $
of sulfur emitted
$1,774,661 $1
654,265
122,276
59,410
98,345
$2,708,957 $1
of sulfur emitted
$2,215,069 $1
796,647
174,631
66,797
108,837
$3.361,981 $2
of sulfur emitted
$2.589,688 $1
934,526
225,008
72,018
117,873
$3,939,113 $2
of sulfur emitted
$2,911,882 $2
1,059,893
274,856
76,831
125,894
$4,449,358 $2
of sulfur emitted
$3,186,375 $2
1 ,164,706
324,519
'81 ,548
133,820
$4,890,980 $3
655
181
5
57
899
,182
355
14
40
76
,668
,531
382
21
48
80
,064
,810
383
25
52
85
,356
,112
502
26
53
86
.779
,326
546
26
53
86
,039
,881
,026
,596
0
,206
,709
,231
,391
,730
,074
,233
,659
,656
,344
,670
,486
,386
,542
,295
,646
,056
,525
,323
,845
,048
,214
.161
,110
,103
,636
,992
.153
,681
,392
,288
,506
$
$
$
$1
$
$1
$
$1
$
$1
$
$1
515
285
60
38
27
928
592
298
107
19
22
,040
683
414
152
18
28
,297
779
550
199
19
32
,582
799
557
248
'23
39
,669
859
618
297
28
47
,851
,928
,472
,808
,546
,892
,646
,432
,874
,546
,335
.112
.299
,417
,306
,961
,311
,451
,446
,394
,881
,952
,492
,550
,269
,836
,680
,696
,722
.791
,725
,384
,554
,837
,158
,532
,477
Source: Research Triangle Institute.
28
-------
more than 70 percent. Beyond tax rates of 10 cents, only small additional
amounts of reduction are induced. For example, a 30-cent tax would only
yield an additional 10-percent reduction.
3.5 Costs
The costs of sulfur emissions tax strategy would consist of the tax
payments plus the costs of control, both of which would initially be paid
by the polluting sources. To the industry, there is no significant conceptual
difference in these costs since they both become part of the costs of pro-
duction and must either be absorbed in profits or shifted to customers and,
ultimately, to consumers. From the perspective of society, however, the
difference between the tax payments and the costs of control is significant.
In a full employment economy, allocation of resources for production of
emissions control equipment implies a reduction in the production of other
goods and services. The tax payments, however, are transfers of income
from industry to government, and imply no reduction in production.
The total annualized costs to all major sources together are shown
in figure 6. Since the tax induces emission control, the total costs of
the tax increase at a decreasing rate.
10
20
6
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
25
30
Figure 6. Total costs induced by a tax on the sulfur emissions
from all major sources combined--1978 (Source: Research Triangle
Institute).
29
-------
3.6 Tax Revenues
From the perspective of government, the tax payments by the polluting
sources are revenues. These revenues are shown in figure 7 for all tax
rates under consideration. At a tax rate of 10 cents, revenues would be
about $1 billion annually or about $5 per capita. A tax of 30 cents would
almost double revenues from those projected for the 10-cent tax. It should
be noted, however, that the net increase in government revenues would not
be the full amount of the emissions tax since emissions tax payments would
reduce the firm's income tax liability by the amount equal to the tax rate
times the total emissions tax liability. For example, if the corporate
tax rate were 50 percent, the net increase in government revenues would be
one-half of the total emissions tax proceeds.
3.7 Cost-Benefit Analysis
Ideally, as discussed in chapter 2, the application of emissions taxes
for environmental quality management would not be based on cost-effectiveness
but rather on cost-benefit analyses. Data on the nature of the benefit
10 15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure 7. Tax revenues from a tax on sulfur emissions from all major sources
combined—1978 (Source: Research Triangle Institute).
30
-------
function for sulfur emissions reductions are,currently quite incomplete.
However, estimates developed from a recent compilation and evaluation of
the damages of residential property, materials, health, and vegetation from
the presence of sulfur oxides in the atmosphere for 1968 indicate that the
national cost averaged about $500 per ton of sulfur*. The authors of that
study point out a number of limitations in the data, but conclude: (1) these
are the best estimates currently available, and (2) for the present it must
be assumed that the marginal and average benefits (or avoidance of damages)
are equal. Using their estimate for 1968 of $500 per ton and assuming that
the damages in 1978 are the same per ton, a tax of 25 cents would equate
the costs and benefits of control at the margin. It should be noted that
the benefits would vary from region to region as would the costs.
3.8 Impacts on Consumer Prices
It is beyond the scope of this study to provide any extensive analysis
of the likely incidence of the sulfur tax. However, a preliminary analysis
is possible.
The effect of a tax on sulfur emissions (or, for that matter, of
regulation on emissions) will be an increase in the marginal costs of
production for every affected firm in the industry whose emissions are
being taxed. Since the horizontal summation of the marginal cost curves
of these firms yields the industry supply curve (assuming the absence of
external economies or diseconomies and a perfectly competitive industry),
the effect of a tax is to shift the industry supply curve upward and to
the left. Assume, for the sake of expositional simplicity, that the sulfur
tax implies a uniform increase (t) in the marginal costs of producing Q,
an output whose production generates sulfur emissions. One can then depict
the effect of the tax as an upward shift in the supply schedule of the
subject industry by the constant amount of the uniform tax-induced increase
(t) in marginal costs. The supply schedule shifts from S to S + t in figure
8. Since the price and output of Q are determined by the intersection of
the supply and demand schedules, the sulfur tax would induce a reduction
in output from QQ to Q,, and an increase in the price per unit of Q from
P0 to P1.
*Larry Barrett and Thomas Waddell, Cost of Air Pollution Damage: A
Status Report, Environmental Protection Agency, Research Triangle Park,
N. C., 1973, p. 61.
31
-------
Figure 8. The incidence of a sulfur emissions tax.
The importance of this analytical device derives from the obvious
fact that the percentage increase in the price of the product whose output
generates pollution depends critically upon the slope of both the supply
and demand schedules. The demand schedule, in general, will become more
horizontal (vertical) according to whether there are (are not) good sub-
stitutes for Q and according to whether the product constitutes a large
(small) share of the average consumer's budget. The supply schedule, on
the other hand, becomes more horizontal (vertical) according to whether
inputs to the production of Q can (cannot) be easily shifted into the pro-
duction of other products.
As the supply and demand schedules are depicted in figure 8, only
(P, - PQ)/t of the uniform increase in the per unit cost of Q production
is passed on to consumers. The remainder, (PQ - ?2)/t, is absorbed by
producers; they are forced to forego the profits (or rents) that would
otherwise accrue to them. If all of the increase in costs is accounted
for by tax payments, the government would collect emissions tax revenues
equal to the area ^1^2 dunn9 eac& Period.
Now, observe the extreme possibilities. Assume that the demand
schedule continues to be represented by D in figure 8 but that it is
32 ft
-------
virtually impossible to shift inputs out of Q production. The result would
be a vertical supply schedule at the quantity Q«. All of the emissions tax-
induced costs would be absorbed by producers as forfeited profits. The
price would remain at P«. At the opposite extreme, assume that all inputs
to Q production are supplied to producers in unlimited quantities at constant
prices (perfectly elastically supplied). The consequence is a horizontal
supply schedule. All of the costs of the emissions tax would devolve onto
the final consumers. Prices would rise by the full amount of the tax-induced
costs (to P3) and consumption would fall (to Q3).
A similar analysis of extreme behavior in demand is possible by assuming
the supply schedule remains stable at S and by allowing the demand schedule
to vary in slope. If, for example, there are nearly perfect nonpolluting
substitutes for Q, consumers will not tolerate any increase in its price;
the demand schedule will be perfectly elastic (horizontal). Producers will
have to absorb all of the cost imposed by the tax, and output will fall to
Q« as the marginal producers of Q go out of business. On the other hand,
if the consumer has no ready access to substitutes for Q or if Q is a very
small share of his budget (so that price increases go virtually unnoticed),
the demand schedule is likely to be perfectly inelastic (vertical) at QQ.
This would enable producers to shift the entire burden of the tax to con-
sumers. The price of Q would rise to P, while the output would remain
constant at QQ.
An initial analysis of these price effects can be accomplished by
assuming what, for consumers, is a worst-case situation—that the entire
cost increase attributable to the tax is passed on to them. This analysis
combines the assumption that inputs to the affected industry are relatively
mobile (supply is elastic) while consumer preference for that product is
relatively inflexible, in the face of price changes. Allowing these assump-
tions, RTI was able to adapt its previously developed model to project the
consumer price increases implied by increases in industry production costs.
In a previous study for EPA*, RTI developed a model for projecting the
impact on consumer prices of increases in industry costs. This model utilizes
*D. A. LeSourd and F. 6. Bunyard, eds., Comprehensive Study of Specified
Air Pollution Sources to Access the Economic Impact of Air Quality Standards,
EPA contract No. 68-02-0088, Research Triangle Institute, Research Triangle
Park, N. C., August 1972.
33
-------
the 1963 national input-output table (at the 364 sector level) plus a
disaggregated consumer demand sector with 80 subcategories of personal
consumption expenditures.*
The limitations of input-output analysis are well known and the impli-
cations of using a 1963 structure of product to represent an economy 15
years hence are obvious and need not be further discussed here. Price
increases may take many forms such as reductions in quality or service;
they may also be distributed discriminately to different customers of the
industry's products. Suffice it to say that the model can only provide a
first approximation of what is a very complex and not well understood pro-
cess. Nevertheless, taken in that light, the model does provide a useful
appraisal of the possible price impacts of a tax on sulfur emissions.
Table 6 shows the initial price increases for selected tax rates,
on a percentage basis using 1970 product prices as the base; these increases
are projected for each of five sources assuming that they shift the entire
cost of the tax in the form of higher product prices. All annualized
industry costs associated with the tax or with pollution abatement were
halved to account for the effect of government cost-sharing through the
corporation tax structure. See appendix F-
Table 6. Projected initial price increases resulting
from a tax on sulfur emissions
(percent increase over 1970 average prices)
Source
Steam-electric
Petroleum refining
Sulfuric acid production
Primary copper smelting
Primary zinc smelting
Primary lead smelting
(cents per
5 10
2.2
0.1
2.6
T.O
0.9
D.7
3.4
0.2
4.0
1.1
1.4
0.8
Tax rate
pound of sulfur
15 20
4.2
0.3
4.5
1-2
1.8
0.9
4.9
0.4
4.9
1.3
1.9
1.1
emitted)
25 30
5.0
0.5
5.2
1.4
2.0
1.2
6.2
0.6
5.5
1.5
2.1
1.3
Source: Research Triangle Institute.
*This index is a Paasche-type measure of consumer price changes, since
the current composition of expenditures is used to weight the components of
the index.
34
-------
FOOD AND TOBACCO
0.15
0
0.15
0
O.ISp-
O.L
CLOTHING .ACCESSORIES, AND JEWELRY
0.75,-
0.15]
OJ
0.25
HOUSING
HOUSEHOLD OPERATION
MEDICAL CARE EXPENSES
CO
UJ Q.I5
< 0
UJ
g 0-25
PERSONAL BUSINESS
TRANSPORTATION
RECREATION
PRIVATE EDUCATION AND RESEARCH
RELIGIOUS AND WELFARE ACTIVITIES
FOREIGN TRAVEL AND OTHER
PERSONAL CONSUMPTION EXPENDITURES. TOTAL
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure 9. Projected impact on consumer prices of a tax on the sulfur
emissions from all major sources—1978 (Source: Research Triangle
Institute).
35
-------
Figure 9 depicts the effect of the intermediate goods price increases
(table 6) on final product prices. Among the six identified intermediate
product categories, higher electricity prices would have the most significant
impact on consumer prices. For the highest tax rate examined (30 cents),
the increase in consumer prices would be about 0.15 percent (fig. 9, bottom
panel). The percentage price increases in figure 9 are small relative to
those in table 6 since the products in the table are only a few of the
intermediate goods used to produce the final goods and services in the
figure. The expenditure category that includes relatively more of these
intermediate goods and that therefore would experience a larger percentage
increase in prices is household operation. The other expenditure categories
are affected to a much smaller extent. Overall, it seems apparent that
the sulfur-tax-induced increases in consumer prices would not be substantial.
36
-------
Chapter 4: SOURCE-BY-SOURCE ANALYSIS OF THE EFFECTIVENESS
AND COSTS OF A TAX ON SULFUR EMISSIONS
4.1 Introduction
This chapter provides an analysis of the projected effectiveness and
costs of a tax on sulfur emissions for each of the major sources examined
in this study—steam-electric utilities, area sources, petroleum refineries,
sulfuric acid plants, and primary nonferrous smelters. A discussion of
the indirect impact of the tax on fuel demand and prices is also included.
Many of the control techniques for reducing sulfur emissions from
the sources being examined in this study result in the recovery of sulfur
or sulfuric acid.* If the market for sulfur or sulfuric acid proves to be
sufficiently large to absorb the quantities recovered, revenues from such
sales would lower control costs. The lower costs should encourage greater
levels of control for a given tax rate. However, there are several cogent
reasons for concluding that ffuture prices for recovered sulfur and sulfuric
acid will go below their currently depressed levels.t With only limited
prospects of future markets for sulfur, some firms may actually face a
dispoal problem (i.e., a negative price) with respect to their recovered
sulfur. Therefore, for the industrial process sources (petroleum refineries;
sulfuric acid producers; copper, lead, and zinc smelters), the sensitivity
of the projected effectiveness and costs of the tax to changes in values
of recovered sulfur and sulfuric acid is also provided.
Finally, because of the difficulty in accounting for all of the
process and operating configurations that influence control costs, the
sensitivity of the projected effectiveness and costs of the tax to alter-
native control cost estimates is also analyzed.
4.2 Steam-Electric Power Plants
Projections of the response of the nation's steam-electric power
plants to a tax on sulfur emissions have been made on a utility-by-utility
basis using the emissions and control data and the fuels availability data
and price information shown in appendix A. The projected responses of all
*Sulfuric acid is about 33 percent elemental sulfur, by weight.
tSee appendix E for further discussion on this topic.
37
-------
the nation's utilities have been summed to obtain industry totals. The
resulting projections, however, are approximations that do not reflect
all of the control alternatives nor fuel availabilities and transportation
strategy costs among utilities. The omitted hardware strategies were
generally those that are currently considered highly experimental and
that are not currently expected to be available by 1978.
4.2.1 Background
In 1968, combustion of fossil fuels by steam-electric utilities
accounted for an estimated 50.6 percent of the estimated nationwide sulfur
emissions from all sources.*
In 1970, installed steam-electric generating capacity of the 953
plants was 260,272.1 thousand kilowatts; actual generation was 1,220.1
billion kilowatthours.t An additional 269 billion kilowatthours were
produced by hydroelectric nuclear generating stations. Utilities are
either investor owned, publicly owned (non-Federal), federally owned, or
owned by cooperatives. This industry is subject to comprehensive rate
and service regulation on the part of governmental commissions at several
levels of government.
4.2.2 Industry Growth
From 1957 through 1970, steam-electric-generated electricity increased
an average of 7.4 percent annually. However, as shown in figure 10, growth
has varied throughout the period.
Generation of electricity by steam-electric power plants is projected
to increase about 3.6 percent annually between 1970 and 1978.1= Actual
generation is therefore projected to be about 1,631.0 billion kilowatthours.
4.2.3 Effectiveness
The response of the nation's steam-electric utilities to a tax on
their emissions of sulfur has been projected for selected tax rates.
*National Air Pollution Control Administration. Natipnwide Inve n tory
of Air Pollutant Emissions,—1968, Raleigh, N. C., August 1970.
tNational Coal Association, Division of Economics and Statistics,
Steam-Electric Plant Factors—1969, Washington, D. C., 1969, p. 118.
^National Economic Research Associates, Inc., Fuels for the Electric
Utility Industry, 1971-1985, Prepared for the Edison Electric Institute,
New York, N. Y., August 19>2.
38
-------
300
1300
INSTALLED CAPACITY
NET GENERATION
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71
Figure 10. Steam-electric power plant trends
(Source: National Coal Association).
One considered control alternative used, magnesia base scrubbing, produces
sulfur which can be sold to offset some control costs. It has been
assumed that the recovered sulfur can be sold at $10 per ton, although,
as discussed in appendix E, there is considerable uncertainty regarding
the 1978 market value of sulfur. However, a brief analysis has confirmed
that this value does not significantly alter the projected effectiveness
and costs of a tax in the case of the steam-electric utilities.
Since sulfur emissions from steam-electric utilities are, in the
absence of flue gas desulfurization, a function of the amount and sulfur
content of the fuels consumed, projections of fuel demands, supplies, and
prices are needed to project fuel utilization and emissions.
As discussed in appendix A, projecting long-run fuel supplies and
price requires a knowledge of many technical, economic, and even political
factors. This research has provided only an initial examination of how
these factors may interrelate in 1978 with projected fuel demands to
determine future fuels usage and the resulting emissions. Using fuels
• : *,•
supply data as presented in appendix A for each major supplying region
and for several fuel sulfur contents, these supplies have been allocated
to utilities by establishing prices consistent with projected supplies.
The future supplies of coal and domestic residual oil were assumed to grow
39
-------
at the maximum rates hypothesized by MITRE.* The supply of imported oil
was assumed to be perfectly elastic at prices given in a recent study.t
Gas was assumed to be available only to current users.
Within the model employed for this study, the demand for coal-supplied
Btu's can be met by purchasing any one of 9 sulfur content coals from 19
producing regions. Residual oil demand can be met by purchasing any one
of 5 sulfur contents from 12 origins, including imports. For gas demand,
1 sulfur content and 12 origins have been used. Prices at the mine or
wellhead plus transportation costs equal the delivered prices at the utility.
An estimate of minimum mine and wellhead prices from a previous study for EPAt
provided initial selling prices for domestically produced fuels. In cases
where these prices caused fuel demand to exceed supply, the prices were
increased until fuel demand was less than or approximately equal to supply
on a national basis. It is recognized that this procedure does not incor-
porate all of the sophistication desirable; yet within the more limited
scope of this study, it does provide a reasonable basis for developing a
structure of relative fuel prices by sulfur content and location.
Sulfur emissions from all steam-electric plants, assuming no additional
controls other than those required by the New Source Performance Standards
(a zero tax) are projected to be about 11 million tons in 1978 (see table 7).
Coal burning accounts for 77 percent of these emissions. The remaining 23
percent is accounted for by residual oil combustion. The contribution by
gas is negligible.
Utilities can control sulfur emissions by two general approaches—fuel
switching^ or installation of control equipment. As discussed in appendix A,
the only fuel switching permitted is from coal to oil. Three control systems
(dry limestone, wet limestone, and magnesia scrubbing) are assumed available.
The utilities are assumed to minimize total costs, defined as the sum of
control costs, delivered fuel premiums, and tax payments. This assumption
*MITRE Corporation, Survey of Coal Availabilities by Sulfur Content,
May 1972.
tBattelle Memorial Institute, EPA Energy Quality Model. September 1972.
this study, "fuel switching" means substitution of a fuel of
a different sulfur content from that projected in the absence of a tax.
It may be either the same type of fuel as purchased without a tax or an
alternative fuel.
40
-------
Table 7. Projected sulfur emissions from
steam-electric utilities—1978*
Annual sulfur
Source emissions
(thousand tons of sulfur)
Coal combustion 8,781
Oil combustion 2,615
Gas combustion 0
Total 11,396
*Assuming only controls required by New
Source Performance Standards.
Source: Research Triangle Institute.
follows the conclusions of a recent EPA-sponsored study which concluded
that, for a number of reasons, utilities can be expected to follow cost-
minimizing behavior.*
The projected response of all steam-electric utilities to the sulfur
tax for selected tax rates are provided in table 8 and in figure 11.
Control costs and fuel prices by sulfur content and origin as used for
this analysis are presented in appendix A. The results displayed here
indicate some important results. Although reductions in emissions are
induced by higher tax rates throughout the range of tax rates considered,
only comparatively small additional reductions are induced at rates above
15 cents per pound. This exponential relationship between emissions tax
rates and emissions reductions implies, for example, that a tax of 15 cents
would induce emissions reductions 46 percent greater than those at a
5-cent tax rate. Doubling the tax to 30 cents would precipitate only a
9-percent increase in reductions over those that would occur at a 15-cent
tax rate.
Institute-of Public Administration, Governmental Approaches to Air
Pollution Control: A Compendium and Annotated Bibliography, Submitted
to Office of Air Programs, Environmental Protection Agency, July 15, 1971
(NTIS: PB-203 111).
41
-------
Table 8. Projected response of all steam-electric power plants
national tax on sulfur emissions--1978
(recovered sulfur valued at $10 per ton)
Reductions Total Annualized
Emissions in emissions annualized control Annual tax
(thousand from zero tax cost cost payment
tons) (thousand tons) (thousands) (thousands) (thousands)
Emissions source
Coal combustion
Oil combustion
Gas combustion
Total from all sources
Coal combustion
Oil combustion
Gas combustion
Total from all sources
Coal combustion
Oil combustion
Gas combustion
Total from all sources
Coal combustion
Oil combustion
Gas combustion
Total from all sources
Coal combustion
Oil combustion
Gas combustion
Total from all sources
Coal combustion
Oil combustion
Gas combustion
Total from all sources
Tax
4,731.4
427.4
0.2
5,159.0
Tax
2,675.8
285.8
0.2
2,961.8
Tax
2,055.6
221.9
0.2
2,277.7
Tax
1,746.3
201.7
0.2
1,948.2
Tax
1,413.1
186.2
0.2
1,599.5
Tax
1,253.6
178.3
0.2
1,432.1
rate: 5 cents per pound of sulfur emissions
5,478.7 $1,056,152 $ 582,990 $473,163
758.3 115,635 72,891 42,744
0.0 21 0 21
6,237.0 $1,171,808 $ 655,881 $515,928
rate: 10 cents per pound of sulfur emissions
7,534.4 $1,595,971 $1,060,758 $535,214
899.8 178,648 121,472 57,176
0.0 42 1 42
8,434.2 $1,774,661 $1,182,231 $592,432
rate: 15 cents per pound of sulfur emissions
8,154.6 $1,986,655 $1,369,909 $616,749
963.6 228,350 161,747 66,604
0.0 64 0 64
9,118.2 $2,215,069 $1,531,656 $683,417
rate: 20 cents per pound of sulfur emissions
8,464.0 $2,316,748 $1,618,123 $698,625
983.7 272,855 192,172 80,684
0.0 85 0 85
9,447.7 $2,589,688 $1,810,295 $779,394
rate: 25 cents per pound of sulfur emissions
8,797.3 $2,596,415 $1,889,803 $706,614
999.2 315,361 222,245 93,116
0.0 106 0 106
9,796.5 $2,911,882 $2,112,048 $799,836
rate: 30 cents per pound of sulfur emissions
8,956.8 $2,830,658 $2,078,415 $752,244
1,007.1 355,590 248,577 107,013
0.0 127 0 127
9,963.9 $3,186,375 $2,326,992 $859,384
Source: Research Triangle Institute,
42
-------
EMISSIONS
REDUCTIONS
(RIGHT SCALE)
EMISSIONS
(LEFT SCALE)
10 15 20
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
25
30
Figure 11. Effectiveness of a tax on sulfur emissions: steam-electric
power plants—1978 (Source: Research Triangle Institute).
Table 9 shows the projected percentage distribution of power
producing unit's fuel and flue gas cleaning hardware choices in response
to increments in the sulfur emissions tax. As the tax is increased, the
percentage of producing units that use coal steadily declines (from more
than 40 percent with a 5-cent tax to about 23 percent with a 30-cent tax).
Although relatively small increases in control hardware are induced among
coal-burning plants, large increments are projected for the dry limestone
scrubbing systems among oil-burning plants.
Table 10 provides data for the projected distribution of fuel demands
by sulfur contents. The table shows the heating values (trillion Btu's)
that are projected to derive from domestic and imported residual oil and
from domestically produced coal. Bearing in mind that the simulation
model presumed unlimited supplies of foreign residual oils at the prices
given in appendix A, the reader will note the substantial shift by power
plants away from coal to residual oil in the presence of increasing emissions
tax rates. For example, at a zero tax rate, coal would constitute more
than 76 percent of the total coat- and oil-heat input to power plant boilers;
at a 15-cent tax rate, about 48 percent; and at a 30-cent tax rate, only
43
-------
Table 9. Effects of the sulfur emissions tax on the
distribution of flue gas desulfurization choices
for steam-electric power producing units*
Tax rate (cents per pound of sulfur emissions)
10
15
20
25
30
Coal combustion
No hardware
Dry limestone
Wet limestone
Magnesia base
35.8
0.9
0.1
3.6
28.1
i:9
0.0
3.7
23.3
2.4
0.0
3.7
21.9
1.8
0.0
4.0
19.2
1.6
0.0
4.2
17.2
1.4
0.0
4.3
Total
Oil combustion
No hardware
Dry limestone
Wet limestone
Magnesia base
Total
40.4 33.9
29.4
59.6 66.3
70.6
27.7
52.1
17.1
0.1
3.0
72.3
25.0
22.9
49.
22.
0.
3.0
75.0
*Generally, boiler(s) with a common size, age, and fuel were treated as the
producing unit. For small plants, data inadequacies required that the entire
plant be treated as the producing unit.
Note: At a zero tax rate, no additional flue gas desulfurization would be
induced.
Source: Research Triangle Institute.
Table 10. Distribution of steam-electric utilities' demand for
coal and residual oil by sulfur content—1978
(trillion Btu)
Projected maximum
domestic supply
(trillion Btu)
Tax rate*
0
5
10
15
20
25
30
Coal
Sulfur content (percent)
0.7 0.9 1.3 1.8 2.3 Z.8 3.3 3.8 5.0
2.4 2.6 1.2 1.8 1.1 1.2 2.6 2.8 1.4
0.0 0.6 0.0 0.2 0.7 1.4 2.6 2.9 1.4
0.6 1.1 1.0 0.1 2.0 0.8 0.1 0.9 1.4
1.6 0.9 1.6 0.1 0.4 0.1 0.1 0.9 1.4
1.7 1.0 1.0 0.1 0.0 0.1 0.1 2.1 JD.1
1.8 0.6 0.4 0.0 0.1 0.2 0.2 2.1 0.1
1.6 0.2 0.0 0.0 0.1 0.3 0.3 2.1 0.0
0.9 0.2 0.0 0.0 0.3 0.8 0.8 1.2 0.0
Totals
17.1
9.8
8.0
7.1
6.2
5.5
4.6
4.2
Residual oil
Sulfur content (percent)
0.4 0.6 1.2 2.3 3.0
1.1 0.7 0.3 0.4 0.1
0.0 0.1 0.1 1.2 1.6
0.1 1.1 2.2 0.7 0.7
0.4 3.5 1.0 0.3 0.6
0.9 4.8 0.0 0.4 0.5
1.3 5.0 0.0 0.4 0.5
3.1 4.1 0.0 0.4 0.5
3.9 3.9 0.0 0.4 0.5
Totals
2.6
3.0
4.8
5.8
6.6
7.2
8.1
8.7
Impliedt
imports
~
0.4
2.2
3.2
4:0
4.6
5.5
6.1
•Cents per pound of sulfur emissions.
tThese are minimum amounts since they do not include the projected residual oil consumption by area sources.
Source: Research Triangle Institute.
-------
33 percent. The demand for imported residual oil is projected in the far
right-hand column of table 10. At intermediate tax rates, these values are
comparable to current imports of residual oil, for all uses, which are
currently on the order of 3.5 trillion Btu's; this implies about a 60-percent
increase in residual oil imports by 1978 if all such imports are diverted
for use in power generation. In section 4.2.5 the assumption of fixed oil
prices is relaxed. The sensitivity analysis performed there indicates the
extreme importance of those prices in determining the distribution of demand.
4.2.4 Costs
The total outlays of the steam-electric utilities were shown in
table 8 for selected tax rates. Figure 12 displays the allocation of the
costs between emissions tax payments and control costs. The latter include
both flue gas desulfurization costs and fuel switching premiums. In
absolute terms, total tax payments rise from about $515 million at a 5-cent
tax rate to about $860 million at a 30-cent tax rate. However, as a
percentage of total emission-control-related costs, the share accounted
tn
K
CD
§
o
u
3
D
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure 12. Total costs induced by a tax on sulfur emissions: steam-electric
power plants—1978 (Source: Research Triangle Institute).
45
-------
for by taxes falls from about 44 percent to about 27 percent at tax rates
of 5 and 30 cents per pound, respectively. Considering that a 30-cent tax
is projected to induce about a 90-percent reduction in emissions (fig. 11)
from power plants, this result implies that a policy goal of sulfur
emissions reductions of that magnitude can be achieved at only a 27-percent
penalty above minimum control costs to the average firm. To society as a
whole, there would be no penalty at all since the taxes are merely
transfers. Consequently, to the extent that regulation would achieve
90-percent reductions less efficiently than a tax (it cannot be more
*'
efficient), all of the additional costs (of achieving that goal) above
an estimated $2.3 billion can be considered deadweight losses.
Table 11 gives the distribution of control costs between those for
stack gas cleaning and those for fuel switching as the tax rate varies.
For a zero tax rate, no additional control is induced, of course. At low
tax rates, the distribution is fairly equal; at high tax rates the costs
of flue gas cleaning become the larger share since many more utilities
are induced to use dry limestone and magnesia base scrubbing systems. Fuel
switching costs include net changes not only in the f.o.b. cost of low
sulfur fuels but also in transportation expenses. At positive tax rates,
the transportation costs are actually projected to be somewhat lower than
those with a zero tax since oil (to which utilities are induced to shift,
Table 11. Percentage distribution of control costs
Tax rate*
5
10
15
20
25
30
Flue gas
desulfurization
50.7
47.4
51.3
57.8
58.9
62.0
Fuel
switchingt
49.3
52.6
48.7
42.2
41.1
38.0
Total
100.0
100.0
100.0
100.0
100.0
100.0
*Cents per pound of sulfur emissions.
tThese costs are delivered fuel costs at the utility.
Source: Research Triangle Institute.
46
-------
with the tax and oil prices used) has lower transportation costs per Btu
than coal, for most utilities. It is likely, however, that the fuel
switching costs are underestimated at high tax rates, for if the supply
of imported oil is not assumed to be perfectly elastic, competitive
bidding for limited supplies will increase the price of oil. This possi-
bility is analyzed further in section 4.2.5 below.
Figure 13 gives the average cost increases per kilowatthour, as
the tax rate is increased. As a percentage of current average electricity
costs.(1.6 cents/kWh), a 5-cent tax would induce a 4.3-percent increase
in average costs; a 15-cent tax, an 8.7-percent increase; and a 30-cent
tax, about a 12-percent increase. These values are based on the assumption
that the demand for electricity is perfectly inelastic and do not include
the effects of the corporate income tax.
4.2.5 Sensitivity Analysis
The projected effectiveness and costs of a tax on the sulfur emissions
of steam-electric utilities are expected to be strongly influenced by fuel
availabilities and prices. As discussed above, using projections of the
fuel prices without a tax, and then applying a tax, many utilities are
u
Q.
UJZ
2.0
1.8
1.6
1.4
1.2
1.0
08
-I Q6
III
5
= OA
Q2
10
20
25
15
TAX RATE .
(CENTS PER POUND OF SULFUR EMISSIONS)
30
Figure 13. Average total (control cost plus tax payments) incremental costs
per kilowatthour induced by a tax on sulfur emissions: steam-electric power
plants—1978 (Source: Research Triangle Institute).
47
-------
induced to switch from coal to oil to meet their fuel demands at minimum
cost. However, such switching will place pressure not only on domestic
oil supplies but also on imports. (See appendix A for preliminary estimates
of these quantities.) For example, at a tax of only 5 cents, all of the
projected supplies of domestic residual oil would be purchased by steam-
electric utilities. However, these supplies would provide only two-thirds
of demand; the remainder would have to be imported. At higher tax rates,
higher imports of oil are required. Furthermore, this assumes that all
residual oils are allocated to steam-electric utilities. Industrial and
commercial sources are also consumers of residual oil. In 1978, their
coal and residual oil requirements are expected to total about 4.1 trillion
Btu's. The distribution of their consumption between the two fuels will
also depend on relative fuel prices.
All oil prices were increased 20 and'then 40 percent to examine the
sensitivity of both emissions reductions and total control costs and tax
payments to these higher oil prices. The results, shown in table 12,
indicate that the oil price increases reduce the projected effectiveness
of the tax by the largest percentage increase in emissions at a tax rate
of 20 cents; i.e., emissions are about 20 and 29 percent higher when oil
prices rise by 20 and 40 percent, respectively. To either side of this
rate, the impacts are less (measured as a percentage of emissions) but,
nevertheless, still significant. Total emission-control-related costs
increase most at low tax rates with the percentage increases tapering
off slightly as the tax rate increases.
Besides inducing a general shift, toward coal, the increases in oil
prices particularly encourage low-sulfur coal consumption. Table 13
Table 12. Sensitivity of effectiveness and total cost of tax on sulfur
emissions to residual fuel oil prices: steam-electric power plants—1978
Change
In all
oil
prices
(percent)
+20
+40
Change In emissions (percent)
Tax rate
(cents per pound of sulfur emissions)
5 10 15 20 25 30
12.2 15.0 18.5 20.4 16.8 7.9
15.8 2Q.7 26.5 29.3 20.5 14.4
Change 1n total cost (percent) -
Tax rate
(cents per pound of sulfur emissions)
.5 10 15 20 25 30
19.7 18.6 17.8 17.3 17.1 16.6
29.3 26.0 24.7 24.1 23.4 22.8
Change In annual Ized control cost (percent)
Tax rate
cents per pound of sulfur emissions)
5 10 15 20 25 30
25.6 20.4 17.4 15.9 17.2 19.8
40.0 28.9 24.0 21.9 24.4 25.9
•Control costs plus tax payments.
Source: Research Triangle Institute.
48
-------
Table 13. Distribution of steam-electric utilities' demand for coal and residual
oil by sulfur content with increased prices for residual oil—1978
(trillion Btu)
Projected maximum
domestic supply
(trillion Btu)
Tax rate*
0
5
10
15
20
25
30
0
5
10
15
20
25
30
Coal
Sulfur content (percent)
0.7 0.9 1.3 1.8 2.3 2.8 3.3 3.8 5.0
2.4 2.6 1.2 1.8 1.1 1.2 2.6 2.8 1.4
Increase 1n all oil prices
0.0 0.6 0.0 0.2 0.8 1.5 2.2 3.3 2.3
0.7 1.3 1.1 1.2 2.9 0.8 0.1 0.4 2.2
2.2 1.5 3.2 0.3 0.4 0.1 0.0 0.5 2.1
2.3 2.4 1.7 0.1 0.0 0.1 0.0 1.8 0.8
3.4 1.4 1.2 0.0 0.1 0.3 0.1 2.4 0.1
3.9 1.2 0.0 0.0 1.6 0.4 0.1 2.5 0.0
3.6 0.6 0.1 0,0 2.0 0.8 0.4 1.7 0.0
Totals
17.1
Residual oil
Sulfur content (percent)
0.4 0.6 1.2 2.3 3.0
1.1 0.7 0.3 0.4 0.1
of 20 percent
10.9
10.7
10.3
9.2
9.0
9.7
9.2
0.0 0.0 0.0 0.8 1.2
0.0 0.7 0.3 0.3 0.8
0.2 1.2 0.3 0.4 0.3
0.4 2.4 0.1 0.4 0.3
0.6 2.5 0.0 0.4 0.3
0.6 1.7 0.0 0.5 0.2
1.0 1.8 0.0 0.5 0.3
Increase in all oil prices of 40 percent
0.0 0.6 0.0 0.2 0.8 1.5 2.2 3.5 2.3
0.7 1.3 1.2 1.5 3.0 0.8 0.1 0.4 2.2
2.5 1.5 3.4 0.6 0.4 0.1 0.0 0.5 2.1
3.9 2.4 1.9 0.2 0.0 0.1 0.0 1.8 0.8
5.2 1.4 1.4 0.0 0.1 0.3 0.1 2.5 0.1
4.9 1.3 0.0 0.0 1.7 0.5 0.1 2.6 0.0
5.0 0.6 0.1 0.0 2.2 0.8 0.4 1.7 0.0
11.1
11.2
11.1
11.1
11.1
11.1
10.8
0.0 0.0 0.0 0.9 0.7
0.0 0.4 0.0 0.5 0.7
0.0 0.7 0.1 0.6 0.3
0.1 0.7 0.1 0.4 0.2
0.2 0.8 0.0 0.5 0.1
0.4 0.8 0.0 0.6 0.1
0.4 0.9 0.0 0.6 0.1
Total s
2.6
2.0
2.1
2.4
3.6
3.8
3.0
3.6
1.6
1.6
1.7
1.5
1.6
1.8
2.0
Impliedt
imports
-.-
0.0
0.0
o.o •
1.0
1.2
0.4
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
*Cents per pound of sulfur emissions.
tThese are minimum amounts since they do not Include the projected residual oil consumption by area sources.
Source: Research Triangle Institute.
-------
displays those projected shifts in the presence of 20 and 40 percent
oil price increases. Table 14 is analogous to table 9 in that it shows
the projected percentage distribution of fuel and flue gas desulfurization
choices for power producing units. Compared to the producers' fuel choices
under the benchmark prices, the number of coal consuming units is markedly
higher, particularly at higher tax rates. Furthermore, the percentages of
units using control hardware rise substantially. The most notable increases
occur in the addition of dry limestone scrubbers to coal burning boilers.
4.3 Industrial, Commercial, and Residential Space Heating
Projections of the consumption of fossil fuels in industrial, commercial,
and residential space heating applications have been made on a State-by-State
basis using the emissions and control cost data shown in appendix A. These
sources are generally referred to as "area sources" because of the geographic-
ally diffused nature of their emissions and because of the difficulty of
identifying individual emissions sources. Because of these limited data and
of the diversity of size among these industrial, commercial, and residential
space heating sources, this analysis provides only a very preliminary exam-
ination of the effectiveness and costs of a tax on the sulfur emissions from
these area sources. Further, for the same reasons, no sensitivity analysis
of the response of the area sources to higher residual oil prices has been
included. However, as compared to steam-electric abilities, area sources
have fewer control options and, therefore, may reasonably be expected to
show larger changes in emissions and total costs than those projected for
utilities.
4.3.1 Background
In 1968, industrial, commercial, and residential space heating accounted
for 22.8 percent of the estimated nationwide sulfur emissions from all
sources.* Because the consumption of residual oil and coal (bearing rela-
tively more sulfur than distillate oil and gas) is concentrated in the
industrial and commercial space heating-sectors, the predominant share of
those emissions can reasonably be attributed to those sources. Table 15
shows the distribution of fuel consumption patterns in 1970, by sources;
the reported values confirm the asserted importance of nonresidential sources
in the consumption of relatively higher sulfur bearing fuels.
*National Air Pollution Control Administration, Nationwide Inventory
of Air Pollutant Emissions—1968, Raleigh, N. C., August 1970.
50
-------
Table 14. Effects of residual oil price increases on the percentage
distributions of flue gas desulfurization choices for
steam-electric power producting units*
Tax rate (cents per pound of sulfur emissions
5 10 15 20 25 30
Coal combustion
No hardware
Dry limestone
Wet limestone
Magnesia base
Total
Oil combustion
No hardware
Dry limestone
Wet limestone
Magnesia base
Total
Coal combustion
No hardware
Dry limestone
Wet limestone
Magnesia base
Total
Oil combustion
No hardware
Dry limestone
Wet limestone
Magnesia base
Total
46.3
1.1
0.1
4.1
Increase in all oil prices of 20 percentt
42.3
2.7
0.0
4.2
51.6 49.2
37.4
4.7
0.0
4.2
46.3
43.1
7.9
0.1
2.6
45.2
32.6
5.9
0.0
5.1
43.6
41.8
35.8
19.8
0.0
2.6
48.5 50.8 53.7 54.8 56.4 58.2
Increase in all oil prices of 40 percentt
48.7
1.2
0.1
4.2
54.2 54.2
53.9
45.8
53.6
33.0
53.1
36.7
9.5
0.1
6.1
52.4
46.4
47.0
47.6
*Generally, boiler(s) with a common size, age, and fuel were treated as the
producing unit. For small plants, data inadequacies required that the entire
plant be treated as the producing unit.
tOver the values presented in appendix A.
Source: Research Triangle Institute.
51
-------
Table 15. Area source fuel consumption--!970
Source
Residential
Industrial
Commercial
Total
Source:
Coal
(thousand
tons)
—
210,552
13,557
224,109
Research Triai
Residual
oil
(thousand
barrel s )
—
44,190
73,650
117,840
igle Institute.
Distillate
oil
(thousand
barrels)
447,691
83,490
113,812
644,993
Gas
(million
cubic
feet)
3,206,100
2,664,100
1,035,900
6,906,100
4.3.2 Industry Growth
The growth in the consumption of fossil fuels by these space heating
units is expected to follow broad demographic and economic trends. For
residential sources, growth was projected to follow the growth in population
projected for each State as projected by the Department of Commerce.* For
commercial and industrial sources, growth was projected to follow the overall
national growth in employment as projected by the Bureau of Labor Statistics.t
4.3.3 Effectiveness
The input demand response of space heating sources to a tax on their
sulfur emissions was projected for several tax rates. As discussed in
appendix A, the only control alternative permitted is switching to lower
sulfur coals or residual oils by the industrial and commercial sources;
residential sources are assumed to remain committed to the fuels presently
used; viz., distillate oil and natural gas.
It is likely that flue gas cleaning could be employed by some of the
larger industrial and commercial sources; however, the lack of data on these
sources precluded inclusion of this alternative in this study. If flue gas
cleaning is likely, then the effectiveness of the tax would be greater and
the total cost less than projected.
Sulfur emissions from area sources, assuming no controls, are projected
to be about 5.7 million tons in 1978 (see table 16). The bulk of emissions
would come from industrial sources, which would account for 78 percent of
*Survey of Current Business, Volume 52, Number 4, April 1972.
til. S. Department of Labor, Bureau of Labor Statistics, Patterns of
U. S. Economic.Growth, BLS Bulletin 1672, Washington, D. C.: U.S. Govern-
ment Printing Office, 1970.
52
-------
Table 16. Projected sulfur emissions
from area sources--1978*
Annual sulfur emissions
Source (thousand tons of sulfur)
Residential 154
Industrial 4,490
Commercial 1,115
Total 5,759
*Assuming no controls.
Source: Research Triangle Institute.
the total, and commercial sources, projected to generate 20 percent of
total emissions.
The projected responses of all area sources to the sulfur tax for
selected tax rates are provided in table 17. Reductions in emissions
from area sources are projected to increase over the entire range of tax
rates considered here. However, the rate of increase falls precipitously
above a tax rate of 10 cents per pound of sulfur emissions. For example,
a tax rate of 10 cents per pound would induce emissions reductions that
are 48 percent greater than those at a 5-cent tax rate. Tripling the
tax rate to 30 cents per pound would encourage only an additional 11 percent
in emissions reductions compared to those that are projected at a 10-cent
tax rate. That sharply exponential response to changes in the tax rate
is shown graphically in figure 14.
Table 18 summarizes the demand for coal and residual oil by the
fossil fuel combustion sources considered in this report (steam-electric
utilities and area sources). The reader is reminded again that these data
are*of a preliminary nature and based on assumptions as described above and
in appendix A. The data in table 18 do show, however, rough magnitudes of
the demand responses for several tax rates at the assumed fuel prices. At
these prices, higher tax rates (which are inducing substantial emissions
reductions) are placing significant pressures on oil supplies, especially
imports. If these supplies are not forthcoming at the assumed prices due
to an upward sloping supply curve for imported residual oil and/or curtail-
ment of shipments from a major source for political or other purposes, the
53
-------
Table 17. Projected response of all area sources to a
national tax on sulfur emissions—1978
Emissions source
Reductions Total Annualized
Emissions in emissions annual control Annual tax
(thousand from zero tax cost cost paymentT
tons) (thousand tons)(thousands) (thousands) (thousands)
Residential
Industrial
Commercial
Total from all
Residential
Industrial
Commercial
Total from all
Residential
Industrial
Commercial
Total from all
Residential
Industrial
Commercial
Total from all
Residential
Industrial
Commercial
Total from all
Tax rate:
153.5
2,203.0
498.2
sources 2,854.7
Tax rate:
153.5
1,063.9
276.9
sources 1,494.3
Tax rate:
153.5
959.5
268.1
sources 1,381.1
Tax rate:
153.5
956.4
267.3
sources 1,377.2
Tax rate:
153.5
747.8
213.9
sources 1,115.2
5 cents per pound of sulfur emissions
0.0 $15,350 $ 0
2,206.3 358,925 138,620
616.5 92,222 42,406
2,822.8 $466,497 $181,026
10 cents per pound of sulfur emissions
0.0 $ 30,701 $ 0
3,345.4 499,202 286,416
837.8 124,362 68,975
4,183.2 $654,265 $355,391
15 cents per pound of sulfur emissions
0.0 $ 46,051 $ 0
3,449.9 598,975 311,142
846.6 151,621 71,202
4,296.5 $796,647 $382,344
20 cents per pound of sulfur emissions
0.0 $ 61,401 $ 0
3,452.9 694,752 312,177
847.4 178,373 71,469
4,300.3 $934,526 $383,646
25 cents per pound Of sulfur emissions
0.0 $ 76,751 $ 0
3,661.5 780,574 406,640
900.7 202,568 95,574
4,562.2 $1,059,893 $502,214
$ 15,350
220,306
49,816
$285,472
$ 30,701
212,786
55,387
$298,874
$ 46,051
287,835
80,420
$414,306
$ 61,401
382,576
106,904
$550,881
$ 76,751
373,934
106,995
$557,680
Tax rate: 30 cents per pound of sulfur emissions
Residential
Industrial
Commercial
Total from all sources
153.5
685.1
192.3
1,030.9
0.0 $ 92,101
S.724..2 850,254
922.4 222,351
4,646.6 $1,164,706
$ 0
439,166
106,987
$546,153
$ 92,101
411,088
115,365
$618,554
Source: Research Triangle Institute.
54
-------
O DC
CO Ul
(0 O-
si 2
x
EMISSIONS
REDUCTIONS
(RIGHT SCALE)
EMISSIONS
(LEFT SCALE)
I
100
80
60
l-
I
oc
O
UJ
QC
CO
40 <2
CO
CO
I
UJ
20
10
20
25
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
30
Figure 14. Effectiveness of a tax on sulfur emissions: area sources—1978
(Source: Research Triangle Institute).
Table 18. Distribution of the combined demand of steam-electric
utilities and area sources for coal and residual oil—1978
Tax ,
rate
0
5
10
15
20
25
30
Utilities
1.4
1.1
0.7
0.2
0.2
0.2
0.2
Coal demand
Area
sources
9.8
8.0
7.1
6.2
5.5
4.6
4.2
(t
Total s
11.2
9.1
7.8
6.8
5.7
4.8
4.4
rill ion Btu)
Residual oil
Utilities
2.7
3.0
3.4
3.9
3.9
3.9
3.9
Area
sources
3.0
4.8
5.8
6.6
7.2
8.1
8.7
demand
Total s
5.7
7.8
9.2
10.5
11.1
12.0
12.6
Implied
imports
3.1
5.2
6.6
7.9
8.5
9.4
10.0
Source: Research Triangle Institute.
55
-------
effectiveness of the tax would be reduced and the cost increased over that
projected here. It should be noted, however, that under regulatory approaches
to achieving emissions reductions, cost increases would also occur.
4.3.4 Costs
The total cost by the area sources are shown in table 17 and figure 15.
The broken curve in figure 15 incorporates both tax payments and fuel
switching premiums. The costs for residential sources increase linearly
with higher tax rates since no fuel switching or emissions control options
for those sources are considered in the simulation model. However, the
total cost function for all area sources combined increases at a decreasing
rate since emissions control among industrial and commercial sources is
induced by the tax.
4.4 Petroleum Refineries
Projections of the response of the nation's petroleum refineries to
a tax on sulfur emissions have been made on a pi ant-by-plant basis, using
the emissions and control data shown in appendix B; these projections are
summed to obtain industry totals. However, the resulting projections
I2OO
_ IOOO
800
? 600
o
bf
400
200
) 10 15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure 15. Total cost induced by a tax on sulfur emissions: area sources--1978
(Source: Research Triangle Institute).
56
-------
are approximations that do not reflect all of the variations in process
types and operating procedures among refineries. In addition, emissions
may be understated, since the available data do not make it possible to
account for the disposition of all of the sulfur contained in the crude
oil processed.
4.4.1 Background
In 1968, petroleum refineries accounted for an estimated 6.3 percent
of the estimated nationwide sulfur emissions from all sources.* Petroleum
products are widely used in other industries, especially as fuel for the
generation of electricity, as feedstock for the petrochemical industry,
and as the basic material for asphalt roofing and paving. Therefore, their
supply and price behavior are important to the economy beyond the influence
of their better known, direct uses as vehicle fuels and heating oils.
There were 263 petroleum refineries in the United States in 1970. The
bulk of refining capacity is concentrated in 30 to 35 firms. Of these, 16
are fully integrated international corporations making up the so-called
large majors of the industry; another 8 firms may be classified as small
majors and are also fully integrated. The remainder of the firms in the
industry are somewhat smaller; they either are not fully integrated or
operate in a limited market. In 1970, petroleum refining capacity was about
12.154 million barrels daily,t and annual production, as measured by actual
runs-to-stills, was 3,967.5 million barrels (see fig. 16).
4.4.2 Industry Growth
From 1950 through 1970, petroleum refining (measured by runs-to-stills)
increased an average of 3.2 percent annually. However, as shown in figure
16, growth has varied throughout the period. For the period 1970-78, the
annual growth rate assumed for petroleum refining is 4 percent.f This
implies domestic refining of approximately 5,268.8 million barrels in 1978.
*National Air Pollution Control Administration, Nationwide Inventory
of Air Pollutant Emissions—1968, Raleigh, N. C., August 1970.
tPetroleum Refinery Listing, Research Triangle Institute.
^Research Triangle Institute, Unpublished Data for the Cost of Clean
A1r. 1973.
57
-------
1950 1952 1954 1956 1958 I960 I962 (964 I966 I968 I970
YEAR
Figure 16. Petroleum refining trends (Source: Research Triangle Institute)
4.4.3 Effectiveness
The response of the nation's petroleum refining industry to a tax
on their emissions of sulfur has been projected for several tax rates
ranging from 0 to 30 cents per pound of sulfur. Since all of the emissions
control alternatives available to refineries and costed for this study
result in the recovery of sulfur, it is necessary to credit revenues from
its sale. However, because of the uncertainties surrounding the future
price of recovered sulfur, the impact of the assumed market price of
recovered sulfur on the decisions by refineries to control emissions is
analyzed separately in section 4.4.5. In this section, a sulfur price of
$10 per ton was used as the most likely 1978 market value for recovered
sulfur from petroleum refineries (see appendix E for an extended discussion
of this choice).
Sulfur emissions from all petroleum refineries, assuming no additional
controls other than those required by New Source Performance Standards (a
zero tax), are projected to be about 772,000 tons in 1978 (see table 19).
These projected emissions are fairly equally distributed across the three
major process sources (Claus plants, catalytic crackers, and fuel c/il
combustion).
Tabular results of the projected response of all petroleum refineries
to the sulfur tax for several tax rates are provided in table 20. This
58
-------
Table 19. Projected sulfur emissions from petroleum refineries—1978*
Annual sulfur emissions
Source (thousand tons of sulfur)
Claus plants 280
Fluid catalytic crackers 256
Fuel combustion processes 221
Thermofor and Houdriflow
catalytic cracker 15
Total 772
*Assuming only controls required by New Source Performance Standards.
Source: Research Triangle Institute.
information, plus the effect of several additional tax rates, is graphically
presented in figure 17. Throughout the range of taxes being considered, the
only economically feasible control option was the extended treatment of the
hydrogen sulfide gas stream via Claus plants or via additions to present
Claus plants. No control is induced for emissions from refinery combustion
processes or from catalytic crackers because of the high costs of the control
alternatives. A large refinery, for example, would find it uneconomical to
control emissions from combustion processes until a tax of about 33 cents is
introduced. Catalytic cracker emissions would not be controlled until a
tax. of about $1.60 is introduced. As a result, only 36 percent of the pro-
jected emissions would be controlled even with a 30-cent tax. However, the
tax payments on emissions from these two sources might encourage the develop-
ment of other, more cost-effective control techniques than those currently
available.
At a tax rate as low as 1 cent, the average large refinery without a
Claus plant would find it more economical to install such a plant rather
than to pay the maximum amount of tax. At a 2- or 3-cent tax rate, large
refineries with two-stage Claus plants are projected to upgrade their plants
to four-stage units.
Reductions in emissions are induced by increasing the tax rate to
about 20 cents per pound. Tax rates between 20 and 30 cents per pound,
however, are projected to induce only small additional reductions in
emissions.
59
-------
Table 20. Projected response of all petroleum refineries
to a national tax on sulfur emissions—1978
(recovered sulfur valued at $10 per ton)
Emissions
Emissions source (thousand
tons)
Catalyst regenerators
Fluid catalytic crackers
Thermofor and Houdriflow
catalytic crackers
Claus plants
Fuel combustion
Total from all sources
Catalyst regenerators
Fluid catalytic crackers
Thermofor and Houdriflow
catalytic crackers
Claus plants
Fuel combustion
Total from all sources
Catalyst regenerators
Fluid catalytic crackers
Thermofor and Houdriflow
catalytic crackers
Claus plants
Fuel combustion
Total from all sources
Catalyst regenerators
Fluid catalytic crackers
Thermofor and Houdriflow
catalytic crackers
Claus plants
Fuel combustion
Total from all sources
Catalyst regenerators
Fluid catalytic crackers
Thermofor and Houdriflow
catalytic crackers
Claus plants
Fuel combustion
Total from all sources
Catalyst regenerators
. Fluid catalytic crackers
Thermofor and Houdriflow
catalytic crackers
Claus plants
Fuel combustion
Total from all sources
270.6
255.5
15.1
116.6
220.9
608.1
270.5
255.5
15.0
46.2
220.8
537.5
270.5
255.5
15.0
18.3
220.8
509.6
270.5
255.5
15.0
8.3
220.8
499.6
270.5
255.5
15.0
5.9
220.8
497.2
270.6
255.5
15.1
4.9
220.9
496.4
Reductions Total Annual 1 zed •_.„,. i +,„
1n emissions annual control ™™' *
from zero tax cost cost i?S«.«L«i
(thousand tons) (thousands) (thousands) (thousands)
Tax rate:
0.0
0.0
0.0
163.7
0.0
163.7
Tax rate:
0.0
0.0
0.0
234.0
0.0
234.0
Tax rate:
0.0
0.0
0.0
261.9
0.0
261.9
Tax rate:
0.0
0.0
0.0
271.9
0.0
271.9
Tax rate:
0.0
0.0
0.0
274.4
0.0
274.4
Tax rate:
0.0
0.0
0.0
275.3
0.0
275.3
5 cents per pound of sulfur emissions
$ 27,061 $ 0 $ 27,061
25,554 0 25,554
1 ,507 0 1 ,507
17,254 5,596 11,659
22,088 0 22,088
$ 66,403 $5,596 $ 60,808
10 cents per pound of sulfur emissions
$ 54,123 $ 0 $ 54,123
51,108 0 51,108
3,015 0 3,015
23,977 14,730 9,247
44,176 0 44,176
$122,276 $14,730 $107,546
15 cents per pound of sulfur emissions
$ 81,183 $ 0 $ 81,183
76,661 0 76,661
4,522 0 4,522
27,184 21,670 5,514
66,264 0 66,264
$174,631 $21,670 $152,961
20 cents per pound of sulfur emissions
$108,244 $ 0 $108,244
102,214 0 102,214
6,030 0 6,030
28,413 25,056 3,357 ;
88,351 0 88,351
$225,008 $25,056 $199,952
25 cents per pound of sulfur emissions
$135,304 $ 0 $135,304
127,768 0 127,768
7,536 0 7,536
29,114 26,161 2,954
110,438 0 110,438
$274,856 $26,161 $248,696
30 cents per pound of sulfur emissions
$162,366 0 $162,366
153,322 0 153,322
9,044 $ 0 9,044
29,640 26,681 2,959
132,525 0 132.525
$324,519 $26,681 $297,837
Source: Research Triangle Institute.
60
-------
oe
u
oe
ui
Q.
CO
o
800
7SO
700
650
600
<2 550
CO
12 500
ui
tt
450
co 400
..-4--.
EMISSIONS REDUCTIONS
(RIGHT SCALE)
EMISSIONS
(LEFT SCALE)
15 20
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure 17. Effectiveness of a tax on sulfur emissions: petroleum
refining—1978 (Source: Research Triangle Institute).
4.4.4 Costs
The total cost to the petroleum refining industry is shown in table 20
for selected tax rates. As shown in figure 18, they increase almost
linearly for all tax rates between 0 and 30 cents because the majority of
the emissions (74 percent) would not be controlled over that range. Since
tax rates up to 30 cents are not projected to induce substantial reductions
in sulfur emissions, the major cost element over the entire range of taxes
would be the tax payments. A remaining defense for the tax, however, is
that the continuing tax liability would, very likely, induce the development
of more cost-effective alternatives for controlling emissions from catalyst
regenerators and fuel combustion than those currently available.
The relative magnitude of these costs on a per-unit-of-product basis
assuming a perfectly inelastic demand for petroleum products and not
including the effects of the corporate income tax is shown in figure 19.
For reference, the 1970 average value at the refinery of all refined
petroleum products, exclusive of excise taxes, was about $5.25 per barrel.
61
-------
400
•***
o: 350
S 300
-t 250
~ 200
S
O ISO
o
UJ
N 100
50
i i i i I i i i i
TOTAL
COST
TAX
FWTMENTS
CONTROL
COST —I
5 10 15 20 25 30
TAX RATE (CENTS PER POUND OF SULFUR EMISSIONS)
Figure 18. Total costs induced by a tax on sulfur emissions: petroleum
refining—1978 (Source: Research Triangle Institute).
.-» .070
0 5 10 15 20 25 30
TAX RATE (CENTS PER POUND OF SULFUR EMISSIONS)
Figure 19. Average total* incremental costs per barrel of refined oil T
induced by a tax on sulfur emissions: petroleum refining—1978 (*control
cost plus tax payments) (Source: Research Triangle Institute).
62
-------
At a tax rate as high as 30 cents, the sum of control cost and tax payments
distributed equally across all refined oil would represent only about one
percent of that value.
4.4.5 Sensitivity Analysis
The projected effectiveness and costs of a tax on the sulfur emissions
of petroleum refineries is influenced not only by the tax rate, but also
by the assumed market value of the recovered sulfur and the estimated con-
trol costs of the sulfur emissions control alternatives. To examine the
sensitivity of the projected effectiveness and costs of the sulfur tax,
a sensitivity analysis was performed. Parametric variations were intro-
duced in the assumed future value for recovered sulfur and in the cost
estimates presented in appendix B. The resulting percentage deviations
in both emissions and total costs from those which obtain in the presence
of the assumed sulfur price of $10 and of control costs as presented in
appendix B are given for variations in the price of sulfur in figure 20
and for variations in control costs in table 20. The base values refer to
o ui
111 3-°°
o*~ o
5 co o 2.00
Q. Z <
22 o
°6 ¥P 1.00
i£l~
op
gfr" QU.
S Ul Ul O
S 3 £ -i-oo
SI
ffi o
-2.00
UJ
w 8 -3.00
TOTAL*'
COSTS
I I
VALUE OF RECOVERED
SULFUR = £ 0 / TON
VALUE OF RECOVERED.
SULFUR = % 20/ TON
05 10 15 20 25
TAX RATE ( CENTS PER POUND OF SULFUR EMISSIONS)
30
Figure 20. Sensitivity of the effectiveness and total costs of a tax on
sulfur emissions to the value of recovered sulfur: petroleum refining— 1978
(The base values refer to both the total control costs and the emission
levels that obtain when the base sulfur prices is assumed equal to $10 per
ton) (Source: Research Triangle Institute).
63
-------
both the total control costs and the emission levels that obtain when the
base sulfur price is assumed equal to $10 per ton.
As is shown in figure 20, reducing or increasing the market value of
recovered sulfur by $10 per ton has greater influence at low tax rates.
However, the magnitude of those responses is small, never exceeding 2.5
percent. For example, if the value of recovered sulfur were $0 per ton
instead of $10 as used in the analysis above, for a tax rate of 10 cents,
^
emissions would be about 0.5 percent and total costs about 2.0 percent
higher than projected above. Sulfur emissions and total cost projections
under alternative market values of sulfur appear to level off very rapidly
as the tax rate is increased, reflecting the decreasing importance of the
price of recovered sulfur in influencing the behavior of refineries. The
general conclusion of this analysis, then, is that substantial variation
in the market price of recovered sulfur is not likely to affect signifi-
cantly the overall predicted effects of a tax on sulfur emissions from
petroleum refineries as discussed in previous sections.
The sensitivity of the emissions and total cost projections to positive
and negative percentage deviations in the control cost estimates are pre-
sented in table 21. In general, the projected effectiveness and costs
are not significantly influenced by these percentage changes in control
costs. However, there is a significant exception. Control costs estimates
Table 21. Sensitivity of the effectiveness and total cost of a tax on sulfur
emissions to the control cost estimates: petroleum refineries—1978
Change in
all control cost
estimates
(percent)
+20
+10
+ 5
- 5
-10
-20
Change -in emissions
(percent)
Tax rate
(cents per pound of sulfur emissions)
5 10 15 20 25 30
2.5 1.8 2.5 1.1 0.4 0.2
0.8 0.9 1.7 0.4 0.2 0.0
0.2 0.5 0.9 0.1 0.1 0.0
-0.3 -0.5 -0.4 -0.2 -0.1 - 0.8
-1.2 -0.9 -0.9 -0.3 -0.2 -18.1
-8.0 -2.9 -1.7 -0.5 -0.3 -27.2
Change in total costs
(percent)
Tax rate
(cents per pound of sulfur emissions)
5 10 15 20 25 30
2.0 2.6 2.5 2.4 2.1 1.8
1.1 1.4 1.3 1.2 1.0 0.9
0.5 0.7 0.7 0.6 0.5 0.5
-0.6 -0.7 -0.7 -0.6 -0.5 -0.5
-1.2 -1.4 -1.4 -1.3 -1.1 -1.2 •-••
-2.9 -3.1 -3.0 -2.5 -2.1 -4.5
Source: Research Triangle Institute.
64
-------
that are 20 percent lower than those developed for this study would induce
petroleum refineries to reduce substantially the sulfur emissions from fuel
combustion. With a tax of 30 cents per pound and control costs for desul-
furization that are 20 percent lower than those projected in appendix B,
emissions from fuel combustion would be reduced by 61 percent or by 134,000
tons from a currently projected level of 221,000 tons from those sources.
Overall, emissions would be 27 percent less than those projected without
the application of control policies.
4.5 Sulfuric Acid Producers
Projections of the response of the nation's sulfuric acid producers
to a tax on sulfur emissions have been made on a pi ant-by-plant basis
using the emissions and control data shown in appendix C; these projections
are summed to obtain industry totals. The resulting cost and emission
estimates are approximations in that they do not fully reflect variations
in process types and operating procedures among sulfuric acid producers
due to the discrete nature of this analysis. However, the results are
regarded as very reasonable first order estimates.
4.5.1 Background
In 1968, sulfuric acid production accounted for an estimated 1.8
percent of the estimated nationwide sulfur emissions from all sources.*
Sulfuric acid is a strong, low priced, inorganic acid utilized in the
production of phosphate fertilizers and other industrial chemicals, in
the processing of petroleum, in the production of synthetic fabrics, in the
pickling of steel, and in many other metallurgical applications.
Most of the nation's approximately 240 sulfuric acid plants are
owned by large, diversified corporations. These plants sometimes sell the
acid commercially but, more often, the plants provide one link in a verti-
cally integrated company whose final product requires sulfuric acid as an
intermediate input. These companies include sulfur and chemical producers,
petroleum refineries, fertilizer plants, and smelters.
Over 97 percent of all sulfuric acid is produced by the contact
process.t The remainder is produced by the obsolescent lead chamber
*National Air Pollution Control Administration, Nationwide Inventory
of Air Pollutant Emissions—1968, Raleigh, N. C., August 1970.
tEngineering Analysis of Emissions Control Technology for Sulfuric
Acid Manufacturing Process, Final Report, Chemical Construction Corporation,
New York, N. Y., For National Air Pollution Control Administration, March 1970.
65
-------
process, currently being phased out. This analysis is limited to a study
of the effects of a tax on sulfur emissions from the 183 plants using the
contact process; it does not include sulfuric acid production at primary
nonferrous smelters or petroleum refineries. Those sources are treated
elsewhere in this study.
In 1970, sulfuric acid capacity was about 94,322 tons daily, and
production for the year was 29,525 million tons (see fig. 21).
4.5.2 Industry Growth
From 1950 through 1970, sulfuric acid production increased by an
average of 4.2 percent annually. However, as shown in figure 21, growth
has varied throughout the period.
Expected growth in sulfuric acid capacity allows for two new 1,500-
ton-per-day plants each year between 1970 and 1978.* Assuming that the
same capacity-to-output relationship exists in 1978 as existed in 1970,
1978 production would be 37.035 million tons. The 1970-78 growth rate,
then, is assumed to be 2.9 percent.
1950 1952 I960 1956 1958 I960 1962 1964 1966 1968 1970
Figure 21. Sulfuric acid production trends (Source: U.S. Department of
Commerce).
*Research Triangle Institute, Unpublished Data for the Cost of Clean
Air, 1973.
66
-------
4.5.3 Effectiveness
The reponse of the nation's sulfuric acid industry to a tax on their
emissions of sulfur was projected for selected tax rates. Since all of the
emissions control alternatives available to this industry and costed for
this study result in the recovery or increased production of sulfuric acid,
it is reasonable to allow credit for revenues from its sale. Within this
section we have used $10 per ton as being the most likely 1978 market value
for additional sulfuric acid production from sulfuric acid producers due to
the tax. However, because of the uncertainties surrounding the future price
of recovered sulfuric acid (see app. E), the impact of the assumed market
price of recovered sulfur on the decisions by sulfuric acid producers to
control emissions is analyzed separately below (see sec. 4.5.5).
Sulfur emissions from all sulfuric acid producers, assuming no additional
control other than those required by New Source Performance Standards (a
zero tax), are projected to be about 377,000 tons in 1978 (see table 22).
Tabular results of the projected response of all sulfuric acid producers
for several tax rates are provided in table 23. This information is graph-
ically presented in figure 22. Control of sulfur emissions would not be
induced below a 5-cent tax rate.
Increasing the tax rate by 10 cents per pound (from 5 to 15 cents) is
expected to effect an 84-percent reduction in emissions from the level (385
thousand tons) that would occur at a 5-cent tax rate. Further doubling of
the tax rate from 15 to 30 cents per pound of sulfur emitted will yield only
a further 22-percent reduction from emissions at the 15-cent tax rate. At
a tax rate of 20 cents, 87 percent of potential emissions are expected to be
controlled. Beyond that rate, virtually no further reductions are induced.
Table 22. Projected emissions from sulfuric acid production—1978
Annual sulfur emissions
bource • (thousand tons of sulfur)
Normal plants, mist 10
Oleum plants, mist 12
All plants, gaseous 354
Total 376
*Assuming only controls required by New Source Performance Standards.
Source: Research Triangle-Institute.
67
-------
Table 23. Projected response of all sulfuric acid plants
to a national tax on sulfur emissions--1978
(recovered sulfuric acid valued at $10 per ton)
Emissions source
Gaseous
Mist
Normal
Oleum
Total from all sources
Gaseous
Mist
Normal
Oleum
Total from all sources
Gaseous
Mist
Normal
Oleum
Total from all sources
Gaseous
Mist
Normal
Oleum
Total from all sources
Gaseous
Mist
Normal
Oleum
Total from all sources
Gaseous
Mist
Normal
Oleum
Total from all sources
Emissions Reductions Total Annualized A , t
(thousand Remissions annual control t
*««e\ from 2e«> tax COSt COSt (thnn«;anri<^
tons) (thousand tons)(thousands)(thousands) (thousands>
Tax rate:
362.6
22.8
10.7
12.1
385.4
Tax rate:
82.9
13.8
10.7
3.1
96.7
Tax rate:
47.4
13.5
10.5
3.0
60.9
Tax rate:
44.4
4.3
1.9
2.4
48.7
Tax rate:
44.0
3.4
1.5
1.9
47.4
Tax rate:
43.9
3.0
1.4
1.6
46.9
5 cents per pound of sulfur emissions
0.
0.
0.
0.
0.
10
279.
9.
0.
9.
288.
15
315.
9.
0.
9.
324.
20
318.
13.
8.
9.
336.
25
318.
19.
9.
10.
337.
30
318.
19.
9.
10.
338.
0
0
0
0
0
cents
7
1
0
1
8
cents
1
3
2
1
4
cents
1
6
9
7
7
cents
5
4
2
2
9
cents
6
9
3
6
5
$36,257
2,289
1,076
1,213
$38,546
per pound
$55,498
3,912
2,152
1,760
$59,410
per pound
$61,505
5,292
3,228
2,064
$66,797
per pound
$66,044
5,974
3,613
2,361
$72,018
per pound
$70,461
6,371
3,781
2,589
$76,831
per pound
$74,854
6,694
3,929
2,765
$81 ,548
$0
0
0
0
$0
of sulfur
$38,927
1.147
0
1,147
$40,U74
of sulfur
$47,269
1,217
65
1,152
$48,486
of sulfur
$48,270
4,255
2,871
1,384
$52,525
of sulfur
$48,447
4,663
3,021
1,642
$53,110
of sulfur
$48,499
4,893
3,071
1,321
$53,392
$36,257
2,289
1,076
1,213
$38,546
emissions
$16,570
2,765
2,152
613
$19,335
emissions
$14,236
4,075
3,163
912
$18,311
emissions
$17,773
1,719
742
977
$19,492
emissions
$22,014
1,708
761
947
$23,722
emissions
$26,356
1,801
858
949
$28,158
Source: Research Triangle Institute.
68
-------
500
SULFUR EMISSIONS REDUCTIONS
(RIGHT SCALE)
u co 900
SULFUR EMISSIONS
(LEFT SCALE)
en
10
20
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure 22. Effectiveness of a tax on sulfur emissions: sulfuric acid
producers—1978 (Source: Research Triangle Institute).
4.5.4 Costs
The total annualized cost to the sulfuric acid producers are shown in
table 23 for selected tax rates. Those same results are given graphically
in figure 23; costs increase at a decreasing rate because relatively small
increases in control costs yield significant emission reductions over the
15- to 30-cent tax range. For example, a 10-percent increase in control
costs (from $48 to $53 million) causes a 23-percent reduction in emissions.
Some insight of the relative magnitude of these costs (control cost
plus tax payments) is shown by allocating them on a per-unit-of-product
basis as shown in figure 24. For reference, the 1970 average value of
sulfuric acid was about $20 per ton. At a tax rate as high as 30 cents,
these costs would represent about 10 percent of that value.
4.5.5 Sensitivity Analysis
The projected effectiveness and costs of a tax on the sulfur emissions
of sulfuric acid'producers is influenced not only by the tax rate but also
69
-------
20
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
25
30
Figure 23. Total costs induced by a tax on sulfur emissions: sulfuric
acid producers--1978 (Source: Research Triangle Institute).
20
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
25
30
Figure 24. Average total* incremental costs per ton of acid production induced
by a tax on sulfur emissions: sulfuric acid producers—1978 (*control cost
plus tax payments) (Source: Research Triangle Institute).
70
-------
by the assumed market value of the recovered sulfurlc acid and the estimated
control costs of the sulfur emissions control alternatives. To examine the
sensitivity of the projected effectiveness and costs of the sulfur tax, a
sensitivity analysis was performed. Parametric variations were introduced
in the assumed future value for recovered sulfur and in the control cost
estimates presented in appendix C. The resulting percentage deviations
in emissions and total costs from those which obtain in the presence of
the assumed sulfur price of $10 and of control co'sts as presented in appendix
C are given for variations in the price of sulfur in figure 25 and for vari-
ations in control costs in table 24.
As shown in figure 25, reducing or increasing the market value of the
recovered sulfuric acid by $10 per ton changes projected total costs by
about 5 percent in the direction opposite the price change for all tax rates
that induce emissions control. For example, if a value of $0 per ton for
recovered sulfur is used instead of $10, then for a tax rate of 10 cents,
S IS
10
w*
2
H:
•10
VALUE OF RECOVERED
SULFUR $0/TON
VALUE OF RECOVER
SULFUR $ 20/TON
10
20
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
25
30
Figure 25. Sensitivity of the effectiveness and total costs of a tax
on sulfur emissions to the value of recovered sulfur: sulfuric acid
producers—1978 (*The base values refer to both the total control costs
and the emission levels that obtain when the base sulfur price is
assumed equal to $10 per ton) (Source: Research Triangle Institute).
71
-------
Table 24. Sensitivity of effectiveness and total costs of tax on sulfur
emissions to the control cost estimates: sulfuric acid producers—1978
Change in all
control cost
estimates
(percent)
+20
+10
+ 5
- 5
-10
-20
Change In emissions
(percent)
Tax rate
(cents per pound of sulfur emissions)
5 10 15 20 25 30
0.0 66.1 14.3 6.4 2.3 1.1
0.0 21.2 8.1 2.7 1.3 0.6
0.0 14.7 0.4 2.1 0.5 0.4
0.0 -10.5 - 5.9 -0.4 -0.4 -0.8
-23.6 -10.6 -15.6 -0.9 -0.6 -1.6
-31.6 -27.8 -18.0 -2.6 -1.3 -1.0
Change in total cost
(percent)
Tax rate
(cents per pound of sulfur
5 10 15 20
0.0 12.5 15.1 15.4
0.0 6.7 7.6 7.7
0.0 3.4 3.9 3.9
0.0 - 3.6 - 3.9 - 3.9
-0.3 - 7.4 - 8.0 - 7.8
-3.6 -15.5 -16.2 -15.6
emissions)
25 30
14.6 13.9
7.3 7.0
3.7 3.5
- 3.7 - 3.5
- 7.4 -14.0
-14.7 - 7.0
Source: Research Triangle Institute.
emissions are 15 percent and total cost 5 percent higher than projected
above. These deviations in the price of sulfur would not substantively
affect the rate of emissions for any tax rate at or above 20 cents. The
largest percentage reduction in emissions, compared to those that occur
when the price of sulfur is $10 per ton, occurs at a tax rate of 15 cents
when the sulfur price is $20 per ton; that percentage deviation is slightly
more than 5 percent. All other such percentage reductions are smaller.
On the other hand, when the recovered sulfur is assumed worthless, emissions
could increase as much as 15 percent above those levels projected when the
value of sulfur is $10 per ton; this most sizable deviation occurs at a
tax rate of 10 cents per pound. At tax rates below 10 cents per pound,
no emissions reductions are stimulated for any of the chosen variations in
sulfur prices.
The sensitivity of the emissions and total cost projections to positive
and negative percentage deviations in the control costs estimates are
presented in table 24. Reductions in control costs of 10 percent would
induce control at 5 cents per pound by some sulfuric acid plants. Emissions
are most influenced in the midtax ranges (10 to 15 cents), whereas total
costs are most affected by changes of +_ 10 and 20 percent in control costs
across all tax rates.
4.6 Primary Nonferrous Smelters
Projections of the response of the nation's primary copper, zinc,
and lead smelters to a tax on sulfur emissions have been made on a
72
-------
smelter-by-smelter basis using the emissions and control data shown in
appendix D. Individual projections were summed to obtain predictions for
the entire industry. The resulting projections, however, are approximations
that do not reflect all of the variations in process types and operating
procedures among smelters.
4.6.1 Background
Copper, zinc, and lead smelters accounted for an estimated 11.7 percent
of the estimated 1968 nationwide sulfur emissions from all sources.* The
metals produced from these smelting operations are used in a myriad of
applications: for galvanizing, for castings, in electrical equipment and
supplies, in brass and bronze products, in storage batteries, and as addi-
tives to gasoline. In 1970, primary smelter production capacities in
copper (15 plants), zinc (7 plants), and lead (6 plants) were 4,662, 2,236,
and 2,114 tons per day, respectively. Actual production figures for 1970,
in thousands of tons, were 1,765 (copper), 881 (zinc), and 572 (lead)
(see fig. 26).
4.6.2 Industry Growth
From 1950 through 1970, copper production grew fairly steadily (1.8
percent annually) whereas the production growth rates in primary zinc and
lead production were both more erratic and lower overall. Copper production
is expected to grow 2.2 percent annually reaching 2.101 million tons
annually by 1978. Both zinc and lead production were projected to remain
at their 1970 levels through 1978, since only negligible growth is
anticipated.t
4.6.3 Effectiveness
The response of the nation's primary nonferrous smelters to a tax on
their emissions of sulfur were projected for several tax rates between 0
and 30 cents per pound of sulfur. Since the emissions control alternatives
costed for primary nonferrous smelters result in the recovery of sulfur,
it is necessary to allow credit for revenues from its sale. Within this
section we have used $0 per ton as being the most likely 1978 market value
for sulfuric acid recovered by nonferrous smelters. This price reflects
*National Air Pollution Control Administration, Nationwide Inventory
of Air Pollutant Emissions—1968, Raleigh, N. C., August 1970.
tResearch Triangle Institute, Unpublished Data for the Cost of Clean
Air. 1973.
73
-------
1,800
I.OOO
1,100
TOO1—l
600
§ 500
|2
K I*-
QO
|g 400
H
-1
F 300
200
I95O 1952 1954 1956 1958 I960 1962 1964 1966 1968 ©TO
YEAR
Figure 26. Primary nonferrous smelting trends (indicated rates are compounded
annual growth rates) (Source: Research Triangle Institute).
74
-------
the fact that most of the nation's smelters are not located in areas where
their recovered acid can be easily sold. However, because of the uncer-
tainties surrounding the future price of recovered sulfuric acid (see app.
E), the impact of the assumed market price of recovered sulfuric acid on
the decisions by smelters to control emissions is analyzed separately
(see sec. 4.6.5).
Sulfur emissions from all nonferrous smelters are projected to be
about 1.7 million tons in 1978, assuming a zero tax and no further control
inducements (see table 25).
Over 90 percent of all projected sulfur emissions from smelters are
from copper smelters, primarily copper smelters without an acid plant.
This percentage is greater than copper's share of the total production
(measured in tons) of primary nonferrous smelters due to differences in
the sulfur content of the three ores and in processing techniques.
Tabular results of the projected response of all primary nonferrous
smelters to the sulfur tax for several tax rates are provided in table 26.
This information is presented graphically in figure 27.
Tax rates below 5 cents per pound are projected to induce primary
nonferrous smelters to reduce substantially their sulfur emissions as
compared to those at the zero tax level. For example, the projections
indicate that a tax of 2 cents would cause most copper smelters to reduce
sulfur emissions to 69 percent of their zero tax level emissions. This
would be a reduction of 55 percent in sulfur emissions from all nonferrous
smelters taken together. A tax of 3 cents would induce additional control
by copper smelters, cause zinc smelters without an acid plant to control,
Table 25. Projected sulfur emissions from primary
nonferrous smelters—1978
~ Annual sulfur emissions
iource (thousand tons of sulfur)
Copper smelters 1,534
Zinc smelters 56
Lead smelters 60
Total 1,650
*Assuming no implementation of emissions standards.
Source: Research Triangle Institute.
75
-------
Table 26. Projected response of all primary nonferrous smelters to a
national tax on sulfur emissions—1978 (recovered sulfur
valued at $0 per ton)
Emissions
source*
Copper total
Cu-A
Cu-B
Cu-C
Cu-D
Zinc total
Zn-A
Zn-B
Zn-C
Lead total
Pb-A
Pb-B
Pb-C
Total all sources
Copper total
Cu-A
Cu-B
Cu-C
Cu-D
Zinc total
Zn-A
Zn-B
Zn-C
Lead total
Pb-A
Pb-B
Pb-C
Reductions Total Annual 1 zed
Emissions 1n emissions annual control
(thousand from zero tax cost cost
tons) (thousand tons) (thousands) (thousands)
Tax rate:
248.2
33.5
30.3
32.2
152.2
7.8
1.8
2.3
3.7
22.6
0.6
1.3
20.7
278.6
Tax rate:
84.2
33.5
30.5
2.8
17.4
3.6
1.8
1.5
0.3
22.6
0.6
1.3
20.7
Annual tax
payment
(thousands)
5 cents per pound of sulfur emissions
1,286.2
720.7
177.3
118.8
269.4
47.9
43.4
4.5
0.0
37.2
11.7
25.5
0.0
1,371.3
10 cents per
1 ,450.5
720.7
177.3
148.2
404.3
52.0
43.4
5.3
3.3
37.2
11.7
25.5
0.0
78,070
32,021
15,644
6,372
24,033
2,563
1,550
637
376
4.465
947
1,438
2.080
85,098
pound of sulfur
88,558
35,369
18,670
7,082
27,437
3.040
1,732
830
478
6,746
1,011
1,576
4,159
53,253
28,674
12,617
3,153
8,809
1,770
1,368
402
0
2,183
884
1.299
0
57,206
emissions
71 ,770
28,674
12,617
6,522
23,957
2,282
1,368
511
403
2,183
884
1,299
0
24,816
3,348
3,027
3,218
15,224
793
182
235
376
2,282
64
138
2,080
27,892
16,789
6,695
6,053
560
3,481
758
364
319
75
4,563
128
276
4,159
Total all sources 110.4
See footnotes last page.
1,539.7
98,344
76,235
22,110
76
-------
Table 26. Projected response of all primary nonferrous smelters to a
national tax on sulfur emissions—1978 (recovered sulfur
valued at $0 per ton) (con.)
Reductions Total Annual i zed
Emissions 1n emissions annual control
Emissions (thousand from zero tax cost cost
source* tons) (thousand tons) (thousands) (thousands)
Copper total
Cu-A
Cu-B
Cu-C
Cu-D
Z1nc total
Zn-A
Zn-B
Zn-C
Lead total
Pb-A
Pb-B
Pb-C
Total all sources
Copper total
Cu-A
Cu-B
Cu-C
Cu-D
Zinc total
Zn-A
Zn-B
Zn-C
Lead total
Pb-A '.
Pb-B ~
Pb-C
Tax rate:
84.0
33.5
30.3
2.8
17.4
3.6
1.8
1.5
0.3
6.9
0.6
1.3
5.0
94.8
Tax rate:
70.5
20.0
30.3
2.8
17.4
3.6
1.8
1.5
0.3
6.9
0.6
1.3
5.0
15 cents per
1,450.4
720.7
177.3
148.2
404.3
52.0
43.4
5.3
3.3
52.9
11.7
25.5
15.7
1,555.6
20 cents per
1,464.0
734.2
177.3
148.2
404.3
52.0
43.4
5.3
3.3
52.9
11.7
25.5
15.7
pound of sulfur
96,953
38,717
21 ,697
7,362
29,177
3,420
1,915
989
516
8,464
1,075
1,714
5,675
108,837
pound of sulfur
104,901
41,617
24,724
7,642
30,918
3,798
2,097
1,147
554
9,173
1,139
1,852
6,182
emissions
71 ,769
28,674
12,617
6,522
23,957
2,281
1.368
511
403
6,335
883
1,299
4,153
80,386
emissions
76,707
33,611
12,617
6,522
23,957
2,282
1,368
511
403
6,336
884
1,299
4,153
Annual tax
payment
(thousands)
25,184
10,043
9,080
841
5,221
1,138
546
477
113
2,128
191
415
1,522
28,451
28,194
8,006
12,106
1,121
6,961
1,516
727
638
151
2,838
255
553
2,030
Total all sources 81.0
See footnotes last page.
1,568.9
117,872
85,325
32,548
77
-------
Table 26. Projected response of all primary nonferrous smelters to a
national tax on sulfur emissions—1978 (recovered sulfur
valued at $0 per ton) (con.)
Emissions
source*
Emissions
(thousand
tons)
Reductions
1n emissions
from zero tax
(thousand tons)
Total
annual
cost
(thousands)
Annual 1 zed
control
cost
(thousands)
Annual tax
payment
(thousands)
Copper total
Cu-A
Cu-B
Cu-C
Cu-D
Z1nc total
Zn-A
Zn-B
Zn-C
Lead total
Pb-A
Pb-B
Pb-C
Total all sources
Copper total
Cu-A
Cu-B
Cu-C
Cu-D
Z1nc total
Zn-A
Zn-B
Zn-C
Lead total
Pb-A
Pb-B
Pb-C
Total all sources
Tax rate: 25 cents per pound of sulfur emissions
68.7
18.2
30.3
2.8
17.4
3.6
1.8
1.5
0.3
6.9
0.6
1.3
5.0
79.2
Tax rate:
68.7
18.2
30.3
2.8
17.4
3.3
1.8
1.2
0.3
6.9
0.6
1.3
5.0
78.9
1,465.7
735.9
177.3
148.2
404.3
52.0
43.4
5.3
3.3
52.9
11.7
25.5
15.7
111,832
43,502
27,750
7,922
32,658
4,178
2,279
1.308
591
9,884
1,203
1,991
6,690
77,487
34,391
12,617
6,522
23,957
2,282
1,368
511
403
6,335
883
1,299
4,153
1,570.6 125,894 86,104
30 cents per pound of sulfur emissions
1,465.7
735.9
177.3
148.2
404.3
52.4
43.4
5.7
3.3
52.9
11.7
25.5
15.7
1,571.0
118,702
45,324
30,777
8.203
34,398
4,527
2,461
1,437
629
10,592
1,266
2,129
7,197
133.821
77,487
34,391
12,617
6,522
23,957
2,466
1,368
695
403
6,335
883
1,299
4,153
86,288
34,346
9,111
15,133
1,401
8,701
1,897
911
798
188
4,549
319
691
3,539
40,792
41,216
10.933
18,160
1.681
10,442
2,060
1,093
741
226
4,256
383
829
3,044
47,532
Lead
*Abbrev1at1ons:
See Appendix D.
Copper
Cu-A Green feed—no add plant
Cu-B Green feed—with acid plant
Cu-C Conventional feed-no add plant
Cu-D Conventional feed-vrith add plant
Z1nc
Zn-A Combination roaster—sintering, no add plant
Zn-B Roaster with sintering—with add plant
Zn-C Roaster with electrolytic purification and add plant
Source: Research Trlange Institute.
Pb-A Downdraft sintering—no add plant
Pb-B Updraft sintering—no add plant
Pb-C Updraft sintering—with add plant
78
-------
w ™
ff
£
CO
I
co
o
b
co
o
co
CO
2E
ui
I l_ I ^ 110° p
^*^ ci ii n ID rajieei/\Aie acru I/*TI/^MO or» ?•
SULFUR EMISSIONS REDUCTKDNS-
(RIGHT SCALE)
SULFUR EMISSIONS
(LEFT SCALE)
o
(K
80
70
to
CO
CO
60
LJ
50 tr
40
CO
30 ^
co
20 2
10
u
o
UI
tr
10
20
25
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
30
Figure 27. Effectiveness of a tax on sulfur emissions—1978 (Source:
Research Triangle Institute).
and some lead smelters would be induced to control. A tax of about 10
cents per pound would cause the last significant reduction in emissions.
At this tax, emissions are projected to be only 6 percent of their zero
tax level. From 10 to 30 cents per pound, further reductions are projected
to be minimal.
4.6.4 Costs
The total cost to the primary nonferrous smelting industry are shown
in table 26 for selected tax rates. As graphically shown in figure 28,
the total cost to the industry, comprising control costs and tax payments,
increase at a generally decreasing rate. This is because of the high levels
of emissions control induced by the tax.
These costs are shown on a per-un1t-of-product basis in figure 29,
assuming a perfectly inelastic demand for the three metals, and exclusive
of the impacts the corporate income tax may have on costs. For reference,
the 1970 average values of copper, zinc, and lead were $1,259, $306, and
79
-------
CO
K
-------
$275 per ton, respectively. It appears that at tax rates sufficient to
induce control of 94 percent of potential emissions (10 cents), the effect
on product prices would be minimal, particularly for lead and zinc.
4.6.5 Sensitivity Analysis
The projected effectiveness and costs of a tax on sulfur emissions
of primary nonferrous smelters is influenced not only by the tax rate but
also by the assumed market value of the recovered sulfur and the estimated
control costs of the sulfur emissions control alternatives. To examine
the sensitivity of the projected effectiveness and costs to the assumed
future value for recovered sulfur and to the cost estimates presented in
appendix D, the percentage deviations in emissions and total costs caused
by changes in the value of sulfur and control costs have been examined.
As shown in figure 30, if the sulfuric acid recovered by smelters
could be sold at $10 per ton, rather than $0 as used above, the difference
in total costs would be fairly significant. Because of the large amounts
of sulfur emissions from nonferrous smelters and because of the relatively
economical methods for emissions control, sale of the recovered acid at
PERCENTAGE DEVIATIONS FROM PROJECTE
BASE VALUES DUE TO VARIATIONS FROM
A BASE RECOVERED SULFURIC ACID PRICE
OF $0/TON
§oi£biro-L
_ oooooooooc
TOTAL
COST
EMISSIONS'
1
^
/
f
^^^f^L
\
^-
j
— n
r «
VALUE OF RECOVERED
SULFURIC ACID=$ 10/TON
) 5 10 15 20 25 30
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure 30. Sensitivity of the effectiveness and total costs of a tax
on sulfur emissions to the value of recovered sulfur--1978 (*The base
values refer to both the total control costs and the emission levels
that obtain when recovered sulfur is worthless) (Source: Research
Triangle Institute).
81
-------
$10 per ton would reduce the costs induced by the tax. Lower costs would
also mean lower emissions. For example, at a tax of 5 cents, and a $10 per
ton value for recovered sulfur, emissions would be about 45 percent less
and total costs 15 percent less than that projected, assuming the recovered
sulfur was worthless.
The sensitivity of the emissions and total costs projections to
changes in the control costs estimates of +5, 10, and 20 percent are
presented in table 27. Emissions are significantly different from those
projected using the cost estimates presented in appendix E for tax rates
of only 15 cents and for increases in the control costs of 10 to 20 percent.
Because the schedules of marginal emissions reduction costs, in some
cases, are perfectly inelastic over certain ranges of costs, it is possible
that what would otherwise appear as unusual predictions would in fact
occur. An example of the effect of this inelasticity may be noted in the
presence of a 10-percent decrease in control cost estimates (see table 27);
a 10- and a 15-cent tax fail to induce additional emissions reductions above
those projected (because the marginal cost curve is vertical over that range)
but a 20-cent tax stimulates a 2-percent further reduction in emissions
(because the marginal cost curve becomes more elastic in that neighborhood
of costs). Total control related outlays are never more than 10 percent
different from those projected under the assumed regime of control costs
unless those costs are raised or lowered 20 percent or more.
• * •• r
i
Table 27. Sensitivity of effectiveness and total cost of tax on sulfur
emissions to control cost estimates: primary nonferrous smelters--!978
Change in
all rnntvnl
cost
estimates
(percent)
+20
+10
+ 5
- 5
-10
-20
Change in emissions
(cents per
5
1.7
0.0
0.0
-16.6
-30.0
-50.0
10
0.0
0.0
0.0
0.0
0.0
-3.8
(percent)
Tax rate
pound of sulfur emissions)
15 20
12.2 16.5
8.2 8.7
0.0 2.7
0.0 0.0
0.0 -2.2
-9.4 -2.2
25
2.3
0.0
0.0
-0.5
-0.5
-0.5
30
0.5
0.0
0.0
0.0
-6.1
-6.1
Change in
total
cost
(percent)
(cents
5
13.4
6.7
3.4
-3.4
-7.3
-15.7
Tax
per pound
10 15
15.
7.
3.
-3.
-7.
-15.
5 14.5
8 7.4
9 3.7
9 -14.5
7 - 7.4
5 -14.9
rate
of sulfur emissions)
20
14.0
7.1
3.6
-3.6
-7.2
-14.6
25
13.7
6.8
3.4
-3.4
-6.8
-13.7
30
12.9
6.5
3.2
-3.2
-6.5
-13.2
Source: Research Triangle Institute.
82
-------
Chapter 5: IMPACT OF SULFUR TAX ON AIR QUALITY IN
TWO HYPOTHETICAL AIR QUALITY CONTROL REGIONS
5.1 Introduction
This chapter briefly examines the effect of a sulfur tax on ambient
sulfur dioxide ($02) concentrations in two hypothetical Air Quality Control
Regions (AQCR's). A proportional model is used to relate annual sulfur
emissions to annual arithmetic mean S0« concentrations. This model assumes
uniform air quality across a region and a proportional relationship
between changes in emissions and air quality. It is recognized that air
quality varies across most regions; quality varies according to location
and temporal and climatic conditions. The use of the proportional model
here is only expositional; diffusion modeling is required to predict the
distribution of air quality within a region. The effects of several sulfur
tax rates on emissions and on S02 concentrations are compared, and the
required tax rates to achieve the National Ambient Air Quality Standards
for S02 in each region are determined.
5.2 Relationship between SOg Concentration and Sulfur Emissions
Assuming ambient pollutant concentration proportional to pollutant
emissions, the concentration, C, resulting from a reduction in emissions
is given by:
C = Cp (1 - r) (1)
where C is the initial ambient pollutant concentration and r is the
P
reduction in emissions expressed as a decimal percent.
Similarly, emissions after reduction, E, can be obtained from:
E = Ep (1 - r) (2)
where E is initial emissions and r is defined above.
p
If an ambient air quality standard, CgJ is specified:
Cs = Cp (1 - r) , (3)
then this equation, upon dividing both sides by C and changing signs,
gives the familiar proportional equation:
C - C
-V-1 =r •
S
83
-------
Thus, Eq. 1 can be used to calculate a new pollutant concentration resulting
from a reduction in emissions, Eq. 2 can be used to calculate the emissions
corresponding to a specified decimal percent reduction, and Eq. 4 can be
used to calculate the required decimal percent reduction in emissions to
achieve a specified ambient air quality standard.
5.3 Hypothetical AQCR's
5.3.1 Hypothetical AQCR A
This AQCR contains a petroleum refinery, a sulfuric acid plant, and
point and area combustion sources of sulfur emissions including steam-
electric power plants. Estimated annual uncontrolled sulfur emissions
are 568,301 tons. Table 28 shows the distribution of emissions by source.
The annual arithmetic mean S02 concentration is 0.085 ppm.
Table 28. Sulfur emissions by source for two hypothetical
Air Quality Control Regions*
(tons per year)
Source
Uncontrolled
Tax Rate (cents per ton of sulfur emissions)
5 10 15 20 25 30
AQCR A
Petroleum refineries
Sulfuric add plants
Primary nonferrous smelters
Lead
Zinc
Copper
Stationary combustion
Total
AQCR B
Petroleum refineries
Sulfuric add plants
Primary nonferrous smelters
Lead
Zinc
Copper
Stationary combustion
Total
47,597
11,294
0
0
0
509,410
568,301
19,776
13,265
10,373
781
0
450,393
494,588
40,750
11,294
0
0
0
182,157
234,201
17,083
13,265
10,373
781
0
148,203
189,705
35,845
3,298
0
0
0
105,975
145,118
15,539
2,345
10,373
79
0
85,948
114,284
34,079
1,357
0
0
0
79,227
114,663
14,690
1,670
2,581
79
0
68,532
87,552
34,079
1,295
0
0
0
61 ,817
97,191
14,465
1,435
2,581
79
0
66,236
84,796
34,079
1,094
0
0
0
61 ,553
96,726
14,465
1,435
2,581
79
0
45,072
63,632
34,079
1,094
0
0
0
56.847
92,020
14,465
1,368
2,581
79
0
42,463
60,956
*Tnese are representative of large metropolitan regions.
84
-------
5.3.2 Hypothetical AQCR B
This AQCR contains a primary nonferrous smelter, a petroleum refinery,
a sulfuric acid plant, and point and area combustion sources of sulfur
emissions including steam-electric power plants. Estimated annual un-
controlled sulfur emissions are 494,588 tons. Table 28 shows the distri-
bution of emissions by source. The annual arithmetic mean SCL concentration
is 0.047 ppm.
5.4 Effect of Sulfur Tax on Emissions and Ambient S00 Concentrations
- i_ -
5.4.1 Sulfur Emissions
The relationship between annual sulfur emissions and tax rate for
the two hypothetical ACQR's is shown in figure 31. Annual sulfur emissions
and the decimal percent reduction in annual sulfur emissions associated
with each tax rate, as well as the annual sulfur emission and decimal
percent reduction required to achieve the National Ambient Air Quality
Standards for S02 (0.03 ppm annual arithmetic mean), are shown in table 29.
5.4.2 Ambient SO,, Concentrations
The reductions associated with each tax rate in table 29 and the
initial S02 concentrations were substituted in Eq. 1 to calculate the
5 10 15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure 31. Effectiveness of a tax on sulfur emissions for two hypothetical
Air Quality Control Regions (Source: Research Triangle Institute).
85
-------
Table 29. Annual sulfur emissions and percent reduction
with various tax rates
Hypothetical AQCR A
Annual sulfur
emissions
(tons /year)
Reduction
(decimal percent)
Hypothetical AQCR B
Annual sulfur
emissions
(tons/year)
Reduction
(decimal percent)
Tax rate (cents per pound of sulfur emissions)
0
568,301
0
494,588
0
5
234,201
0.59t
189,705
0.62t
10
145,118
0.75
114,284
0.77
15
114,663
0.80
87,552
0.82
20
97,191
0.83
84,796
0.83
25
96,726
0.83
63,632
0.87
30
92,020
0.84
60,956
0.88
To achieve
ambient air
quality standard
198,905*
0.65*
316,536*
0.36f
Calculated from Eq. 2.
tCalculated by rearranging Eq. 2 and solving for r.
^Calculated from Eq. 4.
corresponding annual arithmetic mean SOg concentrations which are plotted
in figure 32. Inspection of figure 32 indicates that the standard would
be achieved at tax rates approximating 6 and 2 cents per pound of sulfur
in AQCR's A and B, respectively.
The simplified analysis presented herein illustrates the relationship
between sulfur emissions and ambient S02 concentrations and the effect
of increasing sulfur tax rates on both. However, it must be remembered
that the assumed proportionality describes air quality in terms of one
concentration, which is assumed to prevail throughout the AQCR. Since the
sulfur tax would permit each source to select a cost minimizing mix of
tax payments and emission control expenditures, the air quality standard
may be exceeded in the vicinity of some sources if the tax rate is too
low. To adequately explore the regional effects of a sulfur tax, it would
be necessary to employ a diffusion model capable of examining the
influence of individual sources on air quality.
86
-------
NATIONAL AMBIENT AIR QUALITY STANDARD
15
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure 32. Relationship between tax rates and air quality for two hypothetical
Air Quality Control Regions (Source: Research Triangle Institute).
87
-------
BIBLIOGRAPHY
American Petroleum Institute. Petroleum Facts and Figures. New York, 1971.
Ayres, R. U., and Kneese, A. V. "Production, Consumption, and Externalities."
American Economic Review 59. No. 3 (June 1969):282-297-
Barrett, Larry B., and Waddell, Thomas E. Cost of Air Pollution Damage: A
Status Report. Environmental Protection Agency, National Environmental
Research Center, Research Triangle Park, N.C., 1973.
Bassett, Lowell R., and Borcherding, T. E. "Externalities and Output Taxes."
Southern Economic Journal 36 (April 1970):462-464.
Battelle Memorial Institute, Pacific Northwest Laboratories. A Review
and Comparison of Selected United States Energy Forcasts. Prepared
for the Executive Office of the President, Office of Science and
Technology, Energy Policy Staff, 1969.
Baumol, W. J. "On Taxation and the Control of Externalities." The American
Economic Review 62, No. 3 (June 1972):307-22.
Bechtel Corporation. The Economics of Residual Fuel Oil Desulfurization.
Prepared for U.S. Department of Health, Education, and Welfare,
Public Health Service, June 1964.
Bradshaw, Thomas T. "Closing Circle on Environmental Economics."
Chemical and Engineering News, February 21, 1972, pp. 22-33.
Buchanan, J. M., and Stubblebine, W. C. "Externality." Economica,
November 1962, pp. 371-84.
Buckingham, Paul, and Homan, Harry R. "Sulfur and the Energy Industry.
"Hydrocarbon Processing 67, No. 8 (August 1971):121-125, Fluor Cor-
poration, Los Angeles, Calif.
Burchard, John K.; Rochele, Gary T.; Schofield, William R.; and Smith,
John 0. Some General Economic Considerations of Flue Gas Scrubbing
for Utilities. Environmental Protection Agency, Research Triangle
Park, N.C., 1972.
Chapman, Duan; Mount, Timothy; and Tyrrell, Timothy. Predicting the Past
and Future in Electricity Demand. Department of Agricultural Eco-
npmics, Cornell University, Ithaca, N.Y., 1972.
Coase, R. H. "The Problem of Social Cost." The Journal of Law Economics
3 (October 1960):1-44.
Davis, 0. A., and Whinston, A. B. "Some Notes on Equating Private and
Social Cost." The Southern Economic Journal 32, No. 2 (October 1965):
113-26.
-------
Edison Electric Institute. Fuels for the Electric Utility Industry. 1971-1985.
New York, 1972.
"The Effluent Fee Approach for Controlling Air Pollution." Duke Law Journal,
October 1970, pp. 943-90.
Environmental Protection Agency. "EPA Energy Quality Model Documentation."
Unpublished. Research Triangle Park, N.C., October 3, 1972.
Ethridge, D. W. An Economic Study of the Effect of Municipal Sewer Surcharges
on Industrial Wastes.Report 41, Water Resources Research Institute,
North Carolina State University, Raleigh, N.C., November 1970.
Fanner, M. H., and Bertrand, R. R. Long Range Sulfur Supply and Demand Model.
Prepared by Esso Research and Engineering Company, Government Research
Laboratory, Linden, N.J., for Environmental Protection Agency, Division
of Control Systems, Office of Air Programs, November 1971.
Federal Power Commission. Statistics of Privately Owned Electric Utilities
in the United States -1968, 1969, 1970 - Class A and B Companies, 1970.
Washington, D.C., 704 pp.
Fogel, M. E., Johnston, D, R.; Collins, R. L.; LeSourd, D. A.; Gerstle, R. W.;
and Hill, E. L. Comprehensive Economic Cost Study of Air Pollution Control
Costs for Selected Industries and Selected Regions.Final Report prepared
by the Research Triangle Institute, Research Triangle Park, N.C., on
Contract No. CPA 22-69-79 for National Air Pollution Control Administra-
tion, 1970.
Frank, Ernest H. Sulfur Review and Outlook. First Manhattan Co., New York,
May 24, 1968.
Freeman, A. M., III. The Economics of Pollution Control and Environmental
Quality. New York!General Learning Press, 1971, pp. 1-27.
Graham, Robert E., Jr.; Degraff, Henry L.; and Trott, Edward A., Jr.
"State Projections of"income, Employment, and Population." Survey
of Current Business 52, No. 4 (April 1972):22-48.
Hausgaard, 01af. "Proposed Tax on Sulfur Content of Fossil Fuels." Public
Utilities Fortnightly, September 16, 1971, pp. 27-33.
Heller, Walter W. "Economic Growth and Ecology: An Economist's View."
Monthly Labor Review, November 1971, pp. 14-21.
Hittman Associates, Inc. Cost Nomographs of Selected Sulfur Dioxide
Abatement Methods. Contract No. EHSD-71-43 for Environmental
Protection Agency, Office of Air Programs, Columbia, Md., 1972.
Hittman Associates, Inc. Electrical Power Supply and Demand Forecasts
for the United States through 2050.Prepared for Environmental
Protection Agency (EPA), Columbia, Md., 1972.
89
-------
Hittman Associates, Inc. Study of the Future Supply of Natural Gas for
Electrical Utilities.Prepared for Environmental Protection Agency,
Office of Air Programs, Columbia, Md., February 1972.
Hoskins, W. Lee. "Let the Pricing System Provide the Incentive."
Industrial Water Engineering, April 1971, pp. 8-10.
Hottel, H. C., and Howard, J. B. "Nuclear Power." In New Energy
Technology; Some Facts and Assessments. Cambridge, Mass.: M.I.T.
Press, 1972.
Kapp, K. W. "Environmental Disruption and Social Costs: A Challenge to
Economics." Kyklos 23 (1970):833-48.
M. W. Kellogg Company. Detailed Cost Breakdown for Selected Sulfur Oxide
Control Processes. Prepared for Environmental Protection Agency,
Office of Air Programs, Division of Control Systems, Washington, D.C.,
1971, 21 pp.
M. W. Kellogg Company. Evaluation of SOp Control Processes. Prepared for
Environmental Protection Agency, Office of Air Programs, Division of
Control Systems, Washington, D.C., 1971, 264 pp.
Kneese, A. V. (presiding). "Taxation and Management of Environmental
Quality. First Concurrent Conference Session, National Tax Associ-
ation. Columbus, Ohio, September 1971, pp. 88-147.
Kohn, Robert E. "Linear Programming Model for Air Pollution Control: A
Pilot Study of the St. Louis Airshed." Journal of the Air Pollution
Control Association 20 (February 1970):325-60.
LeSourd, D. A., and Bunyard, F. L., eds. Comprehensive Study of Specified
Air Pollution Sources to Assess the Economic Impact of Air Quality
Standards^Research Triangle Institute, Research Triangle Park, N.C.,
and Environmental Protection Agency. Final Report on Contract No.
68-02-0088 for Environmental Protection Agency, Division of Effects
Research, Research Triangle Park, N.C., 1972.
LeSourd, D. A.; Fogel, M. E.; Schleicher, A. R.; Bingham, T. H.; Gerstle,
R. W.; Hill, E. L.; and Ayer, F. A. Comprehensive Study of Specified
Air Pollution Sources to Assess the Economic Effects of Air Quality
Standards.Research Triangle Institute, Research Triangle Park, N.C.
Final Report on Contract No. CPA 70-60 for Environmental Protection
Agency, Division of Economic Effects Research, Washington, D.C., 1970.
Mac Avoy, Paul W. "The Regulation-induced Shortage of Natural Gas." Journal
of Law and Economics 14, No. 1 (April 1971):167-199.
Makin, John H. "Pollution as a Domestic Distortion in Welfare Theory."
Land Economics, May 1971, pp. 185-88.
90
-------
Meade, J. E. "External Economics and Diseconomies in a Competitive Situation."
The Economic Journal, March 1952, pp. 54-67.
Mishan, E. J. "The Postwar Literature on Externalities: An Interpretative
Essay." Journal of Economic Literature 9 (March 1971):!-28.
Mishan, E. J. "A Survey of Welfare Economics, 1939-59." The Economic
Journal 70 (June 1960):197-256.
Mishan, E. J. "Welfare Criteria for External Effects." The American
Economic Review, September 1961, pp. 594-613.
Morrison, Warren E., and Read!ing, Charles L. An Energy Model for the United
States, Featuring Energy Balances for the Years 1947 to 1965 and Pro-
jections and Forecasts"toi the Years 1980 and 2000. U.S. Department of
the Interior, Bureau of Mines, Washington, D.C., 1968.
National Academy of Science and National Academy of Engineering. Abatement
of Sulfur Oxide Emissions from Stationary Combustion Sources.Ad Hoc
Panel on Control of Sulfur Dioxide from Stationary Combustion Sources,
Committee on Air Quality Management, Committees on Pollution Abatement
and Control, Division of Engineering, National Research Council, Wash-
ington, D.C., 1970.
National Academy of Sciences and National Academy of Engineering. Man,
Materials, and Environment. Prepared for the National Commission on
Materials Policy, Washington, D.C., 1973.
National Air Pollution Control Administration. Air Quality Criteria for
Sulfur Oxides. Publ. No. AP-50. Washington, D.C.:U.S. Government
Printing Office, 1969.
National Coal Association, Division of Economics and Statistics. Steam-
Electric Plant Factors, 1969. Washington, D.C., 1969, 118 pp.
National Economic Research Associates, Inc. Fuels for the Electric Utility
Industry 1971-1985. Prepared for the Edison Electric Institute, New
York, August 1972.
National Economic Research Associates, Inc. Possible Impact of Costs of
Costs of Selected Pollution Control Equipment on the Electric Utility
Industry and Certain Power Intensive Consumer Industries. Prepared for
Council on Environmental Quality, Washington, D.C., January 5, 1972.
National Petroleum Council, Committee on U.S. Energy Outlook. U.S. Energy
Outlook: An Initial Appraisal, 1971-1985, (vols. 1 and 2)"; Washington,
D.C., November 1972.
Plott, C. R. "Externalities and Corrective Taxes." Economica, February 1966,
pp. 84-87.
Private communication with Gary Rochelle, Environmental Protection Agency,
Research Triangle Park, N.C.
91
-------
Robson, F. L., et al. Technological and Economic Feasibility of Advanced
Power Cycles and Methods of Producing Nonpolluting Fuels for Utility
Power stations^Final report on Contract No. CPA 22-69-114 for the
Environmental Protection Agency by VARL, 1970.
Rochelle, G. T. "A Critical Evaluation of Processes for the Removal of S02
from Power Plant Gas." Presented at Air Pollution Control Association
Meeting. Chicago, 111., 1973.
Ruff, Larry E. "The Economic Common Sense of Pollution." The Public Interest
10 (Spring 1970):69-85.
Slack, A. V., and Harrington, R.E. "Removal of Sulfur Dioxide from Power
Plant Stack Gas: Status of Limestone Processes. Presented at Second
International Clean Air Congress of the International Union of Air
Pollution Prevention Association, Dec. 6-11, 1970, Washington, D.C.
Solow, Robert M. "The Economist's Approach to Pollution and Its Control."
Science 173 (Augusts, 1971):498-503.
"Symposium: The future of the U.S. wage-price policy: Weidenbaum, Murray L.,
New initiatives in national wage and price policy; Ackley, Gardner, An
incomes policy for the 1970's; Madden, Carl H., Controls or competition—
what's at issue? Houthakker, Hendrick S., Are controls the answer?" The
Review of Economics and Statistics 54, No. 3 (August 1972):213-17.
Teller, A. "Air-pollution Abatement: Economic Rationality and Reality."
Daedalus, Fall 1967, pp. 1083-98.
Tennessee Valley Authority. Sulfur Oxide Removal from Power Plant Stack
Gas: Use of Limestone irTlJet-scrubbing Processes. Prepared for U.S.
Department of Health, Education, and Welfare, National Air Pollution
Control Administration, Washington, O.C., 104 pp.
Turvey, Ralph. "On Divergences between Social Cost and Private Cost."
Economica, August 1963, pp. 309-13.
Tybout, R. A. "Pricing of Pollution." The Bell Journal of Economics and
Management Science 3, No. 1 (Spring 1972):252-66.
U.S. Congress. House of Representatives. Energy "Demand" Studies: An_
Analysis and Appraisal. Prepared for the Committee on Interior and
Insular Affairs, Washington, D.C., September 1972.
U.S. Department of Health, Education, and Welfare. Control Techniques for
Sulfur Oxide Air Pollutants, Washington, D.C., January 1969.
U.S. Department of Health, Education, and Welfare, A Summary of Electric
Power Plant Fuel Use, Sulfur Emissions, and Fuel Cost (1969 Data),
Environmental Protection Agency, Office of Air Programs, Fuel Policy
Section, Washington, D.C., 1971.
92
-------
U.S. Department of the Interior, Bureau of Mines. Transportation Costs of
Fossil Fuels (interim report). Washington, D.C., 1971.
U.S. Environmental Protection Agency, Office of Air Quality and Standards.
Control Techniques for Sulfur Oxide Air Pollutants. (2nd ed. draft),
Washington, D.C., 1972.
Wellisz, Stanislaw. "On External Diseconomies and the Government-assisted
Invisible Hand." Economica, November 1964, pp. 345-62.
Winger, John G.; Emerson, John D.; and Gunning, Gerald D. Outlook for energy
in the United States. The Chase Manhattan Bank, N.A., New York, 1968.
Winton, John M. "Dark Cloud on Sulfur's Horizon." Chemical Week 108, No. 6
(February 10, 1971): 25-36.
93
-------
Appendix A: SULFUR EMISSION FACTORS AND CONTROL
COSTS FOR FUEL COMBUSTION SOURCES
This appendix describes the emission factors used, the sulfur emissions
sources examined, and the control alternatives costed as part of the
research presented in the text for steam-electric power plants and
area sources.
A.I Emission Factors
In the absence of controls, essentially all the sulfur contained in
fossil fuels is emitted as sulfur oxides. The amount of sulfur emissions
depends upon (1) the quantity of fuel consumed and (2) the quantity of
sulfur per unit of the consumed fuel.
For both the steam-electric power plants and the area sources, the
quantity of fuel consumed is based on projections of the demand for heat
energy expressed in Btu. Projections of the national demand for electricity
developed by the Edison Electric Institute* have been used to project
the demand for Btu by steam-electric power plants. The 1970 Btu consumption
by the area sources on a State-by-State basis have been continued to
1978 using projected population growth rates to project residential fuel
consumption and projections of employment growth rates to derive projections
of commercial and residential fuel consumption.t The estimated 1970
consumption of fuels by State and source are shown in table A.I.
The sulfur content of fuels is normally expressed as a percent
by weight. The amount of sulfur emitted per ton of fuel is 20 Ib for
each 1 percent of sulfur content in the fuel. For example, a typical
coal of 3.5 percent sulfur count releases 70 Ib of sulfur per ton during
combustion. Similarly, for residual oil weighing 7.88 Ib per gal, 1
percent sulfur oil emits 0.0788 Ib of sulfur per gal. Sulfur emissions
per unit of heat contained vary with the heat content of the fuel.
The heat content of coal varies significantly depending on the specific
coal. For residual oil and gas, there is little variability. Data
on the fuels and their heat contents are presented in the section on
control alternatives.
*Edison Electric Institute. Fuels for the Electric Utility Industry.
1971-1985, New York, Edison Electric Institute, 1972.
tU.S. Department of Commerce, Survey of Current Business. Vol. 52,
No. 4, April 1972, p. 22-48.
95
-------
Table A.I. Area source fuel consumption by State, 1970
to
en
State
Maine
New Hampshire
Vermont
Massachusetts
Rhode Island
Connecticut
New York
New Jersey
Pennsylvania
Delaware
Maryland
Dlst. of Columbia
Virginia
West Virginia
North Carolina
South Carolina
Georgia
Florida
Michigan
Ohio
Indiana
Residential
Distillate
011
(thousand
bbl)
8,547
6,279
4,293
46,243
6,640
16,399
76,637
38,743
37,327
2,992
11,346
2,583
11,507
295
12,148
3,961
1,936
3,929
24,368
14,156
15,056
Gas
(million ^
ft3)
3,400
53,400
19,800
225,300
98,900
200,600
5,100
57,400
31 ,700
38,300
18,900
12,400
59,800
7,900
226,600
301 ,400
106,600
Commercial
Distillate Residual
Coal oil oil
thousand (thousand (thousand
tons) bbl) bbl)
1,072 1,461.6
15 693 659.4
591 200.7
57 7,014 19,011.3
i '3,801 2,901.8
Gas
(million
ft3)
970.9
Industrial
Distillate Residual
Coal oil oil
(thousand (thousand (thousand
tons) bbl) bbl)
643.2 2,703.9
135.4 415.8 1,219.9
354.6 371.3
14,471.7 ' 293.3 4,208.4 35,170.9
1 Qn>> Q Included *1th cno 9 7 CQft Q
1,905.9 Mfi>> N> vts 508.2 2,598.9
4,892.4 251.1 2,160.6 5,368.3
80 18,106 36,451.8 58,269.8
8 12,883 8,367.4
17,777.3
600 6,160 8,265.5 46,331.5
307 315.9 1,338.8
1,428 3,134.8 J16$410>4
70 464 5,055.1 }
500 1,313 1,006.5 12,217.6
250 32 342.2
500 1,131 679.5
300 442 154.9
9,680.9
8,668.2
6,108.8
) ,,n 169 1,150.5 19,469.0
> lou
\ 351 833.3 10,602.1
900 2,079 899.8 67,759.4
1,500 1,466 298.9 82,459.5
600 1,416 1,748.9 36,922.9
5,659.6 10,863.6 67,435.8
450.1 7,729.8 15,479.7
Gas
(million
ft3)
681.8
7,789.9
2,159.5
5,477.9
39,746.4
26,933.6
7,711.9 3,696.0 15,291.2 119,294.0
1,168.9 184'2 584'4
856.8 5,799.4
375.3 278.4 9,351.9
4,000.9 787.8 1,862.0
4,936.1 19.2 633.1
2,001.5 678.6 1,257.1
1,300.0 265.2 286.6
526.9 101'4 2>128'4
210.6 1,541.6
4,188.5
14,324.1
14,799.4
31,913.7
24,566.2
27,281.1
50,037.8
31.668.4
7,922.6 1,247.4 1,664.6 100,399.6
14,601.1 879.6 552.9 128,687.7
6,418.4 849.6 3,235.5
94,407.6
-------
Table A.I. Area source fuel consumption by State, 1970 (continued)
State
Residential
Distillate
oil /
(thousand
bbl)
Gas
million
ft3)
Commercial
Coal
(thousand
tons)
Distillate
oil
(thousand
bbl)
Residual
oil
(thousand
bbl)
Gas
(million
ft3)
Industrial
Coal
(thousand
tons)
Distillate
oil
(thousand
bbl)
Residual
oil
(thousand
bbl)
Gas
(million
ft3)
Illinois
Wisconsin
Minnesota
Iowa
Missouri
Kansas
Nebraska
South Dakota
North Dakota
Kentucky
Tennessee
Alabama
Mississippi
Arkansas
Louisiana
Oklahoma
Texas
Montana
Idaho
Kyomlng
Colorado
Utah
19,601
17,417
13,456
6.340
1,145
530
1,866
2,572
2,617
1,950
1,679
566
629
449
618
523
2,470
970
2,792
581
1,170
1,064
295,200
68,800
65,000
62,000
103,200
64,100 )
37,500 '
8,800 \
5,200 '
57,000
30,900
37,300 l
20,600 )
39,700
56,400
51,200
150,100
14,600 |
4,700 J
11,300
55,100
29,900
2,800
1,600
900
200
100
60
200
500
600
100
4
. 250
20
300
160
2,550
1,919
1,118
467
620
44
158
146
220
182
148
53
58
40
57
35
157
37
178
52
53
125
8,873.1
599.6
605.7
158.6
1,225.5
262.9
79.3
9.2
42.1
70.8
9.2
47.6
29.9
-
9.8
100.7
337.9
111.6
71.9
207.4
208.0
407.5
97,085.4
21,472.9
24,231 .6
27,537.7
35,625.5
24,664.1 1
18,216.1 J
4,779.0 1
3,667.1 J
17,352.7
19,994.8
18,809.7 j
8,015.7 I
20,064.2
17,320.4
19,195.0
48,797.6
7,546.7
2,822.3 J
6,626.1
26,946.8
4,995.3
7,741.3
4,096.7
1,584.2
1,694.9
1,615.9
. 227.4
473.2
2,199.6
2,229.5
1,767.6
113.7
> 161.4
212.2
441.4
454.8
1,530.0
1,151.4
670.8
280.2
372.0
26.4
94.8
87.6
132.0
109.2
88.8
31.8
34.8
24.0
34.2
21.0
94.2
22.2
106.8
31.2
31.8
75.0
16,415.2
1,109.3
1,120.5
293.4
2,267.2
486.4
146.7
17.0
77.9
130.9
17.0
88.1
55.3
-
18.1
186.3
625.1
206.5
133.0
383.7
384.8
753.9
125.555.9
44,934.8
32,467.8
31.735.9
35,287.4
54,919.6
17,606.1
2,194.2
981.8
22,973.7
41,021.8
60,727.8
43,348.9
51,774.5
337,363.9
40,691.0
599,958.5
11,170.9
8,919.8
14,932.8
30,338.7
21,223.7
-------
Table A.I. Area source fuel consumption by State, 1970 (continued)
to
00
State
Nevada
New Mexico
Arizona
Washington
Oregon
California
Alaska
Hawaii
Total
Residential
Distillate Qas
(thousand
-------
A. 2 Sulfur Emissions Control Alternatives and Costs
There are two classes of control alternatives: (1) removal of sulfur
oxides from stack gases and (2) use of low-sulfur fuels (fuel switching).
Gas cleaning is technologically more complex, while use of low-sulfur
fuels has economic and resource availability complexities. In this study,
gas cleaning is considered only for the steam-electric power plants and not
for area sources because of the lack of data on boiler sizes and distributions.
A.2.1 Sulfur Oxide Removal from Stack Gases
Sulfur oxide removal from stack gases may be accomplished by adsorption
by a dry or wet medium. For both wet and dry, there are regenerable processes
(in which the sorbent is regenerated for reuse with a concentrated sulfur-
bearing byproduct stream) and throwaway processes (in which the sorbent
is discarded along with the captured sulfur values). Of the many processes
and process variants currently in various stages of development and/or
demonstration and utilization, three were selected for this analysis because
it appears that they will be available by 1978: (1) dry limestone adsorption,
(2) wet limestone slurry scrubbing, and (3) magnesium oxide (magnesia
base) solution scrubbing. These three techniques were chosen because
they are respresentative of a range of control costs and efficiencies.
A discussion of each of these control techniques follows. Their
costs are based on EPA studies. The control efficiencies of the three
systems costed are summarized in table A.2. The costs are summarized in
table A.3.
Table A.2. Steam-electric plant sulfur emissions stack gas control alternatives
Removal Input sulfur
Emissions source Alternative Description efficiency recovered
(percent) (percent)
Coal -burning boiler
Oil -burning boiler
Gas-burning boiler
1
2
3
1
2
3
None
Dry limestone
Wet limestone
Magnesia base
Dry limestone
Wet limestone
Magnesia base
—
35
85
95
55
85
95
~
0
0
92*
0
0
92*
—
*Some sulfur is lost in the bleed stream from magnesia base regenerator which
is used to remove Impurities.
Source: Research Triangle Institute.
99
-------
Table A.3. Sulfur emissions control cost equations for steam-electric
utilities using flue gas desulfurization
(l) ci = (370,000)
(2)
785.33
10.87
(L-A-M)0'28 (L-A-M)0'57
where:
G.J = total annualized cost of control, 10 Btu
L = boiler load factor
A = boiler annual hours
M = boiler capacity in megawatts
S = sulfur content of fuel
B1 = 9.0
B2 = 5.4
1 = coal burning
2 = oil burning
2071.4
F2
= 1240.0
162,000 (Z + B1 • S°'67- M0'02)
M °-35
0.175
L
+
F1 • S + 1.8
A
where:
Q
Cj = total annualized cost of control 10 Btu, j = 1 or 2
L = boiler load factor
A = boiler annual hours
M = boiler capacity in megawatts
S = sulfur content of fuel
B.J - 28,600 1 = wet limestone
B2 = 40,400 2 = magnesia base
F1 = 1.850
F2 = 0.440
Z = [M/150] 1f M/150 = [M/150]
[M/150] + 1 otherwise as: [ ] = "the integer part of"
Sources:
Hittman Associates, Inc., Cost Nomographs of Selected Sulfur Dioxide Abatement
Methods, Contract No. EHSD-71-43 for Environmental Protection Agency office of Air
Programs, Columbia, Md., 1972.
John K. Burchard et al., Some General Economic Considerations of/Flue Gas
Scrubbing for Utilities. Environmental Protection Agency, Research Triangle
Park, N.C., 1972.ATsF: Private communication with Gary Rochele at EPA.
100
-------
A.2.1.1 Dry Limestone Adsorption. The dry limestone adsorption
system is a low-efficiency (35 to 55 percent control) sulfur oxide
collection alternative.
The basic steps involved in this process are (1) grinding limestone
to a fine-grained powder, (2) injecting it into the combustion chamber,
and (3) collecting the reacted limestone products which then contain
sulfur along with other particulates in a high-efficiency electrostatic
precipitator.
Pilot plant operation has revealed several problems with dry limestone
removal of sulfur oxides which should be mentioned. First, the limestone
is not utilized efficiently because the sulfur oxides cannot penetrate
into the interior of the particles of lime (calcined limestone). This
problem may be accentuated by fly ash coating the surface of the lime
particles, particularly in coal-burning boilers, and by fusing of the
lime particles (dead-burning) if the grains are overheated in the boiler.
Inefficient use of the limestone can be partially overcome by injection
of additional limestone. However, this causes additional expense because
(1) more limestone must be used, (2) more fuel must be used to calcine
the limestone (releasing more sulfur oxides), and (3) additional
precipitator capacity and waste disposal handling are required.
A second problem is that under certain conditions some of the initially
collected sulfur oxides may be re-released into the gas stream after
having been adsorbed. Third, the limestone, lime, and sulfite/sulfate
become an added particulate burden. Fourth, the collection efficiency
of standard fly-ash electrostatic precipitators is reduced because sulfur
oxides, which enhance the efficiency of a precipi tor, are reduced.
A.2.1.2 Wet Lime Scrubbing. The chemical reaction involved in
wet lime scrubbing is the same as for dry limestone adsorption, namely
converting gaseous sulfur oxides to calcium sulfite/sulfate. The
reaction takes place in the scrubber, however, rather than in the combustion
chamber. The combustion gasses are ducted through a scrubber where they are
washed by slurry of water and lime or limestone, resulting in removal
of approximately 85 percent of the sulfur oxides (the factor used in this
study). Wet lime scrubbing is considerably more economical of lime than
dry limestone scrubbing, requiring as little as 10 percent above the
theoretical minimum. However, investment costs are higher for the
101
-------
(usually multiple) wet scrubbers and scrubbing solution handling equipment.
Higher costs also result from spent slurry disposal and control, added
fuel costs (or lowered generating efficiency) to restore stack gas temperatures
lowered by the scrubbing, and additional operation and maintenance costs
for the more complex system.
There are three methods of introducing lime into the scrubbing system.
In one method, limestone is injected into the boiler in the same manner
as the dry limestone process, but a scrubber is used as the collector. In
a second method, limestone is introduced into the scrubbing liquor. In a
third method, which has been assumed in this study, precalcined lime is
introduced into the scrubbing liquor. Introduction of lime into the
scrubbing liquor (slurry) reduces the amount of slurry to be handled and
utilizes a greater percentage of the calcium input values since lime is
more reactive than limestone, while avoiding the plugging and scaling
problems resulting from injecting limestone into the boiler.
The total scrubbing system includes the scrubber(s), the scrubbing
liquor handling circuit(s), the reheat system, and the associated ductwork
and piping. In addition, some sort of slurry disposal system must be
included.
The scrubber design assumed in this system is a floating bed (ping-
pong ball) scrubber. The maximum gas throughput for which these scrubbers
3
can be designed is currently 300,000 ft /min, which approximates the gas
flow of a 150 MW generating plant. Therefore, further technological
economies of scale do not exist for plants of greater than 150 MW capacity.
However, some cost savings are possible with the use of multiple units
because volume discounts are usually offered and because there is usually
some relative savings in installation costs. The scrubbing slurry handling
equipment must be designed to adjust lime usage and slurry pH for scrubbing
optimization. One slurry handling system is therefore required for each
scrubbing unit.
The flue gas reheat system is designed to raise the temperature
of the scrubbed gases to provide adequate plume rise. This can be done
with a single system for each boiler rather than a separate system for
each scrubber.
Disposal of waste slurry is assumed to be by discharge to a pond
for settling, with overflow to sewers or a watercourse. Expenses related
102
-------
to water pollution control from the pond overflow or to ultimate disposal
of the dewatered slurry waste have not been included.
A.2.1.3 Magnesia Base Scrubbing. The entire magnesia base scrubbing
system, including stack gas reheating, is identical in concept to lime
scrubbing. Some simplification of the scrubbing liquor handling system
is possible, however, because the liquor is more nearly a solution of magnesium
oxide and hydroxide, rather than a slurry such as with limestone or lime.
This factor also allows a higher control efficiency than for lime scrubbing,
typically 95 percent. However, the high cost of the magnesium oxide reagent
and the ease of regenerating it impose additional equipment requirements
for economical operation.
The first addition is a thermal decomposition system which dries the
magnesium sulfite/sulfate solution and subsequently drives off sulfur
dioxide, leaving the magnesium oxide to be reused. A small amount of
the magnesium oxide (typically 2 percent of the total throughput) is
discarded to prevent buildup of the trace impurities removed from the
gas stream during scrubbing. The purified, concentrated sulfur oxide gas
stream must then be treated in some way. Among the alternatives
which yield marketable byproducts are (1) compression and liquefaction
to product liquid sulfur dioxide for sale, (2) conversion to elemental
sulfur in a Claus unit using natural gas or some other hydrogen source,
or (3) conversion to sulfuric acid in a (contact) sulfuric acid plant.
Conversion to sulfuric acid has been chosen for this study since sulfuric
acid is a more saleable byproduct than liquid S02 and because acid
manufacture does not require a hydrogen source.
A.2.2 Fuel Switching
Sulfur emissions from fuel combustion are in direct proportion to
the sulfur content of the fuels. Since the sulfur content of coal and
residual oil can be specified by customers when purchasing these fuels,
there is the opportunity to reduce sulfur emissions by burning fuels of
lower sulfur contents than those which would be selected in the absence
of a tax on sulfur emissions. Gas has the lowest sulfur content per Btu
of the three principal fuels and is therefore a particularly attractive
alternative. However, gas supplies are currently limited due to a
103
-------
regulation-induced shortage.* Switching to low-sulfur fuels is the
only control technique usually assumed available to area source combustion.
For steam-electric power plants, fuel switching is an attractive alternative
because it does not involve the use of new technologies that may not be
completely proven.
Fuel switching costs consist of price premiums for low sulfur fuel
and, whenever a fuel is used other than that for which the boiler was
designed to burn, boiler conversion costs. In this study, a cost of $11
per billion Btu and $300 per megawatt have been used for area source and
steam-electric boiler conversions, respectively.t Estimation of the fuel
price structure under alternative tax rates would ideally involve the use
of a general equilibrium-type of model of the demand and supply of fuels.
Although development of such a model was beyond the scope of this study,
it was possible to include a reasonably complex set of price and fuel
supply conditions so that the role of prices in allocating fuels could
be explicitly included if not completely satisfactorily modeled. Recent
studies by MITREt and Battellei provided a basis for developing projections
for 1978 of maximum domestic production of coal and residual oils by
source location and sulfur content and by minimum price levels sufficient
to cover costs and to provide reasonable profit levels. The supply of
imported residual oils is assumed to be perfectly elastic. Based on these
prices plus the estimated transportation cost, cost functions have been
developed for each fuel user. The analysis of fuel switching, as an
alternative control system for steam-electric utilities and area sources,
has been based on the fuel cost estimates with consumption of domestically
produced fuels constrained by projected national production. It is
recognized that the supply functions assumed for this study do not allow
for the likelihood of shifts in supply over time due to increases in the
prices of some fuels nor do they recognize the position of the United States
*MacAvoy, Paul W., "The Regulation-induced Shortage of Natural Gas,"
Journal of Law and Economics, Vol. 14, No. 1 (April 1971), pp. 167-98.
tDerived from data presented in Ehrenfeld, J.R. et al, Systematic
Study of Air Pollution from Intermediate-size Fossil-fuel Combustion
Equipment, Wai den Research Corp., Cambridge, Mass., July 1971. /'
fMITRE Corporation, Survey of Coal Availabilities by Sulfur Content,
May 1972.
iBattelle Memorial Institute, Energy Quality Model—1972, unpublished.
104
-------
as a consumer of the world's oil supply. However, over the relatively
short time frame involved and considering the scope of this study, they
represent a reasonable set of working assumptions.
A.2.2.1 Coal. Coal is the major fuel for steam-electric power
generation. The production of coal is from coal reserves that vary with
respect to location, characteristics, and amount. Although total coal
reserves are estimated to be about 2.9 trillion tons, only a small part
of this total is precisely located ("known") and recoverable with current
technology. These currently known recoverable reserves, estimated to be
50 percent of known reserves, are as follows: anthracite, 6,368 million
tons; bituminous, 130,755 million tons; subbituminous and lignite, 59,931
tons.*
Known recoverable reserves of coal are unevenly distributed among
the major coal-producing regions. The region with the largest share of
these reserves is the Northern Rocky Mountain region (table A.4) which
is, of course, located long distances from the nation's major population
centers where most coal is consumed.
The sulfur content of coal, which is critical in determining emissions
from coal combustion, varies by type of coal. All anthracite and most sub-
bituminous coal and lignite reserves have low sulfur content (1 percent
or less sulfur by weight). Almost one-half of all recoverable coal reserves,
however, is bituminous coal whose sulfur content varies considerably. Most
V
of the bituminous reserves (about 57 percent) are of high sulfur content
(over 2.0 percent). Less than one-third of the reserves (about 30 percent)
is of low (1 percent or less) sulfur content. The remainder of the bitumi-
nous coal reserves (about 13 percent) is estimated to be of medium (1.1 to
2.0 percent) sulfur content.
In 1970, the more than 5,000 active bituminous coal and lignite mines
operating at an average of 82 percent capacity produced about 603 million
tons.t However, not all of the 603 million tons are available for consumption
by the steam-electric plants and area sources for the following reasons:
(1) Most coal (71 percent) is cleaned to reduce impurities and enhance
the Btu content; as a result about one-fourth of this tonnage becomes
*MITRE Corporation, Survey of Coal Availabilities by Sulfur Content,
May 1972, pp. 18-19.
tMinerals Yearbook, Vol I, U.S. Department of Interior, Bureau of
Mines, p. 329.
105
-------
Table A.4. Geographic distribution of known
recoverable reserves of coal
CoaLproducIng region
Northern Appalachian 16
Southern Appalachian 12
Eastern Interior 62
Western Interior 10
Northern Rocky Mountains 80
Southern Rocky Mountains 17
West Coast 1
Total 198
Source: MITRE Corporation,Survey of Coal Availa-
bilities By Sulfur Content, p. 18.
refuse. (2) Some coal is exported. (3) Finally, some coal is used
to produce coke for blast furnaces and for the self-generation of power
at steel and rolling mills. These so-called metallurgical coals, usually
captive to the consuming industries, must have a property known as caking
(or coking) and must be of low sulfur and ash content. As a result,
they carry a price premium arid should not be considered as possible fuels
for consumption by the steam-electric power plants and the area sources.
After all these adjustments have been made, MITRE estimates the production
of commercial steam coal in 1970 to have been about 363 million tons.*
Table A.5 shows the estimated 1970 production of commercial coal,
which is all noncaptive steam coal, by region and sulfur content. The
regions east of the Mississippi River dominate production of all coals;
however, western reserves of low sulfur coal are many times greater
than those in the eastern regions.
Table A.6 shows maximum 1978 coal production levels. These estimates
are given by region and sulfur content and assume current production costs
*MITRE Corporation, Survey of Coal Availabilities by Sulfur Content,
May 1972, pp. 29-34.
106
-------
Table A.5.
Production of commercial bituminous and subbituminous coal--1970
(millions of tons)
Region or basin
Northern Appalachian
Southern Appalachian
Eastern Interior
Western Interior
Northern Rocky Mountains
Southern Rocky Mountains
West Coast
Total United States
5.0
0
17
0
0
12
12
0
•^••PM
43
.7
.06
.00
.96
.33
.3
.7
.4
••••^•B
.75
Source: MITRE Corporation,
0.8-1.0 1
1.07
46.00
0.19
0.0
1.7
0.1
49.06
Survey of
1.1-1.5
4.1
16.0
7.8
0.34
0.0
0.0
0.0
28.24
Sulfur content (by weight, dry basi
1.6-2.0 2.1-2.5 2.6-3.0 3.1-3.5
31
14
9.2
0.12
0.0
0.0
0.0
54.32
18
4.3
3.7
0.14
0.0
"0.0
0.0
26.14
14
0.22
7.4
0.56
0.0
0.0
0.0
22.18
7.8
2; 2
47.0
0.56
0.0
0.0
0.0
57.56
Coal Availabilities by Sulfur Content, May
Table A. 6. Projected maximum production of commercial bituminous co<
(millions of tons)
Region or basin
Northern Appalachian
Southern Appalachian
Eastern Interior
Western Interior
Northern Rocky Mountains
Southern Rocky Mountains
West Coast
Total United States
10
0
26
1
0
49
34
1
113
.7
.07
.4
.04
.36
.17
.82
.43
.29
Source: Developed from data
0.8-1.0 1
1.92
54.98
0.85
0.33
31.86
19.92
0.37
110.23
in Survey
Sulfur
.1-1.5
10.85
20.19
13.71
0.82
0.05
0.04
0.0
45.66
of Coal.
content
1.6-2.0
36.94
16.03
14.49
0.35
0.48
0.38
0.0
68.67
s)
3.6-4.0 > 4.0
5.4
0.0
51.0
0.64
0.0
0.0
0.0
57.04
1972, pp.
il— 1978
(percent, by weight, dry basis)
2.1-2.5 2.6-3.0 3.1-3.5 3.6-4.0 >
26.05
4.85
8.17
1.97
0.11
0.0
0.0
41.15
A vaU abilities by
17.18
1.17
27.22
0.94
0.05
0.0
0.0
46.56
10.26
2.2
86.56
2.78
0.01
0.0
0.0
101.81
7.46
0.01
102.68
6.41
0.0
0.0
116.56
0.45
0.0
19.0
5.2
0.0
0.0
0.0
24.65
26-34.
4.0
0.69
0.03
40.25
17.48
0.16
1.00
0.0
59.61
Total
81.88
99.72
146.06
8.08
12.3
14.4
0.5
362.94
Total
111.42
125.86
294.93
31.45
81.89
56.16
1.80
703.51
Sulfur Content, MITRE Corporation,
Way 1972, pp. 26-34.
-------
NORTHERN
APPALACHIAN
SOUTHERN
APPALACHIAN
Figure A.I. Coal-producing districts (Source: Department of the Interior).
They are based on the assumption that production growth will not occur
unless it is maintainable for at least 20 years.*
These projections of 1978 production levels for the major coal-
producing regions have been converted to Btu's. Projections of coal
supply used by Battelle were employed here to derive proportions and
the values in table A.5 across the 19 coal-producing districts (fig. A.I).
It would require 4 years to attain the indicated levels in the eastern
regions and 3 years in the western regions.t For purposes of this
analysis, they have been used to provide upper limits on coal supplies.
Coal prices will influence fuel .selections by the fossil fuel
combustion sources. As shown in figure A.2, coal prices have recently
been rising rapidly. In part, these price increases may be due to
*MITRE Corporation, Survey of Coal Availabilities by Sulfur Content,
May 1972, p. 79.
tMITRE Corporation, Survey of Coal Availabilities by Sulfur Content,
May 1972, p. 81.
108
-------
1950 1952 1954 1956 1958 I960 1962 1964 1966 1968 1970
YEAR
Figure A. 2. Bituminous coal price trends (Source: Department of Labor).
increases in the demand for low sulfur coals as a result of air quality
regulations. For this study it is desirable to have sets of fuel prices
by sulfur content and location which reasonably reflect supply conditions
from which to estimate fuel cost increases due to fuel switching.
Low sulfur coals are in limited supply in the Eastern United
States, most deposits being in the Southern Appalachian Region. Western
low sulfur fuels are more abundant but may have high delivered costs
due to their distance from markets and their lower Btu content.
Initial estimates of the prices of all coals were derived from the
Battelle study.* As they explain, however, their prices were minimum
price estimates. Many prices may be bid upward in the presence of air
quality regulations, regardless of whether emissions standards or taxes
are used to induce compliance.
*Battelle Memorial Institute, Energy Quality Model--1972, unpublished
109
-------
Using these estimates as a floor below which prices are not expected
to go, the emissions response model was run to derive projections of
steam-electric utilities demand for each coal type. Fuel demand and
supply for each sulfur content were then compared on a national basis.
When demand exceeded supply, the regions contributing to the excess were
identified from the disaggregated district tables. Prices of fuels for
which the demand exceeds supply were incremented and the model was rerun.
Several iterations were required, until a price structure was developed
that reflected market conditions. It is recognized that this approach
is quite simplistic when compared with a more sophisticated approach
such as linear programming (which would have resulted in a unique price
structure). Such an approach, however, would have been beyond the scope
of this study. For example, in this model each utility can select from
among four control alternatives, (i.e., three hardware and one no-hardware),
nine sulfur contents, and 19 regions for coal alone; a total of over 500
combinations exist. There are about 1,000 utilities. The linear programming
model implied by the size of the model would have been substantial. Yet
it was recognized that (1) the demand for these fuels was interrelated,
(2) each utility faces a large number of alternatives (not merely "low
sulfur" fuel as implied in many studies), and (3) relative fuel prices
will be critical in the selection from among control alternatives. The
approach employed in this study as discussed above does offer the oppor-
tunity to incorporate a fair degree of sophistication into the analysis
of the response of fuel combustion sources to an emissions tax. The
resulting price levels are presented in table A.7. For ease of calculation,
an average sulfur content was used rather than a range as was presented
earlier.
A.2.2.2 Oil. Distillate and residual fuel oils are important
sources of energy for steam-electric utilities and for area heating
sources. Domestic supplies of those fuel oils are functions of the
supply of crude and production decisions on the proportion of the total
that is to be refined into fuel oil versus other petroleum products. An
increase in the share going to gasoline, for example, tends to reduce the
availability of fuel oil.
Domestic supplies of residual and distillate oil have been insufficient
to meet demand in recent years, as shown in figure A.3. As a result,
110
-------
CRUDE
PETROLEUM
PRODUCTION
U.S. (Left scole]
IMPORTS
(Right scale)
1950 1952 1954 1956 1958 I960 1962 1964 1966 1968 1970
YEAR
Figure A.3. Sources of supply of petroleum (Source: Department of
Interior).
950
DISTILLATE
CONSUMPTION
(Leff scale)
NET IMPORTS
(Right scale)
950 1952 1954 1956 1958 I960 1962 1964 1966 1968 1970
YEAR
Figure A.4. Distillate consumption (Source: U.S. Department of
Interior).
Ill
-------
Table A.7. Coal prices by district
(dollars per billion Btu)
Coal
district
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
^_ A _.
Sulfur content (percent)
0.7
424
430
414
385
395
385
420
460
422
421
422
422
385
385
385
385
395
379
360
Source:
i ^ **^
0.9
394
375
340
352
373
364
415
390
385
390
412
385
343
343
333
333
358
325
315
Adapted
'A
1.3
334
340
291
316
331
315
340
330
364
355
364
364
342
334
331
330
355
320
—
1.8
300
301
264
284
305
275
323
300
330
311
341
330
342
323
330
326
353
318
—
2.3
273
300
' 239
280
282
254
261
259
323
276
320
301
341
___
___
—
348
316
—
2.8
269
298
238
275
280
249
247
235
284
275
309
299
341
___
___
___
348
314
—
from Battell e Memorial Institute,
3.3
249
245
237
270
279
243
224
234
270
275
289
298
341
—
—
—
348
313
—
EPA Energy
3.8
247
240
217
265
278
240
215
227
267
274
287
269
—
—
—
—
—
—
—
Quality
5.0
246
235
216
260
277
243
210
225
266
273
286
261
341
320
325
300
351
,313
—
Model,
imports of both crude and refined oils have increased and do supply a
substantial share of the market. Table A.8 shows the 1970 sources of
residual fuel oil. The importance of imports and their country of
origin are shown. The data may understate the role of the Middle East
and overstate that of Europe, however, since some Middle East crude is
refined in Europe and then shipped to the United States. Trends in the
share of crude and residual oil imports from the Middle East are shown
in Table A.9. A recent study* estimated that the import share of domestic
consumption of fuel oil will double between 1970 and 1980.
A.2.2.2.1 Distillate. Distillate production represents a declining
but substantial share of domestic oil production (figure A.4). While
some distillate is used by steam-electric utilities for peak load power
generation, it represents only a small energy source for these utilities.
However, distillate is an important source of energy for the area sources
where it is used as a power generating fuel and, more commonly,( as a
heating fuel.
*National Petroleum Council, U. S. Energy Outlook, Vol. 2, November
1971, p. 57.
112
-------
Table A.8. Sources of residual fuel oil, 1970
(thousand barrels)
Domestic sources of supply 266,228
Production 257,510
Crude used directly as residual 4,317
Reductions in stocks 4,401
Domestic consumption 804,287
Exports 19,786
Imports 557,845
North America (Canada and Mexico) 14,147
Central America and Caribbean 261,299
South America 220,842
Europe 58,251
Middle East 1.554
Africa 1.616
Asia 136
Source: Minerals Yearbook, U.S. Department of the Interior, Bureau
of Mines, 1970
Table A.9. Crude and residual oil imported
into the United States from the Middle East
Year
1967
1968
1969
1970
1971
"Source:
Crude oil
Residual oil
Amount Share of Imports Amount Share of imports
(thousand (percent) (thousand (percent)
barrels) barrels)
67,977
72,330
61,616
61,892
124,155
Minerals Yearbook,
16.5
15.3
12.0
12.8
20.2
U.S. Department
5,421
4,431
4,872
1,554
4,226
of the
1.4
1.1
1.1
0.3
0.7
Interior, Bureau
of Mines.
113
-------
No attempt has been made in this study to assess future supplies of
distillate. The assumption has been made that distillate will be available
in quantities sufficient to meet projected demands. For this study, it has
been assumed that only area sources consume distillate fuel. The sulfur
content is 0.10 pounds per million Btu regardless of origin, and consumers
are assumed to have perfectly inelastic demand for distillate. Thus, it
has not been necessary to develop any structure of prices since, for any
tax rate, consumption will remain constant and tax payments made on emissions,
A.2.2.2.2 Residuals. Domestic production of residual oil provides
only about one-third of consumption (fig. A.5). Residual fuel oils are
a significant source of energy for steam-electric utilities. In 1970 they
accounted for about 15 percent of total Btu's consumed.* While domestic
production is limited, the supply of foreign residuals is assumed to be
perfectly elastic.
MITREt has provided projections of domestic residual oil production
for the 5 PAD districts* (table A.10). In order to use the Battelle price
and transportation data, these projections were distributed among 12 oil-
producing regions (fig. A.6) using proportions derived from the Battelle
study.
Minimum prices by oil-producing regions were presented in the Battelle
study.
Initial runs of the emissions model indicated that some Battelle
prices were not high enough to ration projected production. These prices
were raised in iterative fashion as they were for coal until quantity
demanded was reduced to levels more consistent with projected production.
The resulting price structure is presented in table A.11. Again it is
recognized that this approach is a considerable simplification of the
market-adjustment process that may actually take place. Yet the. procedure
does provide a set of prices from which relative prices can be obtained
*National Coal Association, Steam-Electric Plant Factors, 1971,
Washington, D. C., 1972, p. 52.
tMITRE Corp., Survey of Coal Availabilities by Sulfur Content. May 1972.
tThese refining districts correspond with groupings originated by the
Petroleum Administration for use during World War II which were called PAW
districts. The PAW districts were later changed to PAD (Petroleum Adminis-
tration for Defense) districts.
114
-------
850
CONSUMPTION
{Leftscale)
NET IMPORTS
(Right scale)
1950 1952 1954 1956 1958 I960 1962 1964 1966 1968 1970
YEAR
Figure A.5. Residual fuel oil consumption (Source: U.S. Department of the
Interior).
Figure A.6. Oil-producing regions (Source: U.S. Department of the Interior)
115
-------
Table A.10. Projected maximum U.S. production of residual oils—1978
(billion Btu)
Sulfur content (percent)
District < 0^4 Ot4.0>82 0.83-1.65 1.66-2.94 > 2.94 Totals
I:
II:
III:
IV:
V:
East
Midwest
Southeast-Southwest
Rocky Mountain
West
Total
8.
121.
854.
97.
39.
1121.
6
1
6
0
8
1
0
104.
530.
16.
40.
690.
0
8
0
1
9
0.
60.
108.
32.
109.
311.
1
4
4
9
3
1
0
31.
207.
9.
121.
269.
8
0
7
3
8
0 0
13.9
0
34.8
45.2
93.9
8.7
331.2
1700.8
190.4
355.7
2586.8
Source: MITRE Corporation, Survey of Coal Availabilities by Sulfur
Content, p. 18.
Table A.11. Oil prices by region
(dollars per billion Btu)
Oil
region
1
2
3
4
5
6
7
8
9
10
11
12
Gulf Coast
Great Lakes
West Coast
East Coast
0.4
445
444
540
545
517
505
517
517
499
489
495
489
445
517
499
517
Sulfur
0.6
377
376
459
459
470
NA
NA
NA
422
412
424
412
377
470
422
490
content
1.2
338
350
399
410
435
NA
NA
NA
375
375
378
365
338
435
375
390
(percent)
2.3
291
332
356
360
415
NA
N4
NA
336
335
354
320
291
415
336
313
3.0
290
331
351
352
406
NA
NA
NA
302
321
345
315
290
406
302
266
Source: Adapted from Battelle Memorial Institute, EPA Energy
Quality Model, September 1972.
116
-------
to provide more realistic projections of the cost of control for fuel
switching than would a single premium for low sulfur oil.
A.2.2.3 Gas. Gas is an important energy source for steam-electric
utilities and the area sources. In addition, gas is also a competitor of
electricity, primarily in the commercial and residential heating markets.
No attempt has been made in this study to project the future supply
of gas. It has been assumed that sufficient supplies of gas will be
available to current users of gas. However, no sources are permitted to
switch to gas from coal or oil.
A.2.2.4 Transportation Costs. Transportation costs are an important
element in the delivered prices of coal, oil, and gas. In the future, they
may assume even larger importance than today since large reserves of low
sulfur coal, while available, are located long distances from the popu-
lation centers where most fuel is consumed. It is necessary, therefore,
to incorporate transportation costs into the model that predicts fuel
prices to be faced by users.
The transportation matrix developed by Battelle for EPA has been
used. This matrix estimates shipping costs from 19 coal-, 12 oil-, and
12 gas-producing regions to each State. The location of each of the
nation's steam-electric utilities has been identified in terms of one
of these destinations. The transportation costs are approximations of
actual costs based on the typical modes used for shipping each fuel,
their per mile costs, and distance from origin to destination.
These costs were put on a Btu basis to make them comparable with
other fuel costs. The average Btu contents of residual oil and gas used
in this study are shown in table A.12.
Table A.12. Oil and gas average Btu contents
Fuel
Average
Btu content
(millions)
Fuel content
Residual oil
Gas
6.287
1.030
Per barrel
Per million
cubic feet
Source: Research Triangle Institute.
117
-------
The Btu contents of coal vary significantly among origins, especially
between the Appalachian and Rocky Mountains. For this reason, it was
necessary to adjust the coal transportation costs or!gin-by-origin. The
coal Btu contents are presented in table A.13.
A.2.3 Emissions Reductions and Costs
Cost of control functions for all steam-electric power plants and
area sources have been developed based on the costs and effectiveness of
the control alternatives costed above, a listing of the nation's utilities
which includes relevant parameters for emissions and control cost estimation
and State-by-State demands for Btu's by the area sources. These functions
are minimum cost functions for achieving sulfur emissions reductions from
these two sources.
The long-run industry total and marginal costs of emissions reductions
are shown in figures A.7 through A.10. The continuity of these estimated
functions is probable, due to the large number of sources and control alter-
natives.
Table A.I3. Coal average Btu contents by origin
Coal -producing region Average Btu
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
!9
Source: MITRE Corporation, Survey
per ton
27
27
24
25
28
27
24
23
23
23
27
26
23
19
20
19
21
19
17
of Coal
(millions)
Availabilities
by Sulfur Content, May 1972, pp. II-2 through II-5.
118
-------
o
fc I
i- d
(/> o
O o
o
UJ
N
2.4
2.2
2.0
1.8
1.6
1.4
L2
1.0
0.8
0.6
0.4
0.2
LTCJ
34567
REDUCTIONS IN SULFUR EMISSIONS
(MILLION TONS PER YEAR)
8
10
Figure A.7. Total cost* of reductions in sulfur emissions: steam-electric
utilities (*does not include emissions tax payments) (Source: Research
Triangle Institute).
345678
REDUCTIONS IN SULFUR EMISSIONS
(MILLION TONS PER YEAR)
10
Figure A.8. Marginal cost* of reductions in sulfur emissions: steam-electric
utilities (*does not include emissions tax payments) (Source: Research
Triangle Institute).
119
-------
$
L
N
i
Z I«S^ 1
" 12345
REDUCTIONS IN SULFUR EMISSIONS
(MILLION TONS PER YEAR)
Figure A.9. Total cost* of reductions in sulfur emissions: area sources
(*does not include emissions tax payments) (Source: Research Triangle
Institute).
'LMC
12345
REDUCTIONS IN SULFUR EMISSIONS
(MILLION TONS PER YEAR)
Figure A.10. Marginal cost* of reductions in sulfur emissions: area sources
(*does not include emissions tax payments) (Source: Research Triangle
Institute).
120
-------
Appendix B: SULFUR EMISSION FACTORS AND CONTROL COSTS FOR
PETROLEUM REFINERIES
This appendix describes the emissions factors used and the control
alternatives included and costed as part of this study to project the effect
of various sulfur tax rates on petroleum refineries.
B.I Emission Factors
A variety of processes or operations in a petroleum refinery may
produce sulfur dioxide ($02) emissions. However, several minor refinery
process sources represent only a small (and generally uncertain) fraction
of total refinery S0« emissions. On the other hand, three major refinery
operations produce significant S0« emissions and are treated as separate
emissions sources: (1) catalyst regenerators, (2) Claus sulfur recovery
plants, and (3) fuel combustion processes; i.e., heaters and boilers. As
indicated in the estimated 1970 refinery sulfur balance shown in table B.I,
the SOg emissions from these three operations account for 5.5, 4.5,
and 3.3 percent, respectively, (for a combined total of about 13 percent)
of the sulfur present in the input crude oil. About 78 percent of the
input sulfur is distributed among marketed products (45.9 percent), recovered
sulfur (26.7 percent), and waste water effluent (5.7 percent); the remaining
8.4 percent is unaccounted for and represents either S02 emissions by other
refinery operations for which emission factors are unknown, or the result
of using an unrepresentative value for the average sulfur weight percentage
of the input crude oil.
B.I.I Catalyst Regenerators
Catalysts used in catalytic crackers lose some of their activity
after extended use and must be either regenerated or replaced. The
regeneration process consists of oxidizing coke—which forms on the catalyst
during cracking—to carbon monoxide. During regeneration, sulfur and
sulfide deposits which also accumulate on the catalyst are oxidized to SO^.
Thus, catalyst regenerators, particularly those associated with Fluid
Catalytic Cracker (FCC) units, are a major source of refinery sulfur
emissions. As indicated in table B.2, sulfur emission rates from
regenerators used with Thermofor Catalytic Cracking (TCC) units
are considerably smaller.
121
-------
Table B.I.
Estimated U.S. petroleum refinery sulfur
balance—1970
Barrels Average sulfur
(thousand)* weight (percent)
Sulfur content
(thousand tons)
Input
Cru<
rude oil
3,967,500
O.Slt
4,747
100
Disposition
Products marketed
Gasoline
Kerosene
Jet fuel
Distillate oilfl
Residual oiln
Petroleum coke
Asphalt
Water effluenti
Sulfur recovered
Refinery emissions§§
Residual oil burned
Claus plant
Fluid catalytic cracker
(FCC) regenerators!
Thermofor catalytic cracker
(TCC) regenerators?
Total
Unaccounted for
2,100,000
96,000
302,000
897,000
206,667
11,300#
26,500#
...
43,323
—
—
---
__.
...
0.03
0.05
0.3 tt
0.3 **
1.6 **
1.8 §
3.0 §
...
—
—
1.8
—
—
...
___
2,178
95
128
408
533
212
795
270
1,269
633
159
212
256
6
4,350
397
45.9
5.7
26.7
13.3
3.3
4.5
5.4
0.1
91.6
8.4
* Mineral Industry Survey, U.S. Department of Interior, December 23, 1971.
t Estimated average for domestic and imported crude processed in U.S. refineries.
t From catalyst coke burning.
§ Estimated.
fl Excludes Imports of 557,000 thousand barrels.
# 103 Ton (coke 5.0 bbl/ton; asphalt 5.5 bbl/ton).
**OAP Data file of Nationwide Emission for 1970, July 1972.
ttDomestlc airline specification; actual may be lower.
tf'Sulfur Content of Crude Oils of the Free World," Bureau of Mines, RI 7059, 1967.
§§Research Triangle Institute.
122
-------
Table B.2. Sulfur emission factors for petroleum refineries
Sulfur emission factors
Emissions source (Ib/thousand barrels of fresh feed
or oil burned, as appropriate)
Catalyst regenerators
Fluid catalytic cracker 262
Thermofor and Houdriflow
catalytic cracker 30
Refineries without Claus plants 720
Refineries with Claus plants 72
Fuel combustion 68
Source: Compilation of Air Pollutant Emission Factors, -
Environmental Protection Agency, Research Triangle Park, N.C.,
February 1972, pp. 9-1, 9-2.
B.I.2 Claus Plants
Many refinery processes produce off-gases which contain hydrogen
sulfide (H«S). All plants strip the H/>S (usually in excess of 95 percent)
from the off-gases before they are burned in process heaters and boilers.
If the refinery does not have a Claus plant to convert the stripped H^S
to sulfur, the H2S stream is flared to the atmosphere and produces large
amounts of S0«. It has been assumed that an average 2-stage Claus plant
can provide about a 90-percent conversion of the input H2$ to elemental
sulfur, with the remaining unconverted sulfur being emitted as S02. An
average 4-stage Claus plant is assumed to provide upwards of 95-percent
conversion of H2S. Recently developed "tail gas modifications" of the
4-stage Claus unit provide a 99.9-percent conversion of input HgS to elemental
sulfur.
At present, not all refineries have Claus plants, but those that do
generally have 2-stage plants.
B.I.3 Fuel Combustion
Much of the fuel required by refinery process heaters and boilers is
produced by the refinery itself. Most of the S02 emissions from refinery
123
-------
combustion sources result from the use of liquid fuels such as low value
distillate and residual oils. Because of their relatively high sulfur
concentrations, these fuels are frequently unsuitable for marketing. As
indicated in table B.I, more than 43 million barrels of residual fuel oil
were burned in U.S. refineries during 1970. Using an estimated sulfur
concentration of 1.8 percent by weight, residual fuel oil combustion in
refineries resulted in the release of about 159,000 tons of sulfur to the
atmosphere during 1970. Combustion of refinery gases also results in S02
emissions; however, these gases are generally scrubbed for removal of
sulfur values prior to burning and thus produce relatively little S02 1n
comparison with residual fuel oil combustion.
Sulfur emissions factors for the three refinery sources Identified
above were derived from the data shown 1n table B.I, and are shown in
table B.2.
B.2 Sulfur Emissions Control Alternatives and Costs
Costs as a function of capacity have been estimated for controlling sulfur
emissions from the three emission sources discussed above. In most cases,
several techniques for controlling emissions at each source—each technique
representing a unique level of control—have been costed. The control cost
estimates were based on previous studies for EPA and private communications
with EPA personnel. Table B.3 summarizes the alternatives available for
controlling sulfur emissions which appear to be most feasible by 1978, and
their removal efficiencies by emissions source.
B.2.1 Catalyst Regenerators
Hydrodesulfurization of catalytic cracker feedstock was costed as the
most economical technique for reducing sulfur emissions from catalyst
regenerators. Catalytic crackers are fed by vacuum gas oil from vacuum
distilling units and/or heavy gas oil from atmospheric crude topping units.
Hydrodesulfurization processes now in commercial use are effective 1n
removing up to about 95 percent of the sulfur 1n these oils. Since most of
the sulfur present 1n catalytic cracker feedstock 1s passed on to the cracked
products and only a small fraction is picked up by the catalyst and emitted
during catalyst regeneration, both product desulfurlzatlon and reduction in
regenerator emissions will result from application of this control technique.
Hydrodesulfurization of vacuum gas oil and heavy gas oil Involve high-
temperature, high-pressure hydrogen treatments in the presence of a catalyst.
124
-------
Table B.3. Sulfur emission control alternatives for
petroleum refineries
Emissions source
Catalyst regenerator
FCC
TCC & HCC
Refineries without
Claus plant
Refineries with
Claus plant
Abbrevia-
tion*
PR-CR
PR-CR
PR-S
PR-S
PR-S
PR-S
PR-S
Removal
Alterna- Control" alternative efficiency
tlve (percent)
1
2
1
2
3
4
5
HydrodesulfuHzation of catalytic
cracker feedstock
Two- stage Claus plant
Four-stage Claus plant
Four- stage Claus plant with
tall gas unit
Add two additional conversion
stages to Claus plant
Add two additional conversion
stages to Claus plant plus
tall gas unit
90
90
90
95
99.5
95
99.5
Fuel oil combustion PR-FC 1 Hydrodesulfurization of residual 90
fuel oil
*Abbrev1at1ons Index to Table B.4.
Source: Research Triangle Institute.
The hydrodesulfurization process generally employs a fixed bed catalytic
reactor and a regenerate catalyst. Sulfur compounds are converted to H2S
in the reactor and drawn off with other overhead gases. The H2S is subsequently
separated from the other gases by amine scrubbing and directed to a sulfur
plant for conversion to elemental sulfur. The liquid bottoms constitute the
desulfurized product which, in this case, will be fed to the catalytic cracker.
In addition to sulfur removal, hydrogen treating of catalytic cracker feedstock
will reduce coke production in the cracker by 25 to 30 percent and increase
gasoline yield. In the present analysis, it was assumed that the hydrodesulfu-
rization process removed 90 percent of sulfur present in the cracker feedstock
and effects a similar reduction in sulfur emissions to the atmosphere from the
catalyst regenerator.
Annualized costs for several refinery capacities are presented in table B.4
along with emissions data.
125
-------
Table B.4. Sulfur emission control costs for petroleum refineries
Con
altemi
by u
proc
PR-Cf
PR-C!
PR-S
PR-S
PR-S
PR-S
PR-S
PR-R
tn
tl
nl
es
ili
I
3
4
5
>1
t
s
1
Control costs ,
(thousand dollars)
Investment
5.000* 50,000* 100.000* 500,000*
$122 $ 560 $ 900 $2.500
122 560 900 2,500
170 430 600 1,460
180 500 720 1.900
360 1,000 1.420 3.800
10 46 120 440
160 380 720 1.900
30 90 135 375
mnuallzed total
cost of control
5,000* 50.000* 100.000* 500.000*
$432 $2.920 $5.200 $24.000
432 2.920 5.200 24.000
23 82 121 421
31 108 160 555
65 227 337 1.168
8 26 39 134
34 119 177 613
59 433 806 3.539
Annual missions
after control
(tons)
5.000* 50,000* 100.000*500,000'
9 84 169 840
11 14 29 143
65 649 1.300 6.495
32 325 650 3,248
1 5 13 65
32 325 650 3,248
1 5 13 65
6 59 119 594
Annual additional recovered
sulfur after control
(tons)
5.000* 50,000* 100.000* 500,000*
532 5,325 10,698 53.250
432 5,325 10,698 53.250
585 5,846 11,691 58,459
618 6.170 12,341 61,706
649 6.490 12,978 64,889
33 324 650 3.247
64 644 1.287 6,430
53 535 1.070 5,346
•Plant capacity, barrels per day.
Source: Developed by Research Triangle Institute for this study from data presented 1n:
1. Research Triangle Institute, Control Technology for Sulfur Oxide Pollutants. 2nd ed.. November 20, 1972.
2. Aalund, L.. Hydrodesulfurlzatlon Technology Takes on the Sulfur Challenge.* 011 and Gas Journal. September 11, 1972, p.79.
3. Barry. B.B.. 'Reduce Claus Sulfur Emission," Hydrocarbon Processing. April 1972. p. 102.
4. "Characterization of Claus Plant Emissions," Preliminary draft of final report prepared for the Environmental Protection Agency by Process
Research, Inc., Cincinnati, Ohio, September 1972.
As shown in the table, a refinery with a capacity of 5,000 bbl/d could
reduce fluid catalytic cracker emissions to 9 tons of sulfur annually with an
annualized cost of $432,000. The precontrol level is 90 tons annually (i.e.,
9 •*- (1 - removal efficiency of 90 percent)). A refinery would control sulfur
emissions from this source when the cost of control plus the tax rate times
the remaining emissions was less than the tax times uncontrolled emissions.
For this example refinery, this condition would obtain for tax rates in excess
of 267 cents per pound of sulfur emissions since:
$432,000 + X (9) = X (90)
X - $5,333 per ton or $2.67 per pound
where
X = tax rate per ton of sulfur emissions.
B.2>2 Claus Plants
Annualized costs for Claus plants were prepared for two cases:
(1) In refineries where there is no present Claus plant, there
are three possible alternatives: (a) a 2-stage Claus plant,
(b) a 4-stage Claus plant, or (c) a 4-stage Claus plant with
a tail gas unit.
(2) In refineries where there is an existing 2-stage (assumed)
Claus plant, there are 2 possible alternatives: (a) addition
126
-------
of two additional conversion stages to the existing plant, or
(b) the addition of 2 or more conversion stages plus a tail
gas unit.
Annualized costs that reflect these alternatives are presented in
table 6.4 along with emissions data.
B.2.3 Fuel Combustion
Direct desulfurization of residual fuel oil was selected as the most
economical technique for controlling sulfur emissions from refinery combustion.
Hydrodesulfurization processes applicable to residual fuel oils are similar
to those used with vacuum gas oil except that they are more severe and require
highly selective, well-designed catalysts which are resistant to metal deposits
Unlike catalysts used in vacuum gas oil hydrodesulfurization, these catalysts
normally cannot be regenerated and must be replaced. There are a variety of
direct hydrodesulfurization processes used commercially to reduce the sulfur
content of marketed residual fuel oils. Sulfur reductions in the 80- to
95-percent range can be achieved by these processes. Despite recent advances,
residual fuel oil desulfurization processes are still relatively expensive and
are used only when final fuels must be low in sulfur or when the feedstock is
of high sulfur content. Because of the high costs involved, desulfurization
of residual fuel oil consumed by refineries is not now being practiced.
Residual fuel oil hydrodesulfurization is assumed to remove and recover
90 percent of the sulfur present in the oil, thereby reducing sulfur emissions
from refinery residual fuel oil combustion sources to 10 percent of the
uncontrolled level. Annualized costs for residual fuel oil desulfurization
are presented in table B.4 along with emissions data.
B.2.4 Emissions Reductions and Costs
Industry cost of control functions have been developed based on the
costs and effectiveness of the control alternatives costed above and a
listing of the nation's petroleum refineries which includes relevant process
parameters for emission and control cost estimation. These functions are
minimum cost functions for achieving sulfur emissions reductions from the
petroleum refining industry. Table B.5 summarizes the refinery data.
The long-run industry total and marginal costs of sulfur emissions
reductions are shown in figures B.I and B.2. The total costs (LTC) increase
at an increasing rate throughout the range for which data are available. The
marginal costs (LMC) increase with increases in emissions reductions. Beyond
reductions of 250,000 tons, the marginal cost function rises quite rapidly.
127
-------
Table B.5. Size distribution of petroleum refineries
(number of refineries)
Capacity
(barrels per day)
000
0-10
11 - 20
21 - 30
31 - 40
41 - 50
51 - 60
61 - 70
71 - 80
81 - 90
91 - 100
101 - 110
111 - 120
121 - 130
131 - 140
141 - 150
151 - 160
161 - 170
171 - 180
181 - 190
191 - 200
201 - 220
221 - 240
241 - 260
261 - 280
281 - 300
301 - 325
326 - 350
351 - 375
376 - 400
401 - 425
426 - 450
Crude
91
38
27
17
21
11
6
9
13
1
4
3
1
5
1
5
3
2
5
2
1
2 .
2
1
Catalytic cracking
Thermofor &
Fluid Houdriflow
16 11
33 10
23 8
14 4
10
9
6
4
2
1
2
1 1
1
1
2
2
1
Source: The Oil and Gas Journal. April 6, 1970, pp. 121-41.
128
-------
o
tr.
o
o
30
25
(0
v> <
O -I
O -I
o
gz
g2
*!">
UJ 5
N *-*
J 5
I I I I
I I I i
i I I
'LTC
i I j I
«* 0 50 100 150 200 250 300
i
REDUCTIONS IN SULFUR EMISSIONS (THOUSAND TONS PER YEAR)
Figure B.I. Total cost* of reductions in sulfur emissions: petroleum
refineries--1978 (*does not include emissions tax payments) (Source:
Research Triangle Institute).
50 100 ISO 200 250 300
< REDUCTIONS IN SULFUR EMISSIONS (THOUSAND TONS PER YEAR)
Figure B.2. Marginal cost* of reductions in sulfur emissions: petroleum
refineriesr-1978 (*does not include emissions tax payments) (Source:
Research Triangle Institute).
129
-------
Appendix C: SULFUR EMISSION FACTORS AND CONTROL COSTS
FOR SULFURIC ACID PRODUCERS
This appendix describes the emission factors used and the control
alternatives included and costed as part of this study to project the effect
of various sulfur tax rates on sulfuric acid plants,
C.I Emission Factors
Sulfur emissions from the production of sulfuric acid by the contact
process are of two types. One type is gaseous, as sulfur dioxide (S02),
which results from the incomplete oxidation of S02 to sulfur trioxide (SO,)
in the conversion step. The unconverted S02 is not absorbed in the weak
sulfuric acid during the absorption step and is emitted in the absorption
tower tail gas. The other type of sulfur emission is acid mist which emerges
from the absorption tower entrained in the tail gas.
The actual rate of S02 emissions from a particular plant depends upon
the conversion efficiency of S02 to S03 and the level of production. For
this analysis, a conversion efficiency of 97 percent was used. This results
in an emission factor of 40 pounds of S02 per ton of 100 percent acid produced.
The quantity of acid mist emissions depends primarily on whether normal
sulfuric acid (£'99 percent acid) or oleum (100 percent acid plus excess
dissolved SO,) is being produced. For normal sulfuric acid production, an
emission factor of 2.5 pounds of acid mist per ton of acid produced was used;
while for oleum production, a factor of 7.5 pounds of acid mist per ton of acid
production was used.
Fifty percent by weight of S02 is sulfur.and approximately 32 percent
by weight of acid mist is sulfur.
New Source Performance Standards limit S02 emissions to a maximum of
4 pounds per ton of acid produced, and limit particulate (acid mist) emissions
to a maximum of 0.15 pound per ton of acid produced. The emission factors
are summarized in table C.I.
C.2 Sulfur Emissions Control Alternatives and Costs
The control techniques for reducing sulfur emissions from sulfuric acid
plants vary depending on whether gaseous or particulate emissions are being
controlled. All the systems costed result in the recovery of sulfur in the
form of increased sulfuric acid production.
131
-------
Table C.I. Sulfur emission factors
for sulfuric acid plants
Emissions source Sulfur emission factors
(Ib/ton of acid production)
Gaseous 20.00
Mist, normal plants 1.28
Mist, oleum plants 1.28
Source: Compilation of Air Pollutant Emission
Factors, U.S. Environmental Protection Agency, Research
Triangle Park, N.C., February 1972, pp. 5-18.
Table C.2 summarizes the alternatives available for controlling sulfur
emissions and their removal efficiencies by emissions source. These control
alternatives represent those that appear most feasible by 1978.
C.2.1 Gaseous Emissions Control
Alternatives for gaseous emissions control have been developed and
technically demonstrated: the best are the dual absorption and the sodium
sulfite scrubbing techniques. The dual absorption technique requires the
interposition of an intermediate absorption tower in the traditional
process configuration. By interposing this second absorption tower, overall
conversion of S02 to SOg is increased from 97 to 99.7 percent. Thus, S02
emissions are reduced by 90 percent and production of sulfuric acid is
increased by nearly 3 percent without additional raw materials.
In the sodium sulfite scrubbing technique, unreacted S02 in the exhaust
gas is reacted with sodium sulfite to form certain thermally reactive crystals
which when heated liberate S02. The liberated S02 is fed back into the acid
plant to increase the yield of the plant by about 3 percent. The control
efficiency of this system is about 95 percent.
New sources are expected to meet New Source Performance Standards for
gaseous emissions (4 pounds S02 per ton of acid) by using dual absorption.
The incremental costs necessary for the new sources to use sodium sulfite
scrubbing instead of dual absorption are slightly less than those that can
132
-------
Table C.2.
Sulfur emissions control alternatives
for sulfuric acid plants
Emissions source
Abbrevia-
tion*
Al terna-
tive
Control
alternative
Removal efficiency
(percent)
Gaseous, all plants SA-A
Gaseous, all plants SA-A
Mist, normal plants SA-B
Mist, normal plants SA-B
Mist, oleum plants SA-C
Mist, oleum plants SA-C
Gaseous, all plants SAN-A
Mist, normal plants SAN-B
EXISTING SOURCES
1 Dual absorption
2 Sodium sulfite scrubbing
1 Dual mesh pad demister
2 Tubular fiber demister
1 Dual mesh pad demister
2 Tubular fiber demister
NEW SOURCES
Sodium sulfite scrubbing
Tubular fiber demister
90
95
90
99.5
75
99.5
60
95
'Abbreviations index to table C.3.
Source: Research Triangle Institute.
be calculated for the existing sources since some economies are achieved in
installing these devices on new plants as compared with adding them to exist-
ing plants.
C.2.2. Mist Control
Acid mists can be controlled with filter devices called demisters. Two
effective demisters are currently available. These are the dual mesh pad
demister and the tubular fiber demister which achieve control efficiencies
of 90 and 99.5 percent, respectively, in nonoleum plants; the corresponding
control efficiencies for oleum plants are 75 and 99.5 percent.
New sources are expected to meet the New Source Performance Standards
for mist emissions (0.15 pound of mist per ton of acid) by dual mesh pad
demisters in the normal acid plant and tubular fiber demisters in the oleum
plants. The control alternatives, their costs, and effectiveness in control-
ling S02 emissions from sulfuric acid production were derived from several
studies for EPA as well as private communications with EPA personnel. These
costs have been estimated on an annualized basis for representative plant
133
-------
sizes (table C.3). Interpolation is used to derive cost estimates for
plant sizes other than those estimated for the representative plant sizes.
As shown in table C.3 for an annualized cost of $67,000, emissions from a
50-ton-per-day plant can be reduced to 19 tons of sulfur annually. The pre-
control level is 190 tons annually (i.e., 19 * (1 - removal efficiency of
90 percent)). A plant would control sulfur emissions using dual absorption
when the cost of control plus the tax rate times the remaining emissions
was less than the tax times uncontrolled emissions. For this example plant,
that would be for tax rates greater than 20 cents per pound of sulfur
emissions since:
$67,000 + X (19) = X (190)
X = $391 per ton or 20 cents per pound
where
X = tax rate per ton of sulfur emissions.
C.2.3. Emissions Reductions and Costs
Industry cost of control functions have been developed based on the
costs and effectiveness of the control alternatives costed above and a listing
of the nation's sulfuric acid plants using the contact process. This listing
includes relevant plant information for emissions and cost estimation; the size
and type distribution is shown in table C.4. The functions are minimum cost
functions for achieving sulfur emissions reductions from the sulfuric acid
industry.
The long-run industry total and marginal costs of sulfur emissions
reductions are shown in figures C.I and C.2. The total costs (LTC) increase
at a moderately increasing rate until the higher levels of emissions
reductions are reached. This is reflected in the marginal cost curve
(LMC) which rises rapidly after reductions of about 285,000 tons are
achieved.
134
-------
Table C.3. Sulfur emissions control costs for sulfuric add
Control
alternative
by unit
process
SA-A(l)
SA-A(2)
SA-Bp)
SA-B(2)
SA-C(l)
SA-C(2)
SAN-A(l)
SAN-B(l)
Control costs
(thousand dollars)
Investment
50*
$242
295
16
43
16
43
272
23
250*
$636
776
40
114
40
114
751
62
750* 1500*
$1.230 $1,864
1,500 2,273
105 194
245 396
105 194
245 396
1 ,500 2,321
105 140
Annual 1 zed
50*
$67
88
5
12
5
12
21
7
250* 750* 15001
Annual emissions
after control
/ *nne I
50* 250* 750* 1500*
EXISTING SOURCES
$186 $383
259 563
13 35
33 74
13 35
33 74
73 180
20 39
$612
937
64
124
64
124
19 94 281 563
9 47 141 281
0.8 3.8 11.5 23.0
0.0 0.2 0.6 1.1
5.7 28.7 86.1 172.7
0.8 3.8 111.5 23.0
NEW SOURCES
325
60
9 47 141 ,281
0.0 0.2 0.6 1.1
Annual
additional recovered
sulfur after
50*
169
178
6.9
7.7
17.3
22.2
9
1
lfnn«
control t
\
i
250* 750* 1500*
844 2
891 2
34.5
38,1
86.1
110.0
47
3.6
,531 5
,672 5
103.3
114.2
258.3
332.9
141
10.9
,063
.344
206.6
228.5
516.1
665.8
281
21.9
*Plant size 1n tons of sulfuric add processed per day.
•fSulfur equivalent of sulfuric add.
Source: Developed by Research Triangle Institute from data presented 1n:
1« Background Information for Proposed New Source Performance Standards. APTD-0711, Environmental Protection
Agency, August 1971.
2. Chenrlco Construction Corporation, Engineering Analysis of Emissions Control Technology for Sulfuric Add
Manufacturing Processes. NAPCA, March 1970 (NTIS No. PB-190 393).
3. Boys, Paul A., Environmental Protection Agency (private communication), November 1972.
4. Buckhardt, D.B., VonBree, Inc. (private communication), October 1972.
5- Walsh, R., Environmental Protection Agency (private communication), September 1972.
6. Carey, D., Environmental Protection Agency (private comnunlcation), September 1972.
LTC,
50 100 ISO 200 250
REDUCTIONS IN SULFUR EMISSIONS
(THOUSAND TONS PER YEAR)
300
350
Figure C.I. Total cost* of reductions in sulfur emissions: sulfuric
acid producers (*does not include emissions tax payments) (Source:
Research Triangle Institute).
135
-------
600
I 500
50
100 ISO 200 250
REDUCTIONS IN SULFUR EMISSIONS
(THOUSAND TONS PER YEAR)
300
350
Figure C.2. Marginal cost* of reductions in sulfur emissions: sulfuric
acid producers—1978 (*does not include emissions tax payments) (Source:
Research Triangle Institute).
Table C.4. Size distribution of sulfuric acid plants
Capacity*
less than 100
100-199
200-299
300-399
400-499
500-749
750-999
1 ,000-1 ,499
1,500 & over
Normal
Plants
16
24
18
12
14
21
14
5
28
Oleum
plants
2
9
10
6
7
10
6
2
4
*Plant size in tons of sulfuric acid processed per day.
Source: Research Triangle Institute.
136
-------
Appendix D: SULFUR EMISSION FACTORS AND CONTROL COSTS FOR PRIMARY
NONFERROUS SMELTERS
This appendix discusses the emission factors used, and the control
alternatives included and costed as part of this study to project the effect
of various sulfur tax rates on primary nonferrous smelters.
D.I Emission Factors
D.I.I Copper Smelters
The three principal sources of sulfur dioxide (S02) emissions from
copper smelters are roasters, reverberatory furnaces, and converters. Total
emissions from all these sources amount to approximately 1,250 pounds of S02
for every ton of ore concentrate charged. Of the 15 operating smelters, 7
do not use a roaster but instead charge the reverberatory furnace directly
with the ore concentrate. The roaster, where used, produces about 40 percent
of total S02 emissions; the reverberatory furnace, 20 percent; and lastly the
converter, 40 percent. If no roaster is used, the reverberatory furnace produces
about 30 percent of the total emitted and the converter the remaining 70 percent.
Sulfur emission factors are shown in table D.I.
Table D.I. Sulfur emission factors for copper smelters
Emission factor
(pounds of sulfur per
Emissions source ton of ore concentrate)
Roaster 249
Reverberatory furnace (w/o roaster) 186
Reverberatory furnace (w/roaster) 125
Converter (w/o roaster) 435
Converter (w/roaster) 249
Source: Arthur D. Little, Inc., Economic Impact of Anticipated
Pollution Abatement Costs on the Primary Copper Industry. September 1962.
137
-------
Table D.2. Sulfur emission factors for zinc smelters
CAM**.* Emission factor
source (pounds of sulfur per ton of ore concentrate)
Roaster (dead roast) 600.0
Sinterer (after roast) Negligible
Sinterer (no roast) 630.0
Reduction Negligible
Source: Arthur D. Little, Inc., Economic Impact of Anticipated
Pollution Abatement Costs on the Zinc Industry, April 1971.
D.I.2 Zinc Smelters
Totally uncontrolled emissions from zinc smelters amount to about
1,100 pounds of S0« per ton of ore charged. Where roasting takes place,
either alone or in conjunction with sintering, it accounts for virtually
100 percent of the SOp emitted. Where the zinc concentrate is processed
solely in a sintering machine (actually it is a combination roaster-sinterer),
this unit process produces 100 percent of all S02 emitted in the smelting
operation. Table D.2 presents sulfur emission factors for zinc smelters.
D.I.3 Lead Smelting
Totally uncontrolled emissions from lead smelters amount to about
660 pounds of S02 per ton of ore charged. Roughly 98 percent of this
total is emitted in the sintering step, with the remaining 2 percent emitted
from blast furnaces. Table D.3 presents emission factors for lead smelters.
D.2 Sulfur Emissions Control Alternatives and Costs
The SOp control methods that are likely to be employed in the near
future are:
(a) sulfuric acid plants,
(b) lime and limestone scrubbing,
(c) amine absorption,
(d) ammonia scrubbing, and
(e) sodium sulfite-bisulfite absorption.
The best choice of a control technique from the standpoint of well-developed
n;
technology is the sulfuric acid plant. However, it requires that the
concentration of SOp in the feed gas be at least 3.5 percent for a single
138
-------
Table D.3. Sulfur emission factors for lead smelters
Emission factor
Emissions Source (pounds of sulfur per ton of concentrate)
Roaster-sinterer 472
Blast furnace or 58
dross reverberatory
furnace
Source: Compilation of Air Pollutant Emission Factors, U.S.
Environmental Protection Agency, pp. 7-8.
contact plant and 5 percent for a double contact plant. Unfortunately,
some tail gas streams, notably those from reverberatory furnaces, are normally
below these minimum concentrations. For this reason, two methods that can be
used on dilute streams are considered in this analysis. One is limestone
scrubbing, which has an 85 to 90 percent efficiency and a throwaway byproduct;
the other is amine scrubbing, which can achieve 99 percent efficiency and
produces concentrated S02 which may either be sold as is or mixed with lean
tail gas streams to enrich the input stream to a sulfuric acid plant. Control
cost estimates were based chiefly on previous studies for EPA and private
communications with EPA personnel.
In certain instances, process modification or replacement was incorporated
into the design of control alternatives. The purpose, where this alternative
was chosen, was to produce waste gas richer in SOp and more amenable to the
application of control techniques that produce a marketable byproduct.
The control alternatives costed represent those that appear most
feasible by 1978. Other alternatives may develop in response to the demand
for more cost-effective methods of smelter sulfur emissions control as
a result of an emissions tax. Those alternatives, of course, could not
be included in the analysis.
D.2.1 Copper Smelters
As stated above, gas streams with an S02 concentration of less than
3.5 percent are not amenable to direct input into a sulfuric acid plant.
139
-------
Available data indicate that roaster off-gases average about 8 percent
S02, reverberatory furnaces 1 to 2 percent, and converters about 7 percent.
Where a tail gas stream of S02 concentration less than 3.5 percent
exists, there are essentially two control alternatives. One is to concentrate
the S02 and then input this into an acid plant. The other is to scrub the
lean gas and forego the recovery of economically valuable sulfur products.
The alternative costed to provide a method for concentrating S02 is
amine absorption. In this technique, the low SCL concentration gas stream
is passed through an absorption bed on which up to 99 percent of the SCL
is absorbed. This bed can then be stripped, yielding an almost pure stream
of S02. This S02 can be fed into an acid plant either alone or in
combination with off-gases from other smelting until processed.
The alternative costed to scrub the weak S02 stream is wet limestone
scrubbing. In this system, the gas stream is passed through a wet
limestone slurry, the end product of which is solid calcium sulfate (CaSOJ
which must be disposed of, there being no commercial use available. An
S02 removal efficiency of 90 percent may be expected from the application of
wet limestone scrubbing.
Wjjere a sufficiently concentrated gas stream was available, a sulfuric
acid plant was costed as the control alternative. It must be remembered,
however, that a sulfuric acid plant does not provide 100 percent control of
the input S0«. For the purposes of this analysis, it was assumed that a
single-absorption sulfuric acid plant will emit approximately 2,000 parts
per million (ppm) S02 in its off-gas stream; while a double-absorption
plant will emit about 500 ppm S02. Table D.4 indicates the expected removal
efficiencies of single- and double- absorption systems as a function of the
input S02 concentrations. Double absorption can reduce S02 emissions below
those effected by single absorption, but, of course, at a significantly higher
cost.
Three additional control alternatives are possible. Two alternatives
are the installation of either (a) single- or (b) double-absorption acid
plants where none now exist; the third alternative is the addition of a
dual absorption capability to an existing single-absorption plant.
For the purposes of this analysis, the 15 existing copper smelters are
characterized by two parameters. These are: (1) the unit processes employed
and (2) the presence or absence of an acid plant (it should be noted that all
140
-------
Table 0.4. Sulfur dioxide removal efficiencies of sulfuric acid plants
as a function of input concentrations for single- and double-absorption
systems
Input S02 Removal efficiency
(percent)
concentration
percent Single absorption Double absorption
3.5
5.0
7.5
10.0
94.3
96.0
97.4
98.0
98.6
99.0
99.3
99.5
Source: Research Triangle Institute.
existing smelter acid plants are of the single contact variety and that no
smelters practice limestone scrubbing). Thus, a copper smelter can have
"green feed" to the reverberatory furnace (i.e., no roaster) plus converters,
or it can have "conventional feed" (i.e., ore concentration to a preliminary
roaster) plus reverberatory furnaces and converters; for each of these plant
types, there may or may not be an existing acid plant. These four plant
types are summarized in table D.5 along with an abbreviation for each type
which will be used in subsequent discussion. Each copper smelter in the
plant inventory has been identified as being one of the four plant types.
Table D.5. Copper smelter plant types
Abbreviated
Plant type designation
Green feed (no acid plant) Cu-A
Green feed (acid) Cu-B
Conventional feed (no acid plant) Cu-C
Conventional feed (acid plant (on roaster or converters)) Cu-D
Source: Research Triangle Institute.
141
-------
Table D,6. Sulfur emission control alternatives for copper smelters
Plant type Alternative
Control alternative
Removal efficiency
(percent)
Cu-A
1
2
Double-absorption acid plant on
converters
Double-absorption acid plant on
68.6
95.6
converters & limestone scrubbing
on reverberatory furnace
Double-absorption acid plant on
converters & amine absorption on
reverberatory furnace (concentrated
S0« to double-absorption acid plant)
97.5
Cu-B'
Limestone scrubber on reverberatory
furnace
Amine absorption on reverberatory
furnace (concentrated S02 to an acid
plant)
95.3
97.2
Cu-C
1
2
Double absorption on converter & roaster 78.6
Double absorption on converter, roaster, 98.2
and reverberatory furnace (amine absorber
concentrates the reverberatory furnace flow)
Cu-D
Double-absorption acid plant on either 78.2
roaster or converters
Double-absorption acid plant on either 97.9
roaster or converter & amine absorber
on reverberatory furnace (concentrated
S02 stream to new double-absorption plant)
Source: Research Triangle Institute.
Sets of S02 control alternatives were selected independently for each
plant type. These alternatives are listed in summary fashion in table D.6.
Table D.7 presents a summary of the costs arid emissions associated with
each control alternative for selective plant sizes.
D.2.2 Zinc Smelters
Zinc smelters can be categorized into three plant types:
(a) Combination roaster-sintering machine (downdraft type)—
no acid plant,
142
-------
Table D.7. Sulfur emission control costs for copper smelters
Control
alternative
by unit
process
Cu-A(l)
Cu-A(2
Cu-A(3)
Cu-B(l)
Cu-B(2)
Cu-C(l)
Cu-C(2)
Cu-D(l)
Cu-D(2)
Control costs
(million dollars)
Investment
100*
$ 6.3
9.2
13.8
2.9
10.0
7.2
15.3
4.5
13.2
200* 600*
$ 9.6 $19.6
14.3 28.6
21.4 42.8
4.5 9.0
15.5 31.0
11.2 22.4
23.7 47.4
7.0 14.0
20.5 41.0
Annual 1zed
100* 200*
1.37 2.17
2.52 4.01
3.00 4.75
1.15 1.84
2.13 3.38
1.52 2.41
3.15 4.99
0.96 1.53
2.65 4.19
600*
4.30
7.00
9.37
3.67
6.75
4.81
9.96
3.05
8.37
Annual emissions
after control
(thousand tons)
100* 200* 600*
13.3 26.5 79.5
1.8 3.7 11.0
1.1 2.1 6.3
2.0 4.0 12.1
1.2 2.4 7.3
9.1 18.1 54.5
0.7 1.5 4.4
9.2 18.4 55.2
0.9 1.8 5.5
Annual
additional recovered
sulfur after control t
(thousand tons)
100* 200* 600*
29.0 58.1 174.2
29.0 58.1 174.2
41.3 82.5 247.5
000
12.4 24.9 74.7
33.2 66.4 199.1
41.5 83.0 249.1
33.0 66.0 198.0
41.3 82.5 247.5
*P1ant size 1n tons of copper processed per day.
tSulfur equivalent of sulfurfc acid.
Source: Research Triangle Institute.
Arthur G. McKee and Co.. Systems Study for Control of Emissions. Primary
Nonferrous Smelting Industry. VII, Tina I report to National Air Pollution Control
Administration, Contract PH86-65-85, June 1969.
on the
^"tractors . .The Impact of A1r Pollution Abatement
Si
(b) Roaster(s) followed by the sintering machine(s) with
the acid plant(s) receiving tail gas from the roaster(s),
(c) Roaster(s) followed by electrolytic purification with
the acid plant(s) receiving the tail gas from the roaster(s).
For the purpose of this analysis, these three plant types were termed
Zn-A, Zn-B, and Zn-C, respectively. Each zinc smelter in the plant inventory
has been identified as being one of the three plant types. The combination
roaster-sinterer emits a tail gas stream with an S02 concentration of 2 to 2.5
percent. This concentration is too low to be used as input to a sulfuric acid
plant. In plants with a separate roaster and sintering machine, the roaster
accounts for practically 100 percent of the S02 emitted. However, not all
roasters can be controlled with acid plants. Older Ropp-type roasters issue
a gas stream with an S02 concentration less than 1 percent. There is, however,
only one plant that presently employs Ropp-type roasters and this plant is
expected to close by 1975; it has not been included in this analysis. All
143
-------
other roasters can be and are presently controlled by single contact sulfuric
acid plants.
Control alternatives for the roaster-sintering machine included replacing
the downdraft machine with an updraft machine which emits two off-gas streams
One stream, which contains about 85 percent of all the sulfur emitted, has
an S02 concentration. The rich gas stream is amenable to control by an acid
plant. The other alternative is to replace the existing downdraft machine
with a recirculating updraft machine with a single off-gas stream having an
S02 concentration of about 5 percent. This single stream may then be input
into an acid plant.
In plants with a roaster and an acid plant, there are two alternatives.
One is to convert the single contact plant to a double contact plant thus
increasing the conversion of S02 + S03. The other alternative is to add
a Wellman Scrubber to treat the tail gas of the acid plant. The Well man
process not only scrubs the S02 out of the acid plant tail gas, but also
upon regeneration yields a high S02 concentration stream which can be used
as input to the acid plant; thus, like the double contact process, the
Wellman process increases the effective yield of the acid-making process.
These alternatives are summarized in table D.8. Table D.9 summarizes the
Table D.8. Sulfur emission control alternatives for zinc smelters
Plant type Alternative Description Removal efficiency
(percent)
Zn-A 1 Convert to updraft sintering and 85.3
place acid plant on rich stream
from sinter.
2 Convert sintering machine to recir- 96.0
culating updraft sintering with
acid on entire off-gas stream.
Zn-B
Zn-C
1
2
1
2
Add double contact to present acid
plant.
Add Wellman Scrubber to present acid
plant.
Add double contact to present acid
plant.
j
Add Wellman Scrubber to present acid
plant.
99.0
99.5
99.7,
99.9
Source: -Research Triangle Institute.
144
-------
Table D.9. Sulfur emission control costs for zinc smelters
Control
alternative
by unit
Zn-A(l)
Zn-A(2)
Zn-B(l)
Zn-B(2)
Zn-C(l)
Zn-C(2)
Control costs
(million dollars)
Investment
100*
$3.9
5.3
0.262
0.320
0.262
0.320
350*
$ 8.6
11.7
0.556
0.679
0.556
0.679
600*
$12.1
16.4
0.768
0.938
0.768
0.938
Annual 1 zed
100*
$0.548
0.717
0.073
0.095
0.073
0.095
350*
$1.300
1.702
0.160
0.226
0.160
0.226
600*
$1.890
2.468
0.225
0.328
0.225
0.328
Annual emissions
after control
(tons)
100*
2,772
759
129
96
50
23
350* 600*
9,669 16,566
2,460 4,521
455 779
337 574
172 294
86 149
Annual additional recovered
sulfur after controlf
(tons)
100*
15,048
17,259
436
469
442
465
350*
52,701
60,390
1,525
1,643
1,544
1,630
600*
90,354
103,521
2,614
2,812
2,647
2,795
*Plant size 1n tons of zinc processed per day.
tSulfur equivalent of sulfuric add.
Source: Research Triangle Institute.
Arthur 6. McKee and Co., Systems Study for Control of Emissions. Primary Nonferrgus Smelting Industry, VII,
Final report to National Air Pollution Control Administration, Contract PH86-65-85, June 1969.
Arthur D. Little, Inc., Economic Impact of Anticipated Pollution Abatement Costs on the Primary Zinc
Industry. September 1962.
costs and emissions associated with each control alternative for selected
plant sizes.
D.2.3. Lead Smelters
Lead smelters can be categorized into three plant types:
(a) Plants with downdraft sintering machines and no acid plant,
(b) Plants with updraft sintering machines and no acid plant,
(c) Plants with updraft sintering machines with an acid plant
on the strong (high SOg concentration) off-gas stream only.
These plant types have been given the abbreviated designations Pb-A,
Pb-B, and Pb-C, respectively. Each lead smelter in the plant inventory has
been identified as being one of the three plant types.
The downdraft machines emit an off-gas stream with an S02 concentration
too low to be used as input into an acid plant. Updraft machines emit two
gas streams: one, the so-called rich gas stream, is amenable to control by
an acid plant; the lean gas stream is not. The alternatives selected for
this study are summarized in table D.10.
D.2.4 Emissions Reductions and Costs.
Industry cost of control functions have been developed based on the
costs and effectiveness of the control alternatives costed above and a
listing of the nation's smelters which includes relevant process parameters
for emissions and control cost estimation. The smelter size distribution is
145
-------
Table D.10. Sulfur emission control alternatives for lead smelters
Plant type Alternative
Description
Removal efficiency
(percent)
Pb-A
Downdraft machine replaced with an 84.6
updraft type and add plant Installed
on rich SOg stream.
Updraft machine of recirculating 94.9
type with an acid plant on combined
sinter off gas.
Updraft sintering machine and acid 96.3
plant on rich stream and limestone
scrubber on lean stream.
Pb-B
1
2
Add plant on rich stream.
Weak sinter gas, redrculated and
84.6
94.9
add plant on combined off gas.
Add plant on rich stream and lime-
stone scrubber on lean stream.
96.3
Pb-C
1
Limestone scrubber on lean stream.
96.3
Source: Research Triangle Institute.
Table D.ll. Sulfur emission control costs for lead smelters
Contro
al ternat
by un1
proces
Pb-A|
Pb-A
Pb-A
Pb-B
Pb-B
Pb-B
I
r— CM CO
1
•fur
t
S
Pb-C(l)
Control costs
(million dollars)
Investment
100* 350* 600*
$2.760 $6.080 $ 8.540
3.650 8.040 11.290
3.750 8.260 11.590
1.180 2.600 3.650
2.070 4.560 6.400
2.170 4.780 6.710
0.990 2.180 3.060
Annuall zed
100* 350* 600*
$0.589 $1.517 $1.829
0.626 1.596 1.939
1.012 2.446 3.129
0.431 0.951 1.340
0.468 1.030 1.450
0.854 1.880 2.640
0.421 0.928 1.300
Annual emissions
after control
(tons)
100* 350* 600*
1,274 4,455 7,623
426 1,485 2,544
317 1,069 1,832
1,274 4,455 7,623
426 1,485 2,544
317 1,069 1,832
317 1,069 1,832
Annual recovered
sulfur after control t
(tons)
100* 350* 600*
6,996 24,486 41,976
7,821 27,423 47,025
6,996 24,486 41,976
6,996 24,486 41,976
7,821 27,423 47,025
6,996 24,486 41,976
000
*P1ant size 1n tons of lead processed per day.
tSulfur equivalent of sulfuric add.
Source: Research Triangle Institute.
Arthur G. McKee and Co., Systems Study for Control of Emissions. Primary Nonferrous
Smelting Industry. VII, Final report to National A1r Pollution Control Administration,
Contract PH86-65-85, June, 1969.
Arthur D. Little, Inc., Economic Impact of Anticipated Pollution Abatement Costs on
the Primary Lead Industry. September 1962.
146
-------
shown in tables D.12 through D.14. These functions are minimum cost functions
for achieving sulfur emissions reductions from the primary nonferrous smelting
industries.
Using lead smelters as an example, a 100 ton per day, type A smelter can
reduce emissions to 1,274 tons annually for an annualized cost of $589,000
(table D.ll). The precontrol level of emissions is 8,273 tons of sulfur
annually (i.e., 1,274 - (1 - removal efficiency of 84.6 percent)). A smelter
would control sulfur emissions by installing an acid plant where the cost of
control plus the tax rate times the remaining emissions was less than the tax
times the uncontrolled emissions. For this example smelter, that would be
for tax rates greater than 4 cents per pound of sulfur emissions since:
$589,000 + TAX (1,274) = TAX (8,273)
TAX = $84 per ton or 4 cents per pound
where
X = tax rate per ton of sulfur emitted.
The long-run industry total and marginal unit costs of emissions
reductions are shown in figures D.I and D.2. Because of the absence of
technology that could provide intermediate levels of control at acceptable
costs and the limited number of plants, the total cost function (LTC) is
probably discontinuous throughout all but the upper ranges. The marginal
costs (LMC) rise slowly until emissions reductions of about 94 percent
(1.5 million tons) are reached. After that point, marginal costs rise quite
sharply.
Table D.12. Size distribution of copper smelters
Capacity*
0 -
100 -
200 -
300 -
400 -
500 -
700 -
99
199
299
399
499
599
799
Plant type
Cu-A Cu-B Cu-C Cu-D
1
.
4 - - 2
2-11
1
2
1
*Plant size 1n tons of copper processed per day.
Source: Research Triangle Institute.
147
-------
Table D.13. Size distribution
of zinc smelters
Table D.14.
Plant type
Capaci ty~
100 -
200 -
300 -
400 -
500 -
199
299
399
499
599
Zn-A Zn-B
1
1
-
1
1
Zn-C
1
1
1
-
-
*Plant size in tons of
zinc processed per day.
Source: Research Triangle
Institute.
Size distribution
of lead smelters
Plant type
Capacity*
100
200
300
400
800
- 199
- 299
- 399
- 499
- 899
Pb-A Pb-B Pb-C
1 1
1
1
1
1
*Plant size in tons of
lead processed per day.
Source: Research Triangle
Institute.
o
(E
O
UJ
N
140
120
i w
O K 100
§ g 80
g o 60
2 -i
40
20
LTCi
200 400 600 800 1000 1200
REDUCTIONS IN SULFUR EMISSIONS
(THOUSAND TONS PER YEAR)
1400 1600
Figure D.I. Total cost* of reductions in sulfur emissions: primary
nonferrous smelters—1978 (*does not include emissions tax payments)
(Source: Research Triangle Institute).
148
-------
(E
O
O
U.
O
CO
O
o
600
500
^400
z
g 300
£200
o.
100
en
< 80
si
N
60
40
20
0
LMC
200 400 600 800 1000 1200
REDUCTION IN SULFUR EMISSIONS
(THOUSANP TONS PER YEAR)
1400
1600
Figure D.2. Marginal cost* of reductions ia sulfur emissions: primary
nonferrous smelters (*does not include emissions tax payments) (Source:
Research Triangle Institute).
149
-------
Appendix E: THE PROJECTED MARKET FOR RECOVERED SULFUR
E.I Introduction
Several of the alternatives for controlling sulfur oxide emissions
result in the recovery of marketable sulfur or sulfuric acid. The
future market for those products may influence the choice of emissions
control and the net cost to the industries controlling sulfur emissions.
Therefore, it is desirable to examine briefly the market for sulfur.
Sulfur is a naturally occurring inorganic mineral with many
industrial uses, especially in the form of sulfuric acid. The largest
use is in fertilizer production which consumes about one-half of
all production. To a lesser extent, it is also used in making steel,
rayon, paper, nonferrous metals, and chemicals.
E.2 Major Sources of Sulfur
Sulfur is mined from subterranean sulfur domes (native sulfur) and
is also produced as a byproduct of some industrial processes, principally
in the refining of sour oil and gas. Imports, primarily from Mexico and
Canada, also supply portions to the U.S. market.
The share of the U.S. sulfur market enjoyed by native sulfur has
been declining fairly regularly over the past 20 years. The smelter
acid share has been fairly constant. The decline in the native sulfur
share results from an increase in byproduct sulfur recovery which has
increased over fourfold in the last 20 years.
E.3 Consumption of Sulfur
The apparent consumption of sulfur in the United States has been
increasing about 2.8 percent annually since 1950, reaching 10.0 million
tons* in 1970.t
The demand for sulfur is derived from the demand for fertilizers
and other products for which sulfur is a major component. In the past,
*Sulfur quantities are usually measured in long tons (2,200 pounds);
however, for purposes of consistency within this report all figures here
have been converted to the more common short ton (2,000 pounds).
tU.S. Department of Interior, Bureau of Mines, Minerals Yearbook,
1970, Washington, D.C., p. 1054.
151
-------
1900
V)
g 1600
CO
400
1950 1955 I960 1965
YEAR
1970
1975
I960
Figure E.I. Sulfur consumption trends (Source: Historical data, Department
of the Interior: Projections, Research Triangle Institute).
there has been a close relationship between gross national product (GNP)
and sulfur consumption in the United States. Given the inelastic
demand for sulfur at the low prices, as reported in several studies,*
and the prospects for continued depressed prices, a reasonable projection
of sulfur demand for use in this study was obtained by extending
this historical relationship to 1980. Based on a projected 1980
GNP of $1,155 billion,t sulfur consumption in 1978 is expected to
be 15.1 million tons (see fig. E.I), 5.1 million tons above the 1970
estimate. With the increasing production of sulfur on a worldwide
basis, it appears unlikely that the export demand will grow. Therefore,
*M. H. Farmer and R. R. Bertrand, Long-range Sulfur Supply and
Demand Model, Report GRU.1GM.71. Esso Research and Engineering Company,
Linden, New Jersey, November 1971, p. 6.
tU.S. Department of Labor, Bureau of Labor Statistics, Patterns
of U.S. Economic Growth (1980 projections of final demand, interindustry
relationships, output, productivity, and employment), Bulletin 1672,
Washington, D.C., U.S. Government Printing Office, 1970, p. 43.
152
-------
Table E.I. Projected sulfur emissions--1978*
Fmiccinnc cn,,v.oQ Sulfur emissions
Emissions source (thousand tons)
Steam electric plants 11,396
Area sources 5,679
Petroleum refineries 772
Sul f uric acid production 376
Primary nonferrous smelters 1,650
19,873
*Assuming only the implementation of the New
Source Performance Standards.
Source: Research Triangle Institute.
the projected U.S. consumption is expected to reasonably reflect the
total market for U.S. sulfur.
As shown in table E.I, the projected 1978 uncontrolled sulfur
emissions from the major sources under consideration in this study
are about 20 million tons, an amount in excess of the projected production
of sulfur. Any significant reduction in sulfur emissions and recovery
and sale of sulfur could account for a substantial share of the projected
growth in production. Such an eventuality will have a depressing effect
on sulfur prices.
E.4 Prices of Sulfur
The price of sulfur is. a function of the interrelationship of sulfur
demand and supply. The demand for sulfur is derived from the demand for
products where sulfur is used as a raw material. The supply of sulfur
is based not only on the costs of mining sulfur but also on the availability
of recovered sulfur which will be recovered, regardless of price, and sold
V.
for whatever price it will bring.
Figure E.2 shows the trend in sulfur prices over the past 20 years.
For a long period beginning in 1947 and extending through 1966, sulfur
prices edged up about 2.4 percent annually. However, beginning in
December 1966, sulfur prices began to rise rapidly, averaging a 2.7
153
-------
140
'O 2468024 6JAJO JAJ OJAJOJAJOJ A
S g> 1967 1968 1969 1970 '71
YEAR
Figure E.2. Sulfur price index (Source: Department of Labor).
percent increase per month for the next 15 months. Between September
and October of 1967, prices jumped 16 percent. In 1968, prices were
relatively stable. Then early in 1969, they began to decline rather
precipitously, finally reaching a level in 1971 of half their 1969 high.
The collapse of sulfur prices in 1969 is attributed to several
causes, among them being:* (1) the drop in fertilizer demand due to
overcapacity, (2) the expansion of pyrite roasting capacity in Europe,
(3) the start-up of a large Frasch process operation in west Texas,
(4) the expansion of modified Frasch process production in Poland,
(5) expanded recovery of elemental sulfur by desulfurization of petroleum,
(6) the increase in the recovery of acid from smelter gas, and (7) the
U.S. business recession.
Published data on absolute value of current prices is unavailable
because the Frasch process producers stopped publishing prices in
1969. However, by using the Bureau of Labor Statistics price index
*M. H. Farmer and R. R. Bertrand, Long-range Sulfur Supply and
Demand Model. Report GRU.1GM.71, Esso Research and Engineering Company,
Linden, New Jersey, November 1971, p. 3.
154
-------
lith the 1968 price of $38 per ton for crude, domestic, dark, bulk
sulfur, as reported in the Oil, Paint, and Drug Reporter, the 1971
price can be estimated as about $20 per ton. It appears that $20
per ton represents the low end of the price range for Frasch sulfur;
prices may be as much as $14 higher depending on the port of delivery.
Recovered sulfur prices are a different matter. One Southwest oil
refiner quotes prices of $14 to $25 per ton f.o.b. the refinery.*
Many large buyers, however, are willing to pay a premium for Frasch
process sulfur to obtain needed quantities and to help assure future
availability.
It appears that future prices will continue to be depressed even
without additional recovery by the large increases expected in Canadian
capacity and continued growth in the amount recovered, especially
from sour gas and from smelter gas. Furthermore, if ocean shipment
of liquid sulfur becomes a reality, Canadian sulfur may be competitive
with domestic sulfur in the largest U.S. sulfur market, the Florida
phosphate industry.
The long-run upper limit on the price of sulfur is determined by both
the cost of obtaining sulfur from its various sources and by the cost of
alternative manufacturing processes that avoid the use of sulfur.t One
report placed the costs of sulfur production between $10 and $43 per ton
depending on source (see table E.2). The lower limit is indeterminate.
In summary, because of the downward pressures on sulfur prices
and the likelihood of additional recovery of sulfur from controlling
sulfur emissions, we believe that valuing the recovered product at a
high price appears unwarranted. Mos-t optimistically, the recovered product
could be sold at prevailing market prices. This outcome would appear
reasonable only if the product is not recovered in amounts large
enough to significantly increase supplies and cause prices to decline.
At the other extreme, if increased recovery of sulfur products results
in substantial increases in supply without significant increases in
quantity demanded, then the total value of sales may actually decrease.
*C. W. Winton, "Dark Cloud on Sulfur's Horizon," Chemical Week.
February 10, 1971, p. 31.
tM. C. Manderson, "The Sulfur Outlook," Chemical Engineering
Progress, November 1968, pp. 47-53.
155
-------
Table E.2. Sulfur production costs by source
Source Cost per long ton
Frasch
Low cost $10
Medium cost 15
High cost 23
Sour gas
Natural gas 15
Refinery 20
Smelter gas 18
Pyri tes 35
Gypsum 35
Other native 35
Utility stack gas 43
Source: M. C. Manderson, "The Sulfur Outlook,"
Chemical Engineering Progress, November 1968, pp. 47-53.
This is a reasonable possibility given the inelastic demand for sulfur.
Finally, in industries that use sulfur as an input and that recover
more of the product due to a tax on their emissions of sulfur, the
recovered product may be used as a substitute for purchased sulfur. In
this case, the recovered sulfur should be valued at prices reflective
of the sulfur cost savings. In many cases, this possibility appears
most likely. For this analysis of a tax on sulfur emissions, the
approximate current price, $20 per ton, and lower prices more reflective
of the likely increase in sulfur supply, $10 and $0, have been selected
as alternative future values for the marketable, recovered sulfur and
for sulfuric acid.
156
-------
Appendix F: THE EFFECT OF THE CORPORATE TAX STRUCTURE ON THE
PROJECTED EMISSION LEVELS.
The purpose of this appendix is to analyze the effect of ignoring
corporate income taxes and tax preferences on the levels of emission
reductions that are projected in this study.*
The following variables will be used:
e = corporate income tax rate, (0 < e < 1);
4 = sulfur tax rate, dollars per pound;
E. = sulfur emissions per period under j control option
J
(j = 1, ..., m), pounds per year;
K. = investment cost of the j control option, dollars;
J +h
V. = variable cost per period under the j control option;
j
d = depreciation rate per period, (0 < d < 1);
n = planning horizon, in years;
r = cost of capital;
PV = present value of the anticipated cash outlay over the
period n, in dollars.
Assume that the emissions tax payments () times the emissions (E.) that would remain after the implementation
J.L. J
of j control option:
TVC. = V. + 4>E. j = 1, .... m (1)
j j J
*See Richard D. Wilson and David W. Minnotte, "Government/Industry
Cost Sharing for Air Pollution Control," Journal of the Air Pollution
Control Association, XIX, No. 10 (October 1969), pp. 761-766, for a
detailed discussion, with examples, of these tax considerations.
157
-------
Total emission-control-related depreciation expenses (DPR.) during
any period will be:
DPR. = d K.
j = 1, ..., m (2)
The model used in this study assumes that TVC,- is constant over
time and that emissions (E.) associated with the jth option are known.
Consequently, during the first year of operation with the control
equipment in place, the plant must make a cash outlay for the capital
cost of the device (if any) and for the variable costs associated with
the control option; during each succeeding period, outlays are required
for only the variable costs. The discounted present value of the costs
associated with the j control option is:
n
DPV(costs). =
j
Z (lir) * <»,
t=o V /
K
j
(3)
:th
The discounted cash value of the benefits of the j control option
is the discounted stream of income tax savings which is:
DP V( benefits). =
e(V. + <£. + d K.)
J J J
(4)
t=
The method of choosing among control options in this study was
equivalent to choosing the option for which Eq. 3 was minimized.*
However, the income tax consideration implies that savings associated
with control options are not considered under the criterion of Eq. 3.
The rational manager will consider net costs; i.e., the difference
between Eqs. 3 and 4:
DPV(net cost).. = (1-e) V + >E) + K, l-e
(5)
*To see this, divide both sides of Eq. 3 by the discount factor.
This yields the annualized expenditure whose present value is given by
Eq. 3:
Annualized Cost. = V, + E. + K
The method of this study was to predict that plants would choose the
control option which minimizes annualized cost; formally equivalent
to choosing the option which minimizes Eq. 3.
158
-------
If Eq. 5 is divided through by 0-0), it is different from Eq. 3
only in the coefficient on K.. Given that the annual depreciation rate,
j
d, is the reciprocal of the number of years, n, over which the equipment
is depreciated, it is readily obvious that that coefficient is greater
than unity.*
Some tentative conclusions seem warranted in comparing Eqs. 3 and
5. First, all costs appear to be overstated by using Eq. 3. Because
of the considerations mentioned in the foregoing paragraph, the over-
statement would appear to be somewhat less than a factor of l/(l-e).
For example, if the corporate income tax rate is 50 percent (e=0.5),
the overstatement of costs predicted by Eq. 3 would be somewhat less
than a factor of 2. However, Eq. 5 also indicates that the effective
emissions tax rate, 4, is overstated by a factor of l/(l-e) in Eq. 3
since all emissions tax payments are deductible expenses for income tax
purposes. Consequently, the behavioral predictions of this report are
probably not too affected, since both the effective tax rate and net
costs were overstated by about the same factor.
*Define a (0 < a < 1) as
a = d
Then the coefficient, a, on K. in Eq. 5 is
J
, r
a = 176ct
If o=l, a=l. If a=0, a-l/(l-e). Therefore, if the income tax rate is
50 percent (e=0.5), the upper bound on a is 2.
159
-------
Appendix G: SOME ASPECTS OF SULFUR EMISSION TAXES IN THE PRESENCE
OF ADVANCING CONTROL TECHNOLOGY FOR POWER PLANTS
G.I Introduction
Most of the conclusions of this report are based on the assumption
that the producer-emitter makes his input choices in a static environment,
with certain knowledge of input costs and a finite set of alternative
inputs. One assumption was that only three hardware control options (dry
limestone, wet limestone, and magnesia base scrubbing) would be available
to the power plant, at a known cost per unit of production capacity in
1978. The purpose of this appendix is to reexamine that assumption and
its effect on the results of this study in the light of other control
alternatives that are expected to be available beginning in 1980.
Whereas regulations define specific reductions that must be achieved
by all plants that comply with these directives, an emissions tax policy
grants decisionmakers the latitude to avoid or to delay installing control
devices. The use of a tax allows them to combine tax payments, to switch
fuel, and to remove sulfur from flue gases to the extent that they deem
economical. This additional flexibility not only induces efficiency in
the reduction of sulfur emissions, resulting in equalization of marginal
emissions reduction costs and minimization of total reduction costs (net
of taxes) among plants, but also adds an element of uncertainty to the
projected pattern of emissions reductions over time. This is a particularly
important consideration in view of imminent improvements in control
technology, for it implies that the plant may economize on resources in the
long run by forfeiting tax payments in the near future to avoid getting
tied into an inefficient, but currently available, control technology.
Specifically, this analysis attempts to illustrate the motivations for
and the extent of delays in control activities that could result from
changes in assumptions regarding flue gas desulfurization equipment costs,
operating characteristics, and availabilities.
Section G.2 reviews some of the technical aspects of expected post-
1978 sulfur oxide control alternatives along with some preliminary estimates
of the expected costs (per unit of capacity) of this equipment. In section
G.3, some of the theoretical aspects of analyzing the producer-emitter's
response to a sulfur emissions tax are discussed. That analysis sets the
framework for the preliminary empirical investigation of these considerations
in section G.4.
161
-------
G.2 Some Expected SOp Control Alternatives for Power Plants, 1980-1985
Besides dry and wet limestone and magnesia base scrubbing—all projected
to be available by 1978—there are several other S02 control options that
currently appear attractive and amenable to practical application between
1980 and 1985. These options may be categorized as: (1) new flue gas
cleaning technologies and (2) alternative low sulfur fuels. Another set
of options, also mentioned here, are those that must be incorporated in
new power plants, either as a nonconventional steam-generation system
(nuclear fission) or as a markedly different method of conventional fossil
fuel combustion (fluidized bed and combined cycle power system). Table G-l
at the end of this section summarized the anticipated costs, technical
references, and dates of availability for the control options discussed here.
G.2.1 SOp Control Alternatives for Existing Power Plants
Two new control hardware options that are both representative and—
among the emerging technologies—economically attractive are expected to
become viable by 1980. They are the citrate and the double alkali S02
removal processes.
G.2.1.1 Citrate SOp Removal Process. The citrate process employs a
scrubbing sodium citrate solution. This solution scrubs the flue gas stream
by dissolving the S02 gas component into the solution which is then reacted
with hydrogen sulfide (H2S). The chemical reaction generates solid sulfur
which becomes a marketable byproduct.
Although the expected removal efficiency of this process, 90 to 95
percent, is about the same or slightly greater than those projected for
wet limestone and magnesia base scrubbers, the anticipated initial cost of
the system is somewhat lower, estimated approximately at $39 per kilowatt
of installed capacity. On an annualized basis, these capital costs, plus
operating, maintenance, and servicing costs, are roughly projected at 1.95
mills per kilowatthour of output (assuming average machine load factors
and the absence of credits for the sale of recovered sulfur).
G.2.1.2 Double Alkali S00 Removal Process. The double alkali system
_ ^ ——
is similar to the wet limestone system in-that its expected removal efficiency
is only slightly higher, 90 percent compared to 85 percent, and in that it
is a "throwaway" process. No marketable sulfur product is generated.
In this process a soluble alkali composed of a sodium salt (like
Na2S03) is used to strip S02 from the flue gas. This sulfur-bearing solution
162
-------
is then regenerated by reacting it with an insoluble alkali such as
lime or limestone. The output from this reaction is a reusable alkaline
scrubbing solution and a throwaway sludge, such as calcium sulfite.
The estimated cost of this process, projected to be available in
1980, is of the order of $24 per installed kilowatt. Annualized costs,
including operation, maintenance, and servicing costs, are projected
at 1.75 mills per kilwatthour under the previously mentioned assumption
of average plant operating factors.
G.2.1.3 Gasified Coal. Current energy research indicates that the
gasification of coal is both practical and attractive from the standpoint
of emissions reduction. This process appears capable of generating a
synthetic gas whose caloric content averages about 950 Btu's per cubic
foot, compared to an average of 1,000 Btu's for natural gas. The
synthetic gas would have a sulfur content of about 0.1 percent. A plant
consuming 3 percent sulfur coal containing a comparable heat input would
have to achieve a 99.5 percent sulfur removal efficiency to achieve the
same emissions reductions that would accompany the use of this synthetic
gas. Current best judgment indicates that the cost of producing gasified
coal, not deemed feasible until 1985, will be about 60 cents per million
Btu's.
G.2.2 SOp Control Options for New Power Plants
Conventional power plants have dual deficiencies: (1) all the
sulfur in the fossil fuel being combusted gets volatized and, hence, becomes
gas entrained, and (2) these plants reflect low thermal efficiencies (i.e.,
high fuel input to power output ratios). Some new power generating techniques
remedy one or the other of these problems. These new power systems include
a modification of conventional technology, fluidized bed combustion; a
new concept in coal fired power systems, COGAS; and nuclear power plants.
G.2.2.1 Fluidized Bed Combustion Process. The fluidized bed process
allows the combustion of coal in a bed which contains an active material
such as a limestone dolomite sorbent. Between 90 and 95 percent of the
sulfur in coal is diverted away from entrainment in the combustion off-gases
as a result of chemical reactions which accompany this process. The pres-
surized (as opposed to atmospheric) fluidized bed combustion process appears
the most economical of those options which follow this general concept.
163
-------
This process is not expected to be commercially available until 1984
at the earliest. Current engineering judgment suggests that the capital
cost of a plant equipped with the fluidized bed process will be in the
neighborhood of $190 per kilowatt (compared to about $120 for current
conventional plants) and will require about 6.30 mills per kilowatthour
to operate (compared to 6.22 for a conventional coal-fired power plant).
G.2.2.2 Combined Cycle Power Systems. Combined cycle power production
techniques combine the efficient use of energy with virtually emission-free
power generation. Of the several such systems under development, the
COGAS process is in the most advanced stages and appears capable of
application by 1985. The acronym COGAS refers to this system's combination
of coal to gas conversion with the advanced power cycle concept. The
COGAS process generates a virtually sulfur-free gas from high sulfur and high
ash content coal; the gas is very low in heat content, on the order of
60 Btu's per cubic foot. Once generated, the low-Btu, pressurized gas is
fired in a combustion turbine. The residual energy in the flue gas
from this combustion is then captured in a heat recovery boiler to which
supplemental fuel is fired to produce superheated steam that is used to
drive a steam turbine. This process promises not only 99-percent sulfur
removal efficiencies but also increased thermal efficiencies (a measure of
output to fuel input requirements) of as much as 55 percent. The projected
investment requirements of such a system are about $127 per kilowatt of
installed capacity; annualized operating costs are estimated at 5.20 mills
per kilowatthour.
G.2.2.3 Nuclear Power Plants. Nuclear fission is a sulfur-emission-
free alternative to fossil fuel combustion for central station power
generation. The basic way in which nuclear power generation differs from
conventional techniques is in the way the steam-generating heat is produced.
All forms of nuclear power generation produce heat through a controlled
nuclear fission process. Conventional reactors consume the fissile material
and promise to create upward pressure on the price of uranium as those
reserves become depleted: 'As. opposed to that, the not yet fully developed
breeder reactor actually generates more fuel than it consumes. The details
of these production technologies are well beyond the scope of this^report.
Suffice it to say that the pressurized water reactor was chosen as the most
164
-------
representative of conventional reactors and that its investment requirements
run about $175 per installed kilowatt; Its annualized cost is on the order
of 6.56 mills per kilowatthour. The breeder reactor would cost about
$240 per installed kilowatt and on an annual ized basis would cost about
6.46 mills per kilowatthour to operate. If fuel prices rise steeply and
if the breeder reactor becomes fully operational, its fuel economy could
obviously stimulate large-scale shifts toward this as the preferred power
production technique.
G.2.3 Summary
Table G.I summarizes the cost, operating, and availability character-
istics of some attractive alternatives that are expected to be available
for reducing sulfur emissions from power generation before the middle
of the next decade. Technical references of those projections are
also cited. For reference, the table also includes current estimates of
costs relating to conventional oil- and coal-fired power plants.
G.3 Theoretical Aspects of Producer Responses to the Emissions Tax
Over Time
Throughout this study it has been assumed that the power plant is
the decisionmaking unit and that the plant's output is determined exog-
enously. Consequently, it is assumed that the plant's objective is to
minimize costs subject to the output constraint. Allowing that constraint,
this section attempts to derive a simple model of the producer's decision
function in a dynamic setting. These general concepts are used in the
following section in conjunction with the results of the computerized model
and the data of table G.I to perform a preliminary analysis of the
implications inherent in the exclusive use of an emissions tax as the policy
instrument of choice in achieving emissions reductions.
G.3.1 A Simple Model of Cost Minimization Without Emissions Control
Pol i ci es
An existing power plant has been assumed in this report to require a
fixed flow of fuel heat input (Btut) each year. That heat input is the
product of the physical flow of fuel (F-t), sayMtons, and the heat content
(h.) per physical unit, say Btu's per ton. The total fuel input from the
^j
chosen fuel in any year, measured in Btu's, can be stated as
Btut = hjFjt (G.I)
165
-------
Table G.I.
Projected costs and availability of sulfur emissions
control alternatives for power production, 1980-1985
Alternative
Initial capital
reqiri rements*
($/k11owatt)
Variable costs
a
Anticipated
date of
Mills/kilo- Cents/ coronerdal b
watthour million Btu availability percent reference
Sulfur
emissions
reduction Techn1cal
Conventional power plants:
Coal f1redc
011 f1redc
Add-on technologies:
Citrate process
Double alkali process
Gasified coal
Advanced fossil -fueled plants:
Flu1d1zed bed combustion
Combined cycle (COGAS) system
Nuclear power plants:
Conventional reactor
Breeder reactor
175
168
39
24
0
190
127
175
240
6.22
6.94
1.95
1.75
6.30
5.20
6.56
6.46
Currently
Currently
1980
1980
60 1985
1983
1985
Currently
1985
None
None
90-95
90
99.5
92
99
100
100
e
e
f
f
9
h
e
1
1
aAll costs are estimated 1n current dollars; the estimates are for an average 500 MW plant.
These are based on comparisons with uncontrolled emissions from the combustion of 3 percent
sulfur coal.
cThese costs do not Include the capital and variable Input requirements of either sulfur or
partlculate control systems; the forms are given as the costs of add-on technologies..
Cost estimates do not Include credits for the sale of recovered sulfur.
eRobson, F.L., et al. Technological and Economic Feasibility of Advanced Power Cycles and
Methods of Producing Non-Polluting ruels for Utility Power Stratas, Final Report submitted to EPA
by VARL under Contract Number CPA 22-69-14, 1970.
Rochelle, G.T.,"A Critical Evaluation of Processes for the Removal of S02 from Power Plant
Gas", Paper prepared for A1r Pollution Control Association Meeting, June 1973.
9Edison Electric Institute, Fuels for the Electric Utility Industry 1971-85. Edison Electric
Institute, 1972.
hArcher, D.H., et al. Evaluation of the F1u1d1zed Bed Combustion Process. Final Report
submitted to EPA by Westlnghouse Research Laboratory, under Contract Number CPA 70-9, 1971.
Wtel, H.C. and J.B. Howard, Chapter 4, "Nuclear Power", 1n New Energy Technology; Some
Facts and Assessments. MIT Press, 1972.
166
-------
where
Btut = annual fuel heat Input 1n Btu's at time t ;
Fjt = physical flow of jth fuel Input at time t ;
h.. = Btu's per physical unit of the jth fuel.
If the plant heat rate (required Btu's per kilowatthour output) is constant
(TOD, one may alternatively express the required fuel input in terms of the
heat rate and output:
Btut = (R*)(kWht) (G>2)
where
FiR" = the plant heat rate (required Btu's per kilowatthour output)
kWht = required plant output in kilowatthours at time t.
By assuming that the decimal percent, by weight, of the jth fuel that
appears as volatized sulfur in the combustion off-gases is fixed at a., sulfur
emissions (SU-t) that would occur without control devices during any year can
be stated as:
SUjt = °j Fjt (G.3)
where
SlLt = annual sulfur emissions using the j fuel at time t ;
otj = decimal percentage sulfur content of the jth fuel, by weight.
If one assumes that the capital cost of the power plant is sunk and
that all variable costs besides fuel are proportional to the size, not output,
of the plant, then the volume of sulfur emissions may be regarded as
determined by the cost minimizing flow volume of fuel. If, for example, the
cheapest fuel is 3 percent sulfur coal and the plant requires f* tons to
meet its heat input requirements, then sulfur emissions are o^ F£, where k
is the index identifying 3 percent sulfur coal and where that product satisfies
the constraint of Eq. G.2. The plant has no economic incentive to reduce
emissions in the absence of emissions control policy.
G.3.2 Cost Minimization Over Time in the Presence of Emissions
Control Policies
Either a regulation or a tax on emissions will force the producer to
reconsider his emissions output decision. Assuming that there are m emission
control technologies—the kth one of which manifests an average decimal
percentage collection efficiency of ek—and further assuming that only one
167
-------
such control device can be in place at any point in time, the producer can
potentially achieve any one of k annual sulfur emission levels (E.kt) for
each fuel consumed where:
and where
ek = the sulfur collection efficiency of the
k control technology;
Ei-n- = annual sulfur emissions with k technology in place
J th
using the yn fuel at time t.
Further assuming that the investment cost (I) of any control option is
fixed and known in relation to the power plant's capacity (kW), one may
state the initial pollution control capital requirement as:
Ik = Ak kW (G.5)
where
KW = kilowatt capacity of the plant;
A. = investment cost per kilowatt for the k control option;
L = initial capital requirement for k control option.
Also by assuming that the variable costs (Vkt) of operating the k
sulfur removal system are proportional to plant output in kilowatthours per
year (kWh), these costs can be expressed as
where
Vkt = Bk kWht (6.6)
\t = annua^ variable costs associated with the k sulfur removal system;
Bk = variable costs per kilowatthour.
The annual tax bill (TAX) due from the power plant is the tax rate (e)
times the flow volume of sulfur emissions from the plant during the year.
™jkt= e Ejkf (G-7)
where
^i_ x, u
TAX..,. = annual tax bill using the j n fuel and ktn control
JKt ^
technology at time t ;
e = sulfur tax rate per ton of emitted sulfur.
168
-------
In the presence of a sulfur emissions tax, the cost minimizing plant
will make a concurrent decision regarding the appropriate combination of
fuel and emissions control devices. If, for every one of the n fuels
available, the decisionmaker anticipates a cost (P^ per Btu that will
obtain over a T-year planning horizon, the alternative anticipated annual
fuel bills (FUELt) for the plant will be
FUELjt = PjhjFjt or P.Btut (G.8)
where
FUEL-t = anticipated annual fuel bill using the jth fuel;
P. = anticipated price per Btu of the jth fuel over the
T-year planning horizon.
The producer will choose the combination of fuel costs and investment
outlays that minimizes the discounted present value (PV) of costs over that
planning horizon. One may state the discounted present value of the cost of
t~h th
using the j fuel and the k control technology T* periods from now,
assuming a current opportunity cost of capital of r dollars per dollar
(0 < r < 1), as:
T*
Fueljt
' (vkt + TAXjkt) (G-9>
t=T* ' ' V
where
T* = the number of periods from the current period when the
installation of the control device is anticipated;
r = the opportunity cost of capital.
The first two terms in this expression represent the discounted present
value of the pollution control capital and fuel outlays, respectively. The
third term is the discounted present value of tax payments on sulfur emissions
during the future period when no hardware controls are anticipated. The
169
-------
last term reflects the discounted present value of the variable costs of
operating the control device, once installed, and of the tax payments on
the emissions that would remain.
The producer will choose the minimum cost combination of fuels and
control hardware subject to the constraint, Eq. G.2, that the fuel input
equals that necessary to produce the required output. For a plant meeting
the constraint in Eq. G.2, one may alternatively express output using
Eq. G.I as
kWh. = 1= h.F.. (G.10)
1 HR J ^
By using Eqs. G.2 through G.7, the dynamic cost minimization objective
of Eq. G.9 may be written alternatively as
T / V4. T*
/ -i \ i i
PV,,. =
* E (w)VjFJt + E fey 9aJFJt
t=0 v ' t=0 V '
+ e(1-ekK F« (s-")
subject to: h. F.t = (HR)(kWhJ.
J J •• ••
Some general observations are already obvious from the model of
Eq. 6.11. First, deferring the investment an increasing number of years
(increasing T*) will reduce the discounted present value of capital costs,
the first term in Eq. G.ll. However, offsetting that is the fact that the
collection efficiency (ek) over those periods is zero, resulting in a higher
tax bill, the third term in Eq. G.ll. Furthermore, such a plan would
probably involve the choice of a low sulfur fuel whose price (P.) would
j
be higher than that of a fuel that is economical when sulfur control
technology is in place. This too would tend to increase the cost of
deferring the control investment by increasing the size of the second term.
On the other hand, the variable costs (B.) of operating a currently available
j
control device would be avoided entirely in the interim; i.e., the fourth
term is zero for the first T* periods. Finally, other things being equal,
170
-------
a higher opportunity cost of capital is likely, ceteris paribus, to induce
the deferment of investment in control technology, since that action would
reduce the impact of pollution control investment costs, the first term
in Eq. G.ll.
By using the data in table G.I the sensitivity of plant behavior to
variations in some of the parameters in Eq. G.ll is analyzed in section G.4.
The general theoretical framework of Eq. G.ll is used there to predict
plant responses under alternative assumptions regarding annualized control
costs, tax rates, collection efficiencies, and the expected number of years
(T*) from 1978 that will elapse before the control process is available.
G.4 A Preliminary Empirical Investigation of Sulfur Emission-Tax-Induced
Delays in the Removal of Sulfur Oxides from Stack Gases
To provide insight into the types of effects that time and new
technologies may have beyond those assumed for this study, three synthetic
plant models were developed. The hypothesized operating parameters of
these plants are given in table G.2. The table also reports approximate
Table G.2. Model plant parameters and projected SO removal costs
/\
Plant size Large, Medium, Small,
1500 MW 800 MM 350 MW
Parameters
Annual Btu Input
(billion Btu's)
Annual kllowatthours
output
(million hours)
Annual uncontrolled
emission rate
(tons)
Annual Ized cost of
control for processes
Wet limestone
Magnesia base
Citrate
Double alkali
110,678.0
13,176.0
134,739.0
:*
18.0
17.8
12.0
11.5
59,028.0
7,027.0
71,861.0
21.5
20.2
14.5
13.2
18,446.0
3,074.0
22,456.0
26.5
25.7
18.5
15.5
*Cents per million Btu.
Source: Research Triangle Institute.
171
-------
annualized costs for two control processes, wet limestone and magnesia base
scrubbing, that are attractive SO control options which were anticipated in
/v
other parts of this study. Also presented are anticipated annualized costs
for two new technologies, the citrate and double alkali processes, expected
to be available during the 1980's. The latter costs are extrapolations
of those reported in table G.I.
The analytical technique used in this portion of the study was first to
generate the costs of fuel switching only and the emissions tax payments that
would occur at different tax rates for each plant size. These costs were
projected using the basic computer model used throughout this study. Then
the computer model was run again for each of three hypothetical plants, this
time allowing the plant to institute cost minimizing combinations of either
magnesia base or wet limestone scrubbing and fuel switching and tax payments.
In following the assumption in other parts of this report that the
opportunity cost of capital is in the neighborhood of 12 percent, the "critical"
level of annualized cost of a new technology available T* years from the
present (for this analysis presumed to be 1978) was calculated as follows.
The present value of the costs of fuel switching and tax payments over T*
periods was computed. Similarly, the present value of the cost minimizing
combination of currently available control options (assumed to be wet limestone
and magnesia base), tax payments, and fuel switching was calculated. The
"critical" value of the annualized costs of new technology options was then
determined by division of the difference between the latter and former
discounted costs by the appropriate discount factor-
These concepts can be stated briefly in algebraic notation. Where the
annualized cost of presently available SO control options, including tax
* /\
payments, is denoted as PT; that of fuel switching and tax payments as FS;
the number of years until the new technology is available as T*; the
opportunity cost of capital as r; and the "critical" value of the annualized
cost of new technology and emissions tax payments as X, the point of
indifference between choosing currently existing options and new options
available T* years from the current year is defined by the following equality
(the assumed planning horizon was 15 years):
172
-------
z^-
Since output from the basic computer model generated FS and PT and
since r was assumed to be 0.12, the "critical" value (X) was calculated by
the following equation:
15 T*
X =
15
s
t=T* (6.13)
The value given by Eq. 6.13 represents the annualized cost of a new
technology plus annual emissions tax payments above which the cost minimizing
plant would choose to install immediately (1978) the present technology and
below which it would choose to wait T* periods until the new technology was
available. In the empirical work, the number of years the plant would have
to wait for new technology was varied over three assumed values: 2, 4, and
6 years. For each of those values of T* and for each tax rate for which FS
and PT were projected, Eq. G.I3 was used to calculate the "critical" value.
Since the control cost portion of X, the sum of control costs and tax
payments, depends upon the volume of emissions that remains after the appli-
cation of the new control technology, alternative assumptions concerning
the control efficiencies of new technologies were also made. The assumed
efficiencies of control for projected SO control technologies were 90, 95,
rt
and 99 percent. The emissions rates in table G.2 then allowed calculation
of the annual cost of tax payment at each tax rate and control efficiency.
This then yielded a net remainder available for annualized costs of owning,
maintaining, and operating the SO control hardware. Using the annual Btu
173
-------
Inputs, also reported in table G.2, finally allowed a computation Indicating
the maximum cost, in cents per million Btu, at which the deferment of current
investment in favor of future (cheaper) control processes would be more
economical than immediate installation of stack gas cleaning devices. The
loci of those critical costs are reported in figures G.I, G.2, and G.3 for
the large, medium, and small hypothetical plants, respectively.
An example of the meaning of the curves in figures G.I, G.2, and G.3
follows. Suppose that a new control process would be available by 1982,
4 years beyond 1978, at an approximate cost of 12 cents per million Btu
for a 1,500 MW power plant. Further assume that the process could achieve
95 percent collection efficiencies. Would the plant wait 4 years to install
the new process and pay the emissions tax penalities in the interim? If
so, over what ranges of tax rates? The middle panels (B) of the figures
answer those questions. For tax rates below about 7 cents and above
approximately 19 cents per pound of emitted sulfur, the hypothetical plant
would choose to follow current options; i.e., to install one of the currently
available technologies immediately. Between those anticipated tax rates,
it would choose to wait 4 years until the new technology is available.
This method can be applied for any number of options among the subject
plant sizes.
The cause of the peaking over the mid-range of taxes (B panels) in
all three figures is that the ratios of PT to FS (annualized present technology
and fuel switching costs, respectively) in Eq. G.13 follows that same pattern,
by causing the corresponding values of X and, in turn, the critical values
presented in those figures to follow the pattern displayed there.
It is interesting to compare the costs of the two new technologies
presented in table G.2 against these figures. Recall that the citrate
process promises control efficiencies over the range of 90 to 95 percent
while the double alkali process is expected to manifest a control efficiency
of about 90 percent. Both are expected to be available by 1980, within 2
years of the beginning of the planning horizon. Using the cost data of
table G.2, one can determine from figure G.I that large plants (1,500 MW)
would be expected to wait until those processes were available for any tax
174
-------
Ul
16-
0
z
o
DC
111
0.
at
UJ
u
V)
o
u
o
111
N
90
5 10 15 20 25 30
TAX RATE (CENTS PER POUND)
PERCENT COLLECTION EFFICIENCY
5 10 15 20 25 30
TAX RATE (CENTS PER POUND)
95 PERCENT COLLECTION EFFICIENCY
5 10 15 20 25 30
TAX RATE (CENTS PER POUND)
99 PERCENT COLLECTION EFFICIENCY
Figure 6.1. Loci of anticipated costs of new SOX control processes (available in T* years
beyond 1978) below which a cost minimizing 1500 MW model power plant would defer SOX stack
gas cleaning.
-------
5 10 15 20 25 30
TAX RATE (CENTS PER POUND)
90 PERCENT COLLECTION EFFICIENCY
_L
I
I
I
I
5 10 15 20 25 30
TAX RATE (CENTS PER POUND)
95 PERCENT COLLECTION EFFICIENCY
I
I
I
5 10 15 20 25 3O
TAX RATE ( CENTS PER POUND)
99 PERCENT COLLECTION EFFICIENCY
Figure G.2. Loci of anticipated costs of new SOX control processes (available in T* years beyond
1978) below which a cost minimizing 800 MW model power plant would defer SOX stack gas cleaning.
-------
24 -
5 10 IS 20 25 30
TAX RATE (CENTS PER POUND)
90 PERCENT COLLECTION EFFICIENCY
5 10 IS 20 25 30
TAX RATE (CENTS PER POUND)
95 PERCENT COLLECTION EFFICIENCY
5 10 15 20 25 30
TAX RATE (CENTS PER POUND)
99 PERCENT COLLECTION EFFICIENCY
Figure G.3. Loci of anticipated costs of new SOX control processes (available in T* years beyond
1978) below which a cost minimizing 350 MW model power plant would defer SO stack gas cleaning.
X
-------
rates in excess of about 7 cents per pound of emitted sulfur at 90 percent
efficiencies and for even lower tax rates if the control efficiency ranges
up to 95 percent. Figures G.2 and G.3 indicate that virtually the same
results would hold for medium- and small-sized plants. The choice of deferring
or not deferring appear, for all three sizes, to be very sensitive in the
range of tax rates between 5 and 10 cents, both reasonable values that have
been suggested by some officials as feasible tax rates.
Quite obviously, this analysis is highly dependent on an array of
assumptions about plant location, expected fuel prices, estimated costs of
control options anticipated in 1978, and a multitude of other parameters.
Yet it is useful in that orders of magnitude are identified and in that the
directional affects of the time until new processes are available, of
collection efficiencies, of tax rates, and of new source control costs
are identified and, at least roughly, quantified. More complete investigations
will await better refined cost models for new control options, improved
knowledge of anticipated fuel costs, and intensive microanalysis of the ways
in which firms, in practice, respond to uncertainty in environmental control
policy parameters.
178
-------
APPENDIX H: STATE-BY-STATE PROJECTIONS OF THE EFFECTIVENESS
AND COSTS OF A TAX ON SULFUR EMISSIONS
The primary purpose of this study was to provide an initial
projection of the effectiveness and costs of a uniform national tax
on sulfur emissions. As demonstrated in ch. 5, the effects of such
a tax on local air quality can be expected to vary from region to
region across the country depending on local climatic conditions
and on the type, size, and geographic distribution of polluting
sources. Since the emissions response model developed for this
study uses a listing of discrete sulfur emissions sources, it is
possible to aggregate the behavior of the sources below a national
level to obtain regional projections of effectiveness and costs.
This has been done on a State-by-State basis in this appendix to
provide insight regarding the possible regional differences in
responses to a tax on sulfur emissions. The reader is cautioned,
however, that these projections are very preliminary.
The projected State-by-State responses to a tax on sulfur emissions
are shown in table H.I. Comparison of the projected responses for four
States chosen for expositional purposes follows.
The cost of sulfur emissions control varies significantly from
State-to-State due to differences in the type, number, and size of
the emissions sources present. Marginal cost of control curves for
Alabama, Indiana, New York, and Texas utilizing the tabular data are
presented in figure H.I. The amount of reduction in emissions induced
by the tax will vary considerably among these four States. For example,
for a marginal cost of $300 per ton, sources in Texas could reduce
emissions by 175,000 tons annually; those in Indiana, 700,000 tons
annually. The remaining emissions for each State for alternative tax
rates are shown in figure H.2.
179
-------
600
500
400
300
200
100
•TEXAS
•LMC I IMC
.•ALABAMA i NEW YORK
f LMC
1 MMANA
100 200 300 400 500 600 700 800 900
REDUCTIONS IN SULFUR EMISSIONS
(THOUSAND TONS PER YEAR)
Figure H.I. Marginal cost of reductions in sulfur
emissions from all major sources combined in
selected States. (Source: Research Triangle
Institute).
15 20 25
TAX RATE
(CENTS PER POUND OF SULFUR EMISSIONS)
Figure H.2. Effectiveness of a tax on the sulfur
emissions from all major sources combined in
selected States. (Source: Research Triangle
Institute).
30
180
-------
Table H.I.
Projected State-by-State responses to a national
tax on sulfur emissions
State
A1 abama
Alaska
Arizona
Arkansas
California
Colorado
Tax rate
(cents per
pound of
sulfur
emissions)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
Emissions
(thousand
tons)
482
262
146
135
128
83
58
1
1
24
1
1
0
0
764
175
38
35
30
30
30
13
12
6
5
4
4
4
212
120
90
75
70
69
68
73
53
32
22
19
18
16
Reductions
1n emissions
from zero tax
(thousand
tons)
_
220
336
348
355
399
424
-
0
723
0
0
1
1
-
589
725
728
734
734
734
-
1
7
8
9
9
9
_
92
122
137
142
143
144
_
20
41
50
53
54
56
Total
annual cost
(thousands)
_
40,744
1,446
88,023
105,421
120,934
131,707
_
92
24,514
232
300
348
392
-
37,164
42,757
46,556
49,977
53,029
55,918
_
1,250
2,222
2,826
3,316
3,770
4,216
_
17,527
30,210
39,825
48,527
56,771
64,721
_
6,207
10,443
13,795
16,413
18,637
20,745
Annual 1 zed
control costs
(thousands)
_
24,070
766
57,227
63,861
89,034
106,388
_
0
19,753
28
28
127
127
-
19,881
35,348
36,076
40,753
38,378
38,512
-
195
931
1,217
1,583
1,660
1,694
-
7,466
14,816
19,294
22,175
24,651
26,340
-
2,336
5,563
8,255
10,410
10,848
13,061
Annual tax
payment
(thousands)
_
16,674
670
30,616
41 ,560
31 ,900
25,319
-
92
4,761
204
272
221
266
-
17,183
7,410
10,480
9,224
14,651
17,506
-
1,415
1,291
1,609
1,732
2,110
2,523
-
10.061
15,995
20,531
26,352
32,119
38,381
-
3,871
4,880
5,540
6,002
7,789
7,683
181
-------
Table H.I.
Projected State-by-State responses to a national
tax on sulfur emissions (con.)
State
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Tax rate
(cents per
pound of
sulfur
emissions)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
Emissions
(thousand
tons)
131
79
37
32
32
26
24
70
51
28
25
21
19
18
35
21
10
9
9
9
9
318
284
129
94
72
58
54
212
127
76
56
49
32
31
14
14
6
2
2
2
2
Reductions
In emissions
from zero tax
(thousand
tons)
—
52
95
100
100
105
107
_
19
42
45
49
51
52
_
14
25
26
26
26
26
_
142
296
331
353
368
372
-
85
136
156
163
181
182
-
0
8
12
12
12 •
12
V
Total
annual cost
(thousands)
$ -
17,482
30,777
42 ,280
53,420
64,135
74,265
«.
6,416
10,593
13,664
16,247
18,637
20,878
—
3,737
5,777
7,572
9,266
10,917
12,528
_
33,998
54,528
66,930
75,727
82,610
88,965
_
15,428
26,467
34,310
40,691
45,418
49,725
_
1,377
2,243
2,694
2.078
3,116
3,323
Annual 1 zed
control costs
(thousands)
$
12,233
26,097
35,251
45,759
53,494
62,392
—
2,473
6,187
7,590
9,404
10,557
11,617
—
2,361
4,418
5,344
6,199
6,525
7,793
_
12,145
35,153
45,114
53,356
59,890
63,215
_
6,836
15,372
21 ,682
24,905
33,236
35,215
_
0
1,014
1,995
830
2,078
2,078
Annual tax
payment
(thousands)
$ -
5,249
4,680
7,029
9,958
10,641
11,873
—
3,944
4,406
6,074
6,843
8,080
9,260
—
1,376
1,359
2,528
3,067
4,392
4,735
—
21 ,853
19,375
21,817
22,371
22,720
25,750
_
8,592
11,095
12,628
15,786
12,182
, 14,510
_
1 ,377
1,229
699
2,075
1,038
1,245
182
-------
Table H.I.
Projected State-by-State responses to a national
tax on sulfur emissions (con.)
State
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Tax rate
(cents per
pound of
sulfur
emissions)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
Emissions
(thousand
tons)
3
2
1
1
1
1
1
823
402
228
218
196
172
137
901
378
256
198
172
137
131
84
55
35
25
19
18
17
45
36
32
25
24
24
24
633
272
157
149
135
94
53
Reductions
1n emissions
from zero tax
(thousand
tons)
0
1
2
2
2
2
2
-
421'
595
609
627
651
686
-
523
644
702
728
763
770
_
29
49
59
65
66
67
_
10
14
20
21
22
22
361
476
484
498
538
579
Total
annual cost
(thousands)
$ 0
214
324
417
500
566
624
-
68,796
115,228
154,442
191,225
225 ,668
253,407
-
82,278
139,086
184,481
224,521
260,845
292 ,003
-
7,066
12,401
16,721
19,793
22,513
25,096
_
4,332
10,148
11,837
14,891
17,828
20,728
34,248
75,820
75,820
92,135
104.588
112,863
Annual 1 zed
control costs
(thousands)
$ 0
89
154
222
221
332
550
-
46,436
85,384
104,833
128,760
155,884
187,161
-
62,071
119,511
142,832
173,594
210,256
231,529
-
3,216
6,958
10,573
13,542
15,078
16,504
-
1,243
2,566
4,658
5,688
6,603
7,234
19,514
39,843
43,534
50.789
69,904
93,697
Annual tax
payment
(thousands)
$ 0
125
170
195
279
234
74
-
24,361
29,744
49,608
62 ,465
69,784
66,245
-
19,907
33,359
41 ,650
50,926
50,589
60,474
-
3,850
5,443
6,148
6,251
7,435
8,592
-
3,089
5,815
7,180
9,202
11,225
13,494
14,734
18,734
32,287
41,346
34,683
19,166
183
-------
Table H.I.
Projected State-by-State responses to a national
tax on sulfur emissions (con.)
State
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Tax rate
(cents per
pound of
sulfur
emissions)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10'
15
20
25
30
Reductions
Emissions 1n emissions Total
(thousand from zero tax annual cost
tons) (thousand (thousands)
tons)
118
108
59
55
52
51
51
23
14
6
5
5
4
4
318
171
104
89
72
56
49
352
170
107
75
69
55
53
779
313
M87
164
122
111
108
248
147
77
64
56
44
36
_
10
59
63
66
66
66
_
9
16
16
17
17
18
_
147
215
230
247
263
269
_
181
245
277
283
297
299
_
466
591
615
657
669
672
-
101
170
184
192
203
212
$
11,085
18,212
23,871
29,183
34,360
39,527
_
2,072
3,253
4,111
4,881
5,573
6,196
_
23,403
40,709
53,625
63,976
72,550
80,127
_
26,651
41,456
52,370
61 ,248
69,091
76,298
-
73,496
120,400
159,530
192,488
222,595
250,965
-
15,333
26,079
33,542
40,117
45,355
50,122
Annual i zed
control costs
(thousands )
$ -
441
6,571
7,592
8,582
8,740
8,742
w
1,151
2,520
2,827
3,348
3,719
4,304
_
12,574
26,245
33,278
41,579
51,114
56,759
-
16,644
27,160
37,011
24,124
48,711
51,182
-
55,194
95,981
123,584
156,595
180,579
474,297
-
5,449
15,416
19,179
22,847
27,878
33,722
Annual tax
payment
(thousands)
$ -
10,645
11,642
16,279
20,600
25,621
30,786
_
921
733
1,284
1,533
1,854
1,892
_
10,829
14,464
20,347
22,397
21,436
23,368
-
10,007
14,298
15,359
37,124
20,380
25,116
-
18,303
24,420
35,948
35,893
42,016
76,666
_ fi' •
9,884
10,663
14,363
17,270
17,477
16,399
184
-------
Table H.I.
Projected State-by-State responses to a national
tax on sulfur emissions (con.)
State
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
Tax rate
(cents per
pound of
sulfur
emissions)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
Emissions
(thousand
tons)
28
20
16
15
13
13
13
318
196
117
87
74
51
46
303
38
35
35
31
31
31
133
24
16
12
9
9
8
117
17
12
12
11
9
9
84
54
29
20
17
11
11
Reductions
1n emissions Total Annual 1 zed
from zero tax annual cost control costs
(thousand (thousands) (thousands)
tons)
.
7
12
13
15
15
15
-
122
201
231
244
267
272
_
268
270
271
275
275
275
-
109
117
121
124
124
125
_
100
105
105
106
108
108
-
31
56
65
68
73
73
$ - $
2,071
3,816
5,302
6,655
7,890
9,110
-
27,218
46,084
40,147
70,037
78,376
85,442
_
18.509
- 22.221
25,771
29,027
32.066
35.093
-
5,682
7,649
9,144
10,306
11,374
12,384
_
* i, 6, 654
" 8,474
9.956
11.382
12,632
13,819
-
5,542
9,649
12,072
13,910
15,363
16,628
_
370
1,106
1,236
1,236
2,115
2,120
-
13,075
28,261
38,506
45,698
58,425
63,523
-
14,842
14,935
15,509
16.884
17,059
17.059
-
4,071
5,228
6,302
7,267
7,749
8,156
-
5,433
6,568
6,866
7,557
5,606
8,943
-
1,932
5,524
7,737
8,853
11,245
11,409
Annual tax
payment
(thousands)
$ -
1,701
2,711
4,066
4,681
5,775
6,990
_
14,143
17,824
20,567
24,339
19,951
21,919
_
3,369
7,207
10,261
12,148
15,007
18,034
_
1,610
2,421
2,842
3,038
3,625
4,228
-
1.222
1,906
3,091
3,826
2,637
4,876
-
3,610
4,125
4,335
5,057
4,110
5,219
185
-------
Table H.I.
Projected State-by-State responses to a national
tax on sulfur emissions (con.)
State
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Tax rate
(cents per
pound of
sulfur
emissions)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
Emissions
(thousand
tons)
354
209
123
104
88
82
70
305
150
111
75
64
34
32
671
318
197
171
155
127
123
1,656
636
270
260
251
129
123
54
39
27
21
17
15
14
1,172
699
392
276
241
204
193
Reductions
in emissions
from zero tax
(thousand
tons)
_
145
231
250
266
272
284
_
256
295
330
331
371'
373
_
353
474
500
516
545
549
_
1,020
1,186
1,396
1,405
1,527
1,533
_
, 16
27
33
37
40
40
-
473
780
896
931
968
979
Total
annual cost
(thousands)
$ -
32,249
52,500
67,330
79,780
90,570
100,030
_
30,135
46,852
58,112
67,495
74,200
79,632
_
57,955
91 ,751
119,748
144,515
166,082
185,769
_
686
235,974
297,306
346,194
372,430
396,268
—
3,861
7,088
9,420
11,265
12,713
14,057
_
112,033
187,572
241 ,353
288,813
331,615
370,038
Annual i zed
control costs
(thousands)
$
17,339
33,755
41 ,901
50,247
55,558
63,840
_
21 ,082
30,403
41 ,300
83,786
62,818
65,855
— p
39,053
71,111
81 ,228
95,310
115,371
124,902
_
108,572
175,168
252,561
279,036
341 ,036
352,373
_
1,044
2,893
4,154
5,435
6,989
7,075
_
64,943
132,062
181,590
215,586
252,637
277,584
Annual tax
payment
(thousands)
$
14,910
18,745
25,430
29,533
35,012
36,191
9,053
16,452
16,812
19,908
11,381
13,776
_
18,903
26,640
38,520
49,205
50,711
60,867
_
234
60,806
44,745
67,158
31 ,454
40,895
«•
2,817
3,985
5,266
5,830
5,724
6,982
_
47,090
55,510
59,763
73,227
78,978
92,454
186
-------
Table H.I.
Projected State-by-State responses to a national
tax on sulfur emissions (con.)
State
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tax rate
(cents per
pound of
sulfur
emissions)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
Emissions
(thousand
tons)
72
24
21
20
19
19
17
3
2
1
1
1
1
1
1,482
707
420
346
280
234
215
24
12
7
7
7
6
6
306
148
83
66
59
53
38
13
9
6
6
4
4
4
Reductions
1n emissions
from zero tax
(thousand
tons)
_
48
50
52
53
54
54
-
1
2
2
2
2
2
775
1,062
1,136
1,202
1,248
1,267
-
12
12
17
17
18
18
_
158
223
240
247
253
268
-
4
7
7
9
9
9
Total
annual cost
(thousands)
$ -
4,035
6,311
8,400
10,319
12,170
13,992
-
280
405
509
606
704
795
139,875
228,005
291 ,985
346,884
395,205
438,083
-
2,224
3,168
4,033
4,826
5,582
6,292
!••' '
27,337
45,109
56,824
67,097
76,207
82 ,836
-
954
1,649
2,218
2,676
3,092
3,492
Annual 1 zed
control costs
(thousands)
$
2,010
2,221
2,596
2,947
3,152
3,202
-
131
222
237
245
244
280
97,959
172,716
216,984
263,623
306,864
337,766
-
1,462
2,285
2,605
2,805
3,152
3,281
-
18,708
34,569
43,376
49,884
55,723
65,837
-
340
690
861
1,351
1,471
1,496
Annual tax
payment
(thousands)
$ -
2,337
4,400
11,382
7,373
9,018
10,791
-
149
149
238
327
426
481
41,915
55,288
75,001
83,261
88,341
100,318
.
762
883
1,429
2,021
2,430
3,011
-
8,629
10,540
13,448
17,213
20,484
16,999
-
614
959
1,357
1,325
1,621
1,996
187
-------
Table H.I.
Projected State-by-State responses to a national
tax on sulfur emissions (con.)
State
Tennessee
Texas
Utah
Vermont
Virginia
Washington
Tax rate
(cents per
pound of
sulfur
emissions)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
Emissions
(thousand
tons)
220
192
142
140
137
136
47
305
181
150
137
134
133
132
34
33
17
17
14
12
11
5
2
1
1
1
1
1
213
119
74
63
55
45
41
120
55
21
17
17
17
17
Reductions
1n emissions
from zero tax
(thousand
tons)
.»
34
83
86
89
90
178
—
125
156
168
171
172
183
_
0
17
17
20
22
23
_
2
3
4
4
4
4
-
94
151
151
159
169
193
_
65
98
102
102
103
103 '
Total
annual cost
(thousands )
$ -
17,811
33,201
47,200
60,982
74,612
84,178
v
23,897
40,077
54,249
67,700
80,994
94,201
_
2,869
5,251
7,002
8,499
9,666
10,730
_
298
450
583
699
800
891
-
19.876
47,547
47,547
58,308
66,849
74,124
_
7,606
10,300
12,294
14,110
15,921
17,712
Annual 1 zed
control costs
(thousands)
$
2,886
8,888
9,784
10,576
10,841
60,506
_
6,184
10,629
13,551
14,751
14,978
15,375
.
131
239
346
3,666
4,450
5,140
_
153
292
337
366 •
387
511
-
12,046
30,179
30,179
40,132
48,368
53,640
-
2,433
6,027
7,006
7,209
7,270
7,377
Annual tax
payment
(thousands)
$ - .
14,925
24,989
37,416
37,416
63,766
23,672
_
17,713
29,448
40,697
52,949
66,015
78,819
_
2,738
2,950
4,594
4,833
5,216
5,590
_
145
158
246
333
413
380
-
7,829
15,104
15,104
18,176
18,581
20,483
-
5,172
4,055
5,098
7,001
8,651
10,335
188
-------
Table H.I.
Projected State-by-State responses to a national
tax on sulfur emissions (con.)
State
West Virginia
Wisconsin
Wyoming
Tax rate
(cents per
pound of
sul fur
emissions)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
0
5
10
15
20
25
30
Emissions
(thousand
tons)
558
249
147
132
124
79
55
383
189
101
84
72
55
49
78
57
37
30
31
22
20
Reductions
in emissions
from zero tax
(thousand
tons)
—
309
410
426
434
479
503
_
194
282
300
312
329
335
_
22
42
57
51
58
59
Total
annual cost
(thousands)
$ -
34,520
59,143
75,361
90,032
102,197
110,057
—
27,738
45,874
58,415
69,119
78,690
86,339
_
7,232
12,510
15,351
18,332
20,849
22,929
Annuall zed
control costs
(thousands)
$
20,764
40,688
46,825
51,318
73,544
87,816
_
16,468
33,416
41,185
47,855
59,020
64,610
_
242
1,541
2,648
2,923
2,941
3,333
Annual tax
payment
(thousands)
$
13,756
18,455
28,536
38,715
28,654
22,242
_
11,270
12,458
17,230
21,165
19,670
21,729
_
5,650
7,137
8,969
11,212
10,275
11,563
Source: Research Triangle Institute.
189
-------
BIBLIOGRAPHIC DATA
SHEET
1. K'.por. N...
EPA-600/5-74-009
i; rent's Acccssii
i ;:It- jr..) Suh: ;t!t
Cost-Effectiveness of a Uniform National Sulfur
Emissions Tax
S. Kfj.tr: Date
November 1973
7. A^horfs, x. H. Bingham; P. C. Cooley; M. E. Fogel; D. R. Johnson;
D. A. LeSourd; A« K. Miedema; R. E. Paddock; M. Simons. Jr. •
8. Per: >:r.ir.p Orpanizuun i-... ;•:
No. 41U-757
9. PiTtorniint; Organization Name anj AJJress
Research Triangle Institute
Research Triangle Park, North Carolina 27709
M. M. Wisler
10. Project T.isk 'iork tn:: N.
09AFD 03
11. Contract Grant No.
68-01-0426
12. Sponsoring Organization Name and Addr
Washington Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Washington. D.C. 20460
13. Type of Report & Perioci
Coverej
Final Report
14.
15. Supplementary Notes
16. Abstracts
The study's main objectives are (1) to calculate the relationship between
specified levels of a uniform national sulfur emissions tax, and the resulting
reduction in projected emissions from primary stationary source emitters; (2) to
estimate the costs of emission-reduction activities that ensue as a consequence
of a given emission-tax level; and (3) to determine the tax revenue generated
in light of alternative emission tax levels, and hence the total gross environ-
mental costs imposed on sulfur emitters.
17. Key %'ords and Document Analysis. 17o. Descriptors
Sulfur Oxides, sulfur emissions tax, air quality, sulfur emissions control
costs, sulfur emissions abatement
17b. Identifiers /Open-Ended Terms
17c. COSAT1 FieM/Group
18. Availability Statement
Release unlimited
19. Security Class (This
Report)
I NT.i.ASSlFlFll
20. Security Class.(I'nis
Papt
21. No. of
189
22. Price
FORM NTIS-35 (RtV,
THIS FORM MAY BE REPRODUCED
USCOMM-OC
*U.S. GOVERNMENT PRINTING OFFICE: 1974 546-317/310 1-3
------- |