vvEPA
United States
Environmental Protection
Agency
Office of Air and Radiation
Office of Policy, Planning and
Evaluation
EPA-410-R-97-002
October 1997
The Benefits and Costs of
the Clean Air Act
1970 to 1990
EPA Report to Congress
Internet Address (URL) ••http://www.epa.gov
Recycled/Recyclable • Printed with Vegetable Oil Based Inks on Recycled Paper (Minimum 30% Postconsumer)
-------
-------
The Benefits and Costs of the
Clean Air Act, 1970 to 1990
Prepared for
U.S. Congress
by
U.S. Environmental Protection Agency
October 1997
-------
-------
Abstract
Section 812 of the Clean Air Act Amendments of 1990 requires the Environmental Protection Agency
(EPA) to periodically assess the effect of the Clean Air Act on the "public health, economy, and environment of
the United States," and to report the findings and results of its assessments to the Congress. Section 812 further
directs EPA to evaluate the benefits and costs of the Clean Air Act's implementation, taking into consideration
the Act's effects on public health, economic growth, the environment, employment, productivity, and the economy
as a whole. This EPA Report to Congress presents the results and conclusions of the first section 812 assess-
ment, a retrospective analysis of the benefits and costs of the Clean Air Act from 1970 to 1990. Future reports
will detail the findings of prospective analyses of the benefits and costs of the Clean Air Act Amendments of
1990, as required by section 812.
This retrospective analysis evaluates the benefits and costs of emissions controls imposed by the Clean Air
Act and associated regulations. The focus is primarily on the criteria pollutants sulfur dioxide, nitrogen oxides,
carbon monoxide, particulate matter, ozone, and lead since essential data were lacking for air toxics. To deter-
mine the range and magnitude of effects of these pollutant emission reductions, EPA compared and contrasted
two regulatory scenarios. The "control scenario" reflects the actual conditions resulting from the historical
implementation of the 1970 and 1977 Clean Air Acts. In contrast, the "no-control" scenario reflects expected
conditions under the assumption that, absent the passage of the 1970 Clean Air Act, the scope, form, and
stringency of air pollution control programs would have remained as they were in 1970. The no-control scenario
represents a hypothesized "baseline" against which to measure the effects of the Clean Air Act. The differences
between the public health, air quality, and economic and environmental conditions resulting from these two
scenarios represent the benefits and costs of the Act's implementation from 1970 to 1990.
To identify and quantify the various public health, economic, and environmental differences between the
control and no-control scenarios, EPA employed a sequence of complex modeling and analytical procedures.
Data for direct compliance costs were used in a general equilibrium macroeconomic model to estimate the
effect of the Clean Air Act on the mix of economic and industrial activity comprising the nation's economy.
These differences in economic activity were used to model the corresponding changes in pollutant emissions,
which in turn provided the basis for modeling resulting differences in air quality conditions. Through the use of
concentration-response functions derived from the scientific literature, changes in air quality provided the basis
for calculating differences in physical effects between the two scenarios (e.g, reductions in the incidence of a
specific adverse health effect, improvements in visibility, or changes in acid deposition rates). Many of the
changes in physical effects were assigned an economic value on the basis of a thorough review and analysis of
relevant studies from the economics, health effects, and air quality literature. The final analytical step involved
aggregating these individual economic values and assessing the related uncertainties to generate a range of
overall benefits estimates.
Comparison of emissions modeling results for the control and no-control scenarios indicates that the Clean
Air Act has yielded significant pollutant emission reductions. The installation of stack gas scrubbers and the use
of fuels with lower sulfur content produced a 40 percent reduction in 1990 sulfur dioxide emissions from elec-
tric utilities; total suspended particulate emissions were 75 percent lower as a result of controls on industrial and
utility smokestacks. Motor vehicle pollution controls adopted under the Act were largely responsible for a 50
percent reduction in carbon monoxide emissions, a 30 percent reduction in emissions of nitrogen oxides, a 45
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
percent reduction in emissions of volatile organic compounds, and a near elimination of lead emissions. Several
of these pollutants (primarily sulfur dioxide, nitrogen oxides, and volatile organic compounds) are precursors
for the formation of ozone, particulates, or acidic aerosols; thus, emissions reductions have also yielded air
quality benefits beyond those directly associated with reduced concentrations of the individual pollutants them-
selves.
The direct benefits of the Clean Air Act from 1970 to 1990 include reduced incidence of a number of
adverse human health effects, improvements in visibility, and avoided damage to agricultural crops. Based on
the assumptions employed, the estimated economic value of these benefits ranges from $5.6 to $49.4 trillion, in
1990 dollars, with a mean, or central tendency estimate, of $22.2 trillion. These estimates do not include a
number of other potentially important benefits which could not be readily quantified, such as ecosystem changes
and air toxics-related human health effects. The estimates are based on the assumption that correlations between
increased air pollution exposures and adverse health outcomes found by epidemiological studies indicate causal
relationships between the pollutant exposures and the adverse health effects.
The direct costs of implementing the Clean Air Act from 1970 to 1990, including annual compliance expen-
ditures in the private sector and program implementation costs in the public sector, totaled $523 billion in 1990
dollars. This point estimate of dkect costs does not reflect several potentially important uncertainties, such as
the degree of accuracy of private sector cost survey results, that could not be readily quantified. The estimate
also does not include several potentially important indirect costs which could not be readily quantified, such as
the possible adverse effects of Clean Air Act implementation on capital formation and technological innova-
tion.
Thus, the retrospective analysis of the benefits and costs of implementing the Clean Air Act from 1970 to
1990 indicates that the mean estimate of total benefits over the period exceeded total costs by more than a factor
of 42. Taking into account the aggregate uncertainty in the estimates, the ratio of benefits to costs ranges from
10.7 to 94.5.
The assumptions and data limitations imposed by the current state of the art in each phase of the modeling
and analytical procedure, and by the state of current research on air pollution's effects, necessarily introduce
some uncertainties in this result. Given the magnitude of difference between the estimated benefits and costs,
however, it is extremely unlikely that eliminating these uncertainties would invalidate the fundamental conclu-
sion that the Clean Air Act's benefits to society have greatly exceeded its costs. Nonetheless, these uncertainties
do serve to highlight the need for additional research into the public health, economic, and environmental
effects of air pollution to reduce potential uncertainties in future prospective analyses of the benefits and costs
of further pollution controls mandated by the Clean Air Act Amendments of 1990.
-------
Contents
Tables xi
Figures xv
Acronyms and Abbreviations ^vii
Acknowledgments . xxm
Executive Summary ES-1
Purpose of the Study ES-1
Study Design ES-1
Study Review ES-1
Summary of Results ES-2
Direct Costs ES-2
Emissions ES-2
Air Quality ES-3
Physical Effects ES-5
Economic Valuation ES-7
Monetized Benefits and Costs ES-8
Alternative Results .' ES-9
Conclusions and Future Directions ; ES-9
Chapter 1: Introduction 1
Background and Purpose 1
Clean Air Act Requirements, 1970 to 1990 1
Section 812 of the Clean Air Act Amendments of 1990 2
Analytical Design and Review 2
Target Variable 2
Key Assumptions 2
Analytic Sequence 3
Review Process 6
Report Organization 6
Chapter 2: Cost and Macroeconomic Effects 7
Direct Compliance Costs 7
Indirect Effects of the CAA !!"!""!!!"!'."!."!!."!!""!!!"! 9
Sectoral Impacts 9
Aggregate Effects 9
Uncertainties and Sensitivities in the Cost and Macroeconomic Analysis 10
Productivity and Technical Change 10
Discount Rates \\
Exclusion of Health Benefits from the Macroeconomic Model 12
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Chapter3: Emissions 13
Sector-Specific Approach 15
Summary of Results 15
Uncertainty in the Emissions Estimates 17
Chapter 4: Air Quality 19
General Methodology 20
Sample Results 21
Carbon Monoxide . 21
Sulfur Dioxide , 22
Nitrogen Dioxide 22
Particulate Matter 23
Ozone 23
Urban Ozone 23
Rural Ozone 24
Acid Deposition 24
Visibility 25
Uncertainty in the Air Quality Estimates 25
Chapter 5: Physical Effects 29
Human Health and Welfare Effects Modeling Approach 29
Air Quality 29
Population 29
Health and Welfare Effects 29
Key Analytical Assumptions 30
Mapping Populations to Monitors 32
Choice of Study .. 33
Variance Within Studies 33
PM-Related Mortality 34
Short-Term Exposure Studies 34
Long-Term Exposure Studies 35
Health Effects Modeling Results 37
Avoided Premature Mortality Estimates 37
Non-Fatal Health Impacts 37
Other Physical Effects 38
Ecological Effects 38
Aquatic and Forest Effects 38
Quantified Agricultural Effects 39
Effects of Air Toxics 39
Uncertainty in the Physical Effects Estimates 41
Chapter 6: Economic Valuation 43
Methods for Valuation of Health and Welfare Effects .43
Mortality 44
Survey-Based Values 45
Chronic Bronchitis 45
Respiratory-Related Ailments 46
Minor Restricted Activity Days 46
Visibility 46
Avoided Cost Estimates 46
Hypertension and Hospital Admissions 46
Household Soiling 47
-------
Contents
Other Valuation Estimates 47
Changes in Children's IQ 47
Work Loss Days and Worker Productivity 48
Agricultural Benefits 48
Valuation Uncertainties 48
Mortity Risk Benefits Transfer 48
Chapter 7: Results and Uncertainty 51
Quantified Uncertainty in the Benefits Analysis 51
Aggregate Monetized Benefits 52
Comparison of Monetized Benefits and Costs 55
Major Sources of Uncertainty 56
Alternative Results 57
PM Mortality Valuation Based on Life-Years Lost 57
Alternative Discount Rates 58
Appendix A: Cost and Macroeconomic Modeling A-l
Introduction A-l
Macroeconomic Modeling A-l
Choice of Macroeconomic Model A-2
Overview of the Jorgenson-Wilcoxen Model A-2
Structure of the Jorgenson-Wilcoxen Model A-3
The Business Sector , A-4
The Household Sector '. . A-4
The Government Sector A-5
The Rest-of-the-World Sector A-5
Environmental Regulation, Investment, and Capital Formation A-5
The General Equilibrium A-5
Configuration of the No-control Scenario A-6
Elimination of Compliance Costs in the No-Control Case A-7
Capital Costs - Stationary Sources A-7
Operating and Maintenance Costs - Stationary Sources A-8
Capital Costs - Mobile Sources A-8
Operating and Maintenance - Mobile Sources A-8
Direct Compliance Expenditures Data A-8
Sources of Cost Data A-8
Cost of Clean Data A-8
EPA Data ; A-8
Commerce Data A-9
Stationary Source Cost Data.. A-9
Capital Expenditures Data A-9
Operation and Maintenance Expenditures Data A-10
Recovered Costs A-10
Mobile Source Cost Data A-ll
Capital Expenditures Data A-ll
Operation and Maintenance Expenditures Data A-ll
Fuel Price Penalty A-ll
Fuel Economy Penalty A-l2
Inspection and Maintenance Programs A-13
Maintenance Credits A-13
Fuel Density Credits A-13
Other Direct Cost Data A-13
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Assessment Results A-14
Compliance Expenditures and Costs A-14
Annualization Method A-16
Discounting Costs and Expenditures A-19
Indirect Economic Effects of the CAA A-20
GNP and Personal Consumption A-20
Prices A-23
Sectoral Effects: Changes in Prices and Output by Industry A-23
Changes in Employment Across Industries A-25
Uncertainties in the Cost Analysis A-26
Potential Sources of Error in the Cost Data A-26
Mobile Source Costs A-28
Stationary Source Cost Estimate Revisions A-29
Endogenous Productivity Growth in the Macro Model A-29
Amortization Period for Stationary Source Plant and Equipment.. A-30
Cost and Macroeconomic Modeling References A-31
Appendix B: Emissions Modeling B-l
Introduction B-l
Comparison of Emissions Projections with Other EPA Data B-l
Control Scenario Projections Versus EPA Trends Projections B-l
No-Control Scenario Projections Versus Historical EPA Trends Data B-3
Industrial Boilers and Processes , B-4
Overview of Approach B-4
Industrial Boilers B-4
Industrial Processes and In-Process Fuel Combustion B-4
Establishment of Control Scenario Emissions B-5
Control Scenario Boiler Emissions B-5
Control Scenario Industrial Process Emissions B-7
Development of Economic Driver Data
for the Control Scenario - Industrial Boilers and Processes B-7
Economic Driver Data for Industrial Boiler Approach B-7
Economic Driver Data for the Industrial Process Approach B-8
No-control Scenario Emissions B-8
Industrial Boiler Emissions of SO2, NOx, and TSP B-8
Industrial Boiler Emissions of CO and VOC , B-9
Industrial Process Emissions B-9
Lead Emissions B-9
Off-Highway Vehicles ; B-10
Overview of Approach B-10
Development of Control Scenario B-ll
No-control Scenario Emissions Estimates B-ll
National and State-Level Off-Highway Emission Estimates B-ll
On-Highway B-l-2
Overview of Approach B-l 3
Personal Travel B-13
Iterative Proportional Fitting (DPF) B-13
Vehicle Ownership Projection (VOP) B-14
Projection of Vehicle Fleet Composition B-14
Activity/Energy Computation B-14
Goods Movement B-l 5
Other Transportation Activities... B-15
Lead Emissions B-15
Estimation of No-control Scenario Emissions B-15
Development of Emission Factors B-15
Allocation of Highway Activity to States B-16
IV
-------
Contents
Development of Highway Pollutant Estimates B-16
Control Scenario Emissions Calculation B-16
No-control Scenario Emissions B-21
Utilities B-24
Overview of Approach B-24
Establishment of Control Scenario Emissions B-24
Key Assumptions in the Development of the ICF Analysis B-24
ARGUS Modeling Assumptions B-26
No-control Scenario Emissions B-27
ICF Estimates of SO2, TSP, and NOx Emissions in the No-control Scenario B-27
ARGUS No-control Scenario B-29
Estimation of Lead Emissions from Utilities B-29
CEUM Sensitivity Case B-30
Commercial/Residential B-30
Control Scenario Emissions B-31
Emissions Data B-32
Energy Data B-33
Economic/Demographic Data B-33
No-control Scenario Emissions B-34
Emissions Data B-34
Energy Data B-34
Economic/Demographic Data B-35
Emissions Modeling References B-39
Appendix C: Air Quality Modeling C-l
Introduction C-l
Carbon Monoxide C-l
Control scenario carbon monoxide profiles C-l
No-control scenario carbon monoxide profiles C-2
Summary differences in carbon monoxide air quality C-4
Key caveats and uncertainties for carbon monoxide C-4
Sulfur Dioxide C-5
Control scenario sulfur dioxide profiles C-5
No-control scenario sulfur dioxide profiles C-5
Summary differences in sulfur dioxide air quality C-6
Key caveats and uncertainties for sulfur dioxide C-6
Nitrogen Oxides C-6
Control scenario nitrogen oxides profiles C-7
No-control scenario nitrogen oxides profiles C-8
Summary differences in nitrogen oxides air quality C-8
Key caveats and uncertainties for nitrogen oxides C-8
Acid Deposition C-8
Control scenario acid deposition profiles C-9
No-control scenario acid deposition profiles C-l 1
Summary differences in acid deposition C-12
Key caveats and uncertainties for acid deposition C-12
Paniculate Matter C-13
Control scenario paniculate matter profiles C-14
No-control scenario paniculate matter profiles C-15
Summary differences in paniculate matter air quality < C-16
Key caveats and uncertainties for paniculate matter C-16
Ozone C-18
Control scenario ozone profiles C-21
No-control scenario ozone profiles C-21
Summary differences in ozone air quality C-23
Key caveats and uncertainties for ozone C-24
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Visibility , C-25
Control scenario visibility. C-25
No-control scenario visibility C-26
Summary differences in visibility C-26
DeciView Haze Index C-26
Modeling Results C-28
Key caveats and uncertainties for visibility C-28
Air Quality Modeling References C-30
Appendix D: Human Health and Welfare Effects of Criteria Pollutants D-l
Introduction D-l
Principles for the Section 812 Benefits Analysis D-l
General Modeling Approach D-2
Quantifying Changes in Pollutant Exposures D-2
Air Quality D-2
Population Distribution D-3
Census Data D-3
Gridding U.S. Population D-4
Allocating Exposure Estimates to the Population D-4
Method One D-4
Method Two D-4
Estimating Human Health Effects of Exposure D-5
Types of Health Studies D-5
Epidemiological Studies D-6
Human Clinical Studies D-7
Issues in Selecting Studies To Estimate Health Effects D-9
Peer-Review of Research D-9
Confounding Factors D-9
Uncertainty D-10
Magnitude of Exposure D-ll
Duration of Exposure D-ll
Thresholds D-ll
Target Population D-ll
Statistical Significance of Exposure-Response Relationships D-12
Relative Risks D-12
Baseline Incidence Data D-12
Estimating Mortality Effects D-13
Using PM as an Indicator D-13
Estimating the Relationship Between PM and Premature Mortality D-13
Prematurity of Mortality: Life-Years Lost as a Unit of Measure D-16
Estimating Morbidity Effects D-19
Overlapping Health Effects D-19
Studies Requiring Adjustments D-19
Concentration-Response Functions: Health Effects D-19
Paniculate Matter D-19
Ozone D-26
Nitrogen Oxides D-34
Carbon Monoxide D-36
Sulfur Dioxide D-38
Estimating Welfare Effects of Exposure D-40
Agricultural Effects '. D-40
Materials Damage D-41
Visibility D-41
Worker Productivity D-41
Ecological Effects D-41
Modeling Results , D-44
Human Health and Welfare Effects References D-48
VI
-------
Contents
Appendix E: Ecological Effects of Criteria Pollutants E-l
Introduction E-l
Benefits From Avoidance of Damages to Aquatic Ecosystems, E-l
Acid Deposition E-2
Background E-2
Current Impacts of Acid Deposition E-2
Effects on Water Chemistry E-2
Effects on Fish Habitat Quality E-4
Economic Damages to Recreational Fishing E-5
Benefits From Acid Deposition Avoidance Under the CAA... E-5
Recreational Fishing E-5
Eutrophication E-6
Atmospheric Deposition and Eutrophication E-7
Valuing Potential Benefits from Eutrophication Avoidance Under the CAA E-7
Mercury E-8
Benefits from Avoided Damages to Wetland Ecosystems E-9
Introduction E-9
Effects of Acidification > E-9
Effects of Nutrient Loading E-10
Summary of Wetland Ecosystem Effects E-ll
Benefits from Avoided Damages to Forests E-l 1
Introduction E-ll
Current Air Pollutant Effects on Forests E-12
Acid Deposition Impacts E-12
Ozone Impacts E-12
Experimental Evidence , E-12
Observational Evidence '. E-l3
Endangered species E-14
Valuation of Benefits From CAA-Avoided Damages to Forests E-14
Background E-14
Commercial Timber Harvesting E-15
Non-marketed Forest Services E-16
Ecosystem Effects References '.. E-l 8
Appendix F: Effects of Criteria Pollutants on Agriculture F-l
Introduction F-l
Ozone Concentration Data F-l
Control and No-control Scenario Ozone Concentration Data F-2
Calculation of the W126 Statistic F-2
Aggregating Ozone Data to the County Level F-3
Yield Change Estimates F-3
Exposure-Response Functions F-3
Minimum/Maximum Exposure-Response Functions F-4
Calculation of Ozone Indices F-4
Calculations of County Weights F-5
Calculation of Percent Change in Yield F-5
Economic Impact Estimates F-5
Agricultural Simulation Model (AGSIM) 1 F-5
Conclusions F-9
Agricultural Effects References F-10
Appendix G: Lead Benefits Analysis G-l
Introduction G-l
Methods Used to Measure and Value Health Effects '. G-2
Health Benefits to Children G-2
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Changes in IQ G-2
Quantifying the Relationship Between Blood Lead Levels and IQ G-2
Valuing Changes in Children's Intelligence G-3
Children with IQs Less Than 70 G-7
Quantifying the Number of Children with IQs Less than 70 G-7
Valuing the Reduction in Number of Children with IQs less than 70 G-8
Changes in Neonatal Mortality G-8
Quantifying the relationship between PbB levels and neonatal mortality G-8
Valuing changes in neonatal mortality G-8
Health Benefits to Men G-8
Hypertension G-9
Quantifying the relationship between PbB levels and hypertension G-9
Valuing reductions in hypertension G-9
Quantifying the relationship between blood lead and blood pressure G-9
Changes In Coronary Heart Disease G-10
Quantifying the relationship between blood pressure and coronary heart disease G-10
Valuing reductions in CHD events G-l 1
Changes in Initial Cerebrovascular Accidents
and Initial Atherothrombotic Brain Infarctions G-12
Quantifying the relationship between blood pressure and first-time stroke G-12
Valuing reductions in strokes G-12
Changes in Premature Mortality G-13
Quantifying the relationship between blood pressure and premature mortality G-13
Valuing reductions in premature mortality , G-13
Health Benefits to Women G-13
Changes in Coronary Heart Disease G-14
Quantifying the relationship between blood pressure and coronary heart disease G-14
Valuing reductions in CHD events G-14
Changes in Atherothrombotic Brain Infarctions and Initial Cerebrovascular Accidents ...G-14
Quantifying the relationship between blood pressure and first-time stroke G-14
Valuing reductions in strokes G-15
Changes in Premature Mortality G-15
Quantifying the relationship between blood pressure and premature mortality G-15
Quantifying Uncertainty G-15
Characterizing Uncertainty Surrounding the Dose-Response Relationships ; G-15
Characterizing Uncertainty Surrounding the Valuation Estimates G-15
Industrial Processes and Boilers and Electric Utilities G-16
Methods Used to Determine Changes in Lead Emissions
from Industrial Processes from 1970 to 1990 G-16
TRIData G-16
Derivation of Industrial Process Emissions Differentials 1970-1990 G-17
Data sources G-17
Estimates of industrial process emissions in the control scenario G-17
Estimates of industrial process emissions in the no-control scenario G-l8
Matching TRI Data to Industrial Process Emissions Differentials G-l8
Methods Used to Determine Changes in Lead Emissions
from Industrial Boilers from 1970 to 1990 G-19
TRIData G-19
Derivation of Industrial Combustion Emissions 1970-1990 G-20
Estimates of combustion emissions under the control scenario G-20
Estimates of combustion emissions under the no-control scenario G-20
Matching TRI Data to Industrial Combustion Emissions Data G-21
Methods Used to Determine Changes in Lead Emissions
from Electric Utilities from 1975 to 1990 G-21
Coal-Use Data G-21
The EPA Interim Emissions Inventory G-21
Matching the Coal-Use Data to the Interim Emissions Inventory G-22
Emissions Factors and Control Efficiencies G-22
Vlll
-------
Use of Air Dispersion Modeling to Estimate Ambient Air Lead Levels G-23
Determination of Blood Lead Levels from Air Lead Concentrations G-23
Relationship Between Air Lead Concentrations and Blood Lead Levels G-23
Children , G-25
Adults G-25
Individuals with initial blood lead levels of 30 ug/dL and greater G-26
Estimates of Initial Blood Lead Concentrations G-26
Combination of Air Concentration Estimates with Population Data G-27
Results • • G~28
Reduction in Health Effects Attributable to Gasoline Lead Reductions G-31
Estimating Changes in Amount of Lead in Gasoline from 1970 to 1990 G-31
Estimating the Change in Blood Lead Levels
from the Change in the Amount of Lead in Gasoline G-31
1970-Forward and 1990-Backward Approaches G-32
Relating Blood Lead Levels to Population Health Effects G-32
Changes in Leaded Gasoline Emissions
and Resulting Decreased Blood Lead Levels and Health Effects G-32
Lead Benefits Analysis References G-36
Appendix H: Air Toxics • H-l
Introduction H-l
Limited Scope of this Assessment -H-l
History of Air Toxics Standards under the Clean Air Act of 1970 H-2
Quantifiable Stationary Source Air Toxics Benefits H-3
EPA Analyses of Cancer Risks from Selected Air Toxic Pollutants H-3
Cancer Risk Estimates from NESHAP Risk Assessments H-4
Non-utility Stationary Source Cancer Incidence Reductions H-4
PES Study H-5
Methodology • H-5
Findings ..;. •„.= H-6
ICF Re-analysis H-7
Methodology H-7
Findings • • H-8
Mobile Source HAP Exposure Reductions H-9
Methodology H-10
Results H-10
Non-Cancer Health Effects , :.H-11
Ecological Effects H-ll
Conclusions — Research Needs H-12
Health Effects H-12
Exposure Assessment : — H-l3
Ecosystem Effects H-13
Economic Valuation H-13
Air Toxics References H-14
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants 1-1
Methods Used to Value Health and Welfare Effects 1-1
Valuation of Specific Health Endpoints 1-3
Valuation of Premature Mortality Avoided 1-3
Valuation of Hospital Admissions Avoided 1-3
Valuation of Chronic Bronchitis Avoided 1-4
Valuation of Other Morbidity Endpoints Avoided 1-6
Valuation of Welfare Effects 1-6
Visibility Valuation 1-6
Results of Valuation of Health and Welfare Effects 1-16
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Uncertainties 1-16
The Effect of Discount Rates 1-20
The Relative Importance of Different Components of Uncertainty 1-20
Economic Benefits Associated with Reducing Premature Mortality 1-21
Economic Valuation References 1-27
Appendix J: Future Directions J-l
Research Implications J-l
Research Topics to Reduce Uncertainty J-l
Research Topics to Improve Comprehensiveness J-3
Future Section 812 Analyses J-4
-------
Tables
Table ES-1 Criteria Pollutant Health Benefits - Distributions of 1990 Incidences of Avoided
Health Effects (In Thousands of Incidences Reduced) for 48 State Population ES-4
Table ES-2 Major Nonmonetized, Adverse Effects Reduced by the Clean Air Act ES-5
Table ES-3 Central Estimates of Economic Value per Unit of Avoided Effect
(In 1990 Dollars) ES-6
Table ES-4 Total Monetized Benefits by Endpoint Category for 48 State Population for 1970
to 1990 Period (In Billions of 1990 Dollars) ES-7
Table ES-5 Alternative Mortality Benefits Mean Estimates for 1970 to 1990 (In Trillions
of 1990 Dollars) Compared to Total 1970 to 1990 Compliance Costs ES-9
Table 1 Estimated Annual CAA Compliance Costs ($Billions) 8
Table 2 Compliance Cost, GNP, and Consumption Impacts Discounted to 1990
($1990 Billions) 11
Table 3 Summary of Sector-Specific Emission Modeling Approaches 14
Table 4 Uncertainties Associated with Emissions Modeling 18
Table 5 Key Uncertainties Associated with Air Quality Modeling 26
Table 6 Human Health Effects of Criteria Pollutants. 31
Table 7 Selected Welfare Effects of Criteria Pollutants 32
Table 8 Percent of Population (of the Continental US) Within 50km of a Monitor (Or in a
County with PM monitors), 1970-1990 33
Table 9 Criteria Pollutants Health Benefits — Distributions of 1990 Avoided Premature
Mortalities (Thousands of Cases Reduced) for 48 State Population 37
Table 10 Criteria Pollutants Health Benefits — Distributions of 1990 Non-Fatal Avoided
Incidence (Thousands of Cases Reduced) for 48 State Population 38
Table 11. Health and Welfare Effects of Hazardous Air Pollutants 40
Table 12 Uncertainties Associated with Physical Effects Modeling 42
Table 13 Health and Welfare Effects Unit Valuation (1990 Dollars) 44
Table 14 Summary of Mortality Valuation Estimates (Millions of $1990) 45
Table 15 Estimating Mortality Risk Based on Wage-Risk Studies: Potential Sources
and Likely Direction of Bias ; 50
Table 16 Present Value of 1970 to 1990 Monetized Benefits by Endpoint Category for 48
State Population (Billions of $1990, Discounted to 1990 at 5 Percent) 52
Table 17 Total Monetized Benefits for 48 State Population (Present Value in Billions of
1990 Dollars, Discounted to 1990 at 5 Percent) 53
Table 18 Quantified Uncertainty Ranges for Monetized Annual Benefits and Benefit/Cost
Ratios, 1970-1990 (In Billions of 1990-Value Dollars) 55
Table 19 Alternative Mortality Benefits Mean Estimates for 1970 to 1990 (in Trillions of
1990 Dollars, Discounted at 5 percent) Compared to Total 1970 to 1990
Compliance Costs 57
Table 20 Effect of Alternative Discount Rates on Present Value of Total Monetized
Benefits/Costs for 1970 to 1990 (In Trillions of 1990 Dollars) 57
Table A-l Key Distinguishing Characteristics of the Jorgenson-Wilcoxen Model. A-3
Table A-2 Definitions of Industries Within the J/W Model A-4
Table A-3 Estimated Capital and O&M Expenditures for Stationary Source Air Pollution
Control (Millions of Current Dollars) A-10
Table A-4 Estimated Recovered Costs for Stationary Source Air Pollution Control
(Millions of Current Dollars) A-ll
Table A-5 Estimated Capital and Operation and Maintenance Expenditures for Mobile
Source Air Pollution Control (Millions of Current Dollars) A-12
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table A-6 O&M Costs and Credits (Millions of Current Dollars) A-12
Table A-7 Other Air Pollution Control Expenditures (Millions of Current Dollars) A-14
Table A-8 Summary of Expenditures and Conversion to 1990 Dollars
(Millions of Dollars) A-15
Table A-9 Annualized Costs, 1973-1990 (Millions of 1990 Dollars; Capital Expenditures
Annualized at 5 Percent) A-16
Table A-10 Amortization of Capital Expenditures for Stationary Sources
(Millions of 1990 Dollars) A-17
Table A-l 1 Amortization of Capital Expenditures for Mobile Sources
(Millions of 1990 Dollars) A-18
Table A-12 Compliance Expenditures and Annualized Costs, 1973 to 1990 ($1990 millions) A-19
Table A-13 Costs Discounted to 1990 ($1990 Millions) A-20
Table A-14 Differences in Gross National Product Between the Control and No-Control
Scenarios A-20
Table A-15 Difference in Personal Consumption Between the Control and No-Control
Scenarios A-21
Table A-16 GNP and Consumption Impacts Discounted to 1990 ($1990 Billions) A-21
Table A-17 Percentage Difference in Energy Prices Between the Control and No-Control
Scenarios A-23
Table A-18 Potential Sources of Error and Their Effect on Total Costs of Compliance A-26
Table A-19 Stationary Source O&M Expenditures as a Percentage of Capital Stock
(Millions of 1990 Dollars) A-27
Table A-20 Comparison of EPA and BEA Stationary Source Expenditure Estimates
(Millions of Current Dollars) A-28
Table A-21 BEA Estimates of Mobile Source Costs A-29
Table A-22 Annualized Costs Assuming 40-Year Stationary Source Capital Amortization
Period, 1973 to 1990 ($1990 Millions) A-30
Table A-23 Effect of Amortization Periods on Annualized Costs Discounted to 1990
(Billions of $1990) A-30
Table B-l Correspondence Between Process Emissions Categories Used by MSCET,
Trends, and J/W Industrial Sectors and Identifier Codes B-6
Table B-2 Fuel Use Changes Between Control and No-control Scenarios B-9
Table B-3 Difference in Control and No-control Scenario Off-Highway Mobile Source
Emissions B-12
Table B-4 Sources of Data for Transportation Sector Control Scenario Activity Projection B-17
Table B-5 Distribution of Households by Demographic Attributes for Control Scenario B-l 8
Table B-6 Economic and Vehicle Usage Data for Vehicle Ownership Projection Control
Scenario B-19
Table B-7 Control Scenario Personal Characteristics B-20
Table B-8 Distribution of Households by Income Class for No-Control Scenario B-21
Table B-9 Economic and Vehicle Usage Data for Vehicle Ownership Projection
No-Control Scenario B-22
Table B-10 Percent Changes in Key Vehicle Characteristics Between the Control and
No-Control Scenarios B-23
Table B-l 1 J/W Estimates of Percentage Increases in National Electricity Generation
Under No-Control Scenario B-29
Table B-12 Trends Source Categories and (1975 to 1985) Scaling Factors for TSP and CO. B-33
Table B-l3 Percentage Change in Real Energy Demand by Households from Control to
No-Control Scenario B-34
Table B-14 Percentage Change in Commercial Energy Demand from Control to
No-Control Scenario B-35
Table B-15 J/W Percent Differential in Economic Variables Used in CRESS B-35
Table B-l6 TSP Emissions Under the Control and No-Control Scenarios by Target Year
(In Thousands of Short Tons) B-36
Table B-17 SO2 Emissions Under the Control and No-Control Scenarios by Target Year
(In Thousands of Short Tons) B-36
Xll
-------
Tables
Table B-18 NOx Emissions Under the Control and No-Control Scenarios by Target Year
(In Thousands of Short Tons) B-37
Table B-19 VOC Emissions Under the Control and No-Control Scenarios by Target Year
(In Thousands of Short Tons) B-37
Table B-20 CO Emissions Under the Control and No-Control Scenarios by Target Year
(In Thousands of Short Tons) B-38
Table B-21 Lead (Pb) Emissions Under the Control and No-Control Scenarios by Target
Year (In Thousands of Short Tons) B-38
Table C-l Summary of CO Monitoring Data C-2
Table C-2 Format of Air Quality Profile Databases C-3
Table C-3 Summary of SO2 Monitoring Data C-5
Table C-4 Summary of NO2 Monitoring Data C-7
Table C-5 Summary of NO Monitoring Data C-7
Table C-6 Summary of TSP Monitoring Data C-14
Table C-7 Summary of PM10 Monitoring Data C-15
Table C-8 Fine Particle (PM2.5) Chemical Composition by U.S. Region C-l6
Table C-9 Coarse Particle (PM2.5 to PM10) Chemical Composition by U.S. Region C-17
Table C-10 PM Control Scenario Air, Quality Profile Filenames C-17
Table C-ll PM No-Control Scenario Air Quality Profile Filenames C-l8
Table C-12 Urban Areas Modeled with OZEPM4 C-19
Table C-13 Summary of Ozone Monitoring Data C-21
Table C-14 Apportionment of Emissions Inventories for SAQM Runs C-22
Table C-15 1990 Control Scenario Visibility Conditions for 30 Southwestern U.S. Cities C-27
Table C-16 1990 No-control Scenario Visibility Conditions for 30 Southwestern U.S. Cities C-27
Table C-17 Summary of Relative Change in Visual Range and DeciView Between 1990
Control and No-Control Scenario Visibility Conditions for 30 Southwestern
U.S. Cities ..C-29
Table D-l Criteria Air Pollutant Monitors in the U.S., 1970- 1990 D-3
Table D-2 Population Coverage in the "Within 50 km" Model Runs
(Percent of Continental U.S. Population) D-4
Table D-3 Population Coverage for "Extrapolated to All U.S." Model Runs (Percent of
Continental U.S. Population) D-5
Table D-4 Human Health Effects of Criteria Pollutants D-6
Table D-5 PM2.5/PM10 Ratios Used to Estimate PM2.5 Data Used With Pope et al. (1995)
Mortality Relationship D-16
Table D-6 Summary of Concentration-Response Functions for Particulate Matter D-20
Table D-7 Summary of Concentration-Response Functions for Ozone D-27
Table D-8 Summary of Concentration-Response Functions for NO2 D-35
Table D-9 Summary of Concentration-Response Functions for Carbon Monoxide D-37
Table D-10 Summary of Concentration-Response Functions for Sulfur Dioxide D-39
Table D-l 1 Selected Welfare Effects of Criteria Pollutants D-40
Table D-12 Summary of Functions Quantifying Welfare Benefits D-42
Table D-13 Criteria Pollutants Health Effects — Extrapolated to 48 State U.S. Population
(Cases Per Year-Mean Estimates) D-45
Table D-l4 Mortality Distribution by Age: Proportion of PM- and Pb-related Premature
Mortalities and Associated Life Expectancies D-46
Table D-l5 Quantified Benefits Which Could Not Be Monetized — Extrapolated to the
Entire 48 State Population D-47
Table E-l Summary of Biological Changes with Surface Water Acidification E-3
Table E-2 Comparison of Population of Acidic National Surface Water Survey (NSWS) by
Chemical Category E-4
Table E-3 Results from Benefits Assessments of Aquatic Ecosystem Use Values from Acid
Deposition Avoidance E-6
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table F-l Agriculture Exposure-Response Functions F-4
Table F-2 Relative No-Control to Control Percent Yield Change (harvested acres) for the
Minimum Scenario • F-6
Table F-3 Relative No-Control to Control Percent Yield Change (harvested acres) for the
Maximum Scenario • F-6
Table F-4 Change in Farm Program Payments, Net Crop Income, Consumer Surplus,
and Net Surplus Due to the CAA (Millions 1990 $) F-8
Table G-l Quantified and Unqualified Health Effects of Lead G-l
Table G-2 Uncertainty Analysis: Distributions Associated With Dose-Response
Coefficients Used to Estimate Lead Health Effects G-16
Table G-3 Air Modeling Parameters. G-24
Table G-4 Estimated Indirect Intake Slopes: Increment of Blood Lead Concentration
(in ug/dL) per Unit of Air Lead Concentration (ug/m3) G-26
Table G-5 Estimated Lead Emissions from Electric Utilities, Industrial Processes,
and Industrial Combustion (in Tons) G-28
Table G-6 Yearly Differences in Number of Health Effects Between the Control and
No-Control Scenarios: Industrial Processes, Boilers, and Electric Utilities
(Holding Other Lead Sources at Constant 1970 Levels) G-29
Table G-7 Yearly Differences in Number of Health Effects Between the Controlled and
Uncontrolled Scenarios: Industrial Processes, Boilers, and Electric Utilities
(Holding Other Lead Sources at Constant 1990 Levels) G-30
Table G-8 Lead Burned in Gasoline (In Tons) , G-33
Table G-9 Yearly Differences in Number of Health Effects Between the Control and
No-Control Scenarios: Lead in Gasoline only (Holding Other Lead Sources at
Constant 1970 Levels) G-34
Table G-10 Yearly Differences in Number of Health Effects Between the Control and
No-Control Scenarios: Lead in Gasoline only (Holding Other Lead Sources at
Constant 1990 Levels) G-35
Table H-l Health and Welfare Effects of Hazardous Air Pollutants H-2
Table H-2 Cancer Incidence Reductions and Monetized Benefits for NESHAPs H-5
Table 1-1 Summary of Mortality Valuation Estimates (Millions of 1990 Dollars) 1-3
Table 1-2 Unit Values Used for Economically Valuing Health and Welfare Endpoints 1-8
Table 1-3 Criteria Pollutants Health and Welfare Benefits — Extrapolated to Entire 48 State
Population Present Value (In 1990 Using 5% Discount Rate) of Benefits from
1970-1990 (In Billions of 1990 Dollars) 1-17
Table 1-4 Present Value of 1970 to 1990 Monetized Benefits by Endpoint Category for
48 State Population (Billions of 1990 Dollars, Discounted to 1990 at 5 Percent) 1-18
Table 1-5 Monte Carlo Simulation Model Results for Target Years, Plus Present Value in
1990 Terms of Total Monetized Benefits for Entire 1970 to 1990 Period
(In Billions of 1990-Value Dollars) 1-18
Table 1-6 Comparison of 1990 (Single Year) Monetized Benefits by Endpoint for 48 State
Population and Monitored Areas (In Millions of 1990 Dollars) 1-19
Table 1-7 Effect of Alternative Discount Rates on Present Value of Total Monetized
Benefits for 1970 to 1990 (In Trillions of 1990 Dollars) 1-20
Table 1-8 Alternative Estimates of the Present Value of Mortality Associated With PM
(Based on Pope et al., 1996, in Trillions of 1990 Dollars) 1-25
xiv
-------
Figures
Figure ES-1 Total Direct Compliance Costs of the CAA (in billions of inflation-adjusted
dollars.) ES-1
Figure ES-2 1990 Control and No-control Scenario Emissions (in millions of short tons) ES-2
Figure ES-3 Total Direct Costs and Monetized Direct Benefits of the Clean Air Act, 1970
to 1990 (in trillions of 1990 dollars) '. ES-8
Figure 1 Summary of Analytical Sequence and Modeled versus Historical Data Basis. 4
Figure 2 Control and No-control Scenario Total SO2 Emission Estimates 16
Figure 3 Control and No-control Scenario Total NOx Emission Estimates 16
Figure 4 Control and No-control Scenario Total VOC Emission Estimates 16
Figure 5 Control and No-control Scenario Total CO Emission Estimates ..16
Figure 6 Control and No-control Scenario Total TSP Emission Estimates 16
Figure 7 Control and No-control Scenario Total Pb Emission Estimates 16
Figure 8 Frequency Distribution of Estimated Ratios for 1990 Control to No-control
Scenario 95th Percentile 1-Hour Average CO Concentrations, by Monitor 21
Figure 9 Frequency Distribution of Estimated Ratios for 1990 Control to No-control
Scenario 95th Percentile 1-Hour Average SO2 Concentrations, by Monitor 22
Figure 10 Frequency Distribution of Estimated Ratios for 1990 Control to No-control
Scenario 95th Percentile 1 -Hour Average NO2 Concentrations, by Monitor 23
Figure 11 Distribution of Estimated Ratios for 1990 Control to No-Control Annual Mean
TSP Concentrations, by Monitored County 23
Figure 12 Distribution of Estimated Ratios for 1990 Control to No-control OZIPM4
Simulated 1-Hour Peak Ozone Concentrations, by Urban Area 23
Figure 13 Distribution of Estimated Ratios for 1990 Control to No-control SAQM Simulated
Daytime Average Ozone Concentrations, by SAQM Monitor 24
Figure 14 Distribution of Estimated Ratios for 1990 Control to No-control RADM Simulated
Daytime Average Ozone Concentrations, by RADM Grid Cell. 24
Figure 15 RADM-Predicted Percent Increase in Total Sulfur Deposition (Wet + Dry) Under the
No-control Scenario 24
Figure 16 RADM-Predicted Percent Increase in Total Nitrogen Deposition (Wet + Dry) Under
the No-control Scenario 25
Figure 17 RADM-Predicted Increase in Visibility Degradation, Expressed in DeciViews, for
Poor Visibility Conditions (90th Percentile) Under the No-control Scenario 25
Figure 18 Monte Carlo Simulation Model Results for Target Years (in billions of 1990 dollars) 54
Figure 19 Distribution of 1990 Monetized Benefits of CAA (in billions of 1990 dollars) 54
Figure 20 Uncertainty Ranges Deriving From Individual Uncertainty Factors 55
Figure A-l Percent Difference in Real Investment Between Control and No-control Scenarios A-22
Figure A-2 Percent Difference in Price of Output by Sector Between Control and No-control
Scenario for 1990 A-22
Figure A-3 Percent Difference in Quantity of Output by Sector Between Control and
No-control Scenario for 1990 A-24
Figure A-4 Percent Difference in Employment by Sector Between Control and No-control
Scenario for 1990 A-24
Figure B-l Comparison of Control, No-control, and Trends SO2 Emission Estimates B-2
Figure B-2 Comparison of Control, No-control, and Trends NOx Emission Estimates B-2
Figure B-3 Comparison of Control, No-control, and Trends VOC Emission Estimates B-2
Figure B-4 Comparison of Control, No-control, and Trends CO Emission Estimates B-2
Figure B-5 Comparison of Control, No-Control, and Trends TSP Emission Estimates B-2
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Figure C-l Frequency Distribution of Estimated Ratios for 1990 Control to No-Control
Scenario 95th Percentile 1-Hour Average CO Concentrations, by Monitor C-4
Figure C-2 Frequency Distribution of Estimated Ratios for 1990 Control to No-control
Scenario 95th Percentile 1-Hour Average SO2 Concentrations, by Monitor C-6
Figure C-3 Frequency Distribution of Estimated Ratios for 1990 Control to No-control
Scenario 95th Percentile 1-Hour Average NO2 Concentrations, by Monitor C-8
Figure C-4 Location of the High Resolution RADM 20-km Grid Nested Inside the 80-km
RADM Domain C-9
Figure C-5 RADM-Predicted 1990 Total Sulfur Deposition (Wet + Dry; in kg/ha) Under the
Control Scenario C-IQ
Figure C-6 RADM-Predicted 1990 Total Nitrogen Deposition (Wet + Dry; in kg/ha) Under the
Control Scenario • C-10
Figure C-7 RADM-Predicted 1990 Total Sulfur Deposition (Wet + Dry; in kg/ha) Under the
No-control Scenario C-ll
Figure C-8 RADM-Predicted 1990 Total Nitrogen Deposition (Wet + Dry; in kg/ha) Under the
No-control Scenario C-ll
Figure C-9 RADM-Predicted Percent Increase in Total Sulfur Deposition (Wet + Dry; in kg/ha)
Under the No-control Scenario C-12
Figure C-10 RADM-Predicted Percent Increase in Total Nitrogen Deposition (Wet + Dry;
in kg/ha) Under the No-control Scenario C-12
Figure C-ll Distribution of Estimated Ratios for 1990 Control to No-Control Annual Mean
TSP Concentrations, by Monitored County - C-18
Figure C-12 RADM and SAQM Modeling Domains, with Rural Ozone Monitor Locations C-20
Figure C-l3 Distribution of Estimated Ratios for 1990 Control to No-control OZIPM4
Simulated 1-Hour Peak Ozone Concentrations, by Urban Area C-23
Figure C-14 Distribution of Estimated Ratios for 1990 Control to No-control RADM-Simulated
Daytime Average Rural Ozone Concentrations, by RADM Grid Cell C-23
Figure C-15 Distribution of Estimated Ratios for 1990 Control to No-control SAQM-Simulated
Daytime Average Ozone Concentrations, by SAQM Monitor C-23
Figure C-16 RADM-Predicted Visibility Degradation, Expressed in Annual Average
DeciView, for Poor Visibility Conditions (90th Percentile) Under the Control
Scenario C-26
Figure C-l7 RADM-Predicted Visibility Degradation, Expressed in Annual Average
DeciView, for Poor Visibility Conditions (90th Percentile) Under the No-Control
Scenario C-26
Figure C-18 RADM-Predicted Increase in Visibility Degradation, Expressed in Annual
Average DeciView, for Poor Visibility Conditions (90th Percentile) Under the
No-Control Scenario C-28
Figure H-l PES Estimated Reductions in HAP-Related Cancer Cases H-7
Figure H-2 ICF Estimated Reductions in Total HAP-Related Cancer Cases Using Upper
Bound Asbestos Incidence and Lower Bound Non-Asbestos HAP Incidence H-8
Figure H-3 ICF Estimated Reduction in Total HAP-Related Cancer Cases Using Upper
Bound Incidence for All HAPs H-8
Figure H-4 National Annual Average Motor Vehicle HAP Exposures (ug/m3) H-11
Figure 1-1 Monte Carlo Simulation Model Results for Target Years
(in billions of 1990 dollars) 1-19
Figure 1-2 Uncertainty Ranges Deriving From Individual Uncertainty Factors 1-21
xvi
-------
Acronyms and Abbreviations
ueq/L
ug/m3
l-ig
urn
ACCACAPERS
AGSIM
AIRS
A13+
ANC
ANL
APPI
AQCR
ARGUS
ASI
ATERIS
.ATLAS
AUSM
BEA
be«
BG/ED
BI
BID
BP
BTU
c.i.
CA
CAA
CAAA90
CAPMS
CARB
CASAC
CDC
CERL
CEUM
CHD
CIPP
CO
C02
COH
COHb
COPD
Council
CPUE
microequivalents per liter
micrograms per cubic meter
micrograms
micrometers, also referred to as microns
SAB Advisory Council on Clean Air Compliance Analysis Physical
Effects Review Subcommittee
AGricultural Simulation Model
EPA Aerometric Information Retrieval System
aluminum
acid neutralizing capacity
Argonne National Laboratories
Argonne Power Plant Inventory
Air Quality Control Region
Argonne Utility Simulation Model
Acid Stress Index
Air Toxic Exposure and Risk Information System
Aggregate Timberland Assessment System
Advanced Utility Simulation Model
Bureau of Economic Analysis
total light extinction
Block Group / Enumeration District
atherothrombotic brain infarction
Background Information Document
blood pressure
British Thermal Unit
confidence interval
cerebrovascular accident
Clean Air Act
Clean Air Act Amendments of 1990
EPA's Criteria Air Pollutant Modeling System
California Air Resources Board
SAB Clean Air Scientific Advisory Committee
Centers for Disease Control (now CDCP, Centers for Disease Control
and Prevention)
EPA/ORD Corvallis Environmental Research Laboratory (old name; see
NERL)
ICF Coal and Electric Utility Model
coronary heart disease
changes in production processes
carbon monoxide
carbon dioxide
coefficient of haze
blood level of carboxyhemoglobin
chronic obstructive pulmonary disease
SAB Advisory Council on Clean Air Compliance Analysis
catch per unit effort
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
CR
CRESS
CSTM
CTG
CV
CVM
D.C.
DBF
DDE
DDT
DFEV,
dL
DOC
DOE
DOI
DRI
dV
DVSAM
EC
EDB
EDC
EFI
El
EIA
EKMA
ELI
EOL
EPA
EPRI
ESEERCO
ESP
FERC
FEV,
FGD
FHWA
FIFRA
FDP
FR
FRP
GDP
GEMS
GM
GNP
GSD
ha
HAP
HAPEM-MS
HNO3
hp
HTCM
ICARUS
concentration-response
Commercial and Residential Simulation System model
Coal Supply and Transportation Model
Control Techniques Guidelines
contingent valuation
contingent valuation method
District of Columbia
diastolic blood pressure
dichlorodiphenyldichloroethylene
dichlorodiphenyltrichloroethane
decrement of forced expiratory volume (in one second)
deciliter
Department of Commerce
Department of Energy
Department of Interior
Data Resources Incorporated
DeciView Haze Index
Disaggregate Vehicle Stock Allocation Model
extinction coefficient
ethylene dibromide
ethylene dichloride
Electronic Fuel Injection
Electronic Ignition
Energy Information Administration
Empirical Kinetics Modeling Approach
Environmental Law Institute
end-of-line
Environmental Protection Agency
Electric Power Research Institute
Empire State Electric Energy Research Corporation
electrostatic precipitator
Federal Energy Regulatory Commission
forced expiratory volume (in one second)
flue gas desulfurization
Federal Highway Administration
Federal Insecticide, Fungicide, and Rodenticide Act
Federal Information Processing System
Federal Register
Forest Response Program
gross domestic product
Graphical Exposure Modeling System
geometric mean
Gross National Product
geometric standard deviation
sulfuric acid
hectares
Hazardous Air Pollutant
Hazardous Air Pollutant Exposure Model - Mobile Source
nitric acid
horsepower
Hedonic Travel-Cost Model
Investigation of Costs and Reliability in Utility Systems
xvui
-------
Acronyms and Abbreviations
ICD-9
ICE
ffic
ffiUBK
IMS
IFF
IQ
ISCLT
J/W
kg
km
Ibs
LRI
m/s
m
m3
Mm
MMBTU
MOBILESa
mpg
MRAD
MSCET
MTD
MVATS
MVMA
Mwe
N
NA
NAAQS
NAPAP
NARSTO
NATICH
NCLAN
NBA
NERA
NERC
NERL
NESHAP
NHANES
NHANES H
NIPA
NMOCs
NO
NO2
NO3-
N0x
NPTS
NSPS
NSWS
O&M
O,
International Classification of Diseases, Ninth Version (1975 Revision)
Industrial Combustion Emissions model
Industrial Economics, Incorporated
EPA's Integrated Exposure Uptake Biokinetic model
Integrated Model Set
iterative proportional fitting
intelligence quotient
Industrial Source Complex Long Term air quality model
Jorgenson / Wilcoxen
kilograms
kilometers
pounds
lower respiratory illness
meters per second
meters
cubic meters
megameters
million BTU
EPA's mobile source emission factor model
miles per gallon
minor restricted activity day
Month and State Current Emission Trends
metric tons per day
EPA's Motor Vehicle-Related Air Toxics Study
Motor Vehicle Manufacturers Association
megawatt equivalent
nitrogen
not available
National Ambient Air Quality Standard
National Acid Precipitation Assessment Program
North American Research Strategy for Tropospheric Ozone
National Air Toxics Information Clearinghouse
National Crop Loss Assessment Network
National Energy Accounts
National Economic Research Associates
North American Electric Reliability Council
EPA/ORD National Exposure Research Laboratory (new name for
CERL)
National Emission Standard for Hazardous Air Pollutants
First National Health and Nutrition Examination Survey
Second National Health and Nutrition Examination Survey
National Income and Product Accounts
nonmethane organic compounds
nitric oxide
nitrogen dioxide
nitrate ion
nitrogen oxides
Nationwide Personal Transportation Survey
New Source Performance Standards
National Surface Water Survey
operating and maintenance
ozone
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
OAQPS
OAR
QMS
OPAR
OPPE
ORD
OZEPM4
PACE
PAN
PAPE
Pb
PbB
PCB
PES
pH
PIC
PM
10
POP
P°Pmi!d
Pernod
ppb
PPH
pphm
ppm
PPRG
PRYL
PURHAPS
PVC
r2
RAD
RADM
RADM/EM
RAMC
RfD
RIA
ROM
RRAD
RUM
s.e.
SAB
SAI
SAQM
SARA
SARMAP
SCC
SEDS
SIC
SIP
SJVAQS
SMSA
EPA/OAR Office of Air Quality Planning and Standards
EPA Office of Air and Radiation
EPA/OAR Office of Mobile Sources
EPA/OAR Office of Policy Analysis and Review
EPA Office of Policy Planning and Evaluation
EPA Office of Research and Development
Ozone Isopleth Plotting with Optional Mechanism-IV
Pollution Abatement Costs and Expenditures survey
peroxyacetyl nitrate
Pollution Abatement Plant and Equipment survey
lead
blood lead level
polychlorinated biphenyl
Pacific Environmental Services
the logarithm of the reciprocal of hydrogen ion concentration, a measure
of acidity
product of incomplete combustion
particulates less than or equal to 10 microns in aerometric diameter
particulates less than or equal to 2.5 microns in aerometric diameter
population
exposed population of exercising mild asthmatics
exposed population of exercising moderate asthmatics
parts per billion
people per household
parts per hundred million
parts per million
Pooling Project Research Group
percentage relative yield loss
PURchased Heat And Power
polyvinyl chloride
statistical correlation coefficient, squared
restricted activity day
Regional Acid Deposition Model
RADM Engineering Model
Resource Allocation and Mine Costing model
reference dose
Regulatory Impact Analysis
Regional Oxidant Model
respiratory restricted activity day
Random Utility Model
standard error
Science Advisory Board
Systems Applications International
SARMAP Air Quality Model
Superfund Amendment Reauthorization Act
SJVAQS/AUSPEX Regional Modeling Adaptation Project
Source Classification Code
State Energy Data System
Standard Industrial Classification
State Implementation Plan
San Joaquin Valley Air Quality Study
Standard Metropolitan Statistical Area
xx
-------
Acronyms and Abbreviations
S02
so42-
SOS/T
SRaw
STAR
TAMM90
TEEMS
TIUS
TRI
TSP
U.S.
UAM
URI
USDA
USEPA
VC
VMT
VOC
VOP
VR
W126
WLD
WTP
sulfur dioxide
sulfate ion
State of Science and Technology (refers to a series of NAPAP reports)
Specific Airway Resistance
Stability Array weather database
Timber Assessment Market Model (revised version)
Transportation Energy and Emissions Modeling System
Truck Inventory and Use Surveys
Toxic Release Inventory
total suspended particulate
United States
Urban Airshed Model
upper respiratory illness
United States Department of Agriculture
United States Environmental Protection Agency
vinyl chloride
vehicle miles traveled
volatile organic compounds
Vehicle Ownership Projection
visual range
index of peak weighted average of cumulative ozone concentrations
Work Loss Day
willingness to pay
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
xxu
-------
Acknowledgments
This project is managed under the direction of Robert D. Brenner, Director of the U.S. EPA Office of Air
and Radiation/Office of Policy Analysis and Review and Richard D. Morgenstern, Associate Assistant Admin-
istrator for Policy Planning and Evaluation, U.S. EPA (currently on leave as Visiting Scholar, Resources for the
Future). The principal project managers are Jim DeMocker, EPA/OAR/OPAR; Al McGartland, Director, EPA/
OPPE/OEE; and Tom Gillis, EPA/OPPE/OEE.
Many EPA staff contributed or reviewed portions of this draft document, including Joel Schwartz, Michael
Shapiro, Peter Preuss, Tracey Woodruff, Diane DeWitt, Dan Axelrad, Joel Scheraga, Anne Grambsch, Jenny
Weinberger, Allyson Siwik, Richard Scheffe, Vasu Kilaru, Amy Vasu, Kathy Kaufmann, Mary Ann Stewart,
Eric Smith, Dennis J. Kotchmar, Warren Freas, Tom Braverman, Bruce Polkowsky, David Mobley, Sharon
Nizich, David Meisenheimer, Fred Dimmick, Harvey Richmond, John Haines, John Bachmann, Ron Evans,
Tom McMullen, Bill Vatavuk, Larry Sorrels, Dave McKee, Susan Stone, Melissa McCullough, Rosalina
Rodriguez, Vickie Boothe, Tom Walton, Michele McKeever, Vicki Atwell, Kelly Rimer, Bob Fegley, Aparna
Koppikar, Les Grant, Judy Graham, Robin Dennis, Dennis Leaf, Ann Watkins, Penny Carey, Joe Somers, Pam
Brodowicz, Byron Bunger, Allen Basala, David Lee, Bill O'Neill, Susan Herrod, and Susan Stendebach. Allyson
Siwik of EPA/OAR/OAQPS and Bob Fegley of EPA/ORD/OSPRE played particularly important roles in coor-
dinating substantive and review contributions from their respective offices.
A number of contractors developed key elements of the analysis and supporting documents. These contrac-
tors include Bob Unsworth, Jim Neumann, Mike Hester, and Jon Discher of Industrial Economics, Incorporated
(lEc); Leland Deck, Ellen Post, Lisa Akeson, Brad Firlie, Susan Keane, Kathleen Cunningham, and John Voyzey
of Abt Associates; Bruce Braine, Patricia Kim, Sandeep Kohli, Anne Button, Barry Galef, Cynde Sears, and
Tony Bansal of ICF Resources; John Langstaff, Michelle Woolfolk, Shelly Eberly, Chris Emery, Till Stoekenius,
and Andy Gray of ICF/Systems Applications International (ICF/SAI); Dale Jorgenson, Peter Wilcoxen, and
Richard Goettle of Jorgenson Associates; Jim Lockhart of the Environmental Law Institute (ELI); Beverly
Goodrich, Rehan Aziz, Noel Roberts, and Lucille Bender of Computer Sciences Corporation; Margaret Sexsmith
of Analytical Sciences, Incorporated; Ken Meardon of Pacific Environmental Services (PES); David South,
Gale Boyd, Melanie Tomkins, and K. Guziel of Argonne National Laboratory (ANL); Don Garner; Rex Brown
and Jacob Ulvila of Decision Science Consortium; and Jim Wilson and Dianne P. Crocker of Pechan Associ-
ates. John Pitcher and H. Glenn Court of STRA managed the technical production of an earlier version of the
draft document. The SARMAP AQM runs were provided by Carol Bohnenkamp of EPA Region 9 and Saffet
Tanrikulu of the California Air Resources Board.
Science Advisory Board review of this report is supervised by Donald G. Barnes, Director of the SAB Staff.
SAB staff coordinating the reviews have included Jack Fowle, Jack Kooyoomjian, Sam Rondberg, Fred Talcott,
and Randall Bond. Diana Pozun provided administrative support.
The SAB Council was chaired by Richard Schmalensee of MIT throughout the development of the present
study. The Council is now chaired by Maureen Cropper of the World Bank as the Council's focus shifts to the
upcoming prospective studies. Members who have participated in the review of this draft report include Morton
Lippmann of New York University Medical Center, William Nordhaus of Yale University, Paul Portney of
Resources for the Future, Kip Viscusi of Harvard University, A. Myrick Freeman of Bowdoin College, Maureen
Cropper, Ronald Cummings of Georgia State University, Daniel Dudek of the Environmental Defense Fund,
Robert Mendelsohn of Yale University, Wayne Kachel of MELE Associates, William Cooper of Michigan
State University, Thomas Tietenberg of Colby College, Paul Lioy of the Robert Wood Johnson School of
Medicine, Roger McClellan of the Chemical Industry Institute of Toxicology, George T. Wolff of General
Motors, Richard Conway of Union Carbide Corporation, and Wallace Gates of the University of Maryland.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
The SAB Council Physical Effects Review Subcommittee was chaired by Morton Lippmann. Members
who have participated in the review include David V. Bates of the University of British Columbia, A. Myrick
Freeman of Bowdoin College, Gardner Brown, Jr. of the University of Washington, Timothy Larson of the
University of Washington, Lester Lave of Carnegie Mellon University, Joseph Meyer of the University of
Wyoming, Robert Rowe of Hagler Bailly, Incorporated, George Taylor of the University of Nevada, Bernard
Weiss of the University of Rochester Medical Center, and George Wolff of the General Motors Research Labo-
ratory.
The SAB Council Air Quality Subcommittee was chaired by George Wolff. Members who have partici-
pated in the review include Benjamin Liu of the University of Minnesota, Peter Mueller of the Electric Power
Research Institute, Warren White of Washington University, Joe Mauderly of the Lovelace Biomedical & En-
vironmental Research Institute, Philip Hopke of Clarkson University, Paulette Middleton of Science Policy
Associates, James H. Price, Jr. of the Texas Natural Resource Conservation Commission, and Harvey Jeffries of
the University of North Carolina, Chapel Hill.
This report could not have been produced without the support of key administrative support staff. The
project managers are grateful to Catrice Jefferson, Nona Smoke, Carolyn Hicks, Eunice Javis, Gloria Booker,
Thelma Butler, Wanda Farrar, Ladonya Langston, Michelle Olawuyi, and Eileen Pritchard for their timely and
tireless support on this project.
XXIV
-------
Executive Summary
Purpose of the Study
Throughout the history of the Clean Air Act, ques-
tions have been raised as to whether the health and
environmental benefits of air pollution control justify
the costs incurred by industry, taxpayers, and consum-
ers. For the most part, questions about the costs and
benefits of individual regulatory standards continue
to be addressed during the regulatory development
process through Regulatory Impact Analyses (RIAs)
and other analyses which evaluate regulatory costs,
benefits, and such issues as scope, stringency, and tim-
ing. There has never been, however, any comprehen-
sive, long-term, scientifically valid and reliable study
which answered the broader question:
"How do the overall health, welfare,
ecological, and economic benefits of Clean
Air Act programs compare to the costs of
these programs?"
To address this void, Congress added to the 1990
Clean Air Act Amendments a requirement under sec-
tion 812 that EPA conduct periodic, scientifically re-
viewed studies to assess the benefits and the costs of
the Clean Air Act. Congress further required EPA to
conduct the assessments to reflect central tendency,
or "best estimate," assumptions rather than the con-
servative assumptions sometimes deemed appropri-
ate for setting protective standards.
This report is the first in this ongoing series of
Reports to Congress. By examining the benefits and
costs of the 1970 and 1977 Amendments, this report
addresses the question of the overall value of
America's historical investment in cleaner air. The
first Prospective Study, now in progress, will evalu-
ate the benefits and costs of the 1990 Amendments.
Study Design
Estimates of the benefits and costs of the histori-
cal Clean Air Act are derived by examining the dif-
ferences in economic, human health, and environmen-
tal outcomes under two alternative scenarios: a "con-
trol scenario" and a "no-control scenario." The con-
trol scenario reflects actual historical implementation
of clean air programs and is based largely on histori-
cal data. The no-control scenario is a hypothetical sce-
nario which reflects the assumption that no air pollu-
tion controls were established beyond those in place
prior to enactment of the 1970 Amendments. Each of
the two scenarios is evaluated by a sequence of eco-
nomic, emissions, air quality, physical effect, eco-
nomic valuation, and uncertainty models to measure
the differences between the scenarios in economic,
human health, and environmental outcomes. Details
of this analytical sequence are presented in Chapter 1
and are summarized in Figure 1 of that chapter.
Study Review
EPA is required, under section 812, to consult both
a panel of outside experts and the Departments of
Labor and Commerce in designing and implementing
the study.
The expert panel was organized in 1991 as the
Advisory Council on Clean Air Act Compliance
Analysis (hereafter "Council") under the auspices of
EPA's Science Advisory Board (SAB). Organizing
the external panel under the auspices of the SAB en-
sured that the peer review of the study would be con-
ducted in a rigorous, objective, and publicly open
manner. Eminent scholars and practitioners with ex-
pertise in economics, human health sciences, envi-
ronmental sciences, and air quality modeling served
on the Council and its technical subcommittees, and
these reviewers met many times throughout the de-
sign and implementation phases of the study. During
this ongoing, in-depth review, the Council provided
valuable advice pertaining to the development and
selection of data, selection of models and assumptions,
evaluation and interpretation of the analytical find-
ings, and characterization of those findings in several
successive drafts of the Report to Congress. The
present report was vastly improved as a result of the
Council's rigorous and constructive review effort.
ES-1
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
With respect to the interagency review process,
EPA expanded the list of consulted agencies and con-
vened a series of meetings during the design and early
implementation phases from 1991 through late 1994.
In late 1994, to ensure that all interested parties and
the public received consistent information about re-
maining analytical issues and emerging results, EPA
decided to use the public SAB review process as the
primary forum for presenting and discussing issues
and results. The Interagency Review Group was there-
fore discontinued as a separate process in late 1994.
A final, brief interagency review, pursuant to Cir-
cular A-19, was organized in August 1997 by the Of-
fice of Management and Budget and conducted fol-
lowing the completion of the extensive expert panel
peer review by the SAB Council. During the course
of the final interagency discussions, it became clear
that several agencies held different views pertaining
to several key assumptions in this study as well as to
the best techniques to apply in the context of environ-
mental program benefit-cost analyses, including the
present study. The concerns include: (1) the extent to
which air quality would have deteriorated from 1970
to 1990 in the absence of the Clean Air Act, (2) the
methods used to estimate the number of premature
deaths and illnesses avoided due to the CAA, (3) the
methods used to estimate the value that individuals
place on avoiding those risks, and (4) the methods
used to value non-health related benefits. However,
due to the court deadline the resulting concerns were
not resolved during this final, brief interagency re-
view. Therefore, this report reflects the findings of
EPA and not necessarily other agencies in the Ad-
ministration. Interagency discussion of some of these
issues will continue in the context of the future pro-
spective section 812 studies and potential regulatory
actions.
Summary of Results
Direct Costs
To comply with the Clean Air Act, businesses,
consumers, and government entities all incurred higher
costs for many goods and services. The costs of pro-
viding goods and services to the economy were higher
primarily due to requirements to install, operate, and
maintain pollution abatement equipment. In addition,
costs were incurred to design and implement regula-
tions, monitor and report regulatory compliance, and
invest in research and development. Ultimately, these
higher costs of production were borne by stockhold-
ers, business owners, consumers, and taxpayers.
Figure ES-1 summarizes the historical data on
Clean Air Act compliance costs by year, adjusted both
for inflation and for the value of long-term invest-
ments in equipment. Further adjusting the direct costs
incurred each year to reflect their equivalent worth in
the year 1990, and then summing these annual results,
yields an estimate of approximately $523 billion for
the total value of 1970 to 1990 direct'expenditures
(see Appendix A for calculations).
Emissions
Emissions were substantially lower by 1990 un-
der the control scenario than under the no-control sce-
nario, as shown in Figure ES-2. Sulfur dioxide (SO2)
emissions were 40 percent lower, primarily due to
utilities installing scrubbers and/or switching to lower
sulfur fuels. Nitrogen oxides (NOX) emissions were
30 percent lower by 1990, mostly because of the in-
stallation of catalytic converters on highway vehicles.
Volatile organic compound (VOC) emissions were 45
percent lower and carbon monoxide (CO) emissions
were 50 percent lower, also primarily due to motor
vehicle controls.
For paniculate matter, it is important to recog-
nize the distinction between reductions in directly
emitted particulate matter and reductions in ambient
concentrations of particulate matter in the atmosphere.
As discussed further in the next section, changes in
particulate matter air quality depend both on changes
in emissions of primary particles (i.e., air pollution
which is already in solid particle form) and on changes
in emissions of gaseous pollutants, such as sulfur di-
oxide and nitrogen oxides, which can be converted to
particulate matter through chemical transformation in
the atmosphere. Emissions of primary particulates
Figure ES-1. Total Estimated Direct Compliance Costs of
the CAA (in billions of inflation-adjusted dollars).
30 ,
20
10
o
1975
1990
ES-2
-------
Executive Summary
Figure ES-2. 1990 Control and No-control Scenario
Emissions (in millions of short tons).
200
TSP
CO
were 75 percent lower under the control scenario by
1990 than under the no-control scenario. This sub-
stantial difference is primarily due to vigorous efforts
in the 1970s to reduce visible emissions from utility
and industrial smokestacks.
Lead (Pb) emissions for 1990 are reduced by about
99 percent from a no-control level of 237,000 tons to
about 3,000 tons under the control scenario.1 The vast
majority of the difference in lead emissions under the
two scenarios .is attributable to reductions in the use
of leaded gasoline.
These reductions were achieved during a period
in which population grew by 22.3 percent and the na-
tional economy grew by 70 percent.
Air Quality
The substantial reductions in air pollutant emis-
sions achieved by the Clean Air Act translate into sig-
nificantly improved air quality throughout the U.S.
For sulfur dioxide, nitrogen oxides, and carbon mon-
oxide, the improvements in air quality under the con-
trol scenario are assumed to be proportional to the
estimated reduction in emissions. This is because, for
these pollutants, changes in ambient concentrations
in a particular area are strongly related to changes in
emissions in that area. While the differences in con-
trol and no-control scenario air quality for each of these
pollutants vary from place to place because of local
variability in emissions reductions, by 1990 the na-
tional average improvements in air quality for these
pollutants were: 40 percent reduction in sulfur diox-
ide, 30 percent reduction in nitrogen oxides, and 50
percent reduction in carbon monoxide.
Ground-level ozone is formed by the chemical re-
action of certain airborne pollutants in the presence
of sunlight. Reductions in ground-level ozone are
therefore achieved through reductions in emissions
of its precursor pollutants, particularly volatile organic
compounds (VOCs) and nitrogen oxides (NOX).2 The
differences in ambient ozone concentrations estimated
under the control scenario vary significantly from one
location to another, primarily because of local differ-
ences in the relative proportion of VOCs and NOX,
weather conditions, and specific precursor emissions
reductions. On a national average basis, ozone con-
centrations in 1990 are about 15 percent lower under
the control scenario. For several reasons, this overall
reduction in ozone is significantly less than the 30
percent reduction in precursor NOX and 45. percent
reduction in precursor VOCs. First, significant natu-
ral (i.e., biogenic) sources of VOCs limit the level of
ozone reduction achieved by reductions in man-made
(i.e., anthropogenic) VOCs. Second, current knowl-
edge of atmospheric photochemistry suggests that
ozone reductions will tend to be proportionally smaller
than reductions in precursor emissions. Finally, the
plume model system used to estimate changes in ur-
ban ozone for this study is incapable of handling long-
range transport of ozone from upwind areas and multi-
day pollution events in a realistic manner.
There are many pollutants which contribute to
ambient concentrations of particulate matter. The rela-
tive contributions of these individual pollutant spe-
cies to ambient particulate matter concentrations vary
from one region of the country to the next, and from
urban areas to rural areas. The most important par-
ticle species, from a human health standpoint, may be
the fine particles which can be respired deep into the
lungs. While some fine particles are directly emitted
by sources, the most important fine particle species
are formed in the atmosphere through chemical con-
version of gaseous pollutants. These species are re-
ferred to as secondary particles. The three most im-
portant secondary particles are (1) sulfates, which
derive primarily from sulfur dioxide emissions; (2)
nitrates, which derive primarily from nitrogen oxides
emissions; and (3) organic aerosols, which can be di-
rectly emitted or can form from volatile organic com-
1 Results for lead are not shown in Figure ES-2 because the absolute levels of lead emissions are measured in thousands, not
millions, of tons and will not be discernible on a graph of this scale.
2 Ambient NOx concentrations are driven by anthropogenic emissions whereas ambient VOCs result from both anthropogenic
and biogenic sources (e.g., terpenes emitted by trees).
ES-3
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table ES-1. Criteria P
of Avoided Health
ff <. <,' "'&'•*
Beaef%^Jstimate|
Endpolnt
, F^Iirtantfs)
Premature Mortality -.-,-- -- -VSMA'
Premature Mortality „ \ •-
Chronic Bronchitis ":;"'
Lost IQ Points
IQlessthan70
Hypertension .„,'' -
Coronary Heart Disease " ~;''
Atherolhrombotie brain infarcttdn
Initial cerebrovascular accident „ s. „, - -tead ' - >"«
Hospital Admissions *,.,_ -s,«i,,,->-,,?A*.<>-'--,'Av ,--
All Respiratory -,-<•'•* ,-r.v,PM*O2idne -•
Chronic Obstructive Pulmonary *"v *- PM^'Qzofie';;'
Disease & Pneumonia' ^
laheraic Heart Disease ,,:
Congestive Heart I'aiiuce - ,'--«
Other RespJratory-Retated Ailments-
Shortness of breath, days -K'.^*'1
Acute Bronchitis
Symptoms
Asthma Attacks
Increase in Respiratory Illness , H02, ,,
Any Symptom $02 ,
Restricted Activi^r and WtaJt toss Days
Minor Restricted ActivitjHDays , ,
Work toss Days ,.-. j,iiiwi'AHV»r,
26"'""'" 264 "'"':'1<)^~
•, '' fj ' /'j>'fj' , s
V, 107^)00 pend5x^i«-ueu»us.
. . .„•<*•-, ""^'^^il^^^l^i^aa (pfe) criteria pdEMt'ants,wh5ehjina.v,coBttibute
^ ** 'v>sN' penffitx D for additional discussion.
pound emissions. This highlights an important and
unique feature of particulate matter as an ambient pol-
lutant: more than any other pollutant, reductions in
particulate matter are actually achieved through re-
ductions in a wide variety of air pollutants. In other
words, controlling particulate matter means control-
ling "air pollution" in a very broad sense. In the present
analysis, reductions hi sulfur dioxide, nitrogen oxides,
volatile organic compounds, and directly-emitted pri-
mary particles achieved by the Clean Air Act result in
a national average reduction in total suspended par-
ticulate matter of about 45 percent by 1990. For the
smaller particles which are of greater concern from a
health effects standpoint (i.e., PM10 and PM25), the
national average reductions were also about 45 per-
cent.
Reductions in sulfur dioxide and nitrogen oxides
also translate into reductions in formation, transport,
and deposition of secondarily formed acidic com-
pounds such as sulfate and nitric acid. These are the
principal pollutants responsible for acid precipitation,
or "acid rain." Under the control scenario, sulfur and
nitrogen deposition are significantly lower by 1990
than under the no-control scenario throughout the 31
eastern states covered by EPA's Regional Acid Depo-
sition Model (RADM). Percentage decreases in sul-
fur deposition range up to more than 40 percent in the
upper Great Lakes and Florida-Southeast Atlantic
Coast areas, primarily because the no-control scenario
projects significant increases in the use of high-sulfur
fuels by utilities in the upper Great Lakes and Gulf
ES-4
-------
Executive Summary
Coast states. Nitrogen deposition is also signifi-
cantly lower under the control scenario, with per-
centage decreases reaching levels of 25 percent or
higher along the Eastern Seaboard, primarily due
to higher projected emissions of motor vehicle ni-
trogen oxides under the no-control scenario.
Finally, decreases in ambient concentrations of
light-scattering pollutants, such as sulfates and ni-
trates, are estimated to lead to perceptible improve-
ments in visibility throughout the eastern states and
southwestern urban areas modeled for this study.
Physical Effects
The lower ambient concentrations of sulfur di-
oxide, nitrogen oxides, particulate matter, carbon
monoxide, ozone and lead under the control sce-
nario yield a substantial variety of human health,
welfare and ecological benefits. For a number of
these benefit categories, quantitative functions are
available from the scientific literature which allow
estimation of the reduction in incidence of adverse
effects. Examples of these categories include the
human mortality and morbidity effects of a num-
ber of pollutants, the neurobehavioral effects among
children caused by exposure to lead, visibility im-
pairment, and effects on yields for some agricul-
tural products.
A number of benefit categories, however, can
not be quantified and/or monetized for a variety of
reasons. In some cases, substantial scientific un-
certainties prevail regarding the existence and mag-
nitude of adverse effects (e.g., the contribution of
ozone to air pollution-related mortality). In other
cases, strong scientific evidence of an effect exists,
but data are still too limited to support quantitative
estimates of incidence reduction (e.g., changes in
lung function associated with long-term exposure
to ozone). Finally, there are effects for which there
is sufficient information to estimate incidence re-
duction, but for which there are no available eco-
nomic value measures; thus reductions in adverse
effects cannot be expressed in monetary terms. Ex-
amples of this last category include relatively small
pulmonary function decrements caused by acute
exposures to ozone and reduced time to onset of
angina pain caused by carbon monoxide exposure.
Table ES-1 provides a summary of the key dif-
ferences in quantified human health outcomes esti-
mated under the control and no-control scenarios.
Results are presented as thousands of cases avoided
in 1990 due to control of the pollutants listed in the
table and reflect reductions estimated for the entire
U.S. population living in the 48 continental states. Epi-
demiological research alone cannot prove whether a
cause-effect relationship exists between an individual
'-Table fi'Sp2.' Major Nontaometized, Adverse- Effects
''''Air'Act.-;,;,; " - >••
Pollutant
Parti<;u!ate
Matter
> "' '
Ozone
Dioxide
,- • >> -,--•>,;
Oxides,,,
Nonmonetfzed-Adverse Effects
Large Qftanges?iti"Palmcmky Function
Other Ctemic ReJpi
Inflatianation -oft hfe^i
Chroflic'Respiratojy Diseases"^" „
ExSrapnlmoRary Effects £ie:y other- drgan-sxstems-5
Forest aad other Ecological Effects , ;
'Pecreased Ttoe-tp^O^aset of An^tia.
,
-Other-Cardiovascular Effects
Hospital dHJJssloas
- Materials Damage v';x"
-Ecological Effects "s'--"
to Stitaut!
De<;i;ea,s,ed Pttoionary,.?
'iMJararaiEottofth'eJL'URg
^
rophicatioB <£&,; Chesape;ake Bay) " ••> -
'^ " '"*' """
,CaM'ovaf4|j.faT Diseases ,, ":'v-,iv
R«|)toductive^Eff^ts-m,;'^;o,men ^-^
Other Nearobe^aViorafc>hys»loglcaTEffect<'siri'?<',>
'^ '&>«( '-
Ecological Effects .....
'l sIQ loss Mil dlreet,"'as op
%!
ES-5
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
In addition to controlling the six cri-
teria pollutants, the 1970 and 1977 Clean
Air Act Amendments led to reductions
in ambient concentrations of a small
number of hazardous air pollutants. Al-
though they are not fully quantified in
this report, control of these pollutants
resulted both from regulatory standards
set specifically to control hazardous air
pollutants and from incidental reductions
achieved through programs aimed at
controlling criteria pollutants.
Existing scientific research suggests
that reductions in both hazardous air
pollutants and criteria pollutants yielded
widespread improvements in the func-
tioning and quality of aquatic and ter-
Table BS-
ftvoided
pollutant and an observed health effect. Although not
universally accepted, this study uses the epidemiologi-
cal findings about correlations between pollution and
observed health effects to estimate changes in the num-
ber of health effects that would occur if pollution lev-
els change. A range is presented along with the mean
estimate for each effect, reflecting uncertainties which
have been quantified in the underlying health effects
literature.
Adverse human health effects of the Clean Air
Act "criteria pollutants" sulfur dioxide,
nitrogen oxides, ozone, particulate mat- •
ter, carbon monoxide, and lead dominate
the quantitative estimates in part be-
cause, although there are important re-
sidual uncertainties, evidence of physi-
cal consequences is greatest for these
pollutants. The Clean Air Act yielded
other benefits, however, which are im-
portant even though they are uncertain
and/or difficult to quantify. These other
benefit categories include (a) all benefits
accruing from reductions in hazardous
air pollutants (also referred to as air
toxics), (b) reductions in damage to cul-
tural resources, buildings, and other ma-
terials, (c) reductions in adverse effects
on wetland, forest, and aquatic ecosys-
tems, and (d) a variety of additional hu-
man health and welfare effects of crite-
ria pollutants. A more complete list of
these nonmonetized effects is presented
in Table ES-2.
restrial ecosystems. In addition to any intrinsic value
to be attributed to these ecological systems, human
welfare is enhanced through improvements in a vari-
ety of ecological services. For example, protection of
freshwater ecosystems achieved through reductions
hi deposition of acidic air pollutants may improve com-
mercial and recreational fishing. Other potential eco-
logical benefits of reduced acid deposition include im-
proved wildlife viewing, maintenance of biodiversity,
and nutrient cycling. Increased growth and produc-
tivity of U.S. forests may have resulted from reduc-
Estimates of Ec'bttoitjic yalue per Unit of
' *' '
tost 10 'Point's ',„;
JiQjJsmha-B 70
Led ,„-,-,-/--
?-Lead
$4i,00()
' "
per «ase
Stroke*-/? *
$200,080 "per «ase-males^
*
Hospital Admissions'
Ischemic HeartJDisea-ss, ,,
Congestive Heart Failure
,EM-
O^one
, per, Case
$10,300 -percale
3$,,30(>, per case
• $S,~!00'"oer case
PM &"Ozojie-
.Respiratory Illness aodjSyroptoros"
•-Acute 8K»ttchitfs,,"
Pljil,
"''
N'Oi,'
Upjiat fiesplratoty Symptoms
tower Respiratory' Symptoms,
Work tossWys * * ' ' [ 'v
^sid1 Restricted Activity £>a'y$.''
"Welfare Benefits - <-«.' -"
PM
PM ,,,,„„,,
PM '
PM
, , ^ *
PM & Ozaiie
- $19 per case
,,5,531,2-. per «ase
$5.'30 °per day
'*'$$&.
, $14,, ,,_
in Dec»Via,w,.,,
$2,50', pat household,
^ ,,,perJ?M-10 ""
, ;, % change
ia'Eeonotnic iSu^ilus
aSed^oii aS&igBJJig a vftlue of $29^,006 for eAob life*year lost are
?£ll'l:,,-' , ',,>'" v''''
are <}Qiapr!si;ed Qfjifyettghratabotic btaia
^^
valttations for i'trolffi'olsffis wflectdiffe'reSces ia lost Batran|s between
' iignpsfete discussion o?valuing'
' '
, , ,,
anjgein1 daily ;w8$*s:$t pat worker p'ar'fOfo'"
-^
ES-6
-------
Executive Summary
tions in ground-level ozone. More vigorous forest eco-
systems in turn yield a variety of benefits, including
increased timber production; improved forest aesthet-
ics for people enjoying outdoor activities such as hunt-
ing, fishing, and camping; and improvements in eco-
logical services such as nutrient cycling and tempo-
rary sequestration of global warming gases. These im-
provements in ecological structure and function have
not been quantified in this assessment.
Economic Valuation
Estimating the reduced incidence of physical ef-
fects provides a valuable measure of health benefits
for individual endpoints. However, to compare or ag-
gregate benefits across endpoints, the benefits must
be monetized. Assigning a monetary value to avoided
incidences of each effect permits a summation, in
terms of dollars, of monetized benefits realized as a
result of the Clean Air Act, and allows that summa-
tion to be compared to the cost of the Clean Air Act.
Before proceeding through this step, it is impor-
tant to recognize the substantial controversies and un-
certainties which pervade attempts to characterize ad-
verse human health and ecological effects of pollu-
tion in dollar terms. To many, dollar-based estimates
of the value of avoiding outcomes such as loss of hu-
man life, pain and suffering, or ecological degrada-
tion do not capture the full and true value to society as
a whole of avoiding or reducing these effects. Adher-
ents to this view tend to favor assessment procedures
which (a) adopt the most technically defensible dol-
lar-based valuation estimates for analytical purposes
but (b) leave the moral dimensions of policy evalua-
tion to those who must decide whether, and how, to
use cost-benefit results in making public policy deci-
sions. This is the paradigm adopted in the present
study. Given the Congressional mandate to perform a
cost-benefit study of the Clean Air Act, the Project
Team has endeavored to apply widely-recognized,
customary techniques of Applied Economics to per-
form this cost-benefit analysis. However, EPA be-
lieves there are social and personal values furthered
by the Clean Air Act which have not been effectively
captured by the dollar-based measures used in this
study. Therefore, EPA strongly encourages readersto
look beyond the dollar-based comparison of costs and
benefits of the Clean Air Act and consider the broader
value of the reductions in adverse health and environ-
mental effects which have been achieved as well as
any additional adverse consequences of regulation
which may not be reflected in the cost estimates re-
ported herein.
For this study, unit valuation estimates are derived
from the economics literature and reported in dollars
per case (or, in some cases, episode or symptom-day)
avoided for health effects and dollars per unit of
g-4< Tpt4JpstiWed-Moiieti|edf-BenBfits t>y EMpoi-nt Category-for-48 State Population
$90 period (i^MMans'.of &9& 'dollars}/' , ""''"'.
.\ '.,..'.' .. > J ''''.'?&, ', ' . S+i&f *••*,%* p,^*"-. ,V .. .
Endpoint „, , PoUutant(s)
Mortality , '--, , " :,;•"'•> PM ' •- •"
Mor^lily /*V',PM-, ,,„ , ,',,,-,
4o5m$t '^ *S^!f W;L^-- \~ : ,f „ - ^
tHy'fierten&iw' '-'''"-.-'- '>L«a ••'---.
' Hospit a! Act^isMotts ,;^» » /' 'PM^Ozone^ Lf^ ^ '
- Reapiratory-llel|^| ^ - „ , -^^H Ozone, M02,
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
avoided damage for human welfare effects. Similar to
estimates of physical effects provided by health stud-
ies, each of the monetary values of benefits applied in
this analysis can be expressed in terms of a mean value
and a range around the mean estimate. This range re-
flects the uncertainty in the economic valuation lit-
erature associated with a given effect. These value
ranges, and the approaches used to derive them, are
described in Chapter 6 and Appendix I for each of the
effects monetized in this study. The mean values of
these ranges are shown in Table ES-3.
Monetized Benefits and Costs
The total monetized economic benefit attributable
to the Clean Air Act is derived by applying the unit
values (or ranges of values) to the stream of
monetizable physical effects estimated for the 1970
to 1990 period. In developing these estimates, steps
are taken to avoid double-counting of benefits. In ad-
dition, a computer simulation model is used to esti-
mate ranges of plausible outcomes for the benefits
estimates reflecting uncertainties in the physical ef-
fects and economic valuation literature (see Chapter
7 and Appendix I for details).
The economic benefit estimation model then gen-
erated a range of economic values for the differences
in physical outcomes under the control and no-con-
trol scenarios for the target years of the benefits analy-
sis: 1975,1980, 1985, and 1990. Linear interpolation
between these target years is used to estimate ben-
efits in intervening years. These yearly results are then
adjusted to their equivalent value in the year 1990 and
summed to yield a range and mean estimate for the
total monetized benefits of the Clean Air Act from
Figure
theCAj
50-
5240 -
_CO
I30'
o>
°20-
.g
£10-
0 -
ES-3. Total Estimatt
\ (in trillions of infl
;d Dkec
ation-ad
— —
t Compliance Costs of
usted dollars).
-^ 95th peicentile
•^ Mean
•^ 5th percentile
Costs Benefits
1970 to 1990. These results are summarized in Table
ES-4.
Combining these benefits results with the cost es-
timates presented earlier yields the following analyti-
cal outcomes.3
• The total monetized benefits of the Clean
Air Act realized during the period from
1970 to 1990 range from 5.6 to 49.4 trillion
dollars, with a central estimate of 22.2 tril-
lion dollars.
• By comparison, the value of direct compli-
ance expenditures over the same period
equals approximately 0.5 trillion dollars.
• Subtracting costs from benefits results in
net, direct, monetized benefits ranging
from 5.1 to 48.9 trillion dollars, with a cen-
tral estimate of 21.7 trillion dollars, for the
1970 to 1990 period.
• The lower bound of this range may go down
and the upper bound may go up if analyti-
cal uncertainties associated with compli-
ance costs, macroeconomic effects, emis-
sions projections, and air quality model-
ing could be quantified and incorporated
in the uncertainty analysis. While the range
already reflects many important uncertain-
ties in the physical effects and economic
valuation steps, the range might also
broaden further if additional uncertainties
in these two steps could be quantified.
• The central estimate of 22.2 trillion dollars
in benefits may be a significant underesti-
mate due to the exclusion of large numbers
of benefits from the monetized benefit es-
timate (e.g., all air toxics effects, ecosystem
effects, numerous human health effects).
Figure ES-3 provides a graphical representation
of the estimated range of total monetized benefits and
compares this range to estimated direct compliance
costs. Clearly, even the lower bound estimate of mon-
etized benefits substantially exceeds the costs of the
historical Clean Air Act. As shown by the yearly data
presented in Chapter 7, monetized benefits consis-
tently and substantially exceeded costs throughout the
1970 to 1990 period.
ES-8
-------
Executive Summary
tables ES;5, ^tern>tiye%6ilaHl^ BeneftiEs, Mean,,, *'-,:
"dollars) feompated WTodht 970 to
-, _ '»-~V r'<"-" - s --.
Cosjs, ?-\
V9J;,^10.'f": '-«•
•.-.,....,-;'- .%..«. ,Q.S , » ;
Alternative Results
The primary results of this analysis, including ag-
gregate cost and benefit estimates which reflect many
elements of the uncertainty associated with them, are
presented above. However, some additional analysis
is required to address an important issue raised by the
EPA Science Advisory Board Council on Clean Air
Act Compliance Analysis (a.k.a. Council) charged
with reviewing the present study. Specifically, the
Council believes it is appropriate to also display al-
ternative premature mortality results based on an ap-
proach which estimates, and assigns a value to, the
loss of life-years (i.e., the reduction, in years of re-
maining life expectancy) resulting from the pollution
exposure. The Council's position is based on the con-
clusion that older individuals are more susceptible to
air pollution-induced mortality. EPA believes, how-
ever, that the simplifying assumptions which must be
adopted to implement a life-years lost approach ren-
der its results less reliable, even for the purposes of
economic efficiency analysis, than a value of statisti-
cal life approach. In addition, EPA is concerned about
any analytical methodology which may be interpreted
to justify conferring less environmental protection on
particular individuals or groups of individuals (e.g.,
the elderly and/or sick). EPA therefore prefers at this
time to continue with its current practice of assigning
the same economic value to incidences of premature
mortality regardless of the age and health status of
those affected, and the primary results presented above
reflect this view. Nevertheless, complete alternative
results based on a value of statistical life-years lost
(VSLY) approach are presented in Chapter 7 and Ap-
pendix I and are summarized below.
Table ES-5 summarizes and compares the results
of the mortality benefits estimates based on the value
of statistical life (VSL) and VSLY approaches. Esti-
mated 1970 to 1990 benefits from PM-related mor-
tality alone and total mortality (i.e., PM plus Lead)
benefits are reported, along with total compliance costs
for the same period. Adding the VSLY-based mortal-
ity benefits estimates to the non-mortality benefits
estimates from Table ES-4 yields the following re-
sults for the overall analysis.
• Alternate Result: The total monetized ben-
efits of the Clean Air Act realized during
the period from 1970 to 1990 range from
4.8 to 28.7 trillion dollars, with a central
estimate of 14.3 trillion dollars.
• Alternate Result: Subtracting costs from
benefits results in net, direct, monetized
benefits ranging from 4.3 to 28.2 trillion
dollars, with a central estimate of 13.7 tril-
lion dollars, for the 1970 to 1990 period.
The results indicate that the choice of valuation
methodology significantly affects the estimated mon-
etized value of historical reductions in air pollution-
related premature mortality. However, the downward
adjustment which would result from applying a VSLY
approach in lieu of a VSL approach does not change
the basic outcome of this study, viz. the estimated
monetized benefits of the historical Clean Air Act
substantially exceed the estimated historical costs of
compliance.
Conclusions and Future
Directions
First and foremost, these results indicate that the
benefits of the Clean Air Act and associated control
programs substantially exceeded costs. Even consid-
ering the large number of important uncertainties per-
meating each step of the analysis, it is extremely un-
likely that the converse could be true.
A second important implication of this study is
that a large proportion of the monetized benefits of
the historical Clean Air Act derive from reducing two
pollutants: lead and paniculate matter4 (see Table ES-
4). Some may argue that, while programs to control
these two pollutants may have yielded measurable
4 Ambient paniculate matter results from emissions of a wide array of precursor pollutants, including sulfur dioxide, nitrogen
oxides, and organic compounds.
——
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
benefits in excess of measurable costs, estimates of
measurable benefits of many other historical Clean
Air Act programs and standards considered in isola-
tion might not have exceeded measurable costs. While
this may or may not be true, this analysis provides no
evidence to support or reject such conjectures. On the
cost side, the historical expenditure data used in this
analysis are not structured in ways which allow attri-
bution of control costs to specific programs or stan-
dards. On the benefit side, most control programs
yielded a variety of benefits, many of which included
reductions in other pollutants such as ambient par-
ticulate matter. For example, new source performance
standards for sulfur dioxide emissions from coal-fired
utility plants yielded benefits beyond those associated
with reducing exposures to gaseous sulfur dioxide.
The reductions in sulfur dioxide emissions also led to
reductions in ambient fine particle sulfates, yielding
human health, ecological, and visibility benefits.
This retrospective study highlights important ar-
eas of uncertainty associated with many of the mon-
etized benefits included in the quantitative analysis
and lists benefit categories which could not be quan-
tified or monetized given the current state of the sci-
ence. Additional research in these areas may reduce
critical uncertainties and/or improve the comprehen-
siveness of future assessments. Particularly important
areas where further research might reduce critical
uncertainties include particulate matter-related mor-
tality incidence, valuation of premature mortality, and
valuation of particulate-related chronic bronchitis and
cardiovascular disease. Additional research on haz-
ardous air pollutants and on air pollution-related
changes in ecosystem structure and function might
help improve the comprehensiveness of future ben-
efit studies. (See Appendix J for further discussion.)
Finally, the results of this retrospective study pro-
vide useful lessons with respect to the value and the
limitations of cost-benefit analysis as a tool for evalu-
ating environmental programs. Cost-benefit analysis
can provide a valuable framework for organizing and
evaluating information on the effects of environmen-
tal programs. When used properly, cost-benefit analy-
sis can help illuminate important effects of changes
in policy and can help set priorities for closing infor-
mation gaps and reducing uncertainty. Such proper
use, however, requires that sufficient levels of time
and resources be provided to permit careful, thorough,
and technically and scientifically sound data-gather-
ing and analysis. When cost-benefit analyses are pre-
sented without effective characterization of the un-
certainties associated with the results, cost-benefit
studies can be used in highly misleading and damag-
ing ways. Given the substantial uncertainties which
permeate cost-benefit assessment of environmental
programs, as demonstrated by the broad range of esti-
mated benefits presented in this study, cost-benefit
analysis is best used to inform, but not dictate, deci-
sions related to environmental protection policies,
programs, and research.
ES-10
-------
1
Introduction
Background and Purpose
As part of the Clean Air Act Amendments of 1990,
Congress established a requirement under section 812
that EPA develop periodic Reports to Congress esti-
mating the benefits and costs of the Clean Air Act
itself. The first such report was to be a retrospective
analysis, with a series of prospective analyses to fol-
low every two years thereafter. This report represents
the retrospective study, covering the period beginning
with passage of the Clean Air Act Amendments of
1970, until 1990 when Congress enacted the most re-
cent comprehensive amendments to the Act.
Since the legislative history associated with sec-
tion 812 is sparse, there is considerable uncertainty
regarding Congressional intent behind the requirement
for periodic cost-benefit evaluations of the Clean Air
Act (CAA). However, EPA believes the principal goal
of these amendments was that EPA should develop,
and periodically exercise, the ability to provide Con-
gress and the public with up-to-date, comprehensive
information about the economic costs, economic ben-
efits, and health, welfare, and ecological effects of
CAA programs. The results of such analyses might
then provide useful information for refinement of CAA
programs during future reauthorizations of the Act.
The retrospective analysis presented in this Re-
port to Congress has been designed to provide an un-
precedented examination of the overall costs and ben-
efits of the historical Clean Air Act. Many other analy-
ses have attempted to identify the isolated effects of
individual standards or programs, but no analysis with
the present degree of validity, breadth and integration
has ever been successfully developed. Despite data
limitations, considerable scientific uncertainties, and
severe resource constraints; the EPA Project Team was
able to develop a broad assessment of the costs and
benefits associated with the major CAA programs of
the 1970 to 1990 period. Beyond the statutory goals
of section 812, EPA intends to use the results of this
study to help support decisions on future investments
in air pollution research. Finally, many of the meth-
odologies and modeling systems developed for the
retrospective study may be applied in the future to the
ongoing series of section 812 prospective studies.
Clean Air Act Requirements,
1970 to 1990
The Clean Air Act establishes a framework for
the attainment and maintenance of clean and health-
ful air quality levels. The Clean Air Act was enacted
in 1970 and amended twice:— in 1977 and most re-
cently in 1990. The 1970 Clean Air Act contained a
number of key provisions. First, EPA was directed to
establish national ambient air quality standards for the
major criteria air pollutants. The states were requked
to develop implementation plans describing how they
would control emission limits from individual sources
to meet and maintain the national standards. Second,
the 1970 CAA contained deadlines and strengthened
enforcement of emission limitations and state plans
with measures involving both the states and the fed-
eral government. Third, the 1970 Act forced new
sources to meet standards based on the best available
technology. Finally, the Clean Air Act of 1970 ad-
dressed hazardous pollutants and automobile exhausts.
The 1977 Clean Air Act Amendments also set new
requirements on clean areas already in attainment with
the national ambient air quality standards. In addition,
the 1977 Amendments set out provisions to help ar-
eas that failed to comply with deadlines for achieve-
ment of the national ambient air quality standards. For
example, permits for new major sources and modifi-
cations were required.
The 1990 Clean Air Act Amendments consider-
ably strengthened the earlier versions of the Act. With
respect to nonattainment, the Act set forth a detailed
and graduated program, reflecting the fact that prob-
lems in some areas are more difficult and complex
than others. The 1990 Act also established a list of
189 regulated hazardous air pollutants and a multi-
step program for controlling emissions of these toxic
air pollutants. Significant control programs were also
established for emissions of acid rain precursors and
stratospheric ozone-depleting chemicals. The biggest
regulatory procedural change in the Act is the new
permit program where all major sources are now re-
quired to obtain an operating permit. Finally, the
amendments considerably expanded the enforcement
provisions of the Clean Air Act, adding administra-
tive penalties and increasing potential civil penalties.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Section 812 of the Clean Air Act
Amendments of 1990
Section 812 of the Clean Air Act Amendments of
1990 requires the EPA to perform a "retrospective"
analysis which assesses the costs and benefits to the
public health, economy and the environment of clean
air legislation enacted prior to the 1990 amendments.
Section 812 directs that EPA shall measure the effects
on "employment, productivity, cost of living, economic
growth, and the overall economy of the United States"
of the Clean Air Act. Section 812 also requires that
EPA consider all of the economic, public health, and
environmental benefits of efforts to comply with air
pollution standards. Finally, section 812 requires EPA
to evaluate the prospective costs and benefits of the
Clean Air Act every two years.
Analytical Design and Review
Target Variable
The retrospective analysis was designed to answer
the following question:
"How do the overall health, welfare,
ecological, and economic benefits of Clean
Air Act programs compare to the costs of
these programs?"
By examining the overall effects of the Clean Air
Act, this analysis complements the Regulatory Impact
Analyses (RIAs) developed by EPA over the years to
evaluate individual regulations. Resources were used
more efficiently by recognizing that these RIAs, and
other EPA analyses, provide complete information
about the costs and benefits of specific rules. Further-
more, in addition to the fact that the RIAs already pro-
vide rule-specific benefit and cost estimates, the broad-
scale approach adopted in the present study precludes
reliable re-estimation of the benefits and costs of in-
dividual standards or programs. On the cost side, this
study relies on aggregated compliance expenditure
data from existing surveys. Unfortunately, these data
do not support reliable allocation of total costs incurred
to specific emissions reductions for the various pol-
lutants emitted from individual facilities. Therefore,
it is infeasible in the context of this study to assign
costs to specific changes in emissions. Further com-
plications emerge on the benefit side. To estimate
benefits, this study calculates the change in incidences
of adverse effects implied by changes in ambient con-
centrations of air pollutants. However, reductions
achieved in emitted pollutants contribute to changes
in ambient concentrations of those, or secondarily
formed, pollutants in ways which are highly complex,
interactive, and often nonlinear. Therefore, even if
costs could be reliably matched to changes in emis-
sions, benefits cannot be reliably matched to changes
in emissions because of the complex, nonlinear rela-
tionships between emissions and the changes in am-
bient concentrations which are used to estimate ben-
efits.
Focusing on the broader target variables of "over-
all costs" and "overall benefits" of the Clean Air Act,
the EPA Project Team adopted an approach based on
construction and comparison of two distinct scenarios:
a "no-control scenario" and a "control scenario." The
no-control scenario essentially freezes federal, state,
and local air pollution controls at the levels of strin-
gency and effectiveness which prevailed in 1970. The
control scenario assumes that all federal, state, and
local rules promulgated pursuant to, or in support of,
the CAA during 1970 to 1990 were implemented. This
analysis then estimates the differences between the
economic and environmental outcomes associated
with these two scenarios. For more information on
the scenarios and their relationship to historical trends,
see Appendix B.
Key Assumptions
Two key assumptions were made during the sce-
nario design process to avoid miring the analytical
process in endless speculation. First, the "no-control"
scenario was defined to reflect the assumption that no
additional air pollution controls were imposed by any
level of government or voluntarily initiated by pri-
vate entities after 1970. Second, it is assumed that the
geographic distribution of population and economic
activity remains the same between the two scenarios.
The first assumption is an obvious oversimplifi-
cation. In the absence of the CAA, one would expect
to see some air pollution abatement activity, either
voluntary or due to state or local regulations. It is con-
ceivable that state and local regulation would have
required air pollution abatement equal to—or even
greater than—that required by the CAA; particularly
since some states, most notably California, have done
so. If one were to assume that state and local regula-
tions would have been equivalent to CAA standards,
then a cost-benefit analysis of the CAA would be a
meaningless exercise since both costs and benefits
would equal zero. Any attempt to predict how state
and local regulations would have differed from the
CAA would be too speculative to support the cred-
ibility of the ensuing analysis. Instead, the no-control
scenario has been structured to reflect the assumption
that states and localities would not have invested fur-
ther in air pollution control programs after 1970 in
the absence of the federal CAA. That is, this analysis
accounts for the costs and benefits of all air pollution
-------
Chapter 1: Introduction
control from 1970 to 1990. Speculation about the pre-
cise fraction of costs and benefits attributable exclu-
sively to the federal CAA is left to others. Neverthe-
less, it is important to note that state and local govern-
ments and private initiatives are responsible for a sig-
nificant portion of these total costs and total benefits.
At the same time, it must also be acknowledged that
the federal CAA played an essential role in achieving
these results by helping minimize the advent of pollu-
tion havens1, establishing greater incentives for pol-
lution control research and development than indi-
vidual state or local rules could provide; organizing
and promoting health and environmental research,
technology transfer and other information management
and dissemination services; addressing critical inter-
state air pollution problems, including the regional fine
particle pollution which is responsible for much of
the estimated monetary benefit of historical air pollu-
tion control; providing financial resources to state and
local government programs; and many other services.
In the end, however, the benefits of historical air pol-
lution controls were achieved through partnerships
among all levels of government and with the active
participation and cooperation of private entities and
individuals.
The second assumption concerns changing demo-
graphic patterns in response to air pollution. In the
hypothetical no-control world, air quality is worse than
that in the historical "control" world particularly in
urban industrial areas. It is possible that in the no-
control case more people, relative to the control case,
would move away from the most heavily polluted ar-
eas. Rather than speculate on the scale of population
movement, the analysis assumes no differences in
demographic patterns between the two scenarios. Simi-
larly, the analysis assumes no changes in the spatial
pattern of economic activity. For example: if, in the
no-control case, an industry is expected to produce
greater output than it did in the control case, that in-
creased output is produced by actual historical plants,
avoiding the need to speculate about the location or
other characteristics of new plants providing additional
productive capacity.
Analytic Sequence
The analysis was designed and implemented in a
sequential manner following seven basic steps which
are summarized below and described in detail later in
this report. The seven major steps were:
direct cost estimation
macroeconomic modeling
emissions modeling
air quality modeling
health and environmental effects estimation
economic valuation
results aggregation and uncertainty character-
ization
By necessity, these components had to be com-
pleted sequentially. The emissions modeling effort had
to be completed entirely before the air quality models
could be configured and run; the air quality modeling
results had to be completed before the health and en-
vironmental consequences of air quality changes could
be derived; and so on. The analytical sequence, and
the modeled versus actual data basis for each analyti-
cal component, are summarized in Figure 1 and de-
scribed in the remainder of this, section.
The first step of the analysis was to estimate the
total direct costs incurred by public and private enti-
ties to comply with post-1970 CAA requirements.
These data were obtained directly from Census Bu-
reau and Bureau of Economic Analysis (BEA) data
on compliance expenditures reported by sources, and
from EPA analyses. These direct cost data were then
adopted as inputs to the macroeconomic model used
to project economic conditions-such as production
levels, prices, employment patterns, and other eco-
nomic indicators-under the two scenarios. To ensure
a consistent basis for scenario comparison, the analy-
sis applied the same macroeconomic modeling sys-
tem to estimate control and no-control scenario eco-
nomic conditions.2 First, a control scenario was con-
structed by running the macroeconomic model using
actual historical data for input factors such as eco-
nomic growth rates during the 1970 to 1990 period.
The model was then re-run for the no-control scenario
by, in essence, returning all post-1970 CAA compli-
ance expenditures to the economy. With these addi-
tional resources available for capital formation, per-
sonal consumption, and other purposes, overall eco-
nomic conditions under the no-control scenario dif-
fered from those of the control scenario. In addition
to providing estimates of the difference in overall eco-
nomic growth and other outcomes under the two sce-
narios, these first two analytical steps were used to
define specific economic conditions used as inputs to
the emissions modeling effort, the first step in the es-
timation of CAA benefits.3
1 "Pollution havens" is a term used to identify individual states or localities which permit comparatively high levels of pollution in
order to attract and hold polluting industries and other activities.
2 Using modeled economic conditions for both scenarios has both advantages and disadvantages. The principal disadvantage is that
historical economic conditions "predicted" by a macroeconomic model will not precisely duplicate actual historical events and condi-
tions. However, this disadvantage is outweighed by the avoidance of distortions and biases which would result from comparing a
modeled no-control scenario with actual historical conditions. By using the same macroeconomic model for both scenarios, model errors
and biases essentially cancel out, yielding more robust estimates of scenario differences, which are what this analysis seeks to evaluate.
3 For example, the macroeconomic model projected different electricity sales levels under the two scenarios, and these sales levels
were used as key input assumptions by the utility sector emissions model. ^^^^
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Figure 1. Summary of Analytical Sequence and Modeled versus Historical Data Basis.
Control Scenario
No-Control Scenario
Compile historical compliance expenditure
data
Develop modeled macroeconomic scenario
based on actual historical economic data
Develop modeled macroeconomic scenario
by rerunning control scenario with
compliance expenditures added back to the
economy
Project emissions by year, pollutant, and
sector using control scenario
macroeconomic projection as input to
sector-specific emissions models
Re-run sector-specific emissions models
using no-control scenario macroeconomic
projection
Develop statistical profiles of historical air
quality for each pollutant based on
historical monitoring data (plus
extrapolations to cover unmonitored areas)
Derive no-control air quality profiles by
adjusting control scenario profiles based on
differences in air quality modeling of
control scenario and no-control scenario
emissions inventories
Estimate physical effects based on
application of concentration-response
functions to historical air quality profiles
Estimate physical effects based on
application of concentration-response
functions to no-control scenario air quality
profiles
Calculate differences in physical outcomes
between control and no-control scenario
Estimate economic value of differences in
physical outcomes between the two
scenarios*
Compare historical, direct compliance costs
with estimated economic value of
monetized benefits, considering additional
benefits which could not be quantified
and/or monetized
* In some cases, economic value is derived directly from physical effects modeling (e.g., agricultural yield loss).
-------
Chapter 1: Introduction
Using appropriate economic indicators from the
macroeconomic model results as inputs, a variety of
emissions models were run to estimate emissions lev-
els under the two scenarios. These emissions models
provided estimates of emissions of six major pollut-
ants4 from each of six key emitting sectors: utilities,
industrial processes, industrial combustion, on-high-
way vehicles, off-highway vehicles, and commercial/
residential sources. The resulting emissions profiles
reflect state-wide total emissions from each pollut-
ant-sector combination for the years 1975,1980,1985,
and 1990.5
The next step toward estimation of benefits in-
volved translating these emissions inventories into
estimates of air quality conditions under each scenario.
Given the complexity, data requirements, and operat-
ing costs of state-of-the-art air quality models-and the
afore-mentioned resource constraints-the EPA Project
Team adopted simplified, linear scaling approaches
for a number of pollutants. However, for ozone and
other pollutants or air quality conditions which involve
substantial non-linear formation effects and/or long-
range atmospheric transport and transformation, the
EPA Project Team invested the time and resources
needed to use more sophisticated modeling systems.
For example, urban area-specific ozone modeling was
conducted for 147 urban areas throughout the 48 con-
tiguous states.
Up to this point of the analysis, both the control
and no-control scenario were based on modeled con-
ditions and outcomes. However, at the air quality
modeling step, the analysis returned to a foundation
based on actual historical conditions and data. Spe-
cifically, actual historical air quality monitoring data
from 1970 to 1990 were used to define the control
scenario. Air quality conditions under the no-control
scenario were then derived by scaling the historical
data adopted for the control scenario by the ratio of
the modeled control and no-control scenario air qual-
ity. This approach took advantage of the richness of
the historical data on air quality, provided a realistic
grounding for the benefit measures, and yet retained
4 These six pollutants are total suspended particulates (TSP), sulfur dioxide (SO2), nitrogen oxides (NOp, carbon monoxide (CO),
volatile organic compounds (VOCs), and lead (Pb). The other CAA criteria pollutant, ozone (Oj), is formed in the atmosphere through
the interaction of sunlight and ozone precursor pollutants such as NOx and VOCs.
5 By definition, 1970 emissions under the two scenarios are identical.
6 The one exception is particulate matter (PM). For PM, air quality profiles for both Total Suspended Particulates (TSP) and
particulates less than or equal to 10 microns in diameter (PM10) were constructed at the county level rather than the individual monitor
level.
7 Or, for PM, by county.
8 For example, changes in forced expiratory volume in one second (FEV,) as a result of exposure to ozone were quantified but
could not be expressed in terms of economic value.
9 Although considerable uncertainties surround the direct cost, macroeconomic modeling, emissions modeling,, and air quality
modeling steps, the ranges of aggregate costs and benefits presented in this analysis do not reflect these uncertainties. While the
uncertainties in these components were assessed qualitatively, and in some cases quantitatively, resource limitations precluded the
multiple macroeconomic model, emissions model, and air quality model runs which would have been required to propagate these
uncertainties through the entire analytical sequence. As a result, complete quantitative measures of the aggregate uncertainty in the cost
and benefit estimates could not be derived. However, the ranges presented do reflect quantitative measures of the uncertainties in the
two most uncertain analytical steps: physical effects estimation and economic valuation. _^
the analytical consistency conferred by using the same
modeling approach for both scenarios. The outputs of
this step of the analysis were statistical profiles for
each pollutant characterizing air quality conditions at
each monitoring site in the lower 48 states.6
The control and no-control scenario air quality
profiles were then used as inputs to a modeling sys-
tem which translates air quality to physical outcomes
-such as mortality, emergency room visits, or crop
yield losses- through the use of concentration-re-
sponse functions. These concentration-response func-
tions were in turn derived from studies found in the
scientific literature on the health and ecological ef-
fects of air pollutants. At this point, estimates were
derived of the differences between the two scenarios
in terms of incidence rates for a broad range of human
health and other effects of air pollution by year, by
pollutant, and by monitor.7
In the next step, economic valuation models or
coefficients were used to estimate the economic value
of the reduction in incidence of those adverse effects
which were amenable to such monetization. For ex-
ample, a distribution of unit values derived from the
economic literature was used to estimate the value of
reductions in mortality risk associated with exposure
to particulate matter. In addition, benefits which could
not be expressed in economic terms were compiled
and are presented herein. In some cases, quantitative
estimates of scenario differences in the incidence of a
nonmonetized effect were calculated.8 In many other
cases, available data and techniques were insufficient
to support anything more than a qualitative character-
ization of the change in effects.
Finally, the costs and monetized benefits were
combined to provide a range of estimates for the par-
tial, net economic benefit of the CAA with the range
reflecting quantified uncertainties associated with the
physical effects and economic valuation steps.9 The
term "partial" is emphasized because only a subset of
the total potential benefits of the CAA could be rep-
resented hi economic terms due to limitations in ancal
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
cal resources, available data and models, and the state
of the science.10 Of paramount concern to the EPA
Project Team was the paucity of concentration-re-
sponse functions needed to translate air quality
changes into measures of ecological effect. In addi-
' tion, significant scientific evidence exists linking air
pollution to a number of adverse human health ef-
fects which could not be effectively quantified and/or
monetized."
Review Process
The CAA requires EPA to consult with an out-
side panel of experts-referred to statutorily as the
Advisory Council on Clean Air Act Compliance
Analysis (the Council)-in developing the section 812
analyses. In addition, EPA is required to consult with
the Department of Labor and the Department of Com-
merce.
The Council was organized in 1991 under the aus-
pices and procedures of EPA's Science Advisory
Board (SAB). Organizing the review committee un-
der the SAB ensured that review of the section 812
studies would be conducted by highly qualified ex-
perts hi an objective, rigorous, and publicly open
manner. The Council has met many times during the
development of the retrospective study to review meth-
odologies and interim results. While the full Council
retains overall review responsibility for the section
peer review by the SAB Council. During the course
of the final interagency discussions, it became clear
that several agencies held different views pertaining
to several key assumptions in this study as well as to
the best techniques to apply in the context of environ-
mental program benefit-cost analyses, including the
present study. The concerns include: (1) the extent to
which air quality would have deteriorated from 1970
to 1990 in the absence of the Clean Air Act, (2) the
methods used to estimate the number of premature
deaths and illnesses avoided due to the CAA, (3) the
methods used to estimate the value that individuals
place on avoiding those risks, and (4) the methods
used to value non-health related benefits. However,
due to the court deadline the resulting concerns were
not resolved during this final, brief interagency re-
view. Therefore, this report reflects the findings of
EPA and not necessarily other agencies in the Ad-
ministration. Interagency discussion of some of these
issues will continue in the context of the future pro-
spective section 812 studies and potential regulatory
actions.
Report Organization
The remainder of the main text of this report sum-
marizes the key methodologies and findings of retro-
spective study. The direct cost estimation and macro-
economic modeling steps are presented in Chapter 2.
The emissions modeling is summarized in Chapter 3.
Chapter 4 presents the air quality'modeling method-
ology and sample results. Chapter 5 describes the ap-
proaches used and principal results obtained through
the physical effects estimation process. Economic
valuation methodologies are described in Chapter 6.
Chapter 7 presents the aggregated results of the cost
and benefit estimates and describes and evaluates
important uncertainties in the results.
Additional details regarding the methodologies
and results are presented in the appendices and in the
referenced supporting documents. Appendix A cov-
ers the direct cost and macroeconomic modeling. Ap-
pendix B provides additional detail on the sector-spe-
cific emissions modeling effort. Details of the air qual-
ity models used and results obtained are presented or
referenced in Appendix C. The effects of the CAA on
human health and visibility; aquatic, wetland, and for-
est ecosystems; and agriculture are presented in Ap-
pendices D, E, and F, respectively. Appendix G pre-
sents details of the lead (Pb) benefits analysis. Air
toxics reduction benefits are discussed in Appendix
H. The methods and assumptions used to value quan-
tified effects of the CAA in economic terms are de-
scribed in Appendix I. Appendix J describes some ar-
eas of research which may increase comprehensive-
ness and reduce uncertainties in effect estimates for
future assessments, and describes plans for future sec-
_tion 812 analyses.
10 It should be noted that there is some uncertainty associated with the estimates of economic costs as well and that some omitted
components of adverse economic consequences of pollution control programs may be significant. For example, some economists
argue that the economic costs of the CAA reported herein may be significantly underestimated to the extent potential adverse effects
of regulation on technological innovation are not captured. Nevertheless, it is clear that the geographic, population, and categorical
coverage of monetary cost effects is significantly greater than coverage of monetized benefits in this analysis.
" For example, while there is strong evidence of a link between exposure to carbon monoxide and reduced time of onset of
angina attack, there are no valuation functions available to estimate the economic loss associated with this effect.
6 -
subcommittees comprised of both Council members
and members of other SAB committees. The Council's
Physical Effects Review Subcommittee met several
times and provided its own review findings to the full
Council. Similarly, the Council's Air Quality Subcom-
mittee, comprised of members and consultants of the
SAB Clean Air Scientific Advisory Committee
(CASAC), held several teleconference meetings to
review methodology proposals and modeling results.
With respect to the interagency review process,
EPA expanded the list of consulted agencies and con-
vened a series of meetings during the design and early
implementation phases from 1991 through late 1994.
In late 1994, to ensure that all interested parties and
the public received consistent information about re-
maining analytical issues and emerging results, EPA
decided to use the public SAB review process as the
primary forum for presenting and discussing issues
and results. The Interagency Review Group was there-
fore discontinued as a separate process in late 1994.
A final, brief interagency review, pursuant to Cir-
cular A-19, was organized in August 1997 by the Of-
fice of Management and Budget and conducted fol-
lowing the completion of the extensive expert panel
-------
2
Cosf and Macroeconornic Effects
The costs of complying with Clean Air Act (CAA)
requirements through the 1970 to 1990 period affected
patterns of industrial production, capital investment,
productivity, consumption, employment, and overall
economic growth. The purpose of the analyses sum-
marized in this chapter was to estimate those direct
costs and the magnitude and significance of resulting
changes to the overall economy. This was accom-
plished by comparing economic indicators under two
alternative scenarios: a control scenario serving as the
historical benchmark, including the historical CAA
as implemented; and a no-control scenario which as-
sumes historical CAA programs did not exist. The
estimated economic consequences of the historical
CAA were taken as the difference between these two
scenarios.
Data used as inputs to the cost analysis can be
classified into two somewhat overlapping categories
based on the information source: survey-based infor-
mation (generally gathered by the Census Bureau) and
information derived from various EPA analyses. For
the most part, cost estimates for stationary air pollu-
tion sources (e.g., factory smokestacks) are based on
surveys of private businesses that attempt to elicit in-
formation on annual pollution control outlays by those
businesses. Estimates of pollution control costs for
mobile sources (e.g., automobiles) are largely based
on EPA analyses, rather than on direct observation
and measurement of compliance expenditures. For
example, to determine one component of the cost of
reducing lead emissions from mobile sources, the
Project Team used an oil refinery production cost
model to calculate the incremental cost required to
produce unleaded (or less-leaded, as appropriate)
rather than leaded gasoline, while maintaining the
octane level produced by leaded gasoline.
As is the case with many policy analyses, a sig-
nificant uncertainty arises in the cost analysis as a
consequence of constructing a hypothetical scenario.
With this retrospective analysis covering almost
twenty years, difficulties arise in projecting alterna-
tive technological development paths. In some cases,
the analytical assumptions used to project the alterna-
tive scenario are not immediately apparent. For ex-
ample, the surveys covering stationary source com-
pliance expenditures require respondents to report
pollution abatement expenditures—implicitly asking
them to determine by how much the company's costs
would decline if there were no CAA compliance re-
quirements. While a response might be relatively
straightforward in the few years following passage of
the CAA, a meaningful response becomes more diffi-
cult after many years of technical change and invest-
ment in less-polluting plant and equipment make it
difficult to determine the degree to which total costs
would differ under a "no CAA" scenario. In cases such
as this, assumptions concerning the alternative hypo-
thetical scenario are made by thousands of individual
survey respondents. Where cost data are derived from
EPA analyses, the hypothetical scenario assumptions
are, at least in theory, more apparent. For example,
when determining the incremental cost caused by pol-
lution-control requirements, one needs to make as-
sumptions (at least implicitly) about what an auto
would look like absent pollution control requirements.
In either case, the need to project hypothetical tech-
nology change for two decades introduces uncertainty
into the assessment results, and this uncertainty may
be difficult to quantify.
The remainder of this chapter summarizes the
basic methods and results of the direct compliance
cost and macroeconomic analyses. Further details re-
garding the modeling methods and assumptions em-
ployed, as well as additional analytical results, are
presented in Appendix A.
Direct Compliance Costs
Compliance with the CAA imposed direct costs
on businesses, consumers, and governmental units; and
triggered other expenditures such as governmental
regulation and monitoring costs and expenditures for
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table 1. Estimated Annual CAA -
Compliance Costs
Exnenditures
Year
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
^current
7.2
8.5
10.6
11.2
11.9
12.0
14.4
16.3
17.0
16.0
15.5
173
19.1
17.8
18.2
18.2
19.0
190
$199O
19.6
21.4
24.4
24.1
24.1
22,6
24.8
25.7
24.4
21.6
20,1
21.6
22,9
20,8
20.6
19.8'
19.8
19 0
' An'ntiallTfeia'Costs
'
2%.
11.0
13.2
13,3
' 14.1
'15.3
15,0
17.3
19.7
19.6
18.6
19:1-
JOfl
'22^5
• 21.1
22.1
22.0
.22.9-
23.6
$1990 'a
...5%
11.0
13,4
43.fr -
•14,6 -;
15.9 v*
15.8-
18.3 ^
2Q,8 S-
20.9*V
' 20U' "
20.7 '
21.9 '
24.4
23.2'
l-24.^f
"*2^73* f/"
-*25:#f?
26.1 '•'
t:
7%
11. 1
13.7'
14.0 ;
415!4'"'
-16*6*
163-;
19.3-
-22:0
22,3;*
'Si'!? '\
22,5;'
23.8
26,5'
25^.
'}i>>. x"1"1
*wv»U
IpxJ^
^JS^
IX .7
research and development by both government and
industry. Although expenditures unadjusted for infla-
tion — that is, expenditures denominated in "current
dollars"— increased steadily from $7 billion to $19
billion per year over the 1973 to 1990 period,12 an-
nual CAA compliance expenditures adjusted for in-
flation were relatively stable, averaging near $25 bil-
lion (in 1990 dollars) during the 1970s and close to
$20 billion during most of the 1980s (see Table 1).
Aggregate compliance expenditures were somewhat
less than one half of one percent of total domestic
output during that period, with the percentage falling
from two thirds of one percent of total output in 1975
to one third of one percent in 1990.
Although useful for many purposes, a summary
of direct annual expenditures may not the best cost
measure to use when comparing costs to benefits.
Capital expenditures are investments, generating a
stream of benefits and opportunity cost13 over the life
of the investment. The appropriate accounting tech-
nique to use for capital expenditures in a cost/benefit
analysis is to annualize the expenditure. This tech-
nique, analogous to calculating the monthly payment
associated with a home mortgage, involves spreading
the cost of the capital equipment over the useful life
of the equipment using a discount rate to account for
the time value of money.
For this cost/benefit analysis, "annualized" costs
reported for any given year are equal to O&M expen-
ditures — including R&D and other similarly recur-
ring expenditures — plus amortized capital costs (i.e.,
depreciation plus interest costs associated with the
existing capital stock) for that year. Stationary source
air pollution control capital costs were amortized over
20 years; mobile source air pollution control costs were
amortized over 10 years.14 All capital expenditures
were annualized using a five percent, inflation-ad-
justed rate of interest. Additionally, annualized costs
were calculated using discount rates of three and seven
percent to determine the sensitivity of the cost results
to changes in the discount rate. Table 1 summarizes
costs annualized at three, five, and seven percent, as
well as annual expenditures.
Total expenditures over the 1973-1990 period,
discounted to 1990 using a five percent (net of infla-
tion) discount rate, amount to 628 billion dollars (in
1990 dollars). Discounting the annualized cost stream
to 1990 (with both annualization and discounting pro-
cedures using a five percent rate) gives total costs of
523 billion dollars (in 1990 dollars). Aggregate annu-
alized costs are less than expenditures because the
annualization procedure spreads some of the capital
cost beyond 1990.15
12 Due to data limitations, the cost analysis for this CAA retrospective starts in 1973, missing costs incurred in 1970-72. This
limitation is not likely to be significant, however, because relatively little in the way of compliance with the "new" provisions of the
1970 CAA was required in the first two years following passage.
13 In this context, "opportunity cost" is defined as the value of alternative investments or other uses of funds foregone as a result of
the investment.
14 Although complete data are available only for the period 1973-1990, EPA's Cost of Clean report includes capital expenditures
for 1972 (see Appendix A for more details and complete citation). Those capital expenditure data have been used here. Therefore,
amortized costs arising from 1972 capital investments are included in the 1973-1990 annualized costs, even though 1972 costs are not
otherwise included in the analysis. Conversely, some capital expenditures incurred in the 1973-1990 period are not reflected in the
1973-1990 annualized costs — those costs are spread through the following two decades, thus falling outside of the scope of this study
(e.g., only one year of depreciation and interest expense is included for 1989 capital expenditures). Similarly, benefits arising from
emission reductions realized after 1990 as a result of capital investments made during the 1970 to 1990 period of this analysis are not
included in the estimates of benefits included in this report.
13 This adjustment is required because many 1970 to 1990 investments in control equipment continue to yield benefits beyond
1990. Annualization of costs beyond 1990 ensures that the costs and benefits of any particular investment are properly scaled and
matched over the lifetime of the investment.
8
-------
Chapter 2: Cost and Macroeconomic Effect
Indirect Effects of the CAA
Through changing production costs, CAA imple-
mentation induced changes in consumer good prices,
and thus in the size and composition of economic out-
put. The Project Team used a general equilibrium
macroeconomic model to ,assess the extent of such
second-order effects. This type of model is useful be-
cause it can capture the feedback effects of an action.
In the section 812 macroeconomic modeling exercise,
the feedback effects arising from expenditure changes
were captured, but the analogous effects arising from
improvements in human health were not captured by
the model. For example, the macroeconomic model
results do not reflect the indirect economic effects of
worker productivity improvements and medical ex-
penditure savings caused by the CAA. Consequently,
the macroeconomic modeling exercise provides lim-
ited and incomplete information on the type and po-
tential scale of indirect economic effects.
The effects estimated by the macroeconomic
model can be grouped into two broad classes: sectoral
impacts (i.e., changes in the composition of economic
output), and aggregate effects (i.e., changes in the
degree of output or of some measure of human wel-
fare). The predicted sectoral effects were used as in-
puts to the emissions models as discussed in Chapter
3. In general, the estimated second-order macroeco-
nomic effects were small relative to the size of the
U.S. economy. See Appendix A for more detail on
data sources, analytical methods, and results for the
macroeconomic modeling performed for this assess-
ment.
Sectoral Impacts
The CAA had variable compliance impacts across
economic sectors. The greatest effects were on the
largest energy producers and consumers, particularly
those sectors which relied most heavily on consump-
tion of fossil fuels (or .energy generated from fossil
fuels). In addition, production costs increased more
for capital-intensive industries than for less capital-
intensive industries under the control scenario due to
a projected increase in interest rates. The interest rate
increase, which resulted in an increase in the cost of
capital, occurred under the control scenario because
CAA-mandated investment in pollution abatement
reduced the level of resources available for other uses,
including capital formation.
Generally, the estimated difference in cost impacts
under the control and no-control scenarios for a par-
ticular economic sector was a function of the relative
energy-intensity and capital-intensity of that sector.
Increased production costs in energy- and capital-in-
tensive sectors under the control scenario were re-
flected hi higher consumer prices, which resulted hi
reductions in the quantity of consumer purchases of
goods and services produced by those sectors. This
reduction in consumer demand under the control sce-
nario led, ultimately, to reductions in output and em-
ployment in those sectors. The sectors most affected
by the CAA were motor vehicles, petroleum refining,
and electricity generation. The electricity generation
sector, for example, incurred a two to four percent
increase in consumer prices by 1990, resulting in a
three to five and a half percent reduction in output.
Many other manufacturing sectors saw an output ef-
fect in the one percent range.
Some other sectors, however, were projected to
increase output under the control scenario. Apart from
the pollution control equipment industry, which was
not separately identified and captured in the macro-
economic modeling performed for this study, two ex-
ample sectors for which output was higher and prices
were lower under the control scenario are food and
furniture. These two sectors showed production cost
and consumer price reductions of one to two percent
relative to other industries under the control scenario,
resulting in output and employment increases of simi-
lar magnitudes.
Aggregate Effects
As noted above, the control and no-control sce-
narios yield different estimated mixes of investment.
In particular, the control scenario was associated with
more pollution control capital expenditure and less
consumer commodity capital expenditure. As a result,
the growth pattern of the economy under the control
scenario differed from the no-control scenario. Under
the control scenario, the macroeconomic model pro-
jected a rate of long-run GNP growth about one twen-
tieth of one percent per year lower than under the no-
control scenario. Aggregating these slower growth
effects of the control scenario over the entire 1970 to
1990 period of this study results, by 1990, in a level
of GNP one percent (or approximately $55 billion)
lower than that projected under the no-control sce-
nario.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Although small relative to the economy as a whole,
the estimated changes hi GNP imply that the poten-
tial impact of the CAA on the economy by 1990 was
greater than that implied by expenditures ($19 billion
in 1990) or annualized costs ($26 billion in 1990, an-
nualized at five percent). Discounting the stream of
1973-1990 GNP effects to 1990 gives an aggregate
impact on production of 1,005 billion dollars (in 1990
dollars discounted at five percent). Of that total, $569
billion represent reductions in household consump-
tion, and another $200 billion represent government
consumption, for an aggregate effect on U.S. consump-
tion of goods and services equal to 769 billion dol-
lars. Both the aggregate GNP effects and aggregate
consumption effects exceed total 1973-1990 expen-
ditures ($628 billion) and annualized costs ($523 bil-
lion, with all dollar quantities in $1990, discounted at
five percent).
Changes in GNP (or, even, changes in the national
product account category "consumption") do not nec-
essarily provide a good indication of changes in so-
cial welfare. Social welfare is not improved, for ex-
ample, by major oil tanker spills even though mea-
sured GNP is increased by the "production" associ-
ated with clean-up activities. Nevertheless, the effects
of the CAA on long-term economic growth would be
expected to have had some effect on economic wel-
fare. One of the characteristics of the macroeconomic
model used by the Project Team is its ability to esti-
mate a measure of social welfare change which is su-
perior to GNP changes. This social welfare measure
estimates the monetary compensation which would be
required to offset the losses hi consumption (broadly
defined) associated with a given policy change. The
model reports a range of results, with the range sensi-
tive to assumptions regarding how cost impacts are
distributed through society. For the CAA, the model
reports an aggregate welfare effect of 493 billion to
621 billion dollars (in 1990 dollars), depending on
the distributional assumptions used. This range does
not differ greatly from the range of results represented
by 1973-1990 expenditures, compliance costs, and
consumption changes.
Uncertainties and Sensitivities in
the Cost and Macroeconomic
Analysis
The cost and macroeconomic analyses for the
present assessment relied upon survey responses, EPA
analyses, and a macroeconomic simulation model.
Although the Project Team believes that the results of
the cost and macroeconomic analyses are reasonably
reliable, it recognizes that every analytical step is sub-
ject to uncertainty. As noted at the beginning of this
chapter, explicit and implicit assumptions regarding
hypothetical technology development paths are cru-
cial to framing the question of the cost impact of the
CAA. In addition, there is no way to verify the accu-
racy of the survey results used;16 alternative, plausible
cost analyses exist that arrive at results that differ from
some of the results derived from EPA analyses; and it
is not clear how the use of a general equilibrium mac-
roeconomic model affects the accuracy of macroeco-
nomic projections in a macroeconomy characterized
by disequilibrium. For many factors engendering un-
certainty, the degree or even the direction of bias is
unknown. In several areas, nevertheless, uncertainties
and/or sensitivities can be identified that may bias the
results of the analysis.
Productivity and Technical Change
An important component of the macroeconomic
model used by the Project Team is its treatment of
technical change and productivity growth. Three fac-
tors associated with productivity and technical change
have been identified which may bias the results of the
macroeconomic simulation: (1) the long-run effects
of reducing the "stock" of technology, (2) the pos-
sible "chilling" effect of regulations on innovation and
technical change, and (3) the role of endogenous pro-
ductivity growth within the macroeconomic model.
The macroeconomic model projected a decrease
in the growth of GNP as a result of CAA compliance.
Decreased growth was due not only to decreased capi-
tal investment, but also to decreased factor productiv-
ity. The annual decrement in productivity can be
thought of as a reduction of the stock of available tech-
nology. That reduction in stock could be expected to
affect macroeconomic activity after 1990, as well as
16 For an example of the difficulties one encounters in assessing the veracity of survey results, see the discussion in Appendix A
on the apparently anomalous growth in stationary source O&M expenditures in relation to the size of the stationary source air
pollution control capital stock.
10
-------
Chapter 2: Cost and Macroecoaomic Effect
during the 1973-1990 period studied by the Project
Team. Thus, to the extent that this effect exists, the
Project Team has underestimated the macroeconomic
impact of the CAA by disregarding the effect of 1973-
1990 productivity change decrements on post-1990
GNP.
Some economists contend that regulations have a
"chilling" effect on technological innovation and,
hence, on productivity growth. Two recent studies by
Gray and Shadbegian,17 which are sometimes cited in
support of this contention, suggest that pollution abate-
ment regulations may decrease productivity levels in
some manufacturing industries. The macroeconomic
model allowed policy-induced productivity change
through the mechanism of price changes and result-
ant factor share changes. To the extent that additional
policy-induced effects on productivity growth exist,
the Project Team has underestimated the impact of
the CAA on productivity growth during the 1973-1990
period, and, thus, has underestimated macroeconomic
impacts during the 1973-1990 period and beyond.
The macroeconomic model allowed productivity
growth to vary with changes in prices generated by
the model. This use of "endogenous" productivity
growth is not universal in the economic growth litera-
ture — that is, many similar macroeconomic models
do not employ analogous forms of productivity growth.
The Project Team tested the sensitivity of the model
results to the use of endogenous productivity growth.
If the model is run without endogenous productivity
growth, then the predicted macroeconomic impacts
(GNP, personal consumption, etc.) of the CAA are
reduced by approximately 20 percent. That is, to the
extent that use of endogenous productivity growth hi
the macroeconomic model is an inaccurate simulation
technique, then the Project Team has overestimated
the macroeconomic impact of the CAA.
Discount Rates
approximates the social rate of tune preference — that
is, the rate of return at which individuals are willing
to defer consumption to the future. A three percent
rate would approximate the social rate of time prefer-
ence (all rates used here are "real", i.e., net of price
inflation impacts). Others believe that a rate that ap-
proximates the opportunity cost of capital (e.g., seven
percent or greater) should be used.18 A third school of
thought holds that some combination of the social rate
of time preference and the opportunity cost of capital
is appropriate, with the combination effected either
by use of an intermediate rate or by use of a multiple-
step procedure employing the social rate of time pref-
erence as the "discount rate," but still accounting for
the opportunity cost of capital.
The Project Team elected to use an intermediate
rate (five percent), but recognizes that analytical re-
sults aggregated across the study period are sensitive
to the discount rate used. Consequently, all cost mea-
sures are presented at three and seven percent, as well
as the base case five percent. Table 2 summarizes
major cost and macroeconomic impact measures ex-
pressed in constant 1990 dollars, and discounted to
1990 at rates of three, five, and seven percent.
-,-417 ':-
8SO - "
, ,62,? ' ", 761 ,
/523, ,„ '657
569;-,
769 RR1
There is a broad range of opinion in the econom-
ics profession regarding the appropriate discount rate
to use in analyses such as the current assessment. Some
economists believe that the appropriate rate is one that
17 Gray, Wayne B., and Ronald J. Shadbegian, "Environmental Regulation and Manufacturing Productivity at the Plant Level,"
Center for Economic Studies Discussion Paper, CES 93-6, March 1993. Gray, Wayne B., and Ronald J. Shadbegian, "Pollution
Abatement Costs, Regulation, and Plant-Level Productivity," National Bureau of Economic Research, Inc., Working Paper Series,
Working Paper No. 4994, January 1995.
18 Some would argue that use of the opportunity cost of capital approach would be inappropriate in the current assessment if the
results of the macroeconomic modeling (such as GNP) were used as the definition of "cost," since the macro model already accounts
for the opportunity cost of capital. The appropriate rate would then be the social rate of time preference.
— —
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Exclusion of Health Benefits from the
Macroeconomic Model
The macroeconomic modeling exercise was de-
signed to capture the second-order macroeconomic
effects arising from CAA compliance expenditures.
Those predicted second-order effects are among the
factors used to drive the emissions estimates and, ul-
timately, the benefits modeled for this assessment. The
benefits of the CAA, however, would also be expected
to induce second-order macroeconomic effects. For
example, increased longevity and decreased incidence
of non-fatal heart attacks and strokes would be ex-
pected to improve macroeconomic performance mea-
sures. The structure of the overall analysis, however,
necessitated that these impacts be excluded from the
macroeconomic simulation.
The first-order CAA beneficial effects have been
included in the benefits analysis for this study, includ-
ing measures that approximate production changes
(e.g., income loss due to illness, or lost or restricted
work days; income loss due to impaired cognitive abil-
ity; and income loss due to reduced worker produc-
tion in certain economic sectors). These measures are
analogous to compliance expenditures in the cost
analysis. The second-order benefits impacts, which
would result from price changes induced by CAA-
related benefits, have not been estimated. It is likely
that the estimated adverse second-order macroeco-
nomic impacts would have been reduced had the im-
pact of CAA benefits been included in the macroeco-
nomic modeling exercise; however, the magnitude of
this potential upward bias in the estimate of adverse
macroeconomic impact was not quantitatively as-
sessed.
12
-------
3
Emissions
This chapter presents estimates of emissions re-
ductions due to the Clean Air Act (CAA) for six crite-
ria air pollutants. Reductions are calculated by esti-
mating, on a sector-by-sector basis, the differences in
emissions between the control and no-control sce-
narios. While the relevant years in this analysis are
1970 through 1990, full reporting of emissions was
only made for the 1975 to 1990 period since 1970
emission levels are, by assumption, identical for the
two scenarios. The criteria pollutants for which emis-
sions are reported in this analysis are: total suspended
particulates (TSP),19 carbon monoxide (CO), volatile
organic compounds (VOC), sulfur dioxide (SO2), ni-
trogen oxides (NOx), and Lead (Pb).
The purpose of the present study is to estimate
the differences in economic and environmental con-
ditions between a scenario reflecting implementation
of historical CAA controls and a scenario which as-
sumes that no additional CAA-related control pro-
grams were introduced after 1970. Because of the fo-
cus on differences in -rather than absolute levels of-
emissions between the scenarios, the various sector-
specific emission models were used to estimate both
the control and no-control scenario emission invento-
ries. This approach ensures that differences between
the scenarios are not distorted by differences between
modeled and actual historical emission estimates.20
Despite the use of models to estimate control sce-
nario emission inventories, the models used were con-
figured and/or calibrated using historical emissions
estimates. The control scenario utility emissions esti-
mates, for example, were based on the ICF CEUM
model which was calibrated using historical emissions
inventory data.21 In other cases, such as the EPA Emis-
sions Trends Report (Trends) methodology22 used to
estimate industrial process emissions, historical data
were used as the basis for control scenario emissions
with little or no subsequent modification. Neverthe-
less, differences in model selection, model configura-
tion, and macroeconomic input data23 result in un-
avoidable, but in this case justifiable, differences be-
tween national total historical emission estimates and
national total control scenario emission estimates for
each pollutant. Comparisons between no-control, con-
trol, and official EPA Trends Report historical emis-
sions inventories are presented in Appendix B.24
19 In 1987, EPA replaced the earlier TSP standard with a standard for paniculate matter of 10 microns or smaller (PM10).
20 By necessity, emission models must be used to estimate the hypothetical no-CAA scenario. If actual historical emissions data
were adopted for the control scenario, differences between the monitoring data and/or models used to develop historical emission
inventories and the models used to develop no-control scenario emission estimates would bias the estimates of the differences between
the scenarios.
21 See ICF Resources, Inc., "Results of Retrospective Electric Utility Clean Air Act Analysis -1980,1985 and 1990," September
30, 1992, Appendix C.
22 EPA, 1994a: U.S. Environmental Protection Agency, "National Air Pollutant Emission Trends, 1900-1993," EPA-454/R-94-
027, Office of Air Quality Planning and Standards, Research Triangle Park, NC, October 1994.
23 The Jorgenson/Wilcoxen macroeconomic model outputs were used to configure both the control and no-control scenario
emission model runs. While this satisfies the primary objective of avoiding "across model" bias between the scenarios, the macroeco-
nomic conditions associated with the control scenario would not be expected to match actual historical economic events and condi-
tions. To the extent actual historical economic conditions are used to estimate official historical emission inventories, conformity
between these historical emissions estimates and control scenario emission estimates would be further reduced.
24 In general, these comparisons show close correspondence between control scenario and Trends estimates with the largest
differences occurring for VOC and CO emissions. The Trends report VOC estimates are generally higher than the control scenario
estimates due primarily to the inclusion of Waste Disposal and Recycling as a VOC source in the Trends report. This inconsistency is
of no consequence since Waste Disposal and Recycling sources were essentially uncontrolled by the historical CAA and therefore do
not appear as a difference between the control and no-control scenarios. The higher CO emission estimates in the Trends Report are
primarily associated with higher off-highway vehicle emissions estimates. Again, since off-highway emissions do not change between
the control and no-control scenario in the present analysis, this inconsistency is of no consequence.
13
-------
The Benefits and Costs of the Clean Air Act, 2970 to 1990
To estimate no-control scenario emissions, sec-
tor-specific historical emissions are adjusted based on
changes in the following two factors: (1) growth by
sector predicted to occur under the no-control scenario;
and (2) the exclusion of controls attributable to spe-
cific provisions of the CAA.
To adjust emissions for economic changes under
the no-control scenario, activity levels that affect emis-
sions from each sector were identified. These activity
levels include, for example, fuel use, industrial activ-
ity, and vehicle miles traveled (VMT). The Jorgenson-
Wilcoxen (J/W) general equilibrium model was used
to estimate changes in general economic conditions,
as well as sector-specific economic outcomes used as
inputs to the individual sector emission models.25
Table 3. Snmmarv of Sector-Secific Bnassid
rtfe:-Approad'hes, ''"*
Sector
» ^ -, »;>,i , ' <>&i^ i' 'x^l' ** " ^
gO analyzed i^(ijij5,tttal*proc;es& emissions basf'djf Breads- ra&fidds, Adjasteil
htstotieal etaiss|o|^^p4AVs-Aectotaichaagfes'in'QUtpitt, attd 1970 cotttroU- -
""' *••*>•*'•
•:t;e'ad;ejj|j|si.oas,calculated for JMastrial^rc^es'ses and-
*" ****'"* '' '"'- emissionfaetorsT;an
a5 System
*"
25 For example, the change in distribution of households by income class predicted by the J/W model was used as input to the
transportation sector model system. Changes in household income resulted in changes in vehicle ownership and usage patterns which,
in turn, influence VMT and emissions. (See Pechan, 1995, p. 43).
-------
Chapter 3: Emissions
The specific outputs from the J/W model used in
this analysis are the percentage changes in gross na-
tional product (GNP), personal consumption, and out-
put for various economic sectors under the control and
no-control scenario for the years 1975, 1980, 1985,
and 1990.26 The sectors for which the results of the J/
W model are used include: industrial processes, elec-
tric utilities, highway vehicles, industrial boilers, and
the commercial/residential sector. For the off-highway
sector, economic growth was not taken into account
as there was no direct correspondence between J/W
sectors and the off-highway vehicle source category
activity.
In addition to adjusting for economic activity
changes, any CAA-related control efficiencies that
were applied to calculate control scenario emissions
were removed for the no-control scenario. In most
instances, emissions were recalculated based on 1970
control levels.
Uncertainty associated with several key model-
ing inputs and processes may contribute to potential
errors hi the emission estimates presented herein. Al-
though the potential errors are likely to contribute in
only a minor way to overall uncertainty hi the esti-
mated monetary benefits of the Clean Air Act, the most
significant emission modeling uncertainties are de-
scribed at the end of this chapter.
Sector-Specific Approach
The approaches used to calculate emissions for
each sector vary based on the complexity of estimat-
ing emissions in the absence of CAA controls, taking
economic activity levels and CAA regulations into
account. For the off-highway vehicle and industrial
process sectors, a relatively simple methodology was
developed. The approaches used for the highway ve-
hicles, electric utilities, industrial boilers, and com-
mercial/residential sectors were more complex be-
cause the J/W model does not address all of the deter-
minants of economic activity in these sectors that
might have changed hi the absence of regulation. The
approaches by sector used to estimate emissions for
the two scenarios are summarized hi Table 3, and are
described in more detail in Appendix B.
Summary of Results
Figure 2 compares the total estimated sulfur di-
oxide emission from all sectors under the control and
no-control scenarios over the period from 1975 to
26 For details regarding the data linkages between the J/W model and the various emission sector models, see Pechan (1995).
15
1990. Figures 3, 4, 5, 6, and 7 provide similar com-
parisons for NOx, VOCs, CO, TSP, and Lead (Pb) re-
spectively.
Additional tables presented in Appendix B pro-
vide further breakdown of the emissions estimates by
individual sector. The essential results are character-
ized below. For most sectors, emission levels under
the control scenario were substantially lower than lev-
els projected under the no-control scenario. For some
pollutants, for example NOx, most of the reductions
achieved under the control scenario offset the growth
in emissions which would have occurred under the
no-control case as a result of increases in population
and economic activity. For other pollutants, particu-
larly lead, most of the difference hi 1990 emissions
projected under the two scenarios reflects significant
improvement relative to 1970 emission levels. Ap-
pendix B also assesses the consistency of the control
and no-control scenario estimates for 1970 to 1990
with pre-1970 historical emissions trends data.
The CAA controls that affected SO2 emitting
sources had the greatest proportional effect on indus-
trial process emissions, which were 60 percent lower
in 1990 than they would have been under the
no-control scenario. SO2 emissions from electric utili-
ties and industrial boilers were each nearly 40 percent
lower in 1990 as a result of the controls. In terms of
absolute tons of emission reductions, controls on elec-
tric utilities account for over 10 million of the total 16
million ton difference between the 1990 control and
no-control scenario SO2 emission estimates.
CAA regulation of the highway vehicles sector
led to the greatest percent reductions hi VOC and NO .
Control scenario emissions of these pollutants hi 1990
were 66 percent and 47 percent lower, respectively,
than the levels estimated under the no-control scenario.
In absolute terms, highway vehicle VOC controls ac-
count for over 15 million of the roughly 17 million
ton difference hi control and no-control scenario emis-
sions.
Differences between control and no-control sce-
nario CO emissions are also most significant for high-
way vehicles. In percentage terms, highway vehicle
CO emissions were 56 percent lower in 1990 under
the control scenario than under the no-control scenario.
Industrial process CO emission estimates under the
control scenario were about half the levels projected
under the no-control scenario. Of the roughly 89 mil-
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Figure 2. Control and No-control Scenario Total SO2
Emission Estimates.
40
30
S 10
1975
(Control I
.No-Control
1980 1985
Year
1990
Figure 5. Control and No-control Scenario Total CO
Emission Estimates.
200
150
.S = 10°
3 §
•1 50
Ol_J_
1975
t Control
. No-Control
1980 1985
Year
1990
1*
Emissions in Short Tons era'
Millions JJ
— to u «. to
00000 .
Control and No-control Scenario Total NOX
Emission Estimates.
.__-
^=^-m m •
1111
1975 1980 1985 1990
Year
v Control 1
^.No-ControB
Figure 6. (
1
40
I 3<>
1 s
•a 1 20
1 10
0
Control and No-control Scenario Total TSP
Emission Estimates.
»-- _ _
-•- m m
I I I i
jl Control
^. No-Control
1975 1980 1985 1990
Year
Figure 4. Control and No-control Scenario Total VOC
Emission Estimates.
40
J 30
i 20
2
10
1975
1980 1985
Year
1990
Figure 7. Control and No-control Scenario Total Pb
Emission Estimates.
200
150
•3 I 100
50
1975
I Control
..No-Control
1980 1985
Year
1990
16
-------
ChapterS: Emissions
lion ton difference in CO emissions between the two
scenarios, 84 million tons are attributable to highway
vehicle controls and the rest is associated with reduc-
tions from industrial process emissions.
For TSP, the highest level of reductions on a per-
centage basis was achieved in the electric utilities sec-
tor. TSP emissions from electric utilities were 93 per-
cent lower in 1990 under the control scenario than
projected under the no-control scenario. TSP emis-
sions from industrial processes were also significantly
lower on a percentage basis under the control scenario,
with the differential reaching 76 percent by 1990.
This is not an unexpected result as air pollution
control regulations in the 1970's focused on solving
the visible particulate problems from large fuel com-
bustors. In terms of absolute tons, electric utilities
account for over 5 million of the 16 million ton differ-
ence between the two scenarios and industrial pro-
cesses account for almost 10 million tons.
The vast majority of the difference in lead emis-
sions under the two scenarios is attributable to reduc-
tions in burning of leaded gasoline. By 1990, reduc-
tions in highway vehicle emissions account for 221
thousand of the total 234 thousand ton difference in
lead emissions. As shown in more detail hi Appendix
B, airborne lead emissions from all sectors were vir-
tually eliminated by 1990.
As described in the following chapter and in Ap-
pendix C, these emissions inventories were used as
inputs to a series of air quality models. These air qual-
ity models were used to estimate air quality condi-
tions under the control and no-control scenarios.
Uncertainty in the Emissions
Estimates
The emissions inventories developed for the con-
trol and no-control scenarios reflect at least two ma-
jor, sources of uncertainty. First, potential errors in the
macroeconomic scenarios used to configure the sec-
tor-specific emissions model contribute to uncertain-
ties in the emissions model outputs. Second, the emis-
sions models themselves rely on emission factors,
source allocation, source location, and other param-
eters which may be erroneous.
An important specific source of potential error
manifest in the present study relates to hypothetical
emission rates from various sources under the no-con-
trol scenario. Emission rates from motor vehicles, for
example, would have been expected to change during
the 1970 to 1990 period due to technological changes
not directly related to implementation of the Clean
Air Act (e.g., advent of electronic fuel injection, or
EFT). However, the lack of emissions data from ve-
hicles with EFI but without catalytic converters com-
pelled the Project Team to use 1970 emission factors
throughout the 1970 to 1990 period for the no-control
scenario. Although this creates a potential bias hi the
emissions inventories, the potential errors from this
and other uncertainties in the emissions inventories
are considered unlikely to contribute significantly to
overall uncertainty in the monetary estimates of Clean
Air Act benefits. This conclusion is based on the de-
monstrably greater influence on the monetary benefit
estimates of uncertainties in other analytical compo-
nents (e.g., concentration-response functions). A list
of the most significant potential errors in the emis-
sions modeling, and their significance relative to over-
all uncertainty in the monetary benefit estimate, is
presented in Table 4.
17
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
! *
Iable4. Uncertainties iSsociatei-widj'^i
' ... fl
Potential Source of Error , ^
' ,'
•>
' r "-f t tf
Use of 1 970 motor vehicle emission factors"'1
for no-control scenario without adjustment .,'
for advent of Electronic Fuel fcjectjon - -
(EFT) and Electronic Ignition (El), ^ ; ,
Use of ARGUS for utility CB atid^VOG-*'-' '
ratherthan CBUM. ,- - ^. ^^^'
1 s* =' ,'^V^"^ '/'.', ''
Use of historical fuel consttmptiOft4o^^v " ;
emissions. - ->~" -f *-> *-^^^-^-
^ ^ "nf *&.!>*••*•'& v f\ " V
Adoption of assomptioa that.utiiit^iinite^V^
inventories remain fixed between the ~>.- -*'-j5-
control and no-control scenarios: •' • "* "^
f, ^^x^^S^^i:^**
- V ;- <" «" ' t'^-0^**''?
.,-.', ^ . -'-ft*"**
,, * ,-- • ! -<»<**•
, , > ^ <*^«<as -Modeling-, - -" ';
^ -B,!as4a'-lEsBniate of
^IntissSon, Eedac^on - -
,,^«Be^fits
-'-ji-;4-%«^f«-% ; • ' ,,,,,'-
p^^jp1^---, ',„:,';;,'.
» , ;< >x<" *-«--»"-'; '• - ,<.^,^
, ,^ ,„ ,-- ,_ ,
' *s*i"--^Xv ,;• •
Unlcftown. .-,* ,---' ,
* ' '> "-"'y^S' /r
'*" ' ^ , \ftyfy > " "
, , ^f^fff^V ,^" *,
1 Ov^c^wtuili&tex x-s-v^ o s
*«>-«;--:,-- ,f,:'.",f~^ie»,,^
>S",,;/X. - -. .- ,^ ,<;-;,|<^^
'-t<*x''"i'''?';; ' ' -'-v*
^ ^wv^"''''V^ ;/ '
'0 Vvg'^V^,-^.1?.^. ,^^, , •- , / s
d^^v/X^^' ^* ' " "
, , f t ^^(.AW^'/A 'f't '' ^ ''"' f ' ' "oV* -5 _
'"''' 'J,.,/-,
''Unknp^nV,i(,at'-}iiely-to be minor |ae ,
t^°o^qyh.elBiing sigiaificance qt-,.,,,,
ckalysts in detetrninina emission
^Negligible,- -: '-; '-'\'- >- '^
Negligible, •- :- ' , , '
;*<•'><••''>""' ' , ^, ^ •
,'• «;,<,;,'«-" •?>--'-- t,^
Un known, b ut likely^tONbe snial
siace.theCAA had«v
1 cpmtprise,a.large',majo!ity-
-------
4
Air Quality
Air quality modeling is the crucial analytical step
which links emissions to changes in atmospheric con-
centrations of pollutants which affect human health
and the environment. It is also one of the more com-
plex and resource-intensive steps, and contributes sig-
nificantly to overall uncertainty hi the bottom-line
estimate of net benefits of air pollution control pro-
grams. The assumptions required to estimate hypo-
thetical no-control scenario air quality conditions are
particularly significant sources of uncertainty in the
estimates of air quality change, especially for those
pollutants which are not linearly related to changes in
associated emissions. Specific uncertainties are de-
scribed in detail at the end of this chapter.
The key challenges faced by air quality modelers
attempting to translate emission inventories into air
quality measures involve modeling of pollutant dis-
persion and atmospheric transport, and modeling of
atmospheric chemistry and pollutant transformation.
These challenges are particularly acute for those pol-
lutants which, rather than being directly emitted, are
formed through secondary formation processes. Ozone
is the paramount example since it is formed in the
atmosphere through complex, nonlinear chemical in-
teractions of precursor pollutants, particularly vola-
tile organic compounds (VOCs) and nitrogen oxides
(NOx). In addition, atmospheric transport and trans-
formation of gaseous sulfur dioxide and nitrogen ox-
ides to particulate sulfates and nitrates, respectively,
contributes significantly to ambient concentrations of
fine particulate matter. In addition to managing the
complex atmospheric chemistry relevant for some
pollutants, air quality modelers also must deal with
uncertainties associated with variable meteorology and
the spatial and temporal distribution of emissions.
Given its comprehensive nature, the present analy-
sis entails all of the aforementioned challenges, and
involves additional complications as well. For many
pollutants which cause a variety of human health and
environmental effects, the concentration-response
functions which have been developed to estimate those
effects require, as inputs, different air quality indica-
tors. For example, adverse human health effects of
particulate matter are primarily associated with the
respirable particle fraction;27 whereas household soil-
ing is a function of total suspended particulates, espe-
cially coarse particles. It is not enough, therefore, to
simply provide a single measure of particulate matter
air quality. Even for pollutants for which particle size
and other characteristics are not an issue, different air
quality indicators are needed which reflect different
periods of cumulative exposure (i.e., "averaging peri-
ods"). For example, 3-month growing season averages
are needed to estimate effects of ozone on yields of
some agricultural crops, whereas adverse human health
effect estimates require ozone concentration profiles
based on a variety of short-term averaging periods.28
Fortunately, in responding to the need for scien-
tifically valid and reliable estimation of air quality
changes, air quality modelers and researchers have
developed a number of highly sophisticated atmo-
spheric dispersion and transformation models. These
models have been employed for years supporting the
development of overall federal clean air programs,
national assessment studies, State Implementation
Plans (SIPs), and individual air toxic source risk as-
sessments. Some of these models, however, require
massive amounts of computing power. For example,
completing the 160 runs of the Regional Acid Depo-
sition Model (RADM) required for the present study
required approximately 1,080 hours of CPU time on a
Cray-YMP supercomputer at EPA's Bay City
Supercomputing Center.
Given the resource-intensity of many state-of-the-
art models, the Project Team was forced to make dif-
ficult choices regarding where to invest the limited
27 Particles with an aerometric diameter of less than or equal to 10 microns.
28 For example, ozone concentration-response data exists for effects associated with 1-hour, 2.5-hour, and 6.6-hour exposures.
_
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
resources available for air quality modeling. With a
mandate to analyze all of the key pollutants affected
by historical Clean Air Act programs, to estimate all
of the significant endpoints associated with those pol-
lutants, and to do so for a 20 year period covering the
entire continental U.S., it was necessary to use sim-
plified approaches for most of the pollutants to be
analyzed. In several cases related to primary emissions
—particularly sulfur dioxide (SO2), nitrogen oxides
(NO ), and carbon monoxide (CO)— simple "roll-up
model" strategies were adopted based on the expecta-
tion that changes in emissions of these pollutants
would be highly correlated with subsequent changes
in air quality.29 Significant pollutants involving sec-
ondary atmospheric formation, nonlinear formation
mechanisms, and/or long-range transport were ana-
lyzed using the best air quality model which was af-
fordable given time and resource limitations. These
models, discussed in detail in Appendix C, included
the Ozone Isopleth Plotting with Optional Mechanism-
IV (OZIPM4) model for urban ozone; various forms
of the above-referenced RADM model for background
ozone, acid deposition, sulfate, nitrate, and visibility
effects in the eastern U.S.; and the SJVAQS/AUSPEX
Regional Modeling Adaptation Project (SARMAP)
Air Quality Model (SAQM) for rural ozone in Cali-
fornia agricultural areas. In addition, a linear scaling
approach was developed and implemented to estimate
visibility changes in large southwestern U.S. urban
areas.
By adopting simplified approaches for some pol-
lutants, the air quality modeling step adds to the over-
all uncertainties and limitations of the present analy-
sis. The limited expanse and density of the U.S. air
quality monitoring network and the limited coverage
by available air quality models of major geographic
areas30 further constrain the achievable scope of the
present study. Under these circumstances, it is impor-
tant to remember the extent and significance of gaps
in geographic coverage for key pollutants when con-
sidering the overall results of this analysis. Key un-
certainties are summarized at the end of this chapter
in Table 5. More extensive discussion of the caveats
and uncertainties associated with the air quality model-
ing step is presented in Appendix C. In addition, in-
formation regarding the specific air quality models
used, the characteristics of the historical monitoring
data used as the basis for the control scenario pro-
files, pollutant-specific modeling strategies and as-
sumptions, references to key supporting documents,
and important caveats and uncertainties are also pre-
sented in Appendix C.
General Methodology
The general methodological approach taken in this
analysis starts with the assumption that actual histori-
cal air quality will be taken to represent the control
scenario. This may seem somewhat inconsistent with
the approach taken in earlier steps of the analysis,
which used modeled macroeconomic conditions as the
basis for estimating macroeconomic effects and emis-
sions. However, the central focus of the overall analy-
sis is to estimate the difference in cost and benefit
outcomes between the control and no-control sce-
narios. It is consistent with this central paradigm to
use actual historical air quality data as the basis for
estimating how air quality might have changed in the
absence of the Clean Air Act.
The initial step, then, for each of the five non-
lead (Pb) criteria pollutants31 was to compile com-
prehensive air quality profiles covering the entke ana-
lytical period from 1970 to 1990. The source for these
data was EPA's Aerometric Information Retrieval
System (AIRS), which is a publicly accessible data-
base of historical air quality data. The vast number of
air quality observations occurring over this twenty year
period from the thousands of monitors in the U.S. in-
dicates the need to represent these observations by
statistical distributions. As documented in detail in
the supporting documents covering SO2, NOx, CO, and
ozone,32 both lognormal and gamma distributional
forms were tested against actual data to determine the
29 It is important to emphasize that the correlation expected is between changes in emissions and changes in air quality. Direct
correlations between the absolute emissions estimates and empirical air quality measurements used in the present analysis may not be
strong due to expected inconsistencies between the geographically local, monitor-proximate emissions densities affecting air quality
data.
30 For example, the regional oxidant models available for the present study do not cover some key Midwestern states, where
human health, agricultural crop, and other effects from ozone may be significant.
31 Lead (Pb), the sixth criteria pollutant, is analyzed separately. The ability to correlate emissions directly with blood lead levels
obviates the need for using air quality modeling as an intermediate step toward estimation of exposure.
32 See SAISO2, NOx, and CO Report (1994) and SAI Ozone Report (1995).
20
-------
Chapter 4: Air Quality
form which provided the best fit to the historical data.33
Based on these tests, one or the other statistical distri-
bution was adopted for the air quality profiles devel-
oped for each pollutant. In addition to reducing the
air quality data to a manageable form, this approach
facilitated transformations of air quality profiles from
one averaging period basis to another.
Once the control scenario profiles based on his-
torical data were developed, no-control scenarios were
derived based on the results of the various air quality
modeling efforts. Again, the focus of the overall analy-
sis is to isolate the difference in outcomes between
the control and no-control scenarios. The no-control
scenario air quality profiles were therefore derived by
adjusting the control scenario profiles upward (or
downward) based on an appropriate measure of the
difference in modeled air quality outcomes. To illus-
trate this approach, consider a simplified example
where the modeled concentration of Pollutant A un-
der the no-control scenario is 0.12 ppm, compared to
a modeled concentration under the control scenario
of 0.10 ppm. An appropriate measure of the differ-
ence between these outcomes, whether it is the 0.02
ppm difference in concentration or the 20 percent per-
centage differential, is then used to ratchet up the con-
trol case profile to derive the no-control case profile.
Generally, the modeled differential is applied across
the entire control case profile to derive the no-control
case profile. As described below in the individual sec-
tions covering particulate matter and ozone, however,
more refined approaches are used where necessary to
take account of differential outcomes for component
species (i.e., particulate matter), long-range transport,
and background levels of pollutants.
Sample Results
The results of the air quality modeling effort in-
clude a vast array of monitor-specific air quality pro-
files for particulate matter (PM10 and TSP),34 SO2,
NO2, NO, CO, and ozone; RADM grid cell-based esti-
mates of sulfur and nitrogen deposition; and estimates
of visibility degradation for eastern U.S. RADM grid
cells and southwestern U.S. urban areas. All of these
data were transferred to the effects modelers for use in
configuring the human health, welfare, and ecosystem
physical effects models. Given the massive quantity
and intermediate nature of the air quality data, they
are not exhaustively reported herein.35 To provide the
reader with some sense of the magnitude of the differ-
ence in modeled air quality conditions under the con-
trol and no-control scenarios, some illustrative results
for 1990 are presented in this chapter and in Appen-
dix C. In addition, maps depicting absolute levels of
control and no-control scenario acid deposition and
visibility are presented to avoid potential confusion
which might arise through examination of percent
change maps alone.36
Carbon Monoxide
Figure 8 provides an illustrative comparison of
1990 control versus no-control scenario CO concen-
trations, expressed as a frequency distribution of the
ratios of 1990 control to no-control scenario 95th per-
centile 1-hour average concentrations at individual CO
monitors. Consistent with the emission changes un-
derlying these air quality results, CO concentrations
under the control scenario tend to be about half those.
projected under the no-control scenario, with most
individual monitor ratios ranging from about 0.40 to
0.60 percent, and a few with ratios in the 0.60 to 0.80
range.
Figure 8. Frequency Distribution of Estimated Ratios for
1990 Control to No-control Scenario 95th Percentile 1-
Hour Average CO Concentrations, by Monitor.
300 , - ,
1200
100
0 L
0.05 0.25 0.45 0.65 0.85 1.05 1.25
Ratio of CAA:No-CAA 95th Percentile 1-Hour Average
33 The statistical tests used to determine goodness of fit are described in the SAI reports.
34 PM data are reported as county-wide values for counties with PM monitors and a sufficient number of monitor observations.
35 The actual air quality profiles, however, are available on disk from EPA. See Appendix C for further information.
36 Large percentage changes can result from even modest absolute changes when they occur in areas with good initial (e.g.,
control scenario) air quality. Considering percentage changes alone might create false impressions regarding absolute changes in air
quality in some areas. For example, Appendix C discusses in detail two such cases: the Upper Great Lakes and Florida-Southeast
Atlantic Coast areas, which show high percentage changes in sulfur deposition and visibility.
21
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
In considering these results, it is important to note
that CO is essentially a "hot spot" pollutant, meaning
that higher concentrations tend to be observed in lo-
calized areas of relatively high emissions. Examples
of such areas include major highways, major inter-
sections, and tunnels. Since CO monitors tend to be
located in order to monitor the high CO concentra-
tions observed in such locations, one might suspect
that using state-wide emissions changes to scale air
quality concentration estimates at strategically located
monitors might create some bias in the estimates.
However, the vast majority of ambient CO is contrib-
uted from on-highway vehicles. In addition, the vast
majority of the change in CO emissions between the
control and no-control scenario occurs due to catalyst
controls on highway vehicles. Since CO hot spots re-
sult primarily from highway vehicles emissions, con-
trolling such vehicles would mean CO concentrations
would be commensurately lowered at CO monitors.
While variability in monitor location relative to ac-
tual hot spots and other factors raise legitimate con-
cerns about assuming ambient concentrations are cor-
related with emission changes at any given monitor,
the Project Team believes that the results observed
provide a reasonable characterization of the aggregate
change in ambient CO concentrations between the two
scenarios.
Sulfur Dioxide
As for CO, no-control scenario SO2 concentra-
tions were derived by scaling control scenario air qual-
ity profiles based on the difference in emissions pre-
dicted under the two scenarios. Unlike CO, SO2 is
predominantly emitted from industrial and utility
sources. This means that emissions, and the changes
in emissions predicted under the two scenarios, will
tend to be concentrated in the vicinity of major point
sources. Again, while state-wide emissions changes
are used to scale SO2 concentrations between the sce-
narios, these state-wide emission changes reflect the
controls placed on these individual point sources.
Therefore, the Project Team again considers the dis-
tribution of control to no-control ratios to be a rea-
sonable characterization of the aggregate results de-
spite the uncertainties associated with estimation of
changes at individual monitors.
Figure 9 provides a histogram of the predicted
control to no-control ratios for SO2 which is similar
to the one presented for CO. The results indicate that,
on an overall basis, SO2 concentrations were reduced
by about one-third. The histogram also shows a much
wider distribution of control to no-control ratios for
individual monitors than was projected for CO. This
result reflects the greater state to state variability in
SO2 emission changes projected in this analysis. This
greater state to state variability in turn is a function of
the variable responses of SO2 point sources to histori-
cal C control requirements.37 This source-specific vari-
ability was not observed for CO because controls were
applied relatively uniformly on highway vehicles.
Figure 9. Frequency Distribution of Estimated Ratios-for
1990 Control to No-control Scenario 95th Percentile 1-
Hour Average SO, Concentrations, by Monitor.
300
0.05 0.25 0.45 0.65 0.85 1.05 1.25
Ratio of CAA:No-CAA 95th Percentile 1-Hour Average
Nitrogen Dioxide
Results for NO2 are presented in Figure 10. These
results are similar to the results observed for CO, and
for a similar reason: the vast majority of change in
NO2 emissions between the two scenarios is related
to control of highway vehicle emissions. While
baseline emissions of NO2 from stationary sources may
be significant, these sources were subject to minimal
controls during the historical period of this analysis.
On an aggregated basis, overall NO2 concentrations
are estimated to be roughly one-third lower under the
control scenario than under the no-control scenario.
37 Figure 9 indicates that six monitors were projected to have higher SO2 concentrations for 1990 under the control scenario than
under the no-control scenario. All six of these monitors are located in Georgia, a state for which higher 1990 utility SO. emissions are
projected in the control scenario due to increased use of higher-sulfur coal. The projected increase in overall Georgia utility consump-
tion of higher sulfur coal under the control case is a result of increased competition for the low-sulfur southern Appalachian coal
projected to occur under the control scenario.
22
-------
Chapter 4: Air Quality
Figure 10. Frequency Distribution of Estimated Ratios for
1990 Control to No-control Scenario 95th Percentile 1-
Hour Average NO2 Concentrations, by Monitor.
300
0.05 0.25 0.45 0.65 0.85 1.05 1.25
Ratio of CAA:No-CAA 95th Percentile 1-Hour Average
Particulate Matter
An indication of the difference in outcomes for
paniculate matter between the two scenarios is pro-
vided by Figure 11. This graph shows the distribution
of control to no-control ratios for annual mean TSP in
1990 for those counties which both had particulate
monitors and a sufficient number of observations from
those monitors.38 While the distribution of results is
relatively wide, reflecting significant county to county
variability in ambient concentration, on a national
aggregate basis particulate matter concentrations un-
Figure 11. Frequency Distribution of Estimated Ratios for
1990 Control to No-control Annual Mean TSP Concentra-
tions, by Monitored County.
50
0.00 0.20 0.40 0.60 0.80 1.00
Ratio of CAA:No-CAA Annual Mean TSP (interval midpoint)
der the control scenario were just over half the level
projected under the no-control scenario. The signifi-
cant county to county variability observed in this case
reflects point source-specific controls on particulate
matter precursors, especially SO2, and the effects of
long-range transport and transformation.
Ozone
Urban Ozone
Figure 12 presents a summary of the results of the
1990 OZIPM4 ozone results for all 147 of the mod-
eled urban areas. In this case, the graph depicts the
distribution of ratios of peak ozone concentrations
estimated for the control and no-control scenarios.
While the vast majority of simulated peak ozone con-
centration ratios fall below 1.00, eight urban areas
show lower simulated peak ozone for the no-control
scenario than for the control scenario. For these eight
urban areas, emissions of precursors were higher un-
der the no-control scenario; however, the high pro-
portion of ambient NOx compared to ambient non-
methane organic compounds (NMOCs) in these areas
results in a decrease in net ozone production in the
vicinity of the monitor when NO emissions increase.39
Figure 12. Distribution of Estimated Ratios for 1990
Control to No-control OZIPM4 Simulated 1-Hour Peak
Ozone Concentrations, by Urban Area.
30
20
10
0.00 0.20 0.40 0.60 0.80 1.00
Ratk>ofCAA:No-CAA Peak Ozone (intervalmidpoint)
1.20
38 Given the relative importance of particulate matter changes to the bottom line estimate of CAA benefits, and the fact that a
substantial portion of the population lives in unmonitored counties, a methodology was developed to allow estimation of particulate
matter benefits for these unmonitored counties. This methodology was based on the use of regional air quality modeling to interpolate
between monitored counties. It is summarized in Appendix C and described in detail in the SAIPM Interpolation Report (1996).
39 Over an unbounded geographic area, NOx reductions generally decrease net ozone production.
23
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Rural Ozone
Figures 13 and 14 present frequency distributions
for control to no-control ratios of average ozone-sea-
'son daytime ozone concentrations at rural monitors
as simulated by SAQM and RADM, respectively.
Figure 13. Distribution of Estimated Ratios for 1990
Control to No-control SAQM Simulated Daytime Average
Ozone Concentrations, by SAQM Monitor.
10
8
6
0.00 0.20 0.40 0.60 0.80 1.00 1.20
Ratio ofCAA:No-CAA Ozone-Season Day time Average Ozone (interval midpoint)
Both the RADM and SAQM results indicate rela-
tively little overall change in rural ozone concentra-
tions. This is primarily because reductions in ozone
precursor emissions were concentrated in populated
areas.
Figure 14. Distribution of Estimated Ratios for 1990
Control to No-control RADM Simulated Daytime Average
Ozone Concentrations, by RADM Grid Cell.
200
ISO
100
50
0.00 0.20 0.40 0.60 0.80 1.00 1.20
Ratio ofCAAiNo-CAA Ozone-Season Day time Average Ozone (mtervalmidpoint)
Acid Deposition
Figure 15 is a contour map showing the estimated
percent increase in sulfur deposition under the no-con-
trol scenario relative to the control scenario for 1990.
Figure 16 provides comparable information for nitro-
gen deposition.
Figure 15. RADM-Predicted Percent Increase in Total
Sulfur Deposition (Wet + Dry) Under the No-control
Scenario.
These results show that acid deposition rates in-
crease significantly under the no-control scenario,
particularly in the Atlantic Coast area and in the vi-
cinity of states for which relatively large increases in
emissions are projected under the no-control scenario
(i.e., Kentucky, Florida, Michigan, Mississippi, Con-
necticut, and Florida).
In the areas associated with large increases in sul-
fur dioxide emissions, rates of sulfur deposition in-
crease to greater than or equal to 40 percent. The high
proportional increase in these areas reflects both the
significant increase in acid deposition precursor emis-
sions in upwind areas and the relatively low deposi-
tion rates observed under the control scenario.40
Along the Atlantic Coast, 1990 nitrogen deposi-
tion rates increase by greater than or equal to 25 per-
cent under the no-control scenario. This is primarily
due to the significant increase in mobile source nitro-
gen oxide emissions along the major urban corridors
of the eastern seaboard.
40 Even small changes in absolute deposition can yield large percentage changes when initial absolute deposition is low. See
Appendix C for further discussion of this issue.
24
-------
Chapter 4: Air Quality
Figure 16. RADM-Predicted Percent Increase in Total
Nitrogen Deposition (Wet + Dry) Under the No-
control Scenario.
Visibility
The difference in modeled 1990 control and
no-control scenario visibility conditions projected by
the RADM/EM for the eastern U.S. is depicted by the
contour map presented in Figure 17. This figure shows
the increase in modeled annual average visibility deg-
radation, in DeciView41 terms, for 1990 when mov-
Figure 17. RADM-Predicted Percent Increase in
Visibility Degradation, Expressed in DeciViews, for
Poor Visibility Conditions (90th Percentile) Under the
No-control Scenario.
ing from the control to the no-control scenario. Since
the DeciView metric is based on perceptible changes
in visibility, these results indicate noticeable deterio-
ration of visibility in the eastern U.S. underthe no-
control scenario.
Visibility changes in 30 southwestern U.S. urban
areas were also estimated using emissions scaling tech-
niques. This analysis also found significant, percep-
tible changes in visibility between the two scenarios.
Details of this analysis, including the specific out-
comes for the 30 individual urban areas, are presented
in Appendix C.
Uncertainty in the Air Quality
Estimates
Uncertainty prevades the projected changes in air
quality presented in this study. These uncertainties
arise due to potential inaccuracies in the emissions
inventories used as air quality modeling inputs and
due to potential errors in the structure and parameter-
ization of the air quality models themselves. In addi-
tion, an important limitation of the present study is
the lack of available data and/or modeling results for
some pollutants in some regions of the country (e.g.,
visibility changes in western U.S. Class I areas such
as the Grand Canyon). The inability to provide com-
prehensive estimates of changes in air quality due to
the Clean Air Act creates a downward bias in the
monetary benefit estimates.
The most important specific sources of uncertainty
are presented in Table 5, and are described further in
Appendix C. While the list of potential errors pre-
sented in Table 5 is not exhaustive, it incorporates the
uncertainties with the greatest potential for contribut-
ing to error in the monetary benefit estimates. Over-
all, the uncertainties in the estimated change in air
quality are considered small relative to uncertainties
contributed by other components of the analysis.
41 The DeciView Haze Index (dV) is a relatively new visibility indicator aimed at measuring visibility changes in terms of human
perception. It is described in detail in Appendix C.
25 ! ~~~ '
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table 5. Key Uncertainties Associate^ wither.Quality* Modeling.
Potential S otirce o f Error " , ' J
V -.*v« f
- • -
Use of OZIPM4 model, wjjich does not , ;
capture long-range and night-time transport of
ozone. Use of aregipnal oxidant model, seich *
as UAM-V, would mitigate errors associated
with neglecting transport.
"" % ' "" s ^ -' < •>'
Use of early biogenic em|ssloS;estfeQtesia-«-;-!
RADM to estimate rural ozone changes, in the
eastern. 31 states. " ' < '""
Use of proxy pollutants to scale up some '
particulate species in some areas", tincertainty
is created to the extent species 'of concern are
not perfectly correlated with the proxy , ', , ,
pollutants. ,„,;-
, ,, , ,-",-
Use of state- wide average emission teduetionf*"
to configure air quality models: In'sorne*1 """^
cases, control programs may haveljeen"*"'^'^'^''
targeted to problem areas, so using sjM§3yj$$,
averages would miss relatively large '
reductions in populated areas. , '--"',
Exclusion of visibility benefits in Class I - ,
areas in the Southwestern U.S.
Potential Bias
- - ,,,,<-»,<.
^i -•/• ^^
"; Significance Rclativje to, Key
ilttc^rtaijaties ia Overall Monetary
'" "\ *' v' j^eneJIt Estimate - -. ..
'-'-#-" "'-'- '""; ,
JSlgni^aijt^but probably not major. - , ,„
'Overall av^r|ge;,oz'one response of U% to
Mb* arid V-OC reductions of '-
approximately, 30% and 45%, -
respectively. Even if dsuke response -
doubled -to 30% , estimate of monetized
benefits of C°AA will not change very
muefc. Sigaiicant benefits X)f ozone - ,,,,
?reductioa, however, could not'b'e"
-monetized, , - ' ";,",*',""',„
,Pro-bab]y minor. Errors «eestitnaWd"fp^
-be-w&nis - 15% to *2$ % of the caon'"! ,
,-;, , -' ' '<•-'-,'->-* ' '
f< pjfeQictio'nSx'
Potentially, significant, Oiven tlie relative
' iiBp'ortalJice'of the estimated chal!'ge$''#i'' -'-
fliief pMicte'eoaceritrations to the ' "''
monetized' benefit estimate, any ' " '! ' '
unce'rtairtty associated with,Sne particles" ,'
isjp-p^entlaily significant,. Ho,we>fer,vl;he ,Si,-
jpotential error is mitigated to some extent
since proxy-pollutant measures -ar-e-app lied
Probably, -minor, , , \ ,
..VV, - -X ':'
-------
Chapter 4: Air Quality
Uncertaigttgs
,
Potential Bias
in Estimate of
Significance Relatiye to Key
Lacjc pf niodel'coverage in western,!'
fotapiji deposition. >:*r'v"'' —
.IJnderestimie.
- j ^^
beiiefitiof reduced acid deposition"!^ the,
17' western statfii? ,^ftuld probably h<3t , ,
slgnifieaatly alter Jbjj ,esjtMate of ,' - • •
monetized benefits. - y<- ; - x ,
,»., ,..,,,,, , _^,,,
,U|<|,9JF spatially .aft'd^eo'
---,-, , , • ,, , --„
'sFote,ntiaily si'gpiffeant Any-effect.^hidi1
"might influence the direction of long-
meteorolo'giealdata resallstte i
:|^coatst-fortesi|)era|ttre effects, oa YOCs^and,-/-
''effect otl,c^a8?ed meteorology aroiiRd major; ,
poiat sources. • -; --«„/, , '^t— - -- , ---
" sv, J v , (, . ^
-benefits of the CA A,
organic,a^p|p,|;^ij.ceatratioiis fixed ojiai|s the
>;effeet ol cfiangesjii this constituent '*
"v' '
"s/.,,' • • < •. ,- . ,
Probably minor, because (a) nitrates were'
,ako^hel'
the dominant species in the' e^efn ILS'.r <•"•
t>y using emissions scaling to estimate <
cl^iiges'iai 'organic 'fe'roSdis since a »•
signifieaht'f'raction of-'0rganie-aero|o1s are
-caused by "bipgenie ga&^phase TOfe" ' ' "" '
-emissions which do not change between '
•thescoiarios: ---, .^,, , „ - „ , ' ""/
Probabiyjininor. MisseVpoteatiarhamau" '
health, welfare, and ecotogical'.benerits^of
te'ducing rurlPozoae-itx agricultural aad ""
other' rural, ^^^mt,^/&tf oao«e-chauges '
are'lllcely,;to be'ifo'all gij^i IJoiited """ *^-~
precurs'of reductiofis'ltj'rftlcalare'afs, , ,
RADM co'iitfoltno-control ratios -are,in,f-,; ,
fact, relative^-, s,raalL --'--- •-• --. v,-*-,
,Unkapwn,
.concentration.:? :'; ; ;"-' ->.-.-.
Probably minor^particjilarly since relative*
chang'es 'in ozone cb-hceritration between • '.-
the scoiarios'were small/ '• - , •
27
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
28
-------
5
Physical Effects
Human Health and Welfare
Effects Modeling Approach
This chapter identifies and, where possible, esti-
mates the principal health and welfare benefits en-
joyed by Americans due to improved air quality re-
sulting from the CAA. Health benefits have resulted
from avoidance of air pollution-related-health effects,
such as premature mortality, respiratory illness, and
heart disease. Welfare benefits accrued where im-
proved air quality averted damage to measurable re-
sources, including agricultural production and visibil-
ity. The analysis of physical effects required a combi-
nation of three components: air quality, population,
and health or welfare effects. As structured in this
study, the 3-step process involved (1) estimating
changes in air quality between the control and no-con-
trol scenarios, (2) estimating the human populations
and natural resources exposed to these changed air
quality conditions, and (3) applying a series of con-
centration-response equations which translated
changes in air quality to changes in physical health
and welfare outcomes for the affected populations.
Air Quality
The Project Team first estimated changes in con-
centrations of criteria air pollutants between the con-
trol scenario, which at this step was based on histori-
cal air quality, and the no-control scenario. Air qual-
ity improvements resulting from the Act were evalu-
ated in terms of both their temporal distribution from
1970 to 1990 and their spatial distribution across the
48 conterminous United States. Generally, air pollu-
tion monitoring data provided baseline ambient air
quality levels for the control scenario. Air quality
modeling was used to generate estimated ambient con-
centrations for the no-control scenario. A variety of
modeling techniques was applied, depending on the
pollutant modeled. These modeling approaches and
results are summarized in Chapter 4 and presented in
detail in Appendix C.
Population
Health and some welfare benefits resulting from
air quality improvements were distributed to individu-
als in proportion to the reduction in exposure. Pre-
dicting individual exposures, then, was a necessary
step in estimating health effects. Evaluating exposure
changes for the present analysis required not only an
understanding of where air quality improved as a re-
sult of the CAA, but also how many individuals were
affected by varying levels of air quality improvements.
Thus, a critical component of the benefits analysis
required that the distribution of the U.S. population
nationwide be established.
Three years of U.S. Census data were used to rep-
resent the geographical distribution of U.S. residents:
1970, 1980, and 1990. Population data was supplied
at the census block group level, with approximately
290,000 block groups nationwide. Allocating air qual-
ity improvements to the population for the other tar-
get years of this study -1975 and 1985 - necessitated
interpolation of the three years of population data.
Linear interpolation was accomplished for each block
group in order to maintain the variability in growth
rates throughout the country.
Health and Welfare Effects
Benefits attributable to the CAA were measured
in terms of the avoided incidence of physical health
effects and measured welfare effects. To quantify such
benefits, it was necessary to identify concentration-
response relationships for each effect being consid-
ered. As detailed in Appendix D, such relationships
were derived from the published science literature. In
the case of health effects, concentration-response func-
tions combined the air quality improvement and popu-
lation distribution data with estimates of the number
of fewer individuals that suffer an adverse health ef-
fect per unit change in air quality. By evaluating each
concentration-response function for every monitored
location throughout the country, and aggregating the
29
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
resulting incidence estimates, it was possible to gen-
erate national estimates of incidence under the con-
trol and no-control scenarios.
In performing this step of the analysis, the Project
Team discovered that it was impossible to estimate
all of the health and welfare benefits which have re-
sulted from the Clean Air Act. While scientific infor-
mation was available to support estimation of some
effects, many other important health and welfare ef-
fects could not be estimated. Furthermore, even though
some physical effects could be quantified, the state of
the science did not support assessment of the economic
value of all of these effects. Table 6 shows the health
effects for which quantitative analysis was prepared,
as well as some of the health effects which could not
be quantified in the analysis. Table 7 provides similar
information for selected welfare effects.
While the 3-step analytical process described
above was applied for most pollutants, health effects
for lead were evaluated using a different methodol-
ogy. Gasoline as a source of lead exposure was ad-
dressed separately from conventional point sources.
Instead of using ambient concentrations of lead re-
sulting from use of leaded gasoline, the concentra-
tion-response functions linked changes in lead releases
directly to changes in the population's mean blood
lead level. The amount of leaded gasoline used each
year was directly related to mean blood lead levels
using a relationship described in the 1985 Lead Regu-
latory Impact Analysis (U.S. EPA, 1985). Health ef-
fects resulting from exposure to point sources of at-
mospheric lead, such as industrial facilities, were con-
sidered using the air concentration distributions mod-
eled around these point sources. Concentration-re-
sponse functions were then used to estimate changes
in blood lead levels in nearby populations.
Most welfare effects were analyzed using the same
basic 3-step process used to analyze health effects,
with one major difference in the concentration-re-
sponse functions used. Instead of quantifying the re-
lationship between a given air quality change and the
number of cases of a physical outcome, welfare ef-
fects were measured in terms of the avoided resource
losses. An example is the reduction in agricultural crop
losses resulting from lower ambient ozone concentra-
tions under the control scenario. These agricultural
benefits were measured in terms of net economic sur-
plus.
Another important welfare effect is the benefit
accruing from improvements in visibility under the
control scenario. Again, a slightly different method-
ological approach was used to evaluate visibility im-
provements. Visibility changes were a direct output
of the models used to estimate changes in air qual-
ity.42 The models provided estimates of changes in
light extinction, which were then translated mathemati-
cally into various specific measures of perceived vis-
ibility change.43 These visibility change measures were
then combined with population data to estimate the
economic value of the visibility changes. Other wel-
fare effects quantified in terms of avoided resource
losses include household soiling damage by PM,0 and
decreased worker productivity due to ozone exposure.
The results of the welfare effects analysis are found
in Chapter 6 and in Appendices D and F.
Because of a lack of available concentration-re-
sponse functions (or a lack of information concerning
affected populations), ecological effects were not
quantified for this analysis. However, Appendix E
provides discussion of many of the important ecologi-
cal benefits which may have accrued due to historical
implementation of the CAA.
Key Analytical Assumptions
Several important analytical assumptions affect
the confidence which can be placed in the results of
the physical effects analysis. The most important of
these assumptions relate to (a) mapping of potentially
exposed populations to the ambient air quality moni-
toring network, (b) choosing among competing scien-
tific studies in developing quantitative estimates of
physical effects, (c) quantifying the contribution to
analytical uncertainty of within-study variances in
effects estimates and, perhaps most important in the
context of the present study, (d) estimating particu-
late matter-related mortality based on the currently
available scientific literature.
Because these resultant uncertainties were caused
by the inadequacy of currently available scientific in-
formation, there is no compelling reason to believe
42 These models, and the specific visibility changes estimated by these models, are described in summary fashion in the previous
chapter and are discussed in detail in Appendix C.
43 These visibility measures are described in Appendix C.
30
-------
ChapterS: Physical Effects
Tablefe-Xfamaa-'HsaltH-Effects-of''CriteriaPoHatanis:. >
Quantified Health Effects
Unquantified Healtii Effects
Ozone s^
, *
ponsiveness to stJnlpli '
Isuraunologic dianges
-, Chrtmio' respiratory disease^,,
'
Hosgii ai adjftissiotts \
•, Easergencyrooni visits "* '' ;
• /Asthma attacks- , , "*•:
Changes in ptttooaaty ftmctioji
Oi'iibnic Sinasafe'&ilTay Beyer
Inflartimadon in the lung,
- - ,
, I-f6»ctton,of -broncfeitfe - - •
laSatQHKitiQn. JB the Juag
.,
AJJ resteicteii apflyity days"
Days *pfvt" **
•$•&,'-, ff' "" •• J1 ^v^ ^
'Decreased pataoaaty ftmctiba
-Respiratory syiitp,tams to ntoa-
and;
Bealtfe e
NmtrotfehavSo'raJ | toeiott
,,,- t,
-swia! .......
-31
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
'J ^^Ji^"-^^•?' iv^VP^"* t ''"• y !* •""•Sjsis'-w > " ''*•••* •• •• v /'*"' -v' ••> v *jfi'*"j--t j ff ft* /"],
Table 7. Selected Welfare Effecls;^(^ti^Jl»|aigf v^,"-^ •—.T-T'*'*"^- *^>-;f.; _^/,;,v
Pollutant
Ozone
Particulate Matter/
TSP/Sul fates
Nitrogen Oxides
Sulfur Dioxide
lj,,«V",, ^ /*? ~';;,yA,, st
,„", •" ,"f,.' --., • ," ' ' " '
Changesyn ,crj>p.yfeJds (foXXcrgiJS)
Decreased-WQMterpWdttctivity //'"^ ,
• - , ';s-';'il-;j^»~,; ', „;,:•?>
.,-• - ' ' f-Kf/X,-' ' ' '
House^gld'js oilffljg'^ ,' , , - '" ~""^! ~%
:'visibint^:r-"v5sTi'-' ^\- 1%>'*S:
;" : """,< v^f^/i- '••'«•«
Visibility ' ' '"5-'%.—-,,; ; ' ' ~
* *~ '*<>*, •• ' * * > <^ >"< '* "'
-• -'- .- , * - v'-.J5?'-";'"i, ,
•^4sv4\~,;i~- : ': ?-:>'!,:>
^ " ' b^ ^^ ,"4.^,^^ -.•• ' ^vV^^Pj
x " J ' •• '-• ''"* ' •>? '
Visibility "'"':-'':;, , ,. ''-I''1'',
-^v^/^^sSj ,^ ^ ,''"c^
-,'''' K^i 1" :,'c;>^! '' ;
,;,;,9s -*->>,,, s v'sX-sA',5-- ;
-; ' &S -f ,,-,/,„,, - , . ,; "!%;"' '--
^ f,f -. us* "^ v -. ' ^^; ' ^ 1"1"1'1' J v'yj-- 'V'^' ^ 'v/
/JUnquanWffed Welfare* 5ffi%<|S • ' ' -/
C hanges', m/^atlier »oro p; yields
; Materials, ^damage ' i/:-- -' ';
'JEffe'ets o« forests ' '^;"'.y/, ',.,..;.;,/ I
',Effedts'csi,w5MIire ' "' "1>;V:v'f" 7 ::L '-
O^KK^^6saial$ damage,-; li>,.'/, ',?*, 'S"?^
/;Effects;ci''w! Wife , • ' ,. . - ' " " r,
h^w"-- /v%j^«*^y-?'#-!;>;£v-.>y>^5 •' ,
-;Bec»i:losses dae to acid depjosjttojiv"
, ,^^^.> ^,' ' ' "• f *"*" Jf ,/ v94("s '•{,£ >
^Effe'ctsW SsiSifS-jdttito'aoiaie ^ ;;-;;'^
^pp^!?i ;- -~~ -, ' ''' ' ' "--' ::'''' - ' '; - "•'", ' -:•
"Bfficf^Qn'^rests1 -„- .- ' '-'"•>'' - '"!"
•Oi<^ losses du^to!aCici;;dep4sitfoTi ',!"; v
'^Materials daHiage1dae"to"a1da?de^ositiori" "
rij&e&on, fisheries due to' aeile^ ; !'-
-- >'' v^X*i>>V:^J ^^ /'^/^" - r .,,>'"*'
dftpositj|0i>^,;,", ,"" "•'.,-!,', •"," -; - - Y- »*
Effects' oti"forests-",> '--'' ;,•'.">"'"
•f, »v ,g ,,,f«^,-. ,^ ,/f~., -r r
that the results of the present analysis are biased in a
particular direction. Some significant uncertainties,
however, may have arisen from interpretation of model
results, underlying data, and supporting scientific stud-
ies. These assumptions and uncertainties are charac-
terized in this report to allow the reader to understand
the degree of uncertainty and the potential for mises-
timation of results. In addition, the overall results are
presented in ranges to reflect the aggregate effect of
uncertainty in key variables. A quantitative assessment
of some of the uncertainties in the present study is
presented in Chapter 7. In addition, the key uncertain-
ties in the physical effects modeling step of this analy-
sis are summarized in Table 12 at the end of this chap-
ter. The remainder of this section discusses each of
the four critical modeling procedures and associated
assumptions.
Mapping Populations to Monitors
The Project Team's method of calculating ben-
efits of air pollution reductions required a correlation
of air quality data changes to exposed populations.
For pollutants with monitor-level data (i.e., SO2, O3,
NO2, CO), it was assumed that all individuals were
exposed to air quality changes estimated at the near-
est monitor. For PM]0, historical air quality data were
available at the county level. All individuals residing
in a county were assumed to be exposed to that
county's PM air quality.44
Many counties did not contain particulate matter
air quality monitors or did not have a sufficient num-
ber of monitor observations to provide reliable esti-
mates of air quality. For those counties, the Project
Team conducted additional analyses to estimate PM10
air quality changes during the study period. For coun-
ties in the eastern 31 states, the grid cell-specific sul-
fate particle concentrations predicted by the RADM
model were used to provide a scaled interpolation
between monitored counties.45 For counties outside
the RADM domain, an alternative method based on
state-wide average concentrations was used. With this
supplemental analysis, estimates were developed of
the health effects of the CAA on almost the entire
continental U.S. population.46 Compliance costs in-
*• In some counties and in the early years of the study period, particulate matter was monitored as TSP rather than as PM,., In,.th<5se
cases, PMIO was estimated by applying TSP:PM10 ratios derived from historical data. This methodology is described in Appendix C.
45 The specific methodology is described in detail in Appendix C.
* While this modeling approach captures the vast majority of the U.S. population, it does not model exposure for everyone. To
improve computational efficiency, those grid cells with populations less than 500 were not modeled; thus, the analysis covered
somewhat more than 97 percent of the population.
— 32
-------
ChapterS: Physical Effects
curred in Alaska and Hawaii were included in this
study, but the benefits of historical air pollution re-
ductions were not. In addition, the CAA yielded ben-
efits to Mexico and Canada that were not captured in
this study.
Air quality monitors are more likely to be found
in high pollution areas rather than low-pollution ar-
eas. Consequently, mapping population to the nearest
monitor regardless of the distance to that monitor al-
most certainly results in an overstatement of health
impacts due to air quality changes for those popula-
tions. The Project Team conducted a sensitivity analy-
sis to illustrate the importance of the "mapping to near-
est monitor" assumption. For comparison to the base
case, which modeled exposure for the 48 state popu-
lation, Table 8 presents the percentage of the total 48-
state population covered in the "50 km" sensitivity
scenario. For most pollutants in most years, 25 per-
cent or more of the population resided more than 50
km from an air quality monitor (or in a county with-
out PM[0 monitors). Estimated health benefits are ap-
proximately linear to population covered — that is, if
the population modeled for a pollutant in a given year
in the sensitivity analysis is 25 percent smaller than
the corresponding population modeled in the base case,
then estimated health benefits are reduced by roughly
25 percent in the sensitivity case. This sensitivity
analysis demonstrates that limiting the benefits analy-
sis to reflect only those living within 50 km of a moni-
tor or within a PM-monitored county would lead to a
substantial underestimate of the historical benefits of
the CAA. Since these alternative results may have led
to severely misleading comparisons of the costs and
benefits of the Act, the Project Team decided to adopt
the full 48-state population estimate as the central case
for this analysis despite the greater uncertainties and
potential biases associated with estimating exposures
from distant monitoring sites.
»Table 8> Percent of Population 74%
:62%
73% 68%^
*73% 68%
70S,
Choice of Study
The Project Team relied on the most recent avail-
able, published scientific literature to ascertain the
relationship between air pollution and human health
and welfare effects. The choice of studies, and the
uncertainties underlying those studies, also created
uncertainties in the results. For example, to the extent
the published literature may collectively overstate the
effects of pollution, EPA's analysis will overstate the
effects of the CAA. Such outcomes may occur be-
cause scientific research which fails to find signifi-
cant relationships is less likely to be published than
research with positive results. On the other hand, his-
tory has shown that it is highly likely that scientific
understanding of the effects of air pollution will im-
prove in the future, resulting in discovery of previ-
ously unknown effects. Important examples of this
phenomenon are the substantial expected health and
welfare benefits of reductions in lead and ambient
paniculate matter, both of which have been shown in
recent studies to impose more severe effects than sci-
entists previously believed. To the extent the present
analysis misses effects of air pollution that have not
yet been subject to adequate scientific inquiry, the
analysis may understate the effects of the CAA.
For some health endpoints, the peer-reviewed sci-
entific literature provides multiple, significantly dif-
fering alternative CR functions. In fact, it is not un-
usual for two equally-reputable studies to differ by a
factor of three or four in implied health impact. The
difference in implied health effects across studies can
be considered an indication of the degree of scientific
uncertainty associated with measurement of that health
effect. Where more than one acceptable study was
available, the Project Team used CR functions from
all relevant studies to infer health effects. That is, the
health effect implied by each study is reported (see
Appendix D), and a range of reported results for a
particular health endpoint can be interpreted as a mea-
sure of the uncertainty of the estimate.
Variance Within Studies
Even where only one CR function was available
for use, the uncertainty associated with application of
that function to estimate physical outcomes can be
evaluated quantitatively. Health effects studies pro-
vided "best estimates" of the relationship between air
quality changes and health effects, and a measure of
the statistical uncertainty of the relationship. In this
analysis, the Project Team used simulation modeling
33
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
techniques to evaluate the overall uncertainty of the
results given uncertainties within individual studies,
across studies examining a given endpoint, and in the
economic valuation coefficients applied to each end-
point. The analysis estimating aggregate quantitative
uncertainty is presented in Chapter 7.
PM-Related Mortality
The most serious human health impact of air pol-
lution is an increase in incidences of premature mor-
tality. In the present study, excess premature mortal-
ity is principally related to increased exposure to lead
(Pb)47 and to paniculate matter (PM) and associated
non-Pb criteria pollutants.48 With respect to PM, a
substantial body of published health science literature
recognizes a correlation between elevated PM con-
centrations and increased mortality rates. However,
there is a diversity of opinion among scientific ex-
perts regarding the reasonableness of applying these
studies to derive quantitative estimates of premature
mortality associated with exposure to PM. While 19
of 21 members of the Science Advisory Board Clean
Air Act Scientific Advisory Committee agree that
present evidence warrants concern and implementa-
tion of a fine particle (PM^) standard to supplement
the PM,0 standard, they also point out that the causal
mechanism has not been clearly established.
For the purposes of the present study, the Project
Team has concluded that the well-established corre-
lation between exposure to elevated PM and prema-
ture mortality is sufficiently compelling to warrant an
assumption of a causal relationship and derivation of
quantitative estimates of a PM-related premature mor-
tality effect. In addition to the assumption of causal-
ity, a number of other factors contribute to uncertainty
in the quantitative estimates of PM-related mortality.49
First, although there is uncertainty regarding the shape
of the CR functions derived from the epidemiological
studies, the present analysis assumes the relationship
to be linear throughout the relevant range of expo-
sures. Second, there is significant variability among
the underlying studies which may reflect, at least in
part, location-specific differences in CR functions.
Transferring CR functions derived from one or more
specific locations to all other locations may contrib-
ute significantly to uncertainty in the effect estimate.
Third, a number of potentially significant biases and
uncertainties specifically associated with each of the
two types of PM-related mortality study further con-
tribute to uncertainty. The remainder of this section
discusses these two groups of studies and their atten-
dant uncertainties and potential biases. (See Appen-
dix D for a more complete discussion of these studies
and their associated uncertainties.)
Short-Term Exposure Studies
Many of the studies examining the relationship
between PM exposure and mortality evaluate changes
in mortality rates several days after a period of el-
evated PM concentrations. In general, significant cor-
relations have been found. These "short-term expo-
sure" or "episodic" studies are unable to address two
important issues: (1) the degree to which the observed
excess mortalities are "premature," and (2) the degree
to which daily mortality rates are correlated with long-
term exposure to elevated PM concentrations (i.e..,
exposures over many years rather than a few days).
Because the episodic mortality studies evaluate
the mortality rate impact only a few days after a high-
pollution event, it is likely that many of the "excess
mortality" cases represented individuals who were
already suffering impaired health, and for whom the
high-pollution event represented an exacerbation of
an already serious condition. Based on the episodic
studies only, however, it is unknown how many of the
victims would have otherwise lived only a few more
days or weeks, or how many would have recovered to
enjoy many years of a healthy life in the absence of
the high-pollution event. For the purpose of cost-ben-
efit analysis, it can be important to determine whether
a pollution event reduces the average lifespan by sev-
eral days or by many years. Although the episodic
mortality studies do not provide an estimate of the
expected life years lost (nor do they address the health
status of victims), some have evaluated the age of the
excess premature mortality cases, and have estimated
that 80 to 85 percent of the victims are age 65 or older.
In addition to causing short-term health problems,
air pollution (measured by elevated annual PM con-
47 Detailed information on methods, sources, and results of the Pb mortality analysis are presented in Appendix G.
48 PM concentrations are highly correlated with concentrations of other criteria pollutants. It is difficult to determine which
pollutant is the causative factor in elevated mortality rates. In this study, the Project Team has used PM as a surrogate for a mix of
criteria pollutants.
49 It should also be noted that some of the morbidity studies, most notably the PM/chronic bronchitis epidemiological studies,
involve many of the same uncertainties.
—
-------
ChapterS: Physical Effects
centrations) can cause longer-term health problems
that may lead to premature mortality. Such long-term
changes in susceptibility to premature mortality in the
future will be missed by efforts to correlate prema-
ture mortalities with near-term episodes of elevated
pollution concentrations. Consequently, excess pre-
mature mortality estimates based on the results of the
"episodic" mortality studies will underestimate the
effect of long-term elevated pollution concentrations
on mortality rates.
Long-Term Exposure Studies
The other type of PM-related mortality study in-
volves examination of the potential relationship be-
tween long-term exposure to PM and annual mortal-
ity rates. These studies are able to avoid some of the
weaknesses of the episodic studies. .In particular, by
investigating changes in annual (rather than daily)
mortality rates, the long-term studies do not predict
most cases of excess premature mortality where mor-
tality is deferred for only a few days; also, the long-
term studies are able to discern changes in mortality
rates due to long-term exposure to elevated air pollu-
tion concentrations. Additionally, the long-term ex-
posure studies are not limited to measuring mortali-
ties that occur within a few days of a high-pollution
event. Consequently, use of the results of the long-
term studies is likely to result in a more complete as-
sessment of the effect of air pollution on mortality
risk.
The long-term exposure studies, however, have
some significant limitations and potential biases. Al-
though studies that are well-executed attempt to con-
trol for those factors that may confound the results of
the study, there is always the possibility of insuffi-
cient or inappropriate adjustment for those factors that
affect long-term mortality rates and may be con-
founded with the factor of interest (e.g., PM concen-
trations). Prospective cohort studies have an advan-
tage over ecologic, or population-based, studies in that
they gather individual-specific information on such
important risk factors as smoking. It is always pos-
sible, however, that a relevant, individual-specific risk
factor may not have been controlled for or that some
factor that is not individual-specific (e.g., climate) was
not adequately controlled for. It is therefore possible
that differences in mortality rates that have been as-
cribed to differences in average PM levels may be due,
in part, to some other factor or factors (e.g., differ-
ences among communities in diet, exercise, ethnicity,
climate, industrial effluents, etc.) that have not been
adequately controlled for.
Another source of uncertainty surrounding the
prospective cohort studies concerns possible histori-
cal trends in PM concentrations and the relevant pe-
riod of exposure, which is as yet unknown. TSP con-
centrations were substantially higher in many loca-
tions for several years prior to the cohort studies and
had declined substantially by the time these studies
were conducted. If this is also true for PM^ 5 and PM10,
it is possible that the larger PM coefficients reported
by the long-term exposure studies (as opposed to the
short-term exposure studies) reflect an upward bias.
If the relevant exposure period extends over a decade
or more, then a coefficient based on PM concentra-
tions at the beginning of the study or in those years
immediately prior to the study could be biased up-
ward if pollution levels had been decreasing mark-
edly for a decade or longer prior to the study.
On the other hand, if a downward trend in PM
concentrations continued throughout the period of the
study, and if a much shorter exposure period is rel-
evant (e.g., contained within the study period itself),
then characterizing PM levels throughout the study
by those levels just prior to the study would tend to
bias the PM coefficient downward. Suppose, for ex-
ample, that PM levels were converging across the dif-
ferent study locations over time, and in particular, into
the study period. (That is, suppose PM levels were
decreasing over time, but decreasing faster in the high-
PM locations than in the low-PM locations, so that at
the beginning of the study period the interlocational
differences in PM concentrations were smaller than
they were a decade earlier.) Suppose also that the rel-
evant exposure period is about one year, rather than
many years. The Pope study characterizes the long-
term PM concentration in each of the study locations
by the median PM concentration in the location dur-
ing the five year period 1979-1983. Study subjects
were followed, however, from 1982 through 1989. If
the difference in median PM concentrations across the
50 study locations during the period 1979-1983 was
greater than the difference during the period 1983-
1988, and if it is PM levels during the period 1983-
1988 that most affect premature mortality during the
study period (rather than PM levels during the period
1979-1983), then the study would have attributed
interlocational differences in mortality to larger
interlocational differences in PM concentrations than
were actually relevant. This would result in a down-
ward bias of the PM coefficient estimated hi the study.
35
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
The relevant exposure period is one of a cluster
of characteristics of the mortality-PM relationship that
are as yet unknown and potentially important. It is
also unknown whether there is a time lag in the PM
effect. Finally, it is unknown whether there may be
cumulative effects of chronic exposure — that is,
whether the relative risk of mortality actually increases
as the period of exposure increases.
Three recent studies have examined the relation-
ship between mortality and long-term exposure to PM:
Pope et al. (1995), Dockery et al. (1993), and Abbey
et al. (1991). The Pope et al. study is considered a
better choice of long-term exposure study than either
of the other two studies. Pope et al. examined a much
larger population and many more locations than ei-
ther the Dockery study or the Abbey study. The
Dockery study covered only six cities. The Abbey
study covered a cohort of only 6,000 people in Cali-
fornia. In particular, the cohort in the Abbey study
was considered substantially too small and too young
to enable the detection of small increases in mortality
risk. The study was therefore omitted from consider-
ation in this analysis. Even though Pope et al. (1995)
reports a smaller premature mortality response to el-
evated PM than Dockery et al. (1993), the results of
the Pope study are nevertheless consistent with those
of the Dockery study.
Pope et al., (1995) is also unique in that it fol-
lowed a largely white and middle class population,
decreasing the likelihood that interlocational differ-
ences in premature mortality were attributable to dif-
ferences in socioeconomic status or related factors.
Furthermore, the generally lower mortality rates and
possibly lower exposures to pollution among this
group, hi comparison to poorer minority populations,
would tend to bias the PM coefficient from this study
downward, counteracting a possible upward bias as-
sociated with historical air quality trends discussed
above.
Another source of downward bias in the PM co-
efficient in Pope et al., (1995) is that intercity move-
ment of cohort members was not considered. Migra-
tion across study cities would result in exposures of
cohort members being more similar than would be
indicated by assigning city-specific annual average
pollution levels to each member of the cohort. The
more intercity migration there is, the more exposure
will tend toward an intercity mean. If this is ignored,
differences in exposure levels, proxied by differences
in city-specific annual average PM levels, will be ex-
aggerated, resulting in a downward bias of the PM
coefficient. This is because a given difference in mor-
tality rates is being associated with a larger difference
in PM levels than is actually the case.
An additional source of uncertainty hi the Pope et
al., study arises from the PM indicator used in the
study. The Pope et al. study examined the health ef-
fects associated with two indices of PM exposure;
sulfate particles and fine particles (PM25). The PM25
relationship is used in this analysis because it is more
consistent with the air quality data selected for this
analysis (PM]0). Because we use a PM2 5 mortality re-
lationship, air quality profiles were developed from
the PM,0 profiles generated for the entire 20 year pe-
riod. The same regional information about the PM]Q
components (sulfate, nitrate, organic particulate and
primary particulate) used to develop the PMIO profiles
was used to develop regional PM25/PM]0 ratios. Al-
though both urban and rural ratios are available, for
computational simplicity, only the regional urban ra-
tios were used to estimate the PM^ 5 profiles from the
PM10 profiles used in the analysis. This reflects the
exposure of the majority of the modeled population
(i.e., the urban population), while introducing some
error hi the exposure changes for the rural popula-
tion. In the east and west, where the rural ratio is larger
than the urban ratio, the change in PM2 exposure will
be underestimated for the rural population. In the cen-
tral region the PM25 change will be overestimated.
These ratios were used in each year during 1970-1990,
introducing another source of uncertainty in the analy-
sis.
After considering the relative advantages and dis-
advantages of the various alternative studies available
in the peer-reviewed literature, the Project Team de-
cided that the long-term exposure studies were pref-
erable for the purposes of the present study, primarily
because the long-term exposure studies appear to pro-
vide a more comprehensive estimate of the premature
mortality incidences attributable to PM exposure.
Among the long-term exposure studies, the Pope et
al., (1995) study appears more likely to mitigate a key
source of potential confounding. For these reasons,
the CR function estimated in Pope et al., (1995) is
considered the most reasonable choice for this analy-
sis and is utilized in spite of the several important re-
sidual uncertainties and potential biases which are sub-
sequently reflected in the PM-related mortality effect
estimate.
36
-------
Chapter 5: Physical Effects
Health Effects Modeling Results
This section provides a summary of the differences
in health effects estimated under the control and no-
control scenarios. Because the differences in air qual-
ity between the two scenarios generally increased from
1970 to 1990, and the affected population grew larger
during that period, the beneficial health effects of the
CAA increased steadily during the 1970 to 1990 pe-
riod. More detailed results are presented in Appendix
D.
Avoided Premature Mortality Estimates
The Project Team determined that, despite their
limitations, the long-term particulate matter exposure
studies provided the superior basis for estimating
mortality effects for the purpose of benefit-cost analy-
sis. Three prospective cohort studies were identified
(Pope et al. (1995), Dockery et al. (1993), and Abbey
et al. (1991)), although the Abbey study was omitted
from consideration because the cohort in that study
was considered insufficient to allow the detection of
small increases in mortality risk. Exposure-response
relationships inferred from the Pope et al. study were
used in the health benefits model to estimate avoided
mortality impacts of the CAA. The Pope et al. study
was selected because it is based on a much larger popu-
lation and a greater number of communities (50) than
is the six-city Dockery et al. Study. The results of the
Pope et al. are consistent with those of the other study,
and are consistent with earlier ecological population
mortality studies. See Appendix D for additional dis-
cussion of the selection of mortality effects studies.
Table 9 presents estimated avoided excess pre-
mature mortalities for 1990 only, with the mean esti-
mate and 90 percent confidence interval. See Appen-
dix D for more detail on results implied by individual
epidemiological studies, and on the temporal pattern
of impacts.50 The model reports a range of results for
each health endpoint. Here, the fifth percentile, mean,
and ninety-fifth percentile estimates are used to char-
acterize the distribution. The total number of avoided
cases of premature mortality due to reduced exposure
to lead (Pb) and particulate matter are presented. Ad-
ditionally, avoided mortality cases are listed by age
cohort of those who have avoided premature mortal-
ity in 1990, along with the expected remaining lifespan
(in years) for the average person in each age cohort.
The average expected remaining lifespan across all
age groups is also indicated. These averages might be
higher if data were available for PM-related mortality
in the under 30 age group and for Pb-related mortality
in the 5-39 age group.
>!>-•.f/V'J--/> ' , - y, -, x, ' > "-
Tablet, Qiteriat'ollutants-Bealth Benefits —
~>TS •*-,?„ - _^~ - j ^ ^ ^L'
Distributions of 1990 Ayoide&^remature/^Qrfalities
(thousands of cases reduced) JsM-S Stafft Population.
, Rejcnaifling
^J&xptKtwaciy
PdttalMtt „,„ Age group > ^6nrs>
*?Ks ' 30**M over „,„
< /^ f?3fs?« <
i, ' ~ y'R-34 V^48
- J^44 „ ;*3¥, <
'/*t s ,45->54^ ' '-«29~
''/7^ SS^ k' "
"*1 ^ 65-74^ ' 14^»w
y; 75-S4 "^:> , ? ""^
~f— ; >84 VTs .. . . \
„ ' ^ < - > ^?%,i,!4*
l*ad ' "Ul ages . "^t-^.
y^.;; ^ijj^ ^^Vy y >trt ,* ^ * An ffify
x" 4Q-44 -..^x x/ 38 _
' y ^••<«! ^ ^S^-S^1 ^^ ^ 29
> ^ ^ S5-64 >-*^% ^2-1
* "^ ••> > /jS5*74 _, ^"^ ^ J^
. ,f^ ^ ,~Av^.f38*
^OT^L|ui ',, ~*r#v
Annual Cases Avoided
^ (thousands)
%ifc M:ejua "
, *' 112 184
""! , 2 3
" »«5 8
?4^? / !1
14 > 23
"26 'v 43
^^rSt
" " 24 4T
'*;? ,- 22
1 ° ,%-s,I'f^
0 ^'4
>>„*< Q -^
7^0 4
1<& V -205
astit
%ilc *
257
^
11
15
33'
~$
/'"?6
59
54
5X-
f^'"1
f!8
15 v
252
o^grematute aiorta!itjes.%y age1
Non-Fatal Health Impacts
The health benefits model reports non-fatal health
effects estimates similarly to estimates of premature
mortalities: as a range of estimates for each quanti-
fied health endpoint, with the range dependent on the
quantified uncertainties in the underlying concentra-
tion-response functions. The range of results for 1990
only is characterized in Table 10 with fifth percentile,
mean, and ninety-fifth percentile estimates. All esti-
mates are expressed as thousands of new cases avoided
in 1990. "Lost IQ Points" represent the aggregate num-
ber of points (in thousands) across the population af-
fected by lead concentrations in 1990. All "Hospital
Admissions" estimates are in thousands of admissions,
regardless of the length of time spent in the hospital.
"Shortness of breath" is expressed as thousands of
' Earlier years are estimated to have had fewer excess premature mortalities.
37
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table 10. Criteria Pollutants J^iOth Benefits --, Distributions of 1990 Non-Fatal Avoided
Incidence (thousands of cases reduced) for 48 State Population.
Endpoint -PollutantCs)
Chronic Bronchitis PM ,
LostlQ Points Lead f ,„
IQ < 70 Lead
Hypertension , > ' Lead
Chronic Heart Disease * Lead
Atherothromboticfarain infarction 'Lead
Initial cerebrovascular accident -Lead <
Hospital Admissions '"
All Respiratory - - " 'JPM&O3
COPD + Pneumonia " \ PM & Q3
Jschernic Heart Disease PM *•
Congestive Heart Failure PM &CO -
Other Respiratory-Related Ailments <"'
Shortness of breath, days PM
Acute Bronchitis , - PM
Upper & Lower ReSp. Symptoms -"*PM
Any of 19 Acute Symptoms '*'"&& & O3
Asthma Attacks , - >- - PM~8£*33
Increase in Respiratory Dine?*' " ^P?^
Any Symptom * ' ' SQ2
* v * < w*> ,
Restricted Activity and WorfcLoss Days ; , v
MRAD PM&CeV,''
Work Loss Day s(WLP) PM
Affected *
Population
{age group)
Alt
children
children
men 2,0-74
40-74
40-74
40-47
"VJ
«u
over 65 ,
over $5
65 and over
<,
children
children,
children
18-6i
asthmatics
all
asthmatics
1S-6S
iS»$5
Annual Effects Avoidei
(thousartck)
5th Mean 95*
%ite %ile
493 674 886
7,440 10,400 13,000
3i 45 60
' 9,740 12,600- 15,600
0 22 64
0 ,,,4 15
• 0 6 19
75 89 103'
52 " 62 72
7 -19 31
28 39 -' 50
d *> =
14,800 6&^00 133,000
0 8,700 2f,600
5,400 . 9,500 13,400
1S,400* 130,000 244,000
170 850" iT&O
4,840 9,800 14,000
26 264 706
v
107!0(» -125,006 143,000
19,400 ' 22,600 25,600
Wut
cases'
points
cases
cases
cases
cases
cases
eases
eases
cases
cases
days
cases
eases
cases
cases,
, < cases
cases^
days
days
The following additional wejfaje benefits yi'ere quantified di«i«tly in OCOBOBMP terms; .howsehold soiJlBg
damage, visibility, deisceasfe^'vAM^et.proitoctrvity, aadagricultaralibenefits (measured it^te.ftas of rtet
surplus). - ' -""'" < >
days: that is, one "case" represents one child experi-
encing shortness of breath for one day. Likewise, "Re-
stricted Activity Days" and "Work Loss Days" are
expressed in person-days.
Other Physical Effects
Human health impacts of criteria pollutants domi-
nate quantitative analyses of the effects of the CAA,
in part because the scientific bases for quantifying air
quality and physical effect relationships are most ad-
vanced for health effects. The CAA yielded other ben-
efits, however, which are important even though they
were sometimes difficult or impossible to quantify
fully given currently available scientific and applied
economic information.
Ecological Effects
The CAA yielded important benefits in the form
of healthier ecological resources, including: stream,
river, lake and estuarine ecosystems; forest and wet-
land ecosystems; and agricultural ecosystems. These
benefits are important because of both the intrinsic
value of these ecological resources and the intimate
linkage between human health and the health and vi-
tality of our sustaining ecosystems. Given the com-
plexity of natural and agricultural ecosystems and the
large spatial and temporal dimensions involved, it has
been difficult or impossible to quantify benefits fully
given currently available scientific and applied eco-
nomic information.
Aquatic and Forest Effects
Beyond the intrinsic value of preserving natural
aquatic (i.e., lakes, streams, rivers, and estuaries), ter-
restrial (i.e., forest and grassland), and wetland eco-
systems and the life they support, protection of eco-
systems from the adverse effects of air pollution can
yield significant benefits to human welfare. The his-
torical reductions in air pollution achieved under the
CAA probably led to significant improvements in the
38
-------
Chapter 5: Physical Effects
health of ecosystems and the myriad ecological ser-
vices they provide. Reductions in acid deposition (SOx
and NOx) and mercury may have reduced adverse ef-
fects on aquatic ecosystems, including finfish, shell-
fish, and amphibian mortality and morbidity, reduced
acidification of poorly buffered systems, and reduced
eutrophication of estuarine systems. Ecological pro-
tection, in turn, can enhance human welfare through
improvements in commercial and recreational fishing,
wildlife viewing, maintenance of biodiversity, im-
provements in drinking water quality, and improve-
ments in visibility.
Wetlands ecosystems are broadly characterized as
transitional areas between terrestrial and aquatic sys-
tems in which the water table is at or near the surface
or the land is periodically covered by shallow water.
Valuable products and services of wetlands include:
flood control, water quality protection and improve-
ment, fish and wildlife habitat, and landscape and bio-
logical diversity. High levels of air pollutants have
the potential to adversely impact wetlands. Reductions
of these pollutants due to compliance with the CAA
have reduced the adverse effects of acidification and
eutrophication of wetlands, which in turn has protected
habitat and drinking water quality.
Forest ecosystems, which cover 33 percent of the
land hi the United States, provide an extensive array
of products and services to humans. Products include
lumber, plywood, paper, fuelwood, mulch, wildlife
(game), water (quality), seeds, edible products (e.g.,
nuts, syrup), drugs, and pesticides. Forest services
include recreation, biological and landscape diversity,
amenity functions (e.g., urban forest), reduced runoff
and erosion, increased soil and nutrient conservation,
pollutant sequestration (e.g., CO2, heavy metals) and
pollutant detoxification (e.g., organochlorines). The
greatest adverse effect on forest systems are imposed
by ozone. No studies have attempted to quantify the
economic benefits associated with all product and ser-
vice functions from any U.S. forest. Some studies have
attempted to estimate the net economic damage from
forest exposure to air pollutants by calculating hypo-
thetical or assumed reductions in growth rates of com-
mercial species. While quantification of forest dam-
ages remains incomplete, available evidence suggests
that recreational, service, and non-use benefits may
be substantial.
For a more comprehensive discussion of the pos-
sible ecological effects of the CAA, see Appendix E.
Quantified Agricultural Effects
Quantification of the effects of the CAA on agri-
culture was limited to the major agronomic crop spe-
cies including barley, corn, soybeans, peanuts, cotton,
wheat, and sorghum. These species account for 70
percent of all cropland in the U.S., and 73 percent of
the nation's agricultural receipts. Ozone is the primary
pollutant affecting agricultural production. Nationwide
crop damages were estimated under the control and
no-control scenarios. Net changes in economic sur-
plus (in 1990 dollars) annually and as a cumulative
present value (discounted at 5%) over the period 1976-
1990 were estimated. Positive surpluses were exhib-
ited in almost all years and were the result of the in-
crease in yields associated with decreased ozone con-
centrations under the control scenario. The present
value (in 1990) of the estimated agricultural benefits
of the CAA ranges from $7.8 billion hi the minimum
response case to approximately $37 billion in the
maximum response case51 (note that discounting 1976-
1990 benefits to 1990 amounts to a compounding of
benefits). Exposure-response relationships and culti-
var mix reflect historical patterns and do not account
for possible substitution of more ozone-resistant cul-
tivars in the no-control scenario. Thus, the upper end
of the range of benefit calculations may overestimate
the actual agricultural benefits of the CAA with re-
spect to these crops. Because numerous crops are ex-
cluded from the analysis, including high value crops
that may be sensitive to ozone, the lower end of the
range is not likely to fully capture the agricultural
benefits of reductions in ozone.
Effects of Air Toxics
In addition to control of criteria pollutants, the
Clean Air Act resulted in control of some air toxics
— defined as non-criteria pollutants which can cause
adverse effects to human health and to ecological re-
sources. Control of these pollutants resulted both from
incidental control due to criteria pollutant programs
and specific controls targeted at air toxics through the
National Emission Standards for Hazardous Air Pol-
lutants (NESHAPs) under Section 112 of the Act.
Air toxics are capable of producing a wide vari-
ety of effects. Table 11 presents the range of potential
human health and ecological effects which can occur
due to air toxics exposure. For several years, the pri-
mary focus of risk assessments and control programs
designed to reduce air toxics has been cancer. Accord-
51 Ranges reflect usage of alternate exposure-response functions.
39~
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table 11. Health and Weiarrf-Bffects o
Effect Category
H H alth
Human Welfare
Ecological
Other Welfare
Quantified Effects , -
Canc'"rMtirt lit •'. <•<«-"- <>'•'
-nciotilltyst^na^'/
source.- - »- *,,,„.,
'!' ,-»,,«#
•. J, J. ., ^ <[
• -'^-:,7.
- ...x. „*.„.... •>$.*, x,,;v_
-'* •* ^^^<^'
3 % c - X ^ %v -.••<, ^; -.
" ' i '" "' ' S " -.,-. : »
•. ' - j A
'" ''---^^^•'V-
" s } s\ •,
JUnquantified Effects
' '' XiV -'"•> - , '**'*
* •atility soa'rce , ,
' 3%n#w£r' effects- *•„
- meuroldgical ,^, ;, r'/f>
'""re&KJda^? *• ' '"' '"'
•f - hematopxii ef ic •,,,,„•'
•? developinental
«, Decreased iacome and ,
-due to ifis^ gd,y|j5C;nes
,Bnfects<.'o» plaits' w^*^-
^Ecosystein eflEfects -^; 't>'i^*
^-L^ss^pf 010 loglc^[ %v--^-;
^B?'- '-;r\
^ s ^ ^ ' " ' '"' ''' J '
V$$j|>|licy - <", . ^,- , ^
Baildiag Deterfojratiba
'Otfe^r P'fflSsiWe EITects
*
''''" ^' ' ^ ' ^ i ^, "•
>•? f ,;'" , < s'<.0<
^ J^ "V -^^- ,-,rf w
' ' '' ^ *4 \''\' 'f ' '
^resulting from decreased - ^
. pjjysie^l peifortnaace
t,B3ffeets on jgiobal plimke , - -
, ---;,-/ -,{,',;,S,,,,,,
• ••:•;•„,, > , ' >
">.',<.?/•. --,, , ,, •. ' ",'--/--;-'
"t'oss of fcfetosteVrgrvtnity
'T_:, \^"^ 'V
, V-...v«*^ - -
ing to present EPA criteria, there are over 100 known
or suspected carcinogens. EPA's 1990 Cancer Risk
study indicated that as many as 1,000 to 3,000 can-
cers annually may be attributable to the air toxics for
which assessments were available (virtually all of this
estimate came from assessments of about a dozen well-
studied pollutants).52
In addition to cancer, these pollutants can cause a
wide variety of health effects, ranging from respira-
tory problems to reproductive and developmental ef-
fects. There has been considerably less work done to
assess the magnitude of non-cancer effects from air
toxics, but one survey study has shown that some pol-
lutants are present in the atmosphere at reference lev-
els that have caused adverse effects in animals.53
Emissions of air toxics can also cause adverse
health effects via non-inhalation exposure routes. Per-
sistent bioaccumulating pollutants, such as mercury
and dioxins, can be deposited into water or soil and
subsequently taken up by living organisms. The pol-
lutants can biomagnify through the food chain and
exist in high concentrations when consumed by hu-
mans hi foods such as fish or beef. The resulting ex-
posures can cause adverse effects hi humans, and can
also disrupt ecosystems by affecting top food chain
species.
Finally, there are a host of other potential eco-
logical and welfare effects associated with air toxics,
for which very little exists in the way of quantitative
analysis. Toxic effects of these pollutants have the
potential to disrupt both terrestrial and aquatic eco-
systems and contribute to adverse welfare effects such
as fish consumption advisories in the Great Lakes.54
52 U.S. EPA, Cancer Risk from Outdoor Exposure to Air Toxics. EPA-450/l-90-004f. Prepared by EPA/OAR/OAQPS.
53 U.S. EPA, 'Toxic Air Pollutants and Noncancer Risks: Screening Studies," External Review Draft, September, 1990.
54 U.S. EPA, Office of Air Quality Planning and Standards. "Deposition of Air Pollutants to the Great Waters, First Report to
Congress." May 1994. EPA-453/R-93-055.
40
-------
Chapter 5: Physical Effects
Unfortunately, the effects of air toxics emissions
reductions could not be quantified for the present
study. Unlike criteria pollutants, there was relatively
little monitoring data available for air toxics, and that
which exists covered only a handful of pollutants.
Emissions inventories were very limited and incon-
sistent, and air quality modeling has only been done
for a few source categories. In addition, the scientific
literature on the effects of air toxics was generally
much weaker than that available for criteria pollut-
ants.
Limitations in the underlying data and analyses
of air toxics led the Project Team to exclude the avail-
able quantitative results from the primary analysis of
CAA costs and benefits. The estimates of cancer inci-
dence benefits of CAA air toxics control which were
developed, but ultimately rejected, are presented in
Appendix H. Also found in Appendix H is a list of
research needs identified by the Project Team which,
if met, would enable at least a partial assessment of
air toxics benefits in future section 812 studies.
Uncertainty In The Physical Effects
Estimates
As discussed above, and in greater detail in Ap-
pendix D, a number of important assumptions and
uncertainties in the physical effects analysis may in-
fluence the estimate of monetary benefits presented
in this study. Several of these key uncertainties, their
potential directional bias, and the potential signifi-
cance of this uncertainty for the overall results of the
analysis are summarized in Table 12.
41
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table 12, Uncertainties Associated wJtb Physical Effects Modeling.
Potential Source of Error
Estimation of PMu from modeled PM<«
and TSP data (to support mortality
estimation)
Extrapolation of health effects to
populations dia ant frommonitors (of
monitored counties in the ease'of PM).
Estimation of degreeof life-shortening _
associated with PMNcelated premature
mortality. \ ,
<, ^
Assumption of zero lag between* " ">
exposure and incidence of PM-related-- -
premature mortality.
Choke of CR function (i.e.* "across-
study" u ncertaint ies)
Uncettaimy associated with CR- <"
functions derived from each individual
study (i.e., "within study" uncertainty) ~
Exdusion of potential UV-B attenuation
benefits associated witbhigher
concentrationsoftropospheric ozone "
undertheno-contro5c4se. „ ov_ -
Exclusion of potential substitution of
ozone-resistant ojltivats in agriculture ~,
analysis.
Exclusion of other agricultural effects
(crops, pollutants)
Exclusion of effects on terrestrial,
wetland, and aquatic ecosystems, and
forests.
No quantification of materials damage
„ Dimtfoni of F&tentfel Btes
In Physical Effects Estimate
Unknown
; Probabteoverestimate. _
" <• " u <-~ ~"v'
Unknown.
vf.:f'Z " "
-' -' M, , *
( *°< ? ^
Ovftrestteiate." - -
< f «ft,A%. *
V- "
Sfe ' >
«*>•
_•* ><>^^-^f-i' &•
t^tikiiowa,-
^""VS * v*^St
Baknown-
* <. •?<> -
"dlerestimate. «, -,
•>••>" "• v"
Ovetestimati " '
lfnderes||jnate, -
UhderestlmatB.
^ ~ ^> s
Ua8erestraiaie ' I ,
Slgnilfcaw* RdsHve to Key Voeewsfntfes JB
Overall Monetary Benefit Estimate
Significant. Estimated PMis profiles are used
to calculate most of the preiaatate'mortaSi^, -
there is significant uncertainty about how fee
fine particle share of overall PM levels vanes
temporally and spatially throughout the 20 year
period. „ .
Probably minor. In addition, this adjustment -
avoids the underestimation which would result
by estimating effects for only those people
living nearmonitors. Potential overestimate
r»ay resttlt to the extent air quality in areas
distant from monitors is sigmf tcantiy better fi>an
in monitored areas This disparity should be
quite minor for regional pollutants, such 'as
ozone and fine parflc slates.
Unknown, possibly Significant When using a
value of hfe^years approach. Varylng:tlB<
estimate of degree of prematurity has no effect
on the aggregate benefit estimate when a value
of stat^af to«l We approach is used since all
«icid"etices of premature mortality are valued'
efttjaljy. Wader the alternative approach based
on valuing tndvidual life-years, the influence
of atenasive values for numbers of average
lite-years lost may bft significant,
Probably minor, the shdtt-temmortality "
stjadtes tndtcate tfeat a agaificantponiOKof fee
premature mortah'ty associated with exposure to
elevated PM concentrations ss very short-term
(i.e., a matter
-------
6
Economic Valuation
Estimating the reduced incidence of physical ef-
fects represents a valuable measure of health benefits
for individual endpoints; however, to compare or ag-
gregate benefits across endpoints, the benefits must
be monetized. Assigning a monetary value to avoided
incidences of each effect permits a summation, in terms
of dollars, of monetized benefits realized as a result
of the CAA, and allows that summation to be com-
pared to the cost of the CAA.
For the present analysis of health and welfare ben-
efits, valuation estimates were obtained from the eco-
nomic literature, and are reported in dollars per case
reduced for health effects and dollars per unit of
avoided damage for welfare effects.55 Similar to esti-
mates of physical effects provided by health studies,
each of the monetary values of benefits applied in this
analysis is reported in terms of a mean value and a
probability distribution around the mean estimate. The
statistical form of the probability distribution used for
the valuation measures varies by endpoint. For ex-
ample, while the estimate of the dollar value of an
avoided premature mortality is described by the
Weibull distribution, the estimate for the value of a
reduced case of acute bronchitis is assumed to be uni-
formly distributed between a minimum and maximum
value.
Methods for Valuation of Health
and Welfare Effects
In environmental benefit-cost analysis, the dollar
value of an environmental benefit (e.g., a health-re-
lated improvement in environmental quality) conferred
on a person is the dollar amount such that the person
would be indifferent between having the environmen-
tal benefit and having the money. In some cases, this
value is measured by studies which estimate the dol-
lar amount required to compensate a person for new
or additional exposure to an adverse effect. Estimates
derived in this manner are referred to as "willingness-
to-accept" (WTA) estimates. In other cases, the value
of a welfare change is measured by estimating the
amount of money a person is willing to pay to elimi-
nate or reduce a current hazard. This welfare change
concept is referred to as "willingness-to-pay" (WTP).
For small changes in risk, WTP and WTA are virtu-
ally identical, primarily because the budget constraints
normally associated with expressions of WTP are not
significant enough to drive a wedge between the esti-
mates. For larger risk changes, however, the WTP and
WTA values may diverge, with WTP normally being
less than WTA because of the budget constraint ef-
fect. While the underlying economic valuation litera-
ture is based on studies which elicited expressions of
WTP and/or WTA, the remainder of this report refers
to all valuation coefficients as WTP estimates. In some
cases (e.g., stroke-related hospital admissions), nei-
ther WTA nor WTP estimates are available and WTP
is approximated by cost of illness (COI) estimates, a
clear underestimate of the true welfare change since
important value components (e.g., pain and suffering
associated with the stroke) are not reflected in the out-
of-pocket costs for the hospital stay.
For most goods, WTP can be observed by exam- •
ining actual market transactions. For example, if a
gallon of bottled drinking water sells for one dollar, it
can be observed that at least some persons are willing
to pay one dollar for such water. For goods that are
not exchanged in the market, such as most environ-
mental "goods," valuation is not so straightforward.
Nevertheless, value may be inferred from observed
behavior, such as through estimation of the WTP for
mortality risk reductions based on observed sales and
prices of safety devices such as smoke detectors. Al-
ternatively, surveys may be used in an attempt to elicit
directly WTP for an environmental improvement.
Wherever possible, this analysis uses estimates
of the mean WTP of the U.S. population to avoid an
environmental effect as the value of avoiding that ef-
fect. In some cases, such estimates are not available,
and the cost of mitigating or avoiding the effect is
used as a rough estimate of the value of avoiding the
effect. For example, if an effect results in hospitaliza-
tion, the avoided medical costs were considered as a
possible estimate of the value of avoiding the effect.
Finally, where even the "avoided cost" estimate is not
available, the analysis relies on other available meth-
ods to provide a rough approximation of WTP. As
noted above, this analysis uses a range of values for
most environmental effects, or endpoints. Table 13
55 The literature reviews and valuation estimate development process is described in detail in Appendix I and in the referenced
supporting reports. ™
43
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table 13. Kbalth and Welfare Effects Unit Valuation
(1990 dollars).
Mortality
Chronic Bronchitis
IQ Changes
Lost IQ fbints
IQ<70
Hypertension
Strokes*
Coronary Heart Disease
Hospital Admissions
Ischemic Heart Disease
Congestive Heart Failure
COPD
Pneumonia
All Respiratory
Respiratory Illness and Symptoms
Acute Bronchitis
Acute Asthma
Acute Respiratory Symptoms
Upper Respiratory Symptoms
Lower Respiratory Symptoms
Shortness of Breath
Work Loss Days
Mild Restricted Activity Days
Welfare Benefits
Visibility
Household Soiling
Decreased Worker Productivity
Agriculture (Net Surplus)
PM&Pb
PM
Pb "
Pb
Pb
Pb
Pb
PM
PM
PM&Oj
PM&03 „
pMj&bj -'
PM
PM&tt . _
PM, Qj, NOj,"S
PM
PM
PM
PM
PM&Oj
DeciView
PM „
Qt
P».
$4,800,000 ,,per case
$260,000 per case,
"" - .',. > *
M&OOO per IQ. point
$42,000 per case'* '
$680 j>erea$e
-$200,000 per case-m ales
$150,000 per«ase»
'< females ,
"' $52,-OvOO per ease
? ' > •'-
$10^300 «per case 6" £
$8,1,00* j^t-case ,(^
/ ; ^;S|J2I -/;
;'-">•" '$6,100^1^ , I
^ ' w^ ^^ ^
$45 per case
^>a ^i ^ $32 percase* , ^
o'i - ; « -$XS_ per ease *'-„•<
~ "$19 per ease v^ "^
"^jl^Zv^wcaas* ' J£
"I5L30 per day ' ""
,,^ "$83 'per (Jay
'--'' •->^' ,'
*^"2^sr*
$2.50 ApeF^juseliold
dianEge
*i **f r
Estimated CLs^^iti' ;A •
f, ,vx. ^. ,\V f:
* Strokes are comprised of atherothrombotid bratuitoritjpiis and cefebroyasi
accidents; both areestimated to have the same monetsrytvalue. ^'{,^'^1 f *
** Decreased productivity valued as change in, daily ?ya,gesf $1 perwoiitSjy>er 10^
decrease in O). ^a-—"^?-^ •>
provides a summary of the mean unit value estimates
used in the analysis. The full range of values can be
found in Appendix I.
Mortality
Some forms of air pollution increase the probabil-
ity that individuals will die prematurely. The concen-
tration-response functions for mortality used hi this
analysis express this increase in mortality risk as cases
of "excess premature mortality" per time pe-
riod (e.g., per year).
The benefit, however, is the avoidance
of small increases in the risk of mortality. If
individuals' WTP to avoid small increases in
risk is summed over enough individuals, the
value of a statistical premature death avoided
can be inferred.56 For expository purposes,
this valuation is expressed as "dollars per
mortality avoided," or "value of a statistical
life" (VSL), even though the actual valuation
is of small changes in mortality risk.
The mortality risk valuation estimate
used in this study is based on an analysis of
26 policy-relevant value-of-life studies (see
Table 14). Five of the 26 studies are contin-
gent valuation (CV) studies, which directly
solicit WTP information from subjects; the
rest are wage-risk studies, which base WTP
estimates on estimates of the additional com-
pensation demanded in the labor market for
riskier jobs. The Project Team used the best
estimate from each of the 26 studies to con-
struct a distribution of mortality risk valua-
tion estimates for the section 812 study. A
Weibull distribution, with a mean of $4.8 mil-
lion and standard deviation of $3.24 million,
provided the best fit to the 26 estimates. There
is considerable uncertainty associated with
this approach, however, which is discussed
in detail later hi this chapter and in Appen-
dix!.
In addition, the Project Team developed
alternative calculations based on a life-years
lost approach. To employ the value of statis-
tical life-year (VSLY) approach, the Project
Team had to first estimate the age distribu-
tion of those lives which would be saved by
reducing air pollution. Based on life expect-
ancy tables, the life-years saved from each statistical
life saved within each age and sex cohort were calcu-
lated. To value these statistical life-years, a concep-
tual model was hypothesized which depicted the rela-
tionship between the value of life and the value of
life-years. As noted earlier in Table 9, the average
number of life-years saved across all age groups
for which data were available are 14 for PM-
related mortality and 38 for Pb-related mortality. The
** Because people are valuing small decreases in the risk of premature mortality, it is expected deaths that are inferred. For
example, suppose that a given reduction in pollution confers on each exposed individual a decrease in mortal risk of 1/100,000. Then
among 100,000 such individuals, one fewer individual can be expected to die prematurely . If each individual's WTP for that risk
reduction is $50, then the implied value of a statistical premature death avoided is $50 x 100,000 = $5 million.
44 '
-------
Chapter 6: Economic Valuation
Surrunary 'of Mortality Valuation Estimates
""
Butter 09»3)
Mller an)
Qegax et af'^SS)"
Marin and'jfcacharopoulos
Kaeisner an
4.1
4.S
7.2
" 7.3
9.1s
average for PM, in particular, differs from the 35-year
expected remaining lifespan derived from existing
wage-risk studies.57
Using the same distribution of value of life esti-
mates used above (i.e. the Weibull distribution with a
mean estimate of $4.8 million), a distribution for the
value of a life-year was then estimated and combined
with the total number of estimated life-years lost. The
details of these calculations are presented in Appen-
dix!
Survey-Based Values
Willingness-to pay for environmental improve-
ment is often elicited through survey methods (such
as the "contingent valuation" method). Use of such
methods in this context is controversial within the
economics profession. In general, economists prefer
to infer WTP from observed behavior. There are times
when such inferences are impossible, however, and
some type of survey technique may be the only means
of eliciting WTP. Economists' beliefs regarding the
reliability of such survey-based data cover a broad
spectrum, from unqualified acceptances of the results
of properly-conducted surveys to outright rejections
of all survey-based valuations.
In this analysis, unit valuations which rely exclu-
sively on the contingent valuation method are chronic
bronchitis, respiratory-related ailments, minor re-
stricted activity days, and visibility. As indicated
above, the value derived for excess premature mortal-
ity stems from 26 studies, of which five use the con-
tingent valuation method. These five studies are within
the range of the remaining 21 labor market studies.
All five report mortality valuations lower than the
central estimate used in this analysis. Excluding the
contingent valuation studies from the mortality valu-
ation estimate would yield a central estimate approxi-
mately ten percent higher than the 4.8 million dollar
value reported above. The endpoints with unit valua-
tions based exclusively on contingent valuation ac-
count for approximately 30 percent of the present value
of total .monetized benefits. Most of the CV-based
benefits are attributable to avoided cases of chronic
bronchitis.
Chronic Bronchitis
The best available estimate of WTP to avoid a
case of chronic bronchitis (CB) comes from "Viscusi
et al.(1991). The case of CB described to the respon-
dents in the "Viscusi study, however, was described by
the authors as a severe case. The Project Team em-
ployed an estimate of WTP to avoid a pollution-re-
lated case of CB that was based on adjusting the WTP
to avoid a severe case, estimated by Viscusi et al.
(1991), to account for the likelihood that an average
case of pollution-related CB is not as severe as the
case described in the Viscusi study.
The central tendency estimate of WTP to avoid a
pollution-related case of chronic bronchitis (CB) used
in this analysis is the mean of a distribution of WTP
estimates. This distribution incorporates the uncer-
tainty from three sources: (1) the WTP to avoid a case
of severe CB, as described by Viscusi et al., 1991; (2)
the severity level of an average pollution-related case
"See, for example, Moore and Viscusi (1988) or Viscusi (1992).
45
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
of CB (relative to that of the case described by Viscusi
et al.(1991); and (3) the elasticity of WTP with re-
spect to severity of the illness. Based on assumptions
about the distributions of each of these three uncer-
tain components, a distribution of WTP to avoid a
pollution-related case of CB was derived by Monte
Carlo methods. The mean of this" distribution, which
was about $260,000, is taken as the central tendency
estimate of WTP to avoid a pollution-related case of
CB. The three underlying distributions, and the gen-
eration of the resulting distribution of WTP, are de-
scribed in Appendix I.
Respiratory-Related Ailments
In general, the valuations assigned to the respira-
tory-related ailments listed in Table 14 represent a
combination of willingness to pay estimates for indi-
vidual symptoms which comprise each ailment. For
example, a willingness to pay estimate to avoid the
combination of specific upper respiratory symptoms
defined in the concentration-response relationship
measured by Pope et al. (1991) is not available. How-
ever, while that study defined upper respiratory symp-
toms as one suite of ailments (runny or stuffy nose;
wet cough; and burning, aching, or red eyes), the valu-
ation literature reported individual WTP estimates for
three closely matching symptoms (head/sinus conges-
tion, cough, and eye irritation). The available WTP
estimates were therefore used as a surrogate to the
values for the precise symptoms defined in the con-
centration-response study.
To capture the uncertainty associated with the
valuation of respiratory-related ailments, this analy-
sis incorporated a range of values reflecting the fact
that an ailment, as defined in the concentration-Re-
sponse relationship, could be comprised of just one
symptom or several. At the high end of the range,khe
valuation represents an aggregate of WTP estimates
for several individual symptoms. The low end repre-
sents the value of avoiding a single mild symptom.
Minor Restricted Activity Days
An individual suffering from a single severe or a
combination of pollution-related symptoms may ex-
perience a Minor Restricted Activity Day (MRAD).
Krupnick and Kopp (1988) argue that mild symptoms
will not be sufficient to result in a MRAD, so that
WTP to avoid a MRAD should exceed WTP to avoid
any single mild symptom. On the other hand, WTP to
avoid a MRAD should not exceed the WTP to avoid a
work loss day (which results when the individual ex-
periences more severe symptoms). No studies are re-
ported to have estimated WTP to avoid a day of mi-
nor restricted activity. Instead, this analysis uses an
estimate derived from WTP estimates for avoiding
combinations of symptoms which may result in a day
of minor restricted activity ($38 per day). The uncer-
tainty range associated with this value extends from
the highest value for a single symptom to the value
for a work loss day. Furthermore, the distribution ac-
knowledges that the actual value is likely to be closer
to the central estimate than either extreme.
Visibility
The value of avoided visibility impairment was
derived from existing contingent valuation studies of
the household WTP to improve visibility, as reported
in the economics literature. These studies were used
to define a single, consistent basis for the valuation of
visibility benefits nationwide. The central tendency
of the benefits estimate is based on an annual WTP of
$14 per household per unit improvement in the
DeciView index, with upper and lower bounds of $21
and $8, respectively, on the uncertainty range of the
estimate.
Avoided Cost Estimates
For some health effects, WTP estimates are not
available, and the Project Team instead used "costs
avoided" as a substitute for WTP. Avoided costs were
used to value the following endpoints: hypertension,
hospital admissions, and household soiling.
Hypertension and Hospital Admissions
Avoided medical costs and the avoided cost of lost
work time were used to value hypertension (high blood
pressure) and hospital admissions (this includes hos-
pital admissions for respiratory ailments as well as
heart disease, heart attacks, and strokes).
For those hospital admissions which were speci-
fied to be the initial hospital admission (in particular,
hospital admissions for coronary heart disease (CHD)
events and stroke), avoided cost estimates should con-
sist of the present discounted value of the stream of
medical expenditures related to the illness, as well as
the present discounted value of the stream of lost earn-
ings related to the illness. While an estimate of present
discounted value of both medical expenditures and
lost earnings was available for stroke ($200,000 for
46
-------
Chapter 6: Economic Valuation
males and $150,000 for females), the best available
estimate for CHD ($52,000) did not include lost earn-
ings. Although no published estimates of the value of
lost earnings due to CHD events are available, one
unpublished study suggests that this value could be
substantial, possibly exceeding the value of medical
expenditures. The estimate of $52,000 for CHD may
therefore be a substantial underestimate. The deriva-
tions of the avoided cost estimates for CHD and stroke
are discussed in Appendix G.
In those cases for which it is unspecified whether
the hospital admission is the initial one or not (that is,
for all hospital admissions endpoints other than CHD
and stroke), it is unclear what portion of medical ex-
penditures and lost earnings after hospital discharge
can reasonably be attributed to pollution exposure and
what portion might have resulted from an individual's
pre-existing condition even in the absence of a par-
ticular pollution-related hospital admission. In such
cases, the estimates of avoided cost include only those
costs associated with the hospital stay, including the
hospital charge, the associated physician charge, and
the lost earnings while in the hospital ($6,100 to
$10,300, depending on the ailment for which hospi-
talization is required).
The estimate of avoided cost for hypertension in-
cluded physician charges, medication costs, and hos-
pitalization costs, as well as the cost of lost work time,
valued at the rate estimated for a work loss day (see
discussion below). Based on this approach, the value
per year of avoiding a case of hypertension is taken to
equal the sum of medical costs per year plus work
loss costs per year; the resulting value is $680 per case
per year.
Presumably, willingness-to-pay to avoid the ef-
fects (and treatment) of hypertension would reflect
the value of avoiding any associated pain and suffer-
ing, and the value placed on dietary changes, etc. Like- "
wise, the value of avoiding a health effect that would
require hospitalization or doctor's care would include
the value of avoiding the pain and suffering caused
by the health effect as well as lost leisure time, in ad-
dition to medical costs and lost work time. Conse-
quently, the valuations for these endpoints used in this
analysis likely represent lower-bound estimates of the
true social values for avoiding such health effects.
Household Soiling
This analysis values benefits for this welfare ef-
fect by considering the avoided costs of cleaning
houses due to particulate matter soiling. The Project
Team's estimate reflects the average household's an-
nual cost of cleaning per ug/m3 particulate matter
($2.50). Considered in this valuation are issues such
as the nature of the particulate matter, and the propor-
tion of households likely to do the cleaning themselves.
Since the avoided costs of cleaning used herein do
not reflect the loss of leisure time (and perhaps work
time) incurred by those who do their own cleaning,
the valuation function likely underestimates true WTP
to avoid additional soiling.
Other Valuation Estimates
Changes in Children's IQ
One of the major effects of lead exposure is per-
manently impaired cognitive development in children.
No ready estimates of society's WTP for improved
cognitive ability are currently available. Two effects
of IQ decrements can be monetized, however: reduc-
tions in expected lifetime income, and increases in
societal expenditures for compensatory education.
These two effects almost certainly understate the WTP
to avoid impaired cognitive development in children,
and probably should be considered lower bound esti-
mates. In the absence of better estimates, however,
the Project Team has assumed that the two monetized
effects represent a useful approximation of WTP.
The effect of IQ on expected lifetime income com-
prises a direct and an indirect effect. The direct effect
is drawn from studies that estimate, all else being
equal, the effect of IQ on income. The indirect effect
occurs as a result of the influence of IQ on educa-
tional attainment: higher IQ leads to more years of
education, and more education leads in turn to higher
expected future income. However, this indirect ben-
efit is mitigated, but not eliminated, by the added costs
of the additional education and by the potential earn-
ings forgone by the student while enrolled in school.58
Combining the direct and indirect influences, the net
effect of higher IQ on expected lifetime income (dis-
58 Theoretically, the indirect effect should be small relative to the direct effect of IQ on future earnings. The empirical research
used to derive values for this analysis, however, implies that the indirect effect is roughly equal in magnitude to the direct effect. One
can infer from this information that there is a market distortion of some sort present (such as imperfect knowledge of the returns to
education), or, perhaps, that individuals make their education "investments" for purposes other than (or in addition to) "maximizing
lifetime income." See Appendix G for further discussion of this issue.
47
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
counted to the present at five percent) is estimated to
be $3,000 per additional IQ point.
In this analysis, it is assumed that part-time com-
pensatory education is required for all children with
IQ less than 70. The Project Team assumed that the
WTP to avoid cases of children with IQ less than 70
can be approximated by the cost ($42,000 per child)
of part-time special education in regular classrooms
from grades one through twelve (as opposed to inde-
pendent special education programs), discounted to
the present at five percent. See Appendix G for more
detail on valuation methods and data sources for IQ
effects and other lead-related health impacts.
Work Loss Days and Worker
Productivity
For this analysis, it was assumed that the median
daily 1990 wage income of 83 dollars was a reason-
able approximation of WTP to avoid a day of lost
work. Although a work loss day may or may not af-
fect the income of the worker, depending on the terms
of employment, it does affect economic output and is
thus a cost to society. Conversely, avoiding the w,0rk
loss day is a benefit.
A decline in worker productivity has been mea-
sured in outdoor workers exposed to ozon,e. Reduced
productivity is measured in terms of the/reduction in
daily income of the average worker engaged in strenu-
ous outdoor labor, estimated at $1 per 10 percent in-
crease in ozone concentration.
Agricultural Benefits
Similar to the other welfare effects, the agricul-
tural benefits analysis estimated benefits in dollars per
unit of avoided damage, based on estimated changes
in crop yields predicted by an agricultural sector
model. This model incorporated agricultural price,
farm policy, and other data for each year. Based on
expected yields, the model estimated the production
levels for each crop, and the economic benefits to con-
sumers, and to producers, associated with these pro-
duction levels. To the extent that alternative exposure-
response relationships were available, a range of po-
tential benefits was calculated (see Appendix F).
Valuation Uncertainties
The Project Team attempted to handle most valu-
ation uncertainties explicitly and quantitatively by
expressing values as distributions (see Appendix I for
a complete description of distributions employed),
using a Monte-Carlo simulation technique to apply
the valuations to physical effects (see Chapter 7) with
the mean of each valuation distribution equal to the
"best estimate" valuation. This approach does not, of
course, guarantee that all uncertainties have been ad-
equately characterized, nor that the valuation estimates
are unbiased. It is possible that the actual WTP to avoid
an air pollution-related impact is outside of the range
of estimates used in this analysis. Nevertheless, the
Project Team believes that the distributions employed
are reasonable approximations of the ranges of uncer-
tainty, and that there is no compelling reason to be-
lieve that the mean values employed are systemati-
cally biased (except for the IQ-related and avoided
cost-based values, both of which probably underesti-
mate WTP).
One particularly important area of uncertainty is
valuation of mortality risk reduction. As noted in Chap-
ter 7, changes in mortality risk are a very important
component of aggregate benefits, and mortality risk
valuation is an extremely large component of the quan-
tified uncertainty. Consequently, any uncertainty con-
cerning mortality risk valuation beyond that addressed
by the quantitative uncertainty assessment (i.e., that
related to the Weibull distribution with a mean value
of $4.8 million) deserves note. One issue merits spe-
cial attention: uncertainties and possible biases related
to the "benefits transfer" from the 26 valuation source
studies to valuation of reductions in PM-related mor-
tality rates.
Mortality Risk Benefits Transfer
Although each of the mortality risk valuation
source studies (see Table 14) estimated the average
WTP for a given reduction in mortality risk, the de-
gree of reduction in risk being valued varied across
studies and is not necessarily the same as the degree
of mortality risk reduction estimated in this analysis.
The transferability of estimates of the value of a sta-
tistical life from the 26 studies to the section 812 ben-
efit analysis rests on the assumption that, within a rea-
sonable range, WTP for reductions in mortality risk
is linear in risk reduction. For example, suppose a study
48
-------
Chapter 6: Economic Valuation
estimates that the average WTP for a reduction in
mortality risk of 1/100,000 is 50 dollars, but that the
actual mortality risk reduction resulting from a given
pollutant reduction is 1/10,000. If WTP for reductions
in mortality risk is linear in risk reduction, then a WTP
of 50 dollars for a reduction of 1/100,000 implies a
WTP of 500 dollars for a risk reduction of 1/10,000
(which is ten times the risk reduction valued in the
study). Under the assumption of linearity, the estimate
of the value of a statistical life does not depend on the
particular amount of risk reduction being valued.
Although the particular amount of mortality risk
reduction being valued in a study may not affect the
transferability of the WTP estimate from the study to
the benefit analysis, the characteristics of the study
subjects and the nature of the mortality risk being val-
ued in the study could be important. Certain charac-
teristics of both the population affected and the mor-
tality risk facing that population are believed to affect
the average WTP to reduce risk. The appropriateness
of the mean of the WTP estimates from the 26 studies
for valuing the mortality-related benefits of reductions
in pollutant concentrations therefore depends not only
on the quality of the studies (i.e., how well they mea-
sure what they are trying to measure), but also on (1)
the extent to which the subjects in the studies are simi-
lar to the population affected by changes in air pollu-
tion and (2) the extent to which the risks being valued
are similar.
The substantial majority of the 26 studies relied
upon are wage-risk (or labor market) studies. Com-
pared with the subjects in these wage-risk studies, the
population most affected by air pollution-related mor-
tality risk changes is likely to be, on average, older
and probably more risk averse. Some evidence sug-
gests that approximately 85 percent of those identi-
fied in short-term ("episodic") studies who die pre-
maturely from PM-related causes are over 65.59 The
average age of subjects in wage-risk studies, in con-
trast, would be well under 65.
The direction of bias resulting from the age dif-
ference is unclear. It could be argued that, because an
older person has fewer expected years left to lose, his
or her WTP to reduce mortality risk would be less
than that of a younger person. This hypothesis is sup-
ported by one empirical study, Jones-Lee et al. (1985),
which found WTP to avoid mortality risk at age 65 to
be about 90 percent of what it is at age 40. On the
other hand, there is reason to believe that those over
65 are, in general, more risk averse than the general
population, while workers in wage-risk studies are
likely to be less risk averse than the general popula-
tion. Although the list of 26 studies used here excludes
studies that consider only much-higher-than-average
occupational risks, there is nevertheless likely to be
some selection bias in the remaining studies—that is,
these studies are likely to be based on samples of
workers who are, on average, more risk-loving than
the general population. In contrast, older people as a
group exhibit more risk-averse behavior.
There is substantial evidence that the income elas-
ticity of WTP for health risk reductions is positive
(although there is uncertainty about the exact value of
this elasticity). Individuals with higher incomes (or
greater wealth) should, then, be willing to pay more
to reduce risk, all else equal, than individuals with
lower incomes or wealth. The comparison between
the (actual and potential) income or wealth of the
workers in the wage-risk studies versus that of the
population of individuals most likely to be affected
by changes in pollution concentrations, however, is
unclear. One could argue that because the elderly are
relatively wealthy, the affected population is also
wealthier, on average, than are the wage-risk study
subjects, who tend to be middle-aged (on average)
blue-collar workers. On the other hand, the workers
in the wage-risk studies will have potentially more
years remaining in which to acquire streams of in-
come from future earnings. In addition, it is possible
that among the elderly it is largely the poor elderly
who are most vulnerable to air pollution-related mor-
tality risk (e.g., because of generally poorer health
care). On net, the potential income comparison is un-
clear.
Although there may be several ways in which job-
related mortality risks differ from air pollution-related
mortality risks, the most important difference may be
that job-related risks are incurred voluntarily whereas
air pollution-related risks are incurred involuntarily.
There is some evidence60 that people will pay more to
reduce involuntarily incurred risks than risks incurred
voluntarily. If this is the case, WTP estimates based
on wage-risk studies may be downward biased esti-
mates of WTP to reduce involuntarily incurred air
pollution-related mortality risks.
59 See Schwartz and Dockery (1992), Ostro et al. (1995), and Chestnut (1995).
""See, for example, Violette and Chestnut, 1983.
49
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Finally, another important difference related to the
nature of the risk may be that some workplace mortal-
ity risks tend to involve sudden, catastrophic events,
whereas air pollution-related risks tend to involve
longer periods of disease and suffering prior to death.
Some evidence suggests that WTP to avoid a risk of a
protracted death involving prolonged suffering and
loss of dignity and personal control is greater than the
WTP to avoid a risk (of identical magnitude) of sud-
den death. To the extent that the mortality risks ad-
dressed in this assessment are associated with longer
periods of illness or greater pain and suffering than
are the risks addressed in the valuation literature, the
WTP measurements employed in the present analysis
would reflect a downward bias.
The potential sources of bias introduced by rely-
ing on wage-risk studies to derive an estimate of the
WTP to reduce air pollution-related mortality risk are
summarized in Table 15. Among these potential bi-
ases, it is disparities in age and income between the
subjects of the wage-risk studies and those affected
by air pollution which have thus far motivated spe-
cific suggestions for quantitative adjustment61; how-
ever, the appropriateness and the proper magnitude of
such potential adjustments remain unclear given pres-
ently available information. These uncertainties are
particularly acute given the possibility that age and
income biases might offset each other in the case of
pollution-related mortality risk aversion. Furthermore,
the other potential biases discussed above, and sum-
marized in Table 16, add additional uncertainty re-
garding the transferability of WTP estimates from
wage-risk studies to environmental policy and pro-
gram assessments.
Table 15. Estimating Mortality Risk Based <3a Wage- /
Risk Studies: Potential Sources and Ukeiy*Direcii&n of
Bias.
Ftctor
Age
Degree of Risk Aversion
Income
Voluntary vs.
Involuntary
Catastrophic vs.
Protracted Death
Likdy Direction of Bia$ in WTS**
EsSatiee ' :
ttocertaia, perhaps upwrf
Downward _ ',
Uncertain
Downward
Uncertain, perhaps downward
61 Chestnut, 1995; IEc, 1992.
50
-------
7
Results and Uncertainty
This chapter presents a summary of the monetized
benefits of the CAA from 1970 to 1990, compares
these with the corresponding costs, explores some of
the major sources of uncertainty in the benefits esti-
mates, and presents alternative results reflecting di-
verging viewpoints on two key variables: PM-related
mortality valuation and the discount rate.
Monetized economic benefits for the 1970 to 1990
period were derived by applying the unit valuations
discussed in Chapter 6 to the stream of physical ef-
fects estimated by the method documented in Chapter
5. The range of estimates for monetized benefits is
based on the quantified uncertainty associated with
the health and welfare effects estimates and the quan-
tified uncertainty associated with the unit valuations
applied to them. Quantitative estimates of uncertain-
ties in earlier steps of the analysis (i.e., estimation of
compliance costs,62 emissions changes, and air qual-
ity changes) could not be adequately developed and
are therefore not applied in the present study. As a
result, the range of estimates for monetized benefits
presented in this chapter is narrower than would be
expected with a complete accounting of the uncertain-
ties in all analytical components. However, the uncer-
tainties in the estimates of physical effects and unit
values are considered to be large relative to these ear-
lier components. The characterization of the uncer-
tainty surrounding unit valuations is discussed in de-
tail in Appendix I. The characterization of the uncer-
tainty surrounding health and welfare effects estimates,
as well as the characterization of overall uncertainty
surrounding monetized benefits, is discussed below.
Quantified Uncertainty in the
Benefits Analysis
Alternative studies published in the scientific lit-
erature which examine the health or welfare conse-
quences of exposure to a given pollutant often obtain
different estimates of the concentration-response (CR)
relationship between the pollutant and the effect. In
some instances the differences among CR functions
estimated by, or derived from, the various studies are
substantial. In addition to sampling error, these dif-
ferences may reflect actual variability of the concen-
tration-response relationship across locations. Instead
of a single CR coefficient characterizing the relation-
ship between an endpoint and a pollutant in the CR
function, there could be a distribution of CR coeffi-
cients which reflect geographic differences.63 Because
it is not feasible to estimate the CR coefficient for a
given endpoint-pollutant combination in each county
in the nation, however, the national benefits analysis
applies the mean of the distribution of CR coefficients
to each county. This mean is estimated based on the
estimates of CR coefficients reported in the available
studies and the information about the uncertainty of
these estimates, also reported in the studies.
Based on the assumption that for each endpoint-
pollutant combination there is a distribution of CR
coefficients, the Project team used a Monte Carlo ap-
proach to estimate the mean of each distribution and
to characterize the uncertainty surrounding each esti-
mate. For most health and welfare effects, only a single
study is considered. In this case, the best estimate of
the mean of the distribution of CR coefficients is the
reported estimate in the study. The uncertainty sur-
rounding the estimate of the mean CR coefficient is
62 Although compliance cost estimation is primarily of concern to the cost side of this analysis, uncertainty in the estimates for
compliance costs does influence the uncertainty in the benefit estimates because compliance cost changes were used to estimate
changes in macroeconomic conditions which, in turn, influenced the estimated changes in emissions, air quality, and physical effects.
63 Geographic variability may result from differences in lifestyle (e.g., time spent indoors vs outdoors), deposition rates, or other
localized factors which influence exposure of the population to a given atmospheric concentration of the pollutant.
51
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
best characterized by the standard error of the reported
estimate. This yields a normal distribution, centered
at the reported estimate of the mean. If two or more
studies are considered for a given endpoint-pollutant
combination, a normal distribution is derived for each
study, centered at the mean estimate reported in the
study. On each iteration of a Monte Carlo procedure,
a CR coefficient is randomly selected from each of
the normal distributions, and the selected values are
averaged. This yields an estimate of the mean CR co-
efficient for that endpoint-pollutant combination. It-
erating this procedure many times results in a distri-
bution of estimates of the mean CR coefficient.
Each estimate randomly selected from this distri-
bution was evaluated for each county in the nation,
and the results were aggregated into an estimate of
the national incidence of the health or welfare effect.
Through repeated sampling from the distribution of
mean CR coefficients, a distribution of the estimated
change in effect outcomes due to the change in air
quality between the control and no-control scenarios
was generated.
Once a distribution of estimated outcomes was
generated for each health and welfare effect, Monte
Carlo methods were used again to characterize the
overall uncertainty surrounding monetized benefits.
For each health and welfare effect in a set of non-
overlapping effects, an estimated incidence was ran-
domly selected from the distribution of estimated in-
cidences for that endpoint, and a unit value was ran-
domly selected from the corresponding distribution
of unit values, on each iteration of the Monte Carlo
procedure. The estimated monetized benefit for that
endpoint produced on that iteration is the product of
these two factors. Repeating the process many times
generated a distribution of estimated monetized ben-
efits by endpoint. Combining the results for the indi-
vidual endpoints using the Monte Carlo procedure
yielded a distribution of total estimated monetized
benefits for each target year (1975, 1980, 1985 and
1990). This technique enabled a representation of
uncertainty in current scientific and economic opin-
ion in these benefits estimates.
Aggregate Monetized Benefits
For each of the target years of the analysis, the
monetized benefits associated with the different health
and welfare effects for that year must be aggregated.
These aggregate benefits by target year must theft be
aggregated across the entire 1970 to 1990 period of
the study to yield a present discounted value of aggre-
gate benefits for the period. The issues involved in
each stage of aggregation, as well as the results of
aggregation, are presented in this section. (The de-
tailed results for the target years are presented in Ap-
pendix I.)
Table 16. Present Value of 1970 to 1990-Monetized Benefits by Bfldpoiat Category for 48 State
Population (billions of $1990, discounted to 1990; at< 5* percent).
Endpoint Pollutants)
Mortality ' • - , PM
Mortality Pb „ „
Chronic Bronchitis • , "PM
IQ (Lost IQPts. * Children w/IQ<70) Pi>
Hypertension Pi? <• -»«„
Hospital Admissions „ ' ' PM^O3rPb, &COv
Respiratory-Related Symptoms, Restricted PM , O3, N,Q'i & S02
Activity, & Decreased Productivity • "
Soiling Damage PM
Visibility * --- partJculates,
Agriculture (Net Surplus) O3
Present Valae
5th %Ue
$2,369
' $m
$409
$27,1
$77
$27
$123
$6
$38
$11
Mean
$16,632
$1,330
$3,313
,$399
$98
$57
$182
$74
"$S4
$23
95ffl %ile
$40,597
$3,910
- <$1 0,401
$551
' $120
$120
$261
$192
$71
$35
52
-------
Chapter 7: Results and Uncertainty
Table 16 presents monetized benefits for each
quantified and monetized health and welfare endpoint
(or group of endpoints), aggregated from 1970 to 1990.
The mean estimate resulting from the Monte Carlo
simulation is presented, along with the measured cred-
ible range (upper and lower fifth percentiles of the
distribution). Aggregating the stream, of monetized
benefits across years involved compounding the stream
of monetized benefits estimated for each year to the
1990 present value (using a five percent discount rate).
Since the present value estimates combine streams
of benefits from 1970 to 1990, the calculation required
monetized estimates for each year. However, Monte
Carlo modeling was carried out only for the four tar-
get years (1975, 1980, 1985 and 1990). In the inter-
vening years, only a central estimate of benefits was
estimated for each health and welfare endpoint (by
multiplying the central incidence estimate for the given
year by the central estimate of the unit valuation). The
resulting annual benefit estimates provided a tempo-
ral trend of monetized benefits across the period re-
sulting from the annual changes in air quality. They
Table 16 offers a comparison of benefits by health
or welfare endpoint. The effect categories listed in
the table are mutually exclusive, allowing the mon-
etized benefits associated with them to be added. It
should be noted, however, that the listed categories
combine estimates that are not mutually exclusive. To
avoid double counting, care was taken to treat the ben-
efits associated with overlapping effects as alterna-
tive estimates. For example, the "Hospital Admis-
sions" category includes admissions for specific ail-
ments (Pneumonia and COPD) as well as the broader
classification of "all respiratory" ailments. Clearly,
benefits accruing from the first two represent a subset
of the last and adding all three together would result
in an overestimate of total monetized benefits. To avoid
this, the sum of benefits from Pneumonia and COPD
was treated as an alternative to the benefits estimated
for all respiratory ailments (the sum of the first two
was averaged with the third). This issue of double-
counting also arose for two other cases of overlap-
ping health effects, both of which have been combined
into the "Respiratory-Related Symptoms, Restricted
Activity, & Decreased Productivity" category in Table
Table 17." YotalMoHetizei Benefits for ^'StateTopulatiatf {-Present Vataejn billioris-'bf 1990$,- - -
'''... -^)S'V'' ^ * ff f/ ' ''f -f ' ~ , . ^Jf^fJ'J' ,' ' ' ' ' ' •.^ ^s If'ffj1^ J •. J ••
''-d&coaated to 990 at 5 perc'entV> "• '
' -, "" J>> "'^ * ••/,<•• •* ' '
••" ™: ••*•>** , ,-„•-. .- - - -
y; , ' _ s\ ,v;,-
•>>*>-> ,,, , ,,,,„, *,'/?<•,>,' j ^,
;pOTALX?&Hiofe~Qfi990».xdued;eUars)
,",, ' ',~'' ' ' '"•' ' Present Value,,, ,,"'', .".-.'.
:,,;,,,,.,., , - -vv«
^;;;-'^;5tti«te,. ,>„.,.
';*;'; 7 $5^600 ":>-'~- '•'•'*
/. • 'is/lean
"'•'' $22300,.-- •
;7«;95tii «e, '"""'
$49,400,-,,,,,-,
did not, however, characterize the uncertainty associ-
ated with the yearly estimates for intervening years.
In an attempt to capture uncertainty associated with
these estimates, the Project Team relied on the ratios
of the 5th percentile to the mean and the 95th percen-
tile to the mean in the target years. In general, these
ratios were fairly constant across the target years, for
a given endpoint. The ratios were interpolated between
the target years, yielding ratios for the intervening
years. Multiplying the ratios for each intervening year
by the central estimate generated for that year pro-
vided estimates of the 5th and 95th percentiles, which
were used to characterize uncertainty about the cen-
tral estimate. Thus, the present value of the stream of
benefits, including the credible range estimates, could
be computed.
16. First, acute bronchitis was treated as an alterna-
tive (i.e., averaged with) the combination of upper and
lower respiratory symptoms, since their definitions of
symptoms overlap. Second, various estimates of re-
stricted activity, with different degrees of severity,
were combined into a single benefit category.
Table 17 reports the estimated total national mon-
etized benefits attributed in this analysis to the CAA
from 1970 to 1990. The benefits, valued in 1990 dol-
lars, range from $5.6 to $49.4 trillion with a central
estimate of $22.2 trillion. The Monte Carlo technique
was used to aggregate monetized benefits across end-
points. For each of several thousand iterations, a ran-
dom draw of the monetized benefits for each endpoint
was selected from the distributions summarized in
53
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table 16 and the individual endpoint estimates were
then summed. This resulted in the distribution of total
national monetized benefits reported above.64
The temporal pattern of benefits during the 1970
to 1990 period is related to the difference in emis-
sions between the control and no-control scenarios and
is magnified by population growth during that period.
As illustrated by Figure 18, quantified annual ben-
efits increased steadily during the study period, with
the greatest increases occurring during the late 1970s.
The mean estimate of quantified annual benefits grew
from 355 billion dollars in 1975 (expressed as infla-
tion-adjusted 1990 dollars) to 930 billion dollars hi
1980,1,155 billion dollars in 1985, and 1,248 billion
dollars in 1990.
Figure 19 depicts the distribution of monetized
benefits for 1990 (similar distributions were gener-
ated for other years in the analysis period). The solid
vertical bars in the figure represent the relative fre-
quency of a given result in the 1990 Monte Carlo
analysis. The largest bar, located above the "<$1,000",
indicates that more Monte Carlo iterations generated
monetized benefits of $900 billion to $1 trillion than
in any other $100 billion range bin, making this the
modal bin. The expected value of the estimate for to-
tal monetized benefit for 1990 (i.e., the mean of the
distribution) is $1.25 trillion. The ninety percent con-
fidence interval, a summary description of the spread
of a distribution, is also noted hi the figure.
Figure 19. Distribution of 1990 Monetized Benefits of
CAA (in billions of 1990 dollars).
Distribution SummarvfSBilllonjrt
5th percentlle = $329
mean = $1,250
95th percentlle = $2,760
v v v R R R R 5
5th percentile T Mean 95th peicentllef
Total Monetary Benefits (S Billions)
Figure 18. Monte Carlo Simulation Model Results for
Target Years
S3.000T
"i"
5 $2,500-
§. $2,000-
a
g $1,500-
o
CD
•a $1,000-
3
tg
$500-
SO-
(in billions of 1990 dollars).
•1 Mean
'3;
•--:-'-•
<°*M
^ Mean
4 Elh%
<
^ %
!>°
4 85lh%
4] Mean
e| Elh%
•— i
Vti
^
_^
iC<
1
4 95th%
4 Mean
^ 51h%
4 «h%
1975 ' 1980 1985 1990
On initial inspection, the estimated $1.25 trillion
value for monetized benefits hi 1990 may seem im-
plausibly large, even though 1990 is the year in which
the differences between outcomes under the control
and no-control scenarios are at their most extreme.
The plausibility of this estimate may seem particu-
larly questionable to some if one considers that the
$1.25 trillion value for 1990 is over five percent of
the estimated $22.8 trillion value for total 1990 assets
of households and nonprofit organizations. Consid-
ered from this perspective, $1.25 trillion may seem to
represent a large share of total wealth, and some might
question whether Americans would really be willing
to pay this much money for the reductions in risk
achieved by the Clean Air Act and related programs,
even if the risk in question involves premature death.
However, in the end it is clear that such comparisons
are overly simplistic and uninformative because they
ignore the magnitude and nature of the welfare change
being measured.
First, with respect to the magnitude of the differ-
ence in estimated social welfare under the two sce-
narios, it is important to recognize how severe air qual-
ity conditions and health risks would be under the
hypothetical no-control scenario. Focusing on ambi-
ent paniculate matter, the pollutant responsible for the
vast majority of the estimated monetary benefits, a
comparison of the estimated annual mean concentra-
tions of total suspended particulates (TSP) projected
in the U.S. under the no-control scenario with esti-
64 Comparing Tables 16 and 17, it can be seen that the sum of benefits across endpoints at a given percentile level does not result
in the total monetized benefits estimate at the same percentile level in Table 17. For example, if the fifth percentile benefits of the
endpoints shown in Tkble 16 were added, the resulting total would be substantially less than $5.6 trillion, the fifth percentile value of
the distribution of aggregate monetized benefits reported in Table 17. This is because the various health and welfare effects are treated
as stochastically independent, so that the probability that the aggregate monetized benefit is less than or equal to the sum of the
separate five percentile values is substantially less than five percent.
54
-------
Chapter 7: Results and Uncertainty
mated annual mean TSP concentrations in other parts
of the world65 indicates that in 1990—
• 60 metropolitan areas in the U.S. would have
had higher TSP concentrations than Moscow,
Russia
• 7 metropolitan areas would be worse than
Bangkok, Thailand
• 6 metropolitan areas would be worse than
Bombay, India
• 2 metropolitan areas would be worse than Ma-
nila, Philippines
• One metropolitan area would be worse than
Delhi, India (one of the most polluted cities
in the world)
Under the control scenario, TSP levels in only 3
metropolitan areas were projected to exceed those in
Moscow, and none exceeded levels found in the other
foreign cities listed above. The principal reason air
quality conditions are so poor under the no-control
scenario is that air pollution control requirements re-
main fixed at their 1970 levels of scope and stringency
while total economic activity, including polluting ac-
tivity, grows by 70 percent and population grows by
22.3 percent between 1970 and 1990. Under the se-
vere air quality conditions projected throughout the
U.S. in 1990 under the no-control case, an additional
205,000 people would be projected to die prematurely
due to the effects of particulate matter, lead, and other
criteria pollutants. This represents a very large increase
in the risk of premature mortality. Since the estimate
that the average loss of life for those who actually
succumb to PM exposure related health effects is ap-
proximately 14 years, and life-shortening due to lead
exposure is even greater, it is no longer surprising that
the estimated value of avoiding these severe condi-
tions is so high.
Second, with respect to the nature of the welfare
change reflected in the monetized benefit estimate,
the concern about the effects of limited budgets con-
straining Americans' collective ability to pay to avoid
these severe no-control scenario conditions is mis-
placed. In reality, what society actually had to pay to
avoid these conditions is measured on the cost side of
the analysis, which sums up the total expenditures
made by manufacturers and others to achieve these
air pollution reductions. The most reasonable estimate
of the value Americans place on avoiding those se-
vere no-control scenario conditions, however, is pro-
vided by measuring the amount of compensation
Americans would have demanded from polluting com-
panies and others to accept, willingly, all of that extra
pollution and its associated risks of premature death.
Under this concept of welfare change measurement,
there is no inherent limit on the amount of money citi-
zens would demand from companies to accept their
pollution and so individual personal wealth does not
constrain this value.
The monetized benefit estimate presented in this
study, therefore, does not necessarily represent an at-
tempt to mirror what Americans would pay out of their
own pockets to reduce air pollution from levels they
never experienced; rather, it provides an estimate of
the value Americans place on the protection they re-
ceived against the dire air pollution conditions which
might have prevailed in the absence of the 1970 and
1977 Glean Air Acts and related programs, "viewed
from this perspective, the estimated monetized ben-
efits presented herein appear entirely plausible.
Comparison of Monetized
Benefits and Costs
Table 18 presents summary quantitative results for
the retrospective assessment. Annual results are pre-
sented for four individual years, with all dollar fig-
ures expressed as inflation-adjusted 1990 dollars. The
final column sums the stream of costs and benefits
from 1970 to 1990, discounted (i.e., compounded) to
1990 at five percent. "Monetized benefits" indicate
both the mean of the Monte Carlo analysis and the
credible range. "Net Benefits" are mean monetized
benefits less annualized costs for each year. The table
also notes the benefit/cost ratios implied by the ben-
efit ranges. The distribution of benefits changes little
(except in scale) from year to year: The mean esti-
mate is somewhat greater than twice the fifth percen-
tile estimate, and the ninety-fifth percentile estimate
is somewhat less than twice the mean estimate. The
distribution shape changes little across years because
the sources of uncertainty (i.e., CR functions and eco-
nomic valuations) and their characterizations are un-
changed from year to year. Some variability is induced
by changes in relative pollutant concentrations over
time, which then change the relative impact of indi-
vidual CR functions.
Several measures of "cost" are available for use
in this analysis (see Chapter 2). The Project Team
65 "Urban Air Pollution in Megacities of the World," UNEP/WHO, 1992a, Published by the World Health Organization and
United Nations Environment Program, Blackwell Publishers, Oxford, England, 1992. "City Air Quality Trends," UNEP/WHO, 1992b,
Published by the United Nations Environment Program, Nairobi, Kenya, 1992.
55
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table 18, Quantified Uncertainty Ranges for Monetized
Annual Benefits and Benefit/Cost Ratios; 19?Q"499G:(ui
billions of 1990-value dollars).
Monetized Benefits
5th peroentik
Mean estimate
95th percentile
Annualized Costs (5%)
Net Benefits
Mean benefits- Costs
Benefit/Cost ratio
5th percentile
Mean estimate
95th oercentile
1975
87
355
799
14
341
6/1
25/1
57/1
1980
235 '
930
2.063 •
21
« '-.v.
909-'
44/V"*
98/1 '
J985
<> j
293
1,155
2,569
25
1,130
ill'
itivT
1990
329
-1,248-
2;7fi2v"
26
.„„,.,„
"•11220-
w.
, PV "
's »•>.,.
5,600'
* 49;4Q_'_'
•3MOO-V
'' ii/i I
"""42/J;,,,
•"94/r
PVsi990 present value reflecting compounding of
-------
Chapter 7: Results and Uncertainty
In this sensitivity analysis, all the inputs to the
Monte Carlo uncertainty analysis are held constant
(at their mean values), allowing only one variable ~
for example, the economic valuation of mortality —
to vary across the range of that variable's uncertainty.
The sensitivity analysis then isolates how this single
source of uncertainty contributes to the total measured
uncertainty in estimated aggregate benefits. The first
uncertainty bar represents the credible range associ-
ated with the total monetized benefits of the Clean
Air Act, as reported above. This captures the multiple
uncertainties in the quantified benefits estimation. The
rest of the uncertainty bars represent the quantified
uncertainty ranges generated by single variables. As
shown in Figure 20, the most important contributors
to aggregate quantified uncertainty are mortality valu-
ation and incidence, followed by chronic bronchitis
valuation and incidence.
Alternative Results
The primary results of this analysis, including
aggregate cost and benefit estimates and the uncer-
tainty associated with them, are presented and dis-
cussed above. However, although the range of net
benefit estimates presented reflects uncertainty in
many important elements of the analysis, there are
two key variables which require further discussion and
analysis: PM-related mortality valuation and the dis-
count rate. This additional treatment is necessary be-
cause reasonable people may disagree with the Project
Team's methodological choices for these two vari-
ables, and these choices might be considered ex ante
to significantly influence the results of the study. The
purpose of this section, therefore, is to present alter-
native quantitative results which reflect, separately,
(1) an alternative approach to valuation of premature
mortality associated with particulate matter exposure,
and (2) alternative values for the discount rate used to
adjust the monetary values of effects occurring in vari-
ous years to a particular reference year (i.e., 1990).
PM Mortality Valuation Based on Life-
Years Lost
The primary analytical results presented earlier
in this chapter assign the same economic value to in-
cidences of premature mortality regardless of the age
and health status of those affected. Although this has
been the traditional practice for benefit-cost studies
conducted within the Agency, this may not be the most
appropriate method for valuation of premature mor-
tality .caused by PM exposure. Some short-term PM
exposure studies suggest that a significantly dispro-
portionate share of PM-related premature mortality
occurs among persons 65 years of age or older. Com-
bining standard life expectancy tables with the lim-
ited available data on age-specific incidence allows
crude approximations of the number of life-years lost
by those who die prematurely as a result of exposure
to PM or, alternatively, the changes in age-specific
life expectancy of those who are exposed to PM.
The ability to estimate, however crudely, changes
in age-specific life expectancy raises the issue of
whether available measures of the economic value of
mortality risk reduction can, and should, be adapted
to measure the value of specific numbers of life-years
saved.66 Although the Agency has on occasion per-
formed sensitivity calculations which adjust mortal-
ity values for those over age 65, the Agency is skepti-
cal that the current state of knowledge and available
analytical tools support using a life-years lost approach
or any other approach which assigns different risk re-
duction values to people of different ages or circum-
stances . This skepticism is mirrored hi the OMB guid-
ance on implementing Executive Order 12866 per-
taining to economic analysis methods, which states
on page 31:
While there are theoretical advantages to
using a value of statistical life-year-extended
approach, current research does not provide
a definitive way of developing estimates of
VSLY that are sensitive to such factors as
current age, latency of effect, life years
remaining, and social valuation of different
risk reductions. In lieu of such information,
there are several options for deriving the
value of a life-year saved from an estimate of
the value of life, but each of these methods
has drawbacks. One approach is to use results
from the wage compensation literature (which
focuses on the effect of age on WTP to avoid
risk of occupational fatality). However, these
results may not be appropriate for other types
of risks. Another approach is to annualize the
VSL using an appropriate rate of discount and
the average life years remaining. This
approach does not provide an independent
estimate of VSLY; it simply rescales the VSL
estimate. Agencies should consider providing
estimates of both VSL and VSLY, while
recognizing the developing state of knowledge
in this area.
While the Agency continues to prefer an approach
which makes no valuation distinctions based on age
or other characteristics of the affected population, al-
ternative results based on a VSLY approach are pre-
66 This issue was extensively discussed during the Science Advisory Board Council review of drafts of the present study. The
Council suggested it would be reasonable and appropriate to show PM mortality benefit estimates based on value of statistical life-
years (VSLY) saved as well as the value of statistical life (VSL) approach traditionally applied by the Agency to all incidences of
premature mortality.
__
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
sented below. The method used to develop VSLY es-
timates is described briefly in Chapter 6 and in more
detail in Appendix I.
Table 19 summarizes and compares the results of
the VSL and VSLY approaches. Estimated 1970 to
1990 benefits from PM-related mortality alone and
total assessment benefits are reported, along with to-
tal compliance costs for the same period, in 1990 dol-
lars discounted to 1990 at five percent. The results
indicate that the choice of valuation methodology sig-
nificantly affects the estimated monetized value of
historical reductions in air pollution-related prema-
ture mortality. However, the downward adjustment
which would result from applying a VSLY approach
in lieu of a VSL approach does not change the basic
outcome of this study, viz. the estimated monetized
benefits of the historical CAA substantially exceed
the historical costs of compliance.
1970 toward 1990 (see Table 18 above), benefit cost
ratios decline as the discount rate increases (because
earlier periods are given greater weight). Overall, the
results of the benefit-cost assessment appear to be
generally insensitive to the choice of discount rate.
tes on
l9Q, dollars).
Atin»»lf zed-Costs ""''''r
-,$&, - -5,%.,
-19-.2;,-. 22.2 ' '25,8
Q'A-"-" • -0.5 '*G3
' '18.8 21.7, 25.1 '
•^Table 19. Alternative Mortality Benefits
Estimates for 1970 to 1990 (in trillionsjof) 990
dollars, discounted at 5 percent) Compared to
Total 1970 to 1990 Compliance Costs/
fffit- P.cHmatirm
Statistical life method ($4.8M/case)
Life-years lost method ($293>0/year)
Total compliance cost _
Tnt
16>6 18!o'
9.1 "1&,1°
„—!. ^ "0.5
Alternative Discount Rates
In some instances, the choice of discount rate can
have an important effect on the results of a benefit-
cost analysis; particularly for those analyses with rela-
tively long time horizons for costs and/or benefits. In
this assessment, the discount rate affects only four
factors: IQ-related benefits estimates (especially esti-
mates of changes in discounted lifetime income), life-
time income losses due to other health effects (e.g.,
stroke), annualized costs (i.e., amortized capital ex-
penditures), and compounding of all costs and ben-
efits to 1990. Table 20 summarizes the effect of alter-
native discount rates on the "best estimate" results of
this analysis. Because monetized benefits exceed costs
for all years in the analysis period, net benefits in-
crease as the discount rate increases. Because the an-
nual benefit/cost ratio increases as one moves from
58
-------
Chapter 7: Results and Uncertainty
59
-------
-------
Appendix A: Cost and Macroeconomic Modeling
Introduction
The purpose of this appendix is to describe in de-
tail the estimation of direct compliance costs associ-
ated with the CAA and the effect of those expendi-
tures on U.S. economic conditions from 1970 to 1990.
The first section of this appendix describes the dy-
namic, general equilibrium macroeconomic model
used to examine economy-wide effects. Two broad
categories of models were considered for use in the
assessment: Macroeconomic forecasting models (e.g.,
the Data Resources Inc. model of the U.S. economy),
and general equilibrium models (e.g., Hazilla and
Kopp [1990], and Jorgenson and Wilcoxen [1990a]).
The project team selected the Jorgenson-Wilcoxen (J/
W) general equilibrium model of the United States
for this analysis (Jorgenson and Wilcoxen [1990a]).
There are two main reasons for choosing a dynamic
general equilibrium approach: To capture both the
direct and indirect economic effects of environmen-
tal regulation, and to capture the long-run dynamics
of the adjustment of the economy. The general equi-
librium framework enabled the project team to assess
shifts in economic activity between industries, includ-
ing changes in distributions of labor, capital, and other
production factors within the economy, and changes
in the distribution of goods and services.
The second section describes the data sources for
direct compliance expenditures and presents estimates
of historical air pollution control expenditures. These
estimates are derived primarily from EPA's 1990 re-
port entitled "Environmental Investments: The Cost
of a Clean Environment"l (hereafter referred to as Cost
of Clean). Specific adjustments to the Cost of Clean
stationary source and mobile source O&M data needed
to adapt these data for use in the present study are
also described. These adjusted expenditure estimates
represent the compliance cost data used as inputs to
1 Environmental Investments: The Cost of a Clean Environment, Report of the Administrator of the Environmental Protection
Agency to the Congress of the United States, EPA-230-11-90-083, November 1990.
2 In the present study, both benefits and costs are driven by of the same macroeconomic projections from the Jorgenson/
Wilcoxen model, to ensure that the estimates are based on a consistent set of economic assumptions.
_
the JAV model to determine macroeconomic effects.
The final section presents a summary of the di-
rect expenditure data, presents direct costs in a form
that can be compared to the benefits estimates found
elsewhere in the study, and discusses indirect effects
arising from compliance expenditures estimated by
the macroeconomic model. The indirect effects re-
ported by the model are sectoral impacts and changes
in aggregate measures of economic activity such as
household consumption and gross national product.
These indirect effects are second-order impacts of
compliance expenditures — a parallel modeling ex-
ercise to estimate second-order economic impacts aris-
ing from the benefits of compliance (e.g., increased
output as a result of improved longevity or fewer
workdays lost as a result of non-fatal heart attacks)
has not been attempted.
Macroeconomic Modeling
EPA analyses of the costs of environmental regu-
lations typically quantify the direct costs of pollution
abatement equipment and related operating and main-
tenance expenses. However, this approach does not
fully account for all of the broader economic conse-
quences of reallocating resources to the production
and use of pollution abatement equipment. A general
equilibrium, macroeconomic model could, in theory,
capture the complex interactions between sectors in
the economy and assess the full economic cost of air
pollution control. This would be particularly useful
for assessing regulations that may produce significant
interaction effects between markets. Another advan-
tage of a general equilibrium, macroeconomic frame-
work is that it is internally consistent. The consistency
of sectoral forecasts with realistic projections of U.S.
economic growth is ensured since they are estimated
within the context of a single model.2 This contrasts
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
with typical EPA analyses that compile cost estimates
from disparate sectoral and partial equilibrium mod-
els.
The economic effects of the CAA may be over-
or underestimated, if general equilibrium effects are
ignored, to the extent that sectors not directly regu-
lated are affected. For example, it is well known that
the CAA imposed significant direct costs on the en-
ergy industry. Economic sectors not directly regulated
will nonetheless be affected by changes in energy
prices. However, an examination of the broader ef-
fects of the CAA on the entire economy might reveal
that the CAA also led to more rapid technological
development and market penetration of environmen-
tally "clean" renewable sources of energy (e.g., pho-
tovoltaics). These effects would partially offset ad-
verse effects on the energy industry, and lead to a dif-
ferent estimate of the total economic cost to society
of the CAA.
The significance of general equilibrium effects in
the context of any particular analysis is an empirical
question. Kokoski and Smith (1987) used a comput-
able general equilibrium model to demonstrate that
partial-equilibrium welfare measures can offer rea-
sonable approximations of the true welfare changes
for large exogenous changes. In contrast, the results
of Jorgenson and Wilcoxen (1990a) and Hazilla and
Kopp (1990) suggest that total pollution abatement in
the U.S. has been a major claimant on productive re-
sources, and the effect on long-run economic growth
may be significant. Again, such conclusions must be
considered in light of the limitations of general equi-
librium models.
Choice of Macroeconomic Model
The adequacy of any model or modeling approach
must be judged in light of the policy questions being
asked. One goal of the present study is to assess the
effects of clean air regulations on macroeconomic
activity. Two broad categories of macroeconomic
models were considered for use hi the assessment:
short run, Keynesian models and long-run, general
equilibrium models.
Recognizing that structural differences exist be-
tween the models, one needs to focus in on the par-
ticular questions that should be answered with any
particular model. The Congressional Budget Office
(1990) noted:
"Both the [Data Resources Incorporated] DRI
and the IPCAEO models show relatively
limited possibilities for increasing energy
efficiency and substituting other goods for
energy in the short run... Both models focus
primarily on short-term responses to higher
energy prices, and neither is very good at
examining how the structure of the economy
could change in response to changing energy
prices. The [Jorgenson-Wilcoxen] model
completes this part of the picture..."3
One strategy for assessing the macroeconomic
effects of the CAA would be to use a DRI-type model
in conjunction with the Jorgenson-Wilcoxen model
to assess both the long-term effects and the short-run
transitions, in much the same way that the Congres-
sional Budget Office used these models to assess the
effects of carbon taxes. However, because of signifi-
cant difficulties in trying to implement the DRI model
in a meaningful way, the project team chose to focus
on the long-run effects of the CAA. Structural changes
(e.g., changes hi employment in the coal sector due to
the CAA) can be identified with the Jorgenson-
Wilcoxen model.
Overview of the Jorgenson-
Wilcoxen Model
The discussion below focuses on those character-
istics of the Jorgenson-Wilcoxen model that have
important implications for its use in the assessment
of environmental regulations (see Table A-l). The J/
W model is a detailed dynamic general equilibrium
model of the U.S. economy designed for medium run
analysis of regulatory and tax policy (Jorgenson and
Wilcoxen [1990a]). It provides projections of key
macroeconomic variables, such as GNP and aggre-
gate consumption, as well as energy flows between
economic sectors. As a result, the model is particu-
larly useful for examining how the structure of the
economy could change hi response to changes in re-
3 The Congressional Budget Office report (1990) refers to an older (1981) version of the Jorgenson model, not the current
(1988) version. The approach to long-run dynamics differs between the two models. The newer Jorgenson-Wilcoxen model contains
both the capital accumulation equation and the capital asset pricing equation. The 1981 version of the model contained only the
capital accumulation equation.
-------
Appendix A: Cost and Macroeconomic Modeling
Table A-l
s of*
macroeponomic model of the U.S. economy.
p estfmated,«$Sn'g historic,,;,,
A-Free'm'obiHty t£f%, siagle-ty-pejof capital-arid-
-.lab^tf b'elween inctoslrfejs ^ : - ' ,.,, „---
' ' "*' * '
<*',-• 'lUgohms representation of savings -aad-. -
-,v*«<> -invesftnertt. - '" -/ -,« * '''-7,,--;-,
• — ,!•, 'Bndogerjoiis model of techniea! change;- --, „ 0
;-';')'; ,, ,jand^remgloy meat ,-o'r the costs *of mo vfiig
--•>,; ,,e^ kal from- one Industry '
source prices. For the purpose of this study, it has five
key features: a detailed treatment of production and
consumption, parameters estimated econometrically
from historical data, an endogenous model of techni-
cal change, a rigorous representation of saving and
investment, and free mobility of labor and capital be-
tween industries.
The first two features, industry and consumer de-
tail and econometric estimation, allow the model to
capture the effects of the CAA at each point in time
for given levels of technology and the size of the
economy's capital stock. A detailed treatment of pro-
duction and consumption is important because the
principal effects of the Clean Air Act fell most heavily
on a handful of industries. The J/W model divides
total U.S. production into 35 industries which allows
the primary economic effects of the CAA to be cap-
tured. Econometric estimation is equally important
because it ensures that the behavior of households and
firms in the model is consistent with the historical
record.
The model's second two features —its represen-
tations of technical change and capital accumulation—
complement the model's intratemporal features by
providing specific information on how the Act affected
technical change and the accumulation of capital.
Many analyses of environmental regulations overlook
or ignore intertemporal effects but these effects can
be very important. Jorgenson and Wilcoxen (1990a)
suggests that the largest cost of all U.S. environmen-
tal regulations together was that,the regulations re-
duced the rate of capital accumulation.
The model's last feature, free mobility of a single
type of capital and a single type of labor, is important
because it limits the model's ability to measure the
short run costs of changes in policy. J/W is a full-
employment model that describes the long-run dynam-
ics of transitions from one equilibrium to another.
Capital and labor are both assumed to be freely mo-
bile between sectors (that is, they can be moved from
one industry to another at zero cost) and to be fully
used at all times. Over the medium to long run, this is
a reasonable assumption, but in the short run it is too
optimistic. In particular, the model will understate the
short run costs of a change in policy because it does
not capture unemployment, underemployment, or the
costs of moving capital from one industry to another.
A single rate of return on capital exists that efficiently
allocates the capital in each period among sectors.
Similarly, a single equilibrium wage rate allocates
labor throughout the economy.
Structure of the Jorgenson-Wilcoxen
Model
The J/W model assesses abroad array of economic
effects of environmental regulations. Direct costs are
captured as increased expenditures on factors of pro-
duction —capital, labor, energy and materials— that
the various industries must make to comply with the
regulations, as well as additional out-of-pocket ex-
penditures that consumers must make. Indirect costs
are captured as general equilibrium effects that occur
throughout the economy as the prices of factors of
production change (e.g., energy prices). Also, the rate
of technological change can respond to changes in the
prices of factors of production, causing changes in
productivity (Jorgenson and Fraumeni, 1981).
The model is divided into four major sectors: the
business, household, government, and rest-of-the-
world sectors. The business sector is further subdi-
vided into 35 industries (see Table A-2).4 Each sector
produces a primary product, and some produce sec-
ondary products. These outputs serve as inputs to the
production processes of the other industries, are used
for investment, satisfy final demands by the house-
hold and government sectors, and are exported. The
model also allows for imports from the rest of the
world.
The 35 industries roughly correspond to a two-digit SIC code classification scheme.
—-
-------
The Benefits and Costs of the Clean Air Act, .1970 to 1990
Table A-2. Definitions of Industries Within
the J/W Model. , ' ,
Industry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
n ' ' " ' •
Agriculture, forestry, and -*> '
fisheries
Metal mining
Coalmining ' * f
Crudepetroleum and natural- gas
NonmetalUc mineral mining
Construction
Food and kindred products ' "
Tobacco manufacturers
Textile mill products
Ap parel and other textile "" „„ *
products
Lumber and wood products
Furniture and fixtures f - ,
Paper and allied products "" ,,
Printing and publishing
Chemicals and allied products
Petroleum refining
Rubber and plastic products
Leather and leather products
Stone, clay, and glass products
Primary metals
Fabricated metal p roduets ->?
M achinery , except electrical
Electrical machinery
Motor vehicles
Other transportation equipment
Instruments , ^f"
Miscellaneous manufacturing
Transportation and warehousing
Communication.
Electric utilities
Gas Utilities
Trade
Finance, insurance, and real
estate
Other services
Government enterprises
"t~
»,
»
"
<
The Business Sector
The model of producer behavior allocates the
value of output of each industry among the inputs of
the 35 commodity groups, capital services, labor ser-
vices, and noncompeting imports. Output supply and
factor demands of each sector are modeled as the re-
sults of choices made by wealth maximizing, price
taking firms which are subject to technological con-
straints. Firms have perfect foresight of all future
prices and interest rates. Production technologies are
represented by econometrically estimated cost func-
tions that fully capture factor substitution possibili-
ties and industry-level biased technological change.
Capital and energy are specified separately in the
factor demand functions of each industry. The ability
of the model to estimate the degree of substitutability
between factor inputs facilitates the assessment of the
effect of environmental regulations. A high degree of
substitutability between inputs implies that the cost
of environmental regulation is low, while a low de-
gree of substitutability implies high costs of environ-
mental regulation. Also, different types of regulations
lead to different responses on the part of producers.
Some regulations require the use of specific types of
equipment. Others regulations restrict the use of par-
ticular factor inputs; for example, through restrictions
on the combustion of certain types of fuels. Both of
these effects can change the rate of productivity growth
in an industry through changes in factor prices.
The Household Sector
In the model of consumer behavior, consumer
choices between labor and leisure and between con-
sumption and saving are determined. A system of in-
dividual, demographically defined household demand
functions are also econometrically estimated. House-
hold consumption is modeled as a three stage optimi-
zation process. In the first stage households allocate
lifetime wealth to full consumption in current and fu-
ture time periods to maximize intertemporal utility.
Lifetime wealth includes financial wealth, discounted
labor income, and the imputed value of leisure. House-
holds have perfect foresight of future prices and in-
terest rates. In the second stage, for each time period
full consumption is allocated between goods and ser-
vices and leisure to maximize intratemporal utility.
This yields an allocation of a household's time en-
dowment between the labor market (giving rise to la-
bor supply and labor income) and leisure time and
demands for goods and services. In the third stage,
personal consumption expenditures are allocated
among capital, labor, noncompeting imports and the
outputs of the 35 production sectors to maximize a
subutility function for goods consumption. As with
the business sector, substitution possibilities exist in
consumption decisions. The model's flexibility en-
ables it to capture the substitution of nonpolluting
products for polluting ones that may be induced by
environmental regulations. Towards this end, pur-
chases of energy and capital services by households
are specified separately within the consumer demand
functions for individual commodities.
A-4
-------
Appendix A: Cost and Macroeconomic Modeling
It is important to be clear regarding the notions of
labor supply and demand within the J/W model, and
what is meant by "employment" throughout this re-
port. Labor demands and supplies are represented as
quality-adjusted hours denominated in constant dol-
lars. The labor market clears in each period; the quan-
tity of labor services offered by households is absorbed
fully by the economy's producing sectors. However,
inferences regarding the number of persons employed
require information on labor quality and work-hours
per person over time and across simulations. Neither
of these are explicitly modeled.
The Government Sector
The behavior of government is constrained by
exogenously specified budget deficits. Government
tax revenues are determined by exogenously speci-
fied tax rates applied to appropriate transactions in
the business and household sectors. Levels of eco-
nomic activity in these sectors are endogenously de-
termined. Capital income from government enterprises
(determined endogenously), and nontax receipts
(given exogenously), are added to tax revenues to
obtain total government revenues. Government expen-
ditures adjust to satisfy the exogenous budget deficit
constraint. ,
The Rest-of-the-World Sector
The current account balance is exogenous, limit-
ing the usefulness of the model to assess trade com-
petitiveness effects. Imports are treated as imperfect
substitutes for similar domestic commodities and com-
pete on price. Export demands are functions of for-
eign incomes and ratios of commodity prices in U.S.
currency to the exchange rate. Import prices, foreign
incomes, and tariff policies are exogenously speci-
fied. Foreign prices of U.S. exports are determined
endogenously by domestic prices and the exchange
rate. The exchange rate adjusts to satisfy the exog-
enous constraint on net exports.
Environmental Regulation, Investment,
and Capital Formation
Environmental regulations have several important
effects on capital formation. At the most obvious level,
regulations often require investment in specific pieces
of pollution abatement equipment. If the economy's
pool of savings were essentially fixed, the need to in-
vest in abatement equipment would reduce, or crowd
out, investment in other kinds of capital on a dollar
for dollar basis. On the other hand, if the supply of
savings were very elastic then abatement investments
might not crowd out other investment at all. In the J/
W model, both the current account and government
budget deficits are fixed exogenously so any change
in the supply of funds for domestic investment must
come from a change in domestic savings. Because
households choose consumption, and hence savings,
to maximize a lifetime utility function, domestic sav-
ings will be somewhat elastic. Thus, abatement in-
vestment will crowd out other investment, although
not on a dollar for dollar basis.
The J/W assumption that the current account does
not change as a result of environmental regulation is
probably unrealistic, but it is not at all clear that this
biases the crowding out effects in any particular di-
rection. By itself, the need to invest in abatement capi-
tal would tend to raise U.S. interest rates and draw in
foreign savings. To the extent this occurred, crowd-
ing out would be reduced. At the same tune, how-
ever, regulation reduces the profitability of domestic
firms. This effect would tend to lower the return on
domestic assets, leading to a reduced supply of for-
eign savings which would exacerbate crowding out.
Which effect dominates is an empirical question be-
yond the scope of this study.
In additional to crowding out ordinary investment,
environmental regulation also has a more subtle ef-
fect on the rate of capital formation. Regulations raise
the prices of intermediate goods used to produce new
capital. This leads to a reduction in the number of capi-
tal goods which can be purchased with a given pool
of savings. This is not crowding out hi the usual sense
of the term, but it is an important means by which
regulation reduces capital formation.5
The General Equilibrium
The J/W framework contains intertemporal and
intratemppral models (Jorgenson and Wilcoxen
[19906]). In any particular time period, all markets
clear. This market clearing process occurs in response
to any changes in the levels of variables that are speci-
5 Wilcoxen (1988) suggests that environmental regulation may actually lead to a "crowding in" phenomenon. Wilcoxen
examined the effects of regulation at the firm level, and introduced costs into the model related to the installation of capital. He found
that when firms shut down their plants to install environmental capital, they take account of the adjustment costs and often concur-
rently replace other older capital equipment. This effect, however, is not captured in the current version of the Jorgenson-Wilcoxen
model.
A-5
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
fied exogenously to the model. The interactions among
sectors determine, for each period, aggregate domes-
tic output, capital accumulation, employment, the
composition of output, the allocation of output across
different household types, and other variables.
The model also produces an intertemporal equi-
librium path from the initial conditions at the start of
the simulation to the stationary state. (A stationary
solution for the model is obtained by merging the
intertemporal and intratemporal models.) The dynam-
ics of the J/W model have two elements: An accumu-
lation equation for capital, and a capital asset pricing
equation. Changes in exogenous variables cause sev-
eral adjustments to occur within the model. First, the
single stock of capital is efficiently allocated among
all sectors, including the household sector. Capital is
assumed to be perfectly malleable and mpbile among
sectors, so that the price of capital services in each
sector is proportional to a single capital service price
for the economy as a whole. The value of capital ser-
vices is equal to capital income. The supply of capital
available in each period is the result of past invest-
ment, i.e., capital at the end of each period is a func-
tion of investment during the period and capital at the
beginning of the period. This capital accumulation
equation is backward-looking and captures the effect
of investments in all past periods on the capital avail-
able in the current period..
The capital asset pricing equation specifies the
price of capital services in terms of the price of in-
vestment goods at the beginning and end of each pe-
riod, the rate of return to capital for the economy as a
whole, the rate of depreciation, and variables describ-
ing the tax structure for income from capital. The cur-
rent price of investment goods incorporates an assump-
tion of perfect foresight or rational expectations. Un-
der this assumption, the price of investment goods in
every period is based on expectations of future capi-
tal service prices and discount rates that are fulfilled
by the solution of the model. This equation for the
investment goods price in each time period is forward-
looking.6
One way to characterize the J/W model —or any
other neoclassical growth model— is that the short-
run supply of capital is perfectly inelastic, since it is
completely determined by past investment. However,
the supply of capital is perfectly elastic in the long
run. The capital stock adjusts to the tune endowment,
while the rate of return depends only on the
intertemporal preferences of the household sector.
A predetermined amount of technical progress
also takes place that serves to lower the cost of sectoral
production. Finally, the quality of labor is enhanced,
giving rise to higher productivity and lower costs of
production.
Given all of these changes, the model solves for a
new price vector and attains a new general equilib-
rium. Across all tune periods, the model solves for
the time paths of the capital stock, household con-
sumption, and prices. The outcomes represent a gen-
eral equilibrium in all time periods and in all markets
covered by the J/W model.
Configuration of the No-control
Scenario
One of the difficulties in describing the no-con-
trol scenario is ascertaining how much environmen-
tal regulation would have been initiated by state and
local governments in the absence of a federal program.
It may reasonably be argued that many state and local
governments would have initiated their own control
programs in the absence of a federal role. This view
is further supported by the fact that many states and
localities have, in fact, issued rules and ordinances
which are significantly more stringent and encompass-
ing than federal minimum requirements. However, it
may also be argued that the federal CAA has moti-
vated a substantial number of stringent state and local
control programs.
Specifying the range and stringency of state and
local programs that would have occurred in the ab-
sence of the federal CAA would be almost entirely
speculative. For example, factors which would com-
plicate developing assumptions about stringency and
scope of unilateral state and local programs include:
(i) die significance of federal funding to support state
and local program development; (ii) the influence of
more severe air pollution episodes which might be
expected in the absence of federally-mandated con-
trols; (iii) the potential emergence of pollution havens,
as well as anti-pollution havens, motivated by local
6 The price of capital assets is also equal to the cost of production, so that changes in the rate of capital accumulation result in an
increase in the cost of producing investment goods. This has to be equilibrated with the discounted value of future rentals in order to
produce an intertemporal equilibrium. The rising cost of producing investment is a cost of adjusting to a new intertemporal equilib-
rium path.
-------
Appendix A: Cost and Macroeconotnic Modeling
political and economic conditions; (iv) the influence
of federally-sponsored research on the development
of pollution effects information and control technolo-
gies; and (v) the need to make specific assumptions
about individual state and local control levels for in-
dividual pollutants to allow estimation of incremen-
tal reductions attributable to federal control programs.
Another complication associated with the no-con-
trol scenario is the treatment of air pollution control
requirements among the major trading partners of the
U.S. Real-world manifestation of a no-control scenario
would imply that public health and environmental
goals were not deemed sufficiently compelling by U.S.
policy makers. Under these conditions, major trading
partners of the U.S. in Japan, Europe, and Canada may
well reach similar policy conclusions. Simply put, if
the U.S. saw no need for air pollution controls, there
is little reason to assume other developed industrial
countries would have either. In this case, some of the
estimated economic benefits of reducing or eliminat-
ing air pollution controls in the U.S. would not mate-
rialize because U.S. manufacturers would not neces-
sarily gain a production cost advantage over foreign
competitors. However, like the question of state and
local programs in the absence of a federal program,
foreign government policies under a no-control sce-
nario would be highly speculative.
Given the severity of these confounding factors,
the only analytically feasible assumptions with respect
to the no-control scenario are that (a) no new control
programs would have been initiated after 1970 by the
states or local governments in the absence of a fed-
eral role, and (b) environmental policies of U.S. trad-
ing partners remain constant regardless of U.S. policy.
Elimination of Compliance Costs in the
No-Control Case
Industries that are affected by environmental regu-
lations can generally respond in three ways: (i) with
process changes (e.g., fluidized bed combustion); (ii)
through input substitution (e.g., switching from high
sulfur coal to low sulfur coal); and (iii) end-of-pipe
abatement (e.g., the use of electrostatic precipitation
to reduce the emissions of particulates by combus-
tion equipment).7 Clean air regulations have typically
led to the latter two responses, especially in the short
run. End-of-pipe abatement is usually the method of
choice for existing facilities, since modifying exist-
ing production processes can be costly. This approach
is also encouraged by EPA's setting of standards based
on the notion of "best available technology" (Free-
man, 1978).
All three possible responses may lead to: (i) un-
anticipated losses to equity owners; (ii) changes in
current output; and (iii) changes in long-run profit-
ability. If firms were initially maximizing profits, then
any of the above three responses will increase its costs.
Fixed costs of investment will be capitalized imme-
diately. This will result in a loss to owners of equity
when regulations are introduced. As far as firms are
concerned, this is just like a lump sum tax on sunk
capital. Such effects will not affect growth or effi-
ciency. However, regulations could also change mar-
ginal costs and therefore current output. In addition,
they could change profits (i.e., the earnings of capi-
tal), and thus affect investment. Both of these effects
will reduce the measured output of the economy.
On the consumption side, environmental regula-
tions change consumers' expectations of their lifetime
wealth. In the no-control scenario of this assessment,
lifetime wealth increases. This causes an increase in
consumption. In fact, with perfect foresight, consump-
tion rises more in earlier time periods. This also re-
sults in a change in savings.
Capital Costs - Stationary Sources
To appropriately model investment in pollution
control requires a recognition that the CAA had two
different effects on capital markets. First, CAA regu-
lations led to the retrofitting of existing capital stock
in order to meet environmental standards. In the no-
control scenario, these expenditures do not occur. In-
stead, the resources that were invested in pollution
abatement equipment to retrofit existing sources are
available to go to other competing investments. Thus,
at each point in time, these resources might go to in-
vestments in capital in the regulated industry, or may
go into investments in other industries, depending
upon relative rates of return on those investments. This
will affect the processes of capital formation and deep-
ening.
Second, the CAA placed restrictions on new
sources of emissions. When making investment deci-
sions, firms take into account the additional cost of
pollution abatement equipment. Effectively, the
Regulation may also affect the rate of investment, and change the rate of capital accumulation.
A-7
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
"price" of investment goods is higher because more
units of capital are required to produce the same
amount of output. In the no-control scenario, there
are no restrictions on new sources and hence no re-
quirements for pollution control expenditures. Effec-
tively, the "price" of investment goods is lower. Thus,
at each point in time, investors are faced with a lower
price of investment goods. This results in a different
profile for investment over time.
Operating and Maintenance Costs - Stationary
Sources
In addition to purchasing pollution abatement
equipment, firms incurred costs to run and maintain
the pollution abatement equipment. In the no-control
scenario, resources used to pay for these operating
and maintenance (O&M) costs are freed up for other
uses. The model assumes that the resources required
to run and maintain pollution control equipment are
in the same proportions as the factor inputs used in
the underlying production technology. For example,
if 1 unit of labor and 2 units of materials are used to
produce 1 unit of output, then one-third of pollution
control O&M costs are allocated to labor and two-
thirds are allocated to materials. These adjustments
were introduced at the sector level. O&M expendi-
tures are exclusive of depreciation charges and offset
by any recovered costs.
Capital Costs - Mobile Sources
Capital costs associated with pollution control
equipment were represented by changing costs for
motor vehicles (sector 24) and other transportation
equipment (sector 26). Prices (unit costs) were reduced
in proportion to the value of the pollution control de-
vices contained in cars, trucks, motorcycles, and air-
craft.
Operating and Maintenance - Mobile Sources
Prices for refined petroleum products (sector 16)
were changed to reflect the resource costs associated
with producing unleaded and reduced lead gasoline
(fuel price penalty), the change in fuel economy for
vehicles equipped with pollution control devices (fuel
economy penalty), and the change in fuel economy
due to the increased fuel density of lower leaded and
no lead gasoline (fuel economy credit). Third, inspec-
tion and maintenance costs and a maintenance credit
associated with the use of unleaded and lower leaded
(i.e., unleaded and lower leaded gasoline is less cor-
rosive, and therefore results in fewer muffler replace-
ments, less spark plug corrosion, and less degrada-
tion of engine oil) were represented as changes in
prices for other services (sector 34).
Direct Compliance Expenditures
Data
Sources of Cost Data
Cost data for this study are derived primarily from
the 1990 Cost of Clean report. EPA publishes cost
data in response to requirements of the Clean Air and
Clean Water Acts. The following subsections describe
Cost of Clean data in detail, as well as adjustments
made to the data and data from other sources.
Cost of Clean Data
EPA is required to compile and publish public
and private costs resulting from enactment of the Clean
Air Act and the Clean Water Act. The 1990 Cost of
Clean report presents estimates of historical pollution
control expenditures for the years 1972 through 1988
and projected future costs for the years 1989 through
2000. This includes federal, state, and local govern-
ments as well as the private sector. Estimates of capi-
tal costs, operation and maintenance (O&M) costs,
and total annualized costs for five categories of envi-
ronmental media, including air, water, land, chemi-
cal, and multi-media, are presented. It should be noted
that these estimates represent direct regulatory imple-
mentation and compliance costs rather than social
costs. The Cost of Clean relied on data from two gov-
ernmental sources, the EPA and the U.S. Department
of Commerce (Commerce).
EPA Data
EPA expenditures were estimated from EPA bud-
get justification documents.8 Estimates of capital and
operating costs resulting from new and forthcoming
regulations were derived from EPA's Regulatory Im-
pact Analyses (RIAs). RIAs have been prepared prior
to the issuance of all major regulations since 1981.
Finally, special analyses conducted by EPA program
offices or contractors were used when other data
sources did not provide adequate or reliable data.
8 The main source of data for EPA expenditures is the Justification of Appropriation Estimates for Committee on Appropriations.
A-8
-------
Appendix A: Cost and Macroeconomic Modeling
Commerce Data
Data collected by Commerce were used exten-
sively in the Cost of Clean for estimates of historical
pollution control expenditures made by government
agencies other than EPA and by the private sector.
Two Commerce agencies, the Bureau of Economic
Analysis (BEA) and the Bureau of the Census (Cen-
sus), have collected capital and operating costs for
compliance with environmental regulations since the
early 1970's. Commerce is, in fact, the primary source
of original survey data for environmental regulation
compliance costs. Commerce publishes a number of
documents that report responses to surveys and com-
prise most of the current domain of known pollution
abatement and control costs in the United States, in-
cluding:
• A series of articles entitled "Pollution Abate-
ment and Control Expenditures" published
annually in the Survey of Current Business
by BEA (BEA articles);
• A series of documents entitled "Pollution
Abatement Costs and Expenditures" pub-
lished annually in the Current Industrial Re-
ports by Census (PACE reports); and
• A series of documents entitled Government
Finances published annually by Census (Gov-
ernment Finances).
BEA articles contain data derived from a number
of sources, including two key agency surveys —the
"Pollution Abatement Costs and Expenditures Sur-
vey" (PACE Survey) and the "Pollution Abatement
Plant and Equipment Survey" (PAPE Survey)—
which are conducted annually by Census for BEA.
Data have been reported for 1972 through 1987.9
PACE reports have been published annually since
1973 with the exception of 1987. Figures for 1987
were estimated on the basis of historical shares within
total manufacturing. These reports contain expendi-
ture estimates derived from surveys of about 20,000
manufacturing establishments. Pollution abatement
expenditures for air, water and solid waste are reported
by state and Standard Industrial Code (SIC) at the four-
digit level. According to Census, surveys conducted
since 1976 have not included establishments with
fewer than 20 employees because early surveys
showed that they contributed only about 2 percent to
the pollution estimates while constituting more than
10 percent of the sample size.
Each year Census conducts a survey of state, lo-
cal, and county governments; and survey results are
published in Government Finances. Census asks gov-
ernment units to report revenue and expenditures, in-
cluding expenditures for pollution control and abate-
ment.
Non-EPA Federal expenditures were estimated
from surveys completed by federal agencies detailing
their pollution control expenditures, which are sub-
mitted to BEA. Private sector air pollution control
expenditures, as well as state and local government
air pollution expenditures, were taken from BEA ar-
ticles.
Stationary Source Cost Data
Capital Expenditures Data
Capital expenditures for stationary air pollution
control are made by factories and electric utilities for
plant and equipment that abate pollutants through end-
of-line (EOL) techniques or that reduce or eliminate
the generation of pollutants through changes in pro-
duction processes (CIPP). For the purposes of this
report EOL and CIPP expenditures are aggregated.10
Table A-3 summarizes capital expenditures for sta-
tionary air pollution control, categorized as "nonfarm
business" or "government enterprise" expenditures.
Nonfarm business capital expenditures consist of
plant and equipment expenditures made by 1) manu-
facturing companies, 2) privately and cooperatively
owned electric utilities, and 3) other nonmanufacturing
companies. "Government enterprise" is, according to
BEA, an agency of the government whose operating
costs, to a substantial extent, are covered by the sale
of goods and services. Here, government enterprise
means specifically government enterprise electric
9 The most recent BEA article used as a source for air pollution control costs in the Cost of Clean was "Pollution Abatement and
Control Expenditures, 1984-87" in Survey of Current Business, June 1989.
10 Survey respondents to the Census annual Pollution Abatement Surveys report the difference between expenditures for CIPP
and what they would have spent for comparable plant and equipment without pollution abatement features. Disaggregated capital
expenditures by private manufacturing establishments can be found in annual issues of Census reports.
_
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table A-3. Estimated Capital and G&M
Expenditures for Stationary .Source Air
*• * -4 •• V
Pollution Control (millions
Nonfarm
Business
^eaz
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
Tap.'
2,172
2,968
3,328
3,914
3,798
3,811
3,977
'4,613
5,051
5,135
5,086
4,155
4,282
4,141
4,090
4,179
4,267
4,760
4,169
OfrM*
1,407--
1,839
2,195
2,607
3463
3,652 '
4,499
5,420
5,988
5,674.
6,149 -'
6,690
6,997
7,116
7,469"
7,313
7,743
8,688
Government
Enterprise
£a|Lf' £
63 :
82 -.-,
104
102' ~"
156 -
197
205
285 ,
<»-398v, ,N
45TV>V;
508%o,
*• "412 ^
41^ *"
' "328 ' '
312
277
'',243*^
-g.^,?XV
226 \ '
?/=** ~*
, -^2^
^ -56-,
"v>45'-
^ 58
' 60
72
,-1%
,-v,H4,8j-
^- ,135--
•'••- I4,i:-
%v-143 -
'147,
- ^C
-•-i4dC
"U30*
T61-'
*'lfo
" 154
Sources:
a. Non-fena capital expenditures for 197&S7 ateftocn Cost
ofClean,Table B-l, line 2. >• - ,- — , ,-,
b. Non-famj O&M expenditures for J973-85 as to* Co**
^fCfea/i.Tdile B-l, Une 8, . •, (, ^^^'^^
c. GovemmeTit enterpnse capital expenditures fpr i STJ-ST'^;*
are from Cort of Clean, Table B-9, line 1. , .*,---,
d. Government enterpnse O&M expenditures for 1973-85 '
are from Cost of Clean, Table B-?, fine 5, ' ' !'
All other reported expenditures ate EPA estimates.
utilities. Government enterprise capital expenditures
are pollution abatement expenditures made by pub-
licly owned electric utilities.11
Operation and Maintenance Expenditures Data
Stationary source O&M expenditures are made
by manufacturing establishments, private and public
electric utilities, and other nonmanufacturing busi-
nesses to operate air pollution abatement equipment.
O&M expenditures for electric utilities are made up
of two parts: 1) expenditures for operating air pollu-
tion equipment and 2) the additional expenditures as-
sociated with switching to alternative fuels that have
lower sulfur content (fuel differential). Expenditures
to operate air pollution abatement equipment are for
the collection and disposal of flyash, bottom ash, sul-
fur and sulfur products, and other products from flue
gases.12 O&M expenditures are net of depreciation
and payments to governmental units, and are summa-
rized in Table A-3. O&M data were disaggregated to
the two digit SIC level for use in the macroeconomic
model.
For both capital and O&M expenditures, histori-
cal survey data were not available for each year
through 1990 prior to publication of Cost of Clean.
For the purpose of the section 812 analysis, EPA pro-
jected 1988-1990 capital expenditures and 1986-1990
O&M expenditures. Those projections were used in
the macroeconomic simulation, and have been retained
as cost estimates to ensure consistency between the
macroeconomic results and the direct cost estimates.
Since completion of the macroeconomic modeling,
however, BEA has published expenditure estimates
through 1990. A comparison of more recent BEA es-
timates with the EPA projections used in the section
812 analysis can be found in the "Uncertainties in the
Cost Analysis" section, below.
Recovered Costs
"Recovered costs" are costs recovered (i.e., rev-
enues realized) by private manufacturing establish-
ments through abatement activities. According to in-
structions provided to survey participants by Census,
recovered costs consist of 1) the value of materials or
energy reclaimed through abatement activities that
were reused in production and 2) revenue that was
obtained from the sale of materials or energy reclaimed
through abatement activities. Estimates of recovered
costs were obtained from the PACE reports and are
summarized in Table A-4. In this analysis, recovered
costs were removed from total stationary source air
pollution control O&M costs — that is, net O&M cost
in any year would be O&M expenditures (see Table
A-3) less recovered costs. Recovered cost data were
disaggregated to the two digit SIC level for use in the
macroeconomic model.
11 BEA calculates these expenditures using numbers obtained from Energy Information Agency (EIA) Form 767 on steam-
electric plant air quality control.
12 Farber, Kit D. and Gary L. Rutledge, "Pollution Abatement and Control Expenditures: Methods and Sources for Current-
Dollar Estimates," Unpublished paper, Bureau of Economic Analysis, U.S. Department of Commerce, October 1989.
A-10
-------
Appendix A: Cost and Macroeconomic Modeling
s'-f*"' -
fable A-avBMJ
(f^lksis-of current dollars)^'",_
"•?- ' '^ ^ x;"'=?.g>
PAfTE* '''Estimated
";, ,, , '•'?'>«!>' >•:
,, 'it' -. ''
',;-,>1973..' ,-4- '""- ,>-:*-/ ' IsV'
- " - -»,-*•<•,-496^'
,1977 " ' -K >v-;" <«*•< 5S'?-VV
,,1^,78 - -'' ,Xv t>;'-"- ,,,N,;617 '";
?S^9 "^,,, ^'7S€k" ""'"'"
,,;l9|o „'-,/--• ^.
',-1983 *''l'f r-"-
'&&£' 'J' '" ->->?«O"J^
s,'-.*•* 'v-oOO ' _, J-J vo/w
^, 767--•->-" ,,, ,,-,^7,68
\,, 860- -' " ,,,,,8.67
i,V Xv
*~s"
"\;biSusfrial-Report|;fey Census*- '
Mobile Source Cost Data
Costs of controlling pollution emissions from
motor vehicles were estimated by calculating the pur-
chase price and O&M cost premiums associated with
vehicles equipped with pollution abatement controls
over the costs for vehicles not equipped with such
controls. These costs were derived using EPA analy-
ses, including EPA RIAs, the Cost of Clean, and other
EPA reports.13 This Appendix summarizes the sec-
tion 812 mobile source compliance cost estimates and
provides references to published data sources where
possible. Further information on specific methods,
analytical steps, and assumptions can be found in
McConnell etal. (1995),14 which provides a detailed
description of the section 812 mobile source cost es-
timation exercise and compares the method and re-
13 A complete listing of sources used in calculating mobile source capital and operating expenditures can be found in Environ-
mental Investments: The Cost of a Clean Environment, Report of the Administrator of the Environmental Protection Agency to the
Congress of the United State, EPA-230-11-90-083, November 1990.
14 Evaluating the Cost of Compliance with Mobile Source Emission Control Requirements: Retrospective Analysis, Resources
for the Future Discussion Paper, 1995. Note that McConnell etal. refer to the section 812 estimates as: Cost of Clean (1993, unpub-
lished).
suits to other similar analyses (including Cost of Clean
(1990)).
Capital Expenditures Data
Capital expenditures for mobile source emission
control are associated primarily with pollution abate-
ment equipment on passenger cars, which comprise
the bulk of all mobile sources of pollution. These capi-
tal costs reflect increasingly stringent regulatory re-
quirements and improvements in pollution control
technologies overtime. Each of the following devices
have been used at one time or another dating back to
the Clean Air Act Amendments of 1965: air pumps,
exhaust-gas recirculation valves, high altitude con-
trols, evaporative emissions controls, and catalysts.
The cost estimates for each component were computed
on a per-vehicle basis by engineering cost analyses
commissioned by EPA. The resulting per-vehicle capi-
tal costs were multiplied by vehicle production esti-
mates to determine annual capital costs. Table A-5
summarizes mobile source capital costs.
Operation and Maintenance Expenditures Data
Costs for operation and maintenance of emission
abatement devices include the costs of maintaining
pollution control equipment plus the cost of vehicle
inspection/maintenance programs. Operating costs per
vehicle were multiplied by total vehicles in use to
determine annual cost. Mobile source O&M costs are
made up of three factors: 1) fuel price penalty, 2) fuel
economy penalty, and 3) inspection and maintenance
program costs as described below. These costs are
mitigated by cost savings in the form of maintenance
economy and fuel density economy. Table A-6 sum-
marizes mobile source O&M expenditures and cost
savings by categories, with net O&M costs summa-
rized above in Table A-5. The following sections de-
scribe the components of the mobile source O&M cost
estimates.
Fuel Price Penalty
Historically, the price of unleaded fuel has been
several cents per gallon higher than the price of leaded
fuel. CAA costs were calculated as the difference be-
A-ll
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table A-5, Estimated Capital and
Operation and Maintenance Expenditures
for Mobile Source Air Pollution Control
(millions of current dollars).
Year
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
Capital" ,
276
242
1,570
1,961
2,248
2,513
2,941
2,949
3,534
3,551
4,331
5,679
6,387
6,886
6,851
7,206
7,053
7,299
O&M* „
<• 1,765 "
'2,35 f~~'
° f 2^282 j
2^?6lf®'
S'lf?$5:* *;
~<" / f 9Q% '
- - " " ' 1,229
1,790
1,389
555
-155.
-326
337"
, -1,394
-l',302\
-1,575 <
" v -1,63,6, •
*" ~~ :U&1A
a. CBpital'oqp,: Cost of Clean, Tables C-2 to O9, tine 3
on each; Tables C-2 A to C-9A, fine 10 on each; «sa V«rted
from SJ986to cuirent dollars.
b, O&M ex p.: EPA analyses based on sources and
methods in: Costs and Benefits of Reducing Lead~in
Gasoline: Final Regulatory Imp act Analysis, TJ.S.'
Environmental Protection Agency, Office of Bjlicy
Analysis, EPA-230-05-SS-O06, "
of Clean.
is likely to understate costs because regulatory require-
ments and market developments cannot be perfectly
anticipated over time. This procedure resulted in esti-
mates that are about ten percent less than estimates in
other EPA reports.15 However, new process technolo-
gies that were developed in the mid-1980s were not
reflected in either the base case or regulatory case runs.
It is reasonable to expect that regulatory requirements
would have encouraged development of technologies
at a faster rate than would have occurred otherwise.
Fuel Economy Penalty
The fuel economy penalty benefit is the cost as-
sociated with the increased/decreased amount of fuel
used by automobiles with air pollution control devices
(all else being equal). An assumption that can be made
is that the addition of devices, such as catalytic con-
1 -TableA-^'0^M'C9'$ts-'«t)4'O-edits (millions!''
Fuell»iic«""vl*)Ec1)aV';"^"' Net Totfil
'1974
*2'35'l'
, IOT
-------
Appendix A: Cost and Macroeconomic Modeling
verters, decrease automobile fuel efficiency.16 If this
assumption is true, air pollution control devices in-
crease the total fuel cost to consumers. An alternative
assumption is that the use of catalytic converters has
increased fuel economy. This increase has been at-
tributed in large measure to the feedback mechanism
built into three-way catalytic converters.17 Under this
assumption, the decrease in total fuel cost to consum-
ers is considered a benefit of the program.
For the purposes of this study, sensitivity analy-
ses were performed using data presented in the Cost
of Clean report. These analyses were conducted to
evaluate the significance of assumptions about the
relationship between mile per gallon (MPG) values
for controlled automobiles and MPG values for un-
controlled cars. Based on results of these and other
analyses, fuel economy was assumed to be equal for
controlled and uncontrolled vehicles from 1976 on-
ward. This may bias the cost estimates although hi an
unknown direction.
Inspection and Maintenance Programs
Inspection and maintenance programs are admin-
istered by a number of states. Although these programs
are required by the Clean Air Act, the details of ad-
ministration were left to the discretion of state or lo-
cal officials. The primary purpose of inspection and
maintenance programs is to identify cars that require
maintenance —including cars that 1) have had poor
maintenance, 2) have been deliberately tampered with
or had pollution control devices removed, or 3) have
used leaded gasoline when unleaded is required— and
force the owners of those cars to make necessary re-
pairs or adjustments.18 Expenditures for inspection and
maintenance were taken from the Cost of Clean.
Beneficial effects of the mobile source control
program associated with maintenance and fuel den-
sity were also identified. These cost savings were in-
cluded in this study as credits to be attributed to the
mobile source control program. Credits were estimated
based on an EPA study,19 where more detailed expla-
nations may be found.
16 Memo from Joel Schwartz (EPA/OPPE) to Joe Somers and Jim DeMocker dated December 12,1991, and entitled "Fuel
Economy Benefits." Schwartz states that since this analysis is relative to a no Clean Air Act baseline, not a 1973 baseline, fuel
economy benefits are not relevant. In the absence of regulation, tuning of engines for maximum economy would presumably be
optimal in the base case as well.
17 Memo from Joseph H. Somers, EPA Office of Mobile Sources, to Anne Grambsch (EPA/OPPE) and Joel Schwartz (EPA/
OPPE) entitled "Fuel Economy Penalties for section 812 Report," December 23,1991.
18 Walsh, Michael P., "Motor Vehicles and Fuels: The Problem," EPA Journal, Vol. 17, No. 1, January/February 1991, p. 12.
19 Schwartz, J., et al. Costs and Benefits of Reducing Lead in Gasoline: Final Regulatory Impact Analysis, U. S. Environmental
Protection Agency, Economic Analysis Division, Office of Policy Analysis, February 1985.
A-13
Maintenance Credits
Catalytic converters require the use of unleaded
fuel, which is less corrosive than leaded gasoline. On
the basis of fleet trials, the use of unleaded or lower
leaded gasoline results in fewer muffler replacements,
less spark plug corrosion, and less degradation of en-
gine oil, thus reducing maintenance costs. Mainte-
nance credits account for the majority of the direct
(non-health) economic benefits of reducing the lead
concentration in gasoline.
Fuel Density Credits
The process of refining unleaded gasoline in-
creases its density. The result is a gasoline that has
higher energy content. Furthermore, unleaded gaso-
line generates more deposits in engine combustion
chambers, resulting in slightly increased compression
and engine efficiency. Higher energy content of un-
leaded gasoline and increased engine efficiency from
the used of unleaded gasoline yield greater fuel
economy and therefore savings in refining, distribu-
tion, and retailing costs.
Other Direct Cost Data
The Cost of Clean report includes several other
categories of cost that are not easily classified as ei-
ther stationary source or mobile source expenditures.
Federal and state governments incur air pollution
abatement costs; additionally, federal and state gov-
ernments incur costs to develop and enforce CAA
regulations. Research and development expenditures
by the federal government, state and local govern-
ments, and (especially) the private sector can be at-
tributed to the CAA. These data are summarized by
year in Table A-7.
Unlike the other private sector expenditure data
used for this analysis, the survey data used as a source
for private sector R&D expenditures cannot be disag-
gregated into industry-specific expenditure totals.
Consequently, private sector R&D expenditures are
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table A-7. Other Air Pollution Control ExpetfditaresTpnipons of
current dollars). ' ' " „ y;,v v,, ,»:,;>,-:
year,
Abatement
State &
Fed." Local*
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
47
56
88
105
106
90
103
95
85
87
136
115
98
67
80
65
70
71
0
0
1
1
1
0
0
0
0
0
4
14
12
14
15
10
12
13
Regulations
; Research
State &
Fed." Local* Private^' |
50
52
66
69
80
93
100
122
108
93
88
101
103
106'
110
120
130
133
115
131
139
135
161
183-
200
207
226
230
239
250
250
307
300
320*
360
343
492 '
. -520--
487
562
675
805
933-
851
79,8, ,
761
691 "•-
665; '
•77r::
'" "83'f"*
887,;;'
' 934 "-'
984
749'
jakCi
~iW '
100" '
los ;;
13 r:
144
146
105,
'130
131
126;;
133"
'16'5
;;%? ,.
:;2t7 ,v
;i^!;;
- 220 --"
"230*-
231- '
f >••
-< V- , -
V, " s " "
6
'* '7 •'
;^;8
'6
' "?"
,, ,8,
-, ;-^-'»
' 5
;c.
" <- 2, <
6;
4*
- 3'
- ~4"
! 2
1
v-V-2
U"2
Tntel
,' '', *A
'836'
-'866-
'897;
t'.tKf^
1,174-
- vl,32f.
,11,448-,
1,410'
'%'34$
^229
:4',297
"" 1*314'
""J$3j$
^5%'
*'??!
,/lv6TO-
1,788
i;54a<
_ Federal government abatement expenditures: 1973-82, "Pollution Abatement-anACoittroi;;^
Expenditures", .s|'"vy pf .Qirrent- Business (BEA) July 188S Table 9 line 13^1^3-87, BEA ' t
June 1989 Table 71ine 13; 1988-90, BEA May 1995 Table 7 lipes,lS.s ";,'.,, ' " '"
b. St«e and local abatement expenditures; 1973-87, Cost of dean, Ta&te B»9lin^5f 1985^0,
BEA May 1995 Table? line 14. ^ c ,t, , , > *•- *•-....
«. Privae sector R&D expenditures: 1973-86, BEA May 1994 Table'-*! v—^^
expenditures in $1987 are converted to current dollars, vsipg to QiDP price df fia
elsewhetein tbisAppendix —netting out public sector I^&l? leaves priv*»
1987-90, BEA May 1995 Table 7 line 20. ' - ,,*,-, -
f. Federal govcrnmentR&D expenditures: 1973-82, BEA My 198fjTafeJe91j)j»*||.;4^
BEAJune I989Toble61ine21; 1988-90, BEAMiy 1995,Tafale-7 fine21. ' *^r>V
g. State and local government R&D expenditures: 1973^87, SgQtgggfe/ut, Table B'9' fit<
1988-90,BEA May 1995Table? Iine22. " ^r"" " ,VlSS', -'-- --«-,—
from more recent issues of the Survey of
Current Business (BEA). Federal govern-
ment expenditures are from BEA (various
issues). Private R&D expenditures were
reported in Cost of Clean. Since publica-
tion of Cost of Clean, however, BEA has
revised its private sector R&D expenditure
series (BEA, 1994 and 1995). Since private
R&D expenditures were not included in the
macroeconomic modeling exercise, the re-
vised series can be (and has been) used
without causing inconsistency with other
portions of the section 812 analysis.
Assessment Results
Compliance Expenditures and
Costs
Compliance with the CAA imposed
direct costs on businesses, consumers, and
governmental units, and triggered other
expenditures such as governmental regula-
tion and monitoring costs and expenditures
for research and development by both gov-
ernment and industry. As shown in Table
A-8, annual CAA compliance expenditures
- including R&D, etc.- over the period
from 1973 to 1990 were remarkably
stable20, ranging from about $20 billion to
$25 billion hi inflation-adjusted 1990 dol-
lars (expenditures are adjusted to 1990 dol-
lars through application of the GDP Implicit
Price Deflator). This is equal to approxi-
mately one third of one percent of total
domestic output during that period, with the
percentage falling from one half of one per-
cent of total output hi 1973 to one quarter
of one percent in 1990.
omitted from the macroeconomic modeling exercise
(the macro model is industry-specific). The R&D ex-
penditures are, however, included in aggregate cost
totals used in the benefit-cost analysis.
The Cost of Clean and the series of articles "Pol-
lution Abatement and Control Expenditures" hi the
Survey of Current Business (various issues) are the
data sources for "Other Air Pollution Control Expen-
ditures." State and local expenditures through 1987
are found in Cost of Clean; 1988-90 expenditures are
Although useful for many purposes, a summary
of direct annual expenditures is not the best cost mea-
sure to use when comparing costs to benefits. Capital
expenditures are investments, generating a stream of
benefits (and opportunity cost) over the life of the in-
vestment. The appropriate accounting technique to use
for capital expenditures in a cost/benefit analysis is to
annualize the expenditure — i.e., to spread the capi-
tal cost over the useful life of the investment, apply-
ing a discount rate to account for the time value of
money.
20 While total expenditures remained relatively constant over the period, the sector-specific data presented in Tables A-3 and A-5
above indicate that capital expenditures for stationary sources fell significantly throughout the period but that this decline was offset
by significant increases in mobile source capital expenditures.
__
-------
Appendix A: Cost andMacroeconomic Modeling
A-15
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Annualization Method
For this cost/benefit analysis, all capital expendi-
tures have been annualized at 3 percent, 5 percent,
and 7 percent (real) rates of interest. Therefore, "an-
nualized" costs reported for any given year are equal
to O&M expenditures (plus R&D, etc., expenditures,
minus recovered costs) plus amortized capital costs
(i.e., depreciation plus interest costs associated with
the pre-existing capital stock) for that year. Station-
ary source air pollution control capital costs are am-
ortized over twenty years; mobile source air pollution
control costs are amortized over ten years. Capital
expenditures are amortized using the formula for an
annuity [that is, r/(l-(l+r)'1) , where r is the rate of
interest and t is the amortization period].21 Multiply-
ing the expenditure by the appropriate annuity factor
gives a constant annual cost to be incurred for t years,
the present value of which is equal to the expenditure.
Due to data limitations, the cost analysis for this
CAA retrospective starts in 1973, missing costs in-
curred in 1970-72. Cost of Clean, however, includes
stationary source capital expenditures for 1972. In this
analysis, amortized costs arising from 1972
capital investments are included in the 1973-
1990 annualized costs, even though 1972
costs are not otherwise included in the analy-
sis. Conversely, only a portion of the (e.g.)
1989 capital expenditures are reflected in the
1990 annualized costs — the remainder of
the costs are spread through the following
two decades, which fall outside of the scope
of this study (similarly, benefits arising from
emission reductions in, e.g., 1995 caused by
1990 capital investments are not captured
by the benefits analysis). Table A-9 presents
CAA compliance costs from 1973 to 1990,
in 1990 dollars, with capital expenditures
amortized at a five percent real interest rate.
"Total" costs are the sum of stationary
source, mobile source, and "other" costs,
minus recovered costs.
The Stationary Source table reports a capital ex-
penditure of $6,521 million for 1972 (in 1990 dol-
lars). The cost is spread over the following twenty
years (which is the assumed useful life of the invest-
ment) using a discount rate of five percent; thus, the
amortization factor to be used is f(20)=0.0802. Mul-
tiplying $6,521 million by 0.0802 gives an annuity of
$523 million. That annuity is noted on the first data
row of the table, signifying that the 1972 expenditure
of $6,521 million implies an annual cost of $523 mil-
lion for the entire twenty-year period of 1973 to 1992
(the years following 1990 are not included on the
tables, since costs incurred in those years are not in-
cluded in this retrospective assessment). The first sum-
mary row near the bottom of the table (labeled "SUM")
reports aggregate annualized capital costs: for 1973
(the first data column), capital costs are $523 million.
Capital expenditures in 1973 amounted to $8,360
million. Using the amortization technique explained
above, one can compute an annualized cost of $671
million, incurred for the twenty-year period of 1974
to 1993. Aggregate annualized capital costs for 1974
include cost flows arising from 1972 and 1973 invest-
TabieA-
", 1973-1990
1973
"1974
1977,
-1978
1986
Tables A-10 and A-11 provide details
of the amortization calculation (using a five
percent interest rate) for stationary sources
and mobile sources, respectively. Similar
calculations were performed to derive the
annualized cost results using discount rates
of three percent and seven percent.
1999
' 199*6
Stalioria'iy
*k
"523,, ''C'936
1*194' ^4,778
2,630
- 3,31-7
3,968"
,5,277
'6!,6iO'
5,768
6,527
6,991
" 7,959
",S,7&t
S',785
' 7,855
8,168
8',593
9,005'-'
'- 9,410
8,477
!;,
-------
Appendix A: Cost and Macroeconomic Modeling
''5*^11 J™£&fg 5-^-*" § * *^Sic§ "^ S § ' ''''''frl^s,; ' "'»'%•;§»'j^'i:§''S ''•'•,%%#$?* S'^S^isi"°""'
t^^£ ,£&. ^%IV^%:^
'"*'^i< ^^ ~^^:^m^^-'£^i^!''-, -•ksiJ^.v ''<-'^rh:,' j&,'B**s<r- >^>i ^ -'^T--"^^'- <•>*:•>"&>"•-/'< - ,'^^>,-;/A g:s:s-srls^
ftg^; vl?|;g|:^|^,,|;|||«|,,-| "'X?L-£:;>V':'-- -'/''"" '^^^"^ 1'^'^ '^"1 ^*1)-J
1&iv il^l.s§:i2i--^;V:B'?-- : - — -l^|tl~l:l^ -:%
S""" ^"i""1 1V' '• ' '"^*«™ '"„','""''-, r,v"-"~'-s^V- V1 ^:i^«^f<5ri"cf Clfv,',\'"W;;;/^_ ~, ^ ••;Ir^"C- - -liv*-
••>§ ••, cS^tS^ ••••<*• "VN. A ' "" ^^•—v-X' ' s "•'•-Wlc/ ^ ^ ••*'' "'/'"••''" \"";'tS^>^ v y'*§\;
^8-'
"•• _
€4v'f%r4
59 » 'P.'-i:'-
,
m -w>; V;JN , - *-'ZSi%:'^ ,r;~ ~'^^^™:^;''!'/":u, ^^§^^>
''" "' ' 'y *
O ' -
* * stf ^ 5 ^ ^ ' ' ' "••'<•• ^ * "^ <'j ' ...V™^ y'" '-"&•'•• '••* * '' ' ^^-""^ «*»^rs^v , ^ * ^y^QLf O"^^^^'
•-•^^""i ® *"- ''::- , ' -*^< ~ '""''-'^-^^ii-' - " ^SosiS-iS-^ .^', ;":':,':l':-"§ ^-^(:&'-'"
•-Q
«, ~ -fct-'-
=! us te :-iV, ^ 'S-,ff .a.
V».'" --"T"-'"-<^t:*. - ^;'~",-f^i' • /"--„«„ , •&-;*&<;5, - 1' '§i;C"^-, E -
K KJt>^.^<5Jl'^ri.v*%.^'^*t. ^, ^^cjS5?''****-^'C>x:'''' JJ^ir\.--"""O
A-17
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
g
c\
»—*
•8
V3
§
g
g
CO
I
§
-
O
s
N
j
•
I
,,>x .- - ' "-«, <=, <*. « ^ r;
<-;\-"**" «- •> - • ,"'? *"• ' •* •* •-, -,**?•-• •-" . ,. „, * ^ ^.» a
m £ S in ffi. J?
to
- *Sf S t3 iT*
jj, "./--/-, « -g d »
'"'"' r^g-o 8 B '
.-.. ^"^;,^;^,^
A-18
-------
Appendix A: Cost and Macroeconomic Modeling
raents: that is, $523 million plus $671 million, or
$1,194 million (see the "SUM" row). Similar calcu-
lations are conducted for every year through 1990, to
derive aggregate annualized capital costs that increase
monotonically from 1973 to 1990, even though capi-
tal expenditures decline after 1975.22
An alternative calculation technique is available
that is procedurally simpler but analytically identical
to that outlined above. Instead of calculating an annu-
ity for each capital expenditure (by multiplying the
expenditure by the annuity factor f), then summing
the annuities associated with all expenditures in pre-
vious years, one can sum all previous expenditures
and multiply the sum (i..e., the capital stock at the
start of the year) by /. The third summary row (la-
beled "K stock") near the bottom of the amortization
summary tables give the pollution control capital stock
at the start of each year. For example, the stationary
sources capital stock in place at the start of 1975 was
$23,533 million (this is the sum of 1972, 1973, and
1974 capital expenditures). Multiplying the capital
stock by the annuity factor 0.0802 gives $1,888 mil-
lion, which is the aggregate annualized stationary
source capital cost for 1975.
One can perform further calculations to decom-
pose the annualized capital costs into "interest" and
"financial depreciation" components.23 For example,
at the start of 1973, the stationary source capital stock
was $6,521 million. A five percent interest rate im-
plies an "interest expense" for 1973 of $326 million.
Given a 1973 annualized cost of $523 million, this
implies a "depreciation expense" for that year of ($523
million minus $326 million =) $197 million. For 1974,
the existing capital stock net of "financial deprecia-
tion" was $14,684 million (that is, the $6,521 million
in place at the start of 1973, plus the investment of
$8,360 million during 1973, minus the depreciation
of $197 million during 1973); five percent of $14,684
million is the interest expense of $734 million. Since
the annualized capital cost for 1974 is $1,194 mil-
lion, depreciation expense is $460 million (i.e., the
difference between annualized cost and the interest
component of annualized cost). This procedure is re-
peated to determine interest and depreciation for each
year through 1990 (see the last three rows of Table A-
11).
The three tables above all present costs (and in-
termediate calculations) assuming a five percent in-
terest rate. As noted above, the Project Team also
employed rates of three percent and seven percent to
calculate costs. Those calculations and intermediate
results are not replicated here. The method employed,
however, is identical to that employed to derive the
five percent results (with the only difference being
the interest rate employed in the annuity factor calcu-
lation). Table A-12 presents a summary of expendi-
tures and annualized costs at the three interest rates.
Table A-T2. -Gompllanee Expenditures
Annualize4 Costs, 19,73-1990- ($1990
, million$). ' '• 's 1""!" '
^'
,1973
Alft«»a:
19,635 10,957',; -H-,042
-21,405""-'-"
1977
1979
'1980
1982
24,062 '7 '*! 5,253
^22;593'''"'i4\963-
"24;S37 , ,17,309-
25,74*1--»--l-9,666-
-24367 ' i9-,590
'21,555,' , I8.-643'
- 19,095"
13,638 13;988;
13,13.9
- 15,7,?6;
,-,,,18,232, 19,
-20;905"22;321
-;- ao^s,^^.
-- 21,909;:
1986
1988:
1989
-21--, 109'
-20,615 ---<22$7!2V
L 22,012
53,161- ;25,364
24,237, •?&'$&'
,24^81 24719^-'
-;-.2349B
,
"26,066 28^17-
jj-^v,X-v,'S'-',,'
Discounting Costs and Expenditures
The stream of costs from 1973 to 1990 can be
expressed as a single cost number by discounting all
costs to a common year. In this analysis, all costs and
benefits are discounted to 1990 (in addition, all costs
and benefits are converted to 1990 dollars, removing
the effects of price inflation).24 There is a broad range
22 Similar calculations were performed for mobile source control capital costs, where the assumed amortization period is ten years.
23 One might, for example, wish to examine the relative importance of the "time value" component of the computed capital costs.
24 Unlike most cost-benefit analyses, where future expected costs and benefits are discounted back to the present, this exercise
brings past costs closer to the present. That is, the discounting procedure used here is actually compounding past costs and benefits.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
of opinion in the economics profession regarding the
appropriate discount rate to use in analyses such as
this. Some economists believe that the appropriate rate
is one that approximates the social rate of time pref-
erence — three percent, for example (all rates used
here are "real", i.e., net of price inflation impacts).
Others believe that a rate that approximates the op-
portunity cost of capital (e.g., seven percent or greater)
should be used. A third school of thought holds that
some combination of the social rate of time prefer-
ence and the opportunity cost of capital is appropri-
ate, with the combination effected either by use of an
intermediate rate or by use of a multiple-step proce-
dure which uses the social rate of time preference as
the "discount rate," but still accounts for the cost of
capital. The section 812 Project Team chose to use a
range of discount rates (three, five, and seven per-
cent) for the analysis.
Expenditures and annualized costs discounted to
1990 are found on Table A-13. Expenditures are dis-
counted at all three rates; annualized costs are dis-
counted at the rate corresponding to that used in the
annualization procedure (i.e., the "annualized at 3%"
cost stream is discounted to 1990 at three percent).
The final row presents the result of an explicit combi-
nation of two rates: Capital costs are annualized at
seven percent, then the entire cost stream is discounted
to 1990 at three percent.
Table A-13. Costs Discounted to 1990 ($1990
millions).
Expenditures
Annualized Costs
Annualized.at 7%
a.^S:
520^75
416,804
476329
5SL
627,621
522^06
7^.
760,751
657,00,3
^ "•
Indirect Economic Effects of the CAA
In addition to imposing direct compliance costs
on the economy, the CAA induced indirect economic
effects, primarily by changing the size and composi-
tion of consumption and investment flows. Although
this analysis does not add these indirect effects to the
direct costs and include them in the comparison to
benefits, they are important to note. This section sum-
marizes the most important indirect economic effects
of the CAA, as estimated by the J/W macroeconomic
simulation.
GNP and Personal Consumption
Under the no-control scenario, the level of GNP
increases by one percent in 1990 relative to the con-
trol case (see Table A-14). During the period 1973-
1990, the percent change in real GNP rises monotoni-
cally from 0.26 percent to 1 .0 percent. The increase
'.Table A'»l¥. flefe^Besfm Otass* -
^National Produet'J|e%een the fcojatrol and
•No- control Scenarios * r
Real % ,
Year
1973
-0.09
-048
-0x10
-0,00
-0-10
ff 0.26
^'0.27
', -it**"** *. 'M A
"**j$?$4'
1980 v'
1981 > >'
1982" ,
1983 /-
1984 " -
^-044
, -044
1987
'y '%M
JsO.,73
"0<74
0.84'
0,95
1,00
1.
in the level of GNP is attributable to a rapid accumu-
lation of capital, which is driven by changes in the
price of investment goods. The capital accumulation
effect is augmented by a decline hi energy prices rela-
tive to the base case. Lower energy prices that corre-
spond to a world with no CAA regulations decreases
costs and increases real household income, thus in-
creasing consumption.
Removing the pollution control component of new
capital is equivalent to lowering the marginal price of
investment goods. Combining this with the windfall
gain of not having to bring existing capital into com-
pliance leads to an initial surge in the economy's rate
of return, raising the level of real investment. The in-
A-20
-------
Appendix A: Cost and Macroeconomic Modeling
vestment effects are summarized in Figure A-l. More
rapid (ordinary) capital accumulation leads to a de-
cline in the rental price of capital services which, in
turn, stimulates the demand for capital services by pro-
ducers and consumers. The capital rental price reduc-
tions also serve to lower the prices of goods and ser-
vices and, so, the overall price level. Obviously, the
more capital intensive sectors exhibit larger price re-
ductions.25 The price effects from investment changes
are compounded by the cost reductions associated with
releasing resources from the operation and mainte-
nance of pollution control equipment and by the eh'mi-
nation of higher prices due to regulations on mobile
sources.
To households, no-control scenario conditions are
manifest as an increase in permanent future real earn-
ings which supports an increase in real consumption
in all periods and, generally, an increase in the de-
mand for leisure (see Table A-15). Households mar-
ginally reduce their offer of labor services as the in-
come effects of
f^^^Hffffffff^^^fuff^^^ffff^t higher real earn-
ings dominate the
substitution ef-
fects of lower
goods prices.
Table A-l|;;6ffference in Personal
Between" the Control-,
' *"* N
?^v
f 'St
/ '
'
v •.>'
>, "V »*,
" '••• , "V
-^at"
-- - ,,-,.
..-1973,,
' '1?74 J
1976?'
1977
" 19¥8;A
- 1979'. .:-
1 19'80 "
1981
1982-'-'
rt^'A-.",*
ISfJJJ
1984
1985
1987
' 1988
" 19&9
" 1990
«»^»»
Norainai ^,
; Change'
«0:6i
, -our -
•8.1ft, „
•8.10
-0,09. x
' -0.11 ,,
" '-0.12 -
'-0.13 -.
" ' ' -0.12
w" * A 1 -3
",V/ ""'^ s ,s
•".•"*< -*0 15
,,Z..0.19"%
* >C^ ^ ,, _,
, ,, f) 1 Q '
!.7'T-0.19 -N
,^/-0.-17
-> --0.17
^-?>0ii8
,',;,; s^<',, > ,
. * * jVS *
Real %'
^haiige
, - Oi33v,,
" J*3k"
0;39': -.
' - o:54/-.
,,„ 0.63'"=
„, ,,0.68'"
- - -0-7i'>>;'
- < -0.74:T'
.*>, t^r
'f\ ''^ '^'?
y,o.)' >
0 'So-1^ *
0,8&'-V
:,,„,; 0.94 •'•-
,,,;t_0;98;:
":'v>|ri,03
^^1104-r
",,VOi,,,,
^
., <^.,
inc iiiuicasc in
consumption is
dampened by an
increase in the
rate of return that
produces greater
investment (and
personal sav-
'. , ings).
Finally, tech-
nical change is a
\7^t"17 1Yf\f\f\ftC\ttt
very imporicini.
aspect of the sup-
, , ply-side adjust-
"" ",;, ments under the
n*x- no-control sce-
#?. , „ nario. Lower fac-
'-, * tor prices in-
- - "** crease the endog-
-; *;- enous rates of
' J; S;
technical change in those industries that are factor-
using. Lower rental prices for capital benefit the capi-
tal-using sectors, lower materials prices benefit the
materials-using sectors, and lower energy prices ben-
efit the energy-using sectors. On balance, a signifi-
cant portion of the increase in economic growth is
attributable to accelerated productivity growth. Un-
der the no-control scenario, economic growth aver-
ages 0.05 percentage points higher over the interval
1973-1990. The increased availability of capital ac-
counts for 60 percent of this increase while faster pro-
ductivity growth accounts for the remaining 40 per-
cent. Thus, the principal effect arising from the costs
associated with CAA initiatives is to slow the
economy's rates of capital accumulation and produc-
tivity growth. This finding is consistent with recent
analyses suggesting a potential association between
higher reported air, water, and solid waste pollution
abatement costs and lower plant-level productivity in
some manufacturing industries (Gray and Shadbegian,
1993 and 1995).
As with the cost and expenditure data presented
above, it is possible to present the stream of GNP and
consumption changes as single values by discounting
the streams to a single year. Table A-16 summarizes
the results of the discounting procedure, and also in-
cludes discounted expenditure and annualized cost
data for reference. Accumulated (and discounted to
1990) losses to GNP over the 1973-1990 period were
half again as large as expenditures during the same
period, and approximately twice as large as annual-
ized costs. Losses in household consumption were
approximately as great as annualized costs.
^A-ii6,,,GNPaftd
ra&countedto 1990 fS1§9"()"bi
GNP , ,„_,„„ '
Household C,6ft'su!rfj>ti(Mi
HH ai
657-
loos • iVsi"'
569" '/$&••
iBacroecoaoMc
Jorgeason'etat(1993),
'
25 Not surprisingly, at the industry level, the principal beneficiaries in the long run of eliminating the costs associated with air
pollution abatement are the most heavily regulated industries. The largest changes in industry prices and outputs occur in the motor
vehicles industry. Other industries that benefit significantly from the elimination of environmental controls are refined petroleum
products, electric utilities, and other transportation equipment. Turning to manufacturing industries, metal mining and the primary
metals have the largest gains in output from elimination of air pollution controls.
A-21
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Figure A-l. Percent Difference in Real Investment Between Control and No-control Scenarios.
i
a
1873 1874 1875 1878 1877 187a 1878 1880 1881 1882 1883 1984 1985 1988 1887 1988 1989 1980
Year
Figure A-2. Percent Difference in Price of Output by Sector Between Control and No-control
Scenario for 1990.
1 23 4 56 78 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Sector
A-22
-------
Appendix A: Cost and Macroeconomic Modeling
Although they have value as descriptors of the
magnitude of changes in economic activity, neither
GNP nor consumption changes are perfect measures
of changes in social welfare. A better measure is
Equivalent Variations (EVs), which measure the
change in income that is equivalent to the change in
(lifetime) welfare due to removal of the CAA. As part
of its macroeconomic exercise, EPA measured the EVs
associated with removal of the CAA. Elimination of
CAA compliance costs (disregarding benefits) repre-
sents a welfare gain of $493 billion to $621 billion,
depending on assumptions used in the analysis.26 This
result does not differ greatly from the range of results
represented by expenditures, anualized costs, and con-
sumption changes.
Prices
One principal consequence of the Clean Air Act
is that it changes prices. The largest price reductions
accrue to the most heavily regulated industries which
are the large energy producers and consumers (see
Table A-17). But these are also the most capital in-
tensive sectors and it is the investment effects that are
the dominant influences in altering the course of the
economy. Focusing on energy prices, under the
no-control scenario the price of coal in 1990 declines
by 1.3 percent, refined petroleum declines by 3.03
oontrol Scenarios ,
'Year
-0.44
P<*,trr»jftnm
-5:99
;EIecitri
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Figure A-3. Percent Difference in Quantity of Output by Sector Between Control and No-
control Scenario for 1990.
1234567
9 10 11 12 13 14 15 16 17 16 19 20 21 22 23 24 25 26 27 26 29 30 31 32 33 34 35
Sector
Figure A-4. Percent Difference in Employment by Sector Between Control and No-control
Scenario for 1990.
™l
: 16 17 16 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Sector
A-24
-------
Appendix A: Cost and Macroeconomic Modeling
that the largest increase hi prices also occurred hi the
motor vehicles sector. The 3.8 percent reduction hi
prices produces an increase in output of 5.3 percent
relative to the base case.
Significant output effects are also seen hi the pe-
troleum refining sector (sector 16) with a 3.2 percent
increase, hi electricity (sector 30) with a 3.0 percent
increase, and in other transportation equipment (sec-
tor 25) with a 1.6 percent increase. The large gains in
output for these industries are mostly due to the de-
cline in their prices. In manufacturing, the sectors
exhibiting the most significant output effects are metal
mining (sector 2) with a 2.0 percent increase, and pri-
mary metals (sector 20) with a 1.8 percent increase.
Twenty of the remaining industries exhibit increase
in output of less than 0.9 percent after pollution con-
trols are removed.
While most sectors increase output under the
no-control scenario, a few sectors decline in size in
the absence of air pollution controls. The most no-
table of these are food and kindred products (sector
7) which decline by 0.5 percent, furniture and fixtures
(sector 12) which decline by 0.6 percent, and rubber
and plastic products (sector 17) which decline by 0.3
percent. These sectors are among the least capital in-
tensive, so the fall in the rental price of capital ser-
vices has little effect on the prices of outputs. Buyers
of the commodities produced by these industries face
higher relative prices and substitute other commodi-
ties in both intermediate and final demand. The rest
of the sectors are largely unaffected by environmen-
tal regulations.
Changes in Employment Across
Industries
The effect of the CAA on employment presents a
much more complicated picture. Although Jorgenson-
Wilcoxen is a full-employment model and cannot be
used to simulate unemployment effects, it is useful
for gaining insights about changes in the patterns of
employment across industries. Percentage changes in
employment by sector for 1990 are presented hi Fig-
ure A-4.
For 1990, the most significant changes in the level
of employment relative to the control scenario occur
in motor vehicles (sector 24) which increases 1.2 per-
cent, other transportation equipment (sector 25) which
increases 0.8 percent, electric utilities (sector 30)
which increases 0.7 percent, and primary metals (sec-
tor 20) which increases 0.6 percent. The level of em-
ployment is higher relative to the control case in 10
other industries.
For a few sectors, the no-control scenario results
in changes hi real wages which cause reductions in
employment. The most notable reductions in employ-
ment under the no-control scenario occur in tobacco
manufacturing (sector 8) which declines 1.2 percent,
furniture and fixtures (sector 12) which declines 0.8
percent, rubber and plastic products (sector 17) which
declines 0.8 percent, food and kindred products (sec-
tor 7) which declines 0.7 percent, stone, clay and glass
products (sector 19) which declines 0.6 percent, and
instruments (sector 26) which declines 0.6 percent.
These sectors are generally those in which the level
of output was lower hi 1990 relative to the control
scenario, since they are among the least capital inten-
sive and the fall in the rental price of capital services
has little effect on the prices of outputs. Buyers of the
commodities produced by these industries face higher
relative prices and substitute other commodities hi
both intermediate and final demand. It is interesting
to note that several of the least capital intensive sec-
tors experience insignificant employment effects in
the short run (1975) under the no-control scenario,
but increasingly adverse effects over the 20-year pe-
riod of analysis. These include food and kindred prod-
ucts, furniture and fixtures, rubber and plastic prod-
ucts, stone, clay and glass products, and instruments.
Examination of the transition of employment in
the economy from the initial equilibrium to 1990 re-
veals that the employment effects of the C AA on motor
vehicles, transportation equipment, electric utilities,
and primary metals persist over the entire period of
analysis. Employment varies from: an increase of 1.7
percent in 1975 to 1.2 percent in 1990 in motor ve-
hicles; an increase of 0.7 hi 1975 to 0.8 percent hi
1990 hi transportation equipment; an increase of 1.2
percent hi 1975 to 0.7 percent hi 1990 hi electric utili-
ties; and an increase of 0.8 percent in 1975 to 0.6 per-
cent in 1990.
A-25
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Uncertainties in the Cost
Analysis
Potential Sources of Error in the Cost
Data
Because of the importance of the Cost of Clean
data for this assessment, the project team investigated
potential sources of error due to the use of industry's
self-reported costs of compliance with air pollution
abatement requirements. Concerns about the accuracy
of responses include (1) misreporting by firms in re-
sponse to federal agency surveys, and (2) omission of
important categories of compliance cost from the data
collected or reported by these federal agencies.27 Table
A-18 contains a summary of the results of the analy-
sis. This analysis is consistent with the findings of
two recent studies comparing combined air, water, and
solid waste pollution abatement costs, as reported in
federal abatement cost surveys, to their observed ef-
fects on productivity levels. These studies suggest that,
since observed productivity decreases exceed those
expected to result from the reported abatement costs,
there may be additional pollution abatement costs not
captured or reported in the survey data, and that total
abatement costs for the three manufacturing indus-
tries studied may be under-reported by as much as a
factor of two in the most extreme case (Gray and
Shadbegian, 1993 and 1995; Gray, 1996).
The major finding from this analysis indicates that
total O&M costs are likely to be under-reported due
to exclusion of private research and development
Table A-l8. Potential Sources of E^or'aitf} Their,Effec,fc.Gtt Total''CoMs^'f Compliance,
Source of Error •
Lack. of Data at Firm Level
' • '-^'Percent -Unknown ,^;
Misallocation of Costs;
Inclusion of OSBA and Other "' - *"•*
Regulatory Costs ,„ c
. , ',<
Over-reported
V^ . ft' '& f-W-S ,„,
- - Percent Unknown
Exclusion of Solid Waste Disposal 'Cqsls
Related to Air Pollution Abatement
Exclusion of Costs: ' "*' - \
Exclusion of Private R&0 Expenses •'
Exclusion of Energy Ose by;Potfu1tib1a'x
Abatement Devices'*' "'. '-'•*'-.^^
Exclusion of Depreciation Expenses^1-
Exclusion of Recovered Costs
aries by year) '
' ^
Omissio n of Smal 1 Pi rms'
NET EFFECT ' '
'•> Energy outlays are part of the data, on ;O^M
considered along with other opealing expea'
accumulattoo process, as the undepreciated ca
ptoduceis and consumeis. ° " " ,v"
, in fhejgvy, njotfel,
D^«ec'ilfe)n'Js>»|)r
-------
Appendix A: Cost andMacroeconomic Modeling
(R&D) expenditures. Note, however, that although
these costs were excluded from those used for the
macroeconomic modeling, they were included in the
overall direct cost estimate of the CAA; see "Other
Direct Costs," above. These costs are excluded from
the macromodeling because they cannot be disaggre-
gated by industry and, more importantly, because there
is no information on what was purchased or obtained
as a result of these expenditures.
Based on the need indicated by the lEc review,
modifications to the BEA data were made to remedy
some of the biases noted above. In particular, recov-
ered costs for stationary source air pollution, e.g. sul-
fur removed using scrubbers that is then sold in the
chemical market, have been accounted for in the data
set used in the model runs.
s as a Perceata'gesef-Capital 'Stock "
I990'a&?lars)--- •• --
1973
'-1976
,
• 1979
1962
1964
. 1965s
-19B8
"14J880 14,684,. ,,4-,77'a
32,77:3,
-41,331'-'
,49,448"
'57,299..
3W2 '6,768
.- SS^SW"' 6,S27
46,612^,6,991
-'58,232', 3.8,791
74,366 *
02,381?'
-"-74,173, ;?
95,879 V^7eB06 8,605
-107,'082 ', '-,-79,713, '8;<
-80,300- 8:,'l
8,5
-8SS42
117,263 ,
i'22,182 -
127,'394
v% 'O',60 •?..v'&'6Q''
"" 0.32 -' o'JsT
-0,2C:'
-vO;1,4-v , 0,15
.09/HrP'tl- r
' 0.07
K stock",l
',-fromTable-'-A-lO., -.^-^^,.- •-- • -- -";;--'>-
K" is tiie.siatioiia^ryYbuirCe control ,capi!&i stock.Jess, ,"••
ied bVimortfeation ,at, sfefrom Table- - -
* »,"•'-< ~ - -- .
'.I'fi&feal twp.cohjrnns ar^iatips: <
'stpcJc^andO&lVlimd'edby^etcapital. >v^"'-'','-".,.
An additional set of concerns relates directly to
reporting of costs by firms. Some have noted an un-
expected temporal pattern of stationary source con-
trol expenditures in the BEA data that might lead one
to question the accuracy of the Census survey re-
sponses. One would expect that stationary source
O&M expenditures over time would be roughly pro-
portional to the accumulated stationary source con-
trol capital stock. Yet, as illustrated in Table A-19,
O&M expenditures as a fraction of accumulated capi-
tal stock decline over time (even if one discounts the
first few years because of the dramatic percentage
increases in capital stock during those years). It is true
that the ratio of O&M expenditures to the depreci-
ated capital stock (in the far right column, labeled "net
K") is reasonably stable after 1981. The depreciation
shown here, however, is ^financial depreciation only,
depicting the declining value of a piece of equipment
over time, rather than a measure of physical asset
shrinkage. Assuming a twenty-year useful lifetime,
all of the stationary source control capital stock put in
place since 1972 could conceivably still be in place in
1990. If anything, one would expect the O&M/K ra-
tio to increase as the capital depreciates (i.e., ages),
until the equipment is scrapped, because aging equip-
ment requires increasing maintenance. Consequently,
one might infer from this information that firms have
systematically under-reported O&M expenditures, or
have over-reported capital expenditures.
The apparent anomaly might be explained by an
examination of the types of O&M expenditures re-
ported. If more than a token percentage of O&M ex-
penditures are unrelated to "operation and mainte-
nance" of pollution control devices, then the observed
O&M/K ratio would not appear unusual.
The Census PACE survey28 required respondents
to report air pollution abatement O&M expenses in
the following categories: salaries and wages; fuel and
electricity; contract work; and materials, leasing, and
"miscellaneous."29 In later versions of the survey,
additional information relating to the types of expenses
to report was provided as a guide to respondents. The
types of expenses listed that are relevant to air pollu-
tion abatement include:
28 Pollution Abatement Costs and Expenditures, various years.
29 Census also requested a reporting of "depreciation" expenses as a component of O&M. BEA, however, removed depreciation
expense from the reported O&M costs because retaining depreciation would have amounted to double-counting, since BEA also
reported capital expenditures.
A-27
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
(1) operating and maintaining pollution abate-
ment equipment;
(2) fuel and power costs for operating pollution
abatement equipment;
(3) parts for pollution abatement equipment re-
placement and repair;
(4) testing and monitoring of emissions;
(5) incremental costs for consumption of envi-
ronmentally preferable materials and fuels;
(6) conducting environmental studies for devel-
opment or expansion;
(7) leasing of pollution abatement equipment;
(8) compliance and environmental auditing;
(9) salaries and wages for time spent completing
environmental reporting requirements; and
(10) developing pollution abatement operating
procedures.30
The magnitude of the expenditures associated with
the first three items should be correlated with the size
of the existing stock of air pollution abatement capi-
tal. Expenditures associated with items four through
ten, however, should be independent of the size of the
existing capital stock (expenditures associated with
item seven, leasing of pollution abatement equipment,
could be negatively correlated with the size of the
capital stock). If items four through ten account for a
non-negligible proportion of total O&M expenditures,
and i/respondents included these cost categories even
though they were not explicitly listed in the survey
instructions before 1991, then one would expect to
see the O&M/K ratio declining during the study pe-
riod. Thus, even though it is possible that O&M ex-
penditures are underreported (or that capital expendi-
tures are overreported), one cannot be certain.
Mobile Source Costs
For the section 812 analysis, EPA used the best
available information on the estimated cost of mobile
source air pollution control. Several other sources of
cost estimates exist, however, including a cost series
produced by the Department of Commerce Bureau of
Economic Analysis (BEA). The BEA cost series is
summarized in Table A-20. The BEA estimates dif-
fer significantly from EPA estimates, particularly with
respect to estimates of capital costs and the "fuel price
penalty" associated with the use of unleaded gaso-
line.
EPA's capital cost estimates are based on esti-
mates of the cost of equipment required by mobile
1987, .„ 4,475 <
l-9?8 ,,,,v«,4,267,/- 7,313
•1989- --«4;760- * 7,743"
1990" ""'"4^69 7, -S.688
BEA Estimates
1986 '4;09ff:'-" ,7,072
19^7 '- '3^82 ' ;5,$,43
1988 3,-120- 6,230'
•r ,',„,,«• ,-5,*',,,"',,
312- '; "<•»
277'"' "'
130
-12,914"
13O37.
312
, - ,I'82
..Ml'1''
1990.-, .4402- 6.799
121 v,;,v 161""" ,, 9,632
229 - - J53U''£ '£,9$St
' "2'0tf """" :154 ••- "'11.255
1995, t
,,,
btltttiaa Aba»eta art. '«ad Coaaol ,
' j for •„
source regulations. BEA's estimates are based on sur-
vey data from the Bureau of Labor Statistics (BLS)
that measures the increase in the per-automobile cost
(relative to the previous model year) due to pollution
control and fuel economy changes for that model year.
The difference in approach is significant: BEA's an-
nual capital cost estimates exceed EPA's by a factor
of (roughly) two. EPA may underestimate costs to the
extent that engineering cost estimates of components
exclude design and development costs for those com-
ponents. The BLS estimates add the incremental an-
nual costs to all past costs to derive total current-year
costs. Such an approach overestimates costs to the
extent that it fails' to account for cost savings due to
changes in component mixes over time.
Some mobile source pollution control devices re-
quired the use of unleaded fuel. Unleaded gasoline is
more costly to produce than is leaded gasoline, and
generally has a greater retail price, thus imposing a
cost on consumers. EPA estimated the "fuel price pen-
alty" by using a petroleum refinery cost model to deter-
mine the expected difference in production cost be-
tween leaded and unleaded gasoline. BEA's "fuel price
penalty" was the difference between the retail price
of unleaded gasoline and that of leaded gasoline.
A detailed description of the data sources, ana-
lytic methods, and assumptions that underlie the EPA
and BEA mobile source cost estimates can be found
in McConnell et al. (1995).
30 Pollution Abatement Costs and Expenditures, 1992, pg. A-9.
-------
Appendix A: Cost andMacroeconomic Modeling
Stationary Source Cost Estimate
Revisions
Endogenous Productivity Growth in the
Macro Model
As noted above, the costs used for stationary
sources in the macro-modeling (and retained in this
cost analysis) were projected for several years in the
late 1980s. Since that time, BEA has released histori-
cal expenditure estimates for those years based on
survey data. A comparison of the expenditure series
can be found in Table A-21. Apparently, EPA's pro-
jections overestimated stationary source compliance
expenditures by approximately $2 billion per year for
the period 1987-1990. Since expenditures from all
sources are estimated to be $18 billion -$19 billion
(current dollars) per year during 1987-1990, this im-
plies that EPA has overestimated compliance expen-
ditures by more than ten percent during this period.
Although a substantial overstatement for those years,
the $2 billion per year overestimate would have little
impact (probably less than two percent) on the dis-
counted present value, in 1990 dollars, of the 1973-
1990 expenditure stream.
" r^/Capitar ' Ne£!j,,,
,,, JExjk" '*
,,1974'
1975-
'---;' 1,118
1,380
2,802
'""' 1,180
' '"""1344
"?Z?"~ry/"/ /<;<
\m
1978
1979
'1980
1981
1982
1983
',' 3,93,5,
\ 4,6,34 '
' 5^63
1985
,1986
1987
1989
1990
7,663
9326''
11,900^
//j-iwio*
- - K36V
13?725
, 16,157
> 15340
-- 14,521
! 1^420-
1,28,9,,
1,1-36
552 -
274,
118"
165~
-.(331)
'(453)
(631)"
(271)''
(719)
•>.•.••
309
1,209
,,2,99,6 ,
3^18
4,235,,-
4,427 ^
4,995 '•'
4,522
3,672
1,792
,2320 ,
-2,252
1-.876
1,972,'
'U70 -
658
420,
183
"(55)
Inspection and maintenance costs less fuel density savings and
°" ''
For each industry in the simulation, the JW model
separates price-induced changes in factor use from
changes resulting strictly from technical change. Thus,
simulated productivity growth for each industry has
two components: (a) an exogenous component that
varies over time, and (b) an endogenous component
that varies with policy changes. Some reviewers have
noted that, although not incorrect, use of endogenous
productivity growth is uncommon in the economic
growth literature. EPA conducted a sensitivity run of
the J/W model, setting endogenous growth parameters
to zero (i.e., removing endogenous productivity
growth from the model).31
Endogenous productivity growth is an important
factor in the J/W model. For example, for the period
1973-1990, removal of the endogenous productivity
growth assumptions reduces household income by 2.9
to 3.0 percent (depending on whether one uses a world
with CAA or one without CAA as the baseline). In
comparison, removal of CAA compliance costs re-
sults in a 0.6 to 0.7 percent change in household in-
come (depending on whether one uses, as a baseline,
a world with or one without endgenous productivity
growth). That is, use of the endogenous productivity
growth assumption has four to five times the impact
of that of CAA compliance costs.
Although very important to the simulated growth
of the economy within any policy setting, the endog-
enous productivity growth assumption is less impor-
tant across policy settings. Under the base (i.e., "with
endogenous productivity growth") scenario, the ag-
gregate welfare effect (measured as EVs, see above)
of CAA compliance costs and indirect effects is esti-
mated to be 493 billion to 621 billion in 1990 dollars.
If one removes the endogenous productivity growth
assumption, the aggregate welfare effect declines to
the range 391 billion to 494 billion in 1990 dollars
(Jorgenson et al., 1993, pg. 6-15), areduction of about
twenty percent.
31 For greater detail, see Jorgenson et al., 1993.
A-29
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Amortization Period for Stationary
Source Plant and Equipment
In developing annualized costs, stationary source
capital expenditues were amortized over a twenty-year
period. That is, it was assumed that plant and equip-
ment would depreciate over twenty years. It is pos-
sible that stationary source plant and equipment has,
on average, a useful lifetime significantly greater than
twenty years. The Project Team tested the sensitivity
of the cost analysis results to changes in stationary
source capital amortization periods.
Table A-22 presents total annualized compliance
costs assuming a 40-year amortization period for sta-
tionary source capital expenditures (all other cost com-
ponents are unchanged from the base analysis). All
costs are in 1990-value dollars, ad" three alternative
discount rates are used in the annualization period.
Table A-23 presents the results discounted to 1990,
and compared to the base case results (i.e., using a
twenty-year amortization period). Doubling the am-
ortization period to 40 years decreases the 1990 present
value of the 1973-1990 cost stream by approximately
40 billion dollars. This represents a change of six per-
cent to nine percent, depending on the discount rate
employed.
tpital Amortiiatidn FeriodV'WS-/;'.
-a-9.90.(millk
1973
1974
1975
1976
1977
1978
1979
1980
1981
.!$§?,„
Am.
1.984.--
-1'9S5
1986
'1987
.19>S
1990
.fttfj.%.
10,8,0'l 'l4899 11,008-
1-2,875
tjjtJ^O1, , »-*^»w>r *s,v »•
rf'^ S f ft" '/y v" /S-/«S'"yr'ai»(jr
-------
Appendix A: Cost and Macroeconomic Modeling
Cost and Macroeconomic
Modeling References
Chase Econometrics Associates, Inc. 1976. "The
Macroeconomic Impacts of Federal Pollution
Control Programs: 1976 Assessment." Report
prepared for the Council on .Environmental
Quality and the Environmental Protection
Agency.
Congressional Budget Office. 1990. Carbon Charges
as a Response to Global Warming: The Ef-
fects of Taxing Fossil Fuels. Washington, DC,
U.S. Government Printing Office.
Data Resources, Inc. 1979. "The Macroeconomic
Impacts of Federal Pollution Control Pro-
grams: 1978 Assessment," Report prepared
for the Environmental Protection Agency and
the Council on Environmental Quality.
Data Resources, Inc. 1981. "The Macroeconomic
Impact of Federal Pollution Control Pro-
grams: 1981 Assessment," Report prepared
for the Environmental Protection Agency.
July 17.
Economic Report of the President. 1995. U.S. Gov-
ernment Printing Office, Washington, DC.
February.
Farber, Kit D. and G. Rutledge. 1989. "Pollution
Abatement and Control Expenditures: Meth-
ods and Sources for Current-Dollar Esti-
mates." Unpublished Paper for U.S. Depart-
ment of Commerce, Bureau of Economic
Analysis. October.
Freeman, A.M. 1978. "Air and Water Pollution
Policy," in P.R. Portney (ed.), Current Issues
in U.S. Environmental Policy. Johns Hopkins
University Press, Baltimore.
Gray, Wayne B. 1996. Personal communication with
Michael Hester of Industrial Economics, Inc.
December 4.
Gray, Wayne B. and Ronald J. Shadbegian. 1993.
"Environmental Regulation and Manufactur-
ing Productivity at the Plant Level," Center
for Economic Studies Discussion Paper, CES
93-6. March.
Gray, Wayne B. and Ronald J. Shadbegian. 1995.
"Pollution Abatement Costs, Regulation, and
Plant-Level Productivity," National Bureau of
Economic Research, Inc., Working Paper
Series, Working Paper No. 4994. January.
Hazilla, M. and R.J. Kopp. 1990. "Social Cost of En-
vironmental Quality Regulations: A General
Equilibrium Analysis," Journal of Political
Economy, Vol. 98, No. 4. August.
Industrial Economics, Incorporated. 1991. "Sources
of Error in Reported Costs of Compliance
with Air Pollution Abatement Requirements,"
memorandum to Jim DeMocker, EPA/OAR.
October 16.
Jorgenson, Dale W. and Barbara M. Fraumeni. 1989.
"The Accumulation of Human and Nonhu-
man Capital, 1948-1984," in R.E. Lipsey and
H.S. Tice, eds., The Measurement of Saving,
Investment, and Wealth. University of Chi-
cago Press, Chicago, II.
Jorgenson, Dale W. and Barbara M. Fraumeni. 1981.
"Relative Prices and Technical Change," in
E. Berndt and B. Field, eds., Modeling and
Measuring Natural Resource Substitution.
MIT Press, Cambridge, MA.
Jorgenson, Dale W., Richard J. Goettle, Daniel
Gaynor, Peter J. Wilcoxen, and Daniel T.
Slesnick. 1993. "The Clean Air Act and the
U.S. Economy," Final report of Results and
Findings to the U.S. EPA. August.
Jorgenson, Dale W. and Peter J. Wilcoxen. 1990a.
"Environmental Regulation and U.S. Eco-
nomic Growth," inRAND Journal of Econom-
ics, Vol. 21, No. 2, pp. 314-340.
Jorgenson, Dale W. and Peter J. Wilcoxen. 1990c.
"Intertemporal General Equilibrium Model-
ing of U.S. Environmental Regulation," in
Journal of Policy Modeling, Vol. 12, No. 4,
pp. 715-744.
Jorgenson, Dale W. and Peter J. Wilcoxen. 1993. "En-
ergy, the Environment and Economic
Growth," in Handbook of Natural Resource
and Energy Economics, Allen V. Kneese and
James L. Sweeney, eds., Volume 3, Chapter
• 27. North-Holland, Amsterdam, forthcoming.
A-31
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Kokoski, Mary F. and V. Kerry Smith. 1987. "A Gen-
eral Equilibrium Analysis of Partial-Equilib-
rium Welfare Measures: The Case of Climate
Change," American Economic Review, Vol.
77, No. 3, pp. 331-341.
McConnell, Virginia, Margaret A. Walls, and Win-
ston Harrington. 1995. "Evaluating the Costs
of Compliance with Mobile Source Emission
Control Requirements: Retrospective Analy-
sis," Resources for the Future Discussion Pa-
per.
Schwartz, Joel. 1991. "Fuel Economy Benefits."
Memorandum to Joe Somers and Jim
DeMocker. December 12.
Somers, J.H. 1991. "Fuel Economy Penalties for Sec-
tion 812 Report." Memorandum to Anne
Grambsch and Joel Schwartz. December 23.
U.S. Department of Commerce. Government Fi-
nances, various issues. Bureau of the Census.
U.S. Department of Commerce. "Pollution Abatement
and Control Expenditures," Survey of Current
Business, various issues. Bureau of Economic
Analysis.
U.S. Department of Commerce. "Pollution Abatement
Costs and Expenditures," Current Industrial
Reports, various issues. Bureau of the Cen-
sus.
U.S. Environmental Protection Agency (EPA). 1985.
Costs and Benefits of Reducing Lead in
Gasolines: Final Regulatory Impact Analy-
sis. Office of Policy Analysis, EPA-230-05-
85-006. February.
U.S. Environmental Protection Agency (EPA). 1990.
Environmental Investments: The Cost of a
Clean Environment, Report to the Congress.
Office of Policy, Planning and Evaluation.
EPA-230-12-90-084. December.
Verleger, Philip K., Jr. 1992. "Clean Air Regulation
and the L.A. Riots," The Wall Street Journal,
Tuesday, May 19. p. A14.
Walsh, M.P. 1991. "Motor Vehicles and Fuels: The
Problem." in EPA Journal, Vol. 17, No. 1, p.
12,
Wilcoxen, Peter J. 1988. The Effects of Environmen-
tal Regulation and Energy Prices on U.S.
Economic Performance, Doctoral thesis pre-
sented to the Department of Economics at
Harvard University, Cambridge, MA. Decem-
ber.
A-32
-------
Appendix B: Emissions Modeling
Introduction
This appendix provides additional details of the
methodologies used to estimate control and no-control
scenario emissions and the results obtained by these
methods. Methodological information and results are
provided for each of the six principal emission sec-
tors: industrial combustion, industrial processes, elec-
tric utilities, on-highway vehicles, off-highway ve-
hicles, and commercial/residential sources.
The initial section of this appendix assesses the
emissions projections presented in this analysis by (1)
comparing the 1970 to 1990 control scenario projec-
tions with recent EPA Trends report estimates for the
same years and (2) comparing the 1970 to 1990 trend
in no-control scenario projections with 1950 to 1970
emissions as reported in Trends. The first compari-
son indicates that control scenario emissions projec-
tions approximate, but do not precisely match, the EPA
Trends data. The reason for this mismatch is discussed
below. The second comparison is useful for demon-
strating that pre-1970 emissions trends would not pro-
vide a satisfactory basis for extrapolating emissions
trends into the 1970 to 1990 period. The inability to
simply extrapolate pre-1970 trends provides further
justification for applying the present modeling meth-
odologies to generate no-control scenario emissions
projections.
The remainder of the appendix provides further
details of the emissions modeling conducted in sup-
port of the present analysis, and is largely adapted
from the draft report "The Impact of the Clean Air
Act on 1970 to 1990 Emissions; section 812 retro-
spective analysis," March 1,1995 by Pechan Associ-
ates. The draft Pechan report surveys the methodolo-
gies and results associated with the sector-specific
emission modeling efforts by Argonne National Labo-
ratory (ANL), ICF Resources Incorporated (ICF), Abt
Associates (Abt), and the Environmental Law Insti-
tute (ELI).
Comparison of Emissions
Projections with Other EPA Data
Control Scenario Projections Versus
EPA Trends Projections
The control scenario emission results are similar,
but not identical, to official EPA historical emission
estimates provided by the EPA National Air Pollut-
ant Emission Trends Reports.1 Comparisons between
the current estimates and the Trends data for SO2, NOx,
VOC, CO, and TSP are presented in Figures B-l, B-
2, B-3, B-4, and B-5 respectively. More detailed tables
providing emission estimates by sector and by target
year for TSP, SO2, NOx, VOC, CO, and Lead are pre-
sented in Tables B-16, B-17, B-18, B-19, B-20, and
B-21, respectively, at the end of this appendix.
Though the EPA Trends and the present study
emission profiles are similar to each other, they should
not be expected to match precisely. This is because
the emission estimates developed for the present study
are based on modeled macroeconomic and emission
sector conditions. Even though the macroeconomic
and sector models themselves are constructed and
calibrated using historical data, modeled replications
-—ef historical trends would not be expected to precisely
-capture actual historical events and conditions which
affect emissions. Relying on modeled historical sce-
narios is considered reasonable for the present analy-
sis since its purpose is to estimate the differences be-
tween conditions with and without the CAA. Com-
paring actual historical emissions with modeled no-
control emissions would lead to an inconsistent basis
for comparisons between scenarios. Using models for
both scenarios allows potential model biases to es-
sentially cancel out.
In general, however, these comparisons show
close correspondence between control scenario and
Trends estimates with the largest differences occur-
1EPA/OAQPS, "National Air'Pollutant Emission Trends 1900 -1994," EPA-454/R-95-011, October 1995.
_
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Figure B-l. Comparison of Control, No-control, and
Trends SO2 Emission Estimates.
40
30
20
1 10
w
t
Control
.No-Control
.TRENDS
1950 1960
1970
Year
1980 1990
Figure B-2. Comparison of Control, No-control, and
Trends NOX Emission Estimates.
40
30
•220
10
1950 1960 1970 1980 1990
Year
Figure B-3. Comparison of Control, No-control, and
Trends VOC Emission Estimates.
40
30
20
10
I I I
I I I
1950 1960 1970
Year
1980 1990
Figure B-4. Comparison of Control, No-control, and
Trends CO Emission Estimates.
200
I 150
hoo
so
1950 1960 1970 1980 1990
Year
Figure B-5. Comparison of Control, No-control, and
Trends TSP Emission Estimates.
40
30
& 1 20
10
1950 1960 1970 1980 1990
Year
B-2
-------
Appendix B: Emissions Modeling
ring for VOC and CO emissions. The Trends report
VOC estimates are generally higher than the control
scenario estimates due to the inclusion of Waste Dis-
posal and Recycling as a VOC source in the Trends
report. This inconsistency is of no consequence since
Waste Disposal and Recycling sources were essen-
tially uncontrolled by the historical CAA and there-
fore do not appear as a difference between the control
and no-control scenarios. The higher CO emission
estimates in the Trends Report are primarily associ-
ated with higher off-highway vehicle emissions esti-
mates. Again, since off-highway emissions do not
change between the control and no-control scenario
in the present analysis, this inconsistency is of no con-
sequence.
No-Control Scenario Projections Versus
Historical EPA Trends Data
Comparisons between the control scenario emis-
sions estimates generated for the present study and
1970 to 1990 emissions estimates obtained from the
Trends Report are useful for assessing the reasonable-
ness of the control scenario estimates. As indicated
above, there is close correspondence between the con-
trol scenario and the Trends Report. It may also be
useful to compare the pre-1970 historical emissions
data from the Trends Report2 with the no-control sce-
nario estimates presented herein to assess whether
these pre-1970 trends can be reasonably extrapolated
to the 1970 to 1990 period. In addition, examination
of any significant changes in emissions trends between
the pre-1970 Trends data and post-1970 ncbeontrol
projections might indicate flaws in^the emissions
modeling conducted for the present study.
ForJSOjrthe 1950 to 1970 Trends data in Figure
—EPl demonstrate the effects of the huge increase in
fossil fuel combustion between 1960 and 1970. This
net increase occurred, despite the obsolescence of coal-
fired locomotives and reductions in coal refuse burn-
ing, largely because utility emissions nearly doubled
between 1950 and 1960, and nearly doubled again
between 1960 and 1970.3 Although no-control sce-
nario projections for the post-1970 period show sig-
nificant additional increases in SO2 emissions, the rate
of growth is markedly slower than during the 1950 to
1970 period.
The Trends data for 1950 to 1970 NO shown in
X
Figure B-2 indicate the steady increase in emissions
resulting from increased combustion of natural gas
and gasoline.4 The post-1970 emissions estimates
derived for the present study reflect a continuation of
this trend.
Emissions of VOCs increased steadily over the
1950 to 1970 period, as shown in Figure B-3, prima-
rily due to increases in industrial production and ve-
hicular travel.5 The no-control scenario emission es-
timates continue this trend throughout the 1970 to 1990
period, with some acceleration of the rate of change
due to the rapid increase in VMT projected under this
scenario.
The Trends data shown in Figure B-4 for CO in-
dicate an overall increase between 1950 and 1970. This
increase occurred despite significant reductions in
emissions from stationary source fuel combustion and
industrial processes because mobile source emissions
nearly doubled during this period.6 Under the no-con-
trol scenario of the present study, additional reduc-
tions from stationary sources are not available to off-
set the transportation-related increases; therefore, the
rate of increase in CO emissions after 1970 under the
no-control scenario reflects the rapid increase in mo-
bile source emissions caused by increases in vehicle
miles traveled.
Finally, Figure B-5 demonstrates a directional
shift in emissions of primary particulates between the
1950 to 1970 Trends data and the post-1970 no-con-
trol scenario. The declining trend from 1950 to 1970
indicated by the Trends data, however, is largely due
to reductions in use of coal-fired locomotives, reduc-
tions hi residential coal-burning, coarse (i.e., visible)
particle emissions controls installed on fossil fuel com-
bustors and industrial processes, and reductions in
forest fires and other open burning.7 Since the reduc-
tions achievable from these sources were largely
2 While 1970 to 1990 Trends data were obtained from more recent Trends reports, the 1950 to 1970 data were obtained from the
November 1991 report since this was the last year the Trends report series included data for this period.
3 U.S. EPA, "National Air Pollutant Emission Estimates, 1940 -1990", EPA-450/4-91-026, November 1991, Table 4, p. 16.
4 U.S. EPA, "National Air Pollutant Emission Estimates, 1940 -1990", EPA-450/4-91-026, November 1991, p. 42.
5 U.S. EPA, "National Air Pollutant Emission Estimates, 1940 -1990", EPA-450/4-91-026, November 1991, p. 42.
6 U.S. EPA, "National Air Pollutant Emission Estimates, 1940 - 1990", EPA-450/4-91-026, November 1991, Table 7, p. 19.
7 U.S. EPA, "National Air Pollutant Emission Estimates, 1940 - 1990", EPA-450/4-91-026, November 1991, Table 3, p. 15.
B-3
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
achieved by 1970, they are no longer available to off-
set the increases observed from other source catego-
ries (e.g., highway vehicles). The no-control scenario
therefore shows a steady increase in overall emissions
of primary particulates after 1975.
The following sections of this appendix summa-
rize the methodologies used to model control and no-
control scenario emissions for each of the six major
emission sectors. Additional details can be found in
the supporting documents listed in the References sec-
tion of this appendix.
Industrial Boilers and Processes
For the purposes of the retrospective analysis, the
industrial sector was divided into two components:
(1) boilers; and (2) industrial processes and process
heaters. The factors affecting emissions from these
two source types are different, and, as a result, sepa-
rate methods were used to calculate control and
no-control scenario emissions in each of the target
years. To analyze the change in emissions from in-
dustrial boilers, ANL used the ICE model (Hogan,
1988). This model was developed under the auspices
of NAPAP to forecast State-level fuel choice and
emissions from conventional, steam raising, industrial
boilers. For the retrospective analysis of industrial
processes and fuel use emissions from process heat-
ers, ELI used the EPA Trends methods and the ANL
MSCET data base (EPA, 1991; Kohout et al., 1990).
The Trends report contains estimates of national emis-
sions for a variety of industrial sources for the time
period of interest. The MSCET data base provided
the spatial distribution used to calculate State-level
emissions.
The distinction between industrial boilers and non-
boiler industrial processes was necessitated by the
structure of the CAA regulations and by the factors
affecting emission levels from these two source types.
Boilers are regulated differently from processes and
process heaters. Emissions from industrial processes
are primarily a function of levels of industrial activ-
ity. The emissions from fuel combustion, however,
are a function of energy use and fuel choice as well as
industrial activity. Fossil fuel emissions in the absence
of the CAA are not proportional to industrial output,
since the level of energy use is a decision variable for
the firm in its production process. Therefore, in the
ICE model simulations used to estimate no-control
scenario boiler emissions, the level (and type) of en-
ergy use were determined first, and then the effects of
emission regulation were taken into account.
Overview of Approach
Industrial Boilers
ICE model inputs include fuel prices, total boiler
fossil fuel demand by industry type, and environmen-
tal control costs. The outputs of the ICE model were
SO2, NOx, and TSP emissions by State, industry, and
boiler size class. The model runs in 5-year increments
and has a current base year of 1985.
The model required boiler demand input data at
the State level. Seven industry types were included in
the ICE model: Standard Industrial Classification (SIC
) codes 20,22,26,28,29, 33, and "other manufactur-
ing." ANL's approach assumed that industrial boiler
fuel use occurs only in the manufacturing sector. The
model also required fuel price data in each of the tar-
get years at the Federal Region level. Prices by grade
of coal and petroleum product, such as sulfur content
and heating value, were used by the model to deter-
mine the cost of compliance, and to determine emis-
sions when the regulations are not binding.
Control costs were computed by engineering sub-
routines in the model. These costs were used by the
ICE model's fuel choice component to determine the
effect of CAA-related costs on the market share of a
particular fuel. This fuel choice decision only applies
to new industrial boilers, since the cost of existing
emission controls are not in the ICE data base and
fuel choice is not re-evaluated for existing boilers.
Industrial Processes and In-Process Fuel
Combustion
The calculation of historical emissions from in-
dustrial processes uses EPA Trends methods to esti-
mate national emissions for the analysis years, then
allocates these emissions to States using the State
shares from the MSCET data base.
MSCET uses a variety of methods to estimate his-
torical emissions for the various industrial sectors. For
industrial process emissions, MSCET is based on his-
torical data on industrial activity to allocate emissions
based on the State level distribution of the polluting
activities. The State level distribution and benchmark
B-4
-------
Appendix B: Emissions Modeling
is based on the 1985 NAPAP Inventory (EPA, 1989).
This approach implies that the MSCET data corre-
sponds directly to the 1985 NAPAP Inventory, and
that, for any State, the sum of the emissions from
Source Classification Codes (SCCs ) that comprise
the MSCET industry sector are equal to the MSCET
data for that State and sector. Data from Trends are
used by MSCET to provide information on changes
in the aggregate level of control for years other than
the 1985 benchmark. Since no direct correspondence
existed between the Trends data and MSCET, a rela-
tionship was developed to link MSCET sectors to
Trends industry categories and to industry categories
in the J/W model, which was used to change activity
levels for the no-control scenario.
Table B-l shows the relationship between the sec-
tor definition used by MSCET, Trends, and the J/W
model. The mapping from MSCET to J/W and Trends
is used to provide the changes in aggregate activity
and emission control for the calculation of no-control
scenario emissions.
Establishment of Control Scenario Emissions
Energy use and corresponding emissions were
broken down between boilers and non-boiler indus-
trial processes. The latter category includes furnaces,
kilns, internal combustion engines (e.g., compressors),
and other non-steam types of process heat. The focus
of this analysis is on boiler emissions, which were
subject to increasingly stringent regulations over the
1970 to 1990 period. (Emissions from some types of
industrial processes were also regulated, but regula-
tion of non-boiler sources was targeted on the emis-
sions from the industrial process itself, not on its fuel
combustion) For this study, ANL assumed that only
boiler fuel use is affected by emission regulations. The
non-steam boiler portion of industrial fuel use is not
directly affected by the CAA. This portion of the
emissions may be affected indirectly by changes in
industry activity level and fuel consumption. The
emissions from non-boiler industrial processes were
calculated separately by ELI.
Control Scenario Boiler Emissions
Control scenario boiler SO2, NOx, and TSP emis-
sions were calculated by the ICE model. The MSCET
data base provided an estimate of historical emissions
for total fossil fuel combustion by industry. Since
MSCET does not identify the two required compo-
nents of boiler and non-boiler emissions, ANL de-
fined the residual of the ICE model control scenario
and MSCET as the non-boiler or in-process fuel use
emissions. For the relevant study period, MSCET pro-
vided a control scenario estimate of total boiler and
non-boiler emissions, which was used to calculate the
control scenario State-level boiler emissions based on
a special run of the ICE model.8
In order to use ICE to model the historical emis-
sions path, it was necessary to construct a new ICE
model base year file and new user input file so that
the model could begin its calculations from 1975 con-
ditions. Construction of the base year file was com-
pleted in two stages, using two different data sources,
as discussed below. The user input file has several
elements, including energy prices and historical boiler
fuel use; its construction is discussed in the next sec-
tion. The model base year file provided the energy
use in boilers and corresponding emission control
regulations (State Implementation Plans -SIPs- for
example) by several categories. These categories in-
clude:
• State;
• Industry group (one of seven);
• Fuel type (natural gas, distillate or residual
fuel oil, and coal);
• Boiler size class (MMBTU/hr, one of eight
categories);
• Utilization rate (one of five categories); and
• Air quality control region (AQCR ).
For the purposes of ANL's analysis, only the first
three categories were assumed to vary. In other words,
for each State, industry, and fuel type combination,
the distribution of boiler size, utilization rate, and
AQCR was assumed to be constant. Over time, how-
ever, changes in the aggregate composition of State,
industry, and fuel type would cause corresponding
changes in the aggregate composition of the other three
characteristics. As mentioned previously, the current
base year file was 1985. The retrospective analysis
required a 1975 base year. Because of data limita-
tions, the approach to construct a new base year was
achieved hi the following two steps: the construction
of a 1980 interim base year file from the 1985 file,
and then the construction of the 1975 file from the
interim 1980 file.
8 MSCET does not provide State-level estimates of TSP, while ICE does. To estimate total regional TSP from fuel combustion,
the Trends model was employed. These national emissions estimates were allocated to the States based on the State-level shares of
TSP from the NAPAP inventory.
—
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
"8
u
§
-a
S3
CO
1
ts
I
•8
I
CO
TJ
a>
{5
1
o
C3
u
«
c
o
W
(U
pq
1
1
PQ
, ass-a -2- --- '-
W O t
1 -1: »" '
' " "»
' '" ' ' ;'
'O" SS' 'rt
SgiS S'^S'-S^S S'—ft^./»§"•• & «-«•*?• "8" fs^-S"
^sl i-| g § | S S »' "- -"i I ,-^J-1 •* - -fri wsra ,£
l^^"li^iI^iS¥^v'>l§rsifg1g&«
^•s:*?!
1
I
i
TJ
g 3 i
- -q.
1321
N"fe«'
k'O °- Oc,,<*-' SI, \%tet
v-^-"". .: x;,x^r ;:ff
g
I
I I
P5
S
Q
i. . s
'-<>'> g"- -
K>^
S f"1 i.
..«i .J..»^-
W3
i '^:II-^
'«
:::|>2ZiS| II';f?§'.-. s.
-»i lr<rrt"'"
JSj,
I *
I | rllptllJsn*
02 osfi ^-§-yS^/;'-§
g' '("CA »«
' *8 *' '^63 ^
"! -II
i-,, OH.^
B-6
-------
Appendix B: Emissions Modeling
Estimates of boiler fossil fuel consumption in
1980 for each State and major fuel type were pro-
vided by Hogan (Hogan, 1988). These estimates are
based on the assumption that the industry mix, size,
utilization, and AQCR distribution within a State are
constant. Through assuming this relationship, the 1985
ICE base year was scaled to match the data for 1980,
thus forming the 1980 interim base year data.
To construct the 1975 base year file, the assump-
tion of a constant industry mix for a State and fuel
type was no longer necessary, since detailed data on
each industry for 1980 and 1975 were available from
PURchased Heat And Power (PURHAPS ) model data
files (Werbos, 1983). These PURHAPS data files were
derived from the Annual Survey of Manufactures:
Fuels and Electric Energy Purchased for Heat and
Power (DOC, 1991). The available data in these files
were for total fuel use not boiler fuel use. To make
use of these data, it was necessary to assume that the
fraction of fuel used in boilers, for any given State
and industry, remained constant from 1975 to 1980.
To the extent that the fraction of boilers' heat versus
process heat applications is a function of the specific
industrial production process, this assumption is rea-
sonable.
Based on the assumption of constant boiler fuel
fraction of total fuel use, the ratio of 1975 to 1980
energy use for each State, industry, and fuel type was
applied to the corresponding record of the 1980 in-
terim base year file to produce 1975 base year files.
Control Scenario Industrial Process Emissions
To estimate boiler emissions of sulfur oxides
(SO ), NOx, and VOC from industrial processes, data
from Trends were used. The percentage change in
national emissions by Trends category was applied to
the appropriate sector from MSCET to obtain State-
level emissions. In some cases there are several cat-
egories in Trends that match directly with MSCET
categories (see Table B-l). In these cases, the Trends
sectors were aggregated and the percentage change
was computed. It was assumed that the level of con-
trol in each industry sector implied by Trends was
uniform across States. The changes in emissions in
each State are not equal to those at the national level,
since the industry composition in each State varies.
Development of Economic Driver
Data lor the Control Scenario -
Industrial Boilers and Processes
The results of the J/W model were the primary
source of activity in the ICE model driver data. These
results were also used by ELI to produce the national
results for industrial processes fromTrends. Both ICE
and Trends use the forecasted change in industrial
activity that results under the no-control scenario.
These data were in the form of industry specific
changes in energy consumption and industrial output,
for boilers and industrial processes.
Economic Driver Data for Industrial
Boiler Approach
Using the 1975 base year file as a starting point,
the ICE model estimated fuel choice and emissions
based on a user input file containing total boiler en-
ergy demand and regional energy prices. The 1975,
interim 1980, and original 1985 base year files con-
tained the required information on energy demand for
each industry group and State, so the data in these
three files were aggregated across fuel type, and other
boiler characteristics (for example, size). These ag-
gregated data provided the energy demand for three
of the target years. Since 1990 State-level data on
energy use by industry group were not available at
the time of the study, the NAPAP base case forecast
for the ICE model for 1990 was used to provide the
demand data for this year.
The user input file for ICE also requires a price
input for each target year. These prices were input by
Federal Region for distillate oil, 4 grades of residual
oil (by sulfur content), natural gas, and 11 grades of
coal (by sulfur content and coal rank, i.e., bituminous
and sub-bituminous). Prices for 1985 and 1990 were
obtained from the NAPAP base case user input file.
The prices for 1975 and 1980 are from U.S. Depart-
ment of Energy (DOE) data on State-level industrial
energy prices (DOE, 1990). Regional prices of natu-
ral gas, distillate oil, steam coal, and residual oil were
constructed by aggregating expenditures across States
within each region and dividing by total British ther-
mal unit (BTU) consumption for the years 1975,1980,
and 1985. Since prices by sulfur content grade are not
reported by this DOE source, ANL assumed that the
sulfur premium implied by the 1985 ICE model input
file was proportional to the average price. Based on
this assumption, the ratio of the regional coal and re-
B-7
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
sidual oil price in 1975 and 1980 to the 1985 price
was applied to the 1985 price in the ICE model base
case file for each grade of fuel. To provide additional
consistency between the NAPAP analysis and ANL's
study, the distillate oil and natural gas prices were
benchmarked to the 1985 ICE model prices as well.
One possible inconsistency arises using this pro-
cedure. The residual oil and natural gas markets are
closely linked, particularly for industrial customers.
These markets, specifically the gas market, underwent
tremendous changes over the study period. To model
the effect of these structural changes on the sulfur pre-
miums in residual oil would require a detailed oil and
gas supply model that was beyond the scope of this
project. Moreover, the CAA regulations themselves
create the potential for sulfur premiums. This poten-
tial effect of the CAA was not captured, though, be-
cause of the assumption of proportional fuel sulfur
premiums on residual fuel oil. The relationship be-
tween market driven sulfur premiums in the coal mar-
ket and the CAA was given additional consideration
in this analysis through the use of an explicit coal sup-
ply model.
The J/W data for industrial energy consumptions
was supplied in the form of percentage change in cost
shares. In order to compute the percentage change in
the quantity of energy used, ANL used the following
identity:
= In (P) + In (£) - In (P x £)
In
The percentage change in E is the percentage
change in cost share, minus the change in price, plus
the change in value of shipments. These calculations
were performed for each energy type and industry
sector in the J/W model. The ICE model requires to-
tal fuel use, so the fuel specific percentages were
weighted by historical fuel consumption to produce
an aggregate change in fuel consumption to apply to
the ICE model input data files.9
ICE also uses energy prices to simulate boiler fuel
choices. The control scenario forecasts of energy
prices in ICE were adjusted based on the percentage
changes in energy prices, by coal, oil and natural gas.
This implicitly assumes that the oil and coal fuel sul-
fur premiums, by region, are proportional to the aver-
age national price. To test this assumption for the coal
market, additional modeling of the coal prices was
performed using the coal market component of the
ARGUS model.
It is possible that in some regions low sulfur coal
prices to the industrial sector may be lower than the
national average. This was not found to be the case.
For example, in 1990, delivered regional industrial
coal prices change by less than two-thirds of one per-
cent. In most cases, the percentage change was near
zero. This result appears to occur because of the highly
regional nature of the coal market. While the artifi-
cial demand for low sulfur coal may fall, power plants
near low sulfur coal reserves now find it advantageous
to buy this local coal, which raises the price back to
an equilibrium level near to that of the control sce-
nario. This is even more likely to be true of industrial
delivered prices, since industrial prices are more af-
fected by transportation costs than are the utility prices.
No additional ICE modeling was performed.
Economic Driver Data for the Industrial
Process Approach
The J/W model was also used to account for ac-
tivity level changes in the calculation of industrial
process emissions under the no-control scenario. The
correspondence between Trends, MSCET, and the J/
W model was used to apply changes in industrial ac-
tivity in each target year to each industrial process.
No-control Scenario Emissions
Industrial Boiler Emissions of SO2, NOx, and TSP
The CAA imposed different regulations, SIPs, and
New Source Performance Standards (NSPS) that ap-
ply to industrial boilers of varying size. The primary
effect of CAA regulations on industrial boilers was
simulated by defining the Air Quality Control Region
(AQCR), the resulting SIPs, and subsequent NSPS for
boilers. The industrial boiler SIP regulations were in-
cluded in the ICE base year file discussed in the pre-
vious section. Since the ICE model estimates new
boiler emissions for each target year, the boiler NSPS
are input through the ICE user files. Industrial NSPS
were implemented in two phases. The 1971 regula-
tions are imposed for the study years 1975 and 1980.
9 ICE uses six of the manufacturing industries from the J/W model directly. The remaining industries' percentage changes were
weighted to produce the "other" category.
-------
Appendix B: Emissions Modeling
The 1984 NSPS revisions are imposed in the study
years 1985 and 1990. For the no-control scenario, ANL
set the SIPs and NSPS to a flag that indicated "no
regulation."
Industrial Boiler Emissions of CO and VOC
Two of the criteria pollutants emitted by indus-
trial fuel combustors, CO and VOC, were not included
as outputs of the ICE model. Therefore, CO and VOC
emissions were analyzed separately using Trends
methods. Control scenario CO and VOC emissions
were taken directly from Trends.
To estimate CO and VOC emissions from indus-
trial combustion for the no-control scenario, fuel use
for industrial manufacturing was adjusted, reflecting
fuel consumption changes estimated by the J/W model.
These changes in the level of fuel consumption by
industrial combustion were also used in ANL's ICE
boiler model. Changes in industrial combustion fuel
use by manufacturing between the control and
no-control scenarios are reported in Table B-2. These
estimates represent an average of several sectors,
which were developed by ANL as part of the model-
ing process for ICE.
No-control scenario emissions were computed
using 1970 emission factors. Since there were no add-
'Table B-2. ?Fuel Use .Changes Between ,. *;
f <'-^','/ ,;' — ''; ' •> •• •> ,;•> 7>_>VgX't VA '
'"Control andfNo-control Scenarios., ]-n ': •'""'-
' YfcaiSv,*
1975 '
V S :;§f ° ' v
•>v> ^ '
-1980 \
"'- ' ,^
198s ;:'•;
i- **^
•*,-'•!
'•^^::
"Vv- „ -s.
•* ,'?,"•>
FuelType
- Coal ";f '^7,
v&SfsJ
' oii^j.^v/"^-
•>
,Gas _ ^"'
W-SV-'V' s s s s •. s s •- v
,£gal ;,--**
~Q{I if:
*Gas „-•» -
%'> •, -y"
*C0-aL,,w,-<
foil- ;"•;
- ,?/- ?•••
-'Gas vf^f , -
' Coal ;;„;„ -
-C6il -,„.,,.,-
jGas '"";>T-i"
Fuel Use Cftaages -
-H"^-^tOs04^- ,-,%„;. ,
/x""" -fMil;>^: ""
;/T^'
-.006*,-; -
V ^f&to " * '~<\\ ' '
':' ,,;,^.oo6i
;- .'-f.0107 '-
"• XV
*^^Q&5:\^
^V±rf-.oo6i ,J,^
•• »»*;0089 ,-::-;>
--^097 - .,,^»
... :X<::0079;>-|^
V S*NX*' ""''
'" ' 4-.Q091-.v-v,.
^-.^-£^^K
on controls for industrial combustion VOC and CO
emissions, it was not necessary to adjust the no-con-
trol scenario for changes in control efficiency.
Emission estimates were regionalized using State-
level emissions data from industrial boilers recorded
hi MSCET. For the control scenario estimates, VOCs
were regionalized using the MSCET State-level shares
for industrial fuel combustion. In the no-control sce-
nario, the State-level shares were held constant. The
control scenario emissions of CO were regionalized
using the control scenario NOx emissions from the ICE
model. This approach assumes that CO emissions are
consistent with NO emissions. The no-control sce-
X
nario CO emission estimates from industrial combus-
tion sources were regionalized using no-control NO
emission estimates from industrial combustion
sources.
Industrial Process Emissions
A wide range of controls were imposed on indus-
trial processes. These emission limits are embodied
in the assumptions of control efficiencies in the Trends
model. Data on national no-control scenario emissions
from industrial processes were provided by EPA.
These data were combined with MSCET to produce
regional-level results.
Lead Emissions
Estimates of lead emissions from industrial boil-
ers and industrial processes were completed by Abt
Associates. The methods used for calculating lead
emissions from industrial processes and industrial
boilers were similar. The starting point was the TRI,
which provides air toxics emissions data for manu-
facturing facilities with more than 10 employees. To
estimate lead emissions from industrial boilers and
processes, 1990 facility-level lead emissions data were
extracted from the TRI. These data were then adjusted
to create estimates of lead emissions from industrial
sources under the control and no-control scenarios for
each of the target years. For the control scenario, lead
emissions for 1975,1980, and 1985 were obtained by
extracting an emission factor and a control efficiency
for each lead-emitting industrial process in the Trends
data base. These emission factors and control efficien-
cies were multiplied by the economic activity data
for each year for each process as reported in Trends
to yield estimated control scenario emissions by in-
dustrial process. Each industrial process was assigned
B-9
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
a code to correspond with energy consumption data
by industrial process compiled in the National Energy
Accounts (NEA) by the Bureau of Economic Analy-
sis, and emissions were summed over all processes to
obtain a total for each target year.
For consistency with the other emission estimates
in this analysis, industrial process no-control scenario
lead emissions were adjusted for changes in indus-
trial output, and for changes in emissions per unit of
output due to control technology applications. Changes
in industrial output were accounted for using results
from the J/W model. Lead-emitting industrial pro-
cesses in the Trends data base were assigned to a J/W
sector. For each sector, the percentage change in eco-
nomic output was used to adjust the economic activ-
ity data for that process from the Trends data base.
These adjusted economic output figures were used
with the 1970 emission factors and control efficien-
cies to derive the estimated no-control scenario lead
emissions for each industrial process in each target
year. The process-level emissions were then aggre-
gated to the NEA-code level as in the control sce-
nario.
The lead emission estimates from industrial pro-
cesses, by NEA code, were used to derive percentage
changes in emissions under the control and no-control
scenarios by NEA code for application to the TRI
emissions data. Since TRI data are reported by SIC
code, NEA codes were "mapped" to the appropriate
SIC codes, and then the percentage change for each
NEA code was used to represent the percentage change
for all SIC codes covered by that NEA code.
To calculate lead emissions from industrial boil-
ers, Abt Associates developed estimates of lead emis-
sions from industrial combustion under the CAA for
each of the target years. The Trends data base con-
tains national aggregate industrial fuel consumption
data by fuel type. For each fuel type, the fuel con-
sumption estimate was disaggregated by the share of
that fuel used by each NEA industrial category. The
Trends data base also contains emission factors for
industrial fuel use, by fuel type, as well as control
efficiencies. The lead emissions from industrial com-
bustion for each NEA category were derived by mul-
tiplying the fuel-specific combustion estimate for each
NEA category by the emission factor and control ef-
ficiency for that fuel type. The result was emissions
of lead by NEA code and by fuel type. Emissions from
all fuel types were then summed by NEA code. The
NEA data were used to disaggregate the industrial fuel
consumption figures, based on the assumption that the
ICE are the same among all industries covered by a
given NEA code.
To estimate no-control scenario lead emissions,
the macroeconomic effect of the CAA and the change
in emissions per unit of output that resulted from spe-
cific pollution control mandates of the CAA were both
taken into account. As hi the control scenario, the na-
tional aggregate industrial fuel consumption estimate
by fuel type was disaggregated by the share of that
fuel used by each NEA industrial category. The fuel
use was then adjusted in two ways: some NEA codes
were specifically modeled by the ICE model, and for
the remaining NEA codes, J/W percentage changes
in fuel use were applied. These fuel use estimates were
then combined with the 1970 emission factors and
control efficiencies for industrial combustion by fuel
type from the Trends data base to obtain no-control
scenario combustion-related lead emissions from in-
dustrial boilers by NEA code. These estimates of to-
tal lead emissions by NEA codes were matched to
SIC codes, and then to the data in the TRI data base.
This approach assumed that an average emission value
was assigned to all reporting TRI facilities in a given
SIC code.
Off-Highway Vehicles
The off-highway vehicle sector includes all trans-
portation sources that are not counted as highway ve-
hicles. Therefore, this sector includes marine vessels,
railroads, aircraft, and off-road internal combustion
engines and vehicles. As a whole, off-highway ve-
hicle emissions are a relatively small fraction of total
national anthropogenic emissions.
Overview of Approach
The process used by ELI to determine the national
level of emissions from the off- highway transporta-
tion sector is similar to the procedure outlined above
for industrial processes. To estimate the emissions of
criteria air pollutants from these sources under the
no-control scenario, the historical activity levels were
held constant, rather than attempting to calculate a
new no-control scenario level of off-highway vehicle
activity. This assumption was necessary since the off-
highway activity indicators (amount of fuel consumed,
and landing and take-off cycles for aircraft) do not
B-10
-------
Appendix. B: Emissions Modeling
have direct correspondence with a given J/W category.
The national no-control scenario emissions of criteria
air pollutants from these sources were simply derived
by recalculating emissions using 1970 emission fac-
tors.
Development of Control Scenario
To estimate control scenario emissions, the analy-
sis relied on Trends methods, using historical activity
indicators, emission factors, and control efficiencies.
Essentially, the estimates of off-highway emissions
under the control scenario represent the historical es-
timates from the Trends data base.
No-control Scenario Emissions Estimates
The calculation of off-highway emissions for the
no-control scenario required the Trends data to be
adjusted to reflect changes in controls and economic
activity in each of the target years. Linking source
activity changes with economic activity for this sec-
tion is not straightforward. The economic activity data
for off-highway engines and vehicles are expressed
either in terms of amount of fuel consumed, or in terms
of landing and take-off cycles for aircraft. Neither of
these off-highway activity indicators has a direct cor-
respondence with a given J/W sector, making the sort
of direct linkage between Trends categories and J/W
sectoral outputs that was used for industrial processes
inappropriate.
In the absence of a link between the economic
factors that are determinants of emissions from this
sector and the available economic activity forecasts,
the no-control scenario emissions of criteria air pol-
lutants from off-highway mobile sources were esti-
mated based on the same historical activity levels used
for the control scenario. Although there were changes
in sectoral output and personal income that might have
had an effect on off-highway vehicle usage, these
changes were deemed to be small and not likely to
have a major effect on the emissions from this sector.
Emission factors for each of the off-highway
sources were also held constant at 1970 levels to cal-
culate no-control scenario emissions for each target
year. The national emissions of criteria air pollutants
from these sources were then recalculated using 1970
emission factors.
National and State-Level Off-Highway
Emission Estimates
Table B-3 summarizes national-level emission
estimates for off-highway sources. The emission es-
timates derived from using the methodology discussed
above yielded results that seem counter-intuitive. The
emissions from off-highway sources, in particular the
emissions from aircraft, are lower in the no-control
scenario than those projected for the control scenario
for most pollutants. This is a result of calculating
emissions using 1970 emission factors, since the 1970
emission factors for aircraft are lower than the air-
craft emission factors in later years.
ELI identified several potential sources of uncer-
tainty in the emission estimates for this sector. First,
the assumption that the total level of off-highway ve-
hicle fuel consumption is constant between the two
scenarios may be flawed. Second, the use of 1970
emission factors in the no-control scenario may fail
to capture significant changes in technology. These
technological changes are implicitly captured in the
control scenario and it is possible that these techno-
logical changes may also have occurred under a
no-control scenario.
One possible response to the biases created by the
use of 1970 emission factors for all years in the
no-control scenario is to test how results might differ
if the emission factors used for the control scenario,
which would include technological change, were also
used for the no-control scenario. However, using this
treatment of emission factors, the emissions projec-
tions from the adopted methodology from non-high-
way sources in the no-control scenario would be iden-
tical to the emissions projections under the control
scenario. The reason for this is that the economic ac-
tivity levels were not adjusted for the calculation of
emissions under the no-control scenario.
In order to disaggregate the national data to a State
level, the methodology used the MSCET data base,
which is described earlier. Emissions of VOC, SOx,
and NOx were regionalized using the State-level shares
from the MSCET methodology. The emissions of TSP
were regionalized by using the State-level shares for
SOx reported by MSCET, and the emissions of CO
were regionalized using the State-level shares for NOx,
also reported by MSCET. The potential bias that this
introduces is likely to be small, due to the relative
homogeneity of off-highway vehicle emission sources.
B-ll
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table B-3, Difference in Control-atid-Kc^cop^ol Scenario Off-KgKway-'Moljile-Souiice
, •' . _,,,<-,-.- «'<*•"_ fl _ '*" , ;,•>•*/•»•- ,}^> ,""• —-''-
Emissions. • '- ' ,..,,. >'*;'J'-- .' /->.« -'- " -"-- , '"''.,,.*»,
TSP
NO.
SOj
CO
VOCs
-8,,^;A
S=-lo* •"*
CpfltrpJ Scenario;
_ ^ NV.V'
No-Conircd SeeaaritiT*
N ^t
Percentage Increase:
Control Scenarios '", , „.*»,•,
No-Control Scenario: , '-.*.-,-
" "' . ,„,- *-
Percentage Iacrea&:-
Control Scenario:
No-Control Scenario: ,, •#,*?
Percentage Increases:" ' ,
Control Scenario: , •; -:;^
,^t. jJV* '
No-Control SceltafJo:
Percentage InaeisSf' ^ '^^,e
, ; "*;"•*
Control Scenario:"'
No-Control Seertarto:
Percentage- Increase!:,, ',>i •"•"'•
' '"' ISTSiv,'^1
2<>8.<
, - .?*/»'» ;•
., . -^me ;>•*.>
v~ 1;'974,$ , ,
%."-,! ' -
^^-^t^Sst?-"
' -3646 ;,
^r-^sd^s ~ ,^
,;.^ O®-"'^*
'- 8,512,8
_1A ^/^ ij~
^* 83*10
, ,- ,•> ,'*-A; •
-v-- -0» „„,
^- ? ' *
-----1374;$' *""'
,-; ,-,*^'t->'"
, 1,385,9 „.„-,
^T-'l*
19«0- ''
;,>- • -,./-.
.- - -251.1 ><">
- 268f8
, ¥i^i:'-
-- - zmtf*""""'"
""''2, 150.5, „
„
'-'•'• -1%'
/< A /•$*"' " '
,.-..53Ivl-' '
"'oS5*% "'•*'
528.6
;'' ,0%-..
,,s VV^^?,^^ /
, ^ A,^^" y^*-1/* "^ '*'
'8,101.4- r
A0T1.2 '
,,v; «-><«'V"'''"
S''l,S7ft«
^V^S", '"' J
,^,»4,41$4f :
,' - « / 'J ' '""••''
!3%
- . 1985"'' '
" ' 268,7' "•'"
v ''' ~261.2
",'' ',<£%.. •
^^»
40S4**'"'
-. --/ -403.0 - -
' ' 's*,' ' '"•• "* S S' ' '
,,,, 4'&~"
"!----'-7;88l.? '"'
',"7?880.2
'*** *'*'
, 0»
"" " 1,33,4.8-,
1,388,6- - '"
,<^>'4*- ' ";
- ,,<• ,,,'^' '
1990,,,,. -
" "'-280^'-" '
,, ,„/,' "'f-
, 266.9
' 4%
V«,x2,085.9
,'s 2,058.'9'": '
,«.,-, - - 4%
- .;„*<,-$.: <"'-.'>"'
392,5, - -
""*' "386.9, ,, ,
*1»V
"""' '8,tX?9,0 ;-?#•
•"'-"/s^zV ''''
0%
4,405.0" ''•**
-«,v- ---t"--!
7 , ,i,485,,8,,.
^^tiF1®-'
Note; _ ^
scenarios divided by the ControlScenario projeetionT
, , - ,,,, \f,,/##/>/><>**'**'/f<
; increa'se'is thedifferen'tial'between
, ;»^; - - ;-,
'XV , ( '^ '^^i-/'-', ^ ' , ,
As with regionalization of industrial process emis-
sions, the State-level shares are held constant between
the two scenarios. To the extent that the distribution
of economic activity between States was not constant
over the period of the analysis, holding State-level
emission shares constant may bias the results, although
the direction and magnitude of the potential bias is
unknown.
On-Highway
This section addresses the highway vehicle por-
tion of the transportation sector. Highway vehicle
emissions depend on fuel type, vehicle type, technol-
ogy, and extent of travel. Emissions from these ve-
hicles have been regulated through Federal emission
standards and enforced through in-use compliance
programs, such as State-run emission inspection pro-
grams. Vehicle activity levels are related to changes
in economic conditions, fuel prices, cost of regula-
tions, and population characteristics. Emissions are a
function of vehicle activity levels and emission rates
per unit activity.
TEEMS was employed by ANL to analyze the
transportation sector. The modeling system links sev-
eral models, disaggregate and aggregate, to produce
State-level estimates of criteria pollutants. The sys-
tem is subdivided into two modules: an activity/en-
ergy module and an emissions module. Each module
contains multiple models. TEEMS has been docu-
mented in several reports and papers (Mintz and Vyas,
1991; Vyas and Saricks, 1986; Saricks, 1985). It has
been used for several policy analyses and assessment
studies for DOE and NAPAP. This section presents
an overview of the approach used to conduct the analy-
sis of the transportation sector. Also included in this
section is a summary of the methodology used by Abt
Associates to estimate changes in lead emissions from
highway vehicles in each target year.
B-12
-------
Appendix B: Emissions Modeling
Overview of Approach
TEEMS has two modules: an activity/energy
module and an emissions module. The activity/energy
module calculates emissions based on: (1) personal
travel; (2) goods movement; and (3) other transporta-
tion activity inputs.
Personal Travel
Personal travel activity and resulting fuel con-
sumption were calculated for each target year using
procedures that disaggregate households by demo-
graphic and economic attributes. Economic driver
data, developed from U.S. Government data and mac-
roeconomic model(s) of the domestic "economy,
formed the basis for household disaggregation. Mod-
eling procedures were employed by ANL to project
movement of households between various attribute
classes, and vehicle holdings were projected in terms
of the number and type of vehicles held by each house-
hold type. National totals were then developed by
aggregating the vehicle holding estimates for each
household type, accounting for the number of house-
holds of that type. Travel estimates, in terms of VMT,
were calculated using the same approach, and based
on the VMT of each household type. The basis for
household transportation activity projection has been
empirically established through analysis of the 1983-
84 Nationwide Personal Transportation Survey
(NPTS) (FHWA, 1986; Mintz and Vyas, 1991). VMT
are projected using this empirical relationship, and es-
timates of the elasticity of VMT to vehicle operating
cost are then made. Energy consumption was esti-
mated in each target year using VMT, shares of VMT
by vehicle type, and exogenously developed vehicle
characteristics.
The following three models and an accounting
procedure were employed to develop target year per-
sonal travel activity projections:
1. The first model projected the target year dis-
tribution of households by then: attributes.
This model employed an iterative proportional
fitting (IFF) technique and projected the num-
ber of households in each cell of the house-
hold matrix - each of which is defined by vari-
ous categories within six household attributes.
2. The second model projected changes in ve-
hicle ownership resulting from changes in
income and cost of vehicle operation. The
model applied estimated ownership changes
to each target year household matrix such that
the control values within each of the house-
hold attributes, excepting vehicle ownership,
remained unchanged.
3. The third model estimated the composition
of household vehicle fleet by type (cars and
trucks), size, technology, and fuel.
4. An accounting procedure applied VMT per
vehicle to vehicle ownership in each combi-
nation of household attributes. VMT and en-
ergy consumption were accumulated by ve-
hicle type, size, and fuel.
Each of these models is described separately in
the following subsections.
Iterative Proportional Fitting (BPF)
This IFF model modified a control scenario ma-
trix of household counts. A household matrix was
developed from the 1983 NPTS data and upgraded to
the year 1985 using published aggregate data. The
procedure used in constructing the 1985 household
matrix has been documented elsewhere (Appendix B
of Mintz and Vyas, 1991). The matrix is defined by
six attributes: (1) residential location (central city,
suburb, rural); (2) household income; (3) age of house-
holder; (4) household size; (5) number of drivers; and
(6) number of vehicles. The household matrix has
3,072 cells, some of which are illogical (such as 1
person, 2 drivers). Illogical cells were replaced with
zeros.
Household shares within each attribute in each
target year were developed exogenously using data
from the Bureau of the Census and selected macro-
economic model runs. The projected total of house-
holds and shares of households in each category of an
attribute were supplied to the IFF model. The model
modified the control scenario household matrix to
match the specified shares and total number of house-
holds.
The IPF model treated household distribution
within each attribute as a set of vectors. These vectors
were scaled to match the specified shares and house-
hold total. Following the initial scaling, a gradual scal-
ing technique was used to move in the direction of the
target shares. The scaling process was repeated until
closure was achieved for all attribute classes. Since
B-13
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
vehicle ownership levels were estimated by the ve-
hicle ownership model (described in the next section),
shares within the sixth household attribute (number
of vehicles held) were not specified, leaving it uncon-
trolled. This flexibility of an uncontrolled attribute
helped to facilitate the model operation. The number
of households in each class of vehicle ownership
within the output matrix represents distribution of
households using the control scenario (1985) relation-
ship of vehicle ownership to other household at-
tributes.
Vehicle Ownership Projection (VOP)
The VOP model projected the changes in vehicle
ownership resulting from changes in the number of
licensed drivers, disposable personal income, and an-
nual fuel cost of vehicle operation. The model is based
on historical household ownership rates. A target per-
driver ownership rate was computed using disposable
income and fuel cost. This target rate represented de-
sired ownership if income and fuel cost were the only
determinants. A parameter representing ownership
responsibilities such as acquisition effort, disposal
effort, parking requirements, and other indirect aspects
was applied to adjust this target. The new ownership
rate was used to estimate the number of household
vehicles.
The household matrix created by the IFF model
was revised to match the projected household vehicle
ownership. Household shares within the first five at-
tributes remain constant while those within the sixth
attribute (i.e., number of vehicles) were variable. A
deviation measure was defined and its value for each
class within the first five attributes was minimized. A
set of simultaneous equations was solved using
Lagrangian multipliers.
Projection of Vehicle Fleet Composition
The composition of household vehicles was pro-
jected for each household matrix cell using a vehicle
choice model called the Disaggregate Vehicle Stock
Allocation Model (DVSAM ). Vehicles are defined
by type (auto, light track), size (small, mid-size, full-
size auto; small pickup, small utility/minivan, stan-
dard pickup, large utility/standard van; or any other
size classification), fuel (gasoline, diesel, methanol,
ethanol, or compressed natural gas), and technology
(stratified charge, direct injection, electric, fuel cell,
or Brayton).
The model computed vehicle composition based
on an individual vehicle's utility to households and
household, needs. A menu of vehicles classified by
the previously mentioned vehicle attributes was sup-
plied to the model. The menu specified characteris-
tics of each vehicle available to households. Vehicles
were characterized by price, operating cost, seating
capacity, curb weight, and horsepower. These vari-
ables formed the basis for computing "utility" (analo-
gous to consumer satisfaction). The household ma-
trix provided demographic and economic attributes
which, when combined with vehicle usage in miles,
define household needs. Vehicle usage (VMT) was
computed as a function of income, number of drivers,
and number of vehicles. A logit model was applied to
compute vehicle ownership shares. Several model.en-
hancements facilitated modeling of limited range ve-
hicles, and representation of supply constraints and/
or regulated market penetration.
Activity/Energy Computation
An accounting procedure was applied to compute
personal travel activity in terms of VMT by vehicle
type. Control scenario VMT per vehicle estimates for
each cell in the household matrix were developed from
the 1983 NPTS. These rates were adjusted within the
procedure on the basis of changes in average vehicle
operating cost per mile for each cell. The vehicle com-
position projection model computes ownership shares
and share-weighted change hi vehicle operating cost.
Elasticity values were applied to this change.
ANL assumed that VMT per vehicle remained
nearly unchanged for a household matrix cell over time
(with the exception of the effect of changes in vehicle
operating cost). In other words, variation of VMT
across household types is far greater than within house-
hold types. VMT per household vehicle remained
stable during the period from 1977 to 1984 (Klinger
and Kuzmyak, 1986). Some increases were observed
in recent years, which were attributed to lower fuel
prices and increased household income (DOC, 1991;
FHWA, 1992). (A portion of the increase could be
attributed to the method of computing average VMT
per vehicle.) The assumption that VMT per vehicle
for each cell remained nearly constant and was elas-
tic relative to vehicle operating cost is reasonable. As
households move from one cell of the matrix to an-
other, they "acquire" the VMT per vehicle rate of that
cell. Thus, this approach accounted for changes in
VMT per vehicle due to increased household afflu-
ence, increased rate of driver licensing, changes in
fuel price, and changes in vehicle technology.
B-14
-------
Appendix B: Emissions Modeling
Goods Movement
Energy and activity demand resulting from move-
ment of 24 aggregate categories of commodities is
estimated by this subcomponent of the TEEMS activ-
ity module. Changes in commodity demand/produc-
tion were provided by growth indexes by two-digit
SIC generated by a macro model. A model that
projects shifts in mode shares among truck, rail, ma-
rine, air, and pipeline modes was used, followed by a
procedure to compute ton miles of travel for each
mode, VMT by fuel type for trucks, and energy con-
sumption by operation type for non-highway modes.
The model used 1985 control scenario data, which
were compiled from railroad waybill sample and pub-
lications, waterborne commerce publications, trans-
portation statistics, and other sources. The procedure
used in developing the 1985 control scenario freight
data has been documented in an ANL report
(Appendix A of Mintz and Vyas, 1991).
This goods movement model was not used for this
retrospective analysis because of funding and time
constraints. A procedure to estimate truck VMT by
fuel type was employed in its place. Published his-
torical VMT values (FHWA, 1988; 1992) were used
along with VMT shares by fuel and truck type from
Truck Inventory and Use Surveys (TIUS) (DOC, 1981;
1984; 1990).
Other Transportation Activities
The activity/energy module also has other mod-
els for developing activity and energy use projections
for air, fleet automobiles, and bus modes. Fleet auto-
mobile activity estimates from an earlier study (Mintz
and Vyas, 1991) were used while other modes were
not analyzed.
Lead Emissions
Estimates of lead emissions in the transportation
sector were developed by Abt Associates based on
changes in reductions of lead in gasoline. This esti-
mation required the estimates of lead in gasoline con-
sumed over the period from 1970 to 1990 and the
amount of lead content hi gasoline that would have
been consumed in the absence of the CAA. These
values were calculated using the quantity of both
leaded and unleaded gasoline sold each year and the
lead concentration in leaded gasoline in each target
year. Data on annual gasoline sales were taken from a
report by ANL that presented gasoline sales for each
State in each target year. For the control scenario, data
on the fraction of gasoline sales represented by leaded
gasoline were used. For the no-control scenario, all of
the gasoline sold was assumed to be leaded. Data on
the lead content of gasoline was obtained from ANL
for 1975 through 1990. For 1970 through 1975, the
analysis assumed that the 1974 lead content was used.
Estimation of No-control Scenario
Emissions
TEEMS emissions projections were carried out
by ANL in the following three steps:
1. Development of emission factors;
2. Allocation of highway activity to States; and
3. Development of highway pollutant estimates.
The following subsections describe the procedures
used for computing highway vehicle emissions.
Development of Emission Factors
EPA's MOBILESa Mobile Source Emission Fac-
tor model was used to provide all of the highway ve-
hicle emission factors used to estimate 1975 to 1990
emission rates (EPA, 1994b). Documentation of the
MOBILESa model is found in the User's Guide for
the MOBILES model.10
Although the actual emission factors used by ANL
are not documented in either the original ANL TEEMS
model report or in the Pechan summary report, the
Project Team provided direction that defined the emis-
sion factors to be used. For the control scenario, ANL
was directed to use the official EPA emission factors
prevailing at the time for each target year. For ex-
ample, the official EPA emission factor being used in
1980 for on-high way vehicle NOx was to be used to
estimate 1980 control scenario on-highway vehicle
NOx emissions. For the no-control scenario, the offi-
cial EPA emission factors used to estimate emissions
in 1970 were to be used throughout the 1970 to 1990
period.
It is important to note that using the 1970 on-high-
way vehicle emission factors to estimate no-control
scenario emissions for the entire 1970 to 1990 period
may bias scenario emission differentials upward. This
is because it is possible that technological changes to
on-highway vehicles unrelated to CAA compliance
10 EPA/OAR/OMS, "User's Guide to MOBILES," EPA-AA-AQAB-94-01, May 1994; see also 58 FR 29409, May 20, 1993.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
strategies may have yielded incidental reductions in
emissions. However, EPA Office of Mobile Sources
(EPA/OMS) experts indicate that the two major tech-
nological changes in vehicles occurring during the
period of the analysis -electronic ignition and elec-
tronic fuel injection- would have yielded negligible
emission reductions in the absence of catalytic con-
verters."
Another potential bias is introduced by assuming
the CAA had no substantial effect on vehicle turn-
over. However, two factors render this potential bias
negligible. First and foremost, under the no-control
scenario retired vehicles would be replaced by new
but equally uncontrolled vehicles. Second, no-control
scenario vehicle use is greater in terms of VMT per
year. This means no-control scenario vehicles would
reach the end of their service lives earlier, offsetting
to some extent the alleged incentive to retire vehicles
later due to costs imposed by CAA control require-
ments.
Allocation of Highway Activity to States
TEEMS' activity module generated national ac-
tivity and energy estimates. These activity totals were
allocated to States through a regionalization algorithm
that used time series data on historical highway activ-
ity shares by State. A trend extrapolation methodol-
ogy was used that stabilizes shifts after 5 years in the
future. For the retrospective analysis, historical high-
way activity shares for each target year were devel-
oped using data published by the Federal Highway
Administration (FHWA) (FHWA, 1988; 1992).
Development of Highway Pollutant Estimates
Highway emission estimates were calculated in
both scenarios for each target year using VMT esti-
mates generated by TEEMS and emission factors from
MOBILE5a. Control scenario activity levels were
adjusted for the no-control scenario using economic
forecasts and historical data.
Control Scenario Emissions Calculation
Control scenario data for the transportation sec-
tor were compiled from several sources. Household
counts and shares of households by six attributes were
obtained from various editions of the Statistical Ab-
stracts of the United States. Household income infor-
mation was obtained from the control scenario run of
the J/W model. Fuel prices were obtained from the
Annual Energy Review (DOE, 1992) while vehicle fuel
economy and aggregate VMT per vehicle were ob-
teinedfromHighway Statistics (FKW A, 1988; 1992).
B-4 lists data sources for the control scenario run.
Table B-5 shows household shares prepared for
the IFF model. The total number of households in-
creased from 63.4 million in 1970 to 93.3 million in
1990. A gradual shift from rural to urban was observed
with movement to suburbs within urban areas. The
effect of economic downturns in 1975 and 1980 was
an increase in share for the lowest income category;
more households moved to the highest income group
from 1970 to 1990, while the lower middle income
group share expanded and the upper middle income
share declined. The rate of household formation was
high during the 1970's, which resulted in increases in
smaller and younger households. The trend in younger
households reversed after 1980 as household forma-
tion slowed. Average household size dropped from
3.2 in 1970 to 2.67 in 1990. The number of licensed
drivers increased throughout the analysis period as
more and more young people were licensed to drive.
Data for the VOP model included disposable in-
come per capita, fuel price, overall personal vehicle
fuel economy, and annual usage in terms of VMT.
Table B-6 shows these data for each year in the analy-
sis period.
Data preparation for the model that projected
household vehicle composition was limited to char-
acterization of existing technology vehicles. Seven
vehicle size and type combinations were character-
ized for 1975 and 1980 while one vehicle, minivan/
small utility, was added for 1985 and 1990. Control
scenario vehicle characteristics are tabulated in Table
B-7. TEEMS' activity and energy computation pro-
cedure was executed to produce personal vehicle travel
and energy consumption estimates.
Commercial track travel was not modeled but,
historical data published by the FHWA (FHWA, 1987;
1991) were used. FHWA publishes track travel by
three categories: 1) 2-axle, 4-tire tracks; 2) single unit
" Telephone conversation between Jim DeMocker, EPA/OAR and EPA/OMS/Ann Arbor Laboratory staff (date unknown).
Nevertheless, the Project Team did consider reviewing emission factors for European automobiles to attempt to estimate no-control
scenario emission factors for 1975 through 1990 reflecting the use of electronic fuel injection and electronic ignition but no catalytic
converter. However, the Project Team concluded that differences in fuel/air mix ratios used in Europe would probably obscure any
differences in emission rates attributable to the use of electronic fuel injection and electronic ignition.
-------
Appendix B: Emissions Modeling
trucks; and 3) combination trucks. All 2-axle, 4-tire
trucks were treated as light-duty trucks. VMT by per-
sonal light trucks were subtracted from the published
totals to arrive at commercial light truck VMT. Die-
sel truck VMT shares of total VMT were obtained
fromTIUS (DOC, 1981; 1984; 1990). TIUS data were
also used to split VMT by single unit and combina-
tion trucks. All combination trucks were assumed to
be the heaviest, class 7 and class 8, while single unit
trucks could be of any size class 3 through 8. Gaso-
line and diesel VMT totals were developed for these
heavy-duty trucks and were kept constant for the con-
trol and no-control scenarios.
>1 Scenario Aj'tifyity^rqjec'tion.' s-*nv,
(Ujcaiioa, t
Statistic a! Abstfacirofttlte Oittittd States, editions 96th,
," 104th, IQSfo, and ,lf3fh^- '""-w*
' ' - ---1 *" ^ s
-
4 v Statistical-Abstracts aad EHWAHIgl^ay, Statistics- ,.,,-,v
>/,'4 -'•
;»j»«..»,. <.
,
"VNews Miricerb'ata'Book
*->•>$: ,..'- "
,,
''' DVSA'M -•. Dfe^ggi^ga't|xVeMcle Stock
' ' '
B-17
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
TableB-5» Distributionof-Hoaseholdsby'DemograpMcMribur^toV.ControlScedatfo.,""*,,, ,„ ,',,„;<;:
Household (Million)
Population (Million)
Attribute
Location
Central City
Suburbs
Rural
Income (1990$)*
<$ 13,000
$13,000 -$33,000
$33 ,000 -$52,500
>$52,500
Age of Householder (YF
<35
35-44
45-64
> = 65
Household Size
1
2
3-4
Licensed Drivers
0
2
> = 3
: 63-4 " 714 -- ; "' ',80,8",,, - 864. ---,933- >,
,.iu,, '•"''' '' ' ,;-,,;Household',Percentage,_,by Year ^_4;;,,;,; •/ '
1970' °'*'t./';' "i975 ~'",;j$M ' 1985,', >"*;';.19^°; ,",'-,
-/-'-*' '_„„,,, : *•' - -^*- „',,.,. " ,:'*"*.' ,-**•';
33.2,,- ' 321)"^' „", :,;-31.9 _, , ' /31.6 -""""' "3L4,vv -
33'.2 • ^i-2^0, -, -' 314 '_,/ '303 --y'^|Q.,3 /'
-25,9 ,,\,, ', "^5,'U'. "'"'' 26.6 ' " t^f25> - ; ','25,5, '
\.v - ,-;^3410:i^"' ' ^,,o,v-;37,2 - -;<-r-!;";-'37,4 '37.7 , ,;' 38.0^,,--,
" ^'5, '**';S^-^ ^ 13'-6' V"T-^ '*"""' '3;,;
- : ' /l8^;'-;; /16.7 ,, --'--'173 , v-,,-,,;204-'"!' '"'" " -22.1
,,,',36.3 :*"' -!"'34.0,, ,' ' ^I3.f ^^W£^-^^'^$&^
- •- -; j'7,'.2;l-«A v\ 19.5 '"'"'T,^'.? """''' " 23.7'v'" '"'""246,, "
„ -•,-, 's/--s J' ^(K _ „ ; >•(*<. *t* 'J/J'Jr>n 17 s , qt Q-N "21 >i ll'> ^ ,/ '/•"•
, , t „,- •3yAJ , ,t'(v ^Jv.f ' , 5I;Jf , ^^"- ;^s^ +'**v| ^ %*•'*«'
/^^3ft*8^"^/V," s ^6*8 *'" %*'^V^12i8 11,2" "v ^ 10-4 /^^
•^5j , -^ ;'5='"'"< ' ' , " ^-v-^---'J">^J'?s""r X,,A ,^ S&fV*^y!' * , ' 'f' '**** *S**'f»/« ' , ,V-* N " ' ",'' ' ^/J'A" ^ ""' "
*\rt •<>•>= ^^'^^ J *m * *\n rt '•- n^ !•> "'>A''f> *'
/ - * £rl *Q ^ (, sAtfaij * 4rf.\) j£\J*Ar^^ ^ £»\1AJ , , r f
^t , -_^ ,48.i,. /-"' ' - 49-2'^';;^ >tj&ti.. ' -^fZ~- '-'— 5^?; '*
Notei *Approximated to 1990 dollars.
B-18
-------
Appendix B: Emissions Modeling
posable Income Fiiel Price
S^?f"^r- '- ^ ' :&2?&;-% "V-;,,^ 1,06 '.;•'"";,';; Jl3,f;'";^-;i.7'?^.
'{-'|'l: ' -.,v,^ ' ,^A-,\ ,„,;--,'- '•' ; - -,-'A*-' -;';:^:,< ,,^,>4.
t'Tc"'-''^ """"^ '"^ "-"'''-o-'-'Vjtrv .-• ••••"•'•'. - •.--,' "",''i'ferv»,-* "• " i ^* *••••• •• '"" ^ V,-\------ *\ iift^.# ""
<-1979 ', '<5-feV, -:«tte?
%ifc;C.;^'":: :X°ft|^: , ' r:$f r'*'" '•, • ^^• "*^'',, 2^212^"
A9wl^'—^3^,p^;-;%fe%ii iL55 r*:4'^^f. "ta?fe;;^^S
„ ^->^<,^ --'.-,^S£S
•••^SS^offlS
V, ,„ ^', ^V^Il^"-T
„ " ' -, '/ "•". ™, -^ ^>ig, , ^; „
B-19
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table B-7. Control Scenario Per&rfalCharacteristies.*
-----?; ,- • --0^-
VafcaleT!i?i» , , G%«b; ,« ,&ttg»ft6,, fuel'*'"''
nUtOCDODU.6' , ^ •,v,v..j45 - 9.9 " '' "
Utility (1145) , - .;r^>-;"^»r-- ' '
MtnivarfSmall ' „ '" '"*w *VV -*•->*- -•- ,
Utility (7-8) ' >--- , -","s>.«*;^>*-X--
„ *»-;, '.v v *°*;' " '• "t"
' - "X
,,„ C^J*
,<, ^
-2;535-
- 3,736'''
'
4,455
, 3,580 '
-, ,4;975,
't'/'^ y ?'*
' :', /?"-,l?»
1980
Engine Fuel
Bower Economy
"-S3 19^
, ,,- 105 W.9
116"-""^'^15'.-1. ,
^^ ^
143 - -12.4 ,
' , 99 'I,'l5.9
144 ""••'•li'A
*"•*»•'•»•<• --><*'•• ^g
VehidcType * Curb . Ejngi»
^eats) '0^^^, |hp)
Automobile' " ' * '- -^^r^^fc-
Small (2-4) - -,2,^25 , ;,75
Compact (4) 2,775 90
Mid-size (5) - 3,I?0 ,10,8*
I^rge{6) ' * '' 3,973 f "'l'35;
Light truck ^ , ,,,^,, ,-,»^ ^fv.,
Std. truck 4i Jtfiff, ^§2
•• ^ 1"fl v) _
Compact 3,495 90
Std.Van/Std. 4,920 ,142
Otaity(ll-15) - "•"<-, •-'.->->
*WW .,_.,..' ,4;12l;"m'
5^^g;^ " '"* ' J^'jf~*^'' - - 299Q
^%ia '"Ccarb^k^ ' Hug-*
^J^conpnly <«w/*,v^w Witiiut InnriftT Economy
"• SS-??""'^^"" , , , 2,133 ^ 75 '-24*9-;? ;-
' *«. ,< ^^-^ ^ " ,',
•\f\ s± , ,,.. 'A> sO' xo<^ (in s' '*y/i Y^ •?':''
t , i"'3 - A J"--> -A/ xU'tv
' lj *if\ s ' ^ fwn ' ins ' v 10 *t '
*^/^ ^*, ^•ft f V«-o ' *"> svJjlAJU lUq v ly-^>
•.••vOs v V> v^ •. ' "" '''5<'^* O-.) ' '' s* ' ' ' '* ' ' S '
*-&§*y^\A A ' ' a 7/\c lart 17 1
i*+>V_. j^ /v«i> v f^l-pfj fiff f ^1 f >J.,
, ' 13vl ^v^iV^;, --s* x' ' 4,OQQ 128 * vi4.sl^ ,
''"'Afcfc*;^ ^ v '3;§fiS' 9,., ^b/?*? 138 12:9
7;' *r^7/- -^7*'' 18>2-
;-/:--
^
- „,-,
•v, -:,
NntP.r, *A verpges for all vdi icle$ of earfi type 'a'nd sizet'
B-20
-------
Appendix B: Emissions Modeling
U Blsttibution of JpfQtfsetohfe by fiicome Class'
v ^Honsefaold Shares (%)» by Year,.,
:'---'^;/-^-rr--.1975. ^ ; ^ j_ggft ---'-j^
1990,
vs^-.r- -26.3
4(3.000-33;o'00-"-,~ . 37,3
--13.-6
-26.2
' 37.6 , ,38:4,
22.6' ;;;:?f.Q
13.6 ---14.3
24.7
38.4 "'
.-y:
22.v
i4:3x^:
No-control Scenario Emissions
The control scenario data were modified to re-
flect no-control scenario emissions using economic
changes predicted by the J/W model, EPA, and ANL.
The J/W model predicted a slight loss of employment
and drop in GNP in terms of nominal dollars. How-
ever, the lower rate of inflation coincided with a real
GNP rise. ANL's information from the model did not
include any indexes for converting nominal income
to real income. ANL assumed real income changes to
be similar to those of real GNP and modified house-
hold shares by income classes accordingly. The model
also predicted a slight drop in refined petroleum price
beginning in 1973. The predicted drop was the larg-
est (5.35 percent) in 1973, reached the lowest level
(2.16 percent) in 1984, then increased to a second peak
(3.44 percent) in 1988, and dropped again from 1989
to 1990. Since these changes were inconsistent with
historical patterns of leaded and unleaded gasoline
price change, ANL developed an estimate of changes
in fuel price resulting from the cost of removal of lead
from gasoline and other infrastructure costs involved
with distributing a new grade of fuel. Subsequently,
EPA provided a set of fuel costs for use in the analy-
sis. Both ANL and EPA fuel prices followed a similar
pattern, although their magnitudes differed. The
no-control scenario was analyzed with EPA fuel
prices. ANL also established a relationship with cost
of regulation/emission control technology, and the
effect of costs on vehicle price and fuel economy di-
rectly from the EPA publication Cost of A. Clean En-
vironment (EPA, 1990). These changes were used in
the analysis.
The IPF model was executed for target years 1975,
1980, 1985, and 1990 using a set of revised house-
hold shares by income class. Table B-8 shows the re-
vised shares. Comparing Table B-8 no-control sce-
nario shares with those in Table B-5 for the control
scenario, there seems to be a slight shift away from
travel by the lowest, income group and toward the
middle income groups.
The vehicle ownership projection model was ex-
ecuted for the above four target years using the data
listed in Table B-9. Changes in fleet characteristics
are summarized in Table B-10.
B-21
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table B-9. Economic aiid Vehicle Usage'Data- for -Vehicle Ownership",
Projection ~ No-control Scenario.. ,";,;, ,"'",'
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
Disposable
Income per"
Capita (84 $)
7;5lt
-' -7,769' '
7,990
.8,463 '
8,297 -
8,406
- 8,600
8,795
9,126 -
9,114 ',
9,158
9,116'-:
9,312 *
9,775
9,97& <,
10,244
' 10,282 "
10,676
10,827
11,019
*-" Fuel 'Price
- - ,, 0,91-''*
' ' l,/,;0.8;8 "' '
•- -ores'
-'""" 0.84""
1 /\,, 1-06
... ;u;oi.
: ::Si,,
v T'VO.9'6
rr,, 141' '
'- >' -- 1 51°"- -
\'f'f '' * if+fri' ^j-WO
;^,/'i.53ft":.
^ t.t 3^0
*s * '* '
„ ' '\f^,,;;-
,«-, ' Lfl6^,,,
0.-84 •
:';-;v-;o;86 "
- - '---ass
„ 7-0.88 '
,-'T|?7- -
-Miles'/ -
! , ' " ,"' '/' "
"„„;,,- '13.5 ' -
,, i'3:§
" ^;';i3.4,
, ' ,1-3*5-:--; '
_ \:'i3ju--,<-
1 ' Uft
"",14-4^'r, x,,
,,- "-,-15.5
kX'.~..Xo
-;;r;;,,,, f6,.s
^•--17:2
17.9' , ,,
- - ••- 18.3 -- ,-
-- -18.4 -----
,-,;,•- :19A ,*-^
,,;-. '/V20.i^ ;„'
,s"/2Q.s'->
: * Ij-^^;-
- ; ,. ,,,-*'
/-;-'"'
10,143 ';i
^ -10,247 '; ~°-
^5^
l°,',lM, -''••'••
' ,9',56'9,
9,736"''^"
' 9,85'4 ' "
9,,?63, <"•?
,;4 16,17,4 ,-,--
' r:;S-:
'9,234 , • --
9,447 - -
- '' 9,45-0"!
', 9,582, •
- 9^07"-:,-^-
' - 9,738- - - ,'-'
''';'l6,201
-io^i4-v!
v:9;,902 .,„„
-,,';'^>, , , ,,
. Thseffiictof rftduttiotlsin veHdejjriceaiid-vfthicle'operatuig cost, andinoreases,-- -,>-•------ ' .'' .
B-22
-------
Appendix B: Emissions Modeling
^fablfB'la.^elc^nt ehlfl|e$ it* K$ Vehicle Characteristics
thedoMtol'aa4H< ~—rvw:"-"s" • -"""""
1975
,. ' ,1980' -;', ,-
,,,-.„, >v,'-"'- ' -'" * • * "
Ptlfce-'-" aapg HP'. Ftife^ ntpgi:,"*gfr..
-2.35 0.01 '0,59
-2^35 0,01 ,,0,59'
- C>*s'v^ N'' " Is?"*'' "'
„, "J "jJkT^ 3 0+ 22
Std
, ;^ -i,>5u -u;ur
^^ "• ;"s
r^.. .Lsa-VO.Or 0:59
-, •• N •?<&' M-."*-'*1
,-2x71 ':0.22 LSP.
;iF
;0 -3.ZS 0,62
^ r j s/<.
ip&ct Aoto -3,25 0.62
,-, ,-294 0.95 ^ °2.77 -
-2,9F 0.95"s:>
e"Alito ,,'/"»3J;5,; ;OL62\,-420 ' ,-,Y:> "2,94 "0.95-_ ^2,77*-
->;;^* ' ' \>;s ' ", : -*;>& v'^" -' - '"'.'.„ , , ,,. " ' *_ J
"'" "'"\ 0.62
.95*-'-2.77,,,
% ^
W " 2,77--
* 0/62' - :-
Utility v'"
B-23
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Utilities
The electric utility industry retrospective analy-
sis was prepared using two different utility simula-
tion models. ICF utilized its CEUM to estimate con-
trol and no-control scenario emissions for SO2, TSP,
and NOx in each of the target years. ANL's ARGUS
model was used to estimate electric utility CO and
VOC emissions for the same period. This mix of mod-
eling approaches was used because, while CEUM was
determined to be a better tool for examining fuel shifts
that were affected by the CAA than ARGUS, the
CEUM model was not initially set-up to evaluate CO
or VOC emissions. Although CEUM can be (and even-
tually was) configured to provide emission estimates
for pollutants other than SO2, NOx, and PM, ARGUS
was already configured to provide VOC and CO emis-
sions. However, it should also be noted that VOC and
CO emissions from utilities are quite low, as efficient
fuel combustion reduces both pollutants. Thus, for this
sector, the presence or absence of the CAA would not
produce any different VOC or CO control techniques.
VOC and CO emission rates for this sector differ pri-
marily based on the fuel and boiler type. Therefore, a
simpler modeling approach was judged to be accept-
able and appropriate for these two pollutants. This
chapter presents the methodology used to estimate
utility emissions under the control and no-control sce-
nario using the CEUM and ARGUS models. The
method used by Abt Associates to estimate lead emis-
sions from utilities is also presented.
Overview of Approach
The CEUM model uses industry capacity data and
specific unit-by-unit characteristics, operating costs
data, electricity demand estimates under the control
and no-control scenario, and historical fuel prices to
estimate SO2, TSP, and NOX emissions for 1980,1985,
and 1990. Changes in electric utility emissions, costs,
and regional coal production were developed using
ICF's CEUM with a calibration to historical electric-
ity generation, fuel use, and emissions. The ARGUS
model, which was used by ANL to estimate utility
VOC and CO emissions, is driven by operating costs,
industry capacity and generation data, demand for
coal, and unit-level operating characteristics. The J/
W model is used to incorporate predicted changes in
electricity demand under the no-control scenario. Fi-
nally, Abt Associates relied upon energy use data, the
Trends data base, and the Interim 1990 Inventory to
calculate utility lead emissions based on coal con-
sumption. The approaches used by each of these three
contractors are discussed individually in the follow-
ing sections.
Establishment of Control Scenario Emissions
A common feature of the approaches taken by ICF
and ANL was to identify conditions that are inputs to
the CEUM and ARGUS models, respectively, in the
control scenario. Later in the analysis, these variables
were revised to reflect no-control scenario conditions.
The next section discusses the specific assumptions
used in the CEUM analysis.
Key Assumptions in the Development of the
ICF Analysis
At EPA's direction, ICF made several assump-
tions in conducting this analysis for purposes of con-
sistency with other ongoing EPA efforts assessing the
effects of the CAA. These include the macroeconomic
assumptions regarding the effects of the CAA on eco-
nomic growth, or more specifically, electricity de-
mand, developed from other EPA commissioned ef-
forts. Each is described briefly below.
Pollution Control Equipment Costs
Only limited actual data were available for this
analysis on the historical capital and operating costs
of pollution control equipment. Accordingly, for this
analysis, the actual capital and operating costs of
scrubbers were estimated using EPA scrubber cost
assumptions adjusted to reflect actual data from a sur-
vey of scrubbed power plants with scrubbers installed
during the 1970s and early 1980s. For those power
plants with actual survey data, actual capital costs were
used. For other pre-1985 scrubbers, ICF relied on the
average costs from the survey data. For particulate
control equipment (primarily electrostatic precipita-
tors, or ESPs), costs were estimated based on limited
actual data, and a 1980 Electric Power Research In-
stitute (EPRI) study of ESP and baghouse costs. Based
on this information, ESPs were estimated to cost an
average of $50 per kilowatt (in 1991 dollars). The
development of more detailed data on actual power
plant pollution control costs was beyond the scope of
ICF's analysis. ICF concluded that such an effort
would not significantly change the national or regional
cost estimates developed by its approach.
B-24
-------
Appendix B: Emissions Modeling
Electricity Demand and Fuel Prices
Consistent with other EPA ongoing analyses, ICF
assumed that the CAA resulted in a reduction in elec-
tricity demand of 3.27 percent in 1980, 2.77 percent
in 1985, and 2.97 percent in 1990. Also consistent
with these studies, ICF assumed that natural gas prices
and oil prices would not be affected by the CAA. Coal
prices were estimated to change in line with increases
and decreases in demand for specific coal supplies
(and consistent with ICF's detailed modeling of coal
supply and demand). The average prices of all residual
oils consumed were also estimated to change due to a
greater use of more expensive lower sulfur residual
oils under the CAA.
Coal, Nuclear, Hydro, and Oil/Gas Capacity
At EPA's direction, ICF's approach was based
on the assumption that no changes in the amount of
nuclear, coal, hydro, or oil/gas stream or combined
cycle capacity would be built or in place in 1980,1985,
or 1990. Given that the driving factors associated with
the actual decisions to build new baseload capacity
were not based solely on economics but entailed fi-
nancial, regulatory, and political factors as well, the
actual effect of the CAA on these build decisions is
very uncertain. To the extent that more coal-fired
power plants would be built and fewer oil/gas-fired
power plants constructed, the actual emissions reduc-
tions associated with the CAA would be greater than
those estimated by ICF, while the estimated costs of
the CAA would be greater (because fewer, lower-cost,
coal-fired power plants would be on line under the
CAA). However, the CAA had virtually no effect on
the costs of constructing new coal-fired power plants
that came on line prior to about 1975 and a relatively
moderate cost effect on coal-fired power plants that
came on line through the early 1980s (since these
power plants were not required to install scrubbers).
Since a large majority of coal-fired power plant ca-
pacity came on line prior to 1975, ICF concluded that
the effect of the CAA on the amount of total coal-
fired capacity was not expected to be very large.
Natural Gas Consumption
The analysis assumed that the amount of natural
gas consumed under the no-control scenario could not
exceed the actual amount of consumption in 1980,
1985, and 1990. In part, because of natural gas price
regulation and the oil price shocks of the 1970s, natu-
ral gas was often unavailable to electric utilities in the
early 1980s. Since the CAA is relatively unrelated to
the questions of supply availability and price regula-
tion of natural gas, ICF assumed that no additional
gas supplies would be available if the CAA had never
been adopted. It is possible, however, that in the ab-
sence of the CAA, industrial and commercial users of
natural gas would have used more oil or coal. To the
extent that this would have occurred, there would have
been more natural gas supplies available to the elec-
tric utility sector. This increase in supply would have
resulted in an increase in the estimated costs of the
CAA, and a corresponding decrease in the estimated
emission reductions. ICF concluded, however, that this
effect would not be very significant.
State and Local Environmental Regulations
At EPA's direction, ICF assumed that there would
be no State and local emission limits or other emis-
sion control requirements under the no-control sce-
nario. Accordingly, ICF assumed that there would be
no SO2, NOx, or TSP emission limits under the
no-control scenario and that all scrubbers, NOx con-
trols, andESPs/baghouses (at coal-fired power plants)
were installed as a result of the CAA. (The more lim-
ited amount of particulate control equipment installed
at oil-fired plants was assumed to have been installed
prior to the passage of the CAA.) In the case of par-
ticulate control equipment, some ESPs and other
equipment were installed at coal plants prior to the
1970 CAA. To the extent that this is the case, the es-
timates of the costs of meeting the CAA have been
overstated. ICF concluded, however, that the amount
of such capacity was not substantial.
Retirement Age
The analysis assumed that unit retirement age was
constant between the control and no-controls sce-
narios. Adoption of this assumption might bias the
emission reduction estimates upward to the extent
turnover rates of older (and presumably higher-emit-
ting) units may be slower under the control scenarios,
because more significant CAA control requirements
focused on new units. However the vast majority of
existing coal and oil capacity was built after 1950 and
it is generally acknowledged that a relatively short
technical plant lifetime would be about 40 years. As
such, even if the no-control scenarios resulted in no
life-extension activity, there would be virtually no
effect over the 1970 to 1990 timeframe of the analy-
sis.
B-25
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
ICF1975 Control Scenario Emissions
The 1975 emissions under both scenarios were
calculated differently than emissions in 1980, 1985,
and 1990. In calculating or estimating 1975 SO2 emis-
sions for the control scenario (i.e., "actual" 1975), the
weighted average emission rates at the State level, in
the year 1975 were estimated, based on plant level
average sulfur content of fuel deliveries from Federal
Energy Regulatory Commission (FERC ) Form 423
and assumed AP-42 sulfur retention in ash. These
weighted average emission rates were then applied to
actual State-level electric utility fuel consumption in
the year 1975 (DOE, 1991). In the case of NOx emis-
sions, first, an estimate of Statewide NOx emissions
in the year 1975 was derived based on the use of the
same NOx emission rates, by fuel type, as developed
for the 1980 no-control scenario modeling runs. These
emission rates were specific to the fuel type (coal, oil,
or natural gas). These Statewide NOx emission rates
or factors were then applied to actual fuel consumed
by electric utilities in the year 1975, in order to obtain
estimated "actual" 1975 emissions. As before, the fuel
consumption at a State level was derived from iheState
Energy Data Report (DOE, 1991). ICF calculated the
weighted average heat content (BTU/lb) by State from
the 1975 FERC Form 423 data and used these figures
with the TSP emission factors (Ibs/ton) to derive emis-
sion rates by State (Ibs/MMBTU). These emission
rates were then applied to 1975 fuel consumption es-
timates obtained from the State Energy Data Report.
For the control scenario 1975 estimates, ICF used the
1975 factors.
For the remaining target years, ICF used the re-
sults of CEUM runs that provided fuel consumption
figures in 1980,1985, and 1990, respectively. Emis-
sions were then calculated using the appropriate emis-
sion factors for each year.
ARGUS Modeling Assumptions
The portion of the electric utility sector analysis
conducted by ANL with the ARGUS model is de-
scribed in this subsection. ARGUS contains four ma-
jor components: BUILD, DISPATCH, the Emissions
and Cost Model, and the Coal Supply and Transpor-
tation Model (CSTM). An overview of ARGUS can
be found in Veselkaef al (1990). Only the DISPATCH
and CSTM modules were used for the present analy-
sis. A brief description of the ARGUS components
used in this analysis is found in the following subsec-
tions.
DISPATCH Module
The DISPATCH module contains a probabilistic
production-cost model called the Investigation of
Costs and Reliability in Utility Systems (ICARUS ).
This module calculates reliability and cost informa-
tion for a utility system. ICARUS represents detailed,
unit-by-unit operating characteristics such as fuel cost,
forced outage rate, scheduled maintenance, heat rate,
and fixed and variable operating and maintenance
(O&M ) costs. These components are used to effi-
ciently compute system reliability (such as loss-of-
load probability and unserved energy) and production
costs.
The input data required by .ICARUS include
monthly load duration curves, annual peak demands,
and, for both new and existing units, unit sizes, capi-
tal costs, fixed and variable O&M costs, fuel types
and costs, heat rates, scheduled maintenance, and
equivalent forced outage rates. The output from
ICARUS includes annual summaries of capacity, gen-
eration, cost, and reliability for the entire generating
system.
CSTM Module
The CSTM module determines the least-cost com-
bination, on a per BTU basis, of coal supply sources
and transportation routes for each demand source.
First, it estimates coal market prices based on regional
demands for coal from all economic sectors. To gen-
erate market prices, CSTM estimates regional coal
production patterns and coal transportation routes. The
CSTM input data are grouped into three major cat-
egories: demand, supply, and transportation. CSTM
uses supply curves from the Resource Allocation and
Mine Costing (RAMC ) Model (DOE, 1982). Every
region has a separate curve for one or more of the 60
different coal types that may be produced in that re-
gion. CSTM modifies the original RAMC supply
curve by dividing the single RAMC curve into two
curves, one representing deep mines and the other rep-
resenting surface mines, but still uses the same ranges
for heating values and mine prices that define the sup-
ply curves in RAMC. Prices fluctuate as a result of
different mining methods, size of mining operations,
reserve characteristics, and depletion effects.
The transportation data defines the network that
connects 32 coal supply origins with 48 demand cen-
ters. Transportation cost is affected by distance, ter-
rain, congestion, variable fuel costs, cost escalators
B-26
-------
Appendix B: Emissions Modeling
for fuels and facility upgrades, and competition.
CSTM first computes the production cost for each coal
supply region and coal type. It then matches supply
sources with transportation routes to find the lowest
delivered costs.
Coal demand for a particular region is based on
the amount, geographic region, economic sector, and
range of coal types. There are 44 domestic demand
regions. CSTM allows demand to be met by one, or a
combination of, different supply regions.
The ARGUS input data for existing units are based
on the Argonne Power Plant Inventory (APPI). APPI
is a data base of operating and planned generating units
in the United States that was current through 1988 at
the time of ANL's analysis. This data base is updated
annually based on information in the regional North
American Electric Reliability Council (NERC ) re-
ports, reports from the Energy Information Adminis-
tration (EIA), and other sources. Unit operating char-
acteristics (fixed O&M, variable O&M, heat rate,
forced outage rate, and scheduled maintenance) are
based on regional data as defined in the EPRI report
on regional systems and other historic data (EPRI,
1981).
ANL used the 1988 inventory to generate a 1990
inventory. The 1990 inventory was then used to gen-
erate a separate unit inventory for the target years
1975,1980 and 1985. The target year inventories were
generated by removing units whose on-line year was
greater than the target year, from their respective in-
ventory. The regional capacity totals in these prelimi-
nary inventories were tabulated by major fuel category
(nuclear, coal, oil and gas steam) and compared to the
regional historic NERC totals. This review identified
capacity differences, especially in 1975 and 1980 in-
ventories. The original plan was to add phantom units
to match the regional historic totals. However, based
on the need for State-level emissions, it was decided
that a more thorough review of the unit inventories
was required.
ANL's detailed review included an examination
of the nuclear and coal units greater than 100 mega-
watt equivalent (MWe) in each target year. Missing
units, with the appropriate unit size and State code,
were added so that the regional totals were compa-
rable. The availability of coal units was based on the
on-line year of the unit as reported in the EIA report
Inventory of Power Plants in the United States (DOE,
1986). The coal units were also checked against the
EIA Cost and Quality Report (EIA, 1985) to verify
the existence of flue gas desulfurization (FGD ) sys-
tems in each of the target years. The nuclear unit in-
ventories were verified with the EIA report An Analy-
sis of Nuclear Power Plant Operating Costs (DOE,
1988). The review also included oil and gas steam
units greater than 100 MWe. The total capacity of the
oil and gas steam units were compared because many
units switched primary fuel from oil to gas during the
relevant time period. The oil and gas units were com-
pared to historic inventories based on information pro-
vided by Applied Economic Research. In addition to
thermal generation, the hydro and exchange energy
was reviewed. For each target year, the hydro genera-
tion and firm purchase and sale capacity data was ad-
justed to reflect the historic levels. These two compo-
nents, hydro and firm purchase and sales, are ac-
counted for first in the loading order. If these vari-
ables are overestimated, there will be less generation
from coal units. Likewise, if they are underestimated,
there will be too much coal generation. The hydro and
firm purchases and sales can vary significantly from
year to year because of weather conditions and other
variables. Therefore, it was important that they be
accurately represented.
No-control Scenario Emissions
In order to calculate utility emissions under the
no-control scenario, inputs to both the CEUM and
ARGUS models were adjusted to reflect no-control
scenario conditions. The changes made to each
model's base year input files are discussed separately
in the following sections.
ICF Estimates of SO2, TSP, and NOx Emissions
in the No-control Scenario
As described earlier, ICF utilized a different meth-
odology to calculate 1975 emission estimates. Rather
than relying on the use of detailed modeling runs, ICF
based the 1975 emission estimation on historic fuel
consumption and sulfur content data in 1975. This
subsection first outlines the process used to calculate
no-control scenario emissions in 1975 and then pre-
sents the methods used for the remaining target years.
1975 Utility SOf NO^ and TSP Emissions
To develop State-level no-control scenario utility
SO2 emissions, ICF developed no-control scenario SO2
emission rates. A reasonable surrogate for these emis-
sion rates is SO2 rates just prior to the implementa-
B-27
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
tion of the SEPs under the CAA. ICF developed 1972
rates (based on the earliest year available for FERC
Form 423) and compared these with 1975 rates. In
each State, the greater of 1972 or 1975 rates was used
in the calculation of SO2 emissions in the absence of
the CAA. To develop State-level no-control scenario
SO2 emissions, no-control scenario fuel consumption
data were needed. ICF assumed that the demand for
electricity in 1975 would be 2.73 percent higher than
the actual energy sales in 1975. This assumption is
identical to the no-control scenario electricity demand
projections derived from the J/W projections. For the
purpose of this analysis, it was further assumed that
this increment in demand would have been met in 1975
from the oil and coal-fired plants in each State. The
increase in consumption of these fuels was assumed
to be in the same proportion as their share in the 1975
total energy mix for electricity generation in that State.
It was assumed that the generation of nuclear, gas-
fired, and other electricity generation would not
change. A sensitivity case without an assumed elec-
tricity demand change was also calculated. (The sen-
sitivity analysis results are presented later in this ap-
pendix.)
For NO emissions under the no-control scenario,
it was also assumed that the 1975 electricity sales
would have been 2.73 percent higher than was the case
in 1975. No-control scenario TSP emissions in 1975
were based on national emission rate numbers from
EPA that were converted to pounds per million BTU
using the average energy content of fuels hi each State.
No-control scenario TSP emissions were calculated
based on 1970 emission factors (Braine, Kohli, and
Kim, 1993).
1980,1985, and 1990 Utility Emissions
For 1980, 1985, and 1990, ICF calculated
no-control scenario emissions based on fuel consump-
tion figures from the CEUM runs, and 1970 emission
factors from EPA.
Electric utility SO2 emission estimates are ap-
proximately 10 million tons (or about 38 percent)
lower by 1990 under the control scenario than under
the no-control scenario. Most of this estimated differ-
ence results from the imposition of emission limits at
existing power plants through the SEPs under the 1970
CAA. Most of these SIPs were effective by 1980 (with
some not fully effective until 1985). Most of the ad-
ditional reductions that occurred during the 1980s were
the result of the electric utility NSPS, which required
the installation of 70 to 90 percent SO2 removal con-
trol equipment.
By contrast, electric utility NO emission esti-
mates under the control scenario are only about 1.2
.million tons, or 14 percent, lower than under the
no-control scenario by 1990. This occurs because,
under the implementation of the 1970 CAA, only a
few existing power plants were subject to NOx emis-
sion limits. Virtually all of the estimated reductions
are the result of NOx NSPS, which generally required
moderate reductions at power plants relative to un-
controlled levels. In addition, electricity demand is
estimated to be about 3 percent lower under the con-
trol scenario. This decrease reduces the utilization of
existing power plants and also contributes to lower
NO emissions (and other pollutants as well).
Electric utility annualized costs (levelized capi-
tal, fuel, and O&M) are estimated to be $0.2 billion
lower in 1980, $1.5 billion higher in 1985, and $1.9
billion higher in 1990 under the control scenario. Note,
however, that this reflects the effects of two offset-
ting factors: (1) the higher utility compliance costs
associated with using lower sulfur fuels, and the in-
creased O&M and capital costs associated with scrub-
bers and particulate control equipment; and (2) lower
utility generating costs (fuel, operating and capital
costs) associated with lower electricity demand re-
quirements. In 1980, the increase in fuel costs due to
higher generation requirements (under the no-control
scenario), was larger than the decrease in capital and
O&M costs and thus yielded a cost increase over the
control case.
However, lower electricity demand for the utility
sector would translate into higher costs in other sec-
tors (as electricity substitutes are used). This effect
was captured to some extent by the original J/W mac-
roeconomic modeling conducted for the present analy-
sis.
Average levelized U.S. electricity rate estimates
are approximately 3 percent higher under the control
scenario during the 1980s. Note that year by year, elec-
tric utility revenue requirements and capital expendi-
tures (not estimated by ICF) would be estimated to
have increased by a greater percentage particularly in
the 1970s and early 1980s as incremental capital ex-
penditures for scrubbers and ESPs were brought into
the rate base.
B-28
-------
Appendix B: Emissions Modeling
Significant shifts in regional coal production are
estimated to have occurred between the control and
no-control scenarios. High sulfur coal producing re-
gions such as Northern Appalachia and the Midwest/
Central West are estimated to have lower production
under the control scenario, while lower sulfur coal
producing regions such as Central and Southern Ap-
palachia are estimated to have higher coal produc-
tion.12
ARGUS No-control Scenario
Regional fuel prices, for the thermal units, were
based on historic information from the EIA Form 423
data for the year 1977,1980 and 1985. The 1977 data
was used for 1975. Fixed and variable O&M costs
were adjusted from the 1988 level, and all cost data
were converted to 1985 dollars.
The load data were based on regional historic
NERC data for each of the target years. The shapes of
the monthly load duration curves are the result of
modifications based on the data in the EPRI report on
regional systems (EPRI, 1981). The shapes were modi-
fied to match the projected 1988 monthly load factors
for the NERC regions. These load shapes were held
constant for all years.
The actual peak-loads were selected from historic
information and used with the existing load duration
curves. The system was dispatched so that the calcu-
lated generation could be compared with historic data.
Discrepancies were resolved by adjusting the peak
load so that the annual generation was on target. This
procedure was repeated for each of the target years.
The electric utilities were expected to have an in-
crease in generation as identified by the J/W data.
Table B-ll identifies the increase in national level
generation by year. The national level increase in gen-
eration was applied to each power pool.
In addition to load changes, coal units with FGD
equipment were modified. These units had their FGD
equipment removed along with a 3 percent decrease
in heat rate, a 2 percentage point decrease in forced
outage rate, and a 50 percent decrease in their fixed
and variable O&M costs. These changes were incor-
, ,
Table B-l 1. J/^;JMmatesr of->; -v'
Percentage
Percentage
-1915
1985
1$9CT,,,
porated into the ARGUS model for each of the target
years. Model runs were then conducted to arrive at
estimates of VOC and CO emissions in the no-control
scenario.
Estimation of Lead Emissions from
Utilities
In order to estimate lead emissions from electric
utilities in each of the target years, data from three
different sources were used. Energy use data for the
control and no-control scenarios were obtained from
the national coal use estimates prepared for the sec-
tion 812 analysis by ICF (Braine and Kim, 1993). The
Trends data base provided emission factors and con-
trol efficiencies, and the Interim 1990 Inventory iden-
tified utility characteristics. The ICF data bases pro-
vided the amount of coal consumed for both the con-
trol and no-control scenarios in each of the target years.
A correspondence between the Interim Inventory and
the ICF data base was achieved through the plant name
variable. Using emission factors for lead and control
efficiencies for electric utilities, estimates of lead
emissions per plant per year, were calculated. These
factors were obtained from the Trends data base. It
was assumed that pollution control on coal-burning
power plants under the no-control scenario would be
the same as the pollution control level in 1970. There-
fore, the control efficiency from 1970 is used as the
basis for the no-control case.
12 At EPA's direction, ICF's analysis did not estimate the effect of shifts in non-utility coal consumption on regional coal
production, nor did it consider the possibility that fewer new coal powerplants might have been built due to the CAA as discussed
earlier. Both of these factors could result in a greater estimated change in total U.S. coal production than estimated herein although the
difference is not likely to be very significant.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
CEUM Sensitivity Case
In addition to comparing actual (control scenario)
historical costs and emissions with the higher elec-
tricity demand under the no-control scenario, ICF also
evaluated emissions in a sensitivity case without the
CAA (i.e., under the no-control scenario) with the
same electricity demand (versus the no-control sce-
nario with higher demand). The purpose of this sensi-
tivity analysis was to isolate the incremental electric
utility compliance costs and reductions in emissions
associated with the CAA from the lower resulting
generation costs and emissions due to lower estimated
electricity demand under the CAA. The incremental
effects of the CAA when compared with this case in-
dicate:
• Estimated reductions in emissions due to the
CAA are somewhat lower if measured against
the sensitivity case without the CAA with the
same electricity demand than the emissions
without the CAA with lower demand. This
occurs because lower electricity demand un-
der the no-control scenario sensitivity results
in lower utilization of existing coal and oil
plants which, in turn, results in lower emis-
sions. As noted above, in some sense, the
changes in emissions represent the effects of
electric utility compliance actions under the
CAA, absent the effect of lower resultant de-
mand for electricity.
• When measured against the sensitivity case
without the CAA (with the same electricity
demand), electric utility annualized costs are
estimated to have increased by about $5 to $6
billion during the 1980 to 1990 period. This
reflects the following cost factors: (1) higher
annualized capital costs associated primarily
with scrubbers and ESPs installed by electric
utilities to comply with the CAA; (2) higher
O&M costs associated with the additional air
pollution control equipment; and (3) higher
fuel costs associated with using lower sulfur
coal and oil in order to meet the emission limit
requirements of the CAA.
Commercial/Residential
The Commercial and Residential Simulation Sys-
tem (CRESS) model was developed by ANL as part
of the Emissions and Control Costs Integrated Model
Set and used in the NAPAP assessment (Methods for
Modeling Future Emissions and Control Costs, State
of Science and Technology, Report 26) (McDonald
and South, 1984). CRESS is designed to project emis-
sions for five pollutants: SOx, NOx, VOC, TSP, and
CO. The CRESS output is aggregated into residential
and commercial subsectors related to both economic
activity and fuel use. The introductory material pro-
vided in this appendix about CRESS describes the base
year as being 1985. It appears in this way because
CRESS was originally developed to operate using the
1985 NAPAP Emission Inventory as its base year data
set. For the five pollutants reported by CRESS, emis-
sion estimates are provided for the following sectors:
* Commercial/institutional
• coal, including point and area categories of
anthracite and bituminous boilers;
• liquid fuel, including boiler and space heat-
ing uses of residual, distillate, LPG, and
other fuels;
• natural gas boilers, space heaters, and in-
ternal combustion engines;
• wood used in boilers and space heaters; and
• other mixed or unclassified fuel use.
4 Residential
• coal, including area sources of anthracite
and bituminous;
• liquid fuel, composed of distillate and re-
sidual oil;
• natural gas; and
• wood.
4 Miscellaneous
• waste disposal, incineration, and open burn-
ing; and
• other, including forest fires, managed and
agricultural burning, structural fires, cut-
back asphalt paving, and internal combus-
tion engine testing.
In addition, VOC emissions are projected for these
source categories:
4 Service stations and gasoline marketing;
4 Dry-cleaning point and area sources; and
B-30
-------
Appendix B: Emissions Modeling
4 Other solvents, including architectural surface
coating, auto-body refinishing, and consumer/
commercial solvent use.
This section describes the use of CRESS to esti-
mate control and no-control scenario emissions from
the commercial/residential sector.
Control Scenario Emissions
For the NAPAP assessment, 1985 CRESS output
corresponded to the 1985 NAPAP Inventory (EPA,
1989), which served as the benchmark for any pro-
jections. The design of CRESS is such that emissions
by NAPAP SCC are input for each State, then pro-
jected to future years by scaling them to economic
data such as energy demand. In estimating emissions,
differences in emission controls associated with new,
replacement, and existing equipment are taken into
account where such differences are considered sig-
nificant. The basic modeling approach is shown in
the following equation:
-. • f'• --
,
^"^ ' *
where:
Q = emissions in year t or the base year, year 0
E = emission factor for the source category b
in the base year, or for a subcategory j sub-
ject to controls in year t (this takes into
account changes in emission rates that may
occur as a result of emission regulations or
technology changes)
D = driver data indicating activity levels in the
base and future years
f = fraction of total activity in year t differen-
tially affected by emission controls
The calculations are carried out in two subroutines,
one for SO2, NOx, TSP and CO, and one for VOC.
Typically SO2, NOx, TSP, and CO emissions are
projected by multiplying the 1985 NAPAP SCC data
or base year data by the ratio of the driver data (activ-
ity level) value in the projection year to its value in
the base year. Because there are few controls on SO
or NO emissions from the sources covered by CRESS,
projected emissions for most sectors are proportional
to the expected activity levels. Thus,
:*> 'rVCfew&^SJ^'* vv ••" * v""?^
JJ1,-? >-- ', ,7
*i -:*^:&n%Z,r'''"""" ^
'' '':^; ' ^¥--''-"-m^W^'
There are a few source types, such as commer-
cial/institutional boilers, for which emission controls
are mandated. These are modeled by multiplying the
1985 emission data by the ratio of the controlled emis-
sion factor to the base-year emission factor. Emission
factors for each source type are weighted by the pro-
portion of base year activity in each subsector to which
controls are expected to apply.
&_;,
where:
g= the fraction of base-year activity accounted
for by existing source b, replacement
source r, or new source n in year t
The effective emission factor (Et,n) for the sector
is calculated by'weighing the portions of sectoral
emissions subject to NSPS controls and those likely
to continue at existing levels. Ah appropriate Internal
Revenue Service-based rate at which new equipment
replaces existing sources is applied to each sector in
the model. This is done to estimate how emissions
might change as older sources are retired and replaced
by new sources that emit at lower rates.
The SO /NO /TSP/CO subroutine varies in new
X X
and replacement emission-source fractions subject to
NSPS controls. These fractions are applied to the
emission-source replacement rates. In addition, ratios
for new source emission factors are varied by State.
However, emission ratios for any pollutant/source type
combination do not vary over the projection period.
The VOC estimation methodology is similar, but
allows variation in emission factors over time. Emis-
sion ratios are calculated from files of replacement
and existing source emission factors weighted by the
replacement rate for each sector and new source fac-
tors by State. These are input for each 5-year projec-
tion interval. For most source categories, VOC con-
B-31
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
trols are not envisioned, and the 1985 NAPAP emis-
sions for the category are simply scaled proportion-
ally to changes in the driver (activity level) data.
For sources to which controls apply, a variation
on the following equation is employed:
(6)
In equation 6, the emission factors for new and
existing sources are effectively weighted by the pro-
portion of total activity in year t to which controls
apply.
In using CRESS for the CAA retrospective analy-
sis, the base year was 1975. CRESS requires emis-
sions information by State and NAPAP source cat-
egory as input. Since detailed information on emis-
sion levels for 1975 by NAPAP source category were
not available, the data were developed from a combi-
nation of sources. The procedure for calculating 1975
emissions based on the 1985 NAPAP inventory is
described below. The emissions module uses these
initial values in conjunction with activity estimates to
project control and no-control scenario emissions.
Emissions Data
Since the starting point for the analysis was 1975,
emissions data by State and SCC for SO2, NOx, VOC,
TSP, and CO were required. Available emissions in-
formation for this year was not at the level of detail
needed by CRESS. The 1985 NAPAP Inventory,
which contains the necessary level of detail, in con-
junction with information from EPA's National Air
Pollutant Emission Estimates, 1940-1990 (Trends') and
ANL's MSCET, was used to construct an emissions
inventory for 1975. The model then uses these emis-
sions as a benchmark for the analysis.
The method for constructing the 1975 emissions
data base was consistent for all pollutants; however,
two different sources of emissions data were neces-
sary in order to obtain time series information on all
pollutants. MSCET contains monthly State-level emis-
sion estimates from 1975 to 1985 by emission source
group for SO2, NOx, and VOC. Therefore, MSCET
information was used for SO2, NOx, and VOC, while
Trends data were used for TSP and CO. Emission
source groups from MSCET were matched with 1985
NAPAP Inventory SCCs. The MSCET methodology
is benchmarked to the 1985 NAPAP Inventory and
uses tune series information from Trends in conjunc-
tion with activity information to estimate State-level
emissions for SO2, NOx, and VOC. Although the level
of detail contained in the NAPAP Inventory could not
be preserved because of the aggregation needed to
match with MSCET emissions sources, MSCET pro-
vided the State-level spatial detail required by CRESS.
Once the 1985 emissions by SCC and State from
the 1985 NAPAP Inventory were matched with emis-
sion source groups and States from the MSCET data
base, an estimate of 1975 emissions was computed
by multiplying the 1985 NAPAP Inventory emissions
value by the ratio of 1975 MSCET emissions to 1985
MSCET emissions. Ratios were computed and applied
for each combination of State, pollutant, and MSCET
emission source group.
This method of constructing an emissions inven-
tory for 1975 utilizes the State estimates from MSCET,
thus capturing the spatial shifts that occurred over the
analysis period. It is assumed that NAPAP provides
the most reliable point and area source information in
terms of the level of 1985 emissions (which is also
the assumption of the MSCET methodology). Note
that if there were a 1-to-l correspondence between
MSCET and NAPAP, this method would be equiva-
lent to using the MSCET methodology directly for
constructing 1975 emission levels.
A similar method was used for TSP and CO, but
since these pollutants are not included in MSCET, the
Trends ratio of 1975 to 1985 emissions for these two
pollutants was used. Thus, for TSP and CO, all States
were assumed to have experienced the same change
in emissions as indicated by the national figures.
It should be noted that in addition to the loss in
spatial detail, the Trends source groups generally
spanned several NAPAP source.categories. The
strength in the Trends information is the consistency
of emissions estimates over time. It is considered to
be the most reliable data for tracking changes in emis-
sions over the time period of the analysis, and was
therefore chosen for developing 1975 estimates for
TSP and CO.
The 15 source categories reported in Trends were
matched with those in the 1985 NAPAP Inventory.
The ratios of 1975 emissions to 1985 emissions by
source category that were applied to the 1985 NAPAP
emissions data are shown in B-12. The 1975 emis-
B-32
-------
Appendix B: Emissions Modeling
-Table Krl^frends Source Categ >%V< *WS* ', ,' ,', :
, Commerci al/Jnsfitiiiacfjriail Fuel t,, v „,, '[^ ] ^
Coisbustioa; - —-x* /, ^ ^^, '• ,/,,','
FuetOil
•<::; 1*43
W'-'. '0.67
Coal ->•->> -: —
""" ' 1.47
>v, -s
I -00
; Frest Fires- -^^>.>,
^Soliff Waste Disposal: " * ; ,r-
'0,62
Incineration-'
,Op?ai Baroing ;',;,""v
*•<•••,•, ; , s > i •• -
tf^»llaaeou^Ofcer Borning - "'-""1.00
.Industrial Processes;" Caving; •
S /^ ' Cs"^',
?Asplialt Pavrajgs-and, Roofing""
036 >•
^ f 2.11 is coMj)'(ifed'asaielat!0 of the 197,5TSP-«sttss»as
of 40^1 agiam s J»r,)®ar to tfeft coiys{jfi&aiilg' 19 SS'onissitSis'of
1? gigagrarns perj
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
population. Forest wildfires and managed open burn-
ing activity are related to 1977 State-level forest acre-
age. The demographic information is assumed to be
invariant to CAA regulations and thus is the same in
the control and no-control scenarios.
Car stock (or vehicle population), the driver vari-
able for the auto body refinishing, is approximated by
State motor vehicle registrations. Highway Statistics,
an annual publication by the FHWA, was the source
for data on State motor vehicle registrations. The three
source categories connected with gasoline marketing
are driven by State-level gasoline sales in gallons. State
gasoline consumption was obtained from the SEDS
data base. Housing starts and 10 percent of the exist-
ing housing stock were combined to form the activity
indicator for architectural surface coating emissions.
Housing data compiled by the U.S. Bureau of the
Census were available in the Statistical Abstract of
the United States (DOC, 1975; 1977; 1982; 1983;
1987; 1993). Regional-level data for 1975 was allo-
cated to the States based on the 1980 State distribu-
tion.
No-control Scenario Emissions
Adjustments to control scenario emissions in each
of the target years to reflect conditions un-
der the no-control scenario were achieved ^^H
through emission factors, energy input data,
and economic/demographic data. The adjust-
ments made to each of these variables to gen-
erate no-control scenario emissions are dis-
cussed individually in the following subsec-
tions.
Energy Data
State-level energy demand for the residential and
commercial sectors for the no-control scenario was
estimated from the J/W model forecast. Final energy
demand estimates for the household sector were cal-
culated by an EPA contractor for the purposes of the
no-control scenario analysis. State allocation of the
national-level estimates was based on historic State
shares, i.e., this assumes that there is no change in the
distribution of energy demand across States as a re-
sult of removing regulations. In addition, the J/W
model estimates an aggregate refined petroleum cat-
egory and does not distinguish among liquid petro-
leum gas, distillate oil, and residual oil. The relative
shares among these three categories of petroleum prod-
ucts remained constant between the control and
no-control scenarios. The information on percentage
change in energy demand by fuel type as provided by
the J/W model is listed in Table B-13.
The differential for commercial sector final en-
ergy demand was calculated from the combination of
four intermediate product flow categories from the J/
W forecast. The National Income and Product Ac-
counts (NIPA) for the commercial sector correspond
to J/W SIC categories 32 through 35:
Table B-B, Percentage Change to Real Energy Demandt>y
Households' from Control to No-control Scenario.
Emissions Data
CAA regulation of the commercial/resi-
dential sector was minimal. For regulated ^ * "
source categories, emission factors were re- •"•mm••
vised to reflect pre-regulation emission rates.
Six commercial/residential source categories were
regulated for VOC emissions: Service Stations Stage
I Emissions, Service Stations Stage II Emissions, Dry
Cleaning (perchloroethylene), Gasoline Marketed, Dry
Cleaning (solvent), and Cutback Asphalt Paving.
Commercial-Institutional boilers were regulated for
SO2 and TSP and internal combustion sources were
regulated for NOX emissions. All NSPS were removed
for these sources to estimate no-control scenario emis-
sions levels.
Year „
1975
1980
1985
1990
C°al
1.48
1.50 .
1.98
2,23
Refined Petroleum
' 1 4.76
" , 3,S4 '
3,90
' 4,33
Ueetrte
3.62
4,26
, , 3,88
4,1$
jSattwa
-"2.42
2.12" ^
2-41 ..
2.77 " ;
(32) Wholesale and Retail Trade;
(33) Finance, Insurance, and Real Estate;
(34) Other Services; and
(35) Government Services.
Percentage change information from the J/W fore-
cast for energy cost shares, value of output, and en-
ergy prices was used to calculate the differential in
commercial sector energy demand for the no-control
scenario. The energy cost share is defined as the cost
B-34
-------
Appendix B: Emissions Modeling
of energy input divided by the value of the output. In
order to calculate the percentage change in commer-
cial sector energy demand, the change in energy price
was subtracted from the percentage change in energy
cost, and added to the change in the value of output.
Each of these variables was available from the J/W
model results. This calculation was performed for each
of the four energy types, and each of the four NEPA
categories. The change in commercial sector energy
demand was obtained by taking the weighted average
of the four NIPA categories. Since data on relative
energy demand for NIPA categories were not readily
available, square footage was used as a proxy for cal-
culating the weights. These data were taken from the
Nonresidential Buildings Energy Consumption Sur-
vey, Commercial Buildings Consumption and Expen-
diture 1986 (EIA, 1989). The resulting estimate for
commercial sector changes in energy demand is pro-
vided in Table B-14.
State-level gasoline sales is one of the activities
forecasted by the transportation sector model. The
percentage change in gasoline sales calculated by the
TEEMS model was used in the no-control scenario as
a CRESS model input.
Table B.-15;
Economic,Variables Us,ed,m CRESS.
^e'ar
%75
•• -^
:19SQ
1985
4990-
, •_ •**y' ^xx, , ' y Motor
, ,\ , , ^Construction.- ' W Vekijctes
,V., --o;70^ ^ „-, -, -, ^_Q4 ,
, f Vi3' •*' i '-.'''
"'" ' 0-1> -:- ;; ^,^t7^
, -^LTl - >- - - ^''°7 "" "
'''"•-* "Q.-29 ,w^,-6.a5,,,,V
-fableB^li
'' "
Refined s
Petroleum
Natural:-'
.0.80
19SO >^--'
,1985
1990'»'-
;,^- , , ,3-36,;'^ - , 1,40>>V
'*?'' ,,,- ,,IM'12, -s;1^-:, '"''-O'.'4p"
The national-level change in commercial sector
energy demand was allocated to the States using his-
toric shares. Implicit is the assumption that removal
of CAA regulations does not alter the State distribu-
tion of energy use.
Economic/Demographic Data
State population was assumed not to vary as a re-
sult of CAA regulations, thus only the economic vari-
ables were revised for the no-control scenario.
No-control scenario housing starts and car stock were
derived from J/W forecast information on construc-
tion and motor vehicles. The differential for catego-
ries 6 (construction) and 24 (motor vehicles and equip-
ment) was applied to control scenario values to ob-
tain no-control scenario levels. The percentage change
from the J/W forecast is given in Table B-15.
B-35
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table B-16. ISP Emissions Under the Control ati4,1Hor^^T'<>l'Scenarios by Target Year (in
thousands of short tons), , ,,'X, '" "'-•*
Sector
Transportation:
Highway Vehicles
Off-Highway Vehicles
Stationary Sources:
Electric Utilities
Industrial Processes
Industrial Boilers
Commercial/Residential
TOTAL*
,*;-,>>,*,
W5
700
270'
1,72£
5,620,
740
2,020"
11,070
w =.,s"*v '-.'j -^ ^,\^
--- 1S8O-- •
' ' 760 - -
•" - "280>v-^
^ , » V •f
*,*,., ^°
JjyssyL,.,'
,-.¥ ,,,430^
-"-•"2,510.-^-»
-¥-8,550- ,,,,
&A.
1585
< - „
-770-
--£70
450"
3,040'
250'
5,680
7,460
'-'" •:'/%:»•.
1990
':"'" "
**° 820 ,
, ,,280 '
-4" ;,;';?,;
"«?.'. 4fc
' ^''"3,680
/ 240
- , -2,550 '
7,390,<-
' 19WS
;':""770-
"**"•" ""260'
3,460
1,1,120
„ „ ,7|p
* ' -2;020
• 18,410
Without the CAA
'"later
*;- ,
w-., --910
270
- -,
' 4M80'
'''12,000
550
2,520
' 20,730
1985
" "--'-'
- 1,030
260
'-'- 5,180
11,7-10
360
' 2,700
21,250
'v^ "S?
* V*,199Q Emfesions
V,' ' ' '-" "'-
1,180 - ,(30%) '
' 270 ' 5% ,- '
^ , ; ; " ' - ,
5,S60 -(93%),
12,960 ' "'(76%) J
,' 400 (41%)
"••- -2/560 -(1%)
23t23'0' v (68%)>,
models designed to stmulate'condj.aotis in, {heabsence of ti» CAA, Tlesa niimbets should,not^b& taferpffiM as actual hfetotlcaJ •
emission estimates. ^ - < -r-.>-^4x>*.*..*. --• - , ' '^
J ' '"'' v "* - -. -.^ -,-.^.^, v^^^h , ,'V •, s
*Totals may differ sSghtly from sums tJae to rounding.
Table B-17. SOi Emissions Under the
of shorttons). . _,,
-,'-,„,,,''', '- ,'>!,<;,
"' ^'"'"?'os by Target-Year (in thousands
-v^>A1Kkr^ <:*,,,* , ,'-/,,-,
v''x:;-:-1990
. i 570' "
- ^-
** t
16,510 -
- 2,460,;-
:--^2,820
'?, ^9°
nn^'44-°*
' 19?S '
'SttfA'"/, -
* - '380
360
5'
,-, 20,6:%'"
- 4560
3,910
#i,oob
,'N,3l^'6'0'
'"-'198®
-,---450
530
,/,
"25,'^b'
.5,940
,,4,UO,,
?'' 810'
'37,460
1985
500
400
"•25,140
5,630
, 4020*'
610
''V&lftO''
^riif : tS?
jg^O Effliaioas
560 ' 1%' ' -
"-, --39,0 " l%-";
,'/,•>,,
-26,730 (38%).-;
'6,130 (60%),
'""4-H,610-- ~ (39%)
710''" ~,(3%>
"- 39,140, ^^C^bk)4'
models designed to simulate ootudifions «i the'a'bsen'c'f of ,tji» CA.A, itiese p wn^er| should not be in'ierpisteilfactual b|sf orioal
emission estimates. ' % " '':" ""/»«•>. - \- - ,'-'" ' ' '-'V;,-,,-," '•* """«---,-,^,,
, tT^S-v, ;„„--».- ''*Mfc ,w--v ,,v, ,,'<,"-„ -,^4,,y;,,,,,,
"Totals may differ slightly froin sums.duejKi foun|lagl
B-36
-------
Appendix B: Emissions Modeling
0fidertb&-€bj$roland^control'Sceaarios-bv Ta^t^fear (aft , ^ ^ „
"t X „ * JH* ^^= ~ - ~ v ( ~ >. ^%v v^ > < ^. ^ v^ £ >
1975
1990 1975 19»0
: 199® Emissions
Higbway Vdiides^' " ?,640 > 'A340"' x '8,61O
Off-Mighway VeMdiess> ' I^S&Ct,, ,2,180^/2,080
Stationary'Soutces: - v " •* * xv ^tj"" '
""' 5,54V- >>-*'(S,450\v,^46(SO 7,050
^ t 750 "',760' " " ^690 710
, Industrial"Pailers* 4,09p ^Qffff^,-^Q/ 3^710
TOTAL* -'-"'^ '22,06V 23,370 22.460- •* &%&>'
9,020 „, 15,060 -^13,160 15,39'0
1,980 " „ 2,150 " '"""
rocesses
5,740 7,150 >-7,780 8,300-
760"" -, 830 \790 1,090"
4,120* "\ 3,6*3 "3,~68Ow&ife<}eveloped Sf»ecifica31y Mihis section'812'asalysl^usiag ,
^-models"designed to simidate conditions sin the absence of tfie <3A A. 'These numbers shonittnot be mtarpusled as aotoal^istorical
'
^otals^roay ^ffep'sSgtitly ftwa saias'durto ri»ri
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table B-20. CO Emissions Und^eFttiCC^ntrol and"Mo-cotJttor§eeaarios by Tamet;Ye'af;£iti tnoasands
ofshorttons), " —^-'^ • s '".'1T7. ', '*" **• v- "'' '„.:;': ""-";
Sector
Transportation:
Highway Vehicles
Off-Highway Vehicles
Electric Utilities
Industrial Processes
Industrial Boilers
Commercial/Residential
TOTAL*
Wkhthc
1975 19SO
83,580 79,9,70, j,
- 8,510 ' 8.100 '
- ? *"'>'• *<.<"•
' ;£*** *"• v
.,^/fc.A - •..{. •/ V"^ "***
240 ' ^ j, 28$^
7,580^^6^^
>*/Jrt , „„,. , ?vx 't'f rttX^
/Z-w3 v* ., 'j /lv
10,230 !l333tf\
110,880 109, HO
, ,,«^,!
.CAA - »-.
- ^1985-,- -
'*'>*&&>$'' S"*
•^2jt§ft
7*880
s-WV/1 ,-•
wp-
f ,,- 290
^-,,4,840
*>J 670,
; 14,140- -
100,300- „
x--%
4 •&&'"
' , '/*
65,430
- 8,080""
* *->,- >t =,-•?,
,^,,,J,^vv^,
370 s
^ifcf.
,« 7>tO
43,150
x92,9,00
-< ?•">'.-.,--
»"• % „,-
"'"''iW*'
•" 90,460'
"VSjSMJ-
'^ > J
'250"
9,240"
720 '
, 10^2^0,,
119,430s*
' WHhott
' 1980
105,530,
8,070
*" ' '290
9,120 ,
710-,,
,,,-13,170-
' '136,880'
the CAA
1985 <
f f vtffl
131,^20^
, ,,7,880,
'/
,,300
,' '8,860"
/,-,-,, -,620
-- 14,200
163,280,,;,
'* ' " DiBSaceate
, , „ , , ' *1a K8W ',
' " ' 199(11 Emissions
,77" ----- 'T
,449,280 '(56%) ,„
4,,,??,?F;,,;, „ ,°*'";
-
•*', * J»y >S ,,#'ifx'"y '& *<•, ~~,
380 - (3%) '
' 10,!80 '(4'9%y"-
?4o "6si"""
13,210 , ,, W""
v;18t,,S6Q,.;, m°!$ '
- Theestiraates of emission levels w'z*1 aa*l''li''tftoittJtieCAA vsreje^^*s-^v?;V 1 1
1 , >- -;,;;;",:)1|>?^)-1-, 0 '"'o'
190 s- - s --90' -' " --- -> 23 '", ' '"*'"*3 ''
- ,„„,,. ,^,v.,^, ,
, Without the CAA' •-
- -1975,,,-, 1980 '^lilS "' '
7' "203 -.-'-207-,-- -- -214
•i-".,-,«v-/,r ' «
J '-> ' ' *7
217 „,,„-,, ,,221,,,,,,, ,-228
,,,-, »,,,,J-i»,,,,,^ „ ,,,,
Difftieace
,, ~ ,-, ; -,-;<„ ,^J9»,
15*Vw ' J jBJwHUwflwwas
223 (99%)""
237 ' ' *-(9-9%5
••/ ,,„,„,,,, ,
',, - ', , ,
Notre; The estimates of emission levels wftft andvtifluna the 'QiA'^ejfe!^ev«loped specifically for'thts section 812 analysis using '-,
models designed to simulate conditions in ths-absence.qfsliie C^A- These numbers should-not be inter|)reted as actual- historical
emission estimates. - - - < '--xxv-j;,, " -«--•-•--•-
*T«als may differ slightly from sums due-to r^umftng,.; /v,
B-38
-------
Appendix E: Emissions Modeling
Emissions Modeling References
Abt Associates Inc. (Abt). 1995. The Impact of the
Clean Air Act on Lead Pollution: Emissions
Reductions, Health Effects, and Economic
Benefits from 1970 to 1990, Final Report,
Bethesda, MD, October.
Argonne National Laboratory (ANL). 1990. Current
Emission Trends for Nitrogen Oxides, Sulfur
Dioxide, and Volatile Organic Chemicals by
Month and State: Methodology and Results,
Argonne, IL, August.
Argonne National Laboratory (ANL). 1992. Retro-
spective Clean Air Act Analysis: Sectoral
Impact on Emissions from 1975 to 1990,
(Draft), Argonne, IL, July.
Braine, Bruce and P. Kim. 1993. Fuel Consumption
and Emission Estimates by State, ICF Re-
sources, Inc., Fairfax, VA, memorandum to
Jim DeMocker, EPA. April 21.
Braine, Bruce, S. Kohli, and P. Kim. 1993.7975 Emis-
sion Estimates with and without the Clean Air
Act, ICF Resources, Inc., Fairfax, VA, memo-
randum to Jim DeMocker, EPA, April 15.
DeMocker, J. Personal Communication with Office
of Mobile Sources Staff, Ann Arbor, Michi-
gan. Date unknown.
Energy Information Administration (EIA). 1982. Es-
timates of U.S. Wood Energy Consumption
from 1949 to 1981. DOE/EIA-0341, U.S.
Department of Energy. August.
Energy Information Administration (EIA). 1985. Cost
and Quality of Fuels for Electric Utility
Plants. DOE/EIA-0091(85), U.S. Department
of Energy.
Energy Information Administration (EIA). 1989.Non-
residential Buildings Energy Consumption
Survey: Commercial Buildings Consumption
and Expenditures 1986. DOE/EIA-0318(86),
U.S. Department of Energy. May.
Energy Information Administration (EIA). 1990. Es-
timates of U.S. Biofuels Consumption 1990.
DOE/EIA-0548(90), U.S. Department of En-
ergy. October.
Electric Power Research Institute (EPRI). 1981. The
EPRI Regional Systems, EPRI-P-1950-SR,
Palo Alto, CA.
Federal Highway Administration (FHWA). 1986.
1983-1984 Nationwide Personal Transporta-
tion Survey, U.S. Department of Transporta-
tion, Washington, DC.
Federal Highway Administration (FHWA). 1988.
Highway Statistics 1987, PB89-127369, U.S.
Department of Transportation, Washington,
DC.
Federal Highway Administration (FHWA). 1992.
Highway Statistics 1991, FHWA-PL-92-025,
U.S. Department of Transportation, Washing-
ton, DC.
Gschwandtner, Gerhard. 1989. Procedures Document
for the Development of National Air Pollut-
ant Emissions Trends Report, E.H. Pechan &
Associates, Inc., Durham, NC. December.
Hogan, Tim. 1988. Industrial Combustion Emissions
Model (Version 6.0) Users Manual, U.S. En-
vironmental Protection Agency, EPA-600/8-
88-007a.
ICF Resources, Inc. 1992. Results of Retrospective
Electric Utility Clean Air Act Analysis -1980,
1985 and 1990, September 30.
Jorgenson, D.W. and P. Wilcoxen. 1989. Environmen-
tal Regulation and U.S. Economic Growth,
Harvard University Press, Cambridge, MA.
Klinger, D. and J.R. Kuzmyak. 1986. Personal Travel
in the United States, Vol. 1:1983-84 Nation-
wide Personal Transportation Study, U.S.
Department of Transportation, Federal High-
way Administration, Washington, DC. Au-
gust.
Kohout, Ed. 1990. Current Emission Trends for Ni-
trogen Oxides, Sulfur Dioxide, and Volatile
Organic Compounds by Month and State:
Methodology and Results," Argonne National
Laboratory, ANL/EAIS/TM-25, Argonne, IL.
B-39
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Lockhart, Jim. 1992. Projecting with and without
Clean Air Act Emissions for the Section 812
Retrospective Analysis: A Methodology Based
Upon the Projection System used in the Of-
fice of Air Quality Planning and Standards
National Air Pollutant Emission Estimate
Reports" (Draft Report), Environmental Law
Institute, November 16.
McDonald, J.F. and D.W. South. 1984. The Commer-
cial and Residential Energy Use and Emis-
sions Simulation System (CRESS): Selection
Process, Structure, and Capabilities, Argonne
National Laboratory, ANL/EAIS/TM-12,
Argonne, IL. October.
Mintz, M.M. and A.D. Vyas. 1991. Forecast of Trans-
portation Energy Demand through the Year
2010, Argonne National Laboratory, ANL/
ESD-9, Argonne, IL. April.
Pechan Associates. 1995.The Impact of the Clean Air
Act on 1970 to 1990 Emissions; Section 812
Retrospective Analysis. Draft Report. March.
Saricks, C.L. 1985. The Transportation Energy and
Emissions Modeling System (TEEMS): Selec-
tion Process, Structure, and Capabilities,
Argonne National Laboratory, ANL/EES-
TM-295, Argonne, IL. November.
Veselka, T.D., et al. 1990. Introduction to the Argonne
Utility Simulation (ARGUS) Model," Argonne
National Laboratory, ANL/EAIS/TM-10,
Argonne, EL. March.
Vyas, A.D. and C.L. Saricks. 1986. Implementation
of the Transportation Energy and Emissions
Modeling System (TEEMS) in Forecasting
Transportation Source Emissions for the 1986
Assessment, Argonne National Laboratory,
ANL/EES-TM-321, Argonne, IL. October.
U.S. Department of Commerce (DOC). 1975. Statis-
tical Abstract of the United States: 1975 (96th
Edition), Bureau of the Census, Washington,
DC, September.
U.S. Department of Commerce (DOC). 1977. Statis-
tical Abstract of the United States: 1977 (98th
Edition), Bureau of the Census, Washington,
DC, September.
U.S. Department of Commerce (DOC). 1981. 7977
Truck Inventory and Use Survey, Bureau of
the Census, TC-77-T-52, Washington, DC,
August.
U.S. Department of Commerce (DOC). 1982. Statis-
tical Abstract of the United State's: 1982-1983
(103rd Edition), Bureau of the Census, Wash-
ington, DC, December.
U.S. Department of Commerce (DOC). 1983. Statis-
tical Abstract of the United States: 1984
(104th Edition), Bureau of the Census, Wash-
ington, DC, December.
U.S. Department of Commerce (DOC). 1984. 1982
Truck Inventory and Use Survey, Bureau of
the Census, TC-82-T-52, Washington, DC,
August.
U.S. Department of Commerce (DOC). 1987. Statis-
tical Abstract of the United States: 1988
(108th Edition), Burea of the Census, Wash-
ington, DC, December.
U.S. Department of Commerce (DOC). 1990. 1987
Truck Inventory and Use Survey," Bureau of
the Census, TC87-T-52, Washington, DC,
August.
U.S. Department of Commerce (DOC). 1991.Annwa/
Survey of Manufactures: Purchased Fuels and
Electric Energy Used for Heat and Power by
Industry Group, Bureau of the Census,
M87(AS)-1, Washington, DC.
U.S. Department of Commerce (DOC). 1993. Statis-
tical Abstract of the United States: 1993
(113th Edition)," Bureau of the Census,
Washington, DC.
U.S. Department of Energy (DOE). 1982. Documen-
tation of the Resource Allocation and Mine
Costing (RAMC) Model. DOE/NBB-0200.
Energy Information Administration.
U.S. Department of Energy (DOE). 1986. Inventory
of Power Plants in the United States 1985.
DOE/EIA-0095(85), Energy Information
Administration, Washington, DC, August.
B-40
-------
Appendix B: Emissions Modeling
U.S. Department of Energy (DOE). 1988. An Analy-
sis of Nuclear Power Plant Operating Costs.
DOE/EIA-0511(88), Energy Information
Administration.
U.S. Department of Energy (DOE). 1990. State En-
ergy Price and Expenditure Report 1988.
DOE/EIA-0376(88), Energy Information
Administration, Washington, DC, September.
U.S. Department of Energy (DOE). 1991. State En-
ergy Data Report: Consumption Estimates -
1960-1989. DOE/EIA-0214(89), Energy In-
formation Administration, Washington, DC,
May.
U.S. Department of Energy (DOE). 1992. Annual
Energy Review 1991. DOE/EIA-0384(91),
Energy Information Administration, Wash-
ington, DC.
U.S. Environmental Protection Agency (EPA). 1985.
Compilation of Air Pollutant Emission Fac-
tors, Volume I: Stationary Point and Area
Sources, AP-42, Fourth Edition, GPO No.
055-000-00251-7, Research Triangle Park,
NC. September.
U.S. Environmental Protection Agency (EPA). 1989.
The 1985 NAPAP Emissions Inventory, EPA-
600/7-89-012a, Research Triangle Park, NC.
November.
U.S. Environmental Protection Agency (EPA). 1990.
The Cost of a Clean Environment, EPA-230-
11-90-083. November.
U.S. Environmental Protection Agency (EPA). 1991.
Office of Air Quality Planning and Standards,
National Air Pollutant Emissions Estimates,
1940-1990, EPA-450/4-91-026, Research
Triangle Park, NC. November.
U.S. Environmental Protection Agency (EPA). 1992.
7990 Toxics Release Inventory, EPA-700-S-
92-002, Washington, DC.
U.S. Environmental Protection Agency (EPA). 1994a.
National Air Pollutant Emission Trends,
1900-1993, EPA-454/R-94-027, Office of Air
Quality Planning and Standards, Research
Triangle Park, NC. October.
U.S. Environmental Protection Agency (EPA). 1994b.
Office of Mobile Sources, User's Guide to
MOBILES (Mobile Source Emission Factor
Model), EPA-AA-AQAB-94-01, Ann Arbor,
MI. May.
U.S. Environmental Protection Agency (EPA). 1995.
National Air Pollutant Emission Trends 1900-
1994, EPA-454/R-95-011. Office of Air Qual-
ity Planning and Standards, Research Triangle
Park, NC. October.
Werbos, Paul J. 1983. A Statistical Analysis of What
Drives Energy Demand: Volume HI of the
PURHAPS Model Documentation, U.S. De-
partment of Energy, Energy Information Ad-
ministration, DOE/EIA-0420/3, Washington,
DC.
B-41
-------
-------
Appendix C: Air Quality Modeling
Introduction
This appendix describes in greater detail the vari-
ous methodologies used to translate differences in
control and no-control scenario emission estimates
into changes in air quality conditions. Summary char-
acterizations of the results of the air quality modeling
efforts for 1990 are provided here and in the main
text. Further details and discussion of key analytical
and modeling issues can be found in a number of sup-
porting documents. These documents, which provide
the analytical basis for the results presented herein,
are:
* ICF Kaiser/Systems Applications Interna-
tional, "Retrospective Analysis of Ozone Air
Quality in the United States", Final Report,
May 1995. (Hereafter referred to as "SAI
Ozone Report (1995).")
* ICF Kaiser/Systems Applications Interna-
tional, "Retrospective Analysis of Paniculate
Matter Air Quality in the United States", Draft
Report, September 1992. (Hereafter referred
to as "SAI PM Report (1992).")
* ICF Kaiser/Systems Applications Interna-
tional, "Retrospective Analysis of Paniculate
Matter Air Quality in the United States", Fi-
nal Report, April 1995. (Hereafter referred to
as "SAI PM Report (1995).")
* ICF Kaiser/Systems Applications Interna-
tional, "PM Interpolation Methodology for
the section-812 retrospective analysis",
Memorandum from J. Langstaff to J.
DeMocker, March 1996. (Hereafter referred
to as "SAI PM Interpolation Memo (1996).")
* ICF Kaiser/Systems Applications Interna-
tional, "Retrospective Analysis of SOy NOX
and CO Air Quality in the United States",
Final Report, November 1994. (Hereafter re-
ferred to as
(1994).")
'SAI SO , NO and CO Report
* ICF Kaiser/Systems Applications Interna-
tional, "Retrospective Analysis of the Impact
of the Clean Air Act on Urban Visibility in
the Southwestern United States", Final Re-
port, October 1994. (Hereafter referred to as
"SAI SW Visibility Report (1994).")
« Dennis, Robin L., US EPA, ORD/NERL,
"Estimation of Regional Air Quality and
Deposition Changes Under Alternative 812
Emissions Scenarios Predicted by the Re-
gional Acid Deposition Model, RADM", Draft
Report, October 1995. (Hereafter referred to
as "RADM Report (1995).")
The remainder of this appendix describes, for each
pollutant or air quality effect of concern, (a) the basis
for development of the control scenario air quality
profiles; (b) the air quality mbdeling approach used
to estimate differences in air quality outcomes for the
control and no-control scenario and the application of
those results to the derivation of the no-control sce-
nario air quality profiles; (c) the key assumptions,
caveats, analytical issues, and limitations associated
with the modeling approach used; and (d) a summary
characterization of the differences in estimated air
quality outcomes for the control and no-control sce-
narios.
Carbon Monoxide
Control scenario carbon monoxide
profiles
As described in the preceding general methodol-
ogy section, the starting point for development of con-
trol scenario air quality profiles was EPA's AIRS da-
C-l
-------
The Benefits and Costs of the Clean Air Act, 1970 to 2990
Table C-1. Summary of CO Monitoring Oata,
Year
1970
1975
1980
1985
1990
Number of
Monitors
82
503
522
472
5Q6
Number of
Counties
54
246 <
250
232
244
.Percent^
Population
*£%?$}'.£&$i * -V* /
Covered"
\<-?"
n/a .,2
C *s"M%\
n/a ,,J^
.,* 5Q%"^r
.tftyv ^ s'
n/a-;>
55,%
** -* / x,.
Number of
ti">-" 1
Samples
/• ^
*%40$,524 "
'' /
' 2,667,525
, :M>Sl399
>3iS$$,286 -
Jst^SS^S. r
Meaii
Number of
Samples per
' Monitor
4982 ,
' '"$,303
**-, 5,846'
y , /
7,486
7,486
Data Source: SAI SO,, NO, and CO Report (1994).
tabase. Hourly CO air quality monitoring data were
compiled for all monitors in the 48 contiguous states
for the study target years of 1970,1975,1980,1985,
and 1990. Although the CO monitoring network was
sparse in 1970, by 1990 506 monitors in 244 counties
provided monitoring coverage for 55 percent of the
population in the conterminous U.S. Table C-l sum-
marizes the CO monitoring data derived from AIRS.
Additional data regarding the EPA Region location,
land use category, location-setting category, and ob-
jective category of the monitors providing these data
are described in the SAI SO2, NOx, and CO Report
(1994).
The next step hi constructing the control scenario
air quality profiles was to calculate moving averages,
for a variety of time periods, of the hourly CO data
for each monitor. For CO, moving averages of 1,3,5,
7, 8, 12, and 24 hours were calculated. Daily maxi-
mum concentrations observed at each monitor for each
of these averaging periods were then calculated. Fi-
nally, profiles were developed to reflect the average
and maximum concentrations for each of the seven
averaging periods. However, profiles were only de-
veloped for a given monitor when at least 10 percent
of its theoretically available samples were actually
available. The purpose of applying this cutoff was to
avoid inclusion of monitors for which available sample
sizes were too small to provide a reliable indication
of historical air quality.
As discussed in the air quality modeling chapter
of the main text, development of representative dis-
tributions for these profiles was then necessary to pro-
vide a manageable char-
acterization of air qual-
ity conditions. Initially,
two-parameter lognor-
mal distributions were
fitted to the profiles
based on substantial evi-
dence that such distribu-
tions are appropriate for
modeling air quality
data. However, given
the relative importance
of accurately modeling
higher percentile obser-
vations (i.e., 90th per-
centile and higher), a
ly •.' — '„ three-parameter model-
millim^^^^^^^^^^^^m ing approach was used
to isolate the effect of
observations equal, or very close, to zero. In this ap-
proach one parameter defines the proportion of data
below a cutoff close to zero and the remaining two
parameters describe the distribution of data above the
cutoff value. Several other studies have already dem-
onstrated good fit to air quality modeling data with a
three-parameter gamma distribution, and both lognor-
mal and gamma distributions using a three-parameter
approach were developed for the present study. As
documented in the SAI SO2, NOx, and CO Report
(1994), a cutoff of 0.05 ppm was applied and both the
three-parameter lognormal and three-parameter
gamma distributions provided a good fit to the em-
pirical data. For CO, the gamma distribution provided
the best fit.
The control scenario air quality profiles are avail-
able on diskette. The filename for the CO Control
Scenario profile database is COCAA.DAT, and adopts
the format presented in Table C-2.
No-control scenario carbon monoxide
profiles
To derive comparably configured profiles repre-
senting CO air quality in the no-control scenario, con-
trol scenario profile means and variances were ad-
justed in proportion to the difference in emissions es-
timated under the two scenarios. Specifically, for all
control scenario air quality observations predicted by
the three-parameter distributions falling above the
"near-zero" cutoff level, comparable no-control esti-
mates were derived by the following equation:
C-2
-------
Appendix C: Air Quality Modeling
tfraat of Air Quality Profile; Databases.,,,
-. ££•'. - . .\ , /-rtfi-i ^X , ' ' ' >'s *.,<#*>»*>, A* ^ -f ; ,,-. '^-ix
CoIUHHlS
Description '' ' ''',
Year (10,75, SO^S^O).
Integra '
Avwalrag flifi,6tlv-3; 5, "T, 8; 12, 24-faours)" '
State n?iS,cpde
,11-13
- Integar '-' -"''
. Cooity FlPlcode -
* 21-30
Real
-Latitude
, -.0 \»t..>#
"32 -'41
43-44™
Hourly int&inittency parameter^*:;/'
, ,, • s
o. Hourly -logaoajial paraitietw it,
Hourlylogaorraal parame,t£sc cr"
> 76 -
Hourly gamma parameter
V, , '---V •-, „
v Hourly gamma parameter ,p
E»aUy,Baa^tognonnaJ parameter ,n*/,i
PMJym«xlog»oiti>al' parameter ',:„
;'feiily'inaK
',, 136- 145
2 =jj*atttud«f1o»gitiide values from collocated moaftor or previous monitor
--/v-f*/- locariotfimomtor paramet&r'ocoarrence ccxJIs 1)
-9 = Mtade/loagitnae misslng'tc^tlnty center substituted) - •- -, -
SQ2fNOs and CO Rsport :{1994).
'"'''•'• *• .*• J ^ j !• . -
where
Z^ = air quality measurement for the no-control scenario,
Xc = air quality measurement for the control scenario,
ENC = emissions estimated for the no-control scenario,
Ec = emissions estimated for the control scenario, and
b = background concentration.
C-3
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
The adjustment for background concentration is
made to hold ambient background concentrations of
the pollutant constant between the control and
no-control scenarios. To the extent background con-
centrations are affected by transport of anthropogenic
pollutants from upwind sites, and to the extent up-
wind emissions may have been controlled under the
control scenario, assuming a fixed background con-
centration represents a conservative assumption in this
analysis. As discussed in the SAISO2, NOx, and CO
Report (1994), the CO background concentration used
for this analysis was 0.2 ppm, which equals the low-
est typical concentration observed in the lower 48
states.
In the SAI SO2, NOx, and CO Report (1994) docu-
menting the CO air quality modeling effort, reference
is made to using county-level emission estimates as
the basis for deriving the no-control profiles. Deriva-
tion of these county-level results is described hi more
detail in the appendix on emissions estimation. It is
important to emphasize here, however, that the county-
level CO emissions data were derived for both the
control and no-control scenarios by simple popula-
tion-weighted disaggregation of state-level emission
totals. Although CO emission estimates were needed
at the county level to support the ozone ah" quality
modeling effort, differences in state-level emissions
estimates are what drive the difference in the control
and no-control air quality profiles for CO. In other
words, the ENCAA to £GW ratios used to derive the
no-control profiles according to Equation (1) above
are essentially based on state-level emissions estimates
for CO.
As for the control scenario air quality profiles,
the no-control scenario air quality profiles are avail-
able on diskette. The filename for the CO No-control
Scenario profile database is CONCAA.DAT. The
same data format described hi Table C-2 is adopted.
Summary differences in carbon
monoxide air quality
While the control and no-control scenario air qual-
ity profiles are too extensive to present hi their en-
tirety in this report, a summary indication of the dif-
ference in control and no-control scenario CO con-
centrations is useful. Figure C-l provides this sum-
mary characterization. Specifically, the air quality
indicator provided is the 95th percentile observation
of 1990 CO concentrations averaged over a 1-hour
period. The graph shows the number of monitors for
Figure C-l. Frequency Distribution of Estimated Ratios
for 1990 Control to No-control Scenario 95th Percentile
1-Hour Average CO Concentrations, by Monitor.
300
0.05 0.25 0.45 0.65
0.85 1.05 1.25
Ratio of CAA:No-CAA 95th Percentile 1-Hour Average
which the ratio of 1990 control to no-control scenario
95th percentile 1-hour average concentrations falls
within a particular range. The x-axis values in the
graph represent the midpoint of each bin. The results
indicate that, by 1990, CO concentrations under a no-
control scenario would have been dramatically higher
than control scenario concentrations.
Key caveats and uncertainties for
carbon monoxide
A number of important uncertainties should be
noted regarding the CO air quality estimates used in
this analysis. First and foremost, CO is a highly local-
ized, "hot spot" pollutant. As such, CO monitors are
often located near heavily-used highways and inter-
sections to capture the peak concentrations associated
with mobile sources. Since this analysis relies on state-
level aggregate changes in CO emissions from all
sources, the representativeness and accuracy of the
predicted CO air quality changes are uncertain. There
is no basis, however, for assuming any systematic bias
which would lead to over- or under-estimation of air
quality conditions due to reliance on state-wide emis-
sion estimates.
A second source of uncertainty is the extent to
which the three-parameter distributions adequately
characterize air quality indicators of concern. Appen-
dix C of the SAI SO2, NOx, and CO Report (1994)
presents a number of graphs comparing the fitted ver-
sus empirical data for one-hour and 12-hour averag-
ing periods. In the case of CO, the gamma distribu-
tion appears to provide a very reasonable fit, though
clearly some uncertainty remains.
C-4
-------
Appendix C; Air Quality Modeling
Finally, a central
premise of this analy-
sis is that changes in
CO emissions should
be well-correlated
with changes in CO
air quality. Strong
correlation between
the state-level emis-
sions estimates used
in this analysis and
empirical air quality
measurements would
not be expected due to
inconsistencies be-
tween the state-level
scale of modeled
emissions versus the
s , frf v
* "V >. V"- '
,,•.
.1970
,'jf NiHttberol
-' Monitors.
Counties
. 340
, ' Fercea't „,',
iPoptitetion
' 6,6015' ";
^
-X' 'Meaa-Nunvber, ^
' ''
monitor-level scale of the air quality data, and between
the modeled control scenario emissions inventories
and actual historical air quality measurements. Under
these circumstances, it is particularly important to
focus on the primary objective of the current analy-
sis, which is to estimate the difference in air quality
outcomes between scenarios which assume the ab-
sence or presence of historical air pollution controls.
In the process of taking differences, some of the un-
certainties are expected to cancel out. No attempt is
made in the overall analysis to predict historical air
quality, or hypothetical air quality in the absence of
the Clean Air Act, in absolute terms.
Sulfur Dioxide
Sulfur dioxide (SO2) emissions lead to several air
quality effects, including secondary formation of fine
particle sulfates, long range transport and deposition
of sulfuric acid, and localized concentrations of gas-
eous sulfur dioxide. The first two effects are addressed
later in this appendix, under the particulate matter and
acid deposition sections. The focus of this section is
estimation of changes in local concentrations of sul-
fur dioxide.
The methodology applied to estimation of local
sulfur dioxide air quality is essentially identical to the
one applied for carbon monoxide. As such, this sec-
tion does not repeat the "roll-up" modeling method-
ological description presented in the CO section, but
instead simply highlights those elements of the sulfur
dioxide modeling which differ from carbon monox-
ide.
Control scenario sulfur dioxide profiles
Unlike the CO monitoring network, the number
of monitors as well as the population coverage of the
SO2 monitoring network shrank during the 1980's.
Table C-3 summarizes the SO2 monitoring data used
as the basis for development of the control scenario
air quality profiles.
As for CO, air quality profiles reflecting average
values and daily maxima for 1, 3, 5, 7, 8, 12, and 24
hour averages were compiled from AIRS for moni-
tors in the lower 48 states which had at least 10 per-
cent of their potential samples available. Applying a
cutoff of 0.1 ppb to isolate the zero and near-zero ob-
servations, three-parameter lognormal and gamma
distributions were fitted to these empirical profiles.
In the case of SO2, the three-parameter lognormal dis-
tribution was found to provide the best fit.
The control scenario SO2 air quality profiles are
available on diskette, contained in a file named
SO2CAA.DAT. The same data format described in
Table C-2 is adopted.
No-control scenario sulfur dioxide
profiles
The no-control air quality profiles for SO2 are
derived using Equation 1, the same equation used for
CO. For SO2, the background concentration was as-
sumed to be zero. Although anthropogenic emissions
contribute only small amounts to total global atmo-
spheric sulfur, measured background concentrations
C-5
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
for the continental U.S. range from only 0.1 to 1.3
ppb. Background SO2 is discussed in more detail in
the supporting document SAISO2, NOx, and CO Re-
port (1994).1
The no-control scenario SO2 air quality profiles
are available on diskette, contained in a file named
SO2NCAA.DAT. The data format is described in
Table C-2.
Summary differences in sulfur dioxide
air quality
As for CO, reporting differences in control and
no-control scenario air quality projections for each
monitor covered in the analysis is impractical due to
the large amount of data involved. However, Figure
C-2 provides an illustration of scenario differences
similar to the one provided for CO. Specifically, the
graph shows the distribution of 1990 control to no-
control scenario 95th percentile 1-hour average con-
centrations ratios at SO2 monitors. By 1990, SO2 con-
centrations under the no-control scenario were sub-
stantially higher than those associated with the con-
trol scenario.
Figure C-2. Frequency Distribution of Estimated Ratios
for 1990 Control to No-control Scenario 95th Percentile
1-Hour Average SO2 Concentrations, by Monitor.
300
•5'200
I
100
0.05 0.25 0.45 0.65 0.85 . 1.05 1.25
Ratio of CAA:No-CAA 95th Percentile 1-Hour Average
Key caveats and uncertainties for sulfur
dioxide
The'height of stacks used to vent flue gases from
utility and industrial fossil fuel-fired boilers has a sig-
nificant effect on the dispersion of sulfur dioxide and
on the formation and long-range transport of second-
ary products such as particulate sulfates. Under a no-
control scenario, it is conceivable that some sources
might have built taller stacks to allow higher emis-
sion rates without creating extremely high ground-
level concentrations of flue gases. On the other hand,
it is also conceivable that, in the absence of post-1970
air pollution control programs, sources might have
built shorter stacks to avoid incurring the higher costs
associated with building and maintaining taller stacks.
To the extent facilities would have adopted different
stack height configurations under a no-control sce-
nario, both local exposures to sulfur dioxides (and
other emissions from fossil fuel combustion) and long-
range transport, deposition, and exposure associated
with secondary formation products may have been
different. However, this analysis assumes that both
the location of individual facilities and the height and
configuration of emission stacks are constant between
the two scenarios. If, in fact, stack heights were raised
under the historical case due to CAA-related concerns,
increases in local SO2 concentrations under the
no-control scenario may be overestimated. However,
this same assumption may at the same time lead to
underestimation under the no-control scenario of long-
range transport and formation of secondary particu-
lates associated with taller stacks. For stacks built
lower under a no-control scenario, local SO2 expo-
sures would have been higher and long-range effects
lower. Finally, the comments on uncertainties for car-
bon monoxide apply as well to SO2.
Nitrogen Oxides
Similarly to sulfur dioxide, emissions of nitro-
gen oxides (NOx) -including nitrogen dioxide (NO2)
and nitrous oxide (NO)- lead to several air quality
effects. These effects include secondary formation of
fine particle nitrates, formation of ground-level ozone,
long range transport and deposition of nitric acid, and
localized concentrations of both NO2 and NO. The
first three effects are addressed later in this appen-
dix, under the particulate matter, ozone, and acid
deposition sections. The focus of this section is esti-
mation of changes in local concentrations of NO2 and
NO.
The methodology applied to estimation of local
nitrogen oxides air quality is essentially identical to
the one applied for carbon monoxide and sulfur diox-
ide. As such, this section does not repeat the "roll-up"
modeling methodological description presented in the
CO section, but instead simply highlights those ele-
1 SAI SO2, NOx, and CO Report (1994), page 4-9.
-------
Appendix C: Air Quality Modeling
ments of the nitrogen oxides modeling which differ
from carbon monoxide.
Control scenario nitrogen oxides
profiles
After peaking around 1980, the number of NO2
and NO monitors, their county coverage, and their
population coverage shrank between 1980 and 1990.
Tables C-4 and C-5 summarize, respectively, the NO2
and NO monitoring data used as the basis for devel-
opment of the control scenario air quality profiles.
As for CO and SO2, air quality profiles reflecting
average values and maxima for 1, 3, 5, 7, 8, 12, and
24 hour NO2 and NO averages were compiled from
AIRS for monitors in the lower 48 states which had at
least 10 percent of their potential samples available.
Applying a cutoff of 0.5 ppb to both NO2 and NO to
isolate the zero and near-zero observations, three-pa-
rameter lognormal and gamma distributions were fit-
ted to these empirical profiles. For NO2 and NO, the
three-parameter gamma distribution was found to pro-
vide the best fit.
The control scenario NO2 and NO air quality pro-
files are available on diskette, contained in files named
NO2CAA.DAT andNOCAA.DAT, respectively. The
same data format described in Table C-2 is adopted.
<"'->-':'~ ttt'ftttff/'t'&Z'rrs-Js'
PerceiJI^.
Population
Cowifed
,
), Number of
Monitor
:•»
; 1,9^128 ',
s
, 'Samples
Moiritor
;i!, 4,101,051
W85
, -' 956,425™", ;
C-7
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
No-control scenario nitrogen oxides
profiles
The no-control air quality profiles for NO2 and
NO are derived using Equation 1, the same equation
used for CO and SO2. As discussed in detail in the
SAISO2, NOX, and CO Report (1994),2 nitrogen ox-
ides are emitted almost entirely from anthropogenic
sources and they do not have long atmospheric resi-
dence times. Therefore, global background concen-
trations are very low, on the order of 0.1 or 0.2 ppb.
For the present analysis, background concentrations
of NO2 and NO were assumed to be zero.
The no-control scenario NO2 and NO air quality
profiles are available on diskette, contained hi files
named NO2NCAA.DAT andNONCAA.DAT, respec-
tively. The data format is described in Table C-2.
Summary differences in nitrogen oxides
air quality
Figure C-3 provides a summary indication of the
differences in control and no-control scenario air qual-
ity for NO2. As for CO and SO2, the graph shows the
distribution of 1990 control to no-control scenario 95th
percentile 1-hour average concentration ratios at NO2
monitors. These ratios indicate that, by 1990, no-con-
trol scenario NO2 concentrations were significantly
higher than they were under the control scenario. The
changes for NO are similar to those for NO2.
Figure C-3. Frequency Distribution of Estimated Ratios
for 1990 Control to No-control Scenario 95th Percentile
1-Hour Average NO2 Concentrations, by Monitor.
300
oL
0.05 0.25 0.45 0.65 0.85 1.05 1.25
Ratio of CAA:No-CAA 95th Percentile 1-Hour Average
Key caveats and uncertainties for
nitrogen oxides
A number of caveats and uncertainties specific to
modeling NOx should be noted. First, stack height and
stack height control strategies likely to have influenced
•local concentrations of SO2 may also have influenced
local concentrations of NO2 and NO. (For a fuller dis-
cussion of the stack heights issue, refer to the section
"Key caveats and uncertainties for SO2") In addition,
the earlier discussion of uncertainties resulting from
the use of state-level emissions and the cancellation
of uncertainties resulting from analyzing only differ-
ences or relative changes also applies to NOx.
Acid Deposition
The focus of air quality modeling efforts described
above for carbon monoxide, sulfur dioxide, and ni-
trogen oxides was to estimate the change in ambient
concentrations of those pollutants as a result of
changes in emissions. Particularly since the emissions
modeling was driven by modeled macroeconomic
conditions, rather than actual historical economic ac-
tivity patterns, neither the emissions inventories nor
the resultant air quality conditions developed for this
analysis would be expected to match historical out-
comes. The need to focus on relative changes, rather
than absolute predictions, becomes even more acute
for estimating air quality outcomes for pollutants sub-
ject to long-range transport, chemical transformation,
and atmospheric deposition. The complexity of the
relationships between emissions, air concentrations,
and deposition is well-described hi the following para-
graph from the RADM report document developed
by Robin Dennis of US EPA's National Exposure
Research Laboratory in support of the present analy-
sis:
"Sulfur, nitrogen, and oxidant species in the
atmosphere can be transported hundreds to
thousands of kilometers by meteorological
forces. During transport the primary
emissions, SOf NO^, and volatile organic
emissions (VOC) are oxidized in the air or in
cloud-water to form new, secondary
compounds, which are acidic, particularly
sulfate and nitric acid, or which add to or
subtract from the ambient levels ofoxidants,
such as ozone. The oxidizers, such as the
hydroxyl radical, hydrogen peroxide and
1 SAI SO2> NOX, and CO Report (1994), page 4-9.
-------
Appendix C: Air Quality Modeling
ozone are produced by reactions ofVOC and
NOx. The sulfur and nitrogen pollutants are
deposited to the earth through either wet or
dry deposition creating a load of pollutants
to the earth's surface... However, the
atmosphere is partly cleansed of oxidants
through a number of physical processes
including deposition (e.g., ozone is removed
by wet and dry deposition). Dry deposition
occurs when particles settle out of the air onto
the earth or when gaseous or fine particle
species directly impact land, plants, or water
or when plant stomata take up gaseous
species, such as SO2. In wet deposition,
pollutants are removed from the atmosphere
by either rain or snow. In addition, fine
particles or secondary aerosols formed by the
gas- and aqueous-phase transformation
processes scatter or absorb visible light and
thus contribute to impairment of visibility. "3
Control scenario acid deposition
profiles
The derivation of control scenario emission in-
ventory inputs to the RADM model is succinctly de-
scribed in this excerpt from the RADM Report (1995):
The RADM model requires a very detailed
emissions inventory in both time and space.
The emissions fields are also day-specific to
account for the temperature effects on the
volatile organics and the wind and
temperature effects on the plume rise of the
major point sources. At the time of the 812
retrospective study RADM runs, these
inventories had been developed for 1985,
using the 1985 NAPAP (National Acid
Precipitation Assessment Program)
inventory, and adjusted for point source
The complexity and nonlinearity
of the relationships between localized
emissions of precursors, such as SO2
and VOC, and subsequent regional
scale air quality and deposition effects
are so substantial that the simple "roll-
up" modeling methodology used for
estimating local ambient concentra-
tions of SO2, NOx, and CO is inad-
equate, even for a broad-scale, aggre-
gate assessment such as the present
study. For sulfur deposition, and for
a number of other effects addressed
in subsequent sections of this appen-
dix, a regional air quality model was
required. After careful review of the
capabilities, geographic coverage,
computing intensity, and resource re-
quirements associated with available
regional air quality models, EPA de-
cided to use various forms of the Re-
gional Acid Deposition Model
(RADM) to estimate these effects.4
Figure C-4 shows the geographic do-
main of the RADM.
Figure C-4. Location of the High Resolution RADM 20-km Grid Nested
Inside the 80-km RADM Domain.
3 Dennis, R. RADM Report (1995), p. 1.
4 For a detailed description of the various forms of the RADM and its evaluation history, see the Dennis, R. RADM Report (1995).
C-9
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
emissions to 1988 for the Eulerian Model
Evaluation Field Study funded by NAPAP.
These RADM emissions inventories had
county-level and detailed SCC and species-
level information incorporated into them to
provide the 80- and 20-km detail. The 812
Study emissions are principally computed at
the state level. While the 1985 812 Study
emissions are close to the NAPAP inventory,
they do not exactly match, nor do they have
the spatial, nor economic sector, nor species
detail within a state needed to run RADM. To
connect the 812 Study emissions to the RADM
emissions, the following approach was
followed: An industry/commercial-level
disaggregation (including mobile sources)
was developedfor the 812 emissions to allow
different sectors in a state to change their
emissions across time without being in lock
step and the detailed NAPAP emissions for
every 80- and 20-km RADM grid-cell were
grouped by state to the same level of industry/
commercial aggregation for an exact
correspondence. Then it was assumed that the
812 Study 1985 control emissions were
effectively the same as the 1985 NAPAP
emissions. Relative changes in emissions
between the 812 1985 control and any other
scenario (e.g., 1985 no-control, or 1990
control, or 1980 no-control, etc.) were then
applied to the 1985 NAPAP state-level
industry/commercial groups in the
appropriate 80- and 20-km grid cells. Thus,
state-level emissions for each group would
retain the same state-level geographic pattern
in the different scenarios years, but the mix
across groups could change with time. In this
\vay, the more detailed emissions required by
RADM were modeled for each scenario year
using the 812 Study emissions data sets.5
Although the focus of the present analysis is to
estimate the differences between the control and no-
control scenarios, it is useful to illustrate the abso-
lute levels of acid deposition associated with the two
scenarios. It is particularly important to demonstrate
the initial deposition conditions to preclude possible
misinterpretations of the maps showing percent
change in deposition. A relatively high percentage
change in a particular region, for example, may oc-
cur when initial deposition is low, even when the
change in deposition is also modest. The RADM-
Figure C-5. RADM-Predicted 1990 Total Sulfur Deposition
(Wet + Dry; in kg/ha) Under the Control Scenario.
Figure C-6. RADM-Predicted 1990 Total Nitrogen Deposi-
tion (Wet + Dry; in kg/ha) Under the Control Scenario.
Dennis, R. RADM Report (1995).
C-10
-------
Appendix C: Air Quality Modeling
Figure C-7. RADM-Predicted 1990 Total Sulfur Deposition
(Wet + Dry; in kg/ha) Under the No-control Scenario.
Figure C-8. RADM-Predicted 1990 Total Nitrogen Deposi-
tion (Wet + Dry; in kg/ha) Under the No-control Scenario.
modeled 1990 control scenario wet and dry sulfur
deposition pattern is shown in Figure C-5. A com-
parable map for nitrogen deposition is presented in
Figure C-6. Maps of the RADM-predicted 1990 no-
control scenario sulfur and nitrogen deposition are
presented in Figures C-7 and C-8, respectively.
No-control scenario acid deposition
profiles
Configuration of the RADM model for the
present analysis —including allocation of emission
inventories to model grid cells, design of meteoro-
logical cases, treatment of biogenic versus anthro-
pogenic emissions, and temporal, spatial, and spe-
cies allocation of emissions— are described in de-
tail in the RADM Report (1995). The remainder of
this section provides a summary description of the
acid deposition modeling effort.
For sulfur deposition, the RADM Engineering
Model (RADM/EM), which focuses on sulfur com-
pounds, was used to derive annual average total (wet
plus dry) deposition of sulfur in kilograms sulfur
per hectare (kg-S/ha) under both the control and
no-control scenarios. The relative changes in an-
nual average total sulfur deposition for each of the
80-km RADM/EM grid cells for 1975,1980,1985,
and 1990 were then compiled.
Nitrogen deposition was calculated in a differ-
ent manner. Since nitrogen effects are not included
in the computationally fast RADM/EM, nitrogen
deposition had to be derived from the full-scale,
15-layer RADM runs. Because of the cost and com-
putational intensity of the 15-layer RADM, nitro-
gen deposition estimates were only developed for
1980 and 1990. As for sulfur deposition, the rela-
tive changes in annual average total (wet plus dry)
nitrogen deposition, expressed as kg-N/ha, were cal-
culated for each 80-km grid cell and for each of the
two scenarios. It is important to note that ammonia
depositin contributes significantly to total nitrogen
deposition. However, the activities of sources as-
sociated with formation and deposition of ammo-
nia, such as livestock farming and wildlife, were
essentially unaffected by Clean Air Act-related con-
trol programs during the 1970 to 1990 period of
this analysis. Therefore, ammonia deposition is held
constant between the two scenarios.
C-ll
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Figure C-9. RADM-Predicted Percent Increase in Total
Sulfur Deposition (Wet + Dry; in kg/ha) Under the No-
control Scenario.
Figure C-10. RADM-Predicted Percent Increase in Total
Nitrogen Deposition (Wet + Dry; in kg/ha) Under the No-
control Scenario.
Summary differences in acid
deposition
Figure C-9 is a contour map showing the esti-
mated percent increase in sulfur deposition under
the no-control scenario relative to the control sce-
nario for 1990. Figure C-10 provides comparable
information for nitrogen deposition. These maps
indicate that by 1990 acid deposition would have
been significantly higher across the RADM domain
under the no-control scenario.
Examination of the percent change sulfur depo-
sition map indicates relatively large percentage
changes in the upper Great Lakes and the Florida-
Southeast Atlantic Coast areas. This result may ap-
pear somewhat surprising to readers familiar with
the historical patterns of acid deposition. However,
a review of the emission data and the control sce-
nario sulfur deposition map reveal the reasons for
this result.
First, Figure C-5 shows that control scenario
deposition rates are relatively low. As described
above, even a small absolute increase hi deposition
leads to a large percentage increase hi areas with
low initial rates of deposition. Second, the scenario
differences in SOx emission rates for these areas
were substantial. For example, 1990 no-control sce-
nario total SOx emissions for Michigan were ap-
proximately 1.8 million tons but control scenario
emissions for the same year were less than 600,000
tons; a reduction of over two-thirds. Similarly, 1990
no-control scenario emissions for Florida were over
2.3 million tons, compared to approximately
800,000 tons under the control scenario; also a re-
duction of about two-thirds. Almost 1 million tons
of the Michigan reduction and approximately 1.3
million tons of the Florida reduction were associ-
ated with utilities. Emission reductions of these
magnitudes would be expected to yield significant
reductions in rates of acid deposition.
Key caveats and uncertainties for acid
deposition
Regional-scale oxidant and deposition model-
ing involves substantial uncertainty. This uncer-
tainty arises from uncertainties in modeling atmo-
spheric chemistry, incomplete meteorological data,
normal seasonal and temporal fluctuations in atmo-
spheric conditions, temporal and spatial variability
C-12
-------
Appendix C: Air Quality MoiMing
in emissions, and many other factors. Uncertainties
specific to the RADM model, and this particular ex-
ercise, are discussed in detail in the RADM Report
(1995). It is important, however, to highlight some of
the potential sources of modeling uncertainty unique
to this analysis.
The first source of uncertainty specific to this
analysis is associated with the spatial and geographic
disaggregation of emissions data. As discussed in the
RADM Report, the RADM model requires emission
inventory inputs which are highly disaggregated over
both time and space. The ideal emissions inventory
fed into the RADM model includes day-specific emis-
sions to account for temperature effects on VOCs and
the significance of localized meteorological conditions
around major point sources. Given the broad-scale,
comprehensive nature of the present study, such de-
tailed emissions inventories were not available. How-
ever, the industry/commercial-level disaggregation ap-
proach developed for the present analysis would not
be expected to introduce any systematic bias, and the
contribution of this disaggregation of emissions would
not be expected to contribute significantly to the over-
all uncertainty of the larger analysis.
The acid deposition estimates included in the
present analysis are limited in that only the eastern 31
of the 48 coterminous states are covered. Although
acid deposition is a problem primarily for the eastern
U.S., acid deposition does occur in states west of the
RADM domain. The magnitude of the benefits of re-
ducing acid deposition in these western states is likely
to be small, however, relative to the overall benefits
of the historical Clean Air Act.
Particulate Matter
Developing air quality profiles for particulate
matter is significantly complicated by the fact that
"particulate matter" is actually an aggregation of dif-
ferent pollutants with varying chemical and aerody-
namic properties. Particulate species include chemi-
cally inert substances, such as wind-blown sand, as
well as toxic substances such as acid aerosols; and
include coarse particles implicated in household soil-
ing as well as fine particles which contribute to hu-
man respiratory effects. In addition, emissions of both
primary particulate matter and precursors of second-
arily-formed particulates are generated by a wide va-
riety of mobile and stationary sources, further com-
plicating specification of particulate air quality mod-
els. Finally, particulate air quality models must take
account of potentially significant background concen-
trations of atmospheric particles.
Modeling multiple species and emission sources,
however, is not the only major challenge related to
particulate matter which is faced in the present study.
Over the 1970 to 1990 period being analyzed, under-
standing of the relative significance of fine versus
coarse particles evolved significantly. Up until the
mid-1980s, particulate air quality data were collected
as Total Suspended Particulates (TSP). However, dur-
ing the 1980s, health scientists concluded that small,
respirable particles, particularly those with an aero-
dynamic diameter of less than or equal to 10 microns
(PM10), were the component of particulate matter pri-
marily responsible for adverse human health effects.
As of 1987, federal health-based ambient air quality
standards for particulate matter were revised to be ex-
pressed in terms of PM10 rather than TSP. Starting in
the mid-1980s, therefore, the U.S. began shifting away
from TSP monitors toward PM10 monitors. As a re-
sult, neither TSP nor PM10 are fully represented by
historical air quality data over the 1970 to 1990 pe-
riod of this analysis. Furthermore, a large number of
U.S. counties have no historical PM monitoring data
at all, making it difficult to estimate changes in ambi-
ent concentrations of this significant pollutant for ar-
eas containing roughly 30 percent of the U.S. popula-
tion.
Given the relative significance of particulate mat-
ter to the bottom-line estimate of net benefits of the
historical Clean Air Act, it was important to develop
methodologies to meet each of these challenges. The
methodologies developed and data used are described
primarily in the two supporting documents SAI PM
Report (1992) and SAI PM Report (1995).6 To sum-
marize the overall approach, historical TSP data were
broken down into principal component species, in-
cluding primary particulates, sulfates, nitrates, organic
particulates, and background particulates. Historical
data were used for the control scenario. To derive the
no-control profiles, the four non-background compo-
nents were scaled up based on corresponding
no-control to control scenario ratios of emissions and/
or modeled atmospheric concentrations. Specifically,
the primary particulate component was scaled up by
the ratio of no-control to control emissions of PM.
6 In addition, SAI memoranda and reports which supplement the results and methodologies used in this analysis are included in
the references.
C-13
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Organic constituents were scaled up by the ratio of
no-control to control VOC emissions. In the eastern
31 states where RADM sulfate and nitrate data were
available, values for SO4 and NO3 from an appropri-
ate RADM grid cell were assigned to the relevant
county and used to scale these components of PM.
For the western states not covered by RADM, sul-
fates were scaled up by the change in SO2 emissions
and nitrates were scaled up the change in NOx emis-
sions. No-control scenario profiles were then con-
structed by adding these scaled components to back-
ground concentrations.
To resolve the problem of variable records of TSP
and PM10 data, both TSP and PM10 profiles were gen-
erated for the entire 20 year period. Missing early year
data for PM,0 were derived by applying region-spe-
cific, land use category-specific PM10 to TSP ratios to
the historical TSP data. Missing recent year TSP data
were derived for those areas where PM10 monitors
replaced TSP monitors by applying the reciprocal of
the relevant PM]0 to TSP ratio. The methodology is
described in detail in the SAI PM Report (1995).
In addition, to increase the geographic coverage
of estimates of air quality, an interpolation methodol-
ogy7 was developed to predict air quality for the con-
trol scenario in counties without measured data. PM
concentrations were estimated by first estimating the
components of PM (i.e., sulfate, nitrate, and organic
particulate, and primary particulate). The methodol-
ogy for developing the concentrations of components
within a county differed depending upon whether the
county was within or outside the RADM domain.
For those counties within the
RADM domain, the RADM modeled
concentrations for 1980 and 1990 were
used to predict sulfate air quality. Re-
lationships based on linear regressions
that related 1980 and 1990 RADM sul-
fate concentrations to estimated sulfate
particulate concentrations were calcu-
lated for counties with AIRS data. Sul-
fate particulate concentrations were
then calculated for all counties in the
domain by applying the regression re-
sults to the RADM grid cell concen-
tration located over the county center.
Statewide average nitrate, VOC, and
primary particulate concentrations
were calculated from measured ambi-
ent TSP and PM]Q to describe these constituents in
counties without data. Control scenario PM profiles
were developed by adding the RADM-estimated sul-
fate particulate levels to the statewide average nitrate,
VOC, and primary particulate levels, and background.
For counties outside the RADM domain, an al-
ternate procedure was used. Using the primary and
secondary particulate estimates for counties with data,
statewide average sulfate, nitrate, VOC, and primary
particulate concentrations were determined. Control
scenario PM10 was predicted by adding the statewide
averages of all primary and secondary particulate, and
background. Using this method, all counties that did
not have monitors and are in the same state are as-
signed the same PM concentration profiles. These in-
terpolated results are clearly less certain than results
based on actual historical monitoring data and are
therefore presented separately.
Control scenario particulate matter
profiles
The number of TSP monitors peaked in 1977 and
declined throughout the 1980s. Table C-6 summarizes
the daily (i.e., 24-hour average) TSP monitoring data
used as the basis for development of the control sce-
nario air quality profiles. Most of the TSP and PM10
monitors collected samples every six days (i.e., 61
samples per year).
Daily PM10 data were also collected for each year
between 1983 and 1990. Table C-7 summarizes the
daily PM10 monitoring data used for the control sce-
nario air quality profiles.
Year,'/
1970
, 1975 ,,
1980
1985
im
; Ntunlier'pf '
Monitors
'/** t* *
751; -
V 3,467--
" 349?*,""
v '*$&.
„ , -
923;;; -
' ' * " / **
* " ' ^ * ^
Number of ' '
Counties ' '"
245 - ,
1,146
1,478'
- 'i:M^- \
410 '
>v* Number of
« ''""Sample!
, 56,804, w ,»
- -221,873-
'' "'234,$SU'T
!""X18W'*'
, ,,„ $4§L~v.
<> i!§$t / r
„ JMfeanjNuiriljer
of Samples per
Monitor
7C.
<.•%.#•,
6cv: •
'*':' 65- ------
r *>.-J
,,,„, ,
.„„/, , "^ '*:
vff. ' ' ""' f V''''*V */ £, '?s* * VV''* J''; ^ v' ^'/ ^ 'J'^*'''* &%/ '''"'
-Data Souircet-J/I PM'Repott (5995),- •'•
7 The interpolation methodology is described in detail in SAI, 1996. Memo from J. Langstaff to J. DeMocker. PM Interpolation
Methodology for the section 812 retrospective analysis. March 1996.
CM4
-------
Appendix C: Air Quality Modeling
Table C-7* J
allow differentiation between urban and
rural locations for coarser particles.
%-,'.- ''\ x
¥ear '*
198S" '
1990
- -""-~»"
^ s / ,*^- " "*
^ Dumber of ,,,-
v; 'Monitors"
-' ' • -v " '
303"
s>;i;249 •*
- - . --<-
'V-J"'~ " "X^*
"* ' \ XV s ^ s/ s
/Niri»fc>rof
•-(Oolintfes '-
,,u.- j-^ , „/
556° "v:
- .- - , -- ,,,
' " * * \$v/f *' '
, Bomber of
"Samples
" 22,031---'""'
- 98,9C&T°
* /, f
* IMteaB Number-*.'
• of Samples per
Monitor
73
s ., < j V v*'
• " < ' --79- -
-Date Source: -5AIPM Repott (1995),, ,„
Further speciation of TSP and PM10 air quality
data serves two purposes in the present analysis. First,
speciation of TSP into PM10 and other fractions al-
lows derivation of PM10:TSP ratios. Such ratios can
then be used to estimate historical PM10 for those years
and monitors which had TSP data but no PM10 data.
The reciprocal ratio is also applied in this analysis to
expand 1985 and 1990 TSP data to cover those areas
which monitored PM]0 but not TSP. The second pur-
pose served by speciation of particulate data is, as
described earlier, to provide a basis for scaling up
concentrations of each species to derive no-control
scenario TSP and PM10 profiles.
To break the TSP and PM10 data down into com-
ponent species, speciation factors were applied to the
PM fractions with aerodynamic diameters below 2.5
microns (PM25) and from 2.5 to 10 microns (PM10).
The PM2 5 speciation factors were drawn from a Na-
tional Acid Precipitation Assessment Program
(NAPAP) report on visibility which reviewed and
consolidated speciation data from a number of stud-
ies.8 These factors are presented in Table C-8. In the
table, fine particle concentrations are based on par-
ticle mass measured after equilibrating to a relative
humidity of 40 to 50 percent; and organics include
fine organic carbon.
To develop speciation factors for coarser particles
(i.e., in the PM2S to PM10 range), SAI performed a
review of the available literature, including Conner et
al. (1991), Wolff and Korsog (1989), Lewis and
Macias (1980), Wolff et al. (1983), Wolff et al. (1991),
and Chow et al. (1994).9 These speciation factors are
summarized in Table C-9. Data were too limited to
The TSP and PM10 control scenario
profiles developed based on this meth-
odology are available on diskette, un-
der the filenames listed in Table C-10.
No-control scenario
particulate matter profiles
^ -'^J To derive the no-control TSP and
mmimiiitiiiiiA PMIO air quality profiles, individual
component species were adjusted to
reflect the relative change in emissions or, in the case
of sulfates and nitrates in the eastern U.S., the rela-
tive change in modeled ambient concentration. The
following excerpt from the SAI PM Report (1995)
describes the specific algorithm used:10
"For the retrospective analysis, the no-CAA
scenario TSP and PMW air quality was
estimated by means of the following
algorithm:
• Apportion CAA scenario TSP and PM10
to size categories and species;
• Adjust for background concentrations;
• Use a linear scaling to adjust the non-
background portions of primary
particulates,sulfate, nitrate, and organic
components based on emissions ratios of
PM, SOy NOx and VOC, and Regional
Acid Deposition Model (RADM) annual
aggregation results for SO4 and NO3;
• Add up the scaled components to estimate
the no-CAA scenario TSP and PM
concentrations."
10
The specific procedures and values used for the
linear rollback, speciation, fine to coarse particle ra-
tio, scaling, and background adjustment steps are de-
scribed in detail in the SAI PM report (1995).1' Table'
C-ll lists the names of the electronic data files con-
taining the TSP and PM10 profiles for the no-control
scenario.
8 J. Trijonis, "Visibility: Existing and Historical Conditions-Causes and Effects," NAPAP Report 24,1990.
9 This literature review, and complete citations of the underlying studies, are presented in the SAI PM Report (1995), pp. 4-2 to
4-6 and pp. R-l to R-2, respectively.
10 SAI PM Report (1995), p. 5-1.
11 SAI PM Report (1995), pp. 5-2 to 5-15.
__
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table C-8. Fine Particle (PM^>Cherrrical Composition by t££ 1
Component
RURAk BAST '" '"' «^^5.
Fine particle concentration
Ammonium sulfate ' " ; '*
Ammonium nitrate
Organics
URBANBAST - ^.^^^
Fine particle concentration -"'
Ammonium sulfate
Ammonium nitrate
Organics
^ ^^uhwi' -"-7
='- .Number dftV
Data Sets, „
Mean ,
.-> Range of;,,,,
l^^?^~x s% * , ><*&5?&>&&KW^, , V'**
v ,.'.,1*'"1'
' r^' *
*&$$',, ' " " ""V-tf
' ' -i'i*' .-
- 36 /-*>-/ ,
"55::-"'
1 "-'-,--,;.
- - - ,24,"""
^- '^S?*^''",', r'*~
' — =v25v,'
25 ,x, ,„
: ";-i7:"":
25"'--, -,
^'"-«-- -
, '^"-',
:;"'-/-35,
' . ,,4/ /"-
V" ' 27, -'
r^^'^Sr**'
• 35< --*„,„
-*- '--|fe,
.,:,:,,is;/r"
42, ;' ''
-,:•/•;, ',6-46- '-,:
''••j ^ 4 1 ** 66is1^^
'-,-„:,„;'?' -^
^,^,34-,-,
— ;;^^,,
29-43
53 ,-,57'
""' 1
.. .,*&.&"'
^^^r
' ""*«&*! J
""" ' fst7'
' 14-41 v'
',;;'",' 13-74,"'"" '"
-Was .'
""" 2-37-, ,»,-,,
-"' 25-79'
Data Sources: SAIPM Report (1995); and I. Ifi-josis, nVisMlity
to,"NAPAP Report 24,1990. "--^"
Summary differences in paniculate
matter air quality
Figure C-ll provides one indication of the esti-
mated change in participate matter air quality between
the control and no-control scenarios. Specifically, the
graph provides data on the estimated ratios of 1990
control to no-control scenario annual mean TSP con-
centrations in monitored counties. The X-axis values
represent the mid-point of the ratio interval bin, and
the Y-axis provides the number of counties falling into
each bin. Figure C-ll indicates that annual average
TSP concentrations would have been substantially
higher in monitored counties under the no-control sce-
nario.
Key caveats and uncertainties for
paniculate matter
There are several important caveats and uncer-
tainties associated with the TSP and PMJO air quality
profiles developed for this study. Although further
C-16
-------
Appendix C: Air Quality Modeling
(3^'C4~^eP^
-ComponeatV
es
'
:---
5<" %?«>
Coarse parflcle^oncetttrfitioa'
IS
IS
, .,,,
• Data S0!if,ce: SAI |SSlle£>ott-<1995). ' -,',
% ' *
o^O. 3pM>Q>attt>i Sceriario^t; Quality PBitife Hie
, \- -Aahual"Meaji
' 2ncffl!kl'est Dally- '
' *' PMlOOtt^UAT
- - *•-, ^ -' - ^ \
r "(3Q" ie£ei.st0 peiKen'dies fom 5 to 95, inaio^d.ng'19 perceatiie'daUi"files .avafiffljfe,
'
C-17
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table C-ll. PM No-Control Scenario Air 'Quality-Profile 71''",',-'•
Filenames. *^*T **'
Component
ISP
ISP
TSP
PMie
PM«
PM««
Indicator
Annual Mean
2nd Highest Daily
(X)tfi Pewentite
Annual Mean
2nd Highest Daily
(X)th Percentile'
Filename '"''#<
T&O'JMEA J>AT< :-4?-
' ' '" TSPNCHLDAH-,-.^ ,
* ' '' ' -•*«•{• £-W-v
TSPNC(X).DAT,,>~,,
•pMlONCME-DA^s.
PMiONtM38&V
PMIONC ./.-
! ,«-.v,,
Iffw "(X)" refers to peicentiles ftoitt 5 to 95, ittdjcaftag 19)
available for TSP and 19 similar files available fop PM«j for exam^
;50thpetcentileTSP air quality data profile foe theuo-coattd;sce»adtfis,iiamed
ITSPNCSO.DAT. '.,.'* -s
reductions in these uncertainties were not possible for
this study given time and resource limitations, the rela-
tive importance of particulate matter reduction con-
tributions towards total benefits of the Clean Air Act
highlights the importance of these uncertainties.
A number of uncertainties were introduced in the
process of speciating and rolling up individual com-
ponents of particulate matter. First, temporal and spa-
tial variability in the size and chemical properties of
particulate emissions are substantial. These charac-
teristics change from day to day at any given loca-
tion. Second, using changes in proxy pollutant emis-
Figure C-ll. Distribution of Estimated Ratios for 1990
Control to No-control Annual Mean TSP Concentra-
tions, by Monitored County.
50
40
I 30
jjj 20
B
2
10
0.00 0.20 0.40 0.60 0.80 1.00
Ratio of CAA:No-CAA Annual Mean TSP (interval midpoint)
sions, such as using SO2 as a surrogate for SO4
in the western states, to roll up individual PM
components may introduce significant uncer-
tainty. Third, even assuming a satisfactorily
high degree of correlation between target and
surrogate pollutants, relying on predicted
changes in emissions at the state level further
compounds the uncertainty. Finally, and per-
haps most important, using PM10 to TSP ratios
derived from late 1980s monitoring data may
lead to significant underestimation of reduc-
tions in fine particulates achieved hi earlier
years. This is because historical Clean Air Act
programs focused extensively on controlling
combustion sources of fine particulates. As a
result, the share of TSP represented by PM10
observed in the late 1980s would be lower due
to implementation of controls on combustion
sources. This would lead, in turn, to underesti-
mation of baseline PM10 concentrations, as a
share of TSP, in the 1970s and early 1980s. If baseline
PM10 concentrations in these early years are underes-
timated, the reductions in PM10 estimated by linear
scaling would also be underestimated.12
Ozone
Nonlinear formation processes, long-range atmo-
spheric transport, multiple precursors, complex atmo-
spheric chemistry, and acute sensitivity to meteoro-
logical conditions combine to pose substantial diffi-
culties in estimating air quality profiles for ozone.
Even in the context of an aggregated, national study
such as this, the location-specific factors controlling
ozone formation preclude the use of roll-up modeling
based on proxy pollutants or application of state-wide
or nation-wide average conditions. Such simplifica-
tions would yield virtually meaningless results for
ozone.
Ideally, large-scale photochemical grid models —
such as the Urban Akshed Model (UAM)— would
be used to develop control and no-control scenario
estimates for ozone concentrations in rural and urban
areas. Such models provide better representations of
the effects of several important factors influencing air
quality projections such as long-range atmospheric
transport of ozone. However, the substantial comput-
ing time and data input requirements for such models
precluded their use for this study.13 Instead, three sepa-
12 See SAI PM Report (1995), p. 5-9.
13 For a description of the extensive data inputs required to operate UAM, see SAI Ozone Report (1995), p. 1-1.
C-18
-------
Appendix, C: Air Quality Modeling
Table C-12«-¥rbii''reas Modeled" with GMk ,
Albuquerque, NM "»>™ • ' "
Anderson, ,13^,, ..',,,, ,,,>v s s- :-'-' >
AppMon, Wl --svwo -—
'Atjahtic City, &J
Augusta;'GA'-SG^-' •'' *'j?"f*
Austin, IX , • _>,„< ,
Bdfirnore, MD '- --- ,, ^
Baton Rouge,' LA -•c"r ',, '"''"'
Beffingftafli.WA ''' *
'
_
BJrHjinghara,,AL - , , „ „
Boston, MA,,,.,, ,,, -,;-.,
"' "
Canton, OH ' ' , - -"'
Cedar Rap'i&s;*! A- ,,,-•--
Chaaipaign.IL \\; *r ' '/,
''''
CharlQtte, NC, ,, , ^, ,-,:
yChattenpogi'TN^GAj,': - ' ^
jColorada, Springs,, 00-
fialitovTX-" vX;' --'-,vs
Davenport, JA-B^ *" ''**'
Decatur^IL""!'-" '""''"'"
'Denver, CO"" s>""1<'
Detroit, Ml' ---"-> ~ ' ,.*
Brt^'PA
ri.C Ml -;«'''' -'-.,<,,;""
>For£C * "
o,'WJ ,
, TN-VA- -;
' i. Toitnsto wn, PA .-• *'*''''
-
,,- Lafayette, IN ",
>,^- tafay'ette, LA """""
'
Lancaster,' PA- A
1 '' ' Lansing, MI '
'- ,LasCruce,s,KM
La's-Vegas^NV
ton, K.Y '
, AL .
' Ifew Orleans, LA
"New^ork^lSfY
,,,
' '''Oklahoma City, 'OK
Qriatiiio, FL -
' ,. Owerisboro, KY \ ,,,
'„,, „ paikersburg, WV-,^
Pascagojik, MS'{-
" "-",' Pensac'6la,'FL* ' - \ -, -
'•'\"" ' *;",,',,Phikdelphia,>A'"'"
"':,- =,v^-'^ --' "
Portsaioath,M,H,; ,
';" '"' Raleigh;'NC - "/" " "'
Reading,?^"" . •
v'~' 6, NV" -'^"',, ;;,"
,,„ RoanoteVA , - ' •
- \/f"lloGh'ester»NY, _
"Salt Lake Ci$Tte
'
San "Diego, CA
, , , San. Fr'a'n'ci,scq, CA ' " T^}^ *
"
SatttaBarSara,CA'
'•-•^'-'Seatfle/WA ,, -'"-"
Sheboygatf, Wl
Shreveportj'Ij-A'^,-,, _
- Springfield, IL ',,,,,,,
' Springfield, MO ->;
, ,, , St,Louis,MO-;'- - , ,-. „
''" "'"'" 'Steabenvi!ie,'QH-WVs "" ,„„,,.„,
,., ^ '•< ' \ • , ,3' xrt/'. '. -^-^^^"
Syracuse, NY ' ,
Tallahassee, FL
, _ '_,,__ ,T«rre Haute, 1$,
' °: , , T.oeson, AZ
Ventura'C";' " s"Vi'ctof)a>-TX"
, Wheeling/WV'-QH ,-\-
'"
rate modeling efforts were conducted to provide ur-
ban and rural ozone profiles for those areas of the lower
48 states in which historical ozone changes attribut-
able to the Clean Air Act may be most significant.
First, for urban areas the Ozone Isopleth Plotting
with Optional Mechanisms-IV (OZDPM4) model was
14 See SAI Ozone Report (1995), p. 1-1.
run for 147 urban areas. Table C-12 lists the urban
areas modeled with OZIPM4. Although it requires
substantially less input data than UAM, the OZIPM4
model provides reasonable evaluations of the relative
reactivity of ozone precursors and ozone formation
mechanisms associated with urban air masses.14 Three
to five meteorological episodes were modeled for each
C-19
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
of the 147 urban areas; and for each of these, four
model runs were performed to simulate the 1980 and
1990 control and no-control scenarios. The outputs of
these model runs were peak ozone concentrations for
each of the target year-scenario combinations. The
differentials between the control and no-control sce-
nario outputs were averaged over meteorological epi-
sodes and then applied to scale up historical air qual-
ity at individual monitors to obtain no-control case
profiles. As for the other pollutants, the control sce-
nario profiles were derived by fitting statistical distri-
butions to actual historical data for individual moni-
tors.
Second, the 15-layer RADM runs for 1980 and
1990 were used to estimate the relative change in ru-
ral ozone distributions for the eastern 31 states. In ad-
dition, a limited number of 20-km grid cell high-reso-
lution RADM runs were conducted to benchmark the
15-layer, 80-km RADM median ozone response and
to estimate high ozone response. The relative changes
in modeled median and 90th percentile rural ozone
were then assumed to be proportional to the changes
in, respectively, the median and 90th percentile ozone
concentrations. The domain of the high-resolution
RADM is shown in Figure C-4 and the general RADM
domain is shown in Figure C-12.
Finally, the SARMAP Air Quality Model
(SAQM) was run for EPA by the California Air Re-
sources Board (CARD) to gauge the differences in
peak ozone concentrations in key California agricul-
tural areas for 1980 and 1990. No-control profiles were
developed for ozone monitors in these areas by as-
suming the relative change in peak'ozone concentra-
tion also applies to the median of the ozone distribu-
tion. The domain of the SAQM is shown in Figure C-
12.
Figure C-12. RADM and SAQM Modeling Domains, with Rural Ozone Monitor Locations.
C-20
-------
Appendix C: Air Quality Modeling
Control scenario ozone profiles
For ozone, air quality profiles were developed
from historical AIRS data and calculated for individual
monitors based on 1, 2, 6, 12, and 24 hour averaging
times. Profiles based on the daily maximum concen-
trations for these averaging times were also calculated.
Given the significance of seasonal and diurnal ozone
formation, twelve separate profiles of hourly ozone
distributions were also developed for six 2-month
periods and for daytime and nighttime hours. The
2-month periods are January-February, March-April,
and so forth. The diurnal/nocturnal profiles are divided
at 7 A.M. and 7 P.M. Local Standard Time. All of
these profiles are based on constructing 1, 2, 6, 12,
and 24-hour moving average profiles from the hourly
ozone data from each monitor.15 A two-parameter
gamma distribution is then fitted to characterize each
of these air quality profiles.16 The functional form of
the gamma distribution, the basis for deriving the
monitor-specific values for mean and variance, and
an analysis of the goodness of fit to the data are pre-
sented in the SAI Ozone Report (1995).
Table C-13 summarizes the ozone monitoring data
used as the basis for the control scenario profiles. The
distribution of these monitors among urban, subur-
-13;,'
/Data. „ ;']';
-"'•
,-Year
199$
.'791
•'-"'- $34 ;''
Number of
Counties
415-
ban, and rural locations is presented in Table C-2 of
the SAI Ozone Report (1995).
Given the substantial number of alternative air
quality profiles for ozone, approximately 20 high-den-
sity disks are required to hold the profiles, even in
compressed data format. Resource limitations there-
fore preclude general distribution of the actual pro-
files. As discussed in the caveats and uncertainties
subsection below, however, the substantial uncertain-
ties associated with model results for any given area
preclude application of these profiles in contexts other
than broad-scale, aggregated assessments such as the
present study. The historical ozone monitoring data
used as the basis for this study are, nevertheless, avail-
able through EPA's Aerometric Information Retrieval
System (AIRS).
No-control scenario ozone profiles
The specific modeling methodologies for the
OZIPM4 runs —including emissions processing, de-
velopment of initial and boundary conditions, meteo-
rological conditions, simulation start and end times,
organic reactivity, and carbon fractions— are de-
scribed in detail in the SAI Ozone Report (1995).
Assumptions and modeling procedures not otherwise
described in the SAI report were conducted in accor-
dance with standard EPA guidance.17
Similarly, the RADM modeling methodology
used to estimate changes in day-time rural ozone dis-
tributions in the eastern 31 states are described in de-
tail in the RADM Report (1995). The referenced re-
port also provides complete citations of the literature
associated with development, standard application
procedures, and evaluation of RADM by the National
Acid Precipitation Assessment Program (NAPAP).
To derive the no-control scenario results for key
California agricultural areas, the California Air Re-
sources Board and US EPA's Region 9 office agreed
to conduct three runs of the SAQM. For the 1990 con-
trol scenario, the 1990 SARMAP base case scenario
adopted for California State Implementation Plan
modeling was adopted.18 Derivation of 1990
IS For the nighttime profiles, only 1, 2, 6, and 12-hour averaged concentrations are derived.
16 Normal and lognormal distributions were also developed and tested for goodness of fit; however, the gamma distribution provided
a better representation of the concentration distribution. See SAI Ozone Report (1995), page 4-2.
17 US EPA, Office of Air Quality Planning and Standards, "Procedures for Applying City-Specific EKMA," EPA-450/4-89-012, 1989.
18 Documentation of the SARMAP Air Quality Model and the SARMAP 1990 base case can be found in the SAQM references listed
at the end of this appendix.
C-21
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
no-control and 1980 control and no-control scenarios
was based on adjusting the aggregate mobile, point,
and area source VOC and NOx emissions associated
with each of these cases. For example, the 1980
no-control results were derived by, first, multiplying
the 1990 SARMAP base case mobile source VOC
emissions by the ratio of 1980 no-control scenario to
1990 control scenario mobile source VOC emissions
derived for the present study. Similar adjustments were
made for point and area sources, and for NOx. The
SAQM was then re-run holding fixed all other condi-
tions associated with the 1990 SARMAP base case,
including meteorology, activity patterns, and other
conditions. The specific emission ratios used to modify
the 1990 SARMAP base case are presented in Table
C-14. The ratios themselves were derived by adding
on-highway and off-highway emissions to represent
the mobile source category; adding utility, industrial
process, and industrial combustion emissions to rep-
resent point sources; and using commercial/residen-
results. This is because OZDPM4 provides only the
maximum hourly ozone concentration. However, to
estimate all the various physical consequences of
changes in ambient ozone concentrations, the current
study requires estimation of the shift in the entire dis-
tribution of ozone concentrations. Since it is daytime
ozone season concentrations which are most sensi-
tive to changes in VOC and NOx emissions, the pre-
dicted shifts in the most important component of the
ozone concentration distribution are reasonably well-
founded. The method adopted for this analysis in-
volved applying the no-control to control peak con-
centration ratio to all concentrations in the distribu-
tion down to a level of 0.04 ppm. The 0.04 ppm level
is considered at the high end of hypothetical ambient
ozone concentrations in the absence of all anthropo-
genic ozone precursor emissions. A ratio of 1.0 is used
for ozone concentrations at or near zero. The method-
ology is described in more detail in the SAI Ozone
Report (1995) on page 4-6.
Table C-14. Apportionment of Bnissidns; Inventories for-SAQM Runs.
VOC
NO,
Category .
Mobile
Area
Point
Mobile
Area
Point
' ' 1980 Control - "•" • •
.to.l^OConfcol Ratio"
'.Co,, , - 1344 ' - - -'-
,-,«<"•>*• '
: ",„ 0.820 -, ,\, 'w
* - ' L284 ""^ ,','*
< 11042"
''' $»
0.731
1 \^j&$^: - '
WSONo-CoatroI Hf"4'-
1990 Control Ratio '„',
•'- 'J-055* .;•„';".
". - "";oM":;
1:439" "°
,'>„:*- U4jrf;;'
0.738' ""*! '
''•"' ' , <'/'/#•' s' ; '
'* " 1.339 ?—*"'<"-•
'' 'Xv^v sss, ,
199&No-Co»ttiolto>
"'. ;1990,e ^ffrf
„, f -,,; ,%:/, JiSvi??;/, . .
- i.f (&"';' : --
1.232 ""
,' ,,, ', 1,677", *"*';':
1.058-- ,-:-
• • • t fiQ
*.*V? . ,,,;,, ,,..
tial emissions to represent area sources. The no-control
scenarios were then derived by adjusting the peak and
median of the control scenario ozone distribution
based on the ratio of S ARMAP-predicted peak ozone
concentrations under the control and no-control sce-
narios.
The relative results of the control and no-control
scenario runs of the OZEPM4, RADM, and SAQM
models were then used to derive the no-control case
air quality profiles. For the urban monitors relying on
OZIPM4 results, only ozone-season daytime concen-
trations could be calculated directly from OZIPM4
Estimating changes in rural ozone concentrations
is required primarily for estimating effects on agri-
cultural crops, trees, and other vegetation. For this
reason, only the differences in daytime, growing sea-
son ozone concentrations are derived for the present
study. As described in detail in the SAI Ozone Report
(1995) on page 4-7, the no-control rural ozone pro-
files are calculated by, first, taking the ratio of the
average daytime growing season ozone concentrations
simulated by RADM or SAQM (whichever is relevant
for that monitor). The ratio of no-control to control
scenario average ozone concentration is then applied
to all the hourly concentrations from that monitor.
C-22
-------
Appendix C: Air Quality Modeling
Profiles based on 1, 2, 6, 12, and 24-hour averages
are then calculated for the control case; and averages
for daytime hours are calculated for the no-control
case.19 Even though the control and no-control sce-
nario off-season profiles are held constant, profiles
for the no-control scenario are developed for all
months of the year since the ozone season varies
throughout the country.
Summary differences in ozone air
quality
Figure C-13 presents a summary of the results of
the 1990 OZIPM4 results for all 147 of the modeled
urban areas. Specifically, the graph depicts a fre-
quency distribution of the ratio of control to no-control
scenario peak ozone. While the vast majority of simu-
lated peak ozone concentration ratios fall below 1.00,
eight urban areas show lower simulated peak ozone
for the no-control scenario than for the control sce-
nario. For these eight urban areas, emissions of pre-
cursors were higher under the no-control scenario;
however, the high proportion of ambient NOx com-
pared to ambient non-methane organic compounds
(NMOCs) in these areas results in a decrease in net
ozone production when NOx emissions increase. Fig-
ures C-14 and C-15 present frequency distributions
for control to no-control ratios of average ozone-sea-
son daytime ozone concentrations at rural monitors
as simulated by RADM and SAQM, respectively.
These figures indicate that, by 1990, no-control
scenario ozone concentrations in the modeled areas
would have been generally higher in both urban and
rural areas. Rural area concentrations differences are
not as great as urban area differences due to (a) the
differentially greater effect of CAA emission controls
in high population density areas, and (b) potential dif-
ferences in the models used for urban and rural areas.
Ozone reductions in both rural and urban areas
projected in this analysis are not as proportionally large
as the estimated reductions in emissions of ozone pre-
cursors for at least four reasons. First, current knowl-
edge of atmospheric photochemistry suggests that
ozone reductions resulting from emissions changes
will be proportionally smaller than the emissions re-
ductions. Second, biogenic emissions of VOCs, an
important ozone precursor, are significant and are held
constant for the control and no-control scenarios of
this analysis. Biogenic emissions are important be-
cause they contribute roughly half of the total
Figure C-13. Distribution of Estimated Ratios for 1990
Control to No-control OZIPM4-Simulated 1-Hour Peak
Ozone Concentrations, by Urban Area.
30
20
I
10
MI a
0.00 0.20 0.40 0.60 0.80 1.00
RatioofCAANo-CAA Peak Ozone (interval midpoint)
1.20
Figure C-14. Distribution of Estimated Ratios for 1990
Control to No-control RADM-Simulated Daytime Aver-
age Rural Ozone Concentrations, by RADM Grid Cell.
200
150
S 50
0.00 0.20 0.40 0.60 0.80 1.00 1.20
Ratio of CAA *Jo-CA A Ozone-Season Daytime Average Ozone (interval midpoint)
Figure C-15. Distribution of Estimated Ratios for 1990
Control to No-control SAQM-Simluated Daytime Aver-
age Ozone Concentrations, by SAQM Monitor.
10
1 8
1:
CM 4
O
I
0.00 0.20 0.40 0.60 0.80 1.00 1.20
RatioofCAA'No-CAA Ozone-Season Daytime Average Ozone (interval midpoint)
19 The no-control scenario nighttime profiles are assumed to be the same as the control scenario profiles.
___
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
(manmade plus natural) VOC emissions nationwide.
Due to this abundance of VOC loading and the inher-
ent nonlinearity of the ozone-precursor response sys-
tem,20 historical reductions in anthropogenic VOC
emissions can yield minimal reductions in ozone, es-
pecially in rural environments. Third, this rural effect
also influences urban areas receiving substantial ozone
transported in from surrounding areas. Consequently,
the effect of emission controls placed in urban areas
often is reduced since much of the urban area ozone
is imported. Thus, the problem is truly regionalized
given the importance of transport, biogenic emissions
and associated urban-rural interactions, all contribut-
ing toward a relatively non-responsive atmospheric
system.21 Finally, physical process characterizations
within OZIPM4 are severely limited and incapable of
handling transport, complex flow phenomena, and
multi-day pollution events in a physically realistic
manner. Consequently, it is possible that the OZDPM4
method used herein produces negative bias tenden-
cies in control estimations. Additional discussion of
uncertainties in the ozone air quality modeling is pre-
sented in the following section.
Key caveats and uncertainties for ozone
There are a number of uncertainties hi the overall
analytical results of the present study contributed by
the ozone air quality modeling in addition to the po-
tential systematic downward bias discussed above.
First, there are substantial uncertainties inherent in any
effort to model ozone formation and dispersion. These
uncertainties are compounded in the present study by
the need to perform city-specific air quality modeling
using OZIPM4, which is less sophisticated than an
Eulerian model such as the Urban Airshed Model.
However, while the absolute ozone predictions for any
given urban area provided by OZIPM4 may be quite
uncertain, the process of aggregating results for a num-
ber of cities and meteorological episodes should sig-
nificantly reduce this uncertainty.22 Urban areas for
which ozone changes may be overpredicted are offset
to some degree by urban areas for which the change
in ozone concentrations may be underpredicted. In
weighing the significance of this source of uncertainty,
it is important to consider the central purpose of the
present study, which is to develop a reasonable esti-
mate of the overall costs and benefits of all historical
Clean Air Act programs. All analyses are based on
relative modeled results, and ratios of the model pre-
dictions for the control and no-control scenarios, rather
than the absolute predictions. As a result of this, the
effect of any bias in the model predictions is greatly
reduced due to partial cancellation.
Additional uncertainty is contributed by other
limitations of the models, the supporting data, and the
scope of the present analysis. Relying on linear inter-
polation between 1970 and modeled 1980 results to
derive results for 1975, and between modeled results
for 1980 and 1990 to derive results for 1985, clearly
adds to the uncertainty associated with the RADM-
based rural ozone estimates. Assuming that changes
hi peak concentration predicted by OZIPM4 and
SAQM can be applied to scale hourly ozone values
throughout the concentration distribution also contrib-
utes to uncertainty. Resource and model limitations
also required that night-time ozone concentrations be
held constant between the scenarios. This leads to an
underestimation of the night-time component of ozone
transport. Finally, changes in rural ozone in areas not
covered by RADM or SAQM could not be estimated.
As a result, potentially significant changes in ambi-
ent ozone in other major agricultural areas, such as in
the mid-west, could not be developed for this analy-
sis. The Project Team considered using an emissions
scaling (i.e., a roll-back) modeling strategy to develop
crude estimates of the potential change in rural ozone
concentrations in monitored areas outside the RADM
and SAQM domains. However, the Project Team con-
cluded that such estimates would be unreliable due to
the nonlinear effect on ozone of precursor emission
changes. Furthermore, the team concluded that
baseline levels of ozone and changes in precursor
emissions in these areas are relatively low. The deci-
sion not to spend scarce project resources on estimat-
ing ozone changes hi these rural areas is further sup-
ported by the relatively modest change in rural ozone
concentrations estimated within the RADM and
SAQM domains.
20 Nonlinear systems are those where a reduction in precursors can result hi a wide range of responses in secondary pollutants
such as ozone. Ozone response often is "flat" or nonresponsive to reductions of VOCs in many rural areas with significant natural
VOC emissions. Also, ozone can increase in response to increases in NOx emissions in certain localized urban areas.
21 Both the 1990 CAA and EPA's and the National Academy of Science's Section 185B Report to Congress recognized the
consequences of biogenics, transport and the need to conduct regionalized assessments, as reflected in organizational structures such
as the Ozone Transport Commission and the North American Research Strategy for Tropospheric Ozone (NARSTO).
22 Note that aggregating individual urban area.results may reduce the effect of uncertainty in individual city projections (i.e.,
overestimated cities would offset underestimated cities). However, aggregation of individual urban area results would not reduce
potential errors caused by systematic biases which arise due to, for example, misestimated emissions inventories.
-------
Appendix C: Air Quality Modeling
Visibility
Two separate modeling approaches were used to
estimate changes in visibility degradation in the east-
ern and southwestern U.S. These are the two regions
of the coterminous U.S. for which Clean Air Act pro-
grams were expected to have yielded the most sig-
nificant reductions in visibility degradation. Visibil-
ity changes in the eastern 31 states were estimated
based on the RADM/EM results for sulfates; and
changes in visibility in 30 southwestern U.S. urban
areas were calculated using a linear emissions scaling
approach. Despite the potential significance of Clean
Air Act-related visibility changes in southwestern U.S.
Class I areas, such as National Parks, resource limita-
tions precluded implementation of the analysis
planned for these areas.
The RADM/EM system includes a post-proces-
sor which computes various measures of visibility
degradation associated with changes in sulfate aero-
sols.23 The basic approach is to allocate the light ex-
tinction budget for the eastern U.S. among various
aerosols, including paniculate sulfates, nitrates, and
organics. The change in light extinction from sulfates
is provided directly by RADM, thereby reflecting the
complex formation and transport mechanisms asso-
ciated with this most significant contributor to light
extinction in the eastern U.S. Nitrates are not estimated
directly by RADM. Instead, RADM-estimated con-
centrations of nitric acid are used as a surrogate to
provide the basis for estimating changes in the par-
ticulate nitrate contribution to light extinction. The
organic fractions were held constant between the two
scenarios. Standard outputs include daylight distribu-
tion of light extinction, visual range, and DeciViews24
for each of RADM's 80-km grid cells. For the present
study, the RADM visibility post-processor was con-
figured to provide the 90th percentile for light extinc-
tion and the 10th percentile for visual range to repre-
sent worst cases; and the 50th percentile for both of
these to represent average cases. More detailed docu-
mentation of the RADM/EM system and the assump-
tions used to configure the visibility calculations are
presented in the RADM Report (1995).
To estimate differences in control and no-control
scenario visibility in southwestern U.S. urban areas,
a modified linear rollback approach was developed
and applied to 30 major urban areas with population
greater than 100.000.25 For each of the 30 urban cen-
ters, seasonal average 1990 air quality data was com-
piled for key pollutants, including NO2 and PM]0, con-
tributing to visibility degradation in southwestern U.S.
coastal and inland cities. PM,0 was then speciated into
its key components using city-specific annual aver-
age PM1Q profile data. After adjusting for regional —
and for some species, city-specific—background lev-
els, concentrations of individual light-attenuating spe-
cies were scaled linearly based on changes in emis-
sions of that pollutant or a proxy pollutant.26 Using
the same approach used for the 1993 EPA Report to
Congress on effects of the 1990 Clean Air Act Amend-
ments on visibility in Class I areas, light extinction
coefficients for each of these species were then mul-
tiplied by their respective concentrations to derive a
city-specific light extinction budget.27 This process
was repeated for pre-1990 control and all no-control
scenarios by scaling 1990 results by the relative change
in annual county-level emissions of SOx, NO , and
PM. Based on the city-specific light extinction bud-
get calculations, measures for total extinction, visual
range, and DeciView were calculated for each sce-
nario and target year.
Control scenario visibility
Unlike the other air quality conditions addressed
in the present study, modeled visibility conditions are
used as the basis for the control scenario rather than
actual historical conditions. However, like the other
air quality benefits of the historical Clean Air Act, it
is the differences between modeled visibility outcomes
for the control and no-control scenarios which are used
23 A complete discussion, including appropriate references to other documents, of the RADM and RADM/EM modeling
conducted for the present study is presented in the subsection on acid deposition earlier in this appendix.
24 The DeciView Haze Index (dV) is a relatively new visibility indicator aimed at measuring visibility changes in terms of human
perception. It is described in detail in the SAI SW Visibility Report (1994), pp. 4-2 to 4-3. See also Pitchford and Malm (1994) for
the complete derivation of the DeciView index.
25 Complete documentation of the linear scaling modeling, speciation methodologies, spatial allocation of emissions and other
data and assumptions are provided by the SAI SW Visibility Report (1994).
26 For example, sulfate (SO4) concentrations were scaled based on changes in sulfur oxide (SO ) emissions.
27 The term "light extinction budget" refers to the apportionment of total light attenuation in an area to the relevant pollutant
species.
C-25
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
to estimate visibility benefits. Nevertheless, 1990 ab-
solute levels of eastern U.S. visibility predicted by
RADM under the control scenario are presented in
Figure C-16 to provide a sense of initial visibility con-
ditions.
For the southwestern urban areas, 1990 control
scenario annual average light extinction budget, vi-
sual range, and DeciView conditions are listed in Table
C-15. These 1990 results are presented to give the
reader a sense of the initial visibility conditions in
absolute, albeit approximate, terms.
No-control scenario visibility
The no-control scenario visibility results for the
eastern U.S. area covered by RADM are presented in
Figure C-17. No-control scenario 1990 outcomes for
the 30 southwestern U.S. urban areas are presented in
Table C-16.
Summary differences in visibility
DeciView Haze Index
The DeciView Haze Index (dV) has recently been
proposed as an indicator of the clarity of the atmo-
sphere that is more closely related to human percep-
tion than visual range (VR) or total extinction (bext)
(Pitchford and Malm, 1994). It is defined by the equa-
tion:
where:
bcxt= total extinction in inverse megameters
(Mm'1)
This index has the value of approximately 0 when
the extinction coefficient is equal to the scattering
coefficient for particle-free air (Rayleigh scattering)
and increases in value by approximately one unit for
each 10 percent increase in bext. Since the apparent
change in visibility is related to the percent change in
b^j (Pitchford et al., 1990), equal changes in dV cor-
respond to approximately equally perceptible changes
in visibility. Recent research indicates that, for most
observers, a "justnoticeable change" invisibility cor-
responds to an increase or decrease of about one to
two dV units.
Figure C-16. RADM-Predicted Visibility Degradation,
Expressed in Annual Average DeciView, for Poor Visibility
Conditions (90th Percentile Under the Control Scenario.
Figure C-17. RADM-Predicted Visibility Degradation,
Expressed in Annual Average DeciView, for Poor Visibility
Conditions (90th Percentile Under the No-control Scenario.
C-26
-------
Appendix C: Air Quality Modeling
,,,; ;?'''"* ,,,,--'-,>~v - -, °v° ° --<,,'-"/;, -^-'-,' ' , ,,,',->,Vv-< •'/'•'•'***...*. ,; •,',^-'7^-- --$^'""^ %" ~
°°!Tgb'le C-15i ;T^90 Coj&itrdrScef.aj'id'VisibiMty^^1' ' ,Tabfe C-16V 1*990 No-<
Conditions, for- 30 SouthwesterH^tfcS. Cities. , 3t) Soufiilvestera U.S. vo,W,,
>- Forf Collins, CtK'-x
, Greefey, CO '
-'Denver, CO
Colorado, Spriags,
'.puebkto!!!,''.
S s fff ,•. 1 •> -1-
,AlDUQUGKMi6» iSiM
- ELPaso, TX1-, f:'r *
f acson, A2T'S-V •"'-'
,Phowx,A2f'
"flight -,-
',; Budget &-
x^UtX1' J1
' W.ff";
- --,-201.7
-,"',,2p84, '
170,1
,„' 113-3 ,U
-:; 126.9,,;-.,-
»2.8-"
^iM"' '
" ' 2U.7 "'
- >-'14t.8 .
'•^mi'x,
.pirn's '":#
; ;-.;,,,
,- -154,6--;"',
84.?,, ,
' U9>IW.'
" *wC* '
147:4" '
- mKr,,,.
:vi»5 ,
j-5 'v ,
";:" *,Q-^,-,
-'---84;% '
,'! '",15^.4
"?M
-:~ $$& .
•"'*iii.Y
--,-,,-109.3
,,$*<&
~?;%s.3 ,
^ ""Visual ^
' Rattg&i" '"'
tl^)
-VsJiT
' 14.9 .
''T 14,4,,, .
17.6- "
, 26,5 '
---23;,€, -,
,26.6--
-,-13:9"
,%,-14^2 '"
-*-*ao3f --
,«,-, -194
v"":24.8- ^
^ -25.5',:
19.4° """
,,»,3JU""
' 2s,a';',;?-,'
•i&i";'"
"':2o;3':-
., ,:l9.0
J!?25,5 '
?•'* 27;$,,,,,,,,
$7.2
-,--3sir ;
19.6,
, , ,b,,, ,- -
1 ™ 36.0
---34rl"'
..32,9-""
*l3a*
'"Islo-"-
«,^ ?3,9
\;Deci
*SJ
29.8''
,,3b.O
''30.4,,,
at*
,24,3 -
'25x4,
24,2, ,
30,7w,
";,3o,-$;:,
, ,27.0^-
\*&>*
--"24T9""
&'
'' ,27.4.,
,,2i,4,,:
,24,8' '
21,2 '
. 26,9''J
' 2W*A
•m*.
»:-,23,r":
20,9 ;
'21.3 ;»
27,3 **
20' -
21,8—
j S J* '•J V ••
V'23>9"^
-'>2fe5V-
"'"25.3"'
' w\
,> ^
>^ V'
-. A">
y
-; \
>-
-;,
,v-
„
*, -
x. , ,-^
/,>
--.
>'-
,.
^. '
;*• '
j
-^-- -"City -,,:*-;;
"Los-Angeles,CA
^wBefflarfinicA^'""''
Anaheimj CA- --*•• -' -'-v'
Veatur^&A'"""
,,-San Dtego,-'CA
"""Santa Barbara, CA , ',„", ,'™,?,
Bakersfield, CA , , ' '' vv
RB*ao,CA ' '--'=';
•MwiifthC^"^''" --
.... . ,.,. ..^.,. ,^ ,,.^^, ,.
Stocktoa,-'CA^' .'"^'T ".
Saa Fraaoiscd, CA,- ,, ,,;
OaKSark'CA '"''''
S^JoKWC """
-Monterey, CA... . . .. .
*•- SSc'imriepto, CA , -
'• Redding, CA ' >>°:"
jleno, NV • ! '"^"'
- Las -Vegas, NV , - ,-> ,«,-„ >4 ,-
, Salt Lake City, UT
'
Pio vo, UT , , „ • ,, i , -,
PoriCoBfas'CO- •'-
' Cieeley, CO
»£fe*«fi'eO
CoIoradoSprsngs.'CO'
PaelJto,1 CO --.---' "•-----'
-M&Bqn^ueJSM * -,
Bi Paso, TX - -
Tucson, AZ",,, ; --,,.»-;-- -
Phoenix, AZ1,-^-'," ,
\~% '«i'x'.Yveit
'"» '2101 '
,-,,,"' 183.2--.'"-'-, -
, 35,6.4
* 'k>:o"-
' . /&M -- „ ,„
,,-'iv;24&i T;,-.
' ' "" ' 197.3 ' ' 'Xv>*
" ^188.6 --; "' "'
,;,-,•"'" "-253.0, ^%^-
*'y 7?J4L4 "'""^
:^-r" - 189.2^;,^, '
->-' ' -"i2g;S''"" , . -
•!"-, 416.^ v-- '
(„, ,,^43,«,_;,
isir""""
159.0
.- - "191-2
v ^-v?/ ";;;;,;;.,
>-.,.,.- - 284,4 ;,,,,,,,,v,^
VISA*^'-'-
, , •> s s^*
,29^9^ ^' ' ""
'£753'" ';',',"
" :,;27't'^'; -:v.*
1 •> f' s >/ '/•-;•••• %
- --429,5 - - ';:v;
' Wrtft.it ^ ^ ' Jj "
.-Visual,,,'
Range, (ian)1
-,->'" 9:0"
-—-8.9 - -
- • «•?./'
,"',. 16-5 '
<- 15,4
143" ' '' '
1&CC
' " 8.6 '
„- 12.5
1^ 1
1S.2 - --
' 15,9--^
-.-11.9
""-ii^'—'
•'w>'::'
•---'--2^'V"-1
',,,-7,2,--
' ' "4.7-
"''-16.1'"'",
.»>*<§£ „
>x~. IS.%
.,;, 25,6
; Kb***-'
-,",,17!l,', ,
, , , / ^ V. '"' ,
' " ^ , jAQ'Q ,
•^iy.1 -
'""" 10,9
"•' 11,0
^'-7.ff»'">
' ,--,,-,-,\v->
,' DedView,,,,,
-;35.1 „,; ,
,,35.2,,,,.
' ,3^'.4 ' ' ""
-' ---33.5
29.7 ,„ ,
':', ' ' -30.4 „ ,
-'l';atW^
35,7 -v
35^5
q'f S
,jl.O, .,
Aft 1
>, ,,,,At.I
-"-,2?.S '-'
-v „ 29.4
'^'32,3 .
-,-»>,, 26.5,, ,
29" 4' J
25.5" 'vv
37.S
41,6 ,-
,'„ ';29.2' '
, /{••,
•'W295'
24.6
'' ',33.5 *''?,'
28.7
34.0 .,'
28,7 '•
• --33.2- '
33.0
.,n- 37:6-,: _ -
'^- ',' ', 4 *-\, --^--,
,,,!
Z
f "•
*
-?.
>rta Source:. SATSW Vixifiilitv Report <19$i).
C-27
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Both VR and dV are measures of the value of bext
at one location in the atmosphere. Both are unaffected
by the actual variability of the compositions and illu-
mination of the atmosphere, so neither is closely linked
to the human perception of a particular scene. The
isolation of these parameters from site-specific varia-
tions and temporal fluctuations of the atmospheric il-
lumination increases their usefulness for comparing
the effects of air quality on visibility across a range of
geographic locations for a range of time periods. Each
parameter attempts to scale the b data so that changes
in air quality can be used to provide an indication of
changes in the human perception of a scene.
Modeling Results
The differences in modeled 1990 control and
no-control scenario visibility conditions projected by
the RADM/EM for the eastern U.S. are presented in
Figure C-18. The map shows the percent increase in
modeled annual average visibility degradation under
poor conditions for 1990 when moving from the con-
trol to the no-control scenario. The results indicate
perceptible differences in visibility between the con-
trol and no-control scenario throughout the RADM
domain. The relatively large increase in visibility im-
pairment in the Gulf Coast area is a reflection of the
Figure C-18. RADM-Predicted Increase in Visibility
Degradation, Expressed in Annual Average DeciView,
for Poor Visibility Conditions (90th Percentile) Under the
No-control Scenario.
significant increases in 1990 sulfate concentrations
associated with the no-control scenario. (See the ear-
lier discussion of effects in this region in the sections
dealing with acid deposition.)
The differences in modeled 1990 control and
no-control scenario visibility conditions in the 30
southwestern U.S. urban areas projected by linear roll-
back modeling are presented in Table C-17. When
reviewing these visibility degradation differentials for
the 30 southwestern U.S. urban areas, it is important
to consider that while estimated differences in visual
range were in many cases very large, changes in the
DeciView Haze Index (dV) may be relatively small.
This is because the perception of visibility degrada-
tion measured by dV may be small when baseline vis-
ibility is high.28 Even so, the results indicate that, by
1990, visibility in southwestern U.S. urban areas
would be noticeably worse under the no-control sce-
nario.
Key caveats and uncertainties for
visibility
There are several sources of uncertainty in the
RADM and southwestern U.S. linear scaling model
analyses. For RADM, the use of nitric acid as a surro-
gate for estimating changes in light-attenuating ni-
trate particles ignores the interaction effects of ni-
trates, sulfates, and ammonia. As a result, increases
in nitrates may be overestimated by the model when
both sulfates and nitric acid increase. However, the
significance of this potential overestimation is miti-
gated to some extent by the relative insignificance
of nitrate-related visibility degradation relative to
sulfates which prevails in the eastern U.S.
Several important uncertainties in the south-
western U.S. urban area visibility analysis are de-
scribed in detail in the SAI SW Visibility Report
(1994). First, the need to use seasonal average con-
ditions leads to underestimation of extreme visibil-
ity impairment episodes associated with high hu-
midity, since particle growth due to water absorp-
tion is highly nonlinear. Second, although the use
of city-specific light extinction and PM speciation
data is significantly better than reliance on regional
averages, uncertainties in city-specific data may
contribute to overall uncertainty in the estimates.
However, overall uncertainty associated with these
factors will be reduced to some extent since over-
estimation of visibility degradation in some cities
28 See SAI SW Visibility Report (1994), page 5-3.
-------
Appendix C: Air Quality Modeling
will be offset by underestimations in other cities. Fi-
nally, the linear scaling used to estimate the pre-1990
control scenarios and the no-control scenarios was
based on changes in county-wide or air basin emis-
sions. Uncertainties associated with apportionment of
state-wide emission changes to individual counties or
air basins may contribute significantly to overall un-
certainty in the visibility change estimates. Such ap-
portionment is particularly difficult for SOx emission
changes, since emission reductions achieved by the
Clean Air Act tended to be at relatively remote utility
and smelter plants. However, sulfates are a relatively
minor source of light attenuation in western urban
areas.
An important overall limitation of the visibility
analysis conducted for the present study is that only
southwestern urban areas and the eastern 31 states
were included. The Clean Air Act may have contrib-
uted toward significant reductions in visibility degra-
dation in other areas. For example, Clean Air Act pro-
grams to reduce ambient particulate matter may have
motivated reductions in silvicultural burning in some
northwestern states. Perhaps the greatest deficiency
in geographic coverage by the present study is the
omission of visibility changes hi Class I areas in the
west.
'•- f V •> '•>*';":' •O'** '\
,,,,table CflX', J-Stftti
fo'f"
;Cities."
",/-ci«r-r*-' -
,,,*,,,',,
Lois /i}gele$,,CA , ,-,,,-,-<
'- San Betnardino, OA'- •'*""- - -
"Riverside;- GA •'"''"'' " -,-,'='>
;~A>aieim:CA ' ' - ^t-"'"
Ventura, CA , , , ,;,/
San Diegb, CA
- Santa" Barbara, CA , l-tr**'"*
-Bate«iteld;CA:;-'^r* "
,,pftSiict,.CA- - - .-...??x..s"'.f/T^.
Mode^C, CA
StocKtoArCA-'"'"'",' - -
,Sao^tanc!sco,-CA ,,.»--
•-oamSn« CA" ' "'.LL^^
^ San- Jose, CA ,'/>#v>,- s.-x A-
''M- "•'
Sa^roeuto, GA' - ' ' , ,-,»-sx
"R'edflmgi.CA , ,T,,',»f ''" ,"!!,-
« ;-, , -.' v ->
Reno, KV -
LasVegaS/NV, ,", , "
' Salt Lake City, OT '" -'• --
.Prove, UT ,,,,/
PortCQiiins,CO--';-
fteefey, CO
^ ^ >-v ' . "*A; s>'
Denver, CO
' Colorado Sjoings, CCJ' ' ;,
-/"- * " " " V-/
, Pueblo* CO
--^ - - ^^ /^ -^
' - AibuQuef5oe>' I^U^f - ^-* ^ ^
El Paso, -TX >• ••'<">*
- Tuos'on'-AZ ' "" " , '
,,Ffi^^2f- ' ;-;;./
Visu'al Range
®/ -
;*-v«7 '
>*'<" «, ,w-.»
';';,^8^rf
', 72 „
,65 ' '
:\.,:^^
>^«' "
'""ei;""",
62 ', -
- •;- 63 ,,,
- -""iJi",^
<",4»ii-
,;r-,.67-:>--:
.- » -59 - ';
55;,:""
" "183
, . 308 ' • '•
' -58-1^'
„ """!*»'? '.
, „ 137
, ,
-39-'-:*-'--1
- '--- IS
w>- -<" in
,,240'"'"*
, ;:v,93 •
s ' ' „ s
,-!»-- i53,, ,,
V 21S:-> ----
' '243"
- ->-DeoiView"-s-
,-- (dV)
'/ ' ' •• -.-;'''A' '"
«5 - -
v^-s - r
„..***• -
* -":V^>, '
,..-5. ,,";
-5
'•1***X$P ,
"•*/X.""'
* "
-5 •
- - -5 , ,
r5--, >/,,-x
-5
-5
-5- , •
,„„„,,
! ' -4
, -, -, ,-;•>, , , x '
""-i"o ; ,
t ,^i.l4,-,, , ,
^/!r7rv'
• '"-''l4V'
' -&,' , ,
^ ,, ,-. ^
I--"-';" S3*"
-6
, .,.,., ,^-,«
-*»".J-2,: -
" --7
'ft ,,, ,
-12
,...xaw~
t£k&JSflOiU^ SAI SW yisibjlit^R&port (1),
"• ^"~- %' " ~"
C-29
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Air Quality Modeling References
Chang. 1995. SARMAP Air Quality Model (SAQM).
Final report to San Joaquin Valley wide Air
Pollution Study Agency.
DaMassa, Tanrikulu, and Ranzier. 1996. Photochemi-
cal Modeling of August 3-6, 1990, Ozone
Episode in Central California Using the
SARMAP Air Quality Model. Part II: Sensi-
tivity and Diagnostic Testing. Preprints, Ninth
Joint Conference on the Applications of Air
Pollution Meteorology with Air Waste Man-
agement Association. January 28 - February
2,1996, Atlanta, Georgia.
Dennis, R. 1995. Estimation of Regional Air Quality
and Deposition Changes Under Alternative
812 Emissions Scenarios Predicted by the
Regional Acid Deposition Model, RADM.
Draft Report for U.S. Environmental Protec-
tion Agency, ORD/NERL. October 1995.
ICF Kaiser/Science Applications International. 1996.
PM Interpolation Methodology for the Sec-
tion 812 Retrospective Analysis. Memoran-
dum from J. Langstaff to Jim DeMocker.
ICF Kaiser/Systems Applications International. 1994.
Retrospective Analysis of the Impact of the
Clean Air Act on Urban Visibility in the South-
western United States. Final Report.
ICF Kaiser/Systems Applications International. 1995.
Retrospective Analysis of Ozone Air Quality
in the United States. Final Report.
ICF Kaiser/Systems Applications International. 1992.
Retrospective Analysis of Paniculate Matter
Air Quality in the United States. Draft Re-
port.
ICF Kaiser/Systems Applications International. 1995.
Retrospective Analysis of Paniculate Matter
Air Quality in the United States. Final Report.
ICF Kaiser/Systems Applications International. 1994.
Retrospective Analysis of SO, NO and CO
Air Quality in the United States. Final Report.
ICF Resources Incorporated. 1992. Results of Retro-
spective Electric Utility Clean Air Act Analy-
sis -1980, 1985, and 1990. September 30.
Pitchford, Marc L. and William C. Malm. 1994. "De-
velopment and Applications of a Standard
Visual Index." Atmospheric Environment, vol.
28, no. 5. pp. 1049-1054.
Seaman and Stauffer. 1995.Development and Design
Testing of the SARMAP Meteorological
Model. Final report to San Joaquin Valley
wide Air Pollution Study Agency.
Seaman, Stauffer, and Lario-Gibbs. 1995. "A Multi-
Scale Four Dimensional Data Assimilation
System Applied in the San Joaquin Valley
During SARMAP. Part I: Modeling Design
and Basic Performance Characteristics."Jour-
nal of Applied Meteorology. Volume 34. In
press.
Tanrikulu, DaMassa, andRanzieri. 1996. Photochemi-
cal Modeling of August 3-6,1990 Ozone Epi-
sode in Central California Using the
SARMAP Air Quality Model. Part I: Model
Formulation, Description and Basic Perfor-
mance. Preprints. Ninth Joint Conference on
the Application of Air Pollution Meteorology
, with Air Waste Management Association.
January 28 - February 2,1996. Atlanta, Geor-
gia.
Trijonis. 1990. Visibility: Existing and Historical
Conditions—Causes and Effects. NAPAP Re-
port 24. 1990.
U.S. Environmental Protection Agency (EPA). 1989.
Procedures for Applying City-Specific EKMA.
EPA-450/4-89-012. Office of Air Quality
Planning and Standards.
C-30
-------
Appendix D: Human Health and Welfare Effects
of Criteria Pollutants
Introduction
In responding to the mandate of section 812, EPA
conducted a comprehensive benefits analysis to iden-
tify and estimate the quantifiable health and welfare
benefits enjoyed by Americans due to improved air
quality resulting from the CAA. Health benefits re-
sulted from avoidance of air pollution-related health
effects, such as mortality, respiratory illness, and heart
disease. Welfare benefits accrued where improved air
quality averted damage to ecological health and mea-
surable resources, such as agricultural production,
building materials, and visibility.
This appendix presents an overview of EPA's
approach for modeling human health and welfare ef-
fects. It provides an outline of the principles used to
guide the benefits analysis, details methods used to
quantify criteria air pollutant exposure nationwide
across the study period (1970 to 1990), and discusses
several critical conceptual and implementation issues
for using health and welfare effect information. Mod-
eling results, estimates of avoided incidences of ad-
verse health and welfare effects, are then presented.
Ecological and agricultural benefits are examined in
more detail in Appendices E and F, respectively. Ap-
pendix I details the approach used to translate health
and welfare effects into monetary benefits.
Principles for the Section 812
Benefits Analysis
Estimating the effects of even modest shifts in
environmental releases involves complex chemical,
environmental, biological, psychological and eco-
nomic processes. The task of estimating the broad
changes associated with adoption and implementation
of the Clean Air Act challenges the limits of scien-
tific knowledge and modeling capability to synthe-
size available information and techniques into a prac-
tical framework. A pragmatic plan for a comprehen-
sive assessment must fairly reflect the complexities
and uncertainties, but still produce a policy-relevant
analysis in a timely fashion. In order to achieve this
ambitious goal, the following principles have been
used to guide the section 812 benefits assessment.
Comprehensiveness: The assessment should in-
clude as many benefit categories as are reasonably
believed to be affected by implementation of the Clean
Air Act. Comprehensiveness requires assessing effects
with which greater levels of scientific confidence are
associated, as well as less well-understood effects. The
degree of relative certainty among effects must be
carefully described hi order to fairly present a broad
portrayal of the physical and social benefits accruing
to the nation from implementing the Act. In addition,
section 812 of the 1990 CAA Amendments explicitly
directs a comprehensive benefits coverage that pro-
hibits a default assumption of zero value for identi-
fied benefits unless a zero value is supported by spe-
cific data.
Quantification Where Feasible: The central goal
of the present study is to evaluate and compare the
benefits and costs of historical CAA-related programs.
Effective comparison of the variety of human health,
welfare, and ecological benefits with the associated
compliance costs requires that these consequences be
measured in terms of a common metric. Expressing
the value of these various effects in economic terms
is the most efficient way to accomplish this objec-
tive, and is consistent with standard practices associ-
ated with economic benefit-cost analysis. Expressing
these effects in economic terms requires quantifying
and presenting estimated effects in both physical and
monetized economic terms. Pursuant to this paradigm,
the emphasis in the present study is largely on cat-
egories having direct and perceptible effects on hu-
man health. That is, the emphasis of the analysis is on
categories such as symptoms and diseases rather than
on physical changes (such as cell level changes) that
do not directly result in a decreased health status no-
ticeable to the individual.
D-l
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Efficient Use of Previous Research Results: Sig-
nificant research effort has been spent to understand
and quantify the complex relationships between air
pollution and human health. The present study has
relied as much as possible on available research re-
sults, making adjustments as necessary to apply the
existing results to the current analysis.
Incorporate Uncertainty: To properly convey the
results of any benefits assessment, it is important to
include an evaluation and characterization of how
much confidence the analysts have in the estimates.
Ideally this would include a formal quantitative as-
sessment of the potential for error, and the sources,
directions, and potential significance of any resultant
biases. A method for considering and reporting un-
certainty must be built into the fundamental design of
the assessment. Such a framework was developed and
applied in the present study, and was supplemented
where necessary by expert judgment regarding the
sources and potential significance of errors in each
analytical step.
General Modeling Approach
Consistent with these principles, the EPA devel-
oped an approach for quantifying the effects of re-
duced pollutant exposure, with particular focus on
those effect categories for which monetary benefits
could be estimated. As described previously, the study
design adopted for the section 812 assessment links a
sequence of analytical models. The macroeconomic
modeling (Appendix A) estimated economy-wide ef-
fects of CAA expenditures. These effects provided a
basis for the modeling of criteria pollutant emissions
under the two scenarios considered (the factual con-
trol scenario and the hypothetical no-control scenario),
as documented hi Appendix B. The emissions esti-
mates were used as input to the air quality models
(Appendix C). Ambient pollutant concentrations es-
timated by the air quality models.were used as inputs
to the health and welfare benefits model, the focus of
this appendix.
The approach developed to model health and wel-
fare benefits is known as a "reduced form" or "em-
bedded model" approach. The concept of a reduced
form model is to use simplified versions of previously
constructed complex models to characterize the im-
pact of a series of linked physical and socioeconomic
processes. The health and welfare benefits model is
characterized as a reduced form model because it re-
lies on summaries of the data output from the air qual-
ity models, which rely on emissions summaries and
summaries of macroeconomic conditions, succes-
sively. Although results of the independent models
are used in series, the models themselves have not
been integrated into the health and welfare benefits
model.
In general, the reduced form health and welfare
benefits model relies on two fundamental inputs: (1)
nationwide changes in pollutant exposures across the
study period, and (2) the association between changes
in exposure and expected changes in specific health
and welfare effects. These inputs are discussed be-
low.
Quantifying Changes in Pollutant
Exposures
Estimating changes hi pollutant exposures re-
quires characterization of nationwide air quality im-
provements across the study period, as well as the
populations exposed to the different levels of improve-
ment.
Air Quality
As discussed in Appendix C, the section 812
analysis estimated ambient concentrations for both the
control and no-control scenarios for the following
pollutants and air quality parameters:
• Particulate matter, less than 10 microns in
diameter (PM10)
• Ozone (O3)
• Nitrogen dioxide (NO2)
• Sulfur dioxide (SO2)
• Carbon monoxide (CO)
• Visibility measures (light extinction and
DeciView)1
• Lead (Pb)
Generally, this analysis adopted actual historical
air pollution monitoring data to represent control sce-
nario air quality. No-control scenario profiles were
1 While the visibility measures listed are not criteria air pollutants, they provide important measures of a significant welfare
effect resulting from air pollution, visibility degradation. Light extinction (which is related to DeciView, a haziness index) results
from light scattered by fine particles in the atmosphere, especially sulfates and ammonium nitrates. As atmospheric concentrations of
such particles increase, light is attenuated and visibility diminishes.
__
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
derived by running the control and no-control scenario
emissions inventories through a suite of air quality
models and then using the differences in these mod-
eled outcomes to adjust the historical profiles. Since
lead was treated differently than the other pollutants,
the analysis of the CAA impacts on atmospheric lead
concentrations is documented in Appendix G.
With respect to the distribution of air quality data
across the two decades considered, it should be noted
that both the number and location of monitors track-
ing air quality changed over tune. Table D-l depicts
the number of monitors for each pollutant across the
period of this analysis. The number of monitors gen-
erally increased throughout the 1970s and leveled off
or declined at varying points during the 1980s, de-
pending on the pollutant.
Tabje D-l. Criteria Air Pollutant Monitors
ir toe u.
Yea? -
1970
- 1975
- 1980
1985
1~990
a., iy-/u - 1.
-_ __
„ " V--**"
; ,; EM,V v
1,1 2ft
1,131
970
; , ,"/ 7»>, x -
'""*" . ,t,- •/->-_ ~
-- ,,/, ,
Pollutant
-'£b-'-'M,v '^
i ",',',fl.J- »«,
5« ', V-J375 - 1,088
527 ,^'*305,v „ 916-
- «z7"-',^«" *',,?53->
01'
1 '82-
'511
' 458-
, 493
For the section 812 modeling, the non-lead pol-
lutants have been characterized as either county-level
or monitor-level pollutants. The distinction was im-
portant for quantifying the population exposed to dif-
ferent levels of air quality improvements, as discussed
below. PM10 is considered a county-level pollutant,
since historical concentrations in monitored counties
have been synthesized into a single concentration for
each, county.2 In contrast, O3, NO2, NO, SO2, and CO
were reported at specific monitor locations, given by
latitude/longitude coordinates. Finally, visibility was
treated as a county-level pollutant in the western U.S.
and a monitor-level pollutant in the eastern U.S.3 Air
quality data for PM10 and ozone were reported for each
year of the study period; data for the remaining pol-
lutants were reported only for 1975, 1980, 1985, and
1990.
In order to reduce the volume of air quality data
necessary to describe pollutant concentrations for two
scenarios nationwide over twenty years, annual con-
centration profiles were reduced to frequency distri-
butions. That is, annual pollutant concentrations for a
variety of averaging times (e.g., 1-hour, 6-hour, daily)
were summarized as a distribution of values across
the year. This approach reduced data management
requirements significantly, while adequately captur-
ing air quality improvements between the control and
no-control scenarios.
Population Distribution
. Health and some welfare benefits resulting from
air quality improvements are distributed to popula-
tions in proportion to the reduction in exposure each
enjoys. Predicting population exposures, then, is a
necessary step in estimating health effects. Doing so
for the section 812 analysis required not only an un-
derstanding of where air quality improved as a result
of the CAA, but also how many individuals were af-
fected by varying levels of air quality improvements.
Thus, a critical component of the benefits analysis
required that the distribution of the U.S. population
nationwide be described in a manner compatible with
the air quality data. Described below is the method
used to allocate U.S. Census data to a symmetrical
grid overlying the country.
Census Data
Three years of U.S. Census data were used to rep-
resent the geographical distribution of U.S. residents:
1970,1980, and 1990. Population data were supplied
at the census block group level, with approximately
2 Two different measures of ambient concentrations of particulate matter were used in the United States during the period 1970
to 1990. Prior to 1987, the indicator for the National Ambient Air Quality Standard for PM was total suspended particulates (TSP). In
1987, the indicator was changed to PM(() (particles less than 10 uM in diameter). Widespread PM10 monitoring did not begin until
1985; prior to that only TSP data is available. Because the recent scientific literature reports primarily the relationship between PM10
and adverse health and welfare effects, PM10 data is preferred, if available. Where only TSP is available, PMIO concentrations were
estimated using PM]0:TSP ratios that vary by area of the country and the urban/rural characterization of the area.
3 In the western U.S., visibility was modeled using a linear-rollback model and extinction budget approach for 30 major urban
centers (SAI, 1994). The modeling results, reported in DeciView, were applied to the counties in the vicinity of the urban centers and
considered to share a common air basin. In the eastern U.S., Regional Acid Deposition Model (RADM) runs provided visibility
estimates in terms of light extinction coefficients. These were modeled across a 60 km. X 60 km. grid, approximately covering the
eastern half of the country. Since the extinction Coefficients were reported at the grid cell centroids, for which the coordinates were
known, visibility in the east was treated as a monitor-level pollutant.
_
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
290,000 block groups nationwide. Allocating air qual-
ity improvements to the population during intermedi-
ate years necessitated interpolation of the three years
of population data. Linear interpolation was performed
at the block group level in order to preserve the vari-
ability in growth rates throughout the country.
Gridding U.S. Population
To ease computational burden, block group popu-
lation estimates were aggregated to a rectangular grid
structure. The grid, comprised of ten kilometer by ten
kilometer gridcells, spanned the entire area of the con-
tentional United States. This grid size generated
46,885 populated gridcells throughout the U.S.
The entire population of each block group was
assumed to reside at the geographical centroid of the
block group area, the coordinates of which were avail-
able from the U.S. Bureau of the Census. Block group
populations were aggregated to gridcells according
to the block group centroids encompassed by each cell.
In addition to the population of each gridcell, the state
and county names for each gridcell were retained,
permitting aggregation of data at the state and county
level, as well as nationwide.
Allocating Exposure Estimates to the Population
Two alternative modeling strategies we're used to
allocate air quality improvements to the U.S. popula-
tion. They differed in terms of both the certainty of
the estimates and the geographic coverage:
Table D-2. Population Coverage in the 4..B%
,,7l.S%
„ ?$?$
„ 73.0&
„„ 100% -
109(1-
70.4%
72,2%
61,5%
74,4%
67.$%
70,6%
* 100%
Method One
Air quality improvements (difference between
control and no-control scenarios) were applied to in-
dividuals living in the vicinity of air quality monitors.
For pollutants with monitor-level data, it was assumed
that the individuals in a gridcell were exposed to air
quality changes estimated at the nearest monitor, as
long as the monitor was within 50 kilometers. Like-
wise, for PM10 (for which data was available at the
county level) the population of each monitored county
was assumed to be exposed to the air quality changes
reported for that county.4 The remainder of the popu-
lation was excluded from the analysis.
Unfortunately, by limiting the quantitative analy-
sis to populations within 50 km of a monitor (or within
a monitored county, for PM), a significant portion of
the U.S. population was left out of the analysis (see
Table D-2). For most pollutants in most years (ex-
cepting lead), less than three-quarters of the popula-
tion lived within 50 km of a monitor (or within a PM-
monitored county). Clearly, an analysis that excluded
25 percent of the population from the benefits calcu-
lations (thus implicitly assuming that the CAA had
no impact on that population) would understate the
physical effects of the CAA. Conversely, ascribing
air pollution reduction benefits to persons living great
distances from air quality monitors is a speculative
exercise, and could overstate benefits.
Method Two
As an alternative modeling strategy, air quality
improvements were applied to almost all individuals
nationwide. Where monitor data were not available
within 50 kilometers, data from the closest monitor,
regardless of distance, were used. Similarly, PM10
concentrations were extrapolated using regional air
quality models to all counties (even those for which
monitoring data was unavailable) and applied to the
populations of those counties.
Although subject to less certain air quality data,
the second alternative extrapolates pollutant exposure
estimates to almost the entire population using the
, closest monitoring data available (see Table D-3).5
This second alternative was chosen as the preferred
approach hi the benefits analysis. The sensitivity of
4 Since the lead (Pb) analysis, which was handled separately from that of the other criteria pollutants, did not require air quality
modeling data, the issue of proximity to monitors is irrelevant. The Pb analysis extended to 100 percent of the population.
5 While this alternative captures the vast majority of the U.S. population, it does not model exposure for everyone. To improve
computational efficiency, those gridcells with populations less than 1,000 were not modeled; these cells account for less than five
percent of the U.S. population.
D-4
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
the benefits estimate to the extrapolation of air qual-
ity data beyond monitored areas is explored in Ap-
pendix I.
'•Table D-^.,ii
,"-f$Extr^ps6l|ted to Aii;tfc|lf Model Ru'ns,(ggrc'
s
CO,
747%
SO* ' '^ 91.4$>
-v> v,— ""-
.100% ; 'joQ%.'L::"'
Estimating Human Health Effects
of Exposure
It is impossible to estimate all of the physical ef-
fects that would have occurred without the Clean Air
Act. While scientific information is available that
makes it possible to estimate certain effects, many
other, potentially very important, health and welfare
effects cannot be estimated at this time. Other physi-
cal effects can be quantified, but it is impossible to
assess the economic value of those endpoints based
on the current economics literature. Table D-4 shows
the health and welfare effects for which quantitative
analysis has been prepared, as well as some of the
health effects that have not been quantified in the
analysis.
In order to translate the reductions in pollutant
exposure estimated to result from the CAA into health
benefits, it is necessary to quantify the relationship
between such exposures and adverse health effects.
As indicated below, this analysis relies on concentra-
tion-response relationships published in the scientific
literature which provide estimates of the number of
fewer individuals that incur an adverse health effect
per unit change hi air quality. Such relationships are
combined with the air quality improvement and popu-
lation distribution data to estimate changes in the in-
cidence of each health endpoint. By evaluating each
concentration-response function for every gridcell
throughout the country, and aggregating the resulting
incidence estimates, it was possible to generate na-
tional estimates of avoided incidence.
It should be noted that a slightly different approach
was used to compute health effects associated with
exposure to gasoline lead. Instead of relating health
outcomes to ambient pollutant concentrations, the
concentration-response functions for lead-induced
effects link changes in health effects directly to
changes in the population's mean blood lead level.
This value is directly related to the concentration of
lead hi gasoline in a particular year. Appendix G docu-
ments both the methods used to characterize mean
blood lead levels and the approach for estimating hu-
man health effects from lead exposure.
The discussion below outlines the types of health
studies considered for this analysis, and issues criti-
cal to selecting specific studies appropriate for use in
the section 812 context. Next, details regarding use of
the results of the studies are explored. Finally, the
concentration-response functions used to model health
benefits from reductions in non-lead criteria pollut-
ants are outlined.
Types of Health Studies
Scientific research about air pollution's adverse
health impacts uses a broad array of methods and pro-
cedures. The research methods used to investigate the
health effects of air pollution have become consider-
ably more sophisticated over time, and will continue
to evolve in the future. This progress is the result of
better available research techniques and data, and the
ability to focus further research more sharply on key
remaining issues based on the contributions of earlier
work.
The available health effects studies that could
potentially be used as the basis of the section 812 as-
sessment are categorized into epidemiology studies
and human clinical studies. Epidemiological research
in air pollution investigates the association between
exposure to air pollution and observed health effects
in the study population. Human clinical studies in-
volve examination of human responses to controlled
conditions in a laboratory setting. Research has been
conducted on health effects from exposure to pollu-
tion using each approach, and studies using these tech-
niques have been considered in various formal regu-
latory proceedings. Each type of study (as it is used
D-5
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table D-4. Human Health Bffects-of Crtteria-l'qlluta^ts,
Pollutant
Participate Matter/
ISP/ Sulfatcs
Carbon Monoxide
Nitrogen Oxides
Sulfur Dioxide
Lead
' AV,»^f- - ,' ,
Quantified Health Effects , •
, ^ * , „ vw^ie' *<*>•>
Mortality* * * > ^ "s S*V
Respiratory symptoms
Minor restricted activity.days
Respiratory restricted activity,,^,.,,
days
Hospital admission^" >;'"" ''' ' '
Asthma attacks - '
Changes in pu toonary fonetiQn
Chronic S inusitis & Hay. Sever
','-•<•
Mortality*
Bronchitis - Chronic and Acate -
Hospital admissions ' "• •"'•>
Lower respiratory illness • , • v
Upper respiratory Bkess^ ^ , s ,
Chest illness^ ,jiv j,,°,,,^
Respiratory symptoms' • • "
Minor restricted- activity days
All restricted activity (fays \ '* 'J,
Moderate or worse asttenastajg.^ s.
(&sthiti.&tiC*s) * * N
Hospital Aamissions^**1^ "*'
congestive hSte'faiiiite5'''''"'
Decreased tiroe-ttfoHsetWIogm*
Respiratory illness N
.-, -•> i
,/ '&&. >^M ^ f
fix exercising asthmatics: ,„ >„,
Changes in pnlmonaiy function
Respiratory symptoms' v>'f"
Combined respQnWdf5$V
respiratory' symptotas and "\~
poteonary fiujctioa change's^
Mortality "' ' "'''"' v'
HyperteasiDn"?; *^ "^ '," '^ '\ *
Non-fttsl coronary heartdisease
Kon-feal stroke^ 5 \*?£^ ^
IQ loss effect OB li^ime earnings
lOlosseffectsonspecial
eduction **^jjsVhV's -
•.w^ "• >V-> \ ••
S -,'"'
^nq»a»tttieaHe«MJi Effects' '"
' fo'stiiattli>"
, Centibacinarfibrosis- .'/T'H - ' ^
vlnflaromatiolx'in the lung
' - !„' ', ' """ "•
•f , , >' -
"^ ^ ± ' i'/"1 \
^ '','#,,, <
''^'v"V^?x ' '''*'*'!'',,
-I /.;*• '
-A--, , '-•"• '-',
"Changes irfpulmoiiajy &nction
x" ^ "w:*,, -
' :-"<•>„'
^ ?><'' "
f ' "/„ * /
v j ?/*/' ,
.. ^ '^' i^j^A'^ '* 0",
*#%*''•:'> „ •*• ,,.,,,, „„,„
/" ^'*'^V^" ^.>vSs^v^V^ %' *
, •'• ,v<-*;^:-%,y ,;,,,/->->;;,, -
Be,ha5*oj;aJ,«fR}cts ,
^the^jiospit^l admissions ' -
, ^ ^ '^--^//^vv^ ,
, Increased airvirw responsiveness ~.,;
v w^*--r, •- J r
s\o^ ^ ^ ' * ^^'''" ''^4^ ^ ^ y't/iy;
^ ' ki-f, '-„*,„ - -
"*" -?»?« „ ' , °"
i?IVh*-\
. . f, ., ,? - -.\ f ;
''Heaifli etfects forindHdiiais in ,
'ajgerfinges Otherthan those
studied ,^,,.^ -„-;,-----
Neurobehavioral function
Other^canliovascular-diseages ' '''"-
< IReproductiye effects '" '' *
^Fefel.effects frprnmatesrnal' •"•*- '
;,' ^xpostue , ,w,.,.,,,,,,,,,
^fijeliiSijuefttaad anti-sO?iaj, v
°^*'*' tieHavioriincbildirej}^,,,
Otner,PessiWeH!ects
Chronic,respiratory,dise,3;|e|,v^
,^x^rapidinonary effects (ag,,'
',""" ' changes in sfrttetwe, ' 'si*?
ftinc^otysfothe^Wgans) ' "
• • A ^'^/;'y--iJ^ ',Vf?*,,,,, ,,,
; ^,«/;,^4<_;,;,;/?>,!;,,, ^, ,„ ,
' \<^ '<<(*> .jv, ^ >f ,
' CKronic respiratory, diseases
- ,other,th,a(5. c|f o,mc ^roophitis
'Inflammation in the^liSiB' /r /
' '^ ,'.' , * ^ * , , S"'**'^ <;/f,<
"'- '/'^B *^-,,.;-\ ;i>,, - - , --
*'<* <••'*''> ,w *,>!,
-<\ *!f •./
"''""'"'' '-, -t,--; /,-, ,-,< , ,,
•. , i' >'t'-,/ , , s~"? t, _,
Otfeer'cajSiovascular'el&ctS' , ',',,,
"Deyelopiaental e&ctsx'-»
'?*?»* '<; V* " v^-S , , ,, ,, , , ,
.;Decreased,pu.lmoaary fiincti'olf '
Jnflamination to the tang , '" '', /v
>-ImcStinord*gfcal-'cbanges _ _ ,
X-:: -:;;:-? -v-;\ .--,,
Respiratory symptoms t» oon*
asthmatics ''' ' &$#;
• Hospital admissions/ - ,., v , „
'f'f
',"',''
f
'* ""/"M,
'' ', , ' , f/ffs- I, X''" " ' ""
-'-"' /' "-'A;« re •„«;••-•
/*;*»'•
• '• '" ,'•„ ,'" /,'„ -, - ,
"•»,.* .*-,?-,-•--«;-, , , ' "s""- ,
" V, <,',/.,, - ,rvv, !*''SsJ1,
* This analysis estimates excess mortality a'sii&'PM i05as,an indicator,of tlie pollataftt mix to wliicfi-
^ „,„* .!!,.<„ }Sfs-mf<-,~-V;lfZsV:n%frt!/^3 o? "< XT- . . ' -V,/, „>-,, «T -*-,<.
individuals were exposed., '
for air pollution research) is described below, and the
relative strengths and weaknesses for the purposes of
the section 812 assessment are examined.
Epidemiological Studies
Epidemiological studies evaluate the relationship
between exposures to ambient air pollution and health
effects in the human population, typically in a "natu-
ral" setting. Statistical techniques (typically variants
of multivariate regression analysis) are used to esti-
mate quantitative concentration-response (or expo-
sure-response) relationships between pollution levels
and health effects.
Epidemiology studies can examine many of the
types of health effects that are difficult to study using
a clinical approach. Epidemiological results are well-
suited for quantitative benefit analyses because they
provide a means to estimate the incidence of health
effects related to varying levels of ambient air pollu-
tion without extensive further modeling effort. These
estimated relationships implicitly take into account
at least some of the complex real-world human activ-
ity patterns, spatial and temporal distributions of air
pollution, synergistic effects of multiple pollutants and
other risk factors, and compensating or mitigating
behavior by the subject population. Suspected rela-
tionships between air pollution and the effects of both
D-6
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
long-term and short-term exposure can be investigated
using an epidemiological approach. In addition, ob-
servable health endpoints are measured, unlike clini-
cal studies which often monitor endpoints that do not
result in observable health effects (e.g. forced expira-
tory volume). Thus, from the point of view of con-
ducting a benefits analysis, the results of epidemio-
logical studies, combined with measures of ambient
pollution levels and the size of the relevant popula-
tion, provide all the essential components for associ-
ating measures of ambient air pollution and health sta-
tus for a population in the airshed being monitored.
Two types of epidemiological studies are consid-
ered for dose-response modeling: individual level
cohort studies and population level ecological stud-
ies. Cohort-based studies track individuals that are
initially disease-free over a certain period of time, with
periodic evaluation of the individuals' health status.
Studies about relatively rare events such as cancer
incidence or mortality can require tracking the indi-
viduals over a long period of time, while more com-
mon events (e.g., respiratory symptoms) occur with
sufficient frequency to evaluate the relationship over
a much shorter time period. An important feature of
cohort studies is that information is known about each
individual, including other potential variables corre-
lated to disease state. These variables, called con-
founders, are important to identify because if they are
not accounted for in the study they may produce a
spurious association between air pollution and health
effect.
A second type of study used in this analysis is a
population-level ecological study. The relationship
between population-wide health information (such as
counts for daily mortality, hospital admissions, or
emergency room visits) and ambient levels of air pol-
lution are evaluated. One particular type of ecologi-
cal study, time-series, has been used frequently in air-
pollution research. An advantage of the time-series
design is that it allows "the population to serve as its
own control" with regard to certain factors such as
race and gender. Other factors that change over time
(tobacco, alcohol and illicit drug use, access to health
care, employment, and nutrition) can also affect health.
However, since such potential confounding factors are
unlikely to vary over time in the same manner as air
pollution levels, or to vary over periods of months to
several years in a given community, these factors are
unlikely to affect the magnitude of the association
between air pollution and variations in short-term
human health responses.
Drawbacks to epidemiological methods include
difficulties associated with adequately characterizing
exposure, measurement errors in the explanatory vari-
ables, the influence of unmeasured variables, and cor-
relations between the pollution variables of concern
and both the included and omitted variables. These
can potentially lead to spurious conclusions. However,
epidemiological studies involve a large number of
people and do not suffer extrapolation problems com-
mon to clinical studies of limited numbers of people
from selected population subgroups.
Human Clinical Studies
Clinical studies of air pollution involve exposing
human subjects to various levels of air pollution in a
carefully controlled and monitored laboratory situa-
tion. The physical condition of the subjects is mea-
sured before, during and after the pollution exposure.
Physical condition measurements can include general
biomedical information (e.g., pulse rate and blood
pressure), physiological effects specifically affected
by the pollutant (e.g., lung function), the onset of
symptoms (e.g., wheezing or chest pain), or the abil-
ity of the individual to perform specific physical or
cognitive tasks (e.g., maximum sustainable speed on
a treadmill). These studies often involve exposing the
individuals to pollutants while exercising, increasing
the amount of pollutants that are actually introduced
into the lungs.
Clinical studies can isolate cause-effect relation-
ships between pollutants and certain human health
effects. Repeated experiments altering the pollutant
level, exercise regime duration and types of partici-
pants can potentially identify effect thresholds, the
impact of recovery (rest) periods, and the differences
in response among population groups. While cost con-
siderations tend to limit the number of participants
and experimental variants examined in a single study,
clinical studies can follow rigorous laboratory scien-
tific protocols, such as the use of placebos (clean air)
to establish a baseline level of effects and precise
measurement of certain health effects of concern.
There are drawbacks to using clinical studies as
the basis for a comprehensive benefits analysis. Clini-
cal studies are appropriate for examining acute symp-
toms caused by short-term exposure to a pollutant.
While this permits examination of some important
health effects from, air pollution, such as
bronchoconstriction in asthmatic individuals caused
by sulfur dioxide, it excludes studying niore severe
D-7
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
effects or effects caused by long term exposure. An-
other drawback is that health effects measured in some
well-designed clinical studies are selected on the ba-
sis of the ability to measure precisely the effect, for
example forced expiratory volume, rather than a larger
symptom. The impact of some clinically measurable
but reversible health effects such as lung function on
future medical condition or lifestyle changes are not
well understood.
Ethical limits on experiments involving humans
also impose important limits to the potential scope of
clinical research. Chronic effects cannot be investi-
gated because people cannot be kept in controlled
conditions for an extended period of time, and be-
cause these effects are generally irreversible. Partici-
pation is generally restricted to healthy subjects, or at
least to exclude people with substantial health condi-
tions that compromise their safe inclusion in the study.
This can cause clinical studies to avoid providing di-
rect evidence about populations of most concern, such
as people who already have serious respiratory dis-
eases. Ethical considerations also limit the exposures
to relatively modest exposure levels, and to examin-
ing only mild health effects that do no permanent dam-
age. Obviously for ethical reasons human clinical evi-
dence cannot be obtained on the possible relationship
between pollution and mortality, heart attack or stroke,
or cancer.
One potential obstacle to using dose-response in-
formation from clinical research methods hi a ben-
efits assessment is the need for an exposure model.
The dose-response functions developed from clinical
research are specific to the population participating
in the study and the exposure conditions used in the
laboratory setting. It is therefore difficult to extrapo-
late results from clinical settings to daily exposures
faced by the whole population. For example, many
clinical studies evaluate effects on exercising individu-
als. Only a small portion of the population engages in
strenuous activity (manual labor or exercise) at any
time. Reflecting these fundamental differences be-
tween the laboratory setting and the "real world" im-
poses a formidable burden on researchers to provide
information about human activity patterns, exercise
levels, and pollution levels. This requirement adds an
additional step in the analytical process, introducing
another source of uncertainty and possible error.
To apply the clinical results to model the general
population, two decisions must be made. First, how
far can the conditions hi the clinical setting be ex-
panded? For example, if the subjects in the clinical
study were healthy male college students, should the
results be applied to the entire population, including
children? Second, how many people in the general
population are exposed to conditions similar to those
used in the clinical setting? Frequently, clinical stud-
ies are conducted at relatively high exercise levels (in-
creasing the dose, or the quantity of pollutants actu-
ally delivered to the lungs). In the general population
few people experience these conditions very often,
and people do not reach these exercise levels with
equal frequencies during the day and night.
In addition, the analyst must determine the num-
ber of people that are exposed to the levels of ambient
conditions seen hi the laboratory. Air quality varies
throughout a city and is typically reported by data from
monitors located at various places throughout the city.
However, people are not exposed to the conditions at
any one monitor all day. As people move around in
the city, they are exposed to ambient air quality con-
ditions represented by different monitors at different
times during the day. To further compound the prob-
lem, air quality also varies between indoors and out-
doors, within a car or garage, and by such factors as
proximity to a roadway or major pollution source (or
sink). The exposure model must account for the am-
bient conditions in the "microenvironments'' that the
population actually experiences.
The issues of study subjects, exercise and mi-
croenvironments can influence the choice of clinical
studies selected for the section 812 assessment. Clini-
cal studies that use exposure regimes and exercise lev-
els more similar to what larger groups of the popula-
tion see are easier to apply in a benefits model than
are more narrow studies. Similarly, studies that use a
diverse group of subjects are easier to apply to the
general population than are more narrow studies.
Given the major advantages of epidemiological
studies—exposures do not need to be modeled and
health effects are observed in a large, more heteroge-
neous population—epidemiological studies are used
as the basis for determining the majority of health ef-
fects and dose-response curves. The diverse activity
patterns, microenvironments, and pollution levels are
already considered in the aggregate through the con-
centration-response functions derived from epidemio-
logical studies. Clinical studies are used if there are
health effects observed hi clinical studies not observed
in epidemiological studies.
D-8
-------
Appendix D: Human. Health and Welfare Effects of Criteria Pollutants
Issues in Selecting Studies To Estimate
Health Effects
A number of issues arise when selecting and link-
ing the individual components of a comprehensive
benefits analysis. The appropriate procedure for han-
dling each issue must be decided within the context
of the current analytical needs, considering the broader
analytical framework. While more sophisticated or
robust studies may be available in some circumstances,
the potential impact on the overall analysis may make
using a simpler, more tractable approach the pragmatic
choice. In considering the overall impact of selecting
a study for use in the section 812 assessment, impor-
tant factors to consider include the likely magnitude
the decision will have on the overall analysis, the bal-
ance between the overall level of analytical rigor and
comprehensiveness in separate pieces of the analysis,
and the effect on the scientific defensibility of the
overall project.
This section discusses ten critical issues in select-
ing health information for use in the section 812 as-
sessment: use of peer-reviewed research, confound-
ing factors, uncertainty, the magnitude of exposure,
duration of exposure, threshold concentrations, the
target population, statistical significance of relation-
ships, relative risks, and the need for baseline inci-
dence data. The previous discussion about the types
of research methods available for the health informa-
tion alluded to some of these issues, as they are po-
tentially important factors in selecting between stud-
ies using different methods. Other issues address how
scientific research is used in the overall analytical
framework.
Peer-Review of Research
Whenever possible, peer reviewed research rather
than unpublished information has been relied upon.
Research that has been reviewed by the EPA's own
peer review processes, such as review by the Clean
Air Science Advisory Committee (CASAC) of the
Science Advisory Board (SAB), has been used when-
ever possible. Research reviewed by other public sci-
entific peer review processes such as the National
Academy of Science, the National Acidic Precipita-
tion Assessment Program, and the Health Effects In-
stitute is also included in this category.
Research published in peer reviewed journals but
not reviewed by CASAC has also been considered for
use in the section 812 assessment, and has been used
if it is determined to be the most appropriate avail-
able study. Research accepted for publication by peer
reviewed journals ("in press") has been considered to
have been published. Indications that EPA intends to
submit research to the CASAC (such as inclusion in a
draft Criteria Document or Staff Paper) provide fur-
ther evidence that the journal-published research
should be used.
Air pollution health research is a very active field
of scientific inquiry, and new results are being pro-
duced constantly. Many research findings are first
released in University Working Papers, dissertations,
government reports, non-reviewed journals and con-
ference proceedings. Some research is published in
abstract form in journals, which does not require peer
review. In order to use the most recent research find-
ings and be as comprehensive as possible, unpublished
research was examined for possible use in the section
812 assessment. Any unpublished research used is
carefully identified in the report, and treated as hav-
ing a higher degree of uncertainty than published re-
sults. The peer review of the section 812 assessment
by the Advisory Council on Clean Air Compliance
Analysis provides one review process for all compo-
nents of the assessment, as well as for the way in which
the components have been used.
Confounding Factors
Confounding can occur when the real cause of
disease is associated with a number of .factors. If only
one contributing factor is evaluated in an epidemio-
logical study, a false association may occur. For ex-
ample, in epidemiology studies of air pollution, it is
important to take into account weather conditions,
because weather is associated with both air pollution
and health outcomes. If only air pollution is evalu-
ated, a false association between air pollution and
health could result; one may incorrectly assume that
a reduction in air pollution is exclusively responsible
for a reduction in a health outcome. Potential con-
founders include weather-related variables, age and
gender mix of the subject population, and pollution
emissions other than those being studied. Studies that
control for a broad range of likely confounders can
offer a more robust conclusion about an individual
pollutant, even if the statistical confidence interval is
larger due to the inclusion of more variables in the
analysis.
D-9
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
In many cases, several pollutants in a "pollutant
mix" are correlated with each other—that is, they tend
to occur simultaneously. Therefore, although there
may be an association between a health effect and each
of several pollutants in the mix, it may not be clear
which pollutant is causally related to the health effect
(or whether more than one pollutant is causally re-
lated). This analysis includes epidemiological mod-
eling of the health effects that have been associated
with exposure to a number of pollutants. In most cases
where the health effect is being modeled for the sev-
eral correlated pollutants of interest, regression coef-
ficients based on PM as a surrogate for the mixture
were chosen hi preference to multiple pollutant mod-
els and single pollutant models. The most important
example of this occurs in estimating mortality effects.
There is substantial evidence that exposure to criteria
pollutants, either individually or collectively, is sig-
nificantly associated with excess mortality. Generally,
this association is related to particulate matter. There-
fore, even though particulate matter cannot be shown
to be the sole pollutant causing pollution-related ex-
cess mortality, it can be used as an indicator of the
pollutant mixture which appears to result in excess
mortality. This analysis estimates excess mortality (for
all criteria pollutants other than lead) using PM as an
indicator of the pollutant mix to which individuals
were exposed. This issue is discussed further below,
where details on estimating mortality effects are ex-
plored.
The one exception to the use of single pollutant
regression models is estimating hospital admissions.
Both PM and ozone are generally found to have a sta-
tistically significant and separate association with
hospital admissions. Using separate regressions (from
single pollutant models) for each pollutant may over-
state the number of effects caused by each pollutant
alone. On the other hand, using PM as a single indica-
tor of the pollutant mix could underestimate the total
hospital admissions caused by different mechanisms.
Separate PM and ozone coefficients for hospital ad-
missions are selected from regression models that
consider the effects of both pollutants simultaneously.
Uncertainty
The stated goal of the section 812 assessment is
to provide a comprehensive estimate of benefits of
the Clean Air Act. To achieve this goal, information
with very different levels of confidence must be used.
Benefit categories are not to be omitted simply be-
cause they are highly uncertain or controversial, but
those benefit categories that are reasonably well un-
derstood must be distinguished from those which are
more tentative.
The ideal approach to characterizing uncertainty
is to conduct a formal quantitative uncertainty analy-
sis. A common approach develops an estimated prob-
ability distribution for each component of the analy-
sis. A Monte Carlo procedure draws randomly from
each of these distributions to generate an estimate of
the result. Evaluating the result for many such ran-
dom combinations, creates a distribution of results that
reflects the joint uncertainties in the analysis.
The most serious obstacle to preparing a formal
quantitative uncertainty analysis is identifying all the
necessary distributions for each component of the
analysis. The Monte Carlo procedure requires that all
components of the model be rerun many times. How-
ever, the section 812 project links the outputs from
independent modeling activities. It would be imprac-
tical to simultaneously rerun the macroeconomic,
emissions, air quality, and exposure models because
of the diverse origins of the models. Therefore, in-
stead of a complete formal uncertainty analysis, the
section 812 assessment includes a less rigorous analy-
sis of the inherent uncertainties hi the modeling ef-
fort. The uncertainty analysis combines quantitative
and qualitative elements designed to sufficiently de-
scribe the implications of the uncertainties. A primary
goal of the sensitivity/uncertainty analysis is to iden-
tify the health effects that make a sizable contribution
to the overall assessment of the monetary benefits.
There may be situations where there are significant
differences hi the available information used to pre-
dict the incidence of a particular health effect (i.e.,
the uncertainty bounds are large). It is important to
alert the reader to situations where using the lower
incidence estimates may portray the health effect as
only modestly contributing to the overall total ben-
efits, but using reasonable alternative higher estimated
incidence figures (or higher monetized values) would
substantially impact not only the monetized value of
the individual health effect, but actually make a no-
ticeable difference in the total benefits assessment.
Consideration of the overall uncertainties inher-
ent in the section 812 assessment has several impor-
tant implications for health study selection. It was im-
portant to carefully examine the balance between the
level of uncertainties in the analysis and the need for
D-10
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
comprehensive coverage of all benefit categories.
There were frequently situations in which a direct
tradeoff existed between more comprehensive cover-
age and the restriction of the analysis to more certain
information. Also, the relationship between the un-
certainty in other parts of the analysis and the uncer-
tainty for each particular health effect was carefully
considered.
Magnitude of Exposure
One component of the section 812 analysis esti-
mates the air pollution levels that would have occurred
hi the absence of the Clean Air Act. These estimates
are larger than currently observed levels of U.S. air
pollution, and perhaps even levels currently observed
elsewhere in the world. This aspect of the analysis
poses difficulties for the application of concentration-
response functions that have been based on exposures
at much lower pollution levels. The shape of the con-
centration-response function much above observed
exposures levels is unknown. It is possible that bio-
logical mechanisms affecting response that are unim-
portant at low levels of exposure may dominate the
form of response at higher levels, introducing
nonlinearity to the mathematical relationship. In gen-
eral, studies that include exposure levels spanning the
range of interest in the section 812 assessment are
preferable to studies at levels outside of the range, or
that only include a narrow part of the range. A pos-
sible drawback to this approach is that studies which
fit this criterion have often been conducted outside
the U.S. The application .of foreign studies to U.S.
populations introduces additional uncertainties regard-
ing the representativeness of the exposed population
and the relative composition of the air pollution mix
for which the single pollutant is an indicator. These
difficult issues were considered in selecting studies
for the benefits analysis.
Duration of Exposure
Selection of health studies for the section 812 as-
sessment must consider the need to match the health
information to the air quality modeling conducted for
the assessment. For example, information on the health
effects from short term (five minute) exposure to sul-
fur dioxide cannot be readily combined with infor-
mation on average daily sulfur dioxide levels. In se-
lecting studies for the benefits analysis, preference was
shown for studies whose duration of exposure matched
one of the averaging times of the air quality data.
Thresholds
Exposure-response relationships are conceptual-
ized as either exhibiting a threshold of exposure be-
low which adverse effects are not expected to occur,
or as having no response threshold, where any expo-
sure level theoretically poses a non-zero risk of re-
sponse to at least one segment of the population. The
methods employed by health researchers to charac-
terize exposure-response relationships may or may not
explicitly analyze the data for the existence of a thresh-
old. Studies may analyze relationships between health
and air pollution without considering a threshold. If a
threshold for population risk exists but is not identi-
fied by researchers, then Clean Air Act benefits could
be overestimated if CAA levels are below the thresh-
old, because the risk reduction from the no-control
scenario could be overstated. On the other hand, if a
threshold is artificially imposed where one does not
exist, the relative benefits of the Clean Air Act may
be underestimated. In general, those studies that ex-
plicitly consider the question of a threshold (whether
a threshold is identified or not) provide stronger evi-
dence; consideration of this question is a positive fea-
ture when selecting studies for this analysis.
Target Population
Many of the studies relevant to quantifying the
benefits of air pollution reductions have focused on
specific sensitive subpopulations suspected to be most
susceptible to the effects of the pollutant. Some of
these effects may be relevant only for the studied sub-
population; effects on other individuals are either un-
known, or not expected to occur. For such studies, the
challenge of the analysis is to identify the size and
characteristics of the subpopulation and match its oc-
currence to exposure. Other studies have examined
specific cohorts who may be less susceptible than the
general population to health effects from air pollu-
tion (e.g., healthy workers), or who differ in age, gen-
der, race, ethnicity or other relevant characteristics
from the target population of the benefits analysis.
Extrapolating results from studies on nonrepresenta-
tive subpopulations to the general population intro-
duces uncertainties to the analysis, but the magnitude
of the uncertainty and its direction are often unknown.
Because of these uncertainties, benefit analyses often
limit the application of the dose-response functions
only to those subpopulations with the characteristics
of the study population. While this approach has merit
in minimizing uncertainty in the analysis, it can also
D-ll
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
severely underestimate benefits if, in fact, similar ef-
fects are likely to occur in other populations. For these
reasons, studies that examine broad, representative
populations are preferable to studies with narrower
scope because they allow application of the functions
to larger numbers of persons without introducing ad-
ditional uncertainty.
Many studies included in the section 812 analy-
sis focus on a particular age cohort of the population
for the identification of health effects. The choice of
age group is often a matter of convenience (e.g., ex-
tensive Medicare data may be available for the eld-
erly population) and not because the effects are, in
reality, restricted to the specific age group (even
though their incidence may vary considerably over
the life span). However, since no information is avail-
able about effects beyond the studied population, this
analysis applies the given concentration-response re-
lationships only to those age groups corresponding to
the cohorts studied. Likewise, some studies were per-
formed on individuals with specific occupations, ac-
tivity patterns, or medical conditions because these
traits relate to the likelihood of effect. In these cases,
application of dose-response functions has been re-
stricted to populations of individuals with these same
characteristics.
Statistical Significance of Exposure-Response
Relationships
The analysis includes as many studies related to a
given health effect as possible, except for studies in-
applicable to the current analysis. For some endpoints,
the group of adequate studies yielded mixed results,
with some showing statistically significant responses
to pollutant concentrations and others with insignifi-
cant associations. Unless study methods have been
judged inadequate, dose-response functions with both
statistically significant and insignificant coefficients
have been included to characterize the possible range
of risk estimates. Excluding studies exclusively on the
basis of significance could create an upward bias in
the estimates by not reflecting research that indicates
there is a small, or even zero, relationship between
pollution and specific health effects. It should be noted,
however, that some studies that found insignificant
effects for a pollutant could not be used because they
did not report the insignificant coefficient values.
In some cases, a single study reported results for
multiple analyses, yielding both significant and non-
significant results, depending on the nature of the in-
put parameters (e.g., for different lag periods or con-
current exposures). In these cases, only significant
results were included.
Relative Risks
Many studies reported only a relative risk value
(defined as the ratio of the incidence of disease in two
groups exposed to two different exposure levels). The
analysis required conversion of these values to their
corresponding regression coefficients when the coef-
ficients were not reported. When converting the rela-
tive risk to a coefficient value, the analysis used the
functional form of the regression equation reported
by the authors of the study.
The coefficients from a number of studies mea-
sured the change in the number of health effects for
the study population rather than a change per indi-
vidual. These coefficients were divided by the size of
the study population to obtain an estimate of change
per individual. The coefficient could then be multi-
plied by the size of the population modeled in the cur-
rent analysis to determine total incidence of health
effects.
Baseline Incidence Data
Certain dose-response functions (those expressed
as a change relative to baseline conditions) require
baseline incidence data associated with ambient lev-
els of pollutants. Incidence data necessary for the cal-
culation of risk and benefits were obtained from na-
tional sources whenever possible, because these data
are most applicable to a national assessment of ben-
efits. The National Center for Health Statistics pro-
vided much of the information on national incidence
rates. However, for some studies, the only available
incidence information come from the studies them-
selves; in these cases, incidence in the study popula-
tion is assumed to represent typical incidence nation-
ally.
Studies were excluded if health endpoints could
not be defined in the U.S. population. For example, in
Pope and Dockery (1992) the authors developed a
unique definition of symptomatic children in Utah
which has no correlation in the incidence data bases
which were available; consequently, the results could
not be applied to the general population.
D-12
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
Estimating Mortality Effects
Using PM as an Indicator
There is substantial evidence that exposure to cri-
teria pollutants, either individually or collectively, is
significantly associated with excess mortality. This
association is most closely and consistently related to
the ambient air concentrations of PM.
Several studies have found small but statistically
significant relationships between ozone and mortal-
ity, while other studies have not found a significant
relationship. There is inconclusive evidence whether
ozone has an effect independent of the effect of other
pollutants (e.g., PM or CO), has a synergistic effect
in combination with other effects, or is a confounder
in the relationship between mortality and other pol-
lutants. For example, in a recent study HEI (1996)
found a significant and relatively stable ozone coeffi-
cient for most of the model specifications presented
in the study. However, the measured ozone effect was
largest and most significant in the winter and autumn,
when ozone levels are low.
This analysis estimates excess mortality (for all
criteria pollutants other than lead) using PM as an in-
dicator of the pollutant mix to which individuals
were exposed. Even if paniculate matter exposure
cannot be shown to be an independent causal factor
of excess mortality, it is, at a minimum, a good indi-
cator measure of the exposure to the pollutant mix-
ture that has been shown to be related to excess mor-
tality. Because PM is used as an indicator, the con-
centration-response functions from single pollutant
models (i.e., statistical models including PM as the
only pollutant) are preferred. To the extent that ozone
is correlated with PM, the effect of ozone, either as an
independent association or acting hi combination with
other pollutants, will be captured by this approach.
Estimating the Relationship Between PM and
Premature Mortality
Long-term exposure versus short-term exposure
studies and the degree of prematurity of mortality.
Both long-term exposure (cohort) studies and short-
term exposure (longitudinal or time-series) studies
have estimated the relationship between exposure to
PM and premature mortality. While there are advan-
tages and disadvantages to each type of study (as dis-
cussed above), the long-term studies may capture more
of the PM-related premature mortality, as well as pre-
mature mortality that is more premature, than the
short-term studies.
The degree of prematurity of pollution-related
death may be an important uncertainty in the effort to
estimate the benefits of reducing pollution concentra-
tions, as discussed in Appendix I. The willingness to
pay to save a few days of life may be significantly
less than the willingness to pay to save a few, or many,
years of life. Evidence concerning the degree of pre-
maturity of pollution-related death would, in this case,
be crucial. Such evidence is, however, still scarce.
There is some limited evidence that the relative risk
of mortality from exposure to PM is higher for older
individuals than for younger individuals. This, com-
bined with the fact that the baseline incidence of mor-
tality consists disproportionately of people 65 and
over, suggests that PM-related mortality is dispropor-
tionately among older individuals. The extent to which
prematurity of death among older individuals is on
the order of days or weeks versus years, however, is
more uncertain. The short-term exposure studies can
provide little information on this. It is possible that
premature deaths on high pollution days would have
occurred only days later, if the individuals were sick
and therefore particularly susceptible. The fact that
the long-term exposure mortality studies found sub-
stantially larger relative risks, however, suggests that
not all of the premature mortality is on the order of
days or even weeks. Shortening of life of such a small
duration would not be detectable in a long-term epi-
demiology study, ensuring that the effects detected in
such studies must represent longer periods of life short-
ening. This suggests that at least some of the prema-
ture mortality associated with exposure to PM may
reduce lifespans by substantially longer amounts of
time.
Even if an individual's PM-related premature
mortality is of very short duration, on the order of
days, however, it may be misleading to characterize
such a PM-related loss as only those few days if the
individual's underlying susceptibility was itself ex-
acerbated by chronic exposure to elevated levels of
pollution. Suppose, for example, that long-term ex-
posure to elevated PM levels compromises the car-
diopulmonary system, making the individual more
susceptible to mortality on peak PM days than he oth-
erwise would have been. If this is the case, then the
underlying susceptibility would itself be either caused
by chronic exposure to elevated PM levels or exacer-
D-13
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
bated by it. Characterizing the individual's loss as a
few days could, in this case, be a substantial underes-
timate.
In addition, the long-term studies estimate sig-
nificantly more PM-related mortality than the annual
sum of the daily estimates from the short-term stud-
ies, suggesting that the short-term studies may be
missing a component of PM-related mortality that is
being observed in the long-term studies. For example,
if chronic exposure to elevated PM levels causes pre-
mature mortality that is not necessarily correlated with
daily PM peak levels, this type of mortality would be
detected in the long-term studies but not necessarily
in the short-term studies. Two of the long-term expo-
sure studies suggest, moreover, that the association
between ambient air pollution and mortality cannot
be explained by the confounding influences of smok-
ing and other personal risk factors.
Uncertainties surround analyses based on epide-
miological studies of PM and mortality. In addition
to the uncertainty about the degree of prematurity of
mortality, there are other uncertainties surrounding
estimates based on epidemiological studies of PM and
mortality. Although epidemiological studies are gen-
erally preferred to human clinical studies, there is
nevertheless uncertainty associated with estimates of
the risk of premature mortality (and morbidity) based
on studies in the epidemiological literature. Consid-
ering all the epidemiological studies of PM and mor-
tality, both short-term and long-term, there is signifi-
cant interstudy variability as well as intrastudy un-
certainty. Some of the difference among estimates
reported by different studies may reflect only sam-
pling error; some of the difference, however, may re-
flect actual differences in the concentration-response
relationship from one location to another. The trans-
ferability of a concentration-response function esti-
mated in one location to other locations is a notable
source of uncertainty.
Although there may be more uncertainty about
the degree of prematurity of mortality captured by
short-term exposure studies than by long-term expo-
sure studies, certain sources of uncertainty associated
with long-term exposure studies require mention. Al-
though studies that are well-executed attempt to con-
trol for those factors that may confound the results of
the study, there is always the possibility of insuffi-
cient or inappropriate adjustment for those factors that
affect long-term mortality rates and may be con-
founded with the factor of interest (e.g., PM concen-
trations). Prospective cohort studies have an advan-
tage over ecologic, or population-based, studies in that
they gather individual-specific information on such
important risk factors as smoking. It is always pos-
sible, however, that a relevant, individual-specific risk
factor may not have been controlled for or that some
factor that is not individual-specific (e.g., climate) was
not adequately controlled for. It is therefore possible
that differences in mortality rates that have been as-
cribed to differences in average PM levels may be
due, in part, to some other factor or factors (e.g., dif-
ferences among communities in diet, exercise,
ethnicity, climate, industrial effluents, etc.) that have
not been adequately controlled for.
Another source of uncertainty surrounding the
prospective cohort studies concerns possible histori-
cal trends in PM concentrations and the relevant pe-
riod of exposure, which is as yet unknown. TSP con-
centrations were substantially higher in many loca-
tions for several years prior to the cohort studies and
had declined substantially by the time these studies
were conducted. If this is also true for PM10 and or
PM^, it is possible that the larger PM10 and or PM25
coefficients reported by the long-term exposure stud-
ies (as opposed to the short-term exposure studies)
reflect an upward bias. If the relevant exposure pe-
riod extends over a decade or more, then a coefficient
based on PM concentrations at the beginning of the
study or in those years immediately prior to the study
could be biased upward if pollution levels had been
decreasing markedly for a decade or longer prior to
the study.
On the other hand, if a downward trend in PM
concentrations continued throughout the period of the
study, and if a much shorter exposure period is rel-
evant (e.g., contained within the study period itself),
then characterizing PM levels throughout the study
by those levels just prior to the study would tend to
bias the PM coefficient downward.
The relevant exposure period is one of a cluster
of characteristics of the mortality-PM relationship that
are as yet unknown and potentially important. It is
also unknown whether there is a time lag in the PM
effect. Finally, it is unknown whether there may be
cumulative effects of chronic exposure — that is,
whether the relative risk of mortality actually increases
as the period of exposure increases.
Estimating the relationship between PM and pre-
mature mortality. The incidence of PM-related mor-
tality used for estimating the benefits of the CAA is
D-14
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
based on the concentration-response relationship re-
ported by one of the two recent long-term exposure
(prospective cohort) studies (Pope et al., 1995, and
Dockery et al., 1993). Because it is based on a much
larger population and many more locations than
Dockery et al. (1993), the concentration-response
function from Pope et al. (1995) was used in this analy-
sis. The results of Pope et al. are consistent with those
of Dockery et al., which reported an even larger re-
sponse, but in only six cities. Moreover, Pope et al. is
also supported by several ecological cross-sectional
studies of annual mortality based on 1960 and 1970
census data (using either TSP or sulfate as indicators
of PM), including the work of Lave and Seskin (1977)
and Lipfert (1984).
Numerous short-term exposure (time series) stud-
ies have also reported a positive and statistically sig-
nificant relationship between PM and mortality. Of
the fourteen studies that estimated the relationship
between daily PM10 concentrations and daily mortal-
ity listed in Table 12-2 of the PM Criteria Document,
twelve reported positive and statistically significant
findings (Pope et al., 1992; Pope and Kalkstein, 1996;
Dockery et al., 1992; Schwartz, 1993a; Ozkaynak et
al., 1994; Kinney et al., 1995: Ito et al., 1995; Ostro et
al., 1996; Saldiva et al., 1995; Styer et al., 1995; Ito
and Thurston, 1996; Schwartz et al., 1996). While
these studies lend substantial support to the hypoth-
esis that there is a relationship between PM10 and
mortality, they may be capturing only the portion of
that relationship involving short-term effects. For this
reason, they are considered in this analysis only as
supporting evidence to the results of the study by Pope
etal.
The Pope et al. study has several further advan-
tages. The population followed in this study was
largely white and middle class, decreasing the likeli-
hood that interlocational differences in premature mor-
tality were due in part to differences in socioeconomic
status or related factors. In addition, the generally
lower mortality rates and possibly lower exposures to
pollution among this group, in comparison to poorer
minority populations, would tend to bias the PM co-
efficient from this study downward, counteracting a
possible upward bias associated with historical air
quality trends discussed above.
Another source of downward bias in the PM co-
efficient in Pope et al. is that intercity movement of
cohort members was not considered in this study.
Migration across study cities would result in expo-
sures of cohort members being more similar than
would be indicated by assigning city-specific annual
average pollution levels to each member of the co-
hort. The more intercity migration there is, the more
exposure will tend toward an intercity mean. If this is
ignored, differences in exposure levels, proxied by
differences in city-specific annual median PM levels,
will be exaggerated, resulting in a downward bias of
the PM coefficient (because a given difference in mor-
tality rates is being associated with a larger differ-
ence in PM levels than is actually the case).
In summary, because long-term exposure studies
appear to have captured more of the PM-related pre-
mature mortality, as well as premature mortality that
is more premature, they are preferable to the short-
term exposure studies. Among the long-term expo-
sure studies, the Pope et al. study has several advan-
tages, as discussed above, which are likely to reduce
the possibility of a key source of confounding and
increase the reliability of the concentration-response
function from that study. For these reasons, the con-
centration-response function estimated in this study
is considered the most reasonable choice for this analy-
sis.
Matching PM Indices in the Air Quality Profiles
and Concentration-Response Function. The Pope et
al. study examined the health effects associated with
two indices of PM exposure: sulfate particles and fine
particles (PM2 5). The reported mortality risk ratios are
slightly larger for PM25 than for sulfates (1.17 versus
1.15 for a comparison between the most polluted and
least polluted cities). The PM2 5 relationship is used in
this analysis because it is more consistent with the
PM10 air quality data selected for the analysis. Esti-
mated changes in PM25 air quality must be matched
with the PM^ mortality relationship. However, only
PM10 profiles were used for the entire 20 year period.
Therefore, the same regional information about the
PMIO components (sulfate, nitrate, organic particulate
and primary particulate) used to develop the PM10 pro-
files were used to develop regional PM2 /PM10 ratios.
Although both urban and rural ratios are available,
for computational simplicity, only the regional urban
ratios were used to estimate the PM25 profiles from
the PM10 profiles used in the analysis. This reflects
the exposure of the majority of the modeled popula-
tion (i.e., the urban population), while introducing
some error in the exposure changes for the rural popu-
lation. In the east and west, where the rural ratio is
larger than the urban ratio, the change in PM25 expo-
sure will be underestimated for the rural population.
D-15
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
In the central region the PM2J change will be overes-
timated. These ratios were used in each year during
1970-1990, introducing another source of uncertainty
in the analysis. Table D-5 summarizes the PM2 /PMu> Ratios'Used td Estimate*
PMz.5 Data Used With Pope et & (1995)
Mortality Relationship.
East Central West National
Urban
Rural
0.59
0 68
0,58
0.53
0.48
0.49'"
»-9'55
0.57
Prematurity of Mortality: Life-Years Lost as a Unit
of Measure
Perhaps the most important health effect that is
examined in this analysis is mortality. Although this
analysis does not take into account the degree of pre-
maturity of death (that is, the ages of those individu-
als who die prematurely from exposure to PM are not
considered), considerable attention has been paid to
this issue and, in particular, to life-years lost as an
alternative to lives lost as a measure of the mortality-
related effects of pollution.
Because life-years lost is of potential interest and
because there is a substantial potential for confusion
in understanding apparently disparate estimates of life-
years lost from pollution exposure, this section at-
tempts to present a clear discussion of the various
possible measures of life-years lost, what they depend
on, and how they are related to each other.
Because the actual number of years any particu-
lar individual is going to live cannot be known, "life-
years lost" by an individual actually refers to an ex-
pected loss of years of life by that individual. The
expected loss of years of life by an individual depends
crucially on whether the expectation is contingent on
the individual only having been exposed to PM or on
the individual actually having died from that expo-
sure.
An ex ante estimate of life-years lost per indi-
vidual is contingent not on the individual having died
prematurely but only on the individual having been
exposed. Suppose, for example, that a 25 year old has
a life expectancy of 50 more years in the absence of
exposure and only 49 more years in the presence of
exposure. Given (chronic) exposure from the age of
25 on, the 25 year old exposed to (some elevated level
of) PM might expect a shortening of life expectancy
of one year, for example. That is one expected life-
year lost due to chronic exposure. This is the life-years
lost that can be expected by every exposed individual.
An ex post estimate of life-years lost per individual
is contingent on the individual actually having died
from exposure to PM. When an individual dies of
exposure to PM, he is said to have lost the number of
years he would have been expected to live, calculated,
for example, from age- and gender-specific life ex-
pectancy tables. Suppose that the life expectancy of
25 year olds is 75 — that is, a 25 year old can expect
to live 50 more years. A 25 year old who dies from
exposure to PM has therefore lost 50 expected years
of life. This is the life-years lost that can be expected
by every 25 year old affected individual (i.e., every
25 year old who actually dies from exposure to PM).
Estimates of the total life-years lost by a popula-
tion exposed to PM depend on several factors, includ-
ing the age distribution and the size of the exposed
population, the magnitude of the change (or changes)
in PM being considered, the relative risk assumed to
be associated with each change in PM, and the length
of time exposure (i.e., the change in PM) is presumed
to occur. A population chronically exposed to a given
increase in PM will lose more life-years than a popu-
lation exposed to the same increase in PM for only a
year or two.6 A population that is generally older will
lose fewer life-years, all else equal, than one that is
generally younger, because older individuals have
fewer (expected) years of life left to lose. And a popu-
lation exposed to a greater increase in PM will lose
more life-years than if it were exposed to a smaller
increase in PM. Finally, the life-years lost by the popu-
lation will increase as the relative risk associated with
the increase in PM increases.
Life-years lost are usually reported as averages
over a population of individuals. The population be-
ing averaged over, however, can make a crucial dif-
6 Even in the absence of cumulative effects of exposure, exposure of a population for many years will result in a greater total
number of pollution-related deaths than exposure for only a year or two, because the same relative risk is applied repeatedly, year
after year, to the population, rather than for only a year or two.
~~~ " D-16 ~
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
ference in the reported average life-years lost, as noted
above. The average life-years lost per exposed indi-
vidual (the ex ante estimate) is just the total life-years
lost by the population of exposed individuals divided
by the number of exposed individuals. This average
will depend on all the factors that the total life-years
lost depends on except the size of the exposed popu-
lation. The average life-years lost by an exposed indi-
vidual is a statistical expectation. It is the average of
the numbers of life-years actually lost by each mem-
ber of the exposed population. Alternatively, it can be
thought of as a weighted average of possible numbers
of years lost, where the weights are the proportions of
the population that lose each number of expected years
of life. Although those individuals who do die prema-
turely from exposure to PM may lose several expected
years of life, most exposed individuals do not actu-
ally die from exposure to PM and therefore lose zero
life-years. The average life-years lost per exposed in-
dividual in a population, alternatively referred to as
the average decrease in life expectancy of the exposed
population, is therefore heavily weighted towards zero.
The average number of life-years lost per individual
who dies of exposure to PM (the ex post measure of
life-years lost) is an average of the numbers of ex-
pected years of life lost by individuals who actually
died prematurely because of PM. Because everyone
who dies prematurely from exposure to PM loses some
positive number of expected years of life, this aver-
age, by definition, does not include zero.
An example of an ex ante measure of life-years
lost is given by a study in the Netherlands (WHO,
1996), which considered a cohort of Dutch males, aged
25-30, and compared the life expectancy of this co-
hort to what it would be hi a hypothetical alternative
scenario in which these individuals are continuously
exposed to concentrations of PM2S that are 10 ug/m3
lower than in the actual scenario, the life expectancy
of this cohort of 25-30 year old Dutch males was cal-
culated to be 50.21 years in the actual scenario, based
on a 1992 life table from the Netherlands. Assuming
that the relative risk of mortality associated with an
increase of 10 ug/m3 PM25 is 1.1 (the average of the
relative risks of 1.14 from Dockery et al., 1993, and
1.07 from Pope et al., 1995), the study authors calcu-
lated death rates in the hypothetical "cleaner" scenario
by dividing the age-specific death rates in the actual
scenario by 1.1. Using these slightly lower death rates,
and assuming that the effect of PM does not begin
until 15 years of exposure, the authors constructed a
life table for the cohort in the hypothetical "cleaner"
scenario. Based on this new life table in a cleaner
world, the life expectancy of the cohort of 25-30 year
old Dutch males was calculated to be 51.32 years in
the hypothetical cleaner scenario. (In calculating life
expectancies in both the "dirty" scenario and the
"clean" scenario, it is assumed that any individual who
does not survive to the next 5-year age group lives
zero more years. For example, a 30 year old individual
either survives to age 35 or dies at age 30.) The change
in life expectancy for this cohort of 25-30 year old
Dutch males, due to a change in PM exposure of 10
ug/m3 for the rest of their lives (until the age of 90),
was therefore 51.32 years - 50.21 years =1.11 years.
That is, the average life-years lost by an exposed in-
dividual hi this population, under these assumptions,
is 1.11 years.
The estimate of 1.11 years of expected life lost
depends on several things, as mentioned above. If the
study authors had used the relative risk from Pope et
al., 1995, alone, (1.07 instead of 1.1), for example,
the change in life expectancy (the ex ante measure of
life-years lost) for this cohort of 25-30 year old Dutch
males would have been 0.80 years. Similarly, chang-
ing the assumption about the duration of exposure also
changes the estimate of ex ante life-years lost. Using
a relative risk of 1.1, but assuming that exposure lasts
only during the first 5 years (i.e., that the death rate in
the first five years, from age 25 through age 30, is
lower but that after that it is the same as in the "dirty"
scenario), the average life-years lost by an exposed
individual in this population is reduced from 1.11 years
to 0.02 years.
By their construction and definitions, the average
life-years lost per exposed individual and the average
life-years lost per affected individual (i.e., per indi-
vidual who dies prematurely from PM) take the same
total number of life-years lost by the exposed popula-
tion and divide them by different denominators. The
average life-years lost per exposed individual divides
the total life-years lost by the total population exposed;
the average life-years lost per affected individual di-
vides the same total life-years lost by only a small
subset of the total population exposed, namely, those
who died from PM. The average per exposed indi-
vidual is therefore much smaller than the average per
affected individual. Because both types of average may
be reported, and both are valid measurements, it is
important to understand that, although the numbers
will be very dissimilar, they are consistent with each
other and are simply different measures of the esti-
mated mortality impact of PM.
D-17
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
To calculate the total (estimated) life-years lost
by a population, it is necessary to follow each age
cohort in the population through their lives in both
scenarios, the "dirty" scenario and the "clean" sce-
nario, and compute the difference in total years lived
between the two scenarios, as WHO (1996) did for
the cohort of Dutch males 25-30 years old. This
method will be referred to as Method 1. In practice,
however, it is not always possible to do this. (Other
changes to the population, such as those from recruit-
ment and immigration, for example, would make such
an exercise difficult.) An alternative method, which
approximates this, is to predict the numbers of indi-
viduals in each age category who will die prematurely
from exposure to PM (i.e., who will die prematurely
in the "dirty" scenario), and multiply each of these
numbers by the corresponding expected number of
years remaining to individuals in that age category,
determined from life expectancy tables. This method
will be referred to as Method 2. Suppose, for example,
that individuals age 25 are expected to live to age 75,
or alternatively, have an expected 50 years of life re-
maining. Suppose that ten 25 year olds are estimated
to die prematurely because of exposure to PM. Their
expected loss of life-years is therefore 50 years each,
or a total of 500 life-years. If the same calculation is
carried out for the individuals dying prematurely in
each age category, the sum is an estimate of the total
life-years lost by the population.
Using Method 1 (and retaining the assumptions
made by WHO, 1996), the average life-years lost per
PM-related death among the cohort of Dutch males is
calculated to be 14.28 years. Using Method 2 it is es-
timated to be 14.43 years.
Although this ex post measure of life-years lost is
much larger than the ex ante measure (1.11 life-years
lost per exposed individual), it only applies to those
individuals who actually die from exposure to PM.
The number of individuals in the age 25-30 Dutch
cohort example who eventually die from exposure to
PM (7,646) is much smaller than the number of indi-
viduals in the age 25-30 Dutch cohort who are ex-
posed to PM (98,177). The total life-years lost can be
calculated either as the number of exposed individu-
als times the expected life-years lost per exposed in-
dividual (98,177*1.11 = 109,192.1) or as the number
of affected individuals times the expected life-years
lost per affected individual (7,646*14.28 = 109,192.1).
To further illustrate the different measures of life-
years lost and the effects of various input assump-
tions on these measures, death rates from the 1992
U.S. Statistical Abstract were used to follow a cohort
of 100,000 U.S. males from birth to age 90 in a "dirty"
scenario and a "clean" scenario, under various assump-
tions. Death rates were available for age less than 1,
ages 1-4, and for ten-year age groups thereafter. The
ten-year age groups were divided into five-year age
groups, applying the death rate for the ten-year group
to each of the corresponding five-year age groups. Ex
ante and ex post measures of life-years lost among
those individuals who survive to the 25-29 year old
category were first calculated under the assumptions
in the WHO (1996) study. These assumptions were
that the relative risk of mortality in the "dirty" sce-
nario versus the "clean" scenario is 1.1; that exposure
does not begin until age 25; that the effect of expo-
sure takes fifteen years; that individuals at the begin-
ning of each age grouping either survive to the next
age grouping or live zero more years; and that all in-
dividuals age 85 live exactly five more years. Under
these assumptions, the expected life-years lost per
exposed individual in the 25-29 year old cohort is 1.32
years. There are 96,947 exposed individuals in this
age cohort. The expected life-years lost per affected
individual (i.e., per PM-related death) is 16.44 years
(Method 1). There are 7,804 affected individuals. The
total life-years lost by individuals in this cohort is
128,329.3 (1.32*96,947 = 16.44* 7,804= 128,329.3).
If the relative risk is changed to 1.07, the expected
life-years lost per exposed individual in the cohort of
25-29 year old U.S. males is reduced from 1.32 to
0.95 years. The expected life-years lost per affected
individual (i.e., per PM-related death) is 16.44 years
(Method 1). Using a relative risk of 1.1 but assuming
no lag (i.e., assuming that exposure starts either at
birth or at age 25 and has an immediate effect), the
expected life-years lost per exposed individual in the
25-29 year old cohort changes from 1.32 to 1.12. The
expected life-years lost per affected individual (i.e.,
per PM-related death) becomes 19.7 years (Method
1).
D-18
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
Estimating Morbidity Effects
In addition to mortality effects, this analysis quan-
tifies effects for a number of non-fatal health end-
points. Several issues arise in implementing the stud-
ies selected for this analysis.
Overlapping Health Effects
Several endpoints reported in the health effects
literature overlap with each other. For example, the
literature reports relationships for hospital admissions
for single respiratory ailments (e.g. pneumonia or
chronic obstructive pulmonary disease) as well as for
all respiratory ailments combined. Similarly, several
studies quantify the occurrence of respiratory symp-
toms where the definitions of symptoms are not unique
(e.g., shortness of breath, upper respiratory symptoms,
and any of 19 symptoms). Measures of restricted ac-
tivity provide a final example of overlapping health
endpoints. Estimates are available for pollution-in-
duced restricted activity days, mild restricted activity
days, activity restriction resulting in work loss. This
analysis models incidence for all endpoints. Double-
counting of benefits is avoided in aggregating eco-
nomic benefits across overlapping endpoints (see
Appendix I).
Studies Requiring Adjustments
Applying concentration-response relationships
reported in the epidemiological literature to the na-
tional scale benefits analysis required by section 812
required a variety of adjustments.
Normalization of coefficients by population. To
be applied nationwide, concentration-response coef-
ficients must reflect the change in risk per person per
unit change in air quality. However, some studies re-
port the concentration-response coefficient, , as the
change in risk for the entire studied population. For
example, Thurston et al. (1994) reported the total num-
ber of respiratory-related hospital admissions/day in
the Toronto, Canada area. To normalize the coeffi-
cient so that it might be applied universally across the
country, it was divided by the population in the geo-
graphical area of study (yielding an estimate of the
change in admissions/person/day due to a change in
pollutant levels).
Within-study meta-analysis. In some cases, stud-
ies reported several estimates of the concentration-
response coefficient, each corresponding to a particu-
lar year or particular study area. For example, Ostro
and Rothschild (1989) report six separate regression
coefficients that correspond to regression models run
for six separate years. This analysis combined the in-
dividual estimates using a fixed coefficient meta-
analysis on the six years of data.
Conversion of coefficients dependent on symptom
status during the previous day. Krupnick et al. (1990)
employed a Markov process to determine the prob-
ability of symptoms that were dependent on symp-
tom status of the previous day. The current analysis
adjusts the regression coefficients produced by the
model in order to eliminate this dependence on previ-
ous day's symptom status.
Concentration-Response Functions:
Health Effects
After selecting studies appropriate for the section
812 analysis, taking into account the considerations
discussed above, the published information was used
to derive a concentration-response function for esti-
mating nationwide benefits for each health effect con-
sidered. In general, these functions combine air qual-
ity changes, the affected population and information
regarding the expected per person change in incidence
per unit change in pollutant level. The following tables
present the functions used in this analysis, incorpo-
rating information needed to apply these functions and
references for information.
Particulate Matter
The concentration-response functions used to
quantify expected changes in health effects associ-
ated with reduced exposure to particulate matter are
summarized in Table D-6. The data profiles selected
for use in this analysis are PM10. In those cases in
which PM10 was not the measure used in a study, this
analysis either converted PM10 air quality data to the
appropriate air quality data (e.g., PM25 or TSP) or,
equivalently, converted the pollutant coefficient from
the study to the corresponding PM10 coefficient, based
on location-specific information whenever possible.
D-19
-------
Benefits and Costs of the Clean Air Act, 1970 to 1990
«5
S
fM
.0
•o
o
m
d>
w
S
i
|
co
\d
A
S
^3
13
.52
o
u
e
•a
1
S
I
I
&
id
I
•a
8
:!
O
(U
Sources
£
nal fo
0.0
0404
dy Pop.
t
I
Baseline
Incidence (pe
00,
th
oin
Heal
Endp
(ICD-9 co
.,
C
coh
etal.
Pope
Ameri
2
o
o
o
«-
dS
s
0
tal
non-accide
deaths by
county*
S •
ffi "
II
ol
5 . -r
l
i
<*>
,
§ 0
£l?s
o "3 A 2
Pope, 199
e Va
Sal
oo
es
•s
3 1 &
''
«S «?
^ « 2
O ^ X
o V
.9
1
o
II
0
o
vo
n
o
tt
S oo
H Tfr
D-20
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
2 .*;
1 Q
"3
2 &
3 w
1
(2
**"
•s
co 2
&
o
oo
o
o
II
(S J_j
' O
o o
^ o
T i
•1 »3
S §
O
II
6
u
oi
O O
*"•" II
II a
._; to
o
I
O
II
eo.
o
II
co.
o
o
«5
O
II
ca.
o
II
co.
M
VO
o
&
2
dy Pop.
1*
« .g
> •"
•a
-a
8
-a
pos Meas.
rom Original
Stndy
if
Ex
I
t
u
2
'
e
(p
ence
0,00
A 2
^ g.
CO
8 >>£> ^~.
"fig's
A
i
to c« — .
1 «
/-s
'Z! S "3 t"-
3 .2 s si
'S w S "^
12
D-21
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
o
II
g
o
II
ca.
i
o
II
co.
•d
V-a
eo.
T3
vo S
o o
-x- 8
vo a
2
o
I
$
I
P-,
t
§
e>
I
1 A o
£8« "8
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
§
CO
§•
Pi
13
8
o\
•3
5
o
II
d
s
o
II
Functional fo
VO
o
II
ex
§
,0
bO
5
II
i?
.§
o
0.005 12
PM10 to TSP:
P
ert
E*c
<
onversi
TSP co
id in
VO
o
1
00
'a
1
•a
o
H i
al
a?
t!
"
i*
A
Health
Endpoint
(ICD-9 code)
•T-
'M *°
1
_
D-23
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
i
CO
•g
CO
S:
3
O>t
eb a &<
S •» fo
o
II
.2
"3
n
o
§>•§
»i
a<
g
J
§.
r
s*
«i
BJ
§
f
g-i
2
s
HI
J.
1
II .2
^l
| I
n
I
fj
^'35?
>»ia 9
s^'i
5. 8 5-
sg
21
li
o
s
vo
I
o
o 7^
W
tr-a . s
d>2 co S
SiS|
ll"l
g
« «5>g>
HI
D-24
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
CO
a
o
00
oo
i5
«§
o\
o
o
p
o
II
o
II
o
II
«
CO
o
II
N
s
o
o
o
•8
"CI
1
a p
*3 ^
*
? 5
•g
0
.s
1
" oo
•B 22
'
dy Pop.
-i
,a
s
HI
(S ra g
, .
5
*
U iT"-
S
Baseline
idence (per
100,000)
Health
Endpoint
9 code)
-v
li
•N
P
N
o -a
O ^
'
i
D-25
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
•^ a o ca
§1=88
.S-3
S iHI'i*
|SI|g
Ozone
The health effects literature includes studies of
the relationships between ozone and a variety of
non-fatal health effects. Many of these relationships
are provided by the same studies that reported the
particulate matter relationships shown above. For
some health endpoints, most notably hospital ad-
missions, multiple studies report alternative esti-
mates of the concentration-response relationship.
The variability between these reported estimates is
incorporated into the Monte Carlo approach used to
combine estimates of avoided health effects with
economic valuations (discussed in Appendix I).
Table D-7 documents the concentration-response
functions used in this analysis.
D-26
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
&Var.
nal fo
It
13
c
.2
'*^
o
I
(U
•fi
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
f
•§>
•1 «
1 co
1 -c
*§ *5
s
1
I
o
II
8
3 S
,
gl 8
il
43 •
3 .S 12
•&| | f
•"lie
D-28
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
I
GO
vo
ON
a
•s
co
•d"
VO
u
Functional fo
o
II
CO.
5
b
o.
o
CO
;
"
I
u
&
VO
I
s
I
Study Pop.
">" S
l!
O CO
s :
.§>
S
• 0
S §
I'a
•0 g
- g
-1
i
.
D-29
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
form*
Fun
•a -a
a §
E""
f
'
g af
III
0 £T
5 B
1 1
.5 1
- s
fi -s " V
-; 5-g ca
SL'g
'' S 5.
i!
vo
.3
I -
i -v
•S S
tx
*1
«
§ "
eq
fine
includes O3
Os
-
S
-0.0054'
ation p
fine
in
a
o.
a.
I "8 II
"5? S3
"S *"
I "8
a
CO
.3
I5?
oo
II1
.a
I
s 8 &
•
D-30
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
o
a\
&
i
o
o
£
•3
1
J
.3
'I
-
„
B a .2 a
pO .0 .0
IS-
t
I I
II II
x >?
I
.»»'
•a
I
111
BgS a
O o 2 &
18
a %
% %
«fl
|||
w u
D-31
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
<*)
o,
£
I
I*
PH
*
flit
Hi
15
§*
1*8
J<2
§ 1
•s »
|i
S-o
.g§
s?.
II
ai
o H
1.2
60
«5
•s
.s
J3
d be
1
?
?•
O "
« ^ o
|i!
5 g <=
o S o
tf
la*.
=3
So II
1H»t
•S ^ S Q •*
•§
•w
ts
?
=
-
10
1 1 1 S
s g a =
O Q
D-32
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
;dures, National Hospital Discharge Survey, 1990. Number of 1990 discharges divided by
Dt. of Commerce, Bureau of the Census, Washington, D.C.
ge in ppb O3.
§
•g
1
e i
n health effei
1
•S
03
CO
OJ
a Pollutant coefficients expressed
i
sd Diagnoses and Pi
i
&
.8
.Is
to
1
Q
i/j
5
£
o\
|
i
1
•g
1
92. Vital and
rom City and
b Centers for Disease Control, 19
i U.S. population (248,709,873) fi
S
2
isitional probabilities. ESEERCO (1994) obtained transitional probabilities for adults from
I
probabilities from 1
1
en
incorporatin;
o"
c Determined the incremental effi
nal study authors.
" U.S. EPA, 1994a.
;a>
o
365 days per year. Number of attacks per person per year = 9.9 from National Center for
£•
0?
MI per year) divided
nick and Kopp, 198
g ex
I'S
mber of attac
ttack rate (nui
e Calculated as baseline asthma a
, 1979 (as cit
t
a
1
th Statistics, National Health Inte
73
ac
a
•a
o
•a
ex
II
0"
~
owing relationship:
1 on the foil
S
CO
«
X>
0"
•a
ex
to A in cases/
f p coefficient and s.e. .converted
sing a two-tailed test of significance)
9
O
O
II
ex .
o
33
o
CO
assumed the
, the analysis
8 Study did not report a s.e. Thus
1
&
r person, using data
GO
1
1
:ort average a
h Ostro and Rothschild (1989) re]
eighted by the inverse of the variance for that coefficient.
£
ear's coefficient wai
1
pa
1
nts for six ye
irate coefficie
' P is a weighted mean using sepz
es the individual year).
CO
CJ
1/var,)], where I indi
)ums]ybs)
f the weights
: of the sum o
' Standard error is the square rool
I
u
•s
•o
on
c
I
s
73
>
I I 1
o
CA
1
I
a
8, -s
'o
S
•§
g,
eo
•8
>»
x>
1
a
•8 .§
I
_
§
I
•§
CQ. oo
3 .g .|
=3
S 5
V S
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Nitrogen Oxides
Nitrogen dioxide (NO2) is the primary focus of health studies on the nitrogen oxides and serves as the basis
for this analysis. The primary pathophysiology of NO2 in humans involves the respiratory system and the con-
centration-response function identified for NO2 describes the relationships between measures of NO2 and respi-
ratory illness.
A number of epidemiological studies of NO2 are available; however, most have either confounded expo-
sures (with other pollutants) or insufficient exposure quantification (e.g., exposure assessment indicates only
absence or presence of a gas stove). Most studies consider NO2 generated by gas stoves or other combustion
sources in homes and are therefore not directly usable in concentration-response functions. However, studies by
Melia et al, 1980 and Hasselblad et al, 1992 provide a reasonable basis for development of a concentration
response function. Table D-8 presents the function obtained from their work. The function relates NO2 to respi-
ratory illness in children.
D-34
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
.1
I
H
W O.
p
O
I
I
I
I.
I
O
1
f
§>«§
09
I
Z-r
D-35
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Carbon Monoxide
Three concentration-response relationships are available for estimating the health effects of carbon monox-
ide. The first relates ambient CO levels to hospital admissions for congestive heart failure (Morris et al., 1995).
The second equation (Allred et al., 1989a,b, 1991) relates the CO level in the bloodstream to the relative change
in time of onset of angina pain upon exertion. The third relates the CO level in the bloodstream to the relative
change in time of onset of silent ischemia. Due to the lack of quantitative information relating silent ischemia to
a meaningful physical health effect, this analysis uses only the first two dose-response functions shown in Table
D-9.
D-36
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
",;>""
xr~-
}
, , -,3,
Pt *
\s1*s!is J-
=;>%r:
V^-v
=^1^ -.'
s^s-x,
^V-, *v. ••
,^< <'
S "Hv"1 >•-•.
•.^.r >
•>-><#*•
?&i '
x ••,&••$<
v!
• !.
'".'•C
*,,-,:„&
vl ; "'•'•X-
-. -'X- 'f '
•:i
"X,,
;S
vE-ij^
*• •j.H^^.
t^wl1-^
HJfrt "
vf
M
O"
-c
''sOt'
"'-I1
^
1!
;:ft
^"
vg*
*^§>
1;
i*1* J\ f'
v *****?>
*?Q^
o
s
0
CO
^ 1?
isi
a •?
C3 «
unctional Form
to
1-1
gg,
^
0
tl
CO B.
0
(£
W a»
II
« 8
•H u
& S
-w
Si
ea Q.
3 -o
"J
r
•jrt 10 S w
„ ON 0 .Jj
c ^^ b ^
o *«
s
oo
u 2
o5 X
" ~
8 2
<] ^
* ^-*
8 1"
^ ^
I
1
!/•>
vo
S • ***
1 1 i||g-
35 *3 g to *S ^
g^t^ S
<<-! >j /-^
0 1 S
sf.^-2'
§18'
1
a' - 1 * i
£| g43^
— f o\
n ON
^5 ^""
J** *Tj«
Jw
S
<2
S?
oo
o
II
0
w
Sen "S
I 53 II ™
•S § 8 feS t
g> 5| ^S 1- i
1 I «Jfs ail
8 -S. «.- *? 5. 5 « «_• S I °°
! 1 lP| 1 fS
TI ° u5oS *«cg
| ^*s :? ^|sl 2|^3
•a -a -a •§ o a-Sgo-g ° 182
f § S § n .S § -S "S ° i' « » "s
« •-=! .O'S^caS _QJ3~aO
S*O £ S^l3'S M ^So
X HH 0 4- •£* >^ ^ p S **
1 1 | ; ll|| ; ill
l « o ^ili^ HI
1 ox 0 U SS5 =5 8 i ?
u
B. * ^ £
a .S >, - ^ » U '
| a ». |^ it^fl ^'"'•J
|.|go"| Ulslals
Q- o*"* £os^ §* ^o 2
co" p, rf s- "
*A ^ 2 *r* -*^ ^
il^llfl
j> j— Jj 2 rt S
g « c -2, ^ .g
— S3 g
g^ Q g
ifi
J^ 1 "B -2 « «
4-» g &0_ i *=3 *O (/-) M £
J |-|||< ^^"§§3
IuoS'O'gS-gll 8(JO
| S £,§ .5 »^
1 -Ja-
il Ji
s?
•o
o\
o\
I
ts
u
o
B-
S
a
£3
a
«§ 5
c
o
'
•3
< =;
u
1
I
cs
I
I
1
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Sulfur Dioxide
This analysis estimated one concentration-response function for SO2 using clinical data from two sources
on the responses of exercising asthmatics to SO2, as measured by the occurrence of respiratory symptoms in
mild and moderate asthmatics (see Table D-10).
D-38
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
Sources
1
I
* -^ go" |) 2,
,g « 2 «!
i £« £
«i to <"* ^ °,^ "^ g —
8 S |
*"* 4)
1 % s 1
ff
*< s
ta
«
!
D-39
-------
The Benefits
d Costs of the Clean Air Act, 1970 to 1990
Estimating Welfare Effects of
Exposure
In addition to avoided incidences of adverse hu-
man health effects, the air quality improvements esti-
mated to result from the CAA yield additional ben-
efits, namely welfare benefits. Table D-10 indicates a
variety of benefits expected to have accrued through
the avoidance of air pollution damage to resources.
As indicated, data supporting quantified estimates of
welfare benefits are more limited than those quanti-
fying the relationship between air pollution exposure
and human health. While evidence exists that a vari-
ety of welfare benefits result from air quality improve-
ments, currently available data supports quantifying
only a limited number of potential effects at this time.
The Table lists the effects quantified in the section
812 analysis; each is discussed below.
mate such benefits using reported relationships be-
tween ozone exposure and yields of a variety of com-
modity crops.
It should be noted that the method used to allo-
cate monitor-level ozone concentrations to estimate
crop exposure differed from that used to estimate
ozone health effects. Instead of assigning concentra-
tions from the nearest monitor, the agricultural ben-
efits analysis estimated ozone concentrations for each
county nationwide. This was necessary because of two
factors specific to the agricultural analysis. First, crop
production is reported at the county level, so changes
in crop yields associated with changes in ozone levels
must be estimated for each county. Second, much of
the nation's agricultural production of "commodity
crops" (corn, wheat, soybeans, etc.) occurs at signifi-
cant distances from the location of the population-
oriented ozone monitors. Thus, an algorithm was used
Table D-l 1. Selected Welfare Effecte^of Orilrai^ Pollutants,, ' '" : --;r ' *f --; ,,, ^ ,'!;'
Pollutant
Ozone
Parficulate Matter^
TSP/Sulfates
Nitrogen Oxides"
SulBir Dioxide
Quantified Welfare Effects, "* v"/
Agriculture - C^angesluf crop y Was '
{for 7 crop's^' ^ „>- ,/-x',^;,
Decreased wofker g*pdtictivief :\/t,,.
Materials Damaga - Household ,
'•* :'«&&*: -^r- • r' .
Visibility •-""w->'-" -*-. "" '
' *" ' '/x? '
Visibility'', , " ,v; -"*'„, - /
'' ,^?>»v- ' ^ j "• , ,'?* '-" '-"
trnquatttieied Welfare Ettects
, Changes In other crop,,yield's „„,, „,
^Matefjals data age ;j,
Ecological- effects on fereste' - <-* •
-Bcological---e'ffects-on wapfe ,
..Othei1 materials, daaiage,, ^ • ,„,
' 'Q:op losses, due to aejd;dep0siti0n
depositioif"" v,""
"Effectsoti fisheries dtjeto,,»id^ ;,
-deposition; v ,„, -,- , '.
-/Bffeetson forest"'"' ^,,^""' "•
"-"Crop.tossesdue'fo'^ijid.depbsition "
deposition ' , ,, ' , "*"'
' Effecfson. 'fisheries dw^'to acid-/- - ,t
,,depositiott"" „, .-. ••• ,,,-'• iV,'
Bffe^ts,pn forest, , 1 '"''
Agricultural Effects
This analysis was able to quantify the benefits to
economic welfare attributable to the increased crop
yields expected from CAA-related air quality improve-
ments. Appendix F describes the method used to esti-
to assign ozone concentrations for the agricultural
analysis for the control and no-control scenarios to
county centroids based on a planar interpolation of
concentrations at the nearest three monitors. Appen-
dix F documents the details of the triangulation of
ozone air quality data.
D-40
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
Materials Damage
Welfare benefits also accrue from avoided air
pollution damage, both aesthetic and structural, to ar-
chitectural materials and to culturally important ar-
ticles. At this time, data limitations preclude the abil-
ity to quantify benefits for all materials whose dete-
rioration may have been promoted and accelerated by
air pollution exposure. However, this analysis does
address one small effect in this category, the soiling
of households by paniculate matter. Table D-l 1 docu-
ments the function used to associate nationwide PM-
10 levels with household willingness to pay to avoid
the cleaning costs incurred for each additional ug/m3
ofPM-10. ;
Visibility
In addition to the health and welfare benefits esti-
mated directly from reduced ambient concentrations
of individual criteria air pollutants, this analysis also
estimates the general visibility improvements attrib-
uted to improved air quality. Visibility effects are
measured in terms of changes in DeciView, a mea-
sure useful for comparing the effects of air quality on
visibility across a range of geographic locations for a
range of time periods. It is directly related to two other
common visibility measures, visual range (measured
in km) and light extinction (measured in km'1); how-
ever, it characterizes visibility in terms of perceptible
changes in haziness independent of baseline condi-
tions.
Visibility conditions under the control and no-
control scenarios were modeled separately for the east-
ern and western U.S. In the east, the Regional Acid
Deposition Model (RADM) generated extinction co-
efficient estimates for each of 1,330 grid cells in the
RADM domain (essentially the eastern half of the
country). The extinction coefficients were translated
to DeciView using the relationship reported in
Pitchford and Malm (1994). In the Western U.S., a
conventional extinction budget approach provided
DeciView estimates for 30 metropolitan areas (SAI,
1994). A linear rollback model provided the corre-
sponding no-control estimates. Visibility estimates for
both portions of the country were generated for the
target years 1975, 1980, 1985, and 1990.
Table D-l2 summarizes the methodology used to
predict visibility benefits attributable to the CAA.
Physical benefits for a given year are reported in terms
of the average DeciView change per person in the
modeled population.
Worker Productivity
Available data permits quantification of a final
human welfare endpoint, worker productivity. Crocker
and Horst (1981) and U.S. EPA (1994c) present evi-
dence regarding the inverse relationship between
ozone exposure and productivity in exposed citrus
workers. This analysis applies the worker productiv-
ity relationship (reported as income elasticity with
respect to ozone) to outdoor workers in the U.S. (ap-
proximately one percent of the population). Table D-
12 details the form of the concentration response func-
tion.
Ecological Effects
It is likely that the air pollution reductions
achieved under the CAA resulted in improvements in
the health of aquatic and terrestrial ecosystems. To
the extent that these ecosystems provide a variety of
services (e.g., fishing, timber production, and recre-
ational opportunities), human welfare benefits also
accrued. However, due to a lack of quantified con-
centration-response relationships (or a lack of infor-
mation concerning affected population), ecological
effects were not quantified in this analysis. Appendix
E provides discussion of many of the important eco-
logical benefits which may have accrued due to his-
torical implementation of the CAA.
D-41
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
S-a
213
co
lllil
2 r". W
i
$
S
•y N^' ' •
•s
'5
a
Betadistributi
mean = $2.52
s.e. = $1.00
o? B
.53 £•§ 8 CO.
1
o
3
'53
1
, ,' .
£
."'-#
I
T3 .
.
0.04
•g
S
1
o
a
>**
u
I
.
•8
a|
lltl
D-42
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
§
I
*
I
i
S
I
O
t
^
S i£
I
1
•8
I
"S
60
o
•a
o
§
g
^H W ^5 -_.
I
*
1
&
•8
CO
t*
o
D-43
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Modeling Results
This section summarizes results of the health and
welfare effects modeling. As indicated previously, the
Project Team adopted a Monte Carlo approach in an
effort to capture uncertainty in the benefits analysis.
With respect to estimating avoided incidence of ad-
verse health and welfare effects, two sources of vari-
ability are considered. The first is the statistical un-
certainty associated with each concentration-response
relationship reported in the literature. In addition to
an estimate of a concentration-response function co-
efficient, studies typically report a standard error of
the reported estimate. The second source of uncer-
tainty lies in the choice of studies, where multiple stud-
ies offer estimates for the same endpoint. Different
published results reported in the scientific literature
typically do not report identical findings; in some in-
stances the differences are substantial. This between-
study variability is captured by considering the range
of estimates for a given endpoint.
Table D-13 summarizes health and welfare effects
for each study included in the analysis. The values
presented are mean estimates of the number of cases
of each endpoint avoided due to implementation of
the CAA. A distribution is associated with each mean
estimate, capturing the uncertainty inherent in the es-
timate of the concentration-response coefficient. The
distribution of estimated effects corresponding to a
given study was generated by randomly sampling from
the distribution of coefficients (given by the estimated
coefficient and its standard error reported in the study)
and evaluating the concentration-response function,
yielding an estimate of avoided incidence for the given
effect. This procedure was repeated many times. While
only the central estimates of the resulting distribu-
tions are presented here, the distributions were retained
for use in monetizing and aggregating economic ben-
efits (see Appendix I).7
As shown, for some health endpoints more than
one concentration-response function was used, each
representing a different study. The alternative con-
centration-response functions provide differing mea-
sures of the effect. These can be used to derive a range
of possible results. In the case of lead (Pb), alterna-
tive functions were not used; rather, two analytical
procedures were implemented (labeled the "backward-
looking" and "forward looking" analyses), giving a
range of results for most Pb endpoints (see Appendix
G for discussion of Pb health effects).
The table presents the results of modeling "all U.S.
population" (although, with the exception of Pb, not
all of the 48 state population is modeled, with up to
five percent being excluded in a given year). The re-
sults depict the pattern of health effects incidence
across years. The accuracy of the scale of incidence
is less certain (due to the extrapolation of air quality
data). These results are almost certainly more accu-
rate than the corresponding "50 km" results, but rely
on the assumption that (for a portion of the popula-
tion) distant air quality monitors provide a reason-
able estimate of local air quality conditions. Thus, the
results presented here are somewhat speculative. It is
likely that the estimated health effects are overstated
for that population group (20 to 30 percent of total
population in the case of PM) for which distant moni-
tors are used. (Note, however, that the scaling of
unmonitored county PM concentrations based on re-
gional-scale grid model projections significantly miti-
gates this potential overestimation in the case of PM;
see Appendix C for details). Conversely, there is an
implied zero health impact for that portion of the popu-
lation (three to four percent in the case of PM) ex-
cluded from the analysis altogether, an understatement
of health impacts for that group.
The results indicate the growth of benefits over
the study period, consistent with increasing improve-
ments in air quality between the control and no-con-
trol scenarios from 1970 to 1990.
The mortality effects documented above can be
disaggregated by age. Table D-14 indicates the esti-
mated proportions of premature mortalities for vari-
ous age groups (Pb-induced mortality estimates for
children, men, and women are grouped). Also pre-
sented is the average life expectancy for each group,
indicating the degree of prematurity of PM and Pb-
related mortality.
Table D-15 presents estimated incidence reduc-
tions for several health effects which could be quanti-
fied but not monetized for this analysis.
7 With the exception of visibility, welfare endpoints estimated economic benefits directly and are therefore included in the
monetary benefits results presented in Appendix I.
__
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
mean- estimates). - *,
' ' '* ' *''
ffects ^BxteapSl . ,
'--,,- '„',,,,-, , ; -; * ",-•>'».,•>-''- - *' -%---, -,s*,~
* " '
Endpolnt Study
MORTALITY
Mortality (long-term exposure) Pope et al., 1 995
Mortality (Pb exposure) -Male Average of Backward & Forward
Mortality (Pb exposure) -Female Average of Backward & Forward
Mortality (Pb exposure) -Infant Average of Backward & Forward
CHRONIC BRONCHITIS
Chronic Bronchitis Schwartz, 1993b
Abbey etal., 1993
OTHER Pb-INDUCED AILMENTS
Lost IQ Points Average of Backward & Forward
IQ < 70 Average of Backward & Forward
Hypertension-Men Average of Backward & Forward
Cor. Heart Disease Average of Backward & Forward
Atherothrombotic brain infarction - Men Average of Backward & Forward
Atherothrombotic brain Infarction - Women Average of Backward & Forward
Initial cerebrovascular accident - Men Average of Backward & Forward
Initial cerebrovascular accident - Women Average of Backward & Forward
HOSPITAL ADMISSIONS
All Respiratory Schwartz, 1 995, Tacoma
Schwartz, 1996, Spokane
Pope, 1991, Salt Lake Valley
Schwartz, 1 995, New Haven
Thurston et al., 1 994, Toronto
COPD + Pneumonia Schwartz, 1 994c
Schwartz, 1996, Spokane
Schwartz, 1994a
Schwartz, 1994b
Ischemic Heart Disease Schwartz and Morris, 1 995
Congestive Heart Failure Schwartz and Morris, 1995
Morris etal., 1995
OTHER RESPIRATORY-RELATED AILMENTS
-Adults
Any of 1 9 Acute Symptoms Krupnick et al., 1 990
-Children
Shortness of breath, days Ostro etal., 1995
Acute Bronchitis Dockery et al., 1 989
Lower Respiratory Symptoms Schwartz et al., 1 994d
Upper Respiratory Symptoms Pope et ai., 1991
•All Ages
Asthma Attacks Ostro et al., 1 991
Whittemore and Kom, 1980;
EPA ,1983
Increase in Respiratory Illness Hasselbladi 1992
Any Symptom Linn et al. (1 987, 1 988, 1 990)
RESTRICTED ACTIVITY AND WORK LOSS DAYS
RAD Ostro, 1987
MRAD Ostro and Rothschild, 1 989
RRAD Ostro and Rothschild, 1 989
Work Loss Days Ostro, 1 987
HUMAN WELFARE
Household Soiling Damage ESEERCO, 1 994
Visibility - East (DeciView chg. per person) Pitchford and Malm, 1 994
Visibility - West (DeciView chg. per person) Pitchford and Malm, 1 994
Pollutant's)
PM,o
Pb
Pb
Pb
PM,o
PMio.
Pb
Pb
Pb
Pb
Pb
Pb
Pb
Pb
PM,0&03
PM,0&O3
™10
PM10&O3
PM10&O3
PM10&03
PM10&O3
PM,0&03
PM,0 & O3
P^o
PM,0
CO
PM10 & O3
PM,0
PM,o
PM10
PM10
PM10
O3
NO2
S02
PM10
PM10&O3
PM,0&O3
PMW
1975 1980
58,764 145,884
822 5,281
231 1,474
456 2,342
198,973 554,632
173,571 454,309
1,028,492 5,031,157
3,780 20,074
830,299 5,276,999
1,313 . 8,444
181 1,128
84 529
260 1 ,635
120 758
32,004 77,827
29,393 69,449
30,982 73,093
23,137 55,096
13,746 32,383
21,898 53,928
19,769 47,294
'16,942 40,882
13,006 30,679
6,348 14,709
5,733 13,365
3,022 8,543
41,631,456 98,876,110
20,752,402 50,758,872
1,936,260 6,255,801
2,994,048 6,100,276
500,395 1,292,922
264,430 548,306
193 . 482
729,306 2,686,813
104,896 319,192
19,170,337 47,445,314
60,871,610 155,799,151
47,669,732 237,799,482
6,966,775 17,213,581
1985
169,642
10,340
2,866
3,933
720,166
564,753
8,559,426
36,520
10,087,115
16,671
2,165
1,020
3,154
1,466
95,435
93,137
86,407
66,385
39,691
64,217
63,116
49,290
37,434
17,289
15,742
17,028
117,275,400
58,575,484
7,644,924
6,977,680
1,557,177
686,953
816
6,113,639
282,846
56,939,271
190,333,140
176,850,171
20,648,906
1990
183,539
12,819
3,537
4,944
741,775
602,990
10,378,268
45,393
12,646,876
21 ,069
2,690
1,255-
3,926
1,804
106,777
119,290
95,486
73,842
46,013
70,528
80,113
55,227
43,410
19,098
17,362
21,835
129,529,717
68,375,216
8,541,833
7,804,860
1,683,854
841,916
1,080
9,776,267
265,650
62,187,720
209,924,785
174,329,691
22,562,752
PM]0 direct economic valuation
DeciView
DeciView
0.4 1.4
2.4 4.9
1.9
5.0
2.0
6.0
Decreased Worker Productivity Crocker & Horst, 1 981 and EPA, 1 994cO3 direct economic valuation
Agriculture (Net Surplus) Minimum Estimate
Maximum Estimate
O3 direct economic valuation
O3 direct economic valuation
D-45
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
•, ' -if ^sstf* " **."*,.) ' '•''V^**'**^'" J "* v * v* " ;'/ ' ""
Table D-14; lyIbrtaIify^isttibutiod^by-Age:'Rcop|>«ioaof PM- a
, °"".'-L>- 12% (15%)
Notes;
• Distributiott of premature mortalities acrossa||s;is fairly edrisistenfacross years',
11 PM-xetated roortality jmdem&esgroated only foiyndi-SiduilsSC
v* v/'-';- >>->'%&%\^\ * ' -^ * *rttt^>^^'^'
consisteftt'witb tha population siudleo by Jpp{» «t ai., iyy5.'
4 Pb-related'ni6rteliQ''ittcideic& was^stlrnated fo|,|iifenfs," women, aged 45-74, and
'tbiee'age groups (40-54, 55-64',' 65'-74^ eacfe wl^a^stlnct co»ce.nttalfon4espo,nsg,. ------
_ . , -- "' ,„•*£%&/' • •'-'.--"'- .,„---'- '-.'"'*,,' ,v,
relationship. l\^.»-,-**r" ',,^«-;'•"•**'':
- -^^ H " . ^^^-o-sr^^-^ ^ ^^,.^^ ^ % * "*"* M
D-46
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
'StiMly
Pnlin<«ary
fever
OS'
53 • i2i "i$r
'•>_,' ' A>J <•'*•• \s
""'%'-:- ,87 " 441
iecreasedjPEV (per year]
mittion
decreas
million ewes/year "
" 1
lirij'eto Oaset 0JfsAfl^aa Allred,
PaSn ' '"'"'"' '' ;
-0.^% 0.7% '0.8%'
'" '*"
"f'f o •.' , s ,
fractional -mcf&nse'm •
time tf m $# j^emginff
_
D-47
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Human Health and Welfare
Effects References
Abbey, D.E, F. Petersen, P. K. Mills, and W. L.
Beeson. 1993. "Long-term Ambient Concen-
trations of Total Suspended Particulates,
Ozone, and Sulfur Dioxide and Respiratory
Symptoms in a Nonsmoking Population."
Archives of Environmental Health 48(1): 33-
46.
Allred, E.N., E.R. Bleecker, B.R. Chaitman, T.E.
Dahms, S.O. Gottlieb, J.D. Hackney, M.
Pagano, R.H. Selvester, S.M. Walden, and J.
Warren. 1989a. "Short-term Effects of Car-
bon Monoxide Exposure on the Exercise Per-
formance of Subjects with Coronary Heart
Disease." New England Journal of Medicine
321: 1426-1432.
Allred, E.N., E.R. Bleecker, B.R. Chaitman, T.E.
Dahms, S.O. Gottlieb, J.D. Hackney, D.
Hayes, M. Pagano, R.H. Selvester, S.M.
Walden, and J. Warren. 1989b. Acute Effects
of Carbon Monoxide Exposure on Individu-
als with Coronary Heart Disease. Health Ef-
fects Institute, Cambridge, MA. Research
Report No 25.
Allred, E.N., E.R. Bleecker, B.R. Chaitman, T.E.
Dahms, S.O. Gottlieb, J.D. Hackney, M.
Pagano, R.H. Selvester, S.M. Walden, and J.
Warren. 1991. "Effects of Carbon Monoxide
on Myocardial Ischemia." Environmental
Health Perspectives 91:89-132.
American Heart Association. 1991. 1992 Heart and
Stroke Facts.
Avol, E.L., W.S. Linn, T.G. Venet, D.A. Shamoo, and
J.D. Hackney. 1984. "Comparative Respira-
tory Effects of Ozone and Ambient Oxidant
Pollution Exposure During Heavy Exercise."
Journal of the Air Pollution Control Associa-
tion 34: 804-809.
Centers for Disease Control (CDC). 1992. Vital and
Health Statistics, Detailed Diagnoses and Pro-
cedures, National Hospital Discharge Survey,
1990. (Series 13, Number 113).
Crocker T. D., and R. L. Horst, Jr. 1981. "Hours of
Work, Labor Productivity, and Environmen-
tal Conditions: a Case Study." The Review of
Economics and Statistics 63:361-368.
Dockery, D.W., F.E. Speizer, D.O. Stram, J.H. Ware,
J.D. Spengler, and E.G. Ferris, Jr. 1989. "Ef-
fects of Inhalable Particles on Respiratory
Health of Children." Am. Rev. Respir. Dis.
139: 587-594.
Dockery, D.W., et al. 1993. "An Association Between
Air Pollution and Mortality in Six U.S. Cit-
ies." The New England Journal of Medicine
329 (24): 1753-1759.
Dockery, D.W., J. Schwartz, and J.D. Spengler. 1992.
Air Pollution and Daily Mortality: Associa-
tions with Particulates and Acid Aerosols.
Environ. Res. 59: 362-373.
Empire State Electric Energy Research Corporation
(ESEERCO). 1994. New York State Environ-
mental Externalities Cost Study. Report 2:
Methodology. Prepared by: RCG/Hagler,
Bailly, Inc., November.
Hasselblad, V., D.M. Eddy, and D.J. Kotchmar. 1992.
"Synthesis of Environmental Evidence: Ni-
trogen Dioxide Epidemiology Studies."/. Air
Waste Mgmt. Assoc 42: 662-671.
Health Effects Institute (HEI). 1996. Air Pollution and
Mortality in Philadelphia, 1974-1988. Con-
tributors: Samet, J.M., S.L. Zeger, K. Berhane,
and J. Xu.
Ito, K. and G.D. Thurston. 1996. Daily PM /Mortal-
ity Associations: An Investigation o!f At-Risk
Sub-Populations. /. Exposure Anal. Environ.
Epidemiol.: in press.
Ito, K, P. Kinney, and G.D. Thurston. 1995. Varia-
tions on PM Concentrations Within Two
Metropolitan Areas and Their Implications for
Health Effects Analyses. In: Phalen, R.F. and
D.V. Bates, eds. Proceedings of the Collo-
quium on Particulate Air Pollution and Hu-
man Mortality and Morbidity, Part II; Janu-
ary 1994; Irvine, CA. Inhalation Toxical. 7:
735-745.
D-48
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
Kinney, P.L., K. Ito, and G.D. Thurston. 1995. A Sen-
sitivity Analysis of Mortality/PM Associa-
tions in Los Angeles. In: Phalen1,0 R.F. and
D.V. Bates, eds. Proceedings of the Collo-
quium on Particulate Air Pollution and Hu-
man Mortality and Morbidity; January 1994;
Irvine, CA. Inhalation Toxicol. 1: 59-69.
Krupnick, A.J. and R.J. Kopp. 1988. "The Health and
Agricultural Benefits of Reductions in Am-
bient Ozone in the United States." Resources
for the Future Discussion Paper QE88-10,
Washington, DC. August.
Krupnick A.J., W. Harrington, and B. Ostro. 1990.
"Ambient Ozone and Acute Health Effects:
Evidence from Daily Data." Journal of Envi-
ronmental Economics and Management 18:1-
18.
Lave, L.B. and E.P. Seskin. 1977. Air Pollution and
Human Health. Johns Hopkins University
Press for Resources for the Future. Baltimore
and London.
Linn, W.S., E.L. Avol, R. Peng, D.A. Shamoo, and
J.D. Hackney. 1987. "Replicated Dose-Re-
sponse Study of Sulfur Dioxide Effects in
Normal, Atopic, and Asthmatic Volunteers."
Am RevRespirDis. 136: 1127-34.
Linn, W.S., E.L. Avol, D.A. Shamoo, R. Peng, C.E.
Spier, M.N. Smith, and J.D. Hackney. 1988.
"Effect of Metaproterenol Sulfate on Mild
Asthmatics' Response to Sulfur Dioxide Ex-
posure and Exercise." Archives of Environ-
mental Health 43(6): 399-406.
Linn, W.S., D.A. Shamoo, R. Peng, K.W. Clark, E.L.
Avol, and J.D. Hackney. 1990. "Responses
to Sulfur Dioxide and Exercise by Medica-
tion-Dependent Asthmatics: Effect of Vary-
ing Medication Levels." Archives of Environ-
mental Health 45(1): 24-30.
Lipfert, F.W. 1984. "Air Pollution and Mortality:
Specification Searches Using SMSA-based
Data." J. Environ. Econ. Manage. 11: 208-
243.
Manuel, E.H., R.L. Horst, K.M. Brennan, W.N. Lanen,
M.C. Duff, and J.K. Tapiero. 1982. Benefits
Analysis of Alternative Secondary National
Ambient Air Quality Standards for Sulfur Di-
oxide and Total Suspended f 'articulates, Vol-
umes I-IV. Prepared for U.S. Environmental
Protection Agency, Office of Air Quality
Planning and Standards, Research Triangle
Park, NC. [Cited in ESEERCO, 1994].
McClelland, G., W. Schulze, D. Waldman, J. Irwin,
D. Schenk, T. Stewart, L. Deck, and M.
Thayer. 1991. Valuing Eastern Visibility: a
Field Test of the Contingent Valuation Method
(Draft). Prepared for U.S. Environmental Pro-
tection Agency, Washington, D.C.
Melia, R.J.W., C du V Florey, S. Chinn, B.D.
Goldstein, A.G.F. Brooks, H.H. John, D.
Clark, I.B. Craighead, and X. Webster. 1980.
"The Relation Between Indoor Air Pollution
from Nitrogen Dioxide and Respiratory Ill-
ness in Primary Schoolchildren." Clinical
Respiratory Physiology 16:7P-8P.
Melia, R.J.W., C du V Florey, R.W. Morris, B.D.
Goldstein, H.H. John, D. Clark, I.E.
Craighead, and J.C. Mackinlay. 1982. "Child-
hood Respiratory Illness and the Home Envi-
ronment: n. Association Between Respiratory
Illness and Nitrogen Dioxide, Temperature,
and Relative Humidity." International Jour-
nal ofEpidemiology 11: 164-169.
Melia, R.J.W., C du V. Florey, and Y. Sittampalarm.
1983. The Relationship Between Respiratory
Illness in Infants and Gas Cooking in the UK:
a Preliminary Report. Air Quality Vth World
Congress: Proceedings of the International
Union of Air Pollution Prevention Associa-
tions. SEPIC (APPA), Paris France, pp. 263-
269. -
Morris, R.D., E.N. Naumova, and R.L. Munasinghe.
1995. "Ambient Air Pollution and Hospital-
ization for Congestive Heart Failure Among
Elderly People in Seven Large U.S. Cities."
American Journal of Public Health 85(10):
1361-1365.
D-49
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Ostro, B.D. 1987. "Air Pollution and Morbidity Re-
visited: a Specification Test." /. Environ.
Econ. Manage. 14: 87-98.
Ostro, B.D. and S. Rothschild. 1989. "Air Pollution
and Acute Respiratory Morbidity: An Obser-
vational Study of Multiple Pollutants." Envi-
ronmental Research 50:238-247.
Ostro, B.D., MJ. Lipsett, M.B. Wiener, and J.C.
Seiner. 1991. "Asthmatic Responses to Air-
borne Acid Aerosols." American Journal of
Public Health 81: 694-702.
Ostro, B.D., J.M. Sanchez, C. Aranda, and G.S.
Eskeland. 1995. "Air Pollution and Mortal-
ity: Results of a Study of Santiago, Chile." /.
Exposure Anal. Environ. Epidemiol, submit-
ted.
Ostro, B.D., J.M. Sanchez, C. Aranda, and G.S.
Eskeland. 1996. Air Pollution and Mortality:
Results from a Study of Santiago, Chile. In:
Lippmann, M. ed. Papers from the ISEA-ISEE
Annual Meeting; September 1994; Research
Triangle Park, NC./. Exposure Anal. Environ.
Epidemiol.: in press.
Ozkaynak, H., J. Xue, P. Severance, R. Burnett, and
M. Raizenne. 1994. Associations Between
Daily Mortality, Ozone, and Particulate Air
Pollution in Toronto, Canada. Presented at:
Colloquium on Particulate Air Pollution and
Human Mortality and Morbidity: Program
and Abstracts; January; Irvine, CA. Irvine,
CA: University of California Irvine, Air Pol-
lution Health Effects Laboratory; p. PI.13;
report no. 94-02.
Pitchford, M.L. and W.C. Malm. 1994. "Development
and Applications of a Standard Visual Index."
Atmospheric Environment 28(5): 1049-1054.
Pope, C.A., HI. 1991. "Respiratory Hospital Admis-
sions Associated with PM Pollution in Utah,
Salt Lake, and Cache Valleys." Arch. Environ.
Health 46 (2): 90-97.
Pope, C.A., HI, and D.W. Dockery. 1992. "Acute
Health Effects of PM Pollution on Symp-
tomatic and Asymptomatic Children." Am.
Rev. Respir. Dis. 145: 1123-1128.
Pope, C.A., IE, D.W. Dockery, J.D. Spengler, and
M.E. Raizenne. 1991. "Respiratory Health
and PM Pollution: a Daily Time Series
Analysis!* Am. Rev. Respir. Dis. 144: 668-
674.
Pope, C.A., HI, and L.S. Kalkstein. 1996. Synoptic
Weather Modeling and Estimates of the Ex-
posure-Response Relationship Between Daily
Mortality and Particulate Air Pollution.
Environ. Health Perspect. 104: in press.
Pope, C.A., HI, J. Schwartz, and M.R. Ransom. 1992.
Daily Mortality and PM Pollution in Utah
Valley. Arch. Environ. Health 47: 211-217.
Pope, C.A., IH, MJ. Thun, M.M. Namboodiri, D.W.
Dockery, J.S. Evans, F.E. Speizer, and C.W.
Heath, Jr. 1995. Particulate Air Pollution as a
Predictor of Mortality in a Prospective Study
of U.S. Adults.Am. J. Respir. Crit. CareMed.
151:669-674. .
Portney, P.R. and J. Mullahy. 1990. "Urban Air Qual-
ity and Chronic Respiratory Disease." Re-
gional Science and Urban Economics 20:407-
418.
Roger, L.J., H.R. Kehrl, M. Hazucha, and D.H.
Horstman. 1985. "Bronchoconstriction in
Asthmatics Exposed to Sulfur Dioxide Dur-
ing Repeated Exercise." J. Appl. Physiol.
59(3): 784-791.
Saldiva, P.H.N., Pope, C.A., III, J. Schwartz, D.W.
Dockery, A.J. Lichtenfels, J.M. Salge, I.
Barone, andG.M. Bohm. 1995. Air Pollution
and Mortality in Elderly People: A Time-Se-
ries Study in Sao Paulo, Brazil. Arch. Environ.
Health 50: 159-163.
Samet, J.M., Lambert, W.E., Skipper, B.J., Gushing,
A.H., Hunt, W.C., Young, S.A., McLaren,
L.C., Schwab, M., and J.D. Spengler. 1993.
"Nitrogen Dioxide and Respiratory Illnesses
hi Infants." Am Rev. Respir. Dis. 148: 1258-
1265.
Schwartz, J. 1993a. Air Pollution and Daily Mortal-
ity in Birmingham, Alabama. Am. J.
Epidemiol. 137: 1136-1147.
D-50
-------
Appendix D: Human Health and Welfare Effects of Criteria Pollutants
Schwartz, J. 1993b. "Particulate Air Pollution and
Chronic Respiratory Disease." Environmen-
tal Research 62:1-13.
Schwartz, J. 1994a. "Air Pollution and Hospital Ad-
missions in Elderly Patients in Birmingham,
Alabama." American Journal of Epidemiol-
ogy 139:589-98.
Schwartz, J. 1994b. "Ah- Pollution and Hospital Ad-
missions for the Elderly in Detroit, Michigan."
American Journal of Respiratory Care Med
150:648-55.
Schwartz, J. 1994c. "PM , Ozone and Hospital Ad-
missions for the Elderly in Minneapolis-St.
Paul, Minnesota." Archives of Environmen-
tal Health 49(5): 366-374.
Schwartz, J. 1994d. "Acute Effects of Summer Air
Pollution on Respiratory Symptom Report-
ing in Chlldren."Am JRespir Crit Care Med.
150: 1234-1242.
Schwartz, J. 1995. "Short Term Fluctuations hi Air
Pollution and Hospital Admissions of the Eld-
erly for Respiratory Disease." Thorax 50:531-
538.
Schwartz, J. 1996. "Air Pollution and Hospital Ad-
missions for Respiratory Disease." Epidemi-
ology 7(1): 1-9.
Schwartz, J., D.W. Dockery, L.M. Neas. 1996a. Is
Daily Mortality Associated Specifically with
Fine Particles? /. Air Waste Manage. Assoc.:
accepted.
Schwartz, J. and R. Morris. 1995. "Air Pollution and
Cardiovascular Hospital Admissions."Am. /.
Epidemiol. 142: 23-35.
Seal, E., W.F. McDonnell, D.E. House, S.A. Salaam,
PJ. Dewitt, S.O. Butler, J. Green, and L.
Raggio. 1993. "The Pulmonary Response of
White and Black Adults to Six Concentrations
of Ozone." Am. Rev. Respir. Dis. 147: 804-
810.
Styer, P., N. McMillan, F. Gao, J. Davis, and J. Sacks.
1995. The Effect of Airborne Particulate
Matter in Daily Death Counts. Environ.
Health Perspect. 103:490-497.
Systems Application International (SAI). 1994. Ret-
rospective Analysis of the Impact of the Clean
Air Acton Urban Visibility in the Southwest-
ern United States. Prepared for the U.S. En-
vironmental Protection Agency, Office of Air
and Radiation. October 31.
Thurston, G., K. Ito, C. Hayes, D. Bates, and M.
Lippmann. 1994. "Respiratory Hospital Ad-
mission and Summertime Haze Air Pollution
in Toronto, Ontario: Consideration of the Role
of Acid Aerosols." Environmental Research
65: 271-290.
U.S. Environmental Protection Agency (U.S. EPA).
1985. Costs and Benefits of Reducing Lead
in Gasoline: Final Regulatory Impact Analy-
sis. Office of Policy Analysis. Washington,
DC. EPA-230-05-85-006.
U.S. Environmental Protection Agency (U.S. EPA).
1986. Air Quality Criteria for Particulate
Matter: Updated Assessment of Scientific and
Technical Information Addendum to the 1982
OAQPS Staff Paper. Prepared by the Office
of Air Quality Planning and Standards, Re-
search Triangle Park, North Carolina. EPA
450/05-86-012.
U.S. Environmental Protection Agency (U.S. EPA).
199la. Acid Rain Benefit Assessment: Draft
Plan for the [Section 812] 1992 Assessment,
Acid Rain Division, Washington, DC.
U.S. Environmental Protection Agency (U.S. EPA).
1991b. Air Quality Criteria for Carbon Mon-
oxide. EPA-600/8-90/045F, U.S. Environ-
mental Protection Agency, Office of Health
and Environmental Assessment, Environmen-
tal Criteria and Assessment Office; Research
Triangle Park, NC.
U.S. Environmental Protection Agency, 1993. Docu-
mentationfor Oz-One Computer Model (Ver-
sion 2.0). Office of Air Quality Planning and
Standards. Prepared by: Mathtech, Inc., un-
der EPA Contract No. 68D830094, WA 59.
July.
D-51
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
U.S. Environmental Protection Agency (U.S. EPA).
1993b. External Draft, Air Quality Criteria for
Ozone and Related Photochemical Oxidants.
Volume n. Office of Health and Environmen-
tal Assessment, Environmental Criteria and
Assessment Office, Research Triangle Park,
NC; EPA/600/AP-93/004b. 3v.
U.S. Environmental Protection Agency (U.S. EPA).
1994a. Review of the National Ambient Air
Quality Standards for Sulfur Oxides: Assess-
ment of Scientific and Technical Information.
Supplement to the 1986 OAQPS Staff Paper
Addendum. Air Quality Management Divi-
sion, Office of Air Quality Planning and Stan-
dards. Research Triangle Park, NC. EPA Re-
port No. EPA-452/R-94-013.
U.S. Environmental Protection Agency (U.S. EPA).
1994b. Supplement to the Second Addendum
(2986) to Air Quality Criteria for Paniculate
Matter and Sulfur Oxides (1982): Assessment
of New Findings on Sulfur Dioxide Acute Ex-
posure Health Effects in Asthmatics (Exter-
nal Review Draft). Environmental Criteria and
Assessment Office. Research Triangle Park,
NC. EPA/600/AP-93/002.
U.S. Environmental Protection Agency (U.S. EPA).
\994c.Documentationfor Oz-One Computer
Model (Version 2.0). Office of Air Quality
Planning and Standards. Prepared by:
Mathtech, Inc., under EPA Contract No.
68D30030, WA 1-29. August.
Ware J.H., D.W. Dockery, A. Sprio m, F.E. Speizer,
and E.G. Ferris, Jr. 1984. "Passive Smoking,
Gas Cooking, and Respiratory Health of Chil-
dren Living in Six Cities." American Review
of Respiratory Disease 129:366-374.
Watson, W. and J. Jaksch. 1982. "Air Pollution:
Household Soiling and Consumer Welfare
Losses." Journal of Environmental Econom-
ics and Management. 9: 248-262.
Whittemore, A. S., and E. L. Korn. 1980. "Asthma
and Air Pollution in the Los Angeles Area."
American Journal of Public Health 70:687-
696.
World Health Organization (WHO). l996.Final Con-
sultation on Updating and Revision of the Air
Quality Guidelines for Europe. Bilthoven,
The Netherlands 28-31 October, 1996 ICP
EHH018VD962.il.
D-52
-------
Appendix E: Ecological Effects of Criteria
Pollutants
Introduction
Benefits to human welfare from air pollution re-
ductions achieved under the CAA can be expected to
arise from likely improvements in the health of aquatic
and terrestrial ecosystems and the myriad of ecologi-
cal services they provide. For example, improvements
in water quality stemming from a reduction in acid
deposition-related air pollutants (e.g., SOx and NOx)
could benefit human welfare through enhancements
in certain consumptive services such as commercial
and recreational fishing, as well as non-consumptive
services such as wildlife viewing, maintenance of
biodiversity, and nutrient cycling. Increased growth
and productivity of U.S. forests could result from re-
duced emissions of ozone-forming precursors, particu-
larly VOCs and NOx, and thus may yield benefits from
increased timber production; greater opportunities for
recreational services such as hunting, camping, wild-
life observation; and nonuse benefits such as nutrient
cycling, temporary CO2 sequestration, and existence
value.
In this Appendix, the potential ecological benefits
from CAA pollutant controls are discussed in the con-
text of three types of ecosystems: aquatic, wetland,
and forest. In describing the potential ecological ben-
efits of the CAA, it is clearly recognized that this dis-
cussion is far from being comprehensive in terms of
the types and magnitude of ecological benefits that
may actually have occurred from the implementation
of the CAA. Rather, this discussion reflects current
limitations in understanding and quantifying the link-
ages which exist between air quality and ecological
services, in addition to limitations hi the subsequent
valuation of these services in monetary terms. This
discussion also does not cover potential benefits from
improvements in other ecological services, namely ag-
riculture and visibility, which are discussed and quan-
tified in other sections of this report. This appendix
is dedicated to a qualitative evaluation of ecological
benefits. However, where possible, the existing body
of scientific literature is drawn upon in an attempt to
provide insights to the possible magnitude of benefits
that may have resulted from CAA-related improve-
ments of selected ecological services. It is important
to note that the inability to fully value ecological ser-
vices results in a significant undervaluation of the
ecological benefits of air pollution reductions. This
undervaluation should not be interpreted as a devalu-
ation.
Benefits From Avoidance of
Damages to Aquatic Ecosystems
Aquatic ecosystems (lakes, streams, rivers, estu-
aries, coastal areas) provide a diverse range of ser-
vices that benefit the welfare of the human popula-
tion. Commercially, aquatic ecosystems provide a
valuable food source to humans (e.g., commercial fish
and shellfish harvesting), are used for the transporta-
tion of goods and services, serve as important drink-
ing water sources, and are used extensively for irriga-
tion and industrial processes (e.g., cooling water, elec-
trical generation). Recreationally, water bodies pro-
vide important services that include recreational fish-
ing, boating, swimming, and wildlife viewing. They
also provide numerous indirect services such as nu-
trient cycling, and the maintenance of biological di-
versity.
Clearly, these and other services of aquatic eco-
systems would not be expected to be equally respon-
sive to changes in air pollution resulting from the
implementation of the CAA. The available scientific
information suggests that the CAA-regulated pollut-
ants that can be most clearly linked to effects on
aquatic resources include SOx and NOx (through acid
deposition and increases in trace element
bioavailability), NOx (through eutrophication of ni-
trogen-limited water bodies), and mercury (through
changes in atmospheric deposition). Potential ben-
efits from each of these processes (acid deposition,
eutrophication, mercury accumulation in fish) are
described separately in the following sections.
E-l
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Acid Deposition
Background
Acid deposition refers to the depositing of strong
acids (e.g., H2SO4, HNO3) and weak acids «NH4)2SO4,
NH^NOj) from the atmosphere to the earth's surface.
Acid deposition can occur in the wet or dry form and
can adversely affect aquatic resources through the
acidification of water bodies and watersheds. Acidi-
fication of aquatic ecosystems is of primary concern
because of the adverse effects of low pH and associ-
ated high aluminum concentrations on fish and other
aquatic organisms. Low pH can produce direct ef-
fects on organisms, through physiological stress and
toxicity processes, and indirect effects, mediated by
population and community changes within aquatic
ecosystems. Acidification can affect many different
aquatic organisms and communities. As pH decreases
to 5.5, species richness in the phytoplankton, zoop-
lankton, and benthic invertebrate communities de-
creases.1 Additional decreases in pH affect species
richness more significantly, and may sometimes af-
fect overall biomass.2 Table E-l presents descrip-
tions of the biological effects of acidification at dif-
ferent pH levels. In evaluating the severity of bio-
logical changes due to acidification, the reversibility
of any changes is an important consideration; biologi-
cal populations and communities may not readily re-
cover from improved water quality under certain cir-
cumstances. Researchers have addressed acidifica-
tion effects through many different experimental pro-
tocols, including laboratory bioassays, particularly
concerning pH, aluminum, and calcium; manipula-
tive whole-system acidification studies in the field;
and comparative, nonmanipulative field studies.
Although acidification affects phytoplankton,
zooplankton, benthic invertebrates, fish, amphibians,
and waterfowl, most acidification research has con-
centrated on fish populations.3 Aluminum, which can
be toxic to organisms, is soluble at low pH and is
leached from watershed soils by acidic deposition.4
Acidification may affect fish in several ways. The
direct physiological effects of low pH and high alu-
minum include increased fish mortality, decreased
growth, and decreased reproductive potential. The
mechanism of toxicity involves impaired ion regula-
tion at the gill.5 Population losses occur frequently
because of recruitment failure,6 specifically due to
increased mortality of early life stages.7 Changes at
other trophic levels may affect fish populations by
altering food availability.8 Fish in poorly buffered,
low pH water bodies may accumulate higher levels of
mercury, a toxic metal, than in less acidic water bod-
ies, due to increased mercury bioavailability. The
primary consequence of mercury accumulation ap-
pears to be hazardous levels to humans and wildlife
who consume fish, rather than direct harm to aquatic
organisms (discussed further below).
The CAA-regulated pollutants that are likely to
have the greatest effect on aquatic ecosystems through
acid deposition and acidification are SO2 and NOx. In
the atmosphere, SO2 and NOx react to form sulfate
and nitrate particulates, which may be dry-deposited;
also the pollutants may react with water and be wet-
deposited as dilute sulfuric and nitric acids. SO2 is
considered the primary cause of acidic deposition,
contributing 75 to 95 percent of the acidity in rainfall
in the eastern United States.9
Current Impacts of Acid Deposition
Effects on Water Chemistry
The effects of acid deposition and resulting acidi-
fication of water bodies was intensively studied as part
of a 10-year, congressionally-mandated study of acid
rain problems in the United States.10 Based on the
NAPAP study, it is estimated that 4 percent of the
lakes and 8 percent of the streams in acid-sensitive
1 J. Baker et al., NAPAP SOS/T 13,1990; Locke, 1993.
2 J. Baker et al., NAPAP SOS/T 13,1990.
3 NAPAP, 1991.
4 J. Baker et al., NAPAP SOS/T 13,1990.
5 J. Baker et al., NAPAP SOS/T 13,1990.
6 Rosseland, 1986.
7 J. Baker et al., NAPAP SOS/T 13,1990.
8 Mills et al., 1987.
9 NAPAP, 1991.
10 NAPAP. 1991.
E-2
-------
Appendix E: Ecological Effects of Criteria Pollutant;
lea! Conges 'w1th,-S,urface-feter AcMificatiqii,, —' '\,
Decrease
5.5to5.0
^e^easft in speties ridaiess ofpjhytbjilaitooti, 20pgl«akt«i, and berithic
sitiv/spectes, bu{'no mrasurable chan
•., ',„ -s, ,';>vv/ "
V', -• **'"<*', '**>-.,
active succes^'may occur for aciS4eiave-fishaiieci«s!
rttttno^,,stape ''^J'' j * •> ' s 'v-xv'' s s' '"o 4, ' 'ff \
'• f^1* 4veise eff^Cdfc^easdd'^lfe'dBcttVB succes^inay oWiir for aciS^eralSve-fish'ap'ecies feg-StWad."
Loss of se»sitiw' sp^i^offita'l&kand <3ace; socfeas blacknose^coand.
ceo^eased reproductive success (rflafce taut'ai^ walleye ..... " "'";- ""
- -'' "™ ""*-•" ""* ..... y"; lm"~-
)owt ia'some waters "
' '"
Loss Qf^B^Bj« of ooiBmtMi,,ifiy4«ebW-^eoies
'aBlj fe'nthSe in
of
''•> ' '
!aketrosat,
'
.-
fitosn,^iid beaAic iayeiteBf^otantunitlBs^cereases in-the total abundance tad biotnass of bentte' V
jS some Wafers,
,, "^ •> •. ,,
\ < '\"*'-. " ^ .... - -jp -^^--^pn—T -r T--va4.— ,™.fl.w^.ni^^Jr^x,w,^t.jA
teuehtenbergianamtAsptqpcfia prio^otita;,^ snaas,- raSst^ectes of clams,-31$ many
'' "'<>'4"- -'""'•*° '"-
m,ost ish
?-;- --- f ,- - •->-( * * ,,x
rt-ftshspe4es-swb ai^wolf toftt-aad,Atlabtic salmon,
Substa'ntiaXdecrease'in-tbe.nurafeer of species, of zoopiakkton aa8"benftio
clferis ^ndtrtany insects and cru&icean.$;; measurable-^crease in tjje totaV
a»dbeMMc!,i|yertebi'8resi»^ostwat«rsr'>'''-\ • ,-, ' ^•"•'~->~v
Mbiaas sucfe Ss spotted sfi&'a
' ''"' "-,
regions of the U.S. are chronically acidic due to natu-
ral and anthropogenic causes. NAPAP defines acidic
conditions as occurring when the acid neutralizing
capacity11 (ANC) is below 0 ueq/L. Furthermore, ap-
proximately 20 percent of the streams and lakes in
these regions are considered to be extremely suscep-
tible to acidity (defined as ANC <50 ^eq/L) and
slightly more than half show some susceptibility to
acidification (defined as ANC <200 ueq/L).
In terms of the role of acid deposition as a causal
mechanism for the acidification of water bodies, it is
estimated that 75 percent of the 1,181 acidic lakes
and 47 percent of the 4,668 streams studied under
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
NAPAP receive their dominant source of acid anions
from atmospheric deposition (see Table E-2). On a
regional basis, the importance of acid deposition var-
ies considerably, which is believed to result from re-
gional differences in SOx and NOx emissions and dif-
ferences in the biogeochemistry of individual water-
sheds. For acidic lakes (ANC <0), the regions that
appear most likely to be influenced by acid deposi-
tion include the Adirondacks and Mid-Atlantic High-
land region, with acid deposition cited as the domi-
Florida, where the vast majority (79 percent) are acidic
primarily due to organic acids, rather than acid depo-
sition.
Effects on Fish Habitat Quality
By combining information on relevant water
chemistry parameters (pH, aluminum, calcium), fish
toxicity models, and historical and current distribu-
tions of fish populations in the lakes and streams in-
Table E-2. Comparison of Population of Acidif %ioiial'- Surface Water Jurvey (NSWS) by'
Chemical Category'1 «,-- ' - -• - '•••<••• •
Region
V
^
New England
Adirondacks
Mid- Atlantic Highlands , ,„
Southeastern Highlands
Horida
Upper Midwest -
West
AH Lakes
Mid-AtlanticHigblands
Mid-Atlantic Coastal Plain
Southeastern Highlands
Florida
All Streanis
, Number ,ojf,,,, JSeposifiori* ,„„ Organic , •/ Acid Mrnev,,, ,- ' Wa&rsted
"Acidic '. \i ."-Efoaiirated ' , Etojainated " ' ' Drainage, - 'Sauate^
Waters ' MJ^'" ' v / ' "" t,/,/-.;'-""* Dominated , Dominated
' """"' r'ViAKBs, ' ,7--:-,: ; t;'v,;!,^-- - - ' ^ "'"" '
* *•* *i *n rr& "• %'f^''J 0 i J _, ^ J"'t V J -VH- %
" s 1 fJ , ,; vv'11" yy / -.- - ^ __ A* ^ -,v i •pim ^ ^ , - ,77, '
*, ' ^ v . ._; ^ "/ & ' , A^Sf "• " ,/•»'*"'
s ^vvv*, '*"^J , ^^f^^J* *^* J „ , ft A\/,i '',, __ A ' *
- 477 ,, W" - ' ",?^,//, ' ",~ ,- -~ ' 4, , -
* 'IV -- "-' *^ ;„;,--' ' r->-. * *•""**.. A •' '->'<"'/'
1334" " _ 44 '•""'''*," -s-54, ' • ~"^','<1^W, '%',',
-"/«,/--* <".*""" - ; "" <,?;,-;---,-' ' '"
"243-,'--*': 50 ---J--- 50 ' — ,;>'//
;-' 677 , / W2i ' " ;%, '^,'";,7:M",,:":L«~:'- '-,---~'
4^a;" : -• - --47 ' „ ,„ 7€ *w ' • 26 , --^ :rt?
-------
Appendix E: Ecological Effects of Criteria Polh
the survival of acid-sensitive species such as minnows.
About two percent and six percent of the lakes in the
New England region are estimated to be unsuitable
for acid-tolerant and acid-sensitive fish species, re-
spectively. A greater proportion of streams in the Mid-
Atlantic Highland region are estimated to be unsuit-
able for acid-tolerant and acid-resistant fish species
(18 percent and 30 percent, respectively); however,
about 44 percent of streams surveyed in this region
are thought to be heavily influenced by acid mine
drainage (Table E-2).
Economic Damages to Recreational Fishing
In an effort to assess some of the impacts from
existing levels of acid deposition to public welfare,
NAPAP investigated the current economic damages'
associated with acid deposition to trout anglers of New
York, Maine, Vermont, and New Hampshire. The
general approach used consisted of linking the catch
per unit effort (CPUE) for four species of trout at in-
dividual lakes (estimated using participation survey
data) to the relevant water quality conditions at these
lakes (namely, the acid stress index or ASI). Using
historical water quality data, critical water quality
conditions (i.e., the ASI values) were estimated for
lakes in the absence of acid deposition and compared
to current conditions reflecting the presence of acid
deposition. Using two types of travel cost models, the
Random Utility Model (RUM) and Hedonic travel-
cost model (HTCM), estimates of the willingness to
pay (WTP) per trip of sampled trout anglers were ob-
tained. Aggregate estimates of the WTP were obtained
across the populations of trout anglers using statisti-
cal weighting factors. Finally, the difference in total
WTP between the current (acid deposition) scenario
and the historical (acid deposition-free) scenarios was
determined.
The resulting estimates of economic damages to
trout anglers in the four state region are relatively
small. Specifically, damage estimates range from $0.3
million to $1.8 million (in 1989 doliars) for the he-
donic travel-cost and random utility models, respec-
tively. By many accounts, these estimates can be con-
sidered to underestimate actual damages to anglers in
these states. First, data limitations precluded the de-
velopment of meaningful WTP estimates for brook
trout anglers, which may be a significant component
of trout fishing in these areas. Second, resource con-
straints necessitated exclusion of a large population
of trout anglers (i.e., those residing in New York City).
Third, the economic damage estimates were limited
to trout anglers, thus excluding potentially similar if
not greater economic damages to anglers fishing for
other coldwater or warmwater fish species. In addi-
tion, the NAPAP analysis was performed in the con-
text of recreational fishing in lakes, thereby exclud-
ing potentially important welfare impacts from recre-
ational fishing in streams. Finally, these estimates do
not address non-use values of lakes in this region.
Benefits From Acid Deposition Avoidance Under
the CAA
It is currently estimated that in the absence of
pollution reductions achieved under the Clean Air Act,
total sulfur emissions to the atmosphere would have
increased by nearly sixteen million tons by 1990, a 40
percent increase above 1990 levels estimated with
CAA controls remaining in place.13 Based on atmo-
spheric transport and deposition modeling, this in-
crease in sulfur emissions corresponds to an approxi-
mate 25 to 35 percent increase in total sulfur deposi-
tion (wet & dry) in large portions of the northeastern
portion of the United States.14 Given sulfur emission
and deposition changes of this magnitude, and the
importance of sulfur emissions in contributing to acid
deposition, one would expect some benefits to human
welfare to be achieved as a result of improved quality
of aquatic ecosystems. To date, however, no formal
benefits assessment of CAA-avoided acid deposition
impacts has been conducted for aquatic ecosystems.
Nevertheless, past benefit assessments involving acid
deposition impacts on aquatic ecosystems provide
some opportunity to gain insights into the relative
magnitude of certain aquatic-based benefits that may
be achieved through pollution reductions under the
CAA.15
Recreational Fishing
NAPAP evaluated the impact of changes in acid
deposition on use values of aquatic ecosystems (i.e.,
recreational fishing).16 In their integrated assessment,'
NAPAP valued the impacts of three different sulfur-
13 U.S. EPA, 1995; Table B-2.
14 U.S. EPA 1995, p. 3-10.
15 See, for example, NAPAP, 1991.
16 NAPAP, 1991.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
induced acid deposition scenarios to trout anglers from
NY, VT, NH and ME.17 The three scenarios evalu-
ated were:
1. No change in acid deposition.
2. A 50 percent reduction in acid deposition.
3. A 30 percent increase in acid deposition.
As described above, equations were developed by
NAPAP to estimate the catch per hour for species at
each lake as a function of the ASI value for each lake
and of the technique of the fishers. Baseline and pre-
dicted changes in CPUE were evaluated for all lakes
modeled in the region. Willingness-to-pay estimates
for CPUE per trip were derived for the baseline and
sulfur emission scenarios using two travel-cost mod-
els, a random utility model and a hedonic travel cost
model. These willingness-to-pay estimates were then
combined with the results of a participation model
that predicted the total number of trips taken by trout
anglers. Total welfare changes were determined over
a 50 year period (from 1990 to 2040).
At current levels of acid deposition, NAPAP esti-
mates that trout anglers in these four states will expe-
rience annual losses by the year 2030 of $5.3 or $27.5
million (in 1989 dollars) for the random utility model
and hedonic travel cost model, respectively (see Table
E-3). If acid deposition increases by 30 percent, which
roughly corresponds to the 25 to 35 percent increase
predicted for the northeast U.S. in the absence of CAA
sulfur controls,18 the resulting economic losses to trout
anglers in 2030 would range from $ 10 million to nearly
$100 million annually (in 1989 dollars) for the RUM
and HTCM, respectively. If deposition decreases by
50 percent, annual benefits to recreational anglers are
estimated to be $14.7 million (RUM) or $4.2 million
(HTCM).
While an estimation of CAA-related benefits to
trout anglers based on the 30 percent increase in acid
deposition scenario has some appeal, a strict transfer
of these benefits to the section 812 retrospective analy-
sis is hindered by several factors. First, the NAPAP
benefits estimates are projected for future conditions
(the year 2030). Therefore, the extent to which the
NAPAP benefits reflect conditions and benefits in
1990 (the focus of the section 812 retrospective as-
sessment) is unclear. Second, the NAPAP and CAA
section 812 analyses operate from different baselines
(1990 for the NAPAP study versus 1970-1990 for the
section 812 study). However, the NAPAP estimates
of annual benefits of $10 to $100 million provide a
rough benchmark for assessing the likely magnitude
of the avoided damages to an important and sensitive
recreational fishery in a four-state area most impacted
by surface water acidification from atmospheric depo-
sition.
Table E-3. Results from
Ecosystem Use Values from1 Acid Bepo,sMon;Avc-idaSSb%.
— J ' <•<><• •"••* ., .,.,.,.. ..^wfl^v^ -?<-/•
Study
NAPAP
(1991)
Use Scenario Modete4f:,,;>. > Mefltod Annual Beneftts
Vafue -""• x *V," '" " '""".'7'-'!«,- - ' '.'
Trout No changp iii alai °
Fishing deposition- Vs? ~
(NY, 50% decrease- in acid''
ME, VT, deposition
NH) _<•_;_ ^
30% increase jtf acid1'
deposition
mm-
HTCM
RUM
4lG'<3lniili
-------
Appendix E: Ecological Effects of Criteria Pollutants
composition.19 Nuisance algal blooms can have nu-
merous economic and biological costs, including wa-
ter quality deterioration affecting biological resources,
toxicity to vertebrates and higher invertebrates, and
decreased recreational and aesthetic value of waters.20
Although severe eutrophication is likely to adversely
affect organisms, especially fish, a moderate increase
in nutrient levels may also increase fish stocks, by
increasing productivity in the food chain.21
Atmospheric Deposition and Eutrophication
The deposition of NOx in aquatic systems and then-
watersheds is one source of nitrogen that may con-
tribute to eutrophication. The relative importance of
NOx deposition as a contributor to aquatic eutrophi-
cation depends on the extent to which the productiv-
ity of an aquatic ecosystem is limited by nitrogen avail-
ability and the relative importance of nitrogen depo-
sition compared to other internal and external sources
of nitrogen to the aquatic ecosystem. Furthermore,
the vulnerability of aquatic ecosystems to eutrophi-
cation is known to vary seasonally and spatially, al-
though these systems are affected by nutrient deposi-
tion throughout the year. In general, freshwater eco-
systems appear to be more often limited by phospho-
rus, rather than nitrogen, and are not as likely to be
heavily impacted by nitrogen deposition compared to
some estuarine and coastal ecosystems.22 In contrast
to acidification of streams and lakes, eutrophication
from atmospheric deposition of nitrogen is more com-
monly found hi coastal and estuarine ecosystems,
which are more frequently nitrogen-limited.23
Unfortunately, there is limited information with
regard to the relative importance of atmospheric depo-
sition as a nitrogen source hi many estuarine and ma-
rine ecosystems. Estimates of the importance of at-
mospheric nitrogen deposition are difficult to make
because of uncertainties in estimating deposition, es-
pecially dry deposition, as well as watershed nitrogen
retention.24 Paerl (1993) reviews the importance of
atmospheric nitrogen deposition as a contributor to
eutrophication of coastal ecosystems; he concludes
that 10 to 50 percent of the total nitrogen loading to
coastal waters is from direct and indirect atmospheric
deposition. Estimates for the economically impor-
tant Chesapeake Bay indicate that about 25 to 40 per-
cent of the nitrogen loadings to the bay occur via at-
mospheric deposition.25 Hinga et al. (1991) estimate
that anthropogenic deposition provides 11 percent of
total anthropogenic inputs of nitrogen in Narragansett
Bay, 33 percent for the New York Bight, and 10 per-
cent for New York Bay. Fisher and Oppenheimer
(1991) estimate that atmospheric nitrogen provides
23 percent of total nitrogen loading to Long Island
Sound and 23 percent to the lower Neuse River in
North Carolina. Information on the importance of
atmospheric nitrogen deposition for most other U.S.
coastal ecosystems is not available hi the literature.
Episodic atmospheric inputs of nitrogen may be an
important source of nitrogen to nutrient-poor marine
ecosystems, such as the North Atlantic near Bermuda
and the North Sea.26
Valuing Potential Benefits from Eutrophication
Avoidance Under the CAA
It is currently estimated that in the absence of
pollution reductions achieved under the Clean Air Act,
total nitrogen emissions to the atmosphere would have
increased by nearly 90 million tons by 1990, a two-
fold increase above 1990 levels estimated with CAA
controls remaining hi place.27 However, the ability
to determine the potential economic benefit from such
a reduction in nitrogen emissions is heavily con-
strained by gaps in our current biological and eco-
nomic knowledge base of aquatic ecosystems.
One water body that has received much study in
the area of nitrogen-induced eutrophication is Chesa-
peake Bay. As previously discussed, it is estimated
that atmospheric deposition of nitrogen contributes
approximately 25 percent to the total nitrogen load-
19 Paerl, 1993.
20 Paerl, 1988.
21 Hansson and Rudstam, 1990; Rosenberg et al., 1990; Paerl, 1993.
22 Hecky and Kilham, 1988; Vitousek and Howarth, 1991.
23 U.S. EPA, 1993; Paerl, 1993.
24 U.S. EPA, 1993.
25 U.S. EPA, 1994.
26 Owens et al., 1992.
27 U.S. EPA, 1995; Table B-3.
E-7
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
ings to the bay.28 In deposition terms, an estimated
15 to more than 25 percent increase in total nitrogen
deposition has been forecast in the Chesapeake Bay
watershed by 1990 in the absence of CAA pollution
controls.29 These results are based on an estimated
40,000 tons of atmospherically deposited nitrogen (as
nitrate and ammonia) to Chesapeake Bay in 1985,30
which means a 20 percent increase in atmospheric
deposition would amount to approximately 8,000 ad-
ditional tons.
One indirect method available to gauge the po-
tential economic relevance of avoidance of such at-
mospheric nitrogen loadings to Chesapeake Bay is
through the avoidance cost of nitrogen controls. How-
ever, such an assessment is difficult due to the site,
facility, and treatment-specific variation in treatment
costs. For example, Camacho (1993) reviewed nitro-
gen treatment costs for chemical treatment of water
from important point sources (mostly public owned
treatment works) and found that costs ranged from
$9,600 to $20,600 per ton (annual costs, 1990 dol-
lars), depending on the facility evaluated. Biological
treatment of nitrogen from point sources was far more
expensive, varying from $4,000 to $36,000 per ton.
For control of non-point source loading, values of ni-
trogen removal practices ranged from $1,000 to
$285,000 per ton.31 Taking chemical addition as one
possible example, the avoided costs of treatment of
8,000 tons of nitrogen would range from about $75
million to about $170 million annually (in 1990 dol-
lars).
Mercury
Mercury, in the form of methyl mercury, is a neu-
rotoxin of concern and can accumulate in tissue of
fish to levels that are hazardous to humans and aquatic-
feeding wildlife in the U.S. In relation to the section
812 CAA retrospective analysis, mercury is of inter-
est for two reasons. First, potential benefits to human
welfare may have occurred as a result of mercury
emission controls implemented under EPA's National
Emission Standards for Hazardous Air Pollutants
(NESHAP). Second, experimental and observational
evidence suggests that acidification of water bodies
enhances mercury accumulation in fish tissues.32
Therefore, CAA-mandated reductions in sulfur and
nitrogen oxide emissions and subsequent acid depo-
sition may have resulted in indirect benefits from a
reduction in mercury accumulation in fish and subse-
quent improvements to human health and welfare.
The accumulation of mercury to hazardous levels
in fish has become a pervasive problem in the U.S.
and Canada. A rapid increase in advisories occurred
during the 1980s, including a blanket advisory affect-
ing 11,000 lakes in Michigan.33 The Ontario Minis-
tries of Environment and Natural Resources (1990)
recommend fish consumption restrictions for 90 per-
cent of the walleye populations, 80 percent of small-
mouth bass populations, and 60 percent of lake trout
populations in 1,218 Ontario lakes because of mer-
cury accumulation. In many instances, mercury has
accumulated to hazardous levels in fish in highly re-
mote water bodies that are free from direct aqueous
discharges of mercury.34 Mass balance studies have
shown that atmospheric deposition of mercury can
account for the accumulation of mercury in fish to
high levels in lakes of these remote regions.35 The
potential impacts of mercury on the health of humans
and fish-eating (piscivorous) wildlife has lead EPA
to recently establish water quality criteria to protect
piscivorous species in the Great Lakes.36
Although mercury accumulation in fish via atmo-
spheric deposition is now widely recognized as a po-
tential hazard to human health and certain wildlife
species, studies establishing quantitative linkages be-
tween sources of mercury emissions, atmospheric
deposition of mercury, and subsequent accumulation
in fish are lacking. Thus at the present time, we are
unable to quantify potential benefits from CAA-
avoided mercury accumulation in fish of U.S. water
28 U.S. EPA, 1993.
29 U.S. EPA 1995, Figure C-6.
30 NERA, 1994.
31 Shuyler, 1992.
32 Bloom et al., 1991; Watras and Bloom, 1992; Miskimmin et al., 1992; Spry and Wiener, 1991; Wiener et al., 1990.
33 Watras et al., 1994.
34 Glass et al., 1990; Sorenson et al., 1990; Grieb et al. 1990; Schofield et al. 1994.
35 Fitzgerald et al. ,1991.
36 U.S. EPA, 1995. , ' ,
E-8
-------
Appendix E: Ecological Effects of Criteria Pollutants
bodies. Given the pervasiveness of the mercury prob-
lem with CAA-pollution controls, potential benefits
to human health and welfare from avoidance of fur-
ther mercury related damages to aquatic ecosystems
could be substantial.
It should also be noted that atmospheric deposi-
tion is a major contributor to surface water loads of
other toxic pollutants as well. For example, scientists
believe that about 35 to 50 percent of the annual load-
ings of a variety of toxic chemicals to the Great Lakes
may be from the air; for lead, atmospheric deposition
currently accounts for an estimated 95 percent of the
total load in the Great Lakes.37 CAA-related reduc-
tions in air emissions of toxic pollutants (such as lead)
undoubtedly reduced the loading of these chemicals
to the Great Lakes and other water bodies; the magni-
tude of the benefits of reducing these exposures to
humans and wildlife is not known.
Benefits from Avoided Damages
to Wetland Ecosystems
Introduction
This review addresses the effects of air pollutants
on wetland ecosystems; the focus is on acidification
and nutrient loading. Valuable service flows of wet-
land ecosystems include flood control, water quality
protection and improvement, wildlife and fish habi-
tat, and biodiversity. The limited scientific evidence
suggests that air pollutants may most affect
biodiversity, in particular because of nutrient loading
through nitrogen deposition.
Wetlands are broadly characterized as transitional
areas between terrestrial and aquatic systems in which
the water table is at or near the surface or the land is
periodically covered by shallow water.38 Types of
wetlands include swamps (forested wetlands), marshes
(herbaceous vegetation), and peatlands, which are
wetlands that accumulate partially decayed vegeta-
tive matter due to limited decomposition.39 Peatlands
"U.S. EPA, 1994.
38 Cowardin et al., 1979.
39 Mitsch and Gosselink, 1986.
40 Mitsch and'Gosselink, 1986.
41 U.S. EPA, 1993.
42 Gorham et al., 1984.
43 Turner et al., NAPAP SOS/T 10, 1990.
include bogs and fens. Bogs receive water solely from
precipitation, are generally dominated by Sphagnum
moss, and are low in nutrients. Fens receive water
from groundwater and precipitation, contain more
marsh-like vegetation, and have higher pH and nutri-
ent levels than bogs.40 Most of the limited work on
the effects of atmospheric deposition on wetlands has
been done in peatlands, specifically in Europe, where
levels of atmospheric deposition are generally much
higher than in the U.S.
The air pollutants of greatest concern with respect
to effects on wetland ecosystems are oxides of nitro-
gen (NOx) and oxides of sulfur (SOx), primarily sul-
fur dioxide (SO2). Air pollutants may affect wetland
ecosystems by acidification of vulnerable wetlands
and by increasing nutrient levels. Acidification in
vulnerable wetlands may affect vegetation adversely,
as appears to have occurred in Europe. In wetlands
where nitrogen levels are low, increased nitrogen
deposition may alter the dynamics of competition
between plant species. Species adapted to low-nitro-
gen levels, including many endangered species, may
decrease in abundance.41
Effects of Acidification
Limited evidence suggests that acidic deposition
and decreased pH may harm certain wetland plants,
alter competitive relations between wetland plants and
cause changes in wetland drainage and water reten-
tion.
Work concerning the possible acidification of
peatlands is inconclusive. Acidic deposition is un-
likely to result in displacement of base cations from
cation exchange sites in bogs, and therefore it will not
cause a drop in pH.42 Peatland sediments are low in
A13+, so mobilization of toxic aluminum is not a con-
cern as it is in forest soils and aquatic ecosystems.43
Acidification might affect certain fen ecosystems.
Gorham et al. (1984) have hypothesized that acidic
deposition could leach base cations from mineral-poor
fens and decrease pH levels. This could result in a
E-9
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
transition to bog vegetation such as Sphagnum and
away from sedge meadow vegetation. At this time,
this remains a hypothesis; however, pH did not de-
crease in a mineral-poor Ontario fen during a four-
year period in which researchers experimentally in-
creased acidic deposition.44
In European wetlands affected by high levels of
deposition for many years, acidic deposition has seri-
ously affected wetland vegetation. Roelofs (1986)
reports that acidification of heath pools in the Nether-
lands has caused a change in species composition with
Sphagnum and rushes replacing the original vegeta-
tion. Likewise, significant declines in Sphagnum in
British bogs have occurred in areas affected by 200
years of atmospheric pollution, including nitrogen
deposition.45 It is unclear how such changes have af-
fected wetland service flows apart from the effects on
biodiversity; however, water retention has decreased
and significant erosion has occurred hi seriously per-
turbed British bogs near Manchester and Liverpool.46
Effects of Nutrient Loading
Atmospheric deposition may affect wetlands by
increasing the level of nutrients, particularly nitrogen,
in wetlands. Sulfur is not a limiting nutrient in
peatlands,47 but nitrogen commonly limits plant
growth.48 The effects of increased nitrogen levels in
wetlands include an increased threat to endangered
plant species and possible large-scale changes in plant
populations and community structure. Endangered
and threatened plant species are common in wetlands,
with wetland species representing 17 percent of the
endangered plant species in the U.S. (U.S. EPA, 1993).
These plants are often specifically adapted to low ni-
trogen levels; examples include isoetids49 and insec-
tivorous plants.50 In eastern Canadian wetlands, na-
tionally rare species are most common in infertile
sites.51 When nitrogen levels increase, other species
adapted to higher levels of nitrogen may competitively
displace these species. Thus, NOx emissions that in-
crease nitrogen levels in nitrogen-poor wetlands may
increase the danger of extinction for threatened and
endangered species.
By changing competitive relations between plant
species, increased nitrogen deposition may broadly
affect community structure in certain wetlands. Com-
mon species that thrive in nitrogen-poor wetlands may
become less abundant. Many nitrogen-poor bogs in
the northern U.S. are dominated by Sphagnum spe-
cies. These species capture low levels of nitrogen from
precipitation. Increased nitrogen levels may directly
harm Sphagnum and cause increased nitrogen to be
available to vascular plants that may out compete Sph-
agnum.52 Studies in Great Britain have documented
large declines in Sphagnum moss because of atmo-
spheric pollution;53 nitrogen loading may play an im-
portant role in these declines. However, Rochefort et
al. (1990) document limited effects of fertilization
from experimentally-increased NO3~ and SO42' depo-
sition on an Ontario mineral-poor fen over a four-year
period, apart from initially increased Sphagnum
growth. Thus, increased nitrogen loading might ad-
versely or beneficially affect wetland plants depend-
ing on baseline nitrogen concentrations in the wet-
land, atmospheric nitrogen loading, and species re-
quirements for and sensitivity to nitrogen.
Increases in nitrogen levels due to NOx emissions
will have the greatest effect on wetlands that are ex-
tremely nitrogen-limited and that receive small
amounts of nitrogen naturally. Since bogs, including
Sphagnum bogs, receive little surface water runoff,
they get most of their nutrient and water loadings
through precipitation. These bogs may receive a total
of approximately 10 kg nitrogen per hectare per year
(kg N/ha/yr), which is one to two orders of magnitude
less nitrogen than other freshwater wetlands and
44 Rochefort et al., 1990.
45 Lee et al., 1986.
44 Lee et al., 1986.
47 Turner et al., NAPAP SOS/T 10,1990.
48 U.S. EPA, 1993.
49 Boston, 1986.
30 Moore et al., 1989.
51 Moore et al., 1989; Wisheu and Keddy, 1989.
52 Lee & Woodin 1988, Aerts et al., 1992.
53 Ferguson et al., 1984; Lee et al., 1986.
E-10
-------
Appendix E: Ecological Effects of Criteria Pollutants
saltmarshes receive.54 As atmospheric deposition of
nitrogen has been estimated to be at least 5.5 to 11.7
kg N/ha/yr,55 changes in NOx emissions would most
likely affect these bogs. The results of a model by
Logofet and Alexandrov (1984) suggest that a tree-
less, nutrient-poor bog may undergo succession to a
forested bog because of the input of greater than 7 kg
N/ha/yr.
As in freshwater wetlands, significantly increased
nitrogen deposition to coastal wetlands will increase
productivity and alter the competitive relationships
between species.56 However, studies showing this
increased productivity have used 100 to 3000 kg N/
ha/yr.57 Therefore, limited changes in NO emissions
may not affect coastal wetland productivity.
Summary of Wetland Ecosystem Effects
The effects of air pollutants on wetlands have re-
ceived little attention, in contrast to the large body of
work on the effects of acid rain on aquatic and forest
ecosystems. Little evidence exists suggesting that
acidification due to atmospheric deposition is a ma-
jor threat to wetlands. In particular, peatlands are
naturally acidic, although mineral-poor fens may be
at risk from acidification. Nitrogen loading may alter
community composition in wetlands naturally low in
nutrients, such as bogs. Nitrogen loading may threaten
rare species adapted to low nitrogen levels. In Britain
and The Netherlands, heavy atmospheric deposition
over a long period appears to have caused serious de-
clines in Sphagnum in peatlands.
Air pollutants appear to most seriously threaten
rare and endangered species, biodiversity, and com-
munity composition in wetlands, particularly bogs.
These changes are difficult to associate with changes
in economic value; even the qualitative nature of the
effects is uncertain. Air pollutants may not signifi-
cantly affect such important wetland service flows as
flood control, water quality protection, and wildlife
M U.S. EPA, 1993.
55 U.S. EPA, 1993.
56 U.S. EPA, 1993.
57 U.S. EPA, 1993.
58 Powell etal. 1993.
59 U.S. DOI, 1993.
60 Shriner et al., NAPAP SOS/T 18,1990.
61 NAPAP, 1991.
62 Shriner et al., NAPAP SOS/T 18,1990.
habitat in most wetlands, so the impacts on the more
readily monetized aspects of the economic value of
wetlands may be limited.
Benefits from Avoided Damages
to Forests
Introduction
Forests occupy 33 percent of the land mass in the
U.S. (some 738 million acres) and provide a wealth
of services to the U.S. population.58 Notable services
provided by forests include timber production, recre-
ational opportunities such as hunting, camping, hik-
ing, and wildlife observation, water quality protec-
tion, nutrient removal and cycling, flood control, ero-
sion control, temporary carbon sequestration, preser-
vation of diversity, and existence values. In 1991,
hunting participation alone accounted for 236 million
recreation days that included 214 million person trips
with estimated expenditures valued at $12.3 billion.59
The Clean Air Act-regulated pollutants of great-
est concern with respect to effects on forest ecosys-
tems are oxides of sulfur (SOx), primarily sulfur di-
oxide (SO2), oxides of nitrogen (NOx), and volatile
organic compounds (VOCs). While extremely high
ambient concentrations of SO2 and NOx may directly
affect vegetation, such effects are uncommon in the
U.S.;60 the indirect effects of these pollutants are of
greater concern. Specifically, emissions of SO2 and
NOx are known to contribute to acid deposition in
portions of the United States, with SO2 contributing
75 to 95 percent of the acidity in rainfall in the east-
ern U.S.61 Acid deposition is of concern to forests
primarily from the acidification of soils (i.e., by re-
ducing seed germination, altering nutrient and heavy
metal availability). Direct foliar damage can occur
from precipitation with extremely low pH levels (i.e.,
3.0-3.6 and below), although these levels are lower
than ambient levels in the U.S.62 VOCs and NO are
E-ll
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
important precursors to ozone formation, which can
affect leaf photosynthesis and senescence and decrease
cold hardiness, thereby causing deleterious impacts
on tree growth, survival and reproduction. Deposi-
tion of NO may also alter the nutrient balance of for-
est soils, which in turn might alter the competitive
relationships between tree species and affect species
composition and diversity.63
Current Air Pollutant Effects on Forests
Acid Deposition Impacts
In 1985, NAPAP organized the Forest Response
Program (FRP) to evaluate the significance of forest
damage caused by acidic deposition, the causal rela-
tionships between air pollutants and forest damage,
and the dynamics of these relationships regionally.
Research was focussed on four forest regions: East-
ern Spruce-Fir, Southern Commercial Forests, East-
ern Hardwoods, and Western Conifers. With the ex-
ception of high-elevation spruce-fir forests, the avail-
able evidence suggests that acidic deposition does not
currently affect these forests and that observed de-
clines in sugar maple and southern pines are not due
to acidic deposition.64
Circumstantial evidence suggests that acidic depo-
sition may affect high-elevation spruce-fir forests in
the northeastern U.S. These forests have extensive
contact with acidic cloud water.65 Experimental evi-
dence suggests that acidic deposition may affect cold
hardiness in red spruce, an important component of
the spruce-fir forest. Significant declines in red spruce
growth and in its importance hi the forest have oc-
curred in New York and northern New England. The
proximate cause of death of red spruce in the region
is pathogens and insects; acidic deposition may inter-
act with these biological stresses and with weather-
induced stress to produce adverse effects hi red spruce.
Ozone may also play a role hi red spruce decline in
this region.66 Available evidence suggests that soil
aluminum and soil pH levels have not affected red
spruce adversely.67
Ozone Impacts
Experimental Evidence *
For practical reasons, the majority of experimen-
tal evidence linking ozone exposure to damage to tree
species has been derived from studies of individual
plants, especially seedling and branch studies.68 Re-
sults from these studies suggest that ozone exposure
can reduce photosynthesis and increase senescence in
leaves. Subsequently, such effects from ozone may
alter the carbohydrate allocation to plant tissues such
as roots, which may affect plant growth and cold har-
diness. Decreases in cold tolerance may be particu-
larly important for trees in northern latitudes and high
elevations. Recent work on quantifying the relation-
ship between ozone exposure and plant responses sug-
gest that seedlings of aspen, ponderosa pine, black
cherry, tulip poplar, sugar maple, and eastern white
pine seedlings may experience biomass reductions of
approximately 10 percent at or near ambient ozone
exposures.69 Because trees are perennials, the effect
of even a 1-2 percent per year loss in seedling biom-
ass (versus 10 to 20 percent yield loss in crops), if
compounded over multiple years under natural field
conditions of competition for resources, could be se-
vere.
Although indicative of short-term relative re-
sponse to ozone exposure, results from these experi-
ments are unable to provide reliable information on
the long-term effects of ozone on forests. This limi-
tation arises because the effects of ozone on forests
will depend on both the response of individual plants
to ozone exposure and the response of populations of
plants, which interact with their environment. Popu-
lation response will be altered by the varying intraspe-
cific genetic susceptibility to ozone. Individual plant
response will also be affected by many environmen-
tal factors, including insect pests, pathogens, plant
symbionts, competing plants, moisture, temperature,
light, and other pollutants. Consistent evidence on
the interaction of ozone with other environmental fac-
tors is lacking. Furthermore, most experimental stud-
<° U.S. EPA, 1993.
M Barnard et al., NAPAP SOS/T 16,1990; NAPAP, 1991.
65 Barnard et al., NAPAP SOS/T 16,1990.
M Shriner et al., NAPAP SOS/T 18,1990.
OT Barnard et al., NAPAP SOS/T 16,1990.
« U.S. EPA, 1996a.
69 Hogsettetal. ,1995.
E-12
-------
Appendix E: Ecological Effects of Criteria Pollutants
ies have only studied exposure for one growing sea-
son; effects on forest species may occur over de-
cades.70 Therefore, considerable uncertainties occur
in scaling across individuals of different ages, from
individuals to populations and communities, and
across time.
Observational Evidence
Studies of the forests of the San Bernardino Moun-
tains provide the strongest case for linking ozone ex-
posure to damages to an entire forest ecosystem. These
forests have been exposed to extremely high ambient
ozone levels over the past 50 years due to then: prox-
imity to the Los Angeles area. The area has been ex-
tensively studied regarding the effects of ozone, as
described in U.S. EPA (1996a). The ecosystem has
been seriously affected by ozone pollution, with the
climax-dominant, but ozone-sensitive ponderosa pine
and Jeffrey pine declining in abundance, replaced by
more ozone-tolerant species. These sensitive species
have experienced decreased growth, survival, and re-
production, and susceptibility to insects. The effects
of ozone on these species have resulted in other eco-
system effects, including the buildup of a large litter
layer, due to increased needle senescence. The de-
cline of the fire-tolerant ponderosa and Jeffrey pines
may seriously affect the fire ecology of the ecosys-
tem, with fire-sensitive species becoming more com-
mon. Ozone concentrations have been declining in
recent decades, and crown injury of ponderosa and
Jeffrey pine has decreased. However, the two species
have continued to decline in abundance, as measured
by total basal area, compared with other species over
the period 1974 to 1988.71 The nature of community
dynamics, particularly in mixed species, uneven aged
stands, indicates that subtle long-term forest responses
(e.g., shifts in species composition) to elevated levels
of a chronic stress like exposure to ozone are more
likely than wide-spread community degradation.72
Limited field studies have been completed in other
forest ecosystems. Foliar injury has been observed in
the Jefferson and George Washington National For-
ests and throughout the Blue Ridge Mountains, in-
cluding areas of the Shenandoah National Park.73 In
the Great Smoky Mountains National Park, surveys
made in the summers from 1987 through 1990 found
95 plant species exhibited foliar injury symptoms con-
sistent with those thought to be caused by ozone.74
Foliar ozone injury has also been documented in Na-
tional Parks and Forests in the Sierra Nevada moun-
tains.75
Growth and productivity of seedlings have been
reported to be affected by ozone for numerous spe-
cies in the Blue Ridge Mountains of Virginia. In the
Shenandoah National Park, Duchelle et al. (1982,
1983) found that tulip poplar, green ash, sweet gum,
black locust, as well as several evergreen species (e.g.,
Eastern hemlock, Table Mountain pine, pitch pine,
and Virginia pine), common milkweed, and common
blackberry all demonstrated growth suppression of
seedlings. Except for the last two species mentioned,
almost no visible injury symptoms accompanied the
growth reductions. Studies of mature trees in the
Appalachian Mountains also indicate that injury as-
sociated with exposure to ozone and other oxidants
has been occurring for many years.76 Researchers have
also found that major decreases in growth occurred
for both symptomatic and asymptomatic trees during
the 1950s and 1960s in the Western U.S.77 The ad-
verse response of a number of fruit and nut trees to
ozone exposure has been reported.78
Monitoring by the USDA Forest Service shows
that growth rates of yellow pine in the Southeast have
been decreasing over the past two decades in natural
stands but not in pine plantations.79 Solid evidence
linking this growth reduction to air pollutants is lack-
70 U.S. EPA, 1996a. •
71 Miller et al., 1989 and Miller et al., 1991.
72 Shaver et al., 1994
73 Hayes and Skelly, 1977; Skelly et al., 1984
74 Neufeld, et al., 1992
75 Peterson and Arbaugh, 1992
76 Benoit et al., 1982
77 Peterson et al., 1987; Peterson and Arbaugh, 1988, 1992; Peterson et al., 1991
78 McCool and Musselman, 1990; Retzlaff et al., 1991, 1992a, b
79 NAPAP, 1991.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
ing, although ozone, in particular, may be a factor.80
Ambient ozone levels in the region are high enough
to damage sensitive tree species, including pine seed-
lings during experimental exposure.81 Due to the com-
mercial importance of yellow pine, the economic im-
pacts of ozone on forest ecosystems in this area could
be significant if ozone is affecting growth.
Although the ecosystem effects occurring in the
San Bernardino forest ecosystem have occurred at very
high ozone exposures, lower ozone exposure else-
where in the U.S. may still affect forests. The EPA
Ozone Staff Paper82 assessed the risk to vegetation,
including forests, under current ambient air quality.
Using a GIS approach, it was found that under the
base year (1990) air quality, a large portion of Cali-
fornia and a few localized areas in North Carolina and
Georgia have seasonal ozone levels above those which
have been reported to produce greater than 17 percent
biomass loss in 50 percent of studied tree seedling
species. A broader multistate region in the east is
estimated to have air quality sufficient to cause 17
percent biomass loss in seedlings, while at least a third
of the country, again mostly in the eastern U.S., most
likely has seasonal exposure levels which could al-
low up to 10 percent yield loss in 50 percent of stud-
ied seedlings. The Staff Paper did not present mon-
etized benefits because of lack of exposure-response
functions.83
Even small changes in the health of ozone-sensi-
tive species may affect competition between sensi-
tive and tolerant species, changing forest stand dy-
namics.84 Depending on the sensitivities of individual
competing species, this could affect timber produc-
tion either positively or negatively, and affect com-
munity composition and, possibly, ecosystem pro-
cesses.
Endangered species
Ozone effects may also reduce the ability of af-
fected areas to provide habitats to endangered spe-
cies. For example, two listed endangered plant spe-
cies, the spreading aven and Roan Mountain bluet,
80 NAPAP, 1991.
81 NAPAP, 1991.
M U.S. EPA, 1996b
83 U.S. EPA, 1996b.
84 U.S. EPA, 1996a.
85 U.S. EPA, 1996b
86 NAPAP. 1991.
are currently found at a small number of sites in east-
em Tennessee and western North Carolina—forested
areas where ozone-related injury is of concern.85 In
addition, ozone-related effects on individual ozone-
sensitive species that provide unique support to other
species can have broader impacts. For example, one
such species is the common milkweed, long known
for its sensitivity to ozone and usefulness as an indi-
cator species of elevated ozone levels, as well as be-
ing the sole food of the monarch butterfly larvae.
Thus, a major risk associated with of the loss of milk-
weed foliage for a season is that it might have signifi-
cant indirect effects on the monarch butterfly popula-
tion. A large number of studies have shown that
ozone-sensitive vegetation exists over much of the
U.S., with many native species located in forests and
Class I areas, which are federally mandated to pre-
serve certain air quality related values.
Valuation of Benefits From CAA-
Avoided Damages to Forests
Background
To quantitatively assess the economic benefits of
avoided damages of relevant CAA pollutants to for-
ests, it is necessary to link estimated changes in air
pollution to measures of forest health and conditions
that can be readily quantified in economic terms. For
commercial timber production, this would require
quantifying the relationship between atmospheric
deposition and measures of forest productivity such
as timber yield. For assessing recreational benefits,
linkages would have to be drawn between air pollu-
tion and vulnerable factors that influence forest-based
recreation (e.g., site-characteristics such as canopy
density, type of tree species, degree of visible tree
damage, etc.). While important strides have been
made in establishing these linkages (e.g., NAPAP
modeling of air pollution effects on forest soil chem-
istry and tree branch physiology), critical gaps in our
ability to predict whole tree and forest responses to
air pollution changes have precluded the establish-
ment of such quantitative linkages.86 Critical knowl-
E-14
-------
Appendix E: Ecological Effects of Criteria Pollutants
edge gaps exist in our ability to extrapolate experi-
mental results from seedling and branch studies to
whole tree and forest responses, to account for key
growth processes of mature trees, to integrate various
mechanisms by which air pollution can affect trees
(e.g., soil acidification, nitrification, and direct foliar
damage, winter stress, etc.), and to account for the
interaction of other stressors on forest health and dy-
namics (susceptibility to insect damage, drought, dis-
ease, fire, nutrient and light competition, etc.).
Despite these constraints to quantifying economic
benefits from air pollution reductions on forest eco-
systems, relevant studies that have attempted to value
air pollution damages on forests are reviewed and
summarized below. In some cases, the relationship
between air pollution and forest response is estimated
using expert judgement (e.g., for NAPAP assessment
from various growth scenarios). In other cases, dam-
age estimates reflect current impacts of air pollution
on forests, and the dose-response relationship is ab-
sent. In the aggregate, this summary provides some
insight into possible CAA-related benefits from
avoided damages to a select and narrowly focussed
group of forest services, but, because of severe data
constraints, does not provide an estimate of the over-
all range of forest-based benefits possible under the
CAA.
Commercial Timber Harvesting
The economic impact of hypothetical growth re-
ductions in northeastern and southeastern trees (both
hardwood and softwood species) was intensively stud-
ied under NAPAP.87 Growth reductions ranging from
5 to 10 percent over a 5 to 10 year period, depending
on the species and location, were assumed to occur as
a result of all forms of air pollution based on expert
opinion derived from a survey by deSteigner and Pye
(1988). Timber market responses to these hypoth-
esized growth declines were modeled until the year
2040 using a revised version of the Timber Assess-
ment Market Model (TAMM90) and the Aggregate
Timberland Assessment System (ATLAS), which was
used to simulate timber inventories on private tim-
berland in the United States. Economic welfare out-
puts included changes in consumer and producer sur-
plus and changes in revenue to southeast stumpage
owners. Results indicate that annualized reductions
in consumer and producer surplus would total $0.5
billion by the year 2000 and $3 billion by the year
2040 (in 1967 dollars). Simulated effects on stump-
age owners' revenues were minimal ($10 to $20 mil-
lion).
In an attempt to estimate the net economic dam-
ages from ozone effects on selected U.S. forests,
NAPAP studied the effect of various assumed reduc-
tions in growth rates of commercial southeastern pine
forests (both natural and planted).88 For both planted
and natural plus planted pines, the following changes
in growth rates were assumed to occur: a two percent
increase, no change, a two percent decrease, a five
percent decrease, and a ten percent decrease. The two
to five percent growth reductions were considered as
possible outcomes from current ozone induced dam-
age to southeastern forests, although no quantitative
linkage between ozone exposure and damages was
established. The ten percent growth reduction sce-
nario was primarily included for evaluating model
sensitivity to severe changes in growth and was con-
sidered out of the range of likely ozone damage esti-
mates. The TAMM and ATLAS models were again
used to simulate timber market responses under
baseline and hypothesized growth change scenarios
from 1985 to 2040. Results indicate that annual
changes in total economic surplus (i.e., the sum of
consumer and producer surplus and timber owner rev-
enues in 1989 dollars) would range from an increase
of $40 million (for the two percent increase in growth
scenario) to a decrease of $110 million (for the ten
percent decrease in growth scenario) for planted and
natural pine model simulations.
In the context of estimated benefits from avoid-
ance of other damages in the absence of the Clean Air
Act from 1970 to 1990,89 the magnitude of economic
damages estimated to the commercial timber indus-
try are comparatively small. For example, economic
damage estimates range up to $3 billion annually for
five to ten percent growth rate reductions in northeast
and southeast forests, and just $110 million for south-
eastern pines. However, in the context of damages to
forest-based services as a whole, the NAPAP-derived
commercial timber damage estimates should be
viewed as representing a lower bound estimate for a
variety of reasons. First, these damage estimates
exclude other categories of possible forest-based ben-
87 Haynes and Kaiser, NAPAP SOS/T 27 Section B, 1990.
88 NAPAP, 1991.
89 Most notably avoided human health effects, which are estimated on the order of $300 to $800 billion annually.
E-15
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
efits, including recreational and non-use values. Sec-
ond, even within the context of timber-related dam-
ages, the NAPAP forest-damage studies focused on a
portion of U.S. forests (northeastern and southeastern
U.S.); a much greater geographic range of forests could
become susceptible to timber-related damages in the
absence of CAA controls. Finally, the NAPAP dam-
age estimates consider only two types of tree species:
planted and naturally grown pines, although these spe-
cies are economically important. Damages to other
commercially harvested tree species, such as mixed
pine and hardwood forests, are therefore excluded.
Non-marketed Forest Services
In an effort to address the potential benefits re-
sulting from avoidance of acid deposition-induced
damages to non-marketed forest-based services (e.g.,
recreation use, existence value), an extensive review
of the economic literature was conducted under the
auspices of NAPAP.90 From their review, NAPAP
could not identify any single study or model that could
be reliably used to quantify economic benefits from
avoided acid deposition-caused damages to non-mar-
keted forest services (such as recreational use) on a
regional or national basis. The primary limitation in
many of the studies reviewed was the absence of a
quantitative linkage between the value of a recreational
user day and important site characteristics which could
be tied to air pollution effects. In addition, most stud-
ies were narrowly focused geographically to specific
sites and did not attempt to value system-wide (larger
scale) damages that could result from acid deposition
over an entire region. Since the availability of nearby
substitution sites will affect the recreational value for
a given site, the benefits from such site-specific stud-
ies may not reflect actual economic damages incurred
from wide-scale air pollution impacts on forests. The
inability of studies to consider additional crowding at
unaffected sites in addition to changes in recreational
participation rates as a function of air pollution dam-
ages was also recognized as an important limitation.
Despite not being able to quantitatively assess the
benefits from avoided acid deposition-induced dam-
ages to nonmarket forest services, several important
concepts emerge from NAPAP's review of recre-
ational benefits, that bear relevance to the section 812
retrospective analysis. First, several studies were iden-
tified that established a relationship between key for-
est site characteristics and the value of recreational
participation. For example, Brown et al. (1989) used
90 Rosenthal, NAPAP SOS/T 27 Section B, 1990.
contingent valuation to evaluate the relationship be-
tween scenic beauty ratings and willingness of
recreationalists to pay at pictured sites. Based on their
interviews with over 1400 recreationalists at ten dif-
ferent sites in Arizona, positive correlations were es-
tablished between scenic beauty rankings determined
from one group of recreationalists and willingness to
pay to recreate determined by a separate group of
recreationalists (r2 ranged from 0.27 to 0.98 depend-
ing on ranking). In another study, Walsh et al. (1989)
developed a functional relationship between reduc-
tion of recreational benefits and tree density changes
that reflected varying levels of bisect damage at six
campgrounds in the Front Range of the Colorado
Rockies. By using both contingent valuation and travel
cost models, Walsh et al. (1989) were able to show
that 10 percent, 20 percent, and 30 percent decreases
in tree densities reduces the total recreational benefits
at their sites by 7 percent, 15 percent and 24 percent,
respectively. Although results from these studies are
limited to the sites from which they were derived, they
do support the intuition that the degree of visible dam-
age to forests is to some extent correlated with the
magnitude of damages to forest-based recreation ex-
pected. This finding supports the notion that the avoid-
ance of damages to forest ecosystems from CAA-in-
duced pollution controls (albeit currently unqualified)
have likely benefited forest-based recreation in the
U.S.
In addition to establishing relationships between
recreational value and visible damage to forest sites,
there is evidence linking air pollution (ozone) effects
on forests to economic damages to non-use values of
forests. For example, D.C. Peterson et al. (1987) val-
ued ozone-induced damages to forests surrounding the
Los Angeles area. Using contingent valuation meth-
ods, D.C. Peterson et al. (1987) surveyed
recreationalists (a random survey of households in the
San Bernardino, Los Angles and Orange counties) and
residents (a sample of property owners within the San
Bernardino and Angeles national forests) for their
willingness to pay to prevent forest scenes from de-
grading one step on a "forest quality ladder" depict-
ing various levels of ozone-induced damages. The
mean willingness to pay to protect further degrada-
tion was $37.61 and $119.48 per household for
recreationalists and residents, respectively. Annual
damages to Los Angeles area residences from.a one-
step drop on the forest quality ladder were estimated
between $27 million and $147 million.
E-16
-------
Appendix E: Ecological Effects of Criteria Pollutants
These estimates cannot be directly translated into
a rough estimate of the potential non-use values of
avoided forest damages. Considering the limited size
of the population generating the estimated benefits of
forest degradation, however, they do provide evidence
that the recreational and non-use benefits may sub-
stantially exceed the commercial timber values.
E-17
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Ecosystem Effects References
Aerts, R., B. Wallen, and N. Malmer. 1992. Growth-
limiting nutrients in Sphagnum-dominated
bogs subject to low and high atmospheric ni-
trogen supply. Journal of Ecology 80: 131-
140.
Baker, L.A., P.R. Kaufmann, A.T. Herlihy, J.M.
Eilers, D.F. Brakke, M.E. Mitch, RJ. Olson,
R.B. Cook, B.M. Ross-Todd, JJ. Beauchamp,
C.B. Johnson, D.D. Brown, and D.J. Blick.
1990. Current Status of Surface Water Acid-
Base Chemistry. NAPAP SOS/T Report 9,
In: Acidic Deposition: State of Science and
Technology, Volume n, National Acid Pre-
cipitation Assessment Program, 722 Jackson
Place NW, Washington, D.C. 20503.
Baker, J.P., D.P. Bernard, S.W. Christensen, M.J. Sale,
J. Freda, K. Heltcher, D. Marmorek, L. Rowe,
P. Scanlon, G. Suter, W. Warren-Hicks, and
P. Welboum. 1990. Biological effects of
changes in surface water acid-base chemis-
try. NAPAP SOS/T Report 13, In: Acidic
Deposition: State of Science and Technology,
Volume n, National Acid Precipitation As-
sessment Program, 722 Jackson Place NW,
Washington, D.C. 20503.
Barnard, J.E., A.A. Lucier, R.T. Brooks, A.H. Johnson,
P.H. Dunn, and D.F. Karnosky. 1990.
Changes in forest health and productivity in
the United States and Canada. NAPAP SOS/
T Report 16, M: Acidic Deposition: State of
Science and Technology, Volume III, Na-
tional Acid Precipitation Assessment Pro-
gram, 722 Jackson Place NW, Washington,
D.C. 20503.
Benoit, L. F., J. M. Skelly, L. D. Moore, and L. S.
Dochinger. 1982. Radial growth reductions
in Pinus strobus L. correlated with foliar
ozone sensitivity as an indicator of ozone-in-
duced losses in eastern forests. Can. J. For.
Res. 12:673-678.
Bloom N. S., C. J. Watras, and J. P. Hurley. 1991.
Impact of acidification on the methylmercury
cycle of remote seepage lakes. Water Air Soil
Pollution 56:477-491.
Boston, H.L. 1986. A discussion of the adaptations
for carbon acquisition in relation to the growth
strategy of aquatic isoetids. Aquatic Botany
26:259-270.
Brown, T.C., M.T. Richards, and T.C. Daniel. 1989.
Scenic Beauty and Recreation Value: Assess-
ing the Relationship, JJT: J- Vining, ed., So-
cial Science and Natural Resources Recre-
ation Management, Westview Press, Boulder,
Colorado.
Camacho, R. 1993. Financial Cost Effectiveness of
Point and Nonpoint Source Nutrient Reduc-
tion Technologies in the Chesapeake Bay
Basin. Report No. 8 of the Chesapeake Bay
Program Nutrient Reductio Strategy Reevalu-
ation. Washington D.C.: U.S. Environmental
Protection Agency, February.
Cowardin, L.M., V. Carter, F.C. Golet, and E.T.
LaRoe. 1979. Classification of wetlands and
deepwater habitats of the United States. U.S.
Fish & Wildlife Service Pub. FWS/OBS-79/
31, Washington, D.C., 103 pp.
deSteigner, J. E. and J. M. Pye. 1988. Using scientific
opinion to conduct forestry air pollution eco-
nomic analysis. In: A. Jobstl, ed., Proceed-
ings of the Symposium on the Economic As-
sessment of Damage Caused to Forests by Air
Pollutants. IUFRO Working Party S4.04-02,
September 13-17, 1988. Gmunden, Austria.
Duchelle, S. F., J. M. Skelly, T. L. Sharick, B. I.
Chevone, Y. Yang, and J. E. Nellessen. 1983.
Effects of ozone on the productivity of natu-
ral vegetation in a high meadow of the
Shenandoah National Park of Virginia. J of
Env. Manage. 17:299-308.
Duchelle, S. F., J. M. Skelly, and B. I. Chevone. 1982.
Oxidant effects on forest tree seedling growth
in the Appalachian Mountains. J Water, Air,
SoilPollut. 18:363-373.
Ferguson, P., R.N. Robinson, M.C. Press, and J.A.
Lee. 1984. Element concentrations in five
Sphagnum species in relation to atmospheric
pollution. Journal of Bryology 13: 107-114.
E-18
-------
Appendix E: Ecological Effects of Criteria PoJtoartts
Fisher, D.C. and M. Oppenheimer. 1991. Atmo-
spheric nitrogen deposition and the Chesa-
peake Bay estuary. Ambio 20:102-108.
Fitzgerald, W.F., R.P. Mason, and G.M. Vandal. 1991.
Atmospheric Cycling and Air-Water Ex-
change of Mercury Over Mid-Continental
Lacustrine Regions. Water, Soil, Air & Soil
Poll. 56:745-767.
Glass, G. E., J.A. Sorenson, K.W. Schmidt, and G.R.
Rapp, Jr. 1990. New Source Identification of
Mercury Contamination in the Great Lakes.
Environ. Sci. Technol. 24: 1059-1069.
Gorham, E., S.E. Bayley, and D.W. Schindler. 1984.
Ecological effects of acid deposition upon
peatlands: A neglected field in "acid-rain"
research. Can. J. Fish. Aquat. Sci. 41: 1256-
1268.
Grieb, T.M., C.T. DriscoU, S.T. Gloss, C.L. Schofield,
G.L. Bowie, and D.B. Porcella. 1990. Fac-
tors affecting mercury accumulation in fish
in the upper Michigan peninsula. Environ.
Toxicol. Chem. 9:9191-930.
Hansson, S. and L.G. Rudstam. 1990. Eutrophica-
tion and Baltic fish communities. Ambio
19:123-125.
Hayes, E. M., and J. M. Skelly. 1977. Transport of
ozone from the northeast U.S. into Virginia
and its effect on eastern white pines. Plant
Dis. Rep. 61:778-782.
Haynes, R.W. andH.F. Kaiser. 1990. Forests: Meth-
ods for Valuing Acidic Deposition/Air Pol-
lution Effects. NAPAP SOS/T Report 27,
Section B2, In: Acidic Deposition: State of
Science and Technology, Volume IV, Na-
tional Acid Precipitation Assessment Pro-
gram, 722 Jackson Place NW, Washington,
D.C.20503.
Hecky, R.E. and P. Kilham. 1988. Nutrient limita-
tion of phytoplankton in freshwater and ma-
rine environments: a review of recent evi-
dence on the effects of enrichment. Limnol-
ogy and Oceanography 33:796-822.
Hinga, K.R., A.A. Keller, and C.A. Oviatt. 1991.
Atmospheric deposition and nitrogen inputs
to coastal waters. Ambio 20:256-260.
Hogsett, W. E., A. A. Herstom, J. A. Laurence, E. H.
Lee, J. E. Weber, and D. T. Tingey. 1995.
Risk characterization of tropospheric ozone
to forests. In: Comparative Risk Analysis and
Priority Setting for Air Pollution Issues. Pro-
ceedings of the 4th U.S.-Dutch International
Symposium. Pittsburgh, PA. Air and Waste
Management Association. 119-145.
Lee, J.A. and S.J. Woodin. 1988. Vegetation struc-
ture and the interception of acidic deposition
by ombrotrophic mires. In: Vegetation Struc-
ture in Relation to Carbon and Nutrient
Economy, J.T.A. Verhoeven, G.W. Heil, and
MJ.A. Werger, Eds. SPB Academic Publish-
ing bv, The Hague, pp. 137-148.
Lee, J.A., M.C. Press, and S.J. Woodin. 1986. Ef-
fects of NO on aquatic ecosystems. In: En-
vironment £&id Quality of Life: Study on the
Need for an NO Long-term Limit Value for
the Protection ofTerrestrial and Aquatic Eco-
systems. Commission of the European Com-
munities, Luxembourg, pp. 99-116.
Locke, A. 1993. Factors influencing community
structure along stress gradients: zooplankton
responses to acidification. Ecology 73:903-
909.
Logofet, D.O. and G.A. Alexandrov. 1984. Model-
ling of matter cycle in a mesotropic bog eco-
system. II. Dynamic model and ecological
succession. Ecol. Modell. 21:259-276
McCool, P. M., and R. C. Musselman. 1990. Impact
of ozone on growth of peach, apricot, and al-
mond. Hortscience 25: 1384-1385.
Miller, P.R., J.R. McBride, S.L. Schilling, and A.P.
Gomez. 1989. Trend of ozone damage to
conifer forests between 1974 and 1988 in the
San Bernardino mountains of southern Cali-
fornia. In: Effects of Air Pollution on West-
ern Forests. R.K. Olson and A.S. Lefohn, Eds.
Pittsburgh, PA: Air and Waste Management
Association, Pittsburgh, PA, pp. 309-324
(Transaction series no. 16).
E-19
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Miller, P.R., J.R. McBride, and S.L. Schilling. 1991.
Chronic ozone injury and associated stresses
affect relative competitive capacity of species
comprising the California mixed conifer for-
est type. In: Memorias del Primer Simposial
Nacional; Agricultura Sostenible: Una Opcion
para Desarollo sin Deterioro Ambiental.
Comision de Estudios Ambientales, Colegio
de Postgraduados, Montecillo, Edo. Mexico,
Mexico, pp. 161-172.
Mills, K.H., S.M. Chalanchuk, L.C. Mohr, and IJ.
Davies. 1987. Responses of fish populations
in Lake 223 to 8 years of experimental acidi-
fication. Can. J. Fish. Aquat. Sci. 44(Suppl.
1): 114-125.
Miskimmin, B.M., J.W.M. Rudd, and C.A. Kelly.
1992. Influence of dissolved organic carbon,
pH, and microbial respiration rates on mer-
cury methylation and demethylation in lake
water. Can. J. Fish. Aquat. Sci. 49:17-22
Mitsch, W.J. and J.G. Gosselink. 1986. Wetlands.
Van Nostrand reinhold. New York.
Moore, D.R.J., P.A. Keddy, C.L. Gaudet, and I.C.
Wisheu. 1989. Conservation of wetlands:
do infertile wetlands deserve a higher prior-
ity? Biological Conservation 47: 203-217.
National Acid Precipitation Assessment Program
(NAPAP). 1991. 1990 Integrated assessment
report. National Acid Precipitation Assess-
ment Program, 722 Jackson Place NW, Wash-
ington, D.C. 20503.
National Economics Research Associates (NERA),
Inc. 1994. The Benefits of Reducing Emis-
sions of Nitrogen Oxides under Phase I of
Title IV of the 1990 Clean Air Act Amend-
ments.
Neufeld, H. S., J. R. Renfro, W. D. Hacker, and D.
Silsbee. 1992. Ozone in Great Smoky Moun-
tains National Park: dynamics and effects on
plants. In: Tropospheric ozone and the envi-
ronment II - effects, modeling and control:
papers from an international specialty con-
ference; November; Atlanta, GA, Pittsburgh,
PA: Air & Waste Management Association;
pp. 594-617. (A&WMA transactions series:
TR-20).
Ontario Ministry of the Envkonment and Ministry of
Natural Resources. 1990. Guide to eating
Ontario sport fish. Public Information Cen-
tre, Environment Ontario, Toronto.
Owens, N.J.P., J.N. Galloway, and R.A. Duce. 1992.
Episodic atmospheric nitrogen deposition to
oligotrophic oceans. Nature 357:397-399.
Paerl, H.W. 1988. Nuisance phytoplankton blooms
in coastal, estuarine, and inland waters.
Limnol. Oceanogr. 33:823-847.
Paerl, H.W. 1993. Emerging role of atmospheric ni-
trogen deposition in coastal eutrophication:
biogeochemical and trophic perspectives.
Can. J. Fish. Aquat. Sci. 50:2254-2269.
Peterson D.C. et al. 1987. Improving accuracy and
reducing costs of environmental benefit as-
sessments. Office of Policy, Planning and
Evaluation, U.S. Environmental Protection
Agency, Washington D.C.
Peterson, D. L., and M. J. Arbaugh. 1988. An evalu-
ation of the effects of ozone injury on radial
growth of ponderosa pine (Pinus ponderosa)
in the southern Sierra Nevada. JAPCA 38:
921-927.
Peterson, D. L., and M. J. Arbaugh. 1992. Conifer-
ous forests of the Colorado front range. Part
B: ponderosa pine second-growth stands. In:
Olson, R. K.; Binkley, D.; Boehm, M., eds.
The response of western forests to air pollu-
tion. New York, NY: Springer-Verlag; pp.
365 and 385-401. (Billings, W. D.; Golley,
F.; Lange, O. L.; Olson, J. S.; Remmert, H.
Ecological studies; analysis and synthesis v.
97).
Peterson, D.L., M. J. Arbaugh, and J. R. Linday. 1991.
Regional growth changes in ozone-stressed
ponderosa pine (Pinus ponderosa) in the Si-
erra Nevada, California, USA. Holocene 1:50-
61.
Peterson, D. L., M. J. Arbaugh, V. A. Wakefield, and
P. R. Miller. 1987. Evidence of growth re-
duction in ozone-injured Jeffrey pine (Pinus
jeffreyi Grev. and Balf.) in Sequoia and Kings
Canyon National Parks. JAPCA 37:906-912.
E-20
-------
Appendix E: Ecological Effects of Criteria Pollutants
Powell, D. S., et al. 1993. Forest Resources of the
United States, 1992. USDA-Forest Service,
Fort Collins, CO. General Technical Report
RM-234.
Retzlaff, W. A., T. M. DeJong, and L. E. Williams.
1992a. Photosynthesis and growth response
of almond to increased atmospheric ozone
partial pressures. J. Environ. Qual. 21-
208-216.
Retzlaff, W. A., L. E. Williams, and T. M. Dejong.
1992b. Photosynthesis, growth, and yield
response of 'Casselman' plum to various
ozone partial pressures during orchard estab-
lishment. J. Am. Soc. Hortic. Sci. 117:
703-710.
Retzlaff, W. A., L. E. Williams, and T. M. DeJong.
1991. The effect of different atmospheric
ozone partial pressures on photosynthesis and
growth of nine fruit and nut tree species. Tree
Physiol. 8: 93-105.
Rochefort, L., D. Vitt, and S. Bayley. 1990. Growth,
production, and decomposition dynamics of
Sphagnum under natural and experimentally
acidified conditions. Ecology 71(5): 1986-
2000.
Roelofs, J.G.M. 1986. The effect of airborne sulphur
and nitrogen deposition on aquatic and ter-
restrial heathland vegetation. Experientia
42:372-377.
Rosenthal,D. 1990. Forest Recreation. NAPAPSOS/
T Report 27, Section B2.4, In: Acidic Depo-
sition: State of Science and Technology, Vol-
ume IV, National Acid Precipitation Assess-
ment
Rosenberg, R., R. Elmgren, S. Fleischer, P. Jonsson,
G. Persson, and H. Dahlin. 1990. Marine
eutrophication case studies in Sweden.
Ambio 19:102-108.
Rosseland, B.O. 1986. Ecological effects of acidifi-
cation on tertiary consumers: fish population
responses. Water Air Soil Pollution 30:451-
460.
Schofield, C.L., C.T. Driscoll, R. K. Munson, C. Yan,
and J.G. Holsapple. 1994. The Mercury Cycle
and fish in the Adirondack Lakes. Environ.
Science & Tech. 28:3:136-143.
Shaver, C. L., K. A. Tonnessen, and T. G. Maniero.
1994. Clearing the air at Great Smoky Moun-
tains National Park. Ecol. App. 4: 690-701.
Shriner D.S., W.W. Heck, S.B. McLaughlin, D.W.
Johnson, P.M. Irving, J.D. Joslin, and C.E.
Peterson. 1990. Response of vegetation to
atmospheric deposition and air pollution.
NAPAP SOS/T Report 18, JJK Acidic Depo-
sition: State of Science and Technology, Vol-
ume in, National Acid Precipitation Assess-
ment Program, 722 Jackson Place NW, Wash-
ington, D.C. 20503.
Shuyler L.R. 1992. "Cost Analysis for Nonpoint
Source Control Strategies in the Chesapeake
Basin." Annapolis MD: U.S. Environmental
Protection Agency Chesapeake Bay Program,
March.
Skelly, J. M., Y. S. Yang, B. I. Chevone, S. J. Long, J.
E.Nellessen, and W.E. Winner. 1984. Ozone
concentrations and their influence on forest
species in the Blue Ridge Mountains of Vir-
ginia. In: Davis, D.D.; Millen, A.A.;
Dochinger, L., eds. Air pollution and the pro-
ductivity of the forest: proceedings of the
symposium; October 1983; Washington, DC.
Arlington, VA; Izaak Walton League of
America Endowment; pp. 143-159.
Sorenson, J.A., G. E. Glass, K. W. Schmidt, and G.R.
Rapp, Jr. 1990. Airborne Mercury Deposition
and Watershed Characteristics in Relation to
Mercury Concentrations in Water, Sediment,
Plankton and Fish of Eighty Northern Min-
nesota Lakes. Environ. Sci. Technol. 24-
1716-1727.
Spry, D.J. and J.G. Wiener. 1991. Metal
bioavailability and toxicity to fish in low-al-
kalinity lakes: a critical review. Environmen-
tal Pollution 71:243-304.
E-21
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Turner, R.S., R.B. Cook, H. Van Miegroet, D.W.
Johnson, J.W. Elwood, O.P. Bricker, S.E.
Lindberg, and G. M. Hornberger. 1990.
Watershed and Lake Processes Affecting Sur-
face Water Acit-Base Chemistry. NAPAP
SOS/T Report 10, In: Acidic Deposition:
State of Science and Technology, Volume II,
National Acid Precipitation Assessment Pro-
gram, 722 Jackson Place NW, Washington,
D.C. 20503.
U.S. DOI, 1993. Fish and Wildlife Service and U.S.
Department of Commerce, 1991 National
Survey of Fishing, Hunting, and Wildlife-
Associated Recreation, U.S. Government
Printing Office, Washington D.C.
U.S. EPA. 1993. Air Quality Criteria for Oxides of
Nitrogen. Office of Health and Environmen-
tal Assessment, Environmental Criteria and
Assessment Office, Research Triangle Park,
NC; EPA report no. EPA600/8-91/049bF. 3v.
U.S. EPA. 1994. Deposition of Air Pollutants to the
Great Waters: First Report to Congress. Of-
fice of Air Quality Planning and Standards.
Research Triangle Park, NC: EPA Report
Number EPA-453/R-93-055.
U.S. EPA. 1995. The Benefits and Costs of the Clean
Air Act 1970 to 1990 — Report to Congress.
U.S. EPA. 1996a. External Draft, Air Quality Crite-
ria for Ozone and Related Photochemical
Oxidants. Volume E. Office of Health and
Environmental Assessment, Environmental
Criteria and Assessment Office, Research
Triangle Park, NC; EPA report no. EPA/600/
AP-93/004af-cf.
U.S. EPA. 1996b. Review of National Ambient Air
Quality Standards for Ozone: Assessment of
Scientific and Technical Information: OAQPS
Staff Paper. Office of Air Quality Planning
and Standards. Research Triangle Park, NC;
EPA report no. EPA-452\R-96-007. June.
Vitousek, P.M. and R.W. Howarth. 1991. Nitrogen
limitation on land and in the sea: How can it
occur? Biogeochemistry 13:87-115.
Walsh, R.G., F.A. Ward, and J.P. Olienyk. 1989. Rec-
reational demand for trees in national forests.
J. Environ. Manage. 28:255-268.
Watras, C.J. and N.S. Bloom. 1992. Mercury and
methylmercury in individual zooplankton:
implications for bioaccumulation. Limnol-
ogy and Oceanography 37:1313-1318.
Watras, C.J., N.S. Bloom, RJ.M. Hudson, S. Gherini,
R. Munson, S.A. Claas, K.A. Morrison, J.
Hurley, J.G. Wiener, W.F. Fitzgerald, R.
Mason, G. Vandal, D. Powell, R. Rada, L.
Rislov, M. Winfrey, J. Elder, D. Krabbenhoft,
A.W. Andren, C. Babiarz, D.B. Porcella, and
J.W. Huckabee. 1994. Sources and fates of
mercury and methylmercury in Wisconsin
lakes. In: Mercury Pollution: Integration and
Synthesis, C.J. Watras and J.W. Huckabee,
Eds. Lewis Publishers, Boca Raton, Florida,
pp. 153-180.
Wiener, J.G., R.E. Martini, T.B. Sheffy, and G.E.
Glass. 1990. Factors influencing mercury
concentrations in walleyes in northern Wis-
consin lakes. Trans. Am. Fish. Soc. 119:862-
870.
Wisheu, 1C. and P.A. Keddy. 1989. The conserva-
tion and management of a threatened coastal
plain plant community in eastern North
America (Nova Scotia, Canada). Biological
Conservation 48: 229-238.
E-22
-------
Appendix F: Effects of Criteria Pollutants on
Agriculture
Introduction
One potential impact of air pollutants on economic
welfare is their effect on agricultural crops, including
annual and perennial species. Pollutants may affect
processes within individual plants that affect growth
and reproduction, thereby affecting yields of agricul-
tural crops. Possible physiological effects of pollut-
ants include the following: decreased photosynthesis;
changes in carbohydrate allocation; increased foliar
leaching; decreased nutrient uptake; increased sensi-
tivity to climatic stress, pests, and pathogens; de-
creased competitive ability; and decreased reproduc-
tive efficiency. These physiological effects, in con-
junction with environmental factors and intraspecies
differences in susceptibility, may affect crop yields.
Primary air pollutants that might damage plants
include SO2, NOx, and volatile organic compounds
(VOCs). These pollutants may have direct effects on
crops, or they may damage crops indirectly by con-
tributing to tropospheric (ground-level) ozone,
peroxyacetyl nitrate (PAN), and/or acid deposition,
all of which damage plants. Tropospheric ozone is
formed by photochemical reactions involving VOCs
and NOx, while SO2 and NOx cause acidic deposition.
While all of these air pollutants may inflict mere-
mental stresses on crop plants, in most cases air pol-
lutants other than ozone are not a significant danger
to crops. Based primarily on EPA's National Acid
Precipitation Assessment Program (NAPAP) conclu-
sions,1 this analysis considers ozone to be the primary
pollutant affecting agricultural production.
This analysis estimates the economic value of the
difference in agricultural production that has resulted
due to the existence of the CAA since 1970. The analy-
sis is restricted to a subset of agricultural commodi-
1 Shriner et al., 1990; NAPAP, 1991.
ties, and excludes those commodity crops for which
ozone response data are not available. Fruits, veg-
etables, ornamentals, and specialty crops are also ex-
cluded from this analysis. To estimate the economic
value of ozone reductions under the CAA, agricul-
tural production levels expected from control scenario
ozone conditions are first compared with those ex-
pected to be associated with ozone levels predicted
under the no-control scenario. Estimated changes in
economic welfare are then calculated based on a com-
parison of estimated economic benefits associated with
each level of production.
Ozone Concentration Data
To estimate the nationwide crop damages as a
result of ozone exposure, the first step is to estimate
the nationwide ozone concentrations under the con-
trol and no-control scenarios. This section describes
the methodology used to estimate ozone concentra-
tions for each county in each of these two scenarios.
First, historical ozone concentration data at the
monitor level were compiled from EPA's AIRS sys-
tem. Differences between the modeled control and no-
control scenario ozone concentrations were then used
to scale historical data to derive no-control scenario
ozone air quality profiles.2 Next, the ozone index used
in the exposure response evaluation was calculated
and applied at the monitor level. For this analysis, the
W126 index, a peak-weighted average of cumulative
ozone concentrations, was selected to conform with
the index currently being used by EPA in ozone
NAAQS benefits analysis. The W126 index is one of
several cumulative statistics, and may correlate more
closely to crop damage than do unweighted indices.3
EPA has not yet made a final determination of the
appropriate index to use in agricultural benefits analy-
descritedntdnn A^endbfc^ qU3Hty Pr°fileS ** ^ C°ntrO1 ^ no'contro1 scenario is summarized in the following section and
3 Lefohn et al., 1988.
F-l
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
sis; thus this analysis should be viewed only as an
indicator of the magnitude of potential benefits.
The third step in ozone concentration estimation
involved the use of triangulation and planar interpo-
lation to arrive at a W126 statistic at the county, rather
than at the monitor, level. For each county centroid,
the closest surrounding triangle of monitors is located
and the W126 is calculated for that county using a
distance-weighted average of the ozone concentration
at each of these monitors.
Control and No-control Scenario Ozone
Concentration Data
The initial estimation of ozone concentrations in
the control and no-control scenarios was performed
by Systems Applications International (SAI). To cre-
ate the control scenario, SAI compiled ozone data from
the EPA's Aerometric Information and Retrieval Sys-
tem (AIRS).4 SAI summarized these data by fitting
gamma distributions to them and providing the alpha
and the beta parameters to these distributions. Each
of these distributions describes a set of ozone con-
centration levels, and the distributions are categorized
by year, season, and averaging time. SAI defines six
distinct "seasons," each composed of a two month
period in the year. This analysis uses those distribu-
tions which describe 1-hour average ozone concen-
trations taken from 7 AM to 7 PM and separated into
seasons. The analysis utilizes only those monitor
records that were modeled in both the control and no-
control scenarios.
To determine the ozone concentrations for the no-
control scenario, SAI utilized the Ozone Isopleth Plot-
ting with Optional Mechanisms-IV (OZIPM4) model.
The input data required for OZIPM4 includes air qual-
ity data, surface and upper-air meteorological data,
and estimates of anthropogenic and biogenic emis-
sions of volatile organic compounds, NOx and CO.5
To create these inputs, SAI used (among other sources)
outputs from the Regional Acid Deposition Model
(RADM) and the SJVAQS/AUSPEX Regional Mod-
eling Adaptation project (SARMAP). Additional de-
tail concerning the development of ozone concentra-
tion data is available in Appendix C and in the SAI
report to EPA.6
4 SAI, ICF Kaiser, 1995.
5 SAI, ICF Kaiser, 1995.
6 SAI, ICF Kaiser, 1995.
7 Lefohn et al.. 1988.
Calculation of the W126 Statistic
Using the SAI ozone concentration distributions,
we calculated a sigmoidally weighted ozone index for
each monitor. The generalized sigmoidal weighting
function used in calculating such indices is presented
in Lefohn and Runeckles (1987) as:
where:
w. = weighting factor for concentration
(unitless)
c. = concentration (ppm)
M= an arbitrary constant
A = an arbitrary constant
The constants M and A are chosen to give different
weights to higher or lower concentrations. The index
used in this analysis is the W126 statistic, which is
calculated as follows:7
and
Missing values are accounted for by multiplying the
resulting W126 statistic by the ratio of the number of
potential observations to the number of actual obser-
vations (i.e., total hours in period/hours of data in pe-
riod).
To calculate W126 indices from the monitor level
gamma distributions, we used an inverse cumulative
density function to calculate a separate representative
air concentration for each hour in the two month sea-
son. These values are then used in the above equation
to obtain a monitor-level W126 statistic.
To ensure that the interpretation of the gamma
distributions in this manner does not generate errors,
we tested our gamma-derived control-scenario W126s
F-2
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
against W126s calculated directly from the AIRS da-
tabase. We found that insignificant error resulted from
the utilization of the gamma distributions to create
W126 statistics.
Aggregating Ozone Data to the County
Level
Because crop production data are available at the
county level, the lowest level of aggregation that could
be used for ozone indices is also the county level.
Therefore, monitor level data needed to be aggregated
to a county level. For each county, we first located
the monitors from which we would be interpolating
data. To identify these monitors, we searched for the
three monitors which formed the closest triangle
around the centroid of the county.8 The closest tri-
angle was defined as that triangle in which the sum of
the distances from the three monitors to the county
centroid was the least. The algorithm stopped search-
ing for closest triangles of monitors when it had
searched all monitors within 500 km of a given county
centroid (an arbitrary distance, selected to reduce com-
putational requirements).
For coastal counties and some rural counties in
some years, monitor triangles around the county cen-
troid do not exist. We assigned the W126 value from
the monitor closest to the centroid to these counties.
Approximately 15 percent of all county-years (36,973
of 248,880 records) were assigned W126s in this man-
ner.
For the remaining 85 percent, after the closest tri-
angle of monitors was found, a "planar interpolation"
was used to calculate the W126 at that county for that
year. One way to picture this process is to plot each of
the three monitors as a point in space. For each moni-
tor, the x axis represents longitude, the y axis repre-
sents latitude and the z axis represents the W126 sta-
tistic. A plane can then be drawn through these three
points in space. Furthermore, using the equation for
the plane, and given the x and y values (latitude and
longitude) for the county centroid, the county
centroid's z value (W126 statistic) can be calculated.
In essence, this procedure calculates a distance-
weighted average of three monitors' W126 index val-
ues and assigns this value to the county centroid.
The result of this data manipulation is a monthly
W126 statistic for each county in the continental
United States for the years 1971-1990. From these
data, yield change estimates were generated, and eco-
nomic impacts were estimated.
Yield Change Estimates
There are several steps involved in generating
yield change estimates. The first is the selection of
relevant ozone exposure-response functions (mini-
mum and maximum) for each crop in the analysis.
Ozone data, triangulated to the county level, are trans-
formed into an index suitable for use in the selected
function(s) to estimate county level predicted yield
losses for both the control and no-control scenarios.
In the next step, the proportion of each county to the
national production of each crop is calculated to per-
mit national aggregation of estimated yield losses.
Finally, the control scenario percentage relative yield
loss (PRYL) is compared to the minimum and maxi-
mum PRYL for the no-control scenario. Each step is
discussed in more detail below.
Exposure-Response Functions
To estimate yield impacts from ozone, exposure-
response functions are required for each crop to be
analyzed. This analysis was restricted to estimating
changes in yields for those commodity crops for which
consistent exposure-response functions are available
and that are included in national agricultural sector
models. To maintain consistency with the current
ozone NAAQS benefits analysis being conducted by
OAQPS, NCLAN-based exposure-response functions
using a Weibull functional form and a 12-hour W126
ozone index were used.
Several crops included in the NCLAN research
program were not evaluated hi this analysis. Non-com-
modity crops that are not modeled in national agri-
cultural sector models were not included in this analy-
sis: lettuce, tomatoes, potatoes, alfalfa, tobacco, tur-
nips, and kidney beans. In addition, one commodity
crop, spring wheat, was excluded because the NCLAN
exposure-response function was only developed for
winter wheat.
The vast majority of monitors had latitude and longitude data available through AIRS. 1,528 of 1,536 monitors were located in
this manner. For the remaining 8 monitors, if in a given year of monitor data another monitor exists in the county of the unfound
monitor, we discarded the unlocated monitor's data. Otherwise, we located that monitor at the county's centroid. We located 5 of the
remaining 8 monitors in this fashion.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Minimum/Maximum
Functions
Exposure-Response
Estimated responsiveness of a given crop to ozone
varies within the NCLAN data. This range of response
is partially explained by the program's evaluation of
several cultivars for some crops; ozone sensitivity
varies across cultivars. In addition, the conditions for
different experiments varied due to variations in lo-
cation, year, and additional treatments included in
some experiments. No one exposure-response func-
tion can be assumed to be representative of all culti-
vars in use, or of all environmental conditions for crop
production. To develop a range of benefits estimates
that reflects this variation in responsiveness, a mini-
mum responsiveness and a maximum responsiveness
function were selected for each crop. In actuality, a
number of different cultivars are planted by produc-
ers, and so ozone response will be a weighted average
of the responsiveness of each cultivar to its ozone con-
dition and its proportion of total acreage. It is impor-
tant to note that these values do not necessarily bound
the analysis, since the number of cultivars evaluated
by NCLAN is small relative to the number grown for
many crops.
for each crop. Two crops, peanuts and sorghum, did
not have multiple NCLAN experiments on which to
base a comparison of the responsiveness of different
cultivars or the variation in response with different
experimental conditions.
Calculation of Ozone Indices
Each NCLAN ozone exposure-response experi-
ment exposed each studied crop over a portion of the
crop's growing season. The duration of the NCLAN
experiments was provided by CERL and was rounded
to the nearest month. The W126 index is cumulative,
and so is sensitive both to the duration over which it
is calculated and to the specific month(s) within a
growing season that are included in it. Because crop-
ping seasons vary across the U.S., the ozone index
used to calculate county-level changes in yield due to
ozone must reflect the local season for each crop. To
determine which portion of the growing season a par-
ticular exposure period pertains to (in order to calcu-
late an exposure index), we developed state-specific
growing seasons based on planting and harvesting data
developed by USDA.9 The W126 index was calcu-
For the crops used
in this study, CERL
conducted an analysis
to identify the ozone
concentration required
to reduce yields by 10
percent for each crop
cultivar using its 12-
hour W126 exposure-
response function. For
each crop, the function
demonstrating the low-
est ozone concentration
at a 10 percent yield
loss represents the
maximum response,
and the function with
the highest concentra-
tion at 10 percent yield
loss represents the mini-
mum response. Table F-
1 reports the minimum
and maximum expo-
sure-response functions
Table B4, ^Agriculture Bxppsitre-Response^unctiais
" "*
Crop
Barley
Corn-Field ., ,-
Com-Eaeld
Cotton
Cotton
Grain"
Sorgnnm «
Peanuts •',
Soybeans
Soybeans
Wheat
Whe,at
Honeer;3780
MdSfeir235
AealaWx
-NC-6" <
" Davis
-,ART
Equation
Min
MiqC, ,
Max
Both
Both ''
Max
Mia -,
,." ',*>
Max
Yield Function
Duration
*'
a
-83
125-
85
•'
112*
Soorcei EPj&iasuLinnpublisIied) fix*ill ftnctioms.
,,>?,>! v~- -*" ,, ,,; .'•...-.....i^l,';,''-
9 USDA, 1984. Some states did not have explicit growing seasons reported for certain crops due to the low production in these
states. In these cases a proxy state growing season was used. In most of these cases the proxy growing season was taken from a state
with an adjoining boundary within the same geographic region.
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
lated using the county level ozone data developed in
the prior section, summed for the number of months
of NCLAN experimental duration, with the exposure
period anchored on the usual harvest month for each
dex. Next, the county weights are applied to the
PRYLs. The change in yield, measured relative to the
hypothetical zero-ozone crop production, is then:
crop
10
Calculations of County Weights
Because the benefits analysis did not require a
regional level of disaggregation and to minimize com-
putational burdens the economic analysis was con-
ducted at a national level. Ozone data and estimated
yield responses, however, were developed at a county
level. To conduct a national analysis, the county level
yield change estimates were weighted to develop a
single national percent relative yield loss for each crop
relative to the control scenario, for both the minimum
and the maximum yield responses. As a part of cal-
culating a percent change hi yield at the national level,
weights for each county and crop were created for
1975, 1980, 1985, and 1990. The weights for these
four years were used to represent the year itself and
the four years immediately following it (e.g., 1975
weights were also used for 1976, 1977, 1978, and
1979). Although weather and other conditions may
change the proportion of counties' production to the
total national production in each year, five year
weights should reflect stable periods of agricultural
policy between each Farm Bill, and are sufficient for
the level of precision needed for this analysis. The
weights were calculated by dividing the production
level of a crop in a county11 by the sum of all states'
reported production for that crop.12 These county
weights were applied to the percent relative yield loss
results for each county, as discussed below.
Calculation of Percent Change in Yield
Ozone exposure-response functions are expressed
in terms of percent relative yield loss (PRYL); the
ozone level being analyzed is compared with "clean"
(charcoal filtered/zero ozone) air. To calculate a per-
cent change in yield between the control and no-con-
trol scenarios, we first calculate a PRYL based on the
county-level control scenario W126 ozone index, and
a PRYL based on the no-control scenario W126 in-
To obtain the change in terms of our (non-zero)
baseline yield, we divide by that yield, and get:
To create the national percent change in yield for
each crop, the results of this equation are summed for
each scenario (maximum and minimum) and for each
year. Tables F-2 and F-3 present the resulting percent
yield changes that were used as inputs to the economic
model.
Economic Impact Estimates
To estimate the economic benefits of controls on
ozone precursor pollutants under the Clean Air Act,
changes in yields due to those controls need to be
evaluated in terms of their effect on agricultural mar-
kets. To do this, yield changes can be incorporated
into an economic model capable of estimating the as-
sociated changes in economic surpluses within the
agricultural economy, preferably one that reflects
changes in producers' production decisions and de-
mand substitution between crops. This type of dy-
namic analysis is needed because even small changes
in yield or price expectations can cause large shifts in
the acreage allocated to specific crops, and the degree
to which alternative crops will be substituted (particu-
larly for feed uses).
Agricultural Simulation Model (AGSIM)
The modeling approach used in this analysis is an
econometric model of the agricultural sector, which
estimates demand and supply under different produc-
tion technologies and policy conditions. The
AGricultural Simulation Model (AGSIM) has been
10 This analysis required "rounding" some months: if a harvest date was specified to be from the 15th to the end of a month the
W126 index was calculated using that month's data; if the harvest date was specified to be from the first to the 14th of a month the
W126 index was calculated using the prior month's data as the final month in the exposure period.
11 USDA, 1995.
12 The national total does not include USDA areas designated "other counties". These areas are groups of counties that for one
reason or another (disclosure rules, low amount of production, etc.) are not individually listed. Because we did not have ozone values
for these groups, we did not use their production levels in the calculation of the total national production.
~ ~
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table F-2. Relative No-control to Control Peroenf'Tfleld €hangeXharvested acresffor the Mimrflura,,'*
Scenario. "
Year
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
• v-*>* ' ,„ 4 ' "'Ci-op,,, ,„ ;, „„-
Bariey
-0.000020
-0.000013
.0.000013
-0.000019
-0.000027
-0.000019
-0.000016
-0.000020
-0.000023
-0.000027
-0.000025
-0.000029
-0.000033
-0.000027
-0.000024
-0.000024
Com
-0.000171
-0.000329 ,-
-0.00016?-,,
-0.000291 '
•-0.000100 -
-0.000200 ,
-0.000071
-0.000070,,,
-0.000617
-o'.ooom -
-0.000132
-0,000158 ' '
-0.000358 '•
•-0.000662,,,
-O.OOOl'SO'
-0.000210
Cotton- ^
-0.011936 '
. -0:017505
*-o:oi3"i-ii--
-0.018692 *
-QJ)lJ2i7,,
,--0.021315 '-'
-0.018552
,.,^,.017295
-0.020842 ' '
-'0.023552 '
, -0.020947
, T,0.0|7.968>
' '-'OIQ34584
,t,-0,Q350!59-
-0;03'i245l,4
-0.037988 ,
', Peanuts '"
"16.006635
-0:024048 ,
, T0.015,1S,0. -
't-,0:oi7<>Q6" -*
-0.013067- -
;«ps02276i,^;
-,iO'.0142'69
-0.01'42QO
-0.028601
-0.01S225,-
-0,017965- -1
'^031605
-Q.043S54"
-0.038085 '.
;^.022094
^0.04,7-3^5
* -'Soybeans
'-0.001166 -
-0,00217-1-
,;-0.001562-'
;,vo;oo248o
-0.0018'98
-0.002397
-OM1951
-0.00210?
-0.003901 - '
-0.002919
' -OVQ02645
-'-0,00289,9^ ,
:0.003776
-0.004563
•--0,00'3769"
-dXtisiis
-' --
/,/-
Sorghiim
'•ra;090fi7 -
^0,001841
-o.ooajus.
-0,001844
-0,001389,
-0.002222'"
-0.0008&
-0.001050 '
-' -0;0023r66,_,
-0.002881
-,0,0017,26:'
,, T0.001564''
-0.001812
-0.0029,22
,< -0.001359
-0.00,1567
'/ -7 , -
,• •" , ;>,,-
Winter Wheat
- -0.005631,
-0.004841- *
4J.005464,
' '"-b'.'b05894 - --
-O.Ob'4^98
,-0.00538-5
- -0.003964,,
, -0.004773 „
„, ,-,0.005904 ,
- -0:006121 -
-0,00'73l6
;'"J^pjJ7597' '"
-0.009669 "
-0.0198t3, <
-0.007605" '
-0.00644^" '••
J *>:;>,:?" __ s/ , , ,c,, ^ "f •*""" ' ,, ^ ' * '' " ^ v •> * /'^-'V)'^ "*" f" e fft /
~^_, •*ff£-'ty.sf*f •> ' •,' ' ^ \_^*'' , "V ^ % \^J'^f'^'' 'J '
Note: TTieie is only one scenario for parfey^p'eamits, and sorgferm;,becaijse there was only'one cxp^ure-iB sponge fcno,fion.^ -'?' '• '-'' "
Table F-3, Relative No-control to ControlPerc&lpieldChange (harvested ao»s) fortJ»,-Ma
, - ' "" " " s ' - - , ^',^5'>^V ••'''' •"'' J '
Scenario. ,„ s ' ,,.,'•->«?_;
Year
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
Barley
-0.000020
-0.000013
-0.000013
-0.000019!
-0.000027
-0.000019
-0.000016
-0.000020
-0.000023
-0.000027
-0.000025
-0.000029
-0,000033
-0.000027
-0.000024
-0;000024
Corn
-0.001139
-0.0022S1
-0.00'1232
-0:002015''
-0.001052
-0.001537 „
-0.000923
-0.000974
-0.003838 ,
-0.001443,,,,
-0.001377xr
-0.00X451 '
-0.002565
-0.004318
-0.001987
-0.002056 <
,-,{, sj^.
Cottoa
-0.02105&"
'-0.032063
-0.02577,3,^
, -0.033075>
-0,031433 '
• A&nn-
.01035058T-
-0.034101 -
-0.040405
^Q.043890-'-'
"J5M3845 „
-0,052426
-0.061295 „
--0.061660 '
-tTd.0^573 '
• 3&effi&~
f~-f ' "" -'•'"' "• "•
--/,) Croj)
"SPftw&ifts ,;
-0.006635- '•
-9,024048
,,,,-0.015150
,,--0.017606"
4XO'13'067
-0-022761
r-0".bl4269
^£06*208"
-0.028601
'-0,019225
,,^,OJ,7$65.--,
^0-03^05^
,JMM&&-
-0.038085'
-0,022094- "
-0-047395
^^bybeans '
"iO,0058Q8
-0.010298;- :-
-0,007764""'
"-0:bll803 -
-0.009592
-0,011845
, -0.009902
-0,010815'"
-0-.01859?
-0,014502'
-0.013384 „
;--0,;G14754 ,
r-o;oi85?8 :-;
' $.0217671, ,
'*'-0:Ol8739 "
'-0,0'1'8670 '
-" '•? -*'s
^Iprghiom-
-:'d;00"0tlt,,.
- -0,001841-
""'0:001118 '-
"-O.'OO'IS'M
--0,001389-,,,
' -0.002222'-
-0,000802
' ^0,001050
-0,002366,-
-0.002881 '"
v-0.001726
^m^--<,
'COOM'2' ''
-0,002922,-,,
''-O'.'d0l3l9'"'
-0.001567
^ j > r'i» ' *f* f
-Winter Wheat
, -OM803^
---- "-0,040303
"^-G,049'6^'
' -0.050308 ' -
,v,,,;-0.0522n -
" ""-D.054128,",
-0.053470 ,
' ' -'d05S46£ '
. -0.063556 -'
" -d,'067612
,-0.072177- ,.;,
-'-,-,-,---0.081225 V-
'" rO,089042 •'
----4?0:-1-20703
-0,086958,, ,
, ,-0.0823,09--,w.
< fff *''v \ v <. ^
Note: There is only one scenario f«badey,'j>eaflufe,,and s,«ghu!ft, because there was-
' »'"
F-6
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
used extensively to evaluate air pollution impacts, as
well as a number of other environmental policy analy-
ses: AGSIM is an econometric-simulation model that
is based on a large set of statistically estimated de-
mand and supply equations for agricultural commodi-
ties produced in the United States. The model is ca-
pable of estimating how farmers will adjust their crop
acreages between commodities when relative profit-
ability changes as a result of crop yield and produc-
tion cost changes. Acreage and yield changes from
various scenarios will affect total production of crops,
which then affects commodity prices and consump-
tion. The commodity price changes, in turn, affect
profitability and cropping patterns in subsequent years.
Federal farm program and conservation reserve ef-
fects are also incorporated into the model.
The initial version of AGSIM (which went
through various acronym revisions) was developed
in 1977.13 The model was developed to permit esti-
mation of aggregate impacts associated with relatively
small changes in crop yields or production costs, which
might result from various policy conditions such as
changes in pesticide input availability, or in this case,
changes in crop exposure to ozone. Subsequent revi-
sions to the model as well as the current specification
are described in Taylor (1993a).14 Several policy ap-
plications of AGSIM were tested and reported in Tay-
lor (1993b)i5 to provide a comparison to the results
of several alternative agricultural sector models. These
tests included an expansion of Conservation Reserve
acreage, reduced target prices, elimination of agricul-
tural programs in the U.S. other than the Conserva-
tion Reserve Program (CRP), and a tax on nitrogenous
fertilizer use in the U.S. The model has been used to
evaluate the effects of changes to the CRP,16 changes
in agricultural price support programs,17 and evalua-
tions of policies concerning pesticide availability.18
AGSIM is designed to estimate changes in the
agricultural sector resulting from policies that affect
either the yields or the costs of crop production.
Changes in economic variables are computed by com-
paring a policy simulation of the model with a baseline
simulation of the model. For this retrospective evalu-
ation, the baseline reflects actual farm programs,
prices, and other parameters since 1970. The model's
author, Dr. C. Robert Taylor, modified AGSIM for
this analysis to reflect production conditions and poli-
cies as they changed through the 20-year span of the
Clean Air Act, from 1970 to 1990. During this pe-
riod, U.S. farm policy parameters changed every five
years with the enactment of each Farm Bill, and pro-
ducer participation varied significantly over the pe-
riod. Over this time, due to policy, weather, techno-
logical development, and other variations, production
levels and prices have varied, as have production tech-
nologies, costs of production, and relevant cultivars.
To reflect these changes, Dr. Taylor re-estimated de-
mand relationships for three periods (1975-79; 1980-
84; and 1985-89) based on the farm policies in effect
in each period, and structured the model to run on a
national level rather than a regional level. The period
from 1970-1975 was not modeled because of data limi-
tations and because there was limited impact from the
CAA on ozone levels during that period.
The AGSIM baseline production and price data
serve as the control scenario baseline. Percent rela-
tive yield losses (PRYLs) between the control and no-
control scenarios are the relevant input parameter for
this analysis, from which AGSEVI calculates new yield
per planted acre values. Based on these values (as well
as on lagged price data, ending stocks from the previ-
"Taylor, C.R R.D. Lacewell, and H. Talpaz. 1979. Use of Extraneous Information with the Econometric Model to Evaluate
Impacts of Pesticide Withdrawals. Western J. of Ag. Econ. 4: 1 -8.
TT • ''Baylor. C.R. 1993a. AGSIM: An Econometric-Simulation Model of Regional Crop and National Livestock Production in the
United States. In: C.R. Taylor, K.H. Reichelderfer, and S.R. Johnson (Eds) Agricultural Sector Models for the United States-
Descriptions and Selected Policy Applications. Ames Iowa: Iowa State University Press.
15 Taylor, C.R. 1993b. Policy Evaluation Exercises with AGSIM. In: C.R. Taylor, K.H. Reichelderfer, and S.R. Johnson (Eds)
Pre"s °r Cd StateS: Descriptions and Selected Policy Applications. Ames Iowa: Iowa State University
S 199?' SAUP,Pl£ fJSST? tSpeCtf °f ^ C,onservation Reserve- I* T-L- Napier (Ed) Implementing the Conservation
T D r, llo, Aty Act °f I985' A"keny-' Iowa: S011 md Water Conservation Society; Taylor, C.R., H.A. Smith, J.B. Johnson
and T.R. Clark. 1994. Aggregate Economic Effects of CRP Land Returning to Production. J. of Soil and Water Conservation 49:325-
32o.
Deterministic vs" Stochastic Evaluation of the Aggregate Effects of Price Support Programs. Agricultural
"Taylor, C.R. G A. Carlson, FT Cooke, K.H. Reichelderfer, and I.R. Starbird. Aggregate Economic Effects of Alternative Boll
Weevil Management Strategies. Agricultural Econ. Res. 35:19-19;Taylor, C.R, J.B. Penson Jr., E.G. Smith, and R.D. Knutson 1991
Impacts of Chemical Use Reduction in the South. S. J. Of Ag. Econ. 23: 15-23. *vuuu,uu. i»» i .
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
ous ye&:, and other variables), AGSIM predicts acre-
age, production, supply, and price parameters for each
crop for each year, as well as calculating yield per
harvested acre. From these results and the demand
relationships embedded in the model, AGSIM calcu-
lates the utilization of each crop (i.e., exports, feed
use, other domestic use, etc.), as well as the change hi
consumer surplus, net crop income, deficiency pay-
ments and other government support payments. Net
surplus is calculated as net crop income plus consumer
surplus, less government payments. The first year of
results is 1976 because AGSIM must have one year
(1975) of lagged data.
Table F-4 presents the net changes in economic
surpluses (in 1990 dollars) annually and as a cumula-
tive present value (discounted at 5 percent) over the
period 1976-1990 due to the Clean Air Act. The posi-
tive surpluses exhibited in almost all years are a re-
sult of the increase in yields associated with lower
ozone levels than those predicted to occur under the
no-control scenario. The present value of the estimated
agricultural benefits of the CAA ranges between $7.8
billion hi the minimum response case to approximately
$37 billion in the maximum response case. This range
represents the impacts that would occur if all of the
acreage planted to a given crop had an ozone response
function similar to either the minimum available re-
sponse function or the maximum available response
function. The available response functions do not nec-
essarily bracket the true range of potential crop re-
sponses, and it is unrealistic to anticipate that all acre-
age will be planted in cultivars with a uniform response
to ozone exposure. These considerations notwithstand-
ing, these values do indicate the likely magnitude of
agricultural benefits associated with control of ozone
precursors under the CAA, but not the precise value
of those benefits. In addition to estimating the present
value of net surplus at a discount rate of five percent,
two alternative discount rates were used. At a three
percent discount rate, the range of net surplus is be-
tween $6.7 billion and $32 billion; at seven percent
discount rate, the range is between $9 billion and $43
billion. For more detail on AGSIM intermediate model
outputs, see Abt Associates (1996).
Table F-4. Change in FanrtProgram PSyirie'nts, -Nefcdrop Income, ,Cptt8:
Surplus Due to the CAAX«i||lions',;
.. . ifsj.j.&,\$?v$ •kaaj''**?•, v*'**'^.''
piusl'andNet „-;;
*•&£&;?,, t f ' ••
Year
1976/77
1977/78
1978/79
1979/80
1980/81
1981/82
1982/83
1983/84
1984/85
1985/86
1986/87
1987/88
1988/89
1989/90
1990M
Cbaugeitt
Farm Program IPaymtntst
Minimum
0
0
43
0
0
112
168
153
-182
289
270
469
557
329
414
Maximum
V C
'0
345
0
0
518
981
1,009
808
1,291'
'1,356
2,033
2,023
1,401-
, 1,927,
- -%"*tfo&&itt^v^
- --'/Net CropIhcbineu'l-A
Minimum
' "\v.3M3
'• - -97,
•' 30
1 -140
&
-99
, 64
, , 231
- ;, ,82
-181-
' 230
-<;3-20%,
. 316 '
-*•« M61
, 180
Maximum
"~"4W
^«25£;
, -^£8,
,,*;,-406-
"^3.78 '
,,~4p6,.
' -^IQZ
' -V-'^gj^
"""W
- ,„> ,.R7Q
,-~t^^966
4r^'i;m.,,
--',1,508^
,/"";$**
\ v5><\5<$*.V ^*• ' 618"
Maximum
„ JPs;;
•j's /<,*$* 1^55/N'
•v*" :-'l,646
"" '"2,000
,,, 2,049'
'2,594
" '2,730
, , 1;9'69 '
,rTJi686
"' -2,054
' " ' 2,265 -
2,999-
2,943--
/&5T2",
'--•*3';047;.
Cumulative Present Value of Net, Surplus at 5 per<«ntJ0£90) t -,->v<-,-?,^;,>, ,;,' ""'"•''"'"' '" ; '
>*>« v. -Change in
- -;,;NetSurjdusv''v'^
Minimum
r',"'r'4??-
,-• ' 253'
379
309 ,
400
, 231
273
• - ' 395 ,,
-14
509
- -42-2
, - 558
--/ ,- 556 ,
" ' 35V
- . m*.
* * ;' 7,763-".
Maxinmm
i,m
'l,-297
1'397
,- ;'1394
.-,. ,,,,1,870
' ';'---a,67Q
'""1,856
„ „ 'i'^17,,
"^VJ»437<
'-1^44
' T875'
'2,M"
2,428,
"*'- --U785
„„ , 1,593'"
'v-, -37,^25'
F-8
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
Conclusions
Agricultural benefits associated with control of
ozone precursors under the Clean Air Act are likely
to be fairly large. Because it is possible that over time
producers have adopted more ozone-resistant culti-
vars, it may be appropriate to consider the lower end
of the range of predicted benefits to be more indica-
tive of the likely total benefits. The estimates devel-
oped in this analysis, however, do not represent all of
the likely benefits accruing to agriculture, in that many
high-value and/or ozone sensitive crops could not be
included in the analysis due to either exposure-re-
sponse data limitations or agricultural sector model-
ing limitations. The second consideration implies that
benefits will likely be larger than estimated. The mini-
mum case may be the most appropriate starting point,
however, due to the first consideration: the current
crop mix is probably biased toward lower ozone re-
sponsiveness. Therefore, we anticipate that cumula-
tive total agricultural benefits from the Clean Air Act
are on the order of ten billion dollars (real terms).
F-9
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Agricultural Effects References
Abt Associates. 1996. Section 812 Retrospective
Analysis: Quantifying Health and Welfare
Benefits (Draft). Prepared by Abt Associates
under Contract No. 68-W4-0029. U.S. EPA,
Office of Policy, Planning, and Evaluation.
Lefohn, Allen S. et. al. 1988. A comparison of indi-
ces that describe the relationship between
exposure to ozone and reduction in the yield
of agricultural crops. Atmospheric Environ-
ment 22: 1229-1240.
Lee, E. Henry et. al. 1994. Attainment and effects is-
sues regarding alternative secondary ozone air
quality standards. J. Environ. Qual. 23:1129-
1140.
National Acid Precipitation Assessment Program
(NAPAP). 1991.1990 Integrated assessment
report. National Acid Precipitation Assess-
mentProgram, 722 Jackson Place NW, Wash-
ington, D.C.20503.
SAI, ICF Kaiser. 1995. Retrospective Analysis of
ozone air quality in the United States: final
report. Prepared by Systems Applications In-
ternational under contract 68-D4-0103. U.S.
EPA, Office of Policy Analysis and Review.
Shriner, D.S., W.W. Heck, S.B. McLaughlin, D.W.
Johnson, P.M. Irving, J.D. Joslin, and C.E.
Peterson. 1990. Response of vegetation to at-
mospheric deposition and air pollution.
NAPAP SOS/T Report 18, IQ: Acidic Depo-
sition: State of Science and Technology, Vol-
ume HI, National Acid Precipitation Assess-
ment Program, 722 Jackson Place NW, Wash-
ington, D.C. 20503.
Taylor, C.R. 1990. Supply Control Aspects of the
Conservation Reserve. In: T.L. Napier (Ed)
Implementing the Conservation Tide of the
Food Security Act of 1985. Ankeny, Iowa:
Soil and Water Conservation Society.
Taylor, C.R. 1993a. AGSEVI: An Econometric-Simu-
lation Model of Regional Crop and National
Livestock Production in the United States. In:
C.R. Taylor, K.H. Reichelderfer, and S.R.
Johnson (Eds) Agricultural Sector Models for
the United States: Descriptions and Selected
Policy Applications. Ames Iowa: Iowa State
University Press.
Taylor, C.R. 1993b. Policy Evaluation Exercises with
AGSDV1. In: C.R. Taylor, K.H. Reichelderfer,
and S.R. Johnson (Eds) Agricultural Sector
Models for the United States: Descriptions
and Selected Policy Applications. Ames Iowa:
Iowa State University Press.
Taylor, C.R. 1994. Deterministic vs. Stochastic Evalu-
ation of the Aggregate Effects of Price Sup-
port Programs. Agricultural Systems 44:461-
_ 474.
Taylor, C.R., G.A. Carlson, F.T. Cooke, K.H.
Reichelderfer, and I.R. Starbird. Aggregate
Economic Effects of Alternative Boll Weevil
Management Strategies. Agricultural Econ.
Res. 35:19-19.
Taylor, C.R., R.D. Lacewell, and H. Talpaz. 1979. Use
of Extraneous Information with the Econo-
metric Model to Evaluate Impacts of Pesti-
cide Withdrawals. Western J. of Ag. Econ.
4:1-8.
Taylor, C.R., J.B. Penson Jr., E.G. Smith, and R.D.
Knutson. 1991. Impacts of Chemical Use
Reduction in the South. S.J. Of Ag. Econ.
23:15-23.
1994. Aggregate Economic Effects of CRP Land Re-
turning to Production. J. of Soil and Water
Conservation 49:325-328.
USDA. 1984. Usual Planting and Harvesting Dates
for U.S. Field Crops. Statistical Reporting
Service Agricultural Handbook No. 628.
USDA. 1995. Crops County Data. National Agricul-
tural Statistics Service Dataset (Electronic
File) 93100A and 93100B.
F-10
-------
Appendix G: Lead Benefits Analysis
Introduction
The scientific understanding of the relationship
between lead and human health is rapidly expanding.
This expansion is documented in numerous EPA stud-
ies on the health effects associated with lead expo-
sure. In a pioneering study, Schwartz et al. (U.S. EPA,
1985) quantified a number of health benefits that
would result from reductions in the lead content of
gasoline. The work was extended by EPA's analysis
of lead in drinking water (U.S. EPA, 1986a) and by
an EPA-funded study of alternative lead National
Ambient Air Quality Standards (U.S. EPA, 1987).
Despite this substantial research, much uncertainty
remains. While the health effects of very high levels
of blood lead (PbB) are quite severe (including con-
vulsions, coma and death from lead toxicity) and have
been known for many years, the effects of lower lead
doses continue to be the subject of intensive scien-
tific investigation. Dose-response functions are avail-
able for only a handful of health endpoints associated
with elevated blood lead levels. Other known or
strongly suspected health endpoints cannot be quan-
tified due to a lack of information on the relationship
between dose and effect. Table G-l presents the health
effects that are quantified in this analysis, as well as
important known health effects that are not quanti-
fied.
Some of the health effects that are quantified in
this analysis have not been estimated in previous EPA
analyses. This is largely due to more recent informa-
tion about the dose-response functions that makes it
possible to expand the health effect coverage beyond
what was done previously. Recent information is avail-
able for previously unqualified health effects, and
new information on previously estimated dose-re-
sponse functions is also available.
of Lead" •
- Population Group
JRps^ men in specified age ranges: , , , s - , ,
-Hy^aSeafttoa^ ,'l>* «-;- -' -,,,
Quantified heatf It' ••
'ff0'rnim'in$pecified'ag$ ranges;
MoaTfa&coronaiyfi
KkMwfatal stroke ,„
QMtotifild l)?alth ejects fwwoiptign . 'in'otljerage
'iaages^ , 'T,?'-- , - >\, v"' ''" '\
O&ercaraiovascuiar diseases °",,'^V^^A«.VVV,
.8epn>daedveeffecls ""^~ '-, ,; ,_ - ?-,,'°v""
JSfeBrobehavioral fro'otion
Children ';' " •"'"','
>IQ-Wssefffe^pi5 Ijfetinte ftarnmgs -,
1Q loss effects' on special «dtKatio»alneeds^:
'' "'
'Fe^al'effects from matenial e
'
G-l
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Methods Used to Measure and
Value Health Effects
The following sections present relevant dose-re-
sponse relationships for three population groups: chil-
dren, men, and women. These sections also discuss
data sources used for the dose-response relationships,
although an extensive review of the literature is not
given.1 In addition, each section includes the meth-
ods used to value the changes hi health effects deter-
mined using these dose-response relationships.
Health Benefits to Children
Changes in IQ
Elevated Pb levels may induce a number of ef-
fects on the human nervous system. Generally, these
neurobehavioral effects are more serious for children
than for adults because of children's rapid rate of de-
velopment. It is believed that neurobehavioral defi-
cits in children may result from both pre-natal'and
post-natal exposure. These nervous system effects may
include hyperactivity, behavioral and attentional dif-
ficulties, delayed mental development, and motor and
perceptual skill deficits. Quantification of certain
manifestations of these effects is possible because
sufficient data exist to estimate a dose-response rela-
tionship and IQ loss. The relationship used in the
analysis is discussed below.
Quantifying the Relationship Between Blood
Lead Levels and IQ
A dose-response relationship for IQ decrements
has been estimated by a meta-analysis of seven re-
search studies.2 Regression coefficients for each study
were used to determine a weighted average linear re-
gression coefficient for the relationship between lead
and IQ. Each regression coefficient was weighted by
the inverse of the variance of each estimate. To deter-
mine an overall coefficient, the regression coefficients
for studies that used natural logarithms of lead as the
exposure index were linearized. In general, the coef-
ficient was linearized in the blood lead range of 10 to
20 ug/dL. However, in one study (Bellinger et al.,
1991), 70 percent of the data were below 10 ug/dL;
thus, the Bellinger data were linearized in the 5 to 15
ug/dL range. For the studies that did not transform
lead concentrations, the regression coefficients were
used directly. Given the typical uncertainty within
individual studies, the variation in the regression co-
efficients among studies was not more than would be
expected. The relationship determined by Schwartz
(1993) suggests that for a 1 ug/dL increase in lead, a
decrease of 0.25 IQ points can be expected. The p-
value (< 0.0001) indicates that this relationship is
highly significant.
To obtain the total change in number of IQ points
for a population of children, the 0.25 points lost per
ug/dL change in blood lead is multiplied by the aver-
age blood lead level for that population. The average
blood lead level modeled in this analysis is a geomet-
ric mean, not an arithmetic mean. To adjust for this,
we use a relationship between the expected value and
the geometric mean of a lognormally distributed ran-
dom variable:
where E(X) is the expected value (mean) of the distri-
bution, GM is the geometric mean, and GSD is the
geometric standard deviation. Taking the natural loga-
rithm of Equation 1 and rearranging gives the ratio
between the expected value and the GM:
(2)
(3)
GM
« exp
For a GSD of 1,6 (the assumed GSD of children's
blood lead levels3), the resulting ratio between E(X)
and GM is 1.117. This ratio is used in equation 5.
' For a detailed review of this literature see U.S. Environmental Protection Agency, (1986b) Air Quality Criteria Document for
Lead, and 1989 Addendum. Environmental Criteria and Assessment Office, Office of Research and Development, March.
1 Schwartz, 1993.
3 Suggested value for sub-populations provided by IEUBK guidance manual (U.S.'EPA, 1994).
G-2
-------
Appendix G: Lead Benefits Analysis
The total lost IQ points for each group was estimated
as:
where (Pop)k represents the number of children (up to
age six) around a given industrial source (in the case
of estimating benefits from reduced industrial emis-
sions) or the total U.S. population of children (in the
case of estimating benefits from reductions in gaso-
line lead emissions).
As shown in equation 5, the population of chil-
dren up to age six is divided by seven to avoid double
counting. If we assume that children are evenly dis-
tributed by age, this division applies this equation to
only children age 0-1. If we did not divide, this equa-
tion would count a child who is age zero in the first
year of the analysis and count that same child 6 more
times in successive years. Dividing by seven does cre-
ate some undercounting because in the first year of
the analysis children from age 1 to 6 are not accounted
for, while presumably they are affected by the lead
exposure.
The analysis assumes a permanent loss of IQ based
on blood lead levels estimated for children six years
and younger. Recent studies4 provide concrete evi-
dence of long-term effects from childhood lead expo-
sure.
Valuing Changes in Children's Intelligence
Available economic research provides little em-
pirical data for society's willingness to pay (WTP) to
avoid a decrease in an infant's IQ. Some research,
however, has addressed monetization of a subset of
the effects of decreased IQ. These effects would rep-
resent components of society's WTP to avoid IQ de-
creases. Employed alone, these monetized effects
should underestimate society's WTP. Nevertheless,
for the purpose of this analysis, these effects are used
to approximate the WTP to avoid IQ decrements.
IQ deficits incurred through lead exposure are
assumed to persist throughout the exposed infant's
lifetime. Two consequences of this IQ decrement are
then considered: the decreased present value of ex-
pected lifetime earnings for the infant, and the in-
creased educational resources expended for a infant
who becomes mentally handicapped or is in need of
compensatory education as a consequence of lead
exposure. The value of foregone earnings is addressed
in this section.
The reduction in IQ has a direct and indirect ef-
fect on earnings. The direct effect is straightforward:
lower IQs decrease job attainment and performance.
Reduced IQ also results in reduced educational attain-
ment, which, in turn, affects earnings and labor force
participation. These effects on earnings are additive
since the studies used for this analysis have controlled
for these effects separately.5 If personal decisions
about the total amount of education and labor force
participation were based entirely on each individual
maximizing the expected present value of lifetime
income, the magnitude of the indirect effect on in-
come of a small change in educational attainment
would be close to zero,6 and certainly less than the
magnitude of the direct effect. However, individuals
make educational decisions based on a number of
considerations in addition to the effect on the present
value of lifetime earnings, such as satisfaction (util-
ity) derived from formal education, non-compensa-
tion aspects of alternative career opportunities, the
ability to pay educational costs, etc. Such consider-
ations could lead to either a positive or negative mar-
ginal return to education. Studies7 of educational at-
tainment and lifetime earnings have generally identi-
fied a positive marginal return to education, suggest-
ing that the educational attainment decision may not
be based simply on expected earnings.
This analysis uses two sets of estimates of the ef-
fects of IQ on earnings. The first estimate, used by
Abt Associates in a previous analysis, is based on sev-
eral older studies. The second estimate is based on
Salkever (1995).
Older Estimate of the Effect oflQ on Earnings:
The Direct Effect oflQ on Wage Rate
Henry Aaron, Zvi Griliches, and Paul Taubman
have reviewed the literature examining the relation-
4 For example, Bellinger (1992).
5 IQ is also correlated with other socio-economic factors which have not been quantified in this analysis.
6 This is a straightforward result of the "envelope theorem" in economics. In this context, the envelope theorem shows that if
individuals select the level of education that maximizes expected income, then the marginal benefit of additional education (i.e., the
partial derivative of income with respect to education) will be zero at that optimal education level.
7 Including Chamberlain and Griliches (1977), Ashenfelter and Ham (1979), and Salkever (1995)
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
ship between IQ and lifetime earnings.8 They found
that the direct effect, (schooling held constant) of IQ
on wage rates ranged from 0.2 percent to 0.75 percent
per IQ point. Perhaps the best of these studies is
Griliches (1977).9 He reported the direct effect of IQ
on wage rates to be slightly more than 0.5 percent per
IQ point. Because this is roughly the median estimate
of the U.S. EPA review of the literature, this estimate
is used.
Older Estimate of the Effect oflQ on Earnings:
The Indirect Effect oflQ on Earnings
From Needleman et al. (1990) it is possible to
estimate the change in years of schooling attained per
one IQ point change. The study's regression coeffi-
cients for the effect of tooth lead on achieved grade
provide an estimate of current grade achieved. How-
ever, many of these children were in college at the
time and are expected to achieve a higher grade level.
Following Schwartz (1990), after adjusting the pub-
lished results for the fact that a higher percentage of
children with low tooth lead were attending college, a
0.59 year difference in expected maximum grade
achieved between the high and low exposure groups
was estimated. It is assumed that educational attain-
ment relates with blood lead levels in proportion to
IQ. The difference in IQ score between the high and
low exposure group was 4.5 points (from Needleman
et al. (1990)). Dividing 0.59/4.5 = 0.131 suggests that
the increase in lead exposure which reduces IQ by
one point may also reduce years of schooling by 0.131
years.
Studies that estimate the relationship between
educational attainment and wage rates (while control-
ling for IQ and other factors) are less common. Cham-
berlain and Griliches (1977) estimate that a one year
increase in schooling would increase wages by 6.4
percent. In a longitudinal study of 799 subjects over 8
years, Ashenfelter and Ham (1979) reported that an
extra year of education increased the average wage
rate over the period by 8.8 percent. We use the aver-
age of these two estimates (7.6 percent) to calculate
the indirect effect of increased schooling on the present
value of lifetime income. Increased wages per IQ point
are calculated using: (7.6 percent wage increase/school
year) x (0.131 school years/IQ) = 1.0 percent increase
in earnings per IQ point.
There is one final indirect effect on earnings.
Changes in IQ affect labor force participation. Fail-
ure to graduate high school, for example, correlates
with participation in the labor force, principally
through higher unemployment rates and earlier retire-
ment ages. Lead is also a strong correlate with atten-
tion span deficits, which likely reduce labor force par-
ticipation. The results of Needleman et al. (1990) re-
lating lead to failure to graduate high school can be
used to estimate changes in earnings due to labor force
participation. Using the odds ratio from Needleman
et al., it was estimated that a one IQ point deficit would
also result in a 4.5 percent increase in the risk of fail-
ing to graduate. Krupnick and Cropper (1989) pro-
vide estimates of labor force participation between
high school graduates and non-graduates, controlling
for age, marital status, children, race, region, and other
socioeconomic status factors. Based on their data,
average participation in the labor force is reduced by
10.6 percent for persons failing to graduate from high
school. Because labor force participation is only one
component of lifetime earnings (i.e., earnings = wage
rate X years of work), this indirect effect of schooling
is additive to the effect on wage rates. Combining this
estimate with the Needleman result of 4.5 percent in-
crease in the risk of failing to graduate high school
per IQ point, indicates that the mean impact of one IQ
point loss is a (10.6 percent x 4.5 percent) = 0.477
percent decrease in expected earnings from reduced
labor force participation.
Combining the direct effect of 0.5 percent with
the two indirect effects (1.0 percent for less schooling
and 0.477 percent for reduced labor force participa-
tion) yields a total of 1.97 percent decrease in earn-
ings for every loss of one IQ point.
Newer Estimate of the Effect of IQ on Earnings:
Salkever (1995)
One of the most recent studies of the effects of IQ
on earnings is Salkever (1995). Such an analysis with
more recent data is valuable because the labor market
has undergone many changes over the quarter cen-
tury hi which earlier studies have appeared. Like the
analysis of the effect of IQ on earnings presented
above, Salkever (1995) estimates this as the sum of
direct and indirect effects. The direct effect is the sum
of effects of IQ test scores on employment and earn-
8 U.S. EPA, 1984.
9 Griliches used a structural equations model to estimate the impact of multiple variables on an outcome of interest. This method
has conceptual advantages over other empirical estimates used in the literature because it successfully controls for the many con-
founding variables that can affect earnings.
G-4
-------
Appendix G: Lead Benefits Analysis
ings for employed persons, holding years of school-
ing constant. The indirect effect works through the
effect of IQ test scores on years of schooling attained,
and the subsequent effect of years of schooling on the
probability of employment, and on earnings for em-
ployed persons.
Salkever (1995) provides updated estimates all of
the necessary parameters using the most recent avail-
able data set, the National Longitudinal Survey of
Youth (NLSY). Three regression equations provide
these parameters. The years of schooling regression
shows the association between IQ scores and highest
grade achieved, holding background variables con-
stant. The employment regression shows the associa-
tion between IQ test scores, highest grade, and back-
ground variables on the probability of receiving earned
income. This regression thus provides an estimate of
the effect of IQ score on employment, holding school-
ing constant, and the effect of years of schooling on
employment, holding IQ constant. The earnings re-
gression shows the association between IQ test scores,
highest grade, and background variables on earnings,
for those with earned income.
These regressions provide parameters needed to
estimate the total effect of a loss of an IQ point on
earnings. The direct effects of IQ on employment and
earnings for employed persons, holding schooling
constant, come from the employment and earnings
regressions. The indirect effect of IQ on employment
through schooling is the product of the effect of IQ on
years of schooling, from the years of schooling re-
gression, and the effect of highest grade on employ-
ment, from the employment regression. The indirect
effects of IQ on earnings for employed persons through
schooling is the product of the effect of IQ on years
of schooling, from the years of schooling regression,
and the effect of highest grade on earnings for em-
ployed persons, from the earnings regression.
The total estimated effect of a loss of an IQ point
on earnings is larger than the previous estimate of 1.97
percent. Based on the Salkever study, the most recent
estimate of the effect of an IQ point loss is now a
reduction in earnings of 1.93 percent for men and 3.22
percent for women, which is a participation-weighted
average of 2.39 percent.
Value of Foregone Earnings
In the next step to monetize intelligence effects,
the percent earnings loss estimate must be combined
with an estimate of the present value of expected life-
10 U.S. Department of Commerce, 1993
time earnings. Data on earnings for employed per-
sons and employment rates as a function of educa-
tional attainment, age, and gender were reported for
the U.S. population in 1992 by the Bureau of the Cen-
sus.10 Assuming this distribution of earnings for em-
ployed persons and labor force participation rates re-
mains constant over time, and further assuming a trend
rate of real wage growth (productivity effect), an an-
nual discount factor, and year-to-year survival prob-
abilities, the current Census data on earnings can be
used to calculate the mean present value of lifetime
earnings of a person born today. This analysis assumed
a person received earned income from age 18 to age
64, and assumed a real wage growth rate of one per-
cent and an annual discount rate of five percent. Men
tend to earn more than women because of higher wage
rates and higher labor force participation. However,
for both men and women, expected lifetime earnings
increase greatly with education.
While the Census data are most likely the best
available basis for projecting lifetime earnings, a num-
ber of uncertainties deserve mention. Labor force par-
ticipation rates of women, the elderly, and other groups
will most likely continue to change over the next de-
cades. Real earnings of women will probably continue
to rise relative to real earnings of men. Unpredictable
fluctuations in the economy's growth rate will prob-
ably affect labor force participation rates and real wage
growth of all groups. Medical advances will probably
raise survival probabilities.
One problem that was addressed was the fact that
the current educational distribution for older persons
today is an especially poor predictor of educational
attainment for those born today, since educational at-
tainment has risen over time. In fact, if one simply
projected educational attainment for a person born
today using this method, this person would lose years
of schooling with age (starting between ages 40 and
50), since average years of schooling declines with
age in a one-time snapshot of the current population.
To address this issue, the analysis assumes education
levels cannot fall as a person ages.
Note that use of earnings is an incomplete mea-
sure of an individual's value to society. Those indi-
viduals who choose not to participate in the labor force
for all of their working years must be accounted for,
since the lost value of their productive services may
not be accurately measured by wage rates. The larg-
est group are those who remain at home doing house-
work and child rearing. Also, volunteer work contrib-
G-5
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
utes significantly to social welfare and rates of
volunteerism tend to increase with educational attain-
ment and income.11 If the opportunity cost of non-
wage compensated work is assumed to be the average
wage earned by persons of the same sex, age, and edu-
cation, the average lifetime earnings estimates would
be significantly higher and could be approximated by
recalculating the tables using full employment rates
for all age/sex/education groups. To be conservative,
only the value of lost wages is considered in this analy-
sis.
The adjusted value of expected lifetime earnings
obtained above is a present value for an individual
entering the labor force at age 18 and working until
age 64. Given a five percent discount rate, the other
assumptions mentioned, and current survival prob-
abilities,12 the present value of lifetime earnings of a
person bom today would be $170,169.
Costs of Additional Education
The increase in lifetime earnings from additional
education is the gross return to education. The gross
return to education, however, does not reflect the cost
of the additional education. The cost of the marginal
education must be subtracted from the gross return in
order to obtain the net increase per IQ point from ad-
ditional education. There are two components of the
cost of marginal education; the dkect cost of the edu-
cation, and the opportunity cost of lost income during
the education. An estimate of the educational cost
component is obtained from the U.S. Department of
Education.13 The marginal cost of education used in
this analysis is assumed to be $5,500 per year. This
figure is derived from the Department of Education's
reported ($5,532) average per-student annual expen-
diture (current plus capital expenditures) in public
primary and secondary schools hi 1989-'90. For com-
parison, the reported annual cost of college education
(tuition, room and board) in 4 year public institutions
is $4,975, and $12,284 for private institutions.
The estimated cost of an additional 0.131 years
of education per IQ point (from the older estimate of
IQ effects) is $721 (i.e., 0.131 x $5,500). Because this
marginal cost occurs at the end of formal education,
it must be discounted to the time the exposure and
damage is modeled to occur (age zero). The average
level of educational attainment in the population over
age 25 is 12.9 years.14 Therefore, the marginal educa-
tional cost is assumed to occur at age 19, resulting in
a discounted present value cost of $285.
The other component of the marginal cost of edu-
cation is the opportunity cost of lost income while in
school. Income loss is frequently cited as a major fac-
tor in the decision to terminate education, and must
be subtracted from the gross returns to education. An
estimate of the loss of income is derived assuming
that people in school are employed part time, but
people out of school are employed full time. The op-
portunity cost of lost income is the difference between
full-time and part-time earnings. The median annual
income of people ages 18-24 employed full-time is
$16,501, and $5,576 for part-time employment.15 The
lost income associated with being in school an addi-
tional 0.131 years is $1,431, which has a present dis-
counted value at age zero of $566.
Salkever found a smaller effect of IQ on educa-
tional attainment (0.1007 years per IQ point, versus
0.131 years), which results in smaller estimated costs.
Using the same method and data described above, the
estimated present value of educational cost per IQ
point is $219, and the income opportunity cost is $435.
Final Estimate of the Effect oflQ on Earnings.
Combining the value of lifetime earnings with the
two estimates of percent wage loss per IQ point yields
a low estimate of $170,169 x 1.97 percent = $3,000
per lost IQ point, and a higher estimate of $4,064 based
on Salkever (1995). Subtracting the education and
opportunity costs reduces these values to $2,505 and
$3,410 per IQ point, respectively. This analysis uses
the midpoint of these two estimates, which is $2,957.
Of course, changing the discount rate would change
this estimate. With an assumed discount rate of seven
percent, the final estimate is only $1,311. With an
assumed discount rate of three percent, the final esti-
mate rises to $6,879.
" U.S. Department of Commerce, 1986. Table No. 651, p. 383.
12 Special education costs for children who do not survive to age 18 are not counted, which results in some underestimation of
benefits. However, most child mortality occurs before the age of 7, when the special education begins, so this under-counting is not
substantive.
13 "Digest of Education Statistics". U.S. Dept. of Education, 1993.
14 "Digest of Education Statistics". U.S. Dept. of Education, 1993.
is "Money Income of Households, Families, and Persons in the United States: 1992". U.S. Department of Commerce, 1993.
~ CMS
-------
Appendix G: Lead Benefits Analysis
Children with IQs Less Than 70
Quantifying the Number of Children with IQs
Less than 70
In addition to the total IQ point decrements that
can be predicted to occur in a population of children
having a specified blood lead distribution, increases
are also expected to occur in the incidence of children
having very low IQ scores as the mean blood lead
level for that population increases. IQ scores are
normalized to have a mean of 100 and a standard de-
viation of fifteen. An IQ score of 70, which is two
standard deviations below the mean, is generally re-
garded as the point below which children require spe-
cial compensatory education tailored to the mentally
handicapped.
The relationship presented here for estimating
changes in the incidence of IQ < 70 was developed to
make use of the most current IQ point decrement func-
tion provided by Schwartz (1993). It is assumed that
for a baseline set of conditions where a population of
children has a blood lead distribution defined by some
geometric mean and geometric standard deviation, that
population also has a normalized IQ point distribu-
tion with a mean of 100 and a standard deviation of
15. For this baseline condition, the proportion of the
population expected to have IQ < 70 is determined
from the standard normal distribution function:
where:
P(IQ<70) = Probability of IQ scores less than
70
z = standard normal variate; com-
puted for an IQ score of 70, with
mean IQ score of 100 and stan-
dard deviation of 15 as:
-
,V-^^(7)
'-
= Standard normal distribution
function:
The integral in the standard normal distribution
function does not have a closed form solution. There-
fore, values for O(z) are usually obtained readily from
software with basic statistical functions or from tables
typically provided in statistics texts. The solution for
p(z) where z = -2 is 0.02275. That is, for the normal-
ized IQ score distribution with mean of 100 and stan-
dard deviation of 15, it is expected that approximately
2.3 percent of children will have IQ scores below 70.
To estimate changes in the proportion of children
with IQ scores below 70 associated with changes in
mean blood lead levels for a population of children,
the following two key assumptions are made:
1. The mean IQ score will change as a result of
changesui the mean blood lead level as:
A/Q = -0.25 xAPbB
where
AIQandAPbB
are the changes in the mean IQ score and in
the mean blood lead levels, respectively, be-
tween the no-control and control scenarios.
This relationship relies on Schwartz' estimate
(1993) of a decrease of 0.25 IQ points for each
ug/dL increase in blood lead. Note that the
mean blood lead level referred to here is the
arithmetic mean (or expected value) for the
distribution obtained as described previously
from the GM and GSD.
2. The standard deviation for the IQ distribution
remains at 15.
Using these assumptions, the change in the pro-
portion of children having IQ <70 can then be deter-
mined for a given change in mean blood lead from:
where,
O^qOQ * 0.25-
'
For a given change in PbB between the control
and no-control scenarios a response in terms of IQ is
calculated. The procedure above yields an estimate
of the percent of the population with IQs less than 70.
This percentile is multiplied by the exposed popula-
tion of children to estimate the increased incidence of
G-7
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
children with low IQs. As in the IQ point loss equa-
tion, the results of this function are applied to chil-
dren age 0-6 and divided by seven to avoid double
counting. (See discussion under equation 5).
This procedure quantifies only the change in the
number of children who pass below the IQ=70 thresh-
old. Any other changes in children's IQ are quanti-
fied using the IQ point loss function described previ-
ously. Treating these two endpoints additively does
not result in double counting, because the value asso-
ciated with the IQ point loss function is the change in
worker productivity while the value associated with
IQs less than 70 is the increased educational costs for
the individual, as discussed below.
Valuing the Reduction in Number of Children
with IQs less than 70
To value the reduction in the number of children
with IQs less than 70, the reduction in education costs
were measured - a clear underestimate of the total
benefits.16 Kakalik et al. (1981), using data from a
study prepared for the Department of Education's
Office of Special Education Programs, estimated that
part-time special education costs for children who re-
mained in regular classrooms cost $3,064 extra per
child per year in 1978. Adjusting for inflation and real
income growth using the GNP price deflator yields
an estimate of $6,318 per child in 1990 dollars. For
the calculations, this incremental estimate of the cost
of part-time special education was used to estimate
the cost per year per child needing special education
as a result of impacts of lead on mental development.
Costs would be incurred from grades one through
twelve. Discounting future expenses at a rate of three
percent yields an expected present value cost of ap-
proximately $52,700 per infant (assuming compen-
satory education begins at age 7 and continues through
age 18). Note that this underestimates the cost, since
Kakalik et al. measured the increased cost to educate
children attending regular school—not a special edu-
cation program.
Changes in Neonatal Mortality
Quantifying the relationship between PbB
levels and neonatal mortality
U.S. EPA (1990c) cites a number of studies link-
ing fetal exposure to lead (via in utero exposure from
maternal intake of lead) to several adverse health ef-
fects. These effects include decreased gestational age,
reduced birth weight, late fetal death, and increases
in infant mortality.17 The Centers for Disease Control
(CDC, 1991a) presents a method to estimate changes
in infant mortality due to changes in maternal blood
lead levels during pregnancy.18 The analysis links two
relationships. The first relationship, between mater-
nal blood lead level and gestational age of the new-
born, was estimated by Dietrich et al. (1987). CDC
then estimated infant mortality as a function of gesta-
tional age, using data from the Linked Birth and In-
fant Death Record Project from the National Center
for Health Statistics. The resulting association is a de-
creased risk of infant mortality of Ifr4 (or 0.0001) for
each 1 ug/dL decrease in maternal blood lead level
during pregnancy. This is the relationship used in the
current analysis.
Valuing changes in neonatal mortality
The central estimate of the monetary benefit as-
sociated with reducing risks of neonatal mortality is
$4.8 million per avoided mortality. This analysis at-
tempts to capture the credible range of uncertainty
associated with this estimate by describing the mon-
etary benefit as a distribution of values: a Weibull
distribution with a mean value of $4.8 million and a
standard deviation of $3.24 million. Appendix I docu-
ments the derivation of this distribution and the sources
of uncertainty in valuing reduced mortality risks.
Health Benefits to Men
In addition to adversely affecting children's
health, lead exposure has also been shown to adversely
affect adults. The health effects in adults that are quan-
tified and included in the benefits analysis are all re-
16 The largest part of this benefit is the parents' willingness to pay to avoid having their child become mentally handicapped,
above and beyond the increased educational costs.
17 Due to unavailability of suitable data, non-fatal health impacts due to decreased gestational age or reduced birth weight have
not been included in this analysis. For example, the benefits from avoided developmental disabilities such as sensory and motor
dysfunction associated with decreased gestational age have not been included.
18 The estimated change in infant mortality due to change in birth weight was not modeled because the data relating prenatal lead
exposure to birth weight are not as strong as data relating lead exposure and gestational age.
_ —
-------
Appendix G: Lead Benefits Analysis
lated to the effects of lead on blood pressure.19 The
estimated relationships between these health effects
and lead exposure differ between men and women.
The quantified health effects include increased inci-
dence of hypertension (estimated for males only), ini-
tial coronary heart disease (CHD), strokes (initial cere-
brovascular accidents and atherothrombotic brain
infarctions), and premature mortality. Other health
effects associated with elevated blood pressure, and
other adult health effects of lead including
neurobehavioral effects, are not included in this analy-
sis. This section describes the quantified health ef-
fects for men; the next section describes the health
effects for women.
Hypertension
Quantifying the relationship between PbB
levels and hypertension
Elevated blood lead has been linked to elevated
blood pressure (BP) in adult males, especially men
aged 40-59 years.20 Further studies have demonstrated
a dose-response relationship for hypertension (defined
as diastolic blood pressure above 90 mm Hg for this
model) in males aged 20-74 years.21 This relation-
ship is:
1
4, 02744- T93*(lnP6B$'
'a
744 -,793*0 n^feBJ}'
a ix
where:
APr(HYP)
PbB
the change in the probability of
hypertension;
blood lead level in the control
scenario; and
blood lead level in the no-control
scenario.
Valuing reductions in hypertension
The best measure of the social costs of hyperten-
sion, society's willingness to pay to avoid the condi-
tion, cannot be quantified without basic research well
beyond the scope of this project. Ideally, the measure
would include all the medical costs associated with
treating hypertension, the individual's willingness to
pay to avoid the worry that hypertension could lead
to a stroke or CHD, and the individual's willingness
to pay to avoid changes in behavior that may be re-
quired to reduce the probability that hypertension leads
to a stroke or CHD. Medical costs of hypertension
can be divided into four categories: physician charges,
medication costs, hospitalization costs and lost work
tune.
This analysis uses recent research results to quan-
tify two components of this benefit category. Krupnick
and Cropper (1989), using data from the National
Medical Care Expenditure Survey, have estimated the
medical costs of hypertension. These costs include
physician care, drugs and hospitalization costs. In
addition, hypertensives have more bed disability days
and work loss days than others of their age and sex.
Krupnick and Cropper estimated the increase in work
loss days at 0.8 per year, and these were valued at the
mean daily wage rate. Adjusting the above costs to
1990 dollars gives an estimate of the annual cost of
each case of hypertension of $681. The estimate is
likely to be an underestimate of the true social benefit
of avoiding a case of hypertension for several rea-
sons. First, a measure of the value of pain, suffering
and stress associated with hypertension is not included.
Second, the direct costs (out-of-pocket expenses) of
diet and behavior modification (e.g., salt-free diets,
etc.) are not valued. These costs are likely to be sig-
nificant, since modifications are typically severe.
Third, the loss of satisfaction associated with the diet
and behavior modifications are ignored. Finally, the
medication for hypertension may produce side effects
including drowsiness, nausea, vomiting, anemia, im-
potence, cancer, and depression. The benefits of avoid-
ing these side effects are not included in this estimate.
Quantifying the relationship between blood lead
and blood pressure
Because blood lead has been identified as a risk
factor in a number of cardiovascular illnesses,22 it is
useful to quantify the effect of changes in blood lead
levels on changes in blood pressure for reasons other
than predicting the probability of hypertension. Based
on results of a meta-analysis of several studies,
Schwartz (1992a) estimated a relationship between a
' .U'S- EPA (1990c) ^ presents evidence of the genotoxicity and/or carcinogenicity of
toxicolog,cal evidence suggests that human cancer effects are possible, dose-response relation-
20 Pirkle et al., 1985.
21 Schwartz, 1988.
22 Shurtleff, 1974; McGee and Gordon, 1976; Pooling Project Research Group, 1978.
G-9
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
change in blood pressure associated with a decrease
hi blood lead from 10 ug/dL to 5 ng/dL.23 The coeffi-
cient reported by Schwartz leads to the following func-
tion relating blood pressure to blood lead for men:
where:
PbB,
PbB,
the change in men's diastolic blood
pressure expected from a change in
PbB;
blood lead level in the control sce-
nario (in ug/dL); and
blood lead level in the no-control
scenario (in ug/dL).
This blood lead to blood pressure relationship is
used to estimate the incidence of initial coronary heart
disease, strokes (atherothrombotic brain infarctions
and initial cerebrovascular accidents) and premature
mortality in men.
Changes In Coronary Heart Disease
Quantifying the relationship between blood
pressure and coronary heart disease
Estimated blood pressure changes can be used to
predict the increased probability of the initial occur-
rence of CHD and stroke.24 Increased blood pressure
would also increase the probability of reoccurrence
of CHD and stroke, but these quantified relationships
are not available. First-time coronary heart disease
events in men can be predicted using an equation with
different coefficients for each of three age groups. For
men between 40 and 59 years old, information from a
1978 study by the Pooling Project Research Group
(PPRG) is used. PPRG (1978) presents a multivariate
model (controlling for smoking and serum cholesterol)
that relates the probability of coronary heart disease
(CHD) to blood pressure. The model used data from
five different epidemiological studies. From this study,
the equation for the change hi 10-year probability of
occurrence of CHD is:
where:
APr(CHD40.59)
DBP,
DBF,
= change in 10-year probabil-
ity of occurrence of CHD event
for men between 40-59 years old,
mean diastolic blood pressure in
the control scenario; and
mean diastolic blood pressure in
the no-control scenario.
The relationship between BP and first-time CHD
in older men was determined from information pre-
sented hi Shurtleff (1974). This study also uses data
from the Framingham Study (McGee and Gordon,
1976) to estimate univariate relationships between BP
and a variety of health effects by sex and for each of
the following age ranges: 45-54, 55-64, and 65-74
years. Single composite analyses for ages 45-74 were
also performed for each sex. For every equation, t-
statistics on the variable blood pressure are signifi-
cant at the 99th percent confidence interval. For men
aged 60 to 64 years old, first-time CHD can be pre-
dicted from the following equation:
a*>
where:
DBP
DBP,
= change in 2 year probability
of occurrence of CHD event for
men from 60 to 64 years old;
mean diastolic blood pressure in
the control scenario; and
mean diastolic blood pressure in
the no-control scenario.
For men aged 65 to 74 years old, the following
equation uses data from Shurtleff (1974) to predict
the probability of first-time CHD:
where:
APr(CHD65.74)
DBP,
DBP,
= change in 2 year probability
of occurrence of CHD event for
men from 65 to 74 years old;
mean diastolic blood pressure in
the control scenario; and
mean diastolic blood pressure in
the no-control scenario.
25 Schwartz, 1992a.
M U.S. EPA, 1987.
G-10
-------
Appendix G: Lead Benefits Analysis
The probability changes calculated using the func-
tions above are used to estimate the number of CHD
events avoided in a given year due to air quality im-
provements attributable to the Clean Air Act. The re-
sulting CHD incidence estimates include both fatal
and non-fatal events. However, because mortality
benefits are independently estimated in this analysis,
it is important to capture only the non-fatal CHD
events. Shurtleff (1974) reported that two-thirds of
all CHD events were non-fatal. This factor was there-
fore applied to the estimate of avoided CHD events
for each age category.
Valuing reductions in CHD events
General methodology
Because of the lack of information on WTP to
avoid an initial CHD event, WTP was estimated by
estimating the associated cost of illness (COI). This
will underestimate WTP, as explained in Appendix I.
Full COI consists of the present discounted value of
all costs associated with the event, including both di-
rect and indirect costs incurred during the hospital stay,
as well as the present discounted values of the streams
of medical expenditures (direct costs) and lost earn-
ings (indirect costs) incurred once the individual leaves
the hospital.
Wittels et al. (1990) estimate the total medical
costs within 5 years of diagnosis of each of several
types of CHD events (including acute myocardial in-
farction, angina pectoris, unstable angina pectoris,
sudden death and nonsudden death) examined in the
Framingham Study. Costs were estimated by multi-
plying the probability of a medical test or treatment
within five years of the initial CHD event (and asso-
ciated with that event) by the estimated price of the
test or treatment. All prices were in 1986 dollars. (It
does not appear that any discounting was used.) The
probabilities of tests or treatments were based on
events examined in the Framingham Study. The au-
thors estimate a total expected cost over a five year
period (in 1986 dollars) of $51,211 for acute myocar-
dial infarction, $24,980 for angina pectoris, and
$40,581 for unstable angina pectoris. Converted to
1990 dollars (using the consumer price index for medi-
cal care, U.S. Bureau of the Census, 1992), this is
$68,337 for acute myocardial infarction, $33,334 for
angina pectoris, and $54,152 for unstable angina pec-
toris. (The figures for sudden death and nonsudden
death are not included because the CHD events in this
analysis exclude those that resulted in death, to avoid
double counting.)
Cropper and Krupnick (1990) suggest, in an un-
published study, that CHD-related lost earnings could
be a significant component of total COI, although the
value of earnings lost may vary substantially with the
age of onset of CHD. They estimate, for example, that
an individual whose first heart attack occurs between
ages 55 and 65 will have an expected annual earnings
loss of $12,388 (hi 1990 dollars), and a present dis-
counted value of lost earnings over a five-year period
of about $53,600, using a five percent discount rate.
This is almost as much as the total medical costs over
5 years estimated by Wittels et al. (1990) for unstable
angina pectoris, and substantially more than the cor-
responding estimate of medical costs for angina pec-
toris. For an individual whose first heart attack oc-
curs between ages 45 and 54, on the other hand, Crop-
per and Krupnick estimate annual average lost earn-
ings of $2,143 (in 1990 dollars), and a present dis-
counted value of lost earnings over a five-year period
of about $9,300, again using a five percent discount
rate. Cropper and Krupnick do not estimate medical
costs for exactly the same disease categories as Wittels
et al., but their research suggests that whether the five-
year COI of a CHD event, including both medical costs
and lost earnings, is lower or higher than the average
of the three estimates reported by Wittels et al. de-
pends oh an individual's age at the onset of CHD.
Combining Cropper and Krupnick's five-year lost
earnings estimates with their estimates for average
annual medical expenditures for ischemic heart dis-
ease summed over five years, for example, yields a
total COI of about $47,000 for a 50 year old and
$72,000 for a 60 year old, compared to the $52,000
average of the three estimates reported by Wittels et
al.
In addition to the variability in estimates of medi-
cal costs and lost earnings arising from CHD, there is
uncertainty regarding the proportion of pollution-re-
lated CHD events associated with the various classes
of CHD. To characterize this uncertainty it was as-
sumed, in the absence of further information, that all
pollution-related CHD events are either acute myo-
cardial infarctions, angina pectoris, or unstable an-
gina pectoris. A distribution of estimates of COI for
pollution-related CHD was generated by Monte Carlo
methods. On each iteration, a value was randomly
selected from each of three continuous uniform dis-
tributions. Each value selected was normalized by
G-ll
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
dividing by the sum of the three values, so that the
three normalized values summed to 1.0. The result-
ing triplet of proportions represents a possible set of
proportions of pollution-related CHD events that are
acute myocardial infarction, angina pectoris, and un-
stable angina pectoris. The corresponding dollar value
for the iteration is a weighted average of the estimated
dollar values for the three types of CHD event (from
Wittels et al.), where the weights are the three ran-
domly selected proportions. The central tendency es-
timate of the COI associated with a case of pollution-
related CHD is the mean of this distribution, about
$52,000.
This estimate is likely to understate full COI be-
cause it does not include lost earnings. It is likely to
underestimate total WTP to an even greater extent
because it does not include WTP to avoid the pain
and suffering associated with the CHD event. It is,
however, substantially greater than an estimate based
only on the direct and indirect costs incurred during
the hospital stay.
The valuation for CHD is additive with the valu-
ation for hypertension despite the fact that the condi-
tions often occur together, because the two values rep-
resent different costs associated with the conditions.
The valuation for hypertension is based on loss of work
days as a result of hypertension and some of the medi-
cal costs associated with treating hypertension. The
valuation for CHD is based on the willingness to pay
to avoid the pain and suffering of the CHD itself.
Therefore, these two valuations can be separated and
added together.
Changes in Initial Cerebrovascular Accidents and
Initial Atherothrombotic Brain Infarctions
Quantifying the relationship between blood
pressure and first-time stroke
Two types of health events are categorized as
strokes: initial cerebrovascular accidents (CA) and
initial atherothrombotic brain infarctions (BI). The risk
has been quantified for the male population between
45 and 74 years old.25 For initial cerebrovascular ac-
cidents, the logistic equation is:
where:
APr(CAmen) =
DBPt =
DBP =
change in 2 year probability of
cerebrovascular accident in men;
mean diastolic blood pressure in
the control scenario; and
mean diastolic blood pressure in
the no-control scenario.
For initial atherothrombotic brain infarctions, the
logistic equation is:
where:
DBP
DBP,
change in 2 year probability of
brain infarction in men;
mean diastolic blood pressure in
the control scenario; and
mean diastolic blood pressure in
the no-control scenario.
APrfCA J=
1
1
Similar to CHD events, this analysis estimates
only non-fatal strokes (to avoid double-counting with
premature mortality). Shurtleff (1974) reported that
70 percent of strokes were non-fatal. This factor was
applied to the estimates of both CA and BI.
Valuing reductions in strokes
Taylor et al. (1996) estimate the lifetime cost of
stroke, including the present discounted value (in 1990
dollars) of the stream of medical expenditures and the
present discounted value of the stream of lost earn-
ings, using a five percent discount rate. Estimates are
given for each of three separate categories of stroke,
separately for males and females at ages 25, 45, 65,
and 85. For all three types of stroke, the indirect costs
(lost earnings) substantially exceed the direct costs at
the two younger ages, and are about the same as or
smaller than direct costs at the older ages. .
Both types of stroke considered in this analysis
fall within the third category, ischemic stroke, con-
sidered by Taylor et al. To derive a dollar value of
avoiding an initial ischemic stroke for males, a dollar
value for avoiding ischemic stroke among males age
55 was interpolated from the values for males ages 45
and 65; similarly, a dollar value for avoiding ischemic
stroke among males age 75 was interpolated from the
values for males ages 65 and 85. Of males in the United
Shurtleff, 1974.
G-12
-------
Appendix G: Lead Benefits Analysis
States between the ages of 45 and 74 (the age group
for which lead-related stroke is predicted), 41.2 per-
cent are ages 45-54 and the remaining 58.8 percent
are ages 55-74. The value of an avoided stroke among
males was calculated as the weighted average of the
values for males in age group 45-54 and males in age
group 55-74 , where the weights are the above per-
cents. The value for age group 45-54 is the average of
the values for ages 45 and 55; the value for age group
55-74 is the average of the values for ages 55, 65 and
75. The resulting average value of an avoided stroke
among males aged 45-74 is about $200,000.
Changes in Premature Mortality
Quantifying the relationship between blood
pressure and premature mortality
Information also exists to predict the increased
probability of premature death from all causes as a
function of elevated blood pressure. U.S. EPA (1987)
used population mean values for serum cholesterol
and smoking to reduce results from a 12 year follow-
up of men aged 40-54 in the Framingham Study
(McGee and Gordon, 1976) to an equation in one ex-
planatory variable:
* ~
.. .3L
where:
APr(MORT40 54) = the change in 12 year prob-
ability of death for men aged 40-
54;
= mean diastolic blood pressure in
the control scenario; and
= mean diastolic blood pressure hi
the no-control scenario.
DBP
Information from Shurtleff (1974) can be used to
estimate .the probability of premature death in men
older than 54 years old. This study has a 2 year follow
up period, so a 2 year probability is estimated. For
men aged 55 to 64 years old, mortality can be pre-
dicted by the following equation:
where:
APr(MORT55 M)= the change in 2 year prob-
ability of death in men aged 55-
64;
= mean diastolic blood pressure in
the control scenario; and
= mean diastolic blood pressure in
the no-control scenario.
DBP
For men aged 65 to 74 years old, premature mor-
tality can be predicted by the following equation:
-ti?f(MORT65£fc
V "'-'I
-J -4. g3.05
where:
APr(MORT65 ?4) = the change in 2 year prob-
ability of death in men aged 55-
64;
DBPt = mean diastolic blood pressure in
the control scenario; and
DBP2 = mean diastolic blood pressure in
the no-control scenario.
Valuing reductions in premature mortality
As discussed above, premature mortality is val-
ued at $4.8 million per case (discussed further in Ap-
pendix I). Because this valuation is based on the will-
ingness to pay to the risk of death, and the CHD valu-
ation is based on the willingness to pay to avoid the
pain and suffering of a CHD event (defined as a CHD
event that does not end in death, to avoid double count-
ing), these two endpoints are additive as well.
Health Benefits to Women
Available evidence suggests the possibility of
health benefits from reducing women's exposure to
lead. Recent expanded analysis of data from the sec-
ond National Health and Nutrition Examination Sur-
vey26 (NHANES H) by Schwartz (1990) indicates a
significant association between blood pressure and
blood lead in women. Another study, by Rabinowitz
et al. (1987), found a small but demonstrable associa-
tion between maternal blood lead and pregnancy hy-
pertension and blood pressure at time of deli very.
26 The Second National Health and Nutrition Examination Survey (NHANES II) was conducted by the U.S. Department of
Health and Human Services from 1976 to 1980 and provides researchers with a comprehensive set of nutritional, demographic and
health data for the U.S. population.
G-13
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
The effect of lead exposure on the blood pressure
of women, relative to the effect on men, is examined
in a review of ten published studies.27 All of the re-
viewed studies included data for men, and some in-
cluded data for women. A concordance procedure was
used to combine data from each study to predict the
decrease hi diastolic BP associated with a decrease
from 10 ug/dL to 5 ug/dL PbB. The results suggest
that the effect on blood pressure for women of this
decrease in blood lead is 60 percent of the effect of
the same change observed in men. Thus, for women,
Equation can be rewritten as:
=(0.6X1.4) X In
(21)
where:
ADBP,
PbB,
PbB,
the change hi women's diastolic
blood pressure expected from a
change hi PbB;
blood lead level in the control
scenario; and
blood lead level in the no-control
scenario.
-v'l
:«
Although women are at risk of having lead-in-
duced hypertension, there is not a dose-response func-
tion for hypertension in women available at this time.
Omitting the hypertension benefits for women cre-
ates an underestimate of the total benefits, but the
impact on the total benefits estimation will likely be
small. Lead raises blood pressure in women less than
in men, so the probability of causing hypertension is
likely to be less than in men, and the total value of
hypertension in men is a small portion of the overall
estimated benefits.
Changes in Coronary Heart Disease
Quantifying the relationship between blood
pressure and coronary Heart disease
Elevated blood pressure in women results in the
same effects as for men (the occurrence of CHD, two
types of stroke, and premature death). However, the
general relationships between BP and these health
effects are not identical to the dose-response functions
estimated for men. All relationships presented here
have been estimated for women aged 45 to 74 years
old using information from Shurtleff (1974). First-time
CHD in women can be estimated from the following
equation:
where:
APr(CHDwomen) = change hi 2 year probability
of occurrence of CHD event for
women aged 45-74;
DBF = mean diastolic blood pressure in
the control scenario; and
DBF = mean diastolic blood pressure in
the no-control scenario.
Again, non-fatal CHD events were estimated by
assuming that two-thirds of the estimated events were
not fatal (Shurtleff, 1974).
Valuing reductions in CHD events
Values of reducing CHD events for women are
assumed to be equal to those calculated for men
(above): $52,000 per CHD event.
Changes in Atherothrombotic Brain Infarctions
and Initial Cerebrovascular Accidents
Quantifying the relationship between blood
pressure and first-time stroke
For initial atherothrombotic brain infarctions hi
women, the logistic equation is:
where:
APr(BI )=
N women'
DBF =
DBF =
change in 2 year probability of
. .
brain infarction in women aged
45-74;
mean diastolic blood pressure in
the control scenario; and
mean diastolic blood pressure in
the no-control scenario.
The relationship between BP and initial cere-
brovascular accidents can be predicted by the follow-
ing logistic equation:
, + g9.Q7737'-
27 Schwartz, 1992b.
G-14
-------
Appendix G: Lead Benefits Analysis
where:
APr(CA
^ \i
DBF
DBF.
= change hi 2 year probability
of cerebrovascular accident in
women aged 45-74;
mean diastolic blood pressure hi
the control scenario; and
mean diastolic blood pressure in
the no-control scenario.
The predicted incidences of avoided BI and CA
were multiplied by 70 percent to estimate only non-
fatal strokes (Shurtleff, 1974).
Valuing reductions in strokes
The value of avoiding an initial cerebrovascular
accident or an initial atherothrombotic brain infarc-
tion for women was calculated hi the same way as for
men (see above). Of women in the United States be-
tween the ages of 45 and 74 (the age group for which
lead-related stroke was predicted), 38.2 percent are
ages 45-54 and the remaining 61.8 percent are ages
55-74. Using these percentages, and the gender- and
age-specific values in Taylor et al. (1996) the average
value among women ages 45-74 of avoiding either
type of stroke was estimated to be about $150,000.
Changes in Premature Mortality
Quantifying the relationship between blood
pressure and premature mortality
The risk of premature mortality in women can be
estimated by the following equation:
where:
APr(MORT
DBP
DBP
n) = the change in 2 year prob-
ability of death for women aged
45-74;
mean diastolic blood pressure in
the control scenario; and
mean diastolic blood pressure in
the no-control scenario.
Valuing reductions in premature mortality
The value of reducing premature mortality for
women is assumed to be equal to that estimated for
all premature mortality, $4.8 million per incident (see
Appendix I).
Quantifying Uncertainty
Characterizing Uncertainty Surrounding the Dose-
Response Relationships
The dose-response functions described for each
health endpoint considered above generally quantify
the adverse health effects expected due to increased
exposure to lead. For children, these relationships are
described directly in terms of changes in blood lead.
For adults, effects are estimated in terms of changes
in blood pressure (which are related to changes in
blood lead levels). As with any estimate, uncertainty
is associated with the dose-response relationships.
Consistent with the approach outlined in Appen-
dix D for the non-lead criteria air pollutants, this analy-
sis attempts to capture the uncertainty associated with
these relationships. This is accomplished by estimat-
ing a distribution associated with each dose-response
coefficient using the information reported in the lit-
erature. This analysis assumes these distributions to
be normal. For each of the coefficients used to relate
adverse health effects to lead exposure, Table G-2
summarizes the means and standard deviations of the
normal distributions used in this analysis.
Characterizing Uncertainty Surrounding the
Valuation Estimates
The procedure for quantifying uncertainty asso-
ciated with the valuation estimates is similar to that
used to characterize the dose-response coefficient es-
timates. The valuation distributions for health effects
considered in the lead analysis are documented in
Appendix I.
G-15
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table G-2. Uncertainty Analysis: Distribution's ,Associa«:i
Response Coefficients Used to Estimate Lead'Hea||h Effects
Health Effect
Blood Lead-Blood Pressure Coefficieat
(Adults)
Adult Males
Mortality (ages 40-54)
Mortality (ages 55-64)
Mortality (ages 65-74)
Chronic Heart Disease (ages 40-59)_
Chronic Heart Disease (ages ^0-64J
Chronic Heart Disease (ages 65-74)
Cerebrovascular Accidents
Atherothromhotic Brain Infarc|iojns-
Hypertension
Adult Females
Mortality (ages 45-74)
Chronic He art Disease
Cerebrovascular Accidents
Atherothrombotic Brain Infarctions
Children
Infant Mortality
Lost IQ Points
I<70 (cases)
distributions describing Dose-
Response Coefficients--- ^
Standard 'J-
eviation
0.01866"
, ^
- 0.00547
"-0:06533
0-00667
10.003586
^.--
Q.Q231C;:,
-0,02031' ,,:
, . ...... *&< IA , ,
:«-,. * 0.007 11
,
0'. 7*93
1 0,245-'
v-telies'''btt!!losf.'I,Q Poiat
Industrial Processes and Boilers
and Electric Utilities
This section describes the methods and data
sources used to estimate changes in blood lead levels
due to changes in lead emissions from industrial pro-
cesses and boilers between 1970 and 1990 and from
electric utilities between 1975 and 1990. The estimates
of the changes in health effects resulting from changes
in lead emissions due to the CAA are also presented.
Methods Used to
Determine Changes in
Lead Emissions from
Industrial Processes
from 1970 to 1990
This analysis used several
sources to determine the
changes in facility-specific
emissions of lead from indus-
trial processes. To summarize,
the analysis extracted 1990 fa-
cility-specific lead emissions
data from the Toxics Release
Inventory (TRI), which pro-
vides recent emissions data for
over 20,000 U.S. manufactur-
ing facilities. This study then
adjusted these data by the rela-
tive changes in lead emissions
between 1970 and 1990; these
relative changes were derived
from several data sources de-
scribed below. This method
yielded facility-specific emis-
sions for five year intervals be-
tween 1970 and 1990 for both
the controlled and uncontrolled
scenarios. The five-year values
were interpolated to derive an-
nual changes for each year be-
tween 1970 and 1990. Specific
details on this approach are
given below.
TRI Data
The Toxics Release Inven-
tory (TRI) is mandated by the
Superfund Amendment Reau-
thorization Act (SARA) Title
III Section 313 and requires that U.S. manufacturing
facilities with more than 10 employees file annual re-
ports documenting multimedia environmental releases
and off-site transfers for over 300 chemicals. Facili-
ties report both stack and fugitive releases to air. Re-
ported releases are generally estimates rather than
precise quantifications. Emissions data can be pre-
sented as numerical point estimates, or, if releases are
below 1,000 pounds, as an estimated range of emis-
sions.
"'so
0100385
0.00754-
G-16
-------
Appendix G: Lead Benefits Analysis
From the TRI data base, this analysis extracted
data from the reporting year 1990 for all facilities re-
porting emissions of lead to air, as either stack or fu-
gitive emissions. Data were reported as annual emis-
sions (in pounds per year). Where emissions are re-
ported as a range, this analysis used the upper bound
of the range to represent the emissions.28 TRI facili-
ties also report their location by latitude and longi-
tude. In order to later match facilities emitting lead
with Census data on surrounding exposed populations,
this analysis uses the latitudes and longitudes of lead-
emitting facilities.
Derivation of Industrial Process Emissions
Differentials 1970-1990
The TRI database is the Agency's single best
source of consistently reported release data; however,
the database does not include information for most of
the years modeled in this analysis. Furthermore, this
analysis required estimates of hypothetical emissions
in the absence of the CAA. Therefore, estimates were
created for the emissions of lead from industrial
sources under the CAA, and in the absence of the CAA,
for the years 1970, 1975, 1980, 1985, and 1990. The
percent changes, or differentials, reflected by these
estimates were then applied to the 1990 TRI data to
obtain facility-level release estimates for the years of
interest for the control and no-control scenarios.
The method for creating these differentials cap-
tured the two potential causes of the differences be-
tween emissions from industrial sources regulated by
the CAA and emissions from those same sources in
the absence of the CAA. The first cause of the differ-
ence in emissions is a change in overall industrial
output, resulting from the macroeconomic impact of
the CAA. The second element is a change in emis-
sions per unit of output, which results from the adop-
tion of cleaner processes and the application of emis-
sions control technology mandated by the CAA. The
methods used to project the effects of these two causes,
described below, were designed to be as consistent as
possible with other emissions projection methods for
other segments of the CAA retrospective analysis.
Data sources
Data for the differentials estimates were taken
from the following sources:
• the Jorgenson/Wilcoxen (J/W) model projec-
tions, conducted as part of the section 812
analysis. This data source addresses the first
cause of changes in emissions: the macroeco-
nomic changes that resulted from the imple-
mentation of the 1970 CAA. The J/W model
calculated the change in economic output for
each of thirty-five industrial sectors, roughly
analogous to two-digit standard industrial
classification (SIC) codes, that resulted from
the CAA's implementation. The specific out-
put used from the J/W model in this analysis
was the percentage change in economic out-
put for the various industrial sectors, rather
than any absolute measure of economic ac-
tivity.
• the 1991 OAQPS Trends database. This data
base is an emissions projection system that
was used to produce the report, "The National
Air Pollutant Emission Estimates, 1940-
1990." It contains information on economic
activity, national level emissions and emis-
sion controls, by industrial process, from 1970
through 1990. Three different elements were
extracted from the Trends database: the emis-
sions of lead per unit economic output for
various industrial processes for the years
1970-1990; annual economic output data for
these industrial processes; and the emission
calculation formula.
• the National Energy Accounts (NBA), com-
piled by the Bureau of Economic Analysis.
This database records the historical levels of
industrial energy consumption, disaggregated
by fuel type at the approximately three-digit
SIC code level.
The manner in which these data were combined
to derive lead emissions estimates is described be-
low.
Estimates of industrial process emissions in the
control scenario
Emissions data for industrial processes were esti-
mated for the years 1970,1975,1980,1985, and 1990.
For each of these years, this analysis extracted an
emission factor and a control efficiency for each lead-
» Ranges are infrequently reported and are either reported as 0-500 Ibs. or 500-1000 Ibs. The infrequency of the incidence of a
facility reporting a range and the relatively small quantities of lead released by those facilities means any overestimation of benefits
that results from using the upper limit of the range is extremely minor.
-------
Tlie Benefits and Costs of the Clean Air Act, 1970 to 1990
emitting industrial process in the Trends database.
Emissions factors are expressed as amount of lead
emitted per unit of economic activity, and control ef-
ficiencies are reported as the percent that emissions
are reduced through the application of pollution con-
trol technology to the process. The year-specific emis-
sion factors and control efficiencies were multiplied
by the economic activity data for that year, for that
process, as reported in the Trends database, using the
following equation found in the Trends report:
Emissions = (Economic Activity) x
(Emission Factor) x (1 - Control Efficiency)
This calculation yielded the estimated control sce-
nario emissions, by industrial process. Industrial pro-
cesses were then assigned to an NBA code. Finally,
all processes assigned to a given NEA code were
summed to give a total emissions estimate for that
NEA code.
Estimates of industrial process emissions in the
no-control scenario
The results from the J/W model were used to es-
timate process emissions in the no-control scenario.
As stated above, the J/W model provides percent
changes in economic outputs by industrial sector. To
use these values, lead-emitting industrial processes
(in the Trends database) were assigned to a J/W sec-
tor. The percent change for that sector from the J/W
model was then used to adjust the economic activity
data for that process from the Trends database. These
adjusted economic output figures were used together
with 1970 emission factors and control efficiencies to
derive the estimated lead emissions for each indus-
trial process hi the no-control scenario. The 1970
emission factors and control efficiencies were used
for all years in the analysis (1970, 1975, 1980, 1985
and 1990) in the no-control scenario; this assumes that
emissions per unit economic output and control effi-
ciencies would have been constant over tune in the
absence of the CAA. This is the same approach that
was used to project the changes in emissions from
industrial processes for other criteria pollutants in other
portions of the CAA retrospective analysis. The pro-
cess-level emissions were then aggregated to the NEA-
code level, as in the controlled scenario.
Matching TRI Data to Industrial Process
Emissions Differentials
The methods described in the preceding section
yielded emissions estimates from industrial processes
in the control and no-control scenarios, by NEA code.
We used these estimates to derive percent changes in
emissions between control and no-control scenarios,
by NEA code, for application to the TRI emissions
data. However, since TRI data are reported by SIC
code, we first mapped" NEA codes to the appropriate
SIC codes, and used the percent change for each NEA
code to represent the percent change for all SIC codes
covered by that NEA code.
It should be noted that the Trends data base cov-
ers only the most important sources of lead in air, not
all sources; as a result, not all SIC codes reporting
lead emissions in TRI correspond to an NEA code for
which emission differentials have been estimated.
However, we assume that the TRI emissions sources
that have a match are the most important sources of
lead air emissions. In fact, although only 48 out of
519 legitimate SIC codes reporting lead emissions in
TRI have matching differentials, these SIC codes ac-
count for over 69 percent of the lead emissions re-
ported in TRI. The remaining 31 percent of the emis-
sions are distributed relatively evenly among the re-
maining 471 SIC codes, each of which contributes a
small amount to total emissions.
For the 31 percent of the emissions without dif-
ferentials, this analysis has no information regarding
the change in the lead emissions over time or between
the control and no-control scenarios; therefore, we are
unable to predict benefits attributable to the CAA for
these emission sources. Although excluding these
sources may lead us to underestimate total benefits,
we believe these sources are unlikely to contribute
significantly to the difference between control and no-
control scenarios. The Trends data focus on the point
sources of lead emissions of greatest concern to the
Project Team and of greatest regulatory activity. If a
process within an SIC code does not appear in the
Trends, it is unlikely to have had specific CAA con-
trols instituted over the past 20 years. A lack of con-
trol efficiencies for smaller sources prevents them
from being included.
It should also be noted that the total industrial
process emissions of lead estimated in the 1990 Trends
report actually exceeds the reported lead emissions in
G-18
-------
Appendix G: Lead Benefits Analysis
TRI, despite the fact that TRI covers more SIC codes.
This is probably attributable in part to the fact that
TRI covers only a subset of the facilities contributing
to economic output in an SIC code. TRI reporting rules
only require facilities with greater than 10 employees
and who use certain amounts of lead in their processes
to submit information to TRI, while the Trends report
attempted to estimate emissions from all sources con-
tributing to the economic output for the industrial sec-
tor, regardless of size. However, the components of
the Trends data base used in this analysis (i.e., emis-
sions factors, economic output data) represent typical
conditions at average facilities; they do not allow for
the representation of the distribution of emissions
across particular facilities. In contrast, a major strength
of the TRI is its match of emissions data with geo-
graphical information. Because the distribution of
emissions geographically determines the size of ex-
posed populations, this analysis used the TRI data,
rather than Trends data, to characterize lead release
quantities, and used the Trends figures only to char-
acterize relative emissions and changes over time,
rather than to estimate total quantities.
Because the Trends data are intended only as an
estimate of emissions using typical conditions at av-
erage facilities, and do not capture the differences in
facility-level emissions, the data do not provide suffi-
cient information to make specific quantitative adjust-
ments to the TRI-based benefits estimates to account
for the overall higher emissions estimates in Trends.
However, since Trends does generally suggest that
there are many more sources than are accounted for
by TRI, it is possible that our benefits calculations
may be underestimated.
Some additional assumptions were necessary
when matching the TRI lead release data and the dif-
ferentials from the Trends data. Ideally, we would like
to know whether the facilities present at a given loca-
tion, as reported hi the 1990 TRI, were present and
operating in earlier years; whether facilities operat-
ing in 1970 have ceased to operate; and whether new
facilities would have been constructed in the no-con-
trol situation. Unfortunately, data do not exist in an
accessible form at this level of detail for the years
1970 through 1990. Therefore, for the purposes of this
exercise, we have assumed that the locations and num-
bers of the 1990 sources are the same as they were in
1970.
Methods Used to Determine Changes in
Lead Emissions from Industrial Boilers
from 1970 to 1990
Several sources were used to determine the change
in lead emissions from industrial boilers. TRI
locational data, Trends database national fuel con-
sumption levels and emissions factors, and NEA and
SIC codes were used to derive the emissions for the
control and no-control scenarios.
TRI Data
The TRI does not appear generally to contain com-
bustion emissions data. In general, the emissions data
are from process sources. We reached this conclusion
based on two pieces of information:
(1) TRI reporting requirements: TRI has three
reporting requirements: (a) the facility must fall in
SIC codes 20-39; (b) the facility must employ more
than 10 persons; and (c) the facility must manufac-
ture or process more than 25,000 pounds of a TRI
chemical, or otherwise use more than 10,000 pounds
Firms must submit reports only for the chemical that
exceeds the thresholds given in item (c), but they must
report all releases of that chemical, including releases
from uses that would not qualify alone. If the TRI
chemical is part of a blended substance and the quan-
tity of the TRI chemical in the blend exceeds the
threshold, it must be reported. For industrial boilers,
if the amount of lead in the fuel were to exceed the
10,000 pounds threshold, then the firm would be re-
quired to report all emissions of lead from combus-
tion of fuel. There is an exemption, however, for in-
gredients present in small proportions. If the amount
of lead in the oil were less than 0.1 percent (1,000
ppm), then the firm would not be required to report
the emissions.
The conclusion from the above information is that
most firms burning used oil are probably not report-
ing lead combustion emissions to TRI because these
releases fall outside the TRI reporting requirements.
The concentration at which lead is typically found is
used oil (100 ppm) (NRDC, 1991) is much less than
the minimum concentration required for reporting
(1,000 ppm).
(2) Use data from the TRI data base: The hypoth-
esis that firms do not report lead combustion was con-
firmed by an analysis of the data submitted by the
G-19
-------
Tlie Benefits and Costs of the Clean Air Act, 1970 to 1990
firms reporting lead use to TRI. On the TRI submis-
sion forms, firms must indicate how the chemical is
used. Our analysis of category codes submitted by
firms reporting lead emissions showed the following
four use category reports: as a formulation compo-
nent; as a reactant; as an article component; and re-
packaging only. None of these category codes sug-
gest that the source of the reported lead release is com-
bustion. Therefore, we may conclude that all of the
lead emissions reported in TRI are process emissions.
Based on these analyses, the Project Team could
not use the TRI release data to evaluate releases of
lead from industrial combustion. However, the TRI
geographical information was used to locate indus-
trial facilities by longitude and latitude in order to
combine combustion data with population informa-
tion. For combustion emissions, the calculations in-
cluded all TRI reporting facilities, not just those who
reported lead emissions. The assignment of combus-
tion emissions to these facilities is described below.
Derivation of Industrial Combustion Emissions
1970-1990
As with industrial process emissions, estimates
were created for the emissions of lead from industrial
combustion under the CAA, and in the absence of the
CAA, for the years 1970,1975,1980,1985, and 1990.
These emissions estimates were used, in combination
with the TRI data base geographic information, to
obtain facility-level release estimates for the years of
interest for the control and no-control scenarios. The
method for deriving these emissions estimates in-
cluded both the macroeconomic impact of the CAA
and the change in emissions per unit of output that
resulted from specific pollution control mandates of
the CAA. The same data sources were used to derive
combustion differentials as were used to derive pro-
cess differentials. The particular data elements and
the methods by which these data were combined to
derive lead emissions estimates from industrial com-
bustion are described below.
Estimates of combustion emissions under the
control scenario
The Trends database contains a national aggre-
gate industrial fuel consumption estimate, by fuel type
(coal, natural gas, oil). For each fuel type, the fuel
consumption estimate was disaggregated by the share
of that fuel used by each NEA industrial category,
using the NEA data base. It should be noted that the
NEA includes data only for the years 1970 through
1985. For 1990, the 1985 figures were used to disag-
gregate the national-level consumption figure into
NEA industrial categories.
The Trends database also contains emissions fac-
tors for industrial fuel use, by fuel type, as well as
control efficiencies. The lead emissions from indus-
trial combustion for each NEA category was derived
by multiplying the fuel-specific combustion estimate
for each NEA category by the emission factor and
control efficiency for that fuel type. The result was
emissions of lead by NEA code and by fuel type.
Emissions from all fuel types were then summed by
NEA code. By using the NEA data to disaggregate
the industrial fuel consumption figures, the analysis
assumes that the industrial combustion emissions are
the same among all industries covered by a given NEA
code, an assumption which may bias the analysis.
Estimates of combustion emissions under the
no-control scenario
As in the control scenario, the national aggregate
industrial fuel consumption estimate, by fuel type
(coal, natural gas, oil), was disaggregated by the share
of that fuel used by each NEA industrial category.
The fuel use was then adjusted by one of two factors:
(1) seven of the NEA codes were specifically mod-
eled by the Industrial Combustion Emissions (ICE)
model — for these sectors, the ICE modeled percent
changes were used instead of J/W percent changes; or
(2) the remaining NEA codes were matched to J/W
sectors — the J/W percent changes were then applied
to those matched NEA codes. These fuel use estimates
were then combined with the 1970 emission factors
and control efficiencies for industrial combustion by
fuel type from the Trends database to obtain combus-
tion-related lead emissions from industrial boilers in
the no-control scenario, by NEA code.
The process-specific data in the Trends database,
and the energy use data in the NEA, are much more
disaggregated than the J/W sectoral projections. For
the purpose of the analysis, it was assumed that all of
the specific industrial processes in the Trends data-
base and industrial categories in the NEA data set as-
signed to a given J/W sector changed at the same rate
as the entire J/W sector. For example, if the economic
activity in the J/W Sector 20, "Primary Metals,"
changed by one percent between the control and no-
control scenarios, then the analysis assumed that eco-
nomic activity in each industrial process assigned to
G-20
-------
Appendix G: Lead Benefits Anafysis
the Primary Metals sector also increased by one per-
cent. This approach assumes that the economic ac-
tivities of specific industries within a sector are equally
affected by the imposition of the CAA. This assump-
tion is consistent with the projection of the change in
emissions from industrial processes for the other cri-
teria air pollutants, which were calculated using a simi-
lar process.
Matching TRI Data to Industrial Combustion
Emissions Data
Because of the structure of the TRI reporting re-
quirements, it does not appear that TRI generally con-
tains releases from combustion sources. Although TRI
may incidentally contain lead combustion emissions,
TRI would contain data on such releases only if the
reporting facility also used more than 10,000 pounds
of lead per year for manufacturing or processing. As
a result, the combustion releases, estimated using the
methods described above, do not have corresponding
data in the TRI data base. Therefore, we devised a
different method for estimating benefits from changes
in combustion releases.
The first step in the method was to divide the es-
timates of total releases of lead from industrial com-
bustion, by NBA code, by an estimate of the number
of facilities in each NBA code. The number of facili-
ties in each NBA category was estimated using the
1987 Census of Manufactures. This Census, conducted
by the U.S. Department of Commerce, tallies the num-
ber of facilities by four-digit SIC code; these SIC codes
were matched to the NBA codes.
Dividing total lead emissions emitted by number
of facilities yielded the average yearly lead emissions
from industrial combustion for each SIC code. We
then assigned this average value to all reporting TRI
facilities in the SIC code. The consequence of this
approach is that the modeling of combustion from
industrial facilities includes substantially more sources
than the modeling of industrial process emissions;
combustion emissions are assigned to essentially all
facilities reporting to TRI, while the process emis-
sions are only evaluated for facilities actually report-
ing lead air emissions from processes.
One unavoidable drawback to this approach is that
it cannot capture differences in release quantities
among facilities within an SIC code. Furthermore, this
approach does not capture all combustion emissions
because we assign average emissions only to facili-
ties that report to TRI. TRI facilities account for be-
tween two percent and 50 percent of all facilities listed
in the Census of Manufacturers, depending on the SIC
code. Because of the inability to place the remaining
facilities geographically, this analysis excludes the
consideration of emissions from non-TRI facilities.
Methods Used to Determine Changes in
Lead Emissions from Electric Utilities
from 1975 to 1990
The estimation of lead emissions from electric
utilities required data from three different sources.
Energy use data for the control and no-control sce-
narios were obtained from the national coal use esti-
mates prepared for the section 812 analysis by ICF
Incorporated. The OAQPS Trends Database provided
emissions factors and control efficiencies. Individual
plant latitudes, longitudes, and stack information were
collected from the EPA Interim Emissions Inventory.
This analysis combines these three sets of data and
estimates annual lead emissions at the plant level for
coal burning electric utilities in the control and no-
control scenarios. This section describes the sources
and the methods used to create the final data set.
Coal-Use Data
The energy use data obtained from national coal-
use estimates provide plant level energy consumption
information for 822 electric utilities. The data set were
separated into four distinct sets for the years 1975,
1980, 1985, and 1990. Each set of data provided the
state where the plants are located, the plant names,
and the amount of coal consumed, for both the con-
trol and no-control scenarios. The four data sets were
combined into one comprehensive set by matching
the plants' names and states.
The EPA Interim Emissions Inventory
The EPA Office of Air Quality Planning and Stan-
dards Technical Support Division provided the 1991
EPA Interim Emissions Inventory. The Interim Inven-
tory contains data for all electric utility and industrial
plants in the United States including latitude, longi-
tude, stack height, stack diameter, stack velocity, and
stack temperature. The additional stack parameter data
allowed the use of plant-specific parameters in the air
modeling for electric utilities rather than average pa-
rameters for all facilities as was done for industrial
emissions.
G-21
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Matching the Coal-Use Data to the Interim
Emissions Inventory
The combination of the Interim Emissions Inven-
tory and the coal-use data required two steps. First,
the Interim Emissions Inventory had to be pared down
to include only electric utility data, and to narrow the
information provided for each utility. Second, the two
databases had to be combined. One difficulty in com-
bining them was the lack of a common data field that
would allow a quick and complete matching process.
Electric utility plants were identified in the In-
terim Emissions Inventory by SIC code (code 4911).
The associated stack information file, which lists the
size of every stack on every plant, was reduced to in-
clude only the tallest stack for each plant. This pro-
vides a reasonable estimate of the stack height at which
most emissions occur. The air modeling assumes that
each electric utility releases its emissions from the
largest stack that exists at that plant.
Next, the procedure matched the abridged Interim
Emissions Inventory file with the coal use data. Due
to the lack of a common data field between the two
sets, this process required several phases. Both data
sets had name fields, but these fields utilized differ-
ent naming conventions for the plants. Therefore the
name fields were matched directly, with individual
words in the names, and then with abridged words
from the names. Abridged word matches were double
checked by ensuring that the names were indeed simi-
lar and by verifying that the state fields matched. Fi-
nally some matches were made by hand.
Only 27 unmatched plants with positive coal use
remained. There were 493 matched plants with posi-
tive coal usage and these were included in the final
data set.29 To eliminate under-counting of emissions,
the emissions from the 27 unmatched plants were al-
located to matched plants within the states where the
unmatched plants were located. Allocations were
weighted according to the emission level for each
matched plant within that state in the year in which
the allocation was being made.
This equation produces estimates of the emissions
per plant per year in both the control and the no-con-
trol scenarios.
» Plants with zero coal usage were not immediately excluded from the analysis due to the possibility of analyzing lead emissions
from oil combustion at these plants. However, OAQPS has suggested that oil combustion comprises under two percent of the total
lead emitted from electric utilities. For this reason, the electric utility analysis focused entirely on coal.
» The actual figure cited is 12.1 metric pounds per 1,000 tons. A metric pound is one two-thousandth of a metric ton.
31 U.S. EPA, 1991a .
— ' CW2
Emissions Factors and Control Efficiencies
At this stage, the electric utilities data set con-
tained coal consumption by plant by year in the con-
trol and no-control cases as well as air modeling pa-
rameters. Using emission factors for lead and control
efficiencies for electric utilities, estimates of lead emis-
sions per plant per year could now be calculated. As
in the industrial source analysis, the emission factors
and control efficiencies come from the 1991 OAQPS
Trends database.
Control efficiencies are available for coal-fired
electric utilities in each year between 1975 and 1990.
As in the industrial source analysis, it is assumed that
pollution control on coal-burning power plants with-
out the CAA would be the same as the pollution con-
trol level in 1970. Therefore, the control efficiency
from 1970 is used in the no-control analysis.
The emission factor obtained from the Trends
database is expressed in terms of lead emitted per ton
of coal burned (6,050 grams per 1,000 tons).30 The
combined data set, though, contains quantity of coal
burned per plant per year in energy units (trillions of
BTUs). To reconcile this difference, a conversion fac-
tor was obtained from a 1992 DOE report titled Cost
and Quality of Fuels for Electric Utility Plants 1991.
The conversion factor used (20.93 million BTUs per
ton of coal) is the average BTU per pound of coal
burned for all domestic electric utility plants in 1990.
Data for a small subset of other years were also pro-
vided in the DOE report, but they did not differ sig-
nificantly from the 1990 number. Therefore, the 1990
conversion factor (637.3 pounds of lead per trillion
BTU) is assumed valid over the entire study period.
The final equation for lead emissions looks quite simi-
lar to the equation used in the industrial source analy-
sis.31 The only change is that "Economic Activity"
has been replaced by "Coal Consumed" for this par-
ticular analysis:
(Coal Consumed)*. *' %,
(Emission Factor) x (I - Control Efficiency)'" (26)
-------
Appendix G: Lead Benefits Analysis
Use of Air Dispersion Modeling to
Estimate Ambient Air Lead Levels
To link estimates of lead emissions to blood lead
levels of populations living in the vicinity of a facil-
ity, the lead benefits model first uses air dispersion
modeling to estimate air lead concentrations surround-
ing facilities that emit lead into the air. The air con-
centrations are then linked to blood lead levels.
This analysis uses the Industrial Source Complex
Long Term (ISCLT) air dispersion model, a steady-
state Gaussian plume model, to estimate long-term
lead concentrations downwind of a source. The con-
centration is modeled as a function of site parameters
(stack height, stack velocity).32 The general form of
the concentration equation from a point source at a
distance r greater than one meter away is as follows:33
, !ft~,~yyv
*-t _*'.'.- V - — '
,.. v|28)
where,
C*
Q
f
0
s
u =
V
K
concentration at distance r (ug/m3),
pollutant emission rate (g/sec),
frequency of occurrence of wind speed
and direction,
sector width (radians),
smoothing function used to smooth
discontinuities at sector boundaries,
mean wind speed (m/sec),
standard deviation of vertical concentra-
tion distribution (m),
vertical term (m),
scaling coefficient for unit agreement.
For each facility modeled in the lead benefits
model, a 21 by 21 kilometer grid around the source is
specified. The model stores data in 1 km by 1 km cells
and calculates the air lead concentrations for each of
the 441 cells surrounding a given facility. Fugitive
sources are modeled similarly, the only difference
being a modified form of Equation 28.
For facility-specific weather data, the model used
Stability Array (STAR) data. The STAR data contain
information on typical wind speed and direction for
thousands of weather stations in the U.S. For each
facility, the model accesses the STAR data for the
weather station nearest the source. Standard default
parameters are used for the other parameters because
facility-specific data are not available for them (ex-
cept for utilities). Table G-3 lists default parameters
for the ISCLT, and summarizes sources for other pa-
rameters.
Industrial process emissions were modeled as ei-
ther point or fugitive sources, depending on how they
were reported in TRI. All industrial combustion emis-
sions were modeled as "fugitive" emissions. This is a
more appropriate model scenario for boiler emissions
than a 10 meter stack scenario. All electric utility
sources were modeled as point sources.
The model tracks all lead emissions to a given
grid cell. That is, if the plumes of two or more sources
overlap in a given cell, the air concentration in the
given cell is determined from the sum of all of the
contributing sources.
Determination of Blood Lead Levels
from Air Lead Concentrations
Once the air lead concentrations surrounding a
given plant are estimated, the model estimates blood
lead levels for children and adults living in those ar-
eas. This section describes the methods and data
sources used to derive blood lead levels from esti-
mated air lead concentrations.
Relationship Between Air Lead Concentrations
and Blood Lead Levels
The rates at which lead is absorbed from air de-
pend on the age of the exposed individual, distance
from the facility, the initial concentration of blood
lead, and other factors. In addition, rates determined
from empirical data may differ depending on whether
or not the analyses from which rates are derived have
controlled for factors such as lead exposure through
deposition on dust and soil (i.e., "indirect exposure").
Especially when children constitute the exposed group,
the inclusion of indirect exposure results in higher air
lead to blood lead slopes. In both cases, the slope re-
32 Ideally, reported stack and fugitive air releases would be modeled using site-specific data (such as source area or stack height).
However, since TRI does not contain such facility-specific information, default values are used to model TRI facilities.
33 This equation is from U.S. EPA (1992). The equation is for a specific wind speed, direction, and category (ijk). Each facility
has several combinations of these that must be added to arrive at a total concentration at that point. The equation for area sources is
similar.
G-23
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table G-3, Air Modeling Parameters. ' -
Parameter
Stack height
Exit -velocity
Stack diameter
Exit gas temperature
Area source size
Area source height
Lead em ission rate
Frequency of \vlnd speed
Scctorwidth
Wind speed
Smoothing function
,?
Industrial *
Source
Value
10m
0,01 tats
Ink • '
293* 1C >
>
10 mj- - '•'
3m
site-specific* »
site*specifitf"
215"
site-specific
calculated ,
Electric ,-,
"tftUfty, - ' s
Value" , >.
site-specific or 11 5,0
m*
' •? * '/
4ite-sp,e,cific or 22'.5', ,
" Jtt/S* <.-,"
site-specific or 5,15 ,-
mt * ' "
site«specific or,
•"''" ' ,- , 427,5* "'
iprit*'" "* '
* 3a»- -"* , ,
" site;specific , ,
- sife-spe'cifiCj ,,,-,- ^
y.,s ' p
--,--22,5° --'*'-,
site-specific-" '
,f >*>•>""
calculated' „.' -t
, , - cslcSiated^- ;,
; "'" Soiiree/,,, -J '' , " v~
*- - - -- Comment •*'
,?/*•*. ' , ".,•• '•">'
' -"^"
Industrial- U.S.'EPA (1992)' titles »-'U,5, •/.
BP.AOL9&H?) ,- ' •'•' * '**'*- '.*-.
Industrial - U,s" EPA (1992) , ptMt'es »-"UsS, ;";
' ' - '>*.-' ' '* " ' - - *"' *'*'•*-
-In'dastml - U ''.
tadusttial -, U.S,'EP^ (1992) Utilities "U.S. ,„
-EPAd991b)-- A'*'~ • ' ''$" -;'' --•'
U.S.'ipA<1992)' ,* > ^''"" ••-'-•',, . ..
U.S. EPA (1992) ••"'' ,-• '-'• , ,
; Industrtkl"- f RJSWs/yr) "'""''',, " • £','
VtiHsta). • ,-'-.-
STAR data ' " '"'' '"'',„--"'-'
••A,//*' /, ,
-660* divided by 16 wind directions - "' „„ -,-
.^STARdata (Hi/sec) '„ - , ""'^';
, ,,, f t^ f" •,,-.>?'S'jf*f ;??*%* % ,
' ""^ s v^A** " ^<"'^J \ <»<;''* J
>f , , ^ ' •"• //f ''' vp/ JJ
* average value for electric utilities, utilized for,uptieswithOBt«ftisMorinatipn
lationship is expressed as the change in blood lead
(Hg/dL) per change in air concentration (j-ig/m3).
In performing this analysis, a choice had to be
made between the use of air lead'.blood lead relation-
ships that account for inhalation exposure ("direct"
slopes) and those that account for exposure to lead
deposited from air onto soil and dust ("indirect"
slopes). The choice of which slopes to use considered
both the effects on the estimate of benefits over time
(from 1970 to 1990) and the estimate of the differ-
ence in benefits between the control and no-control
scenarios. The indirect slope is more comprehensive
in its coverage of the types of exposures that will re-
sult from air releases, and thus captures more of the
health effects predicted to occur from lead exposures,
especially to children. For this reason, indirect slopes
are preferred to dkect slopes, especially when com-
paring the control and no-control scenarios: using only
the direct slope would underestimate the benefits of
avoiding deposition that controls confer. However,
indirect slopes may capture effects from exposure to
soil and dust lead deposited from both current air re-
leases and historic air releases. Since lead's dissipa-
tion from soil is slow relative to its removal from air,
the reservoir of lead in soil and dust is unlikely to
change at the same rate as the reductions in air lead
concentrations. Therefore, using indirect slopes to
represent a change in blood lead over time due to re-
duced air emissions may overestimate the change in
blood lead, and thus overestimate the benefits of re-
ductions over time, to the extent that the indirect slope
captures exposure to the total reservoir of soil and dust
lead, rather than only recently deposited lead.
Given that the focus of this analysis is the differ-
ence between the control and no-control scenarios, it
is important to capture both the benefits from reduced
lead deposition that result from the CAA, and the di-
rect benefits from reduced air concentrations. There-
fore, this analysis modeled changes in blood lead lev-
els using indirect slopes. It should be kept in mind
G-24
-------
Appendix G: Lead Benefits Analysis
that this choice may overestimate blood lead changes
over time forboth the control and no-control scenarios.
The relationship between concentrations of lead
in ambient air and blood lead concentrations has been
evaluated by a variety of methods. These include ex-
perimental studies of adult volunteers, as well as epi-
demiological studies of different populations of chil-
dren and adults. The discussion below describes the
slopes used in this analysis for children and adults,
and for individuals with blood lead values greater than
30 ug/dL.
Children
U.S. EPA (1986b) reports that slopes which in-
clude both direct (inhalation) and indirect (via soil,
dust, etc.) air lead contributions vary widely, but typi-
cally range from three to five ug/dL increment in
children's blood lead per ug/m3 increment in air lead
concentration (roughly double the slope due to inhaled
air lead alone). Since hand dust levels can play a sig-
nificant role in blood lead levels (U.S. EPA, 1986b),
this higher slope may be due to mouthing behavior of
children that brings them into contact with dust and
soil.
Specific values for estimating contribution of air
lead to blood lead, including indirect pathways, are
cited in U.S. EPA (1986b); slope values (ranging from
-2.63 to 31.2) and data sources for these values are
presented in Table 11-36 of U.S. EPA (1986b). The
median of these values is 4.0 ug/dL per ug/m3, which
matches the midpoint of the range of typical slope
values. This analysis used this value to represent the
relationship between air lead concentrations and blood
lead concentrations for children living in the vicinity
of point sources of lead emissions.
The use of this slope assumes that indirect expo-
sure" principally measures indirect effects of lead
emissions to air (through deposition to dust and soil).
However, it is possible that these slopes include other
exposures not related to ah- lead. In many cases re-
searchers have measured other possible exposures,
such as water and food, and have confirmed that the
most significant contribution conies from soil and dust
lead, which is assumed to result from air deposition
of lead. Those studies that measured lead in tap water
showed that mean levels were generally low or not
significantly related to blood lead. Landrigan et al.
(1975) measured lead in pottery and food; lead in pot-
tery was found in only 2.8 percent of homes, and food
and water made no more than a negligible contribu-
tion to lead uptake. Lead in paint was measured in
some studies.34 Landrigan and Baker (1981) measured
lead in paint at levels greater than one percent in about
one fourth to one third of the houses in each area stud-
ied. Brunekreef et al. (1981) measured high levels of
paint in some houses, but excluded these data points
from the analysis.
Despite the possibility of confounding factors, this
analysis uses the median value determined above (4.0
ug/dL per ug/m3) as the appropriate slope for chil-
dren living within five kilometers of the point source.
Five kilometers is chosen as the cut off point because
the data from most of the studies cited collected the
majority of their data points near lead smelters.35 Fur-
thermore, these slopes, although measured primarily
hi the vicinity of smelters, are assumed applicable to
all point sources that emit lead into the air.
Adults
For adult males and females, the air lead/ blood
lead slopes that include indirect effects due to soil and
dust differ very little from slopes that include only
direct effects. This result is expected since the higher
indirect slope values estimated for children are as-
sumed to be as a result of mouthing behavior typical
of young children.
U.S. EPA (1986b) describes several population
studies that estimate indirect slopes for men; these
slopes range from -0.1 to 3.1 ug/dL per ug/m3.36 Snee
(1981) determined a weighted average of these stud-
ies and one other study.37 The average slope, weighted
by the inverse of each study's variance, is 1.0 ug/dL
per ug/m3. However, the Azar study measured the di-
rect relationship between air lead and blood lead. Ex-
cluding the Azar study from the weighted average,
the average slope is 1.1 ug/m3. Excluding the highest
and lowest slopes from this group (from Goldsmith,
34 Landrigan and Baker, 1981; Brunekreef et al., 1981.
35 U.S. EPA, 1986b, Table 11-36.
36 Johnson et al., 1976; Nordman, 1975; Goldsmith, 1974; Tsuchiya et al., 1975; Fugas et al., 1973.
37 Azar et al., 1975.
'G-25
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
1974 and Tsuchiya et al., 1975), both of which had
difficulties,38 the resulting slope is 1.4 ug/dL per ug/
m3.
Slopes for females range from 0.6 to 2.4 for gen-
eral atmospheric conditions.39 Snee determined an
average slope for women of 0.9 ug/dL per ug/m3,
weighted by the inverse of the variances of the stud-
ies. Excluding the slope for women from Goldsmith
(1974), the resulting slope for women is 1.0 ug/dL
per ug/m3.
These values are adjusted by a factor of 1.3 to
account for the resorption of lead from bone tissue
(according to Chamberlain, 1983), thus deriving an
adjusted slope estimate of 1.8 ug/dL blood lead per
ug/m3 increment in air lead concentration for men and
1.3 for women. These are the slope estimates used in
this analysis.
Individuals with initial blood lead levels of 30
ug/dL and greater
For individuals with high blood lead levels, the
air lead to blood lead uptake slopes have been shown
to be much shallower, as described by U.S. EPA
(1986b). An appropriate change in blood lead per
change in air lead is 0.5 ug/dL per ug/m3 for indi-
viduals that have initial blood lead levels in the range
of 30 to 40 ug/dL. This value is based on cross-sec-
tional and experimental studies.40 For individuals with
initial blood lead levels greater than 40 ug/dL, an ap-
Table G-4. Estimated Indirect Intake Slopes:
- ^ ° i^"1^
Unit of Air Lead Concentration (ug/m).
propriate range of slopes is 0.03 to 0.2, as determined
by occupational studies listed in Table 11-37 of U.S.
EPA (1986b). The median value of these studies is
0.07. These two slopes (0.5 for the population with
blood lead levels between 30 and 40 ug/dL and 0.07
for blood lead levels greater than 40 ug/dL) are used
for both children and adults in this analysis. These
relationships are summarized in Table G-4.
Estimates of Initial Blood Lead Concentrations
The benefits model requkes an initial distribution
of blood lead levels in the exposed populations to
model health benefits of reducing lead air emissions.
The model estimates the new distribution of blood lead
levels that would exist after a given change in air con-
centrations using the slopes described above. Finally,
the model estimates the difference between the two
distributions. This analysis begins with an initial 1970,
no-control scenario blood lead distribution from which
all subsequent changes are modeled. This approach
requires an estimate of the blood lead distributions in
the U.S. population in 1970. Unfortunately, there are
no actual national blood lead distribution estimates
for 1970. Although the first NHANES study covered
1970, blood lead data were not collected in this study.41
Nonetheless, a 1970 distribution of blood lead was
estimated using NHANES II data (from 1976-1980),
combined with estimates of typical changes in blood
lead levels from 1970-1976 observed in localized
screening studies.
Adult Moles
Adult Fern ales
Individuals with blood lead
levels < 30 jig/d!>
v" , ^ if, ' 'J /s, /S' ,,/
13
~ '*' 13 '''*" f"
;-, - -_,'.'• ,fy^-
- '4.0 , < " , -'•''<
;#Bdividuais with blood lead,
, vj£veJs,3,0»,40f»g/aL' ' ,
fV" .-'".^jr
- .-•"'•',1-0.5'
'•*%'
. "- ,,,,/;,,, '0,5,,
•-'"'• lBdj,yiduaIs with bMo8"
lead leyfels > 4<), pg/dL :
0.07, ,,,,,,-
0.07 l,,,-,,.
"-"•^'"V o:oT
38 Goldsmith (1974) refrigerated (rather than froze) the blood samples, and did not analyze the samples until 8 or 9 months after
thev were taken, and restricted the analysis to one determination for each blood sample. Tsuchiya et al. (1975) measured air lead
concentrations after blood samples were taken; blood was drawn in August and September of 1971, whereas air samples were taken
during the 13 month period from September 1971 to September 1972.
39 Tepper and Levin, 1975; Johnson et al., 1976; Nordman, 1975; Goldsmith, 1974; Daines et al., 1972.
40 U.S. EPA, 1986b.
41 NCHS, 1993a. . —
G-26
-------
Appendix G: Lead Benefits Analysis
A major drawback to this approach is the uncer-
tainty in deriving the 1970 estimates. Another draw-
back to beginning with the 1970 level and modeling
changes from that point is the analysis only represents
changes in lead exposure from air; reductions from
other sources of lead exposure are not accounted for.
The purpose of this analysis is to identify changes at-
tributable to the CAA mandates; changes from other
sources of lead exposure should not be considered.
However, due to nonlinear nature of the lead concen-
tration-response functions (see above), the overall
exposure context in which the air lead exposure re-
ductions take place will influence the estimate of ben-
efits from those reductions. Specifically, at higher
blood lead levels, the slope of the concentration-re-
sponse curve is shallower than at lower levels. As a
result, a given change in the mean blood lead level
may result in a smaller change in the health effect if
the change occurs from a relatively high starting level.
On the other hand, if one accounts for the fact that
other sources of lead exposure are reduced at the same
time that the given air reductions occur, then those air
emissions reductions may result in greater changes in
health risk.
This issue is of concern even though the analysis
focuses on the difference between the control and no-
control scenarios, since the health benefit implications
of the emissions differentials between the scenarios
will depend on the point on the blood lead distribu-
tion curve at which the differences are considered.
That is, a difference between a mean blood lead of 25
ug/dL and one of 20 ug/dL may have different health
implications than a difference between 15 ug/dL and
10 ug/dL, even though the absolute value of the dif-
ference is the same (5 ug/dL).
An alternative method is to "start" with a 1990
blood lead level and to "back-calculate" benefits by
representing the differentials as increases over the
1990 levels, rather than decreases from 1970 levels.
The advantage of this approach is that it accounts for
reductions in lead exposure from other sources, as rep-
resented by current blood lead levels. Its disadvan-
tage is that it holds other sources constant to (lower)
1990 levels, and thus the modeling may underesti-
mate actual blood lead distributions in earlier years,
and thereby overestimate benefits from controls dur-
ing those years. This analysis presents the results of
both approaches, indicated as "forward-looking" and
"backward-looking".
Combination of Air Concentration
Estimates with Population Data
The modeled air lead concentrations at various
distances from the sources were combined with popu-
lation data from the Census Bureau to arrive at an
estimate of the number of cases of health effects for
each of the years from 1970 to 1990 in both the con-
trol and no-control scenarios. The primary census in-
formation was accessed from the Graphical Exposure
Modeling System Database (GEMS), an EPA main-
frame database system. The following data were ob-
tained from GEMS for the years 1970,1980, and 1990:
total population for each Block Group/Enumeration
District (BG/ED); state and county FDPS codes asso-
ciated with each BG/ED; latitude and longitude of each
BG/ED; and population of males under 5 and females
under 5 for each BG/ED. The intervening five year
intervals (1975 and 1985) were estimated using the
Intercensal County Estimates from the Census, which
estimate annual populations on a county by county
basis. The decennial Census data and the Intercensal
County Estimates data sets were related by county
FIPS codes; the population in each BG/ED was as-
sumed to grow or shrink at the same rate as the county
population as a whole.
Since the concentration-response data are particu-
lar to specific sex and adult age groups, additional
population data were also required to determine the
sizes of affected subpopulations. For 1990 age and
sex, the U.S. Census, 1992 was used, with age groups
tallied as necessary. For 1980 age and sex, the U.S.
Census, 1982 was used, with age groups also tallied
as necessary. The 1970 age and sex breakdowns were
obtained through personal communication with the
Census Bureau.42 The age and sex percentages were
interpolated for intervening years.
Pregnant women are often a subpopulation of in-
terest for lead effects. Although pregnant women
themselves may be harmed by exposure to lead, this
analysis was concerned with pregnant women because
of possible effects on their fetuses who will be born
March
communication> Karl Kuellmer> Abt Associates and the Bureau of Census, Population, Age and Sex telephone staff,
G-27
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
and evince effects as young children. To estimate the number of exposed fetuses who were born during the years
of interest,43 birth rates for 1970, 1980 and 1990 were obtained from the Census Bureau.44 These birth rates
were used to interpolate for years between 1970 and 1980, and for the years between 1980 and 1990.
Results
For both the control and no-control scenarios, Table G-5 shows estimated lead emissions from electric
utilities, industrial processes, and industrial combustion. Tables G-6 and G-7 show the differences in health
impacts between the two scenarios (for industrial processes, industrial combustion and electric utilities only) for
the "forward-looking" and "backward-looking" analyses. The modeled population for each year is also pre-
sented.
Table G-5. Estimated Lead-Emissions froin Electric'UiJMesJndustrialProcesseslahd
Industrial Combustion
* ;V^/' '
j^'H-i"' "\ ^ y
Electric Utilities*
Control Scenario
Electric Utilities'
Industrial Processes Control
Scenario s
Indus trial Processes .,-"
Industrial Combustion - ••*•
Control Scenario
Industrial Combustioix , -^
•"""" ,j#HJ"'
*\K. * ,s, -J *J*>5'
"^ * x : , ^ u
-.-«-" ^ vv> ,
£&'
7,7'gsr
.,.-»•*" ".**<#>
'• , . 7;?89
A*!' ^4$3ff'
r?-**
V^F^iSTS '„
&>&'' ' i a*f ,
^/'V3Sl',
' ' . ,.•,**£*
\- '"*.'$$&»
.v*^Wt-'-
.»•¥•
"'."- ,^-^ta24-
/ ^ • ,
^?^^4-354
, 4r'1>'&fr2f*'V
'^:;,^,A$7*
-"""* -w^;-
",' '" ' ^;1980
-'" "' ',<$&,*
"' $&$'•
1,032"
'•-'-"*" ,.,,-i/t
"-"V,l,,^50
• ^'^4i,88G
'"'^ -4,653
<.'«!*" ,,„,,,
,/-,i9SS'""
~\;^wt
•, &/'#'' '
,'•><••"$&<)
^V-.**'*' '"
v-;, ,,670
'' J' ,-, ''' *
• ' ""''5,6>6,
' •'-^'";T4>'"
-' "-'"43*4
'^''""",,,V&H:.
"~^ ^:,,i^"'
-=-"^:^64
, ,<&>?>'•,
658>,
^''"^^05,
'/< '"•••
--""-';"/;187^
;';;;i'-X»59^
" Appropriate data on electric mOiies 'do
43 Note that we do not record the number of pregnancies, since the valuation only applies if the child is born and lives to exhibit
the effect. Neither are we concerned with whether the births are single or multiple births, since each fetus is at risk, whether a
pregnant women carries one or more fetuses.
44 Personal communication, Karl Kuellmer, Abt Associates and the Bureau of Census, Population, Fertility/Births telephone
staff.
~ " ' G-28 "
-------
Appendix G: Lead Benefits Analysis
f Dj&araura ml^w^^of;U^^m^^^^n^^n!Kol a#i$^''','
::JtadusttMJ&QC&^ (HqlMag.Qf^es LeagQ
1980
Women (45t7
1 ,'--• v -> ,-.
'^Total
f
0.1
0.8
mi .'
3.9
, 6.3-
1.4
1.8
'W
v—"Total
„-,» ..... 0.0,
- :;,-,;x«> ^.fi 1
\\;; : -0)2 '
Strokes-:
s'
P^74^: /;-,-
-'^..T.Qtal
, 0.1
- 0
,1,1 -,-
--•"'0.5
Y" 0.7 -
;;-,-„-,-», <,*••
--44 ,;^,w
-2.7 ;>«-,
0.9
44 >
L8
3,790'
630
.; , 3
;,v-60, "--120,,,
125
217
G-29
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table G-7. Yearly Differences in Number of Health Effects^elweett the" Oontrol;'ari(i,No-control- •<*<
Scenarios: Industrial Processes, Boilers, and '':
Constant 1990 Levels),
Health Effect - •' '
Mortality „ ... - *4
Men (40-54) . , ,"..,,<••>-*
Men (55-64)
Men (65-74) , , - *'•*'/„-
Women (45-74)
Infants ' - ~~ ,,
' Total
Coronary Heart Disease v
Men (40-54) ' ^,,,-
Men (55-64) •„" .
Men (65-74)
Women (45-74) - • ,'\ . *
Total
Strokes <*• • " ' '„,»->-<
Cerebrovascular Accident (men 45-74)'
Cerebrovascular Accident - .%„,,- •
(women 45-74) , ,, " •*>*Vs,,,y;,v
Brain Infarction (men 45-74) , r
Brain Infarction (women 45-74) v, ,
-- * S"- -Total
Hypertension (men 20-74) •...
IQ Decrement . - - ^ ' , ,,
Lost IQ Points v v.
IQ<70 (cases) s * :* '* , -"
- - ,
Population Exposed (millions^ s ,- f-v' '" • •
„,-•'-"'•
> 19»
'..-o!"-1
'^*',^->,- 0,3
'*r* ;,,,'0'.2
' : oTi, -
,*v ^:^ ^
" " ' „*'-
';),. '""&
•A/'.' L0.8 -
t- ~* 0-4
.\; "0,1,,,,
";"„>„,, -02,
*'f)L2L ,,,
' '" . ,. 0,9
:*'" '"^"
*"*" - 0.2 '
0:1"
,,,„ , //y '"
r' ••%;.,
' d.i *<
,;:;^.-..-o.-5,
'-'- " 422 "*'
— «-_ '„„,,,- =
" , 630'"'"
*rfW- '' 0
, ,,,,,v- "" ^ ^
;r,^,,;~,'i88
• '•; V"^'",", ,
1980 1985 ,;<^i^0: „,
• ' ,-,.'•' "' '>"-•"' ,'f "." '" -
' - • ; _ 6.9, • "" ' ., jCl .5 • " ' ^ , , ,12s-5 • ••
,5,1 ,a;3 ^; ; 8:'2 :--•
2.0 , -3.5'" "^ - 3,9
• 3^-^- - - - : ;;^5 >--'". 6;- - ^ •
„- , O'.Oni,, 'fIDJ32, ' 0.0(E: 'y"~
- »-v*lf .9» - •"- 29,7- :>- " ', 3,t 0
•\-j,^--,;/, -, >^;,/^-! ""* •'
& *%' ' t"^ R^ ^ 1jS^(l J
- 3.4, -;>,,;|,-.fi "* f :„,$,&
'/4.4-- ^ ' 7,6' - 8.0 *',f
-- •• 5JSL_, '%^/-' ,%| ",
'221' - •;• '36.6""""%^3|,3 - -
';:;,"'' 5^0 ., - -8/1 ;:^'V,,V8^ \^,
2.6, --;;;;, ,4/t-;:";/';, 4.2 -"-
" ' -,--- - ,,,, ,v-->-~ ' ' ",,<"-'. »••
' ! o ^ -^ "\ " f f'f j\ £, ^''/' ' ^ T s
-i«O •" ' 'vv^v^fr.O ' , •**. / •*•*' J
i £: ' *> *7 ^ *7
^ .n ' ^L-. / _, ^ jfr_. / •
•» •• V •• ' ' s f ' ' ^ •.^ '^/tf f
' ';,-12»0- -, -•-•
" IA inn ' T> 7nV\ 7^ Qhrt
-.-!<*, ouu ^•A(j>'/uy ^zj^yuu
- ' ~ ?3f ' ' ',s/ , , '„ ,- ,-"'3^ , 61 -''
.i*s^ - - ' V*-^""
- J"jt,7/,:;; ,,,2p7;3^:i;'v^l7 ,r
G-30
-------
Appendix G: Lead Benefits Analysis
Reduction in Health Effects
Attributable to Gasoline Lead
Reductions
Estimating Changes in Amount of Lead
in Gasoline from 1970 to 1990
The relationship between the national mean blood
lead level and lead in gasoline is calculated as a func-
tion of the amount of lead in gasoline consumed. Thus,
to calculate the health benefits from gasoline lead re-
ductions, necessary inputs are estimates of lead in
gasoline consumed over the period 1970 to 1990 and
the amount of lead in gasoline that would have been
consumed in the absence of the Clean Air Act. These
values are calculated using the quantity of both leaded
and unleaded gasoline sold each year and the con-
centration of lead in leaded and unleaded gasoline for
each year in the period of interest. For each year, the
relationship is expressed as:
, *'*'' -.--X \
where:
LEAD
SOLD =
Pb
Pb
leaded
Pb
unleaded
average lead per day in gasoline
sold in a given year (metric tons/
day),
total quantity of gasoline sold
(million gal/yr),
fraction of total gasoline sales
represented by leaded gasoline
(dimensionless),
lead content of leaded gasoline
(g/gal), and
lead content of unleaded gasoline
(g/gal).
Gasoline Sales (SOLD): Data on annual gasoline
sales were taken from a report by Argonne National
Laboratories (1993) which presented gasoline sales
for each state in five year intervals over the period
1970-1990. This analysis used linear interpolation to
estimate the gasoline sales for years between the re-
ported years. These data were summed to obtain na-
tional sales figures.
Fraction of Total Sales Comprised of Leaded Gaso-
line (FRACpb): For the control scenario, this analysis
used information reported by Kolb and Longo (1991)
for the fraction of the gasoline sales represented by
leaded gasoline for the years 1970 through 1988. For
1989 and 1990, data were taken from DOE (1990 and
1991, respectively). For the no-control scenario, all of
the gasoline sold was assumed to be leaded for all years.
Lead Content of Gasoline (Pb, . . and Pb , . J:
J l leaded unleaded'
Argonne National Laboratory in Argonne, Illinois was
the source for the data on the lead content of leaded
and unleaded gasoline for the period 1974-1990.
Argonne compiled these data from historical sales data
submitted to EPA, from Clean Air Act regulations on
lead content, and from recent Motor Vehicle Manu-
facturers Association (MVMA) surveys. For 1970
through 1973, this analysis assumed the lead content
of gasoline to be at the 1974 level. For the no-control
scenario, this analysis used the 1974 lead content in
leaded gasoline as the lead content in all gasoline for
each year.
Estimating the Change in Blood Lead
Levels from the Change in the Amount of
Lead in Gasoline
Several studies have found positive correlations
between gasoline lead content and blood lead levels.45
Data from the National Health and Nutrition Examina-
tion Survey (NHANESII) have been used by other re-
searchers who determined similar positive correlations
between gasoline lead and blood lead levels.46
The current analysis used a direct relationship be-
tween consumption of lead in gasoline and blood lead
levels to estimate changes in blood lead levels result-
ing from Clean Air Act regulation of the lead content
of gasoline. This relationship was based on regression
analyses of the reduction of leaded gasoline presented
in the 1985 Regulatory Impact Analysis (RIA).47 Sev-
eral multiple regressions were performed in the RIA to
relate gasoline usage with individuals' blood lead lev-
45 U.S. EPA, 1985; Billick et al., 1979; Billick et al., 1982.
46 Janney, 1982; Annest et al., 1983; Centers for Disease Control, 1993; National Center for Health Statistics, 1993b.
47 U.S. EPA, 1985.
G-31
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
els, which were taken from NHANES II. These re-
gressions of blood lead on gasoline usage controlled
for such variables as age, sex, degree of urbanization,
alcohol consumption, smoking, occupational expo-
sure, dietary factors, region of the country, educational
attainment, and income. The regressions suggested
that a decrease of 100 metric tons per day (MTD) of
lead used in gasoline is associated with a decrease in
mean blood lead concentration of 2.14 ug/dL for
whites and 2.04 ug/dL for blacks. In both of these
regressions, gasoline use was found to be a highly sig-
nificant predictor of blood lead (p < 0.0001).48
To determine a single gasoline usage-blood lead
slope for the entire population of the U.S., this analy-
sis used the average of the slopes for blacks and for
whites, weighted by the percentage of blacks and
whites in the U.S. during the time period of the analy-
sis.49 The resulting relationship is 2.13 ug/dL blood
lead per 100 metric tons of lead in gasoline consumed
per day. The same relationship was used to model
changes in both children's and adults' blood lead lev-
els. The U.S. EPA (1985) analyzed data from a study
of black children hi Chicago during the tune period
1976 to 1980 and determined a slope of 2.08 ug/dL
per 100 MTD. This slope for children is very similar
to the one used in this analysis.
1970-Forward and 1990-Backward Approaches
As with the industrial processes and boilers analy-
sis, this analysis used two different approaches to de-
termine mean blood lead levels based on changes in
lead concentrations hi gasoline. In the 1970-forward
approach, the calculations began with the estimated
blood lead level for 1970. The change in blood lead
level from one year to the next was based upon the
change in the amount of lead in gasoline sold, as dis-
cussed above, for both the control and no-control sce-
narios. For example, to calculate the blood lead level
for 1971, the calculated change in blood lead from
1970 to 1971 was added to the 1970 value. This pro-
cess was repeated for each succeeding year up to 1990.
The 1990-backward approach began with a mean
blood lead level in 1990 for the control scenario. For
the no-control scenario, the starting blood lead was
estimated from the 1990 level used hi the control sce-
nario, plus an additional blood lead increment result-
ing from the difference between the 1990 consump-
tion of lead in gasoline under the two scenarios. Again,
the difference in mean blood lead levels from one year
to the next was based on the change in gasoline lead
for the corresponding years. For example, the differ-
ence In blood lead levels between 1990 and 1989 was
subtracted from the 1990 level to determine the 1989
level. The process was continued for each year back
to 1970.
Relating Blood Lead Levels to Population Health
Effects
The mean blood lead levels calculated using the
methods described above were used in the dose-re-
sponse functions for various health effects (e.g., hy-
pertension, chronic heart disease, mortality). This in-
formation was then combined with data on the resi-
dent population of the 48 conterminous states in each
year to determine the total incidence of these health
effects attributable to lead in gasoline. A Department
of Commerce Publication (1991) was used to obtain
the total population in 1970, 1980, and 1983-1990,
while a different publication was the source of the
1975 population values.50 Linear interpolation was
used to estimate the populations in years for which
specific data were not available.
For certain health effects, it was necessary to know
the size of various age groups within the population.
Two different sources were used to estimate the pro-
portions of the population in the age groups of inter-
est. A U.S. Census summary (U.S. Dept. of Com-
merce, 1990) was used for information for 1990 for
children and adults and for 1980 for adults, and Cen-
sus Telephone Staff (U.S. Dept. of Commerce, 1994)
provided information for 1980 for children and 1970
for children and adults. The populations for the inter-
vening years were estimated by linear interpolation.
Changes in Leaded Gasoline Emissions and
Resulting Decreased Blood Lead Levels and Health
Effects
Table G-8 shows the estimated quantity of lead
burned in gasoline in the five year intervals from 1970
to 1990. Tables G-9 and G-10 show the difference in
48 U.S. EPA, 1985.
49 U.S. Department of Commerce, 1992. Although the percentages of blacks and whites changed slightly over this time period
(1970-1990), the change did not affect the value of the weighted slope.
30 U.S. Dept. of Commerce, 1976.
G-32
-------
Appendix G: Lead Benefits Anafysis
health impacts between the two scenarios (for lead in
gasoline only) for the "forward-looking" and "back-
ward-looking" analyses. In general, health effect ben-
efits resulting from gasoline lead reductions exceed
those predicted from lead reductions at the point
sources examined (i.e., industrial processes and boil-
ers and electric utilities) by three orders of magni-
tude.
, ,;,^;,,( ,->H;-^,^\
CoHtrOLScenadii-^'*',,,,, ,'
'- ^ '
No-cgntirol Scenario "•-„
• iw?'>;
"17^100
-174100 '•
1975
- 179,200'
--202:600-
"" ^ SJ^ "^
y 86',4'OQ", L-22.0C
v-", 206,900--'"- 2H40
< im
K)"' 2,300'"
,0,' 322,'900 '
G-33
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Health Effect
Mortality
Men (40-54)
Men (55-64)
Men (65-74)
Women (45-74)
Infants
Total
Coronary Heart Disease
Men (40-54)
Men (55-64)
Men (65-74)
Women (45-74)
* Total
Strokes „•"*'„•
Cerebrovascular Accident (men 45-74)
Cerebrovascular Accident . .^>
(women 45-74)
Brain Infarction (men 45-74) ,. f,
Brain Infarction (women 45-74),
Hypertension (men 20-74)
IQ Decrement
Lost IQ Points
IQ<70 (cases)
Population Exposed (millions) , ,
;««- - 1985
' 8"f ' '-,, * ' 520 v
., 12,400'
».. - • -1,220
» •
442*"'••"'''«& ,/" 965
'"*"' <..-'«"'"
, 85
'.^677,000 -w'4.200,000 "7340,000- 9,740^00
'" 36,500
-, 237-;
J
G-34
-------
Appendix G: Lead Benefits Analysis
Table CM& Year^'Biffereaces it). Number of JIeal|h EfffcfsBetwe^i~the,Co|itro] asdl^
Sdhtrol SoeWIo^;£eaa4f!-akwiirte- only "(Holding Other UaiTSb'ttrces at-Coflfeint 1990vw
L^S):-- *:^v';; --w,v.;:*!;!," ' "^!^^W7^ """^^c."'!"^^^^
-, "*'--';•?»,-• "* -.>;'/-*, ' '' '"''"•*>,''' ~H-,» - - - ' A" ,--'j-^.^'" ' , ,„, ' v';j- '"-i,^
1975
1980
1985 1990
'/- 7,950
Men(65-74,X
,,,342 ; > 2,250,
; -2,480
,! 4,030
\; ,1:649
Ro^priary Heait Dfseasfe • -' -vv -,, v:,
J '%£*'''> ' ' ^^ , V * V" * ^ J "^
'.S^W,- aw™, , - V,^, ---,-" 4
280' *tvw , 4,'690 - - - - *".<&£ JO
Total
.:, 1^20; - 2,579; -
'-^-^aiQ ^'*J^9CU
^- f -787 '""5,180 • -.40.700, ,''- '13.-900
225
,,,v., ^4?:^)';,.',. •'-?,«
"Brim1Infarction ^mai'^74) """"
Brain' Mfarctioi''(women 4:
" ' Total
_740
*'' >-V>"
"837;
,3,514"
'f', „ 3,720
'**
;: - g,890
"' 984,000" '
42:300.000 t5;6001
Jiost IQ joints - - -,-'''«-•*-•>- - •«,.,,
1;03Q,OQO
.^ 8,580^00- 10,400,000
-^-36,500 "" °'45tr
PopulS.tiipii Bxp6Bed-{millions->C"'"''
. ,,V ' Wfl.,,. ,™ - •> ;v^ / ,,^-, ;iX' V -,
,
.225
'
<• ,
^; " "24T
G-35
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Lead Benefits Analysis
References
Abt Associates, Inc. 1992. The Medical Costs of Five
Illnesses Related to Exposure to Pollutants.
Prepared for: Nicholas Bouwes, Regulatory
Impacts Branch, Economics and Technology
Division, Office of Pollution Prevention and
Toxics, U.S. Environmental Protection
Agency, Washington, D.C.
Abt Associates, Inc. 1995. The Impact of the Clean
Air Act on Lead Pollution: Emissions Reduc-
tions, Health Effects, and Economic Benefits
From 1970 to 1990, Draft. Prepared for Eco-
nomic Analysis and Innovations Division,
Office of Policy Planning and Evaluation,
U.S. EPA. January 19.
Annest, J.L., J.L. Pirkle, D. Makuc, J.W. Neese, D.D.
Bayse, and M.G. Kovar. 1983. "Chronologi-
cal Trend in Blood Lead Levels Between 1976
and 1980." New England Journal of Medi-
cine 308:1373-1377.
Argonne National Laboratories (Argonne). 1993.
National Gasoline Sales Data, 1970-1990.
Ashenfelter, O. and J. Ham. 1979. "Education, Un-
employment and Earnings." /. Political
Economy 87(5): S99-S131.
Azar, R.D., et al. 1975. An Epidemiologic Approach
to Community Air Lead Exposure Using Per-
sonal Air Samplers. In: Griffin, T.B. and
Knelson, J.H., eds. Lead. Stuttgart, West Ger-
many: GeorgThieme Publishers; pp .254-290.
(Coulston, F. and Korte, K, eds. Environmen-
tal Quality and Safety: Supplement V. 2).
Bellinger, D., J. Sloman, A. Leviton, M. Rabinowitz,
H.L. Needleman, and C. Waternaux. 1991.
"Low-level Lead Exposure and Children's
Cognitive Function in the Preschool Years."
Pediatrics 87(2): 219-227.
Bellinger, D.C. 1992. "Lead Exposure, Intelligence
and Academic Achievement. Pediatrics
90(6): 855.
Billick, I.H., A.S. Curran, and D.R. Shier. 1979.
"Analysis of Pediatric Blood Lead Levels in
New York City for 1910-1916."Environmen-
tal Health Perspectives 31: 183-190.
Billick, I.H., etal. 1982. Predictions of Pediatric Blood
Lead Levels from Gasoline Consumption. U.S.
Department of Housing and Urban Develop-
ment. [Cited in U.S. EPA, 1985.]
Brunekreef, B.D., et al. 1981. "The Arnhem Lead
Study: 1. Lead Uptake by 1- to 3-year-old
Children Living in the Vicinity of a Second-
ary Lead Smelter in Arnhem, the Nether-
lands." Environ. Res. 25: 441-448.
Centers for Disease Control (CDC). 1985. Prevent-
ing Lead Poisoning in Young Children. U.S.
Department of Health and Human Services,
Public Health Service, Centers for Disease
Control, Atlanta, GA.
Centers for Disease Control (CDC). 1991a. Strategic
Plan for Elimination of Childhood Lead Poi-
soning. U.S. Department of Health and Hu-
man Services, Public Health Service, Centers
for Disease Control, Atlanta, GA. February.
Centers for Disease Control (CDC). 1991b. Prevent-
ing Lead Poisoning in Young Children. U.S.
Department of Health and Human Services,
Public Health Service, Centers for Disease
Control, Atlanta, GA. October.
Centers for Disease Control (CDC). 1993. Personal
communication between Abt Associates and
Jim Pirkle. November 16.
Chamberlain, A.C. 1983. "Effect of Airborne Lead
on Blood Lead."Atmos. Environ. 17:677-692.
Daines, R.H., et al. 1972. "Air Levels of Lead Inside
and Outside of Homes." Ind. Med. Surg. 41:
26-28.
Dietrich, K.N., K.M. Krafft, R. Shukla, R.L.
Bornschein, and P.A. Succop. 1987. The
Neurobehavioral Effects of Prenatal and Early
Postnatal Lead Exposure. In: Toxic Sub-
stances and Mental Retardation:
Neurobehavioral Toxicology and Teratology,
G-36
-------
Appendix G: Lead Benefits Analysis
S.R. Schroeder, Ed. American Association of
Mental Deficiency, Washington, DC, pp. 71-
95 (Monograph No. 8).
Elixhauser, A., R. M. Andrews, and S. Fox. 1993.
Clinical Classifications for Health Policy
Research: Discharge Statistics by Principal
Diagnosis and Procedure. Agency for Health
Care Policy and Research, Center for Gen-
eral Health Services Intramural Research,
U.S. Department of Health and Human Ser-
vices.
Environmental Law Institute (ELI). 1992. Projecting
With and Without Clean Air Act Emissions
for the Section 812 Retrospective Analysis:
A Methodology Based upon the Projection
System Used in the OAQPS "National Air
Pollutant Emission Estimates: Reports."
[Jorgenson/Wilcoxen Model Projections], Jim
Lockhart.
Fugas, M., et al. 1973. "Concentration Levels and
Particle Size Distribution of Lead in the Air
of an Urban and an Industrial Area as a Basis
for the Calculation of Population Exposure."
In: Earth, D., et al. eds. Environmental Health
Aspects of Lead: Proceedings, International
Symposium; October 1972; Amsterdam, The
Netherlands. Luxembourg: Commission of
the European Communities, pp. 961-968.
Goldsmith, J.R. 1974. Food Chain and Health Impli-
cations of Airborne Lead. Sacramento, CA:
State of California, Air Resources Board;
Report No. ARB-R-102-74-36.
Griliches, Zvi. 1977. "Estimating the Returns to
Schooling: Some Econometric Problems."
Econometrica 45:1-22.
Hasselblad, V. 1995. Personal Communication be-
tween V. Hasselblad and Abt Associates, Feb-
ruary 28, 1995.
Janney, A. 1982. The Relationship Between Gasoline
Lead Emissions and Blood Poisoning in
Americans. Prepared for U.S. EPA, Office of
Policy Analysis. [Cited in U.S. EPA, 1985 ]
Johnson, D.E., et al. 1976. Base Line Levels of Plati-
num and Palladium in Human Tissue. U.S.
EPA, Health Effects Research Laboratory,
Research Triangle Park, N.C. EPA-600/1-76-
019.
Kakalik, J., et al. 1981. The Cost of Special Educa-
tion. Rand Corporation Report N-1791-ED.
Kolb, J. and K. Longo. 1991. Memorandum to Joel
Schwartz, U.S. EPA, Washington, DC, No-
vember 5.
Krupnick, AJ. and M.L. Cropper. 1989. Valuing
Chronic Morbidity Damages: Medical Costs
and Labor Market Effects. Draft Final Report
to U.S. Environmental Protection Agency,
Office of Policy Planning and Evaluation.
June 26.
Landrigan, P.J., et al. 1975. "Epidemic Lead Absorp-
tion near an Ore Smelter: the Role of Particu-
late Lead." N. Engl. J. Med. 292: 123-129.
Landrigan, PJ. and E.L. Baker. 1981. "Exposure of
Children to Heavy Metals from Smelter: Epi-
demiology and Toxic Consequences."
Environ. Res. 25: 204-224.
McGee and Gordon. 1976. The Results of the
Framingham Study Applied to Four Other
U.S.-based Epidemiologic Studies of Coro-
nary Heart Disease. The Framingham Study:
An Epidemiological Investigation of Cardio-
vascular Disease. Section 31, April.
National Center for Health Statistics (NCHS). 1993a.
Facsimile received by Abt Associates from
Margaret McDowell regarding the types of
laboratory tests conducted during NHANES
I. December 14.
National Center for Health Statistics (NCHS). 1993b.
Personal communication between Abt Asso-
ciates and NCHS Public Information Special-
ist. November 3.
National Energy Accounts, Bureau of Economic
Analysis.
G-37
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Needleman, H.L., et al. 1990. "The Long Term Ef-
fects of Exposure to Low Doses of Lead in
Children." New England Journal of Medicine
322(2): 83-88.
NHANES, National Health and Nutrition Examina-
tion Survey.
NHANES n, National Health and Nutrition Exami-
nation Survey, 1976-1980.
Nordman, C.H. 1975. Environmental Lead Exposure
in Finland: a Study on Selected Population
Groups [dissertation]. Helsinki, Finland: Uni-
versity of Helsinki.
Oliver, T. 1911. "Lead Poisoning and the Race." Brit-
ish Medical Journal 1(2628): 1096-1098.
[Cited in USEPA (1990).]
Piomelli et al. 1984. "Management of Childhood Lead
Poisoning." Pediatrics 4: 105.
Pirkle, J.L., J. Schwartz, J.R. Landis, and W.R. Harlan.
1985. "The Relationship Between Blood Lead
Levels and Blood Pressure and its Cardiovas-
cular Risk Implications." American Journal
of Epidemiology 121: 246-258.
Pirkle, J. L., et al. 1994. "Decline in Blood Lead Lev-
els in the United States, the National Health
and Nutrition Examination Survey
(NHANES)." JAMA 272(4): 284.
Pooling Project Research Group. 1978. "Relationship
of Blood Pressure, Serum Cholesterol, Smok-
ing Habit, Relative Weight and ECG Abnor-
malities to Incidence of Major Coronary
Events: Final Report of the Pooling Project."
Journal of Chronic Disease. Vol. 31.
Rabinowitz, M., D. Bellinger, A. Leviton, H.
Needleman, andS. Schoenbaum. 1987. "Preg-
nancy Hypertension, Blood Pressure During
Labor, and Blood Lead Levels." Hyperten-
sion 10(4): October.
Salkever, D.S. 1995. "Updated Estimates of Earnings
Benefits from Reduced Exposure of Children
to Environmental Lead." Environmental Re-
search 70: 1-6.
Schwartz, J. 1988. "The Relationship Between Blood
Lead and Blood Pressure in the Nhanes II
Survey "EnvironmentalHealth Perspectives.
78:15-22.
Schwartz, J. 1990. "Lead, Blood Pressure, and Car-
diovascular Disease in Men and Women."
Environmental Health Perspectives, in press.
Schwartz, J. 1992a. "Blood Lead and Blood Pressure:
a Meta-analysis." Presented at the Annual
Meeting of Collegium Ramazzini. November.
Schwartz, J. 1992b. "Chapter 13: Lead, Blood Pres-
sure and Cardiovascular Disease." In: Human
Lead Exposure, H. L. Needleman, Ed. CRC
Press.
Schwartz, J. 1993. "Beyond LOEL's, p Values, and
Vote Counting: Methods for Looking at the
Shapes and Strengths of Associations."
Neurotoxicology 14(2/3): October.
Shurtleff, D. 1974. Some Characteristics Related to
the Incidence of Cardiovascular Disease and
Death. The Framingham Study: An Epidemio-
logical Investigation of Cardiovascular Dis-
ease. Section 30, February.
Silbergeld, E.K., J. Schwartz, and K. Mahaffey. 1988.
"Lead and Osteoporosis: Mobilization of Lead
from Bone in Postmenopausal Women." En-
vironmental Research 47: 79-94.
Snee, R.D. 1981. "Evaluation of Studies of the Rela-
tionship Between Blood Lead and Air Lead."
Int. Arch. Occup. Environ. Health 48: 219-
242.
Taylor, T.N., P.H. Davis, J.C. Tomer, J. Holmes, J.W.
Meyer, and M. F. Jacobson. 1996. "Lifetime
Cost of Stroke in the United States." Stroke
27(9): 1459-1466.
Tepper, L.B. and L.S. Levin. 1975. "A Survey of Air
and Population Lead Levels in Selected
American Communities." In: Griffin, T.B.;
Knelson, J.H., eds. Lead. Stuttgart, West Ger-
many: Georg Thieme Publishers; pp. 152-196.
(Coulston, F.; Korte, f., eds. Environmental
Quality and Safety: Supplement v. 2).
G-38
-------
Appendix G: Lead Benefits Analysis
Tsuchiya, K., et al. 1975. "Study of Lead Concentra-
tions in Atmosphere and Population in Japan."
In: Griffin, T.B. and Knelson, J.H.. eds. Lead.
Stuttgart, West Germany: Georg Thieme Pub-
lishers; pp.95-145. (Coulston, F.; Korte, F.,
eds/ Environmental Quality and Safety:
Supplement v. 2)
U.S. Census. 1982. United States Summary, General
Population Characteristics, Table 41: Single
Years of Age by Race, Spanish Origin, and
Sex: 1980.
U.S. Census. 1992. United States Summary, General
Population Characteristics, Table 13: Single
Years by Sex, Race, and Hispanic Origin:
1990.
U.S. Department of Commerce. 1976. Statistical Ab-
stract of the United States: 95th Edition. Bu-
reau of the Census. Washington, DC.
U.S. Department of Commerce. 1980. U.S. Census,
United States Summary, General Population
Characteristics.
U.S. Department of Commerce. 1986. Statistical Ab-
stract of the United States: 105th Edition.
Bureau of the Census. Washington, DC.
U.S. Department of Commerce. 1987. Census of
Manufacturers.
U.S. Department of Commerce. 1990. Earnings by
Occupation and Education: 1990. Bureau of
the Census. Washington, DC.
U.S. Department of Commerce. 1990. U.S. Census,
United States Summary, General Population
Characteristics.
U.S. Department of Commerce. 1991. Statistical Ab-
stract of the United States: lllth Edition.
Bureau of the Census. Washington, DC.
U.S. Department of Commerce. 1992. Statistical Ab-
stract of the United States: 112th Edition.
Bureau of the Census. Washington, DC.
U.S. Department of Commerce. 1993. Money Income
of Households, Families, and Persons in the
United States: 1992. Bureau of the Census,
Series P60-184.
U.S. Department of Commerce. 1993. Personal Com-
munication between Bureau of Census, Popu-
lation, Age and Sex Telephone Staff and Karl
Kuellmer of Abt Associates on December 8,
1993.
U.S. Department of Commerce. 1994. City and County
Databook: 1994. Bureau of the Census. Wash-
ington, DC.
U.S. Department of Commerce. 1994. Personal Com-
munication between Bureau of Census, Popu-
lation, Age and Sex Telephone Staff and Karl
Kuellmer of Abt Associates on February 7,
1994.
U.S. Department of Education. 1993. Digest of Edu-
cational Statistics, 1993. National Center for
Educaton Statistics, Office of Educational
Research and Improvement. D.Ed. publica-
tion number NCES 93-292.
U.S. Department of Energy (DOE). 1990. Petroleum
Supply Annual, 1989, Volume 1. DOE publi-
cation number EIA-0340(89)/1
U.S. Department of Energy (DOE). 1991. Petroleum
Supply Annual, 1990, Volume 1. DOE publi-
cation number EIA-0340(90)/1
U.S. Department of Energy (DOE). 1992. Cost and
Quality of Fuels for Electric Utility Plants
1991. DOE/EIA-0191(91) Energy Informa-
tion Administration, August.
U.S. Environmental Protection Agency (U.S. EPA).
1984. A Survey of the Literature Regarding
the Relationship Between Measures oflQ and
Income. Prepared by ICF, Inc. Report to U.S.
Environmental Protection Agency, Office of
Policy Analysis, June.
U.S. Environmental Protection Agency (U.S. EPA).
1985. Costs and Benefits of Reducing Lead
in Gasoline: Final Regulatory Impact Analy-
sis. Prepared by U.S. Environmental Protec-
tion Agency, Office of Policy Analysis, Eco-
nomic Analysis Division. February.
G-39
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
U.S. Environmental Protection Agency (U.S. EPA).
1986a. Reducing Lead in Drinking Water: A
Benefit Analysis. Prepared by U.S. Environ-
mental Protection Agency, Office of Policy
Planning and Evaluation, Draft Final Report.
December.
U.S. Environmental Protection Agency (U.S. EPA).
19S6b.Air Quality Criteria for Lead: Volume
III. Environmental Criteria and Assessment
Office, Research Triangle Park, NC. EPA-
600/8-83/028cF. June.
U.S. Environmental Protection Agency (U.S. EPA).
1987. Methodology for Valuing Health Risks
of Ambient Lead Exposure. Prepared by
Mathtech, Inc. for U.S. Environmental Pro-
tection Agency, Office of Air Quality Plan-
ning and Standards, Ambient Standards
Branch, Contract No. 68-02-4323.
U.S. Environmental Protection Agency (U.S. EPA).
1990a. AIRS Facility Subsystem Source Clas-
sification Codes and Emission Factor List-
ing for Criteria Air Pollutants. Office of Air
Quality Planning and Standards, Research
Triangle Park, NC. EPA-450/4-90-003.
March.
U.S. Environmental Protection Agency (U.S. EPA).
1990b. National Air Pollutant Emission Esti-
mates 1940-1988. Office of Air Quality Plan-
ning and Standards, Technical Support Divi-
sion, National Air Data Branch. Research
Triangle Park, NC. EPA-450/4-90-001.
U.S. Environmental Protection Agency (U.S. EPA).
1990c. Review of the National Ambient Air
Quality Standards for Lead: Assessment of
Scientific and Technical Information. OAQPS
Staff Paper, Air Quality Management Divi-
sion, Research Triangle Park, NC. December.
U.S. Environmental Protection Agency (U.S. EPA).
1991a. National Air Quality and Emissions
Trends Report, 1989. Office of Air Quality
Planning and Standards. Research Triangle
Park, NC. EPA-450/4-91-003.
U.S. Environmental Protection Agency (U.S. EPA).
1991b. The Interim Emissions Inventory. Of-
fice of Air Quality Planning and Standards,
Technical Support Division, Source Recep-
tor Analysis Branch. Research Triangle Park,
NC.
U.S. Environmental Protection Agency (U.S. EPA).
1992. 7990 Toxics Release Inventory. Office
of Pollution Prevention and Toxics, Washing-
ton, DC. EPA-700-S-92-002.
U.S. Environmental Protection Agency (U.S. EPA).
1994. Guidance Manual for the Integrated Ex-
posure Uptake Biokinetic Model for Lead in
Children. February. EPA 540-R-93-081.
U.S. Environmental Protection Agency (U.S. EPA)
database. Graphical Exposure Modeling Sys-
tem Database (GEMS).
Wallsten and Whitfield. 1986. Assessing the Risks to
Young Children of Three Effects Associated
with Elevated Blood Lead Levels. Argonne
National Laboratory. December.
Wittels, E.H., J.W. Hay, and A.M. Gotto, Jr. 1990.
"Medical Costs of Coronary Artery Disease
in the United States," The American Journal
of Cardiology 65: 432-440.
G-40
-------
Appendix H: Air Toxics
Introduction
Air toxics are defined as air pollutants other than
those six criteria pollutants for which EPA sets ac-
ceptable concentrations in ambient air. The SARA 313
Toxic Release Inventory (TRI), covering 328 of the
approximately 3000 potentially hazardous compounds
detected in air, estimated that approximately 1.2 mil-
lion tons of ah- toxics were released to the atmosphere
in 1987 from U.S. stationary sources alone. While the
TRI estimate tends to understate emissions of toxics
for a number of reasons, it does show that large quan-
tities of toxics are emitted into the atmosphere annu-
ally.
Effects of ah- toxics emissions are divided into
three categories for study and assessment: cancer;
"noncancer" effects, e.g. a wide variety of serious
health effects such as abnormal development, birth
defects, neurological impairment, or reproductive
impairment, etc.; and ecological effects. Each year,
these air toxics emissions contribute to significant
adverse effects on human health, human welfare, and
ecosystems. In EPA's 1987 Unfinished Business Re-
port1 cancer and noncancer air toxics risk estimates
were considered sufficiently high, relative to risks
addressed by other EPA programs, that the air toxics
program area was among the few rated "high risk".
Limited Scope of this
Assessment
The effects of air toxics emissions are difficult to
quantify. The adverse health effects of toxics are of-
ten irreversible, not mitigated or eliminated by reduc-
tion in ongoing exposure, and involve particularly
painful and/or protracted disease. Therefore these ef-
fects are not readily studied and quantified in human
clinical studies, in contrast to, for example, ambient
ozone. In addition, epidemiological studies of effects
in exposed populations are often confounded by si-
multaneous exposure of subjects to a variety of pol-
lutants. Therefore, the effects of air toxics are often
quantified by extrapolating data from animal studies
to human exposure and expressed as risk per unit of
exposure. Incidence of noncancer effects, for example,
often are difficult to translate into monetized benefits.
Similarly, the quantification of ecological effects
due to emissions of ah- toxics is hampered by lack of
sufficient information regarding contribution of
sources to exposure, associations between exposure
to mixtures of toxics and various ecological endpoints,
and economic valuation for ecological endpoints.
The ah" toxics portion of this study is, of neces-
sity, separate and more qualitative in nature than the
benefit analysis conducted for the criteria air pollut-
ants. Limitations in the quantitative analyses of air
toxics effects led the Project Team to decide to ex-
clude the available quantitative results from the pri-
mary analysis of CAA costs and benefits. Table H-l
presents the range of potential human health and eco-
logical effects that can occur due to ah" toxics expo-
sure. As indicated, this appendix presents quantita-
tive estimates of benefits of CAA air toxics control
for the cancer mortality endpoint for only nonutility
stationary source and mobile source categories.
Noncancer effects and ecological effects are described
qualitatively.
1 U.S. EPA. Office of Policy Planning and Evaluation. Unfinished Business: A Comparative Assessment of Environmental
Problems. February 1987.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table H-l. Health and Welfare Effects of Hazardous Air Pollutants/,,
Effect Category
Hum an Health
Human Welfare
Ecological
Other Welfare
Quantified Effects
Cancer Mortality
- nonutility stationary
source
- mobile source ^
ff
- / y
, «, *-
Unqualified Effects f
Caacer Mortality * , ^
- utijUy source,/'
-areaso"urcH ' (. >
Noncancer effects " * •>
V?*'- , , ,";''
- neurological
--'respiratory ' *7#
- repro'dueiiye;
- •hemfiopoelJC^, „. t " , <
"•- develop mental -
- anmunological * , /
- 'organ toxjcity, ' ^ , *'
Decreased income and
; -recreation, ' , ' ',",
' * . 4 * -V^V
opportunities due to >
fish, advisories' % V
Odors 'v"I '?* "•"' <
Effects/on wildlife ' ,, ,
Effects on plants'^ •'
Ecosystem effects » '
Loss of biological!
.^diversity ; "(/, ;- ,
> , ^ ' **
Visibility ; , ^"\:i,
Materials Damage?'
# v ^ ,"'" f
- - /< >; •- "i / '-"1
.Other Possible^ Effects /
*^^ ,„ f, * f ft ' stt,*
~ ' / */*-//
s »'' ' ^ ?'
^ # '
s f V f f •"&
> >' 4, '% ' , ^
•'•'•' Sty' * /v
',.-. - --"''. '•""7'-'
/^" ^^ ^ ^ ' ^
""/'-"' y ' X'
* '' » *•»'"'/,« ' >f/<,
? / Vv
/<:, ,/' "*, ' * , «
/ ^ ^
! ^,'/^' x, X' "
, DeereasetijnjEome resulting
'x |rpm decrefsed physical,;
' performance , //"'' f
v ^
, ,»^> * ' /
Effects.OA global^Iihtate'^
',; '.•"'^-•'V
»• * ' ?
, , / / /
vy^ ^ ^ * •> ^ ? ^
*^.*" x ' ' ^ - •>-
' 7 ^ '
-V ', v ,/
'• "' ' '/,'*,'&' :,,'<• '
'* r<<°%%>
5 * ^ M
V/'
History of Air Toxics Standards
under the Clean Air Act of 1970
The 1970 Clean Air Act required the EPA to list a
chemical as a hazardous air pollutant if it met the leg-
islative definition provided:
"The term 'hazardous air pollutant' means
an air pollutant to which no ambient air
quality standard is applicable and which in
the judgment of the Administrator may cause,
or contribute to, an increase in mortality or
an increase in serious irreversible, or
incapacitating reversible, illness."2
Once a HAP was listed, the EPA Administrator
was required to:
2 42 U.S.C. §1857(a)(l).
3 42 U.S.C. §1857(b).
"establish any such standard at the level
which in his judgment provides an ample
margin of safety to protect the public health
from such hazardous air pollutant."3
In other words the EPA had to first determine that
a chemical was a HAP, and then regulate the emis-
sions of each HAP based solely on human health ef-
fects and with an ample margin of safety. This regu-
latory mandate proved extremely difficult for EPA to
fulfill, for reasons discussed below, and the result was
that only seven HAPs were regulated over a period of
20 years.
Listing chemicals became a difficult task because
of debates within and outside of the EPA surrounding
issues of how much data are needed and which meth-
-------
Appendix tt: Air Toxics
odologies should be used to list a chemical as a HAP.
An even more difficult issue was how to define the
Congressional mandate to provide an "ample margin
of safety." For carcinogens, there is generally no
threshold of exposure considered to be without risk.
What level of risk, then, is acceptable, and how should
it be calculated? The EPA struggled to provide an-
swers to these questions, and was challenged in court.
The end result was a 1987 ruling by the D.C. Circuit
Court that provided the EPA with a legal framework
with which to determine an "ample margin of safety."
This framework was interpreted and used by the EPA
in its 1989 benzene regulations.
Quantifiable Stationary Source
Air Toxics Benefits
One might be tempted to presume that the few
federal HAP standards set would have achieved rela-
tively substantial reductions in quantifiable risk. While
some standards set under section 112 of the Clean Air
Act appear to have achieved significant reductions in
cancer incidence, the coverage, quantification, and
monetization of the full range of potential adverse
effects remains severely limited. This fact serves to
highlight the inadequacy of current methods of evalu-
ating HAP control benefits. This limited ability to es-
timate the total human health and ecological benefits
of HAP reductions is an important area for future re-
search. Thus the quantifiable benefits for CAA air
toxics control presented here are limited in scope.
There are three sources of information that pro-
vide a picture of potential stationary source air toxics
benefits of the CAA. EPA's Cancer Risk studies at-
tempted to broadly assess the magnitude and nature
of the air toxics problem by developing quantitative
estimates of cancer risks posed by selected air toxics
and their sources. Secondly, risk assessments con-
ducted in conjunction with the promulgation of Na-
tional Emissions Standards for Hazardous Air Pollut-
ants (NESHAPs) offer a snapshot of potential mon-
etized cancer mortality benefits. Finally, the Project
Team attempted to estimate historical non-utility sta-
tionary source HAP-related direct inhalation cancer
incidence reductions. Results from each of these stud-
ies are presented below.
EPA Analyses of Cancer Risks from
Selected Air Toxic Pollutants
The Agency conducted two efforts to broadly as-
sess the magnitude and nature of the air toxics prob-
lem. The 1985 report entitled, "The Air Toxics Prob-
lem in the United States: An Analysis of Cancer Risks
for Selected Pollutants"4 otherwise known as the "Six
Month Study," was intended to serve as a "scoping"
study to provide a quick assessment of the air toxics
problem utilizing only readily available data on com-
pound potencies, emissions, and ambient pollutant
concentrations. The Agency updated this analysis of
cancer risks in the 1990 report entitled "Cancer Risk
from Outdoor Exposure to Air Toxics" referred to here
as the " 1990 Cancer Risk study."5
For the pollutant and source categories examined,
the 1990 Cancer Risk study estimated the total na-
tionwide cancer incidence due to outdoor concentra-
tions of air toxics to range from 1,700 to as many as
2,700 excess cancer cases per year, with 14 compounds
accounting for approximately 95 percent of the an-
nual cancer cases. Additionally, point sources con-
tribute 25 percent of annual cases and area sources
contribute 75 percent of annual cases. Mobile sources
account for 56 percent of the nationwide total.6
The Six Month study indicates that the criteria air
pollutant programs appear to have done more to re-
duce air toxics levels during the 1970 to 1990 period
than have regulatory actions aimed at specific toxic
compounds promulgated during the same period.
Metals and polynuclear compounds usually are emit-
ted as particulate matter and most of the volatile or-
ganic compounds are ozone precursors. As such, they
are regulated under State Implementation Plan (SIP)
and New Source Performance Standard (NSPS) pro-
grams and Title II motor vehicle regulations. A num-
ber of reports cited indicate significant reductions in
air toxics emissions attributable to actions taken un-
D. , -- - °fflf,e of Air Quality Planning and Standards. The Air Toxics Problem in the United States: An Analysis of Cancer
Risks for Selected Pollutants. May 1985. EPA-450/ 1-85-001.
™ A X^ffice °f Air Quality Panning and Standards. Cancer Risk from Outdoor Exposure to Air Toxics. September 1990
JbrA-4DO/l-90-U04a.
6 The 1990 Cancer Risk study reported approximately 500 - 900 more cancer cases per year than the Six Month Study due
primarily to the inclusion of more pollutants, better accounting of emissions sources, and, in some cases, increases in unit risk
estimates.
H-3
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
der SIP, NSPS and mobile source programs. Addi-
tionally, EPA conducted a comparison of air quality
and emissions data for 1970 with the estimates of can-
cer incidence for 1980.7 Methods, assumptions and
pollutants included were held constant over the pe-
riod. The analysis showed a significant decrease in
incidence during the decade due to improvements in
air quality, presumably related to general regulatory
programs. For the 16 pollutants studied, estimated
nationwide cancer incidence decreased from 3600 in
1970 to 1600 in 1980. The 1990 Cancer Risk Study
did not attempt to update this analysis.
Although it is difficult to draw quantitative con-
clusions from these two studies regarding the ben-
efits of CAA air toxics control, it is apparent that the
pollutant-specific and source category-specific
NESHAPs were not structured to reduce significant
air toxic emissions from area and mobile sources. In
fact, the 1990 Cancer Risk Study indicates that con-
siderable cancer risk remained prior to passage of the
1990 CAA Amendments: as many as 2,700 excess
cancer cases annually. Some studies indicate that the
criteria air pollutant program played a critical role
during the 1970 to 1990 period in achieving air toxic
emission reductions and therefore decreasing cancer
risk.
Cancer Risk Estimates from NESHAP
Risk Assessments
In looking back at the estimated effects of the HAP
standards, EPA found that the effects of the NESHAPs
were not quantified completely. These estimates oc-
curred at a time when emission estimation and risk
assessment methodologies for HAPs were first being
developed. One consequence is that because emissions
were not fully characterized, air toxics exposures could
not be completely assessed. Additionally, most assess-
ments only focused on the specific HAP being listed
under the CAA and did not assess the reduction of
other pollutants, which are currently considered HAPs.
For example, while the vinyl chloride standard reduces
emissions of ethylene dichloride, these emission re-
ductions were not assessed in the risk assessment. In
a different context, reductions of HAP may also
achieve reductions of VOC and PM. The benefits of
such reductions generally were also not evaluated. In
addition, EPA generally did not assess the potential
exposure to high, short-term concentrations of HAP
and therefore did not know whether toxic effects from
acute exposures would have been predicted and pos-
sibly addressed by the HAP standards.
In addition, people living near emission sources
of concern are often exposed to a mix of pollutants at
once. Some pollutants have been shown to act syner-
gistically together to create a health risk greater than
the risk that would be expected by simply adding the
two exposure levels together. More research is needed
to understand the effects of multiple-pollutant expo-
sures. Finally, HAP risks tend to be distributed un-
evenly across exposed populations, with particularly
high exposures occurring closest to emission sources.
It should be noted that HAP exposure to specific popu-
lations may tend to fall disproportionately among the
poor and minorities, who are more likely to live in
close proximity to emitting facilities.
With the above caveats in mind, Table H-2 pro-
vides information about maximum individual risk
taken from the Federal Register notices for the
NESHAPs promulgated before the 1990 amendments
to the Clean Air Act. The benefits are calculated by
multiplying the estimated annual incidence reduction
by the $4.8 million valuation per statistical life (1990
dollars). These benefit estimates provide a snapshot
of potential monetized benefits for the year in which
each NESHAP was promulgated. Of course these es-
timates do not include air toxics benefits for other
health and ecological benefit categories, or air toxics
benefits from co-control of criteria air pollutants. All
uncertainties associated with the original estimates
remain.
Non-utility Stationary Source
Cancer Incidence Reductions
The Project Team commissioned two studies to
estimate reductions in cancer incidence due to pre-
1990 NESHAPs: the PES Study and the ICF Re-analy-
sis. The methodology used for most air pollutant evalu-
ations involved a "back calculation" for the estima-
tion of incidence reductions. However, the EPA has
elected not to rely on the results of this analysis given
critical methodological flaws. Despite the Project
Team's concerns, the methodology and results of the
two studies are presented below in the interest of full
disclosure and to guide efforts to develop a more valid
7 Hunt, W.F., Faoro, R.B. and Curran, T.C., "Estimation of Cancer Incidence Cases and Rates for Selected Toxic Air Pollutants
Using Ambient Air Pollution Data, 1970 vs. 1980". U.S. EPA. April 1985.
__
-------
Appendix H: Air Toxics
/Tatife H-2L Cancer J|jcideace!ReaBcdons and ^lonetizeXBenefits f or rsfeSHAPs,' '"'' ' " - --".;^'"
';«V- -''/!:<, -'• ''• J'V« ," -''\ ' -";%•-,/ ,,,',>-, ...,;s, ~ - ,,;ft;^;-,;,, „„,,-, - ..„ > „,/, - „,„ |^
*;.*'.„ ^/"^"f / **;?^'?f.tZ*3- "'^^ ,,, " ',' ,,^-,-,-,£ , - "~-£; . • "" '' T , "^..-i,1"",,, ' ''' ''" '" ' * ' ! * ' , -^ , s "
' 'V* .. •"*
'*<-, .-^
1 Pollutant
, * , ?,,* *
; benzene r
^be'feie ^
benzene5*---
- benzene ',
'benzene'™"
arsenic v
^S.
"•• •"' , ..;•
- arsenic ' -
asbestost
; vinyl 7, ,
chloride
'$ , ,
"• "'v'1' , ' -1 -1 %s ' '""
->- '~''»: >: V-
-Source ,„,;:
- - Category
'* •;,!->-.;\,>',
'c6kfe,byi-''
"* product " - -
-storage -, '„',",„
''-vessels' •'' , •
waste ,,
. operations
, transfer'
'operations
primary ' '
'•;1'1- V ^
,,1984-0^ s'/
^ ^^ ^ 'i,.
-: *„• • ' <-
1982^ . ^ „
'19'86 i
,-, ^\
^^ ^ ^ ^ ^ /^ ^
j,9,87; ;,;;, ,„-,
^ s/; ^''^ ,
J9'86 - , ,'; '
W86 r -, ;Vv;
f/ -1 ^ A""' '••i'j
.. ^' •. •- y *&'
"1973 '•••••'* '••-;,,
1975 -,'": " ,„ ••
" >f' '* fa
'-Pre-Reg-5"
Maximum
Individual
Risk- „,/:':",
i.sxiot3^ /-
•'%-' '- '
}7xl6*! , , ,
/ ' '-'^'^^ < ^
4,5x10,-*
=' ^ /, ,
, ; " ;ij
2K10-3; - '' ;
'-" -' , -, '/''',
^xio?,- ";,
L3x!0'*to*'
,SxlOf -. '
7x10140'-'"-
>xiof', £">,"
> ' * -v -;j.
- ,/• ,&*-*„•> ' ^
s f ""* ^'
t ','
Post -Reg ;
Maximum
' Individual ^
Rislk ,, ' *T,
^ ^ ^ '^ ^ /
-C-Sxld-4 ;:- ;'''.,
; < t^;
,„' *'*"' - /• '"
ixlO^ ;,
' ' -'•*„
•* ' , ''- f~*'
i:95 ^ '"",
^ X' ". -.} * ;
,0,01;t^0^;06^ ^
f>]~ ' ^ ' "**
A << - '^ '•''• •.^sss
\rf*~?*s •• ' ^
• "-*,,-, '-"-
r?0°98 '"' ",; ''/
, ' " "•'•<'' '• ',1-
-o:09i!; ''*'< ':'•*
b:ri7to^-, "-"
"o.d'oS' ^ 4~
'100^ ,-'":':' '';-'
, ( -" ",;,;, •. ' ",„&,
-1015 --,„„ '.<--
-^•fL- •''<-' -«/,
Benefits in
, $million per!;
-year' ', >:~ „ '
:(1990$)"'/ -
•:'LS. '-"-'' '^
"'-'-"'-,'•-*' ,,>'
':?A!>:^
1 'f ' %f
;*o.o5,to-, ^ i-
"43' •-' ^^
">, -* '"','^' '
-2:6;i'4'" <„-
' ' "¥f,;'
•" x^^. v v
"4;? >-"'-• /--»-
, '"-'">"- ,''<:<,
* ' ''~ ^,,',\
f?i:i ;*%
v ' ' •* '
cCoa'tbo'.^ *
, - , ,„ , ^^ ^^
'-•* '-».' "';-!" '
480,4:-'- ---:
»,50i4;4<;,if,'>X:,
and reliable analysis of the health-related benefits of
HAP reductions in the upcoming section 812 Prospec-
tive studies.
PES Study
Methodology
The first attempt to estimate, for this study, his-
torical non-utility stationary source HAP-related di-
rect inhalation cancer incidence reductions was con-
ducted by Pacific Environmental Services (PES). The
basic approach used in the PES study was to adjust
the cancer incidence estimates developed for EPA's
1990 Cancer Risk study to reflect the changes in emis-
sions of, and exposures to, 14 key HAPs: arsenic, as-
bestos, benzene, 1,3-butadiene, carbon tetrachloride,
chloroform, hexavalent chromium, dioxin, ethylene
dichloride, ethylene dibromide, formaldehyde, gaso-
line vapors, products of incomplete combustion
(PICs), and vinyl chloride.
The first step was to compile baseline incidence
levels, defined as cancer cases per million population,
for each of the 14 pollutants. The point estimates of
incidence from the 1990 Cancer Risk study were used
for this purpose. For some source categories, the "best
point estimate" from the 1990 Cancer Risk study was
used, for others a mid-point was selected.8 These
baseline incidence levels were based on measured
ambient concentrations of the pollutant, modeled con-
centrations, or both.
The second step involved allocating baseline in-
cidence levels to the individual source categories
known to emit the relevant pollutant. In some cases,
8 For some of the source categories, the original NESHAP/Air Toxic Exposure and Risk Information System (NESHAP/
ATERIS) estimates of incidence were not available, in which case the baseline incidence was obtained from the 1989 National Air
Toxics Information Clearinghouse( NATICH) Database Report. (See PES, "Draft Summary of Methodology Used for Cancer from
Stationary Sources," memorandum from Ken Meardon, PES to Vasu Kilaru, US EPA, March 22, 1993, p. 2.)
H-5
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
adjustments were made to reflect differences among
the vintages of source category-specific data.9 All
baseline incidence estimates were ultimately ex-
pressed relative to a 1985 base year.10 The assump-
tion was then made that source-category incidence
rates were proportional to the level of emissions from
that source category.
Next, levels of control for each source category-
specific incidence rate were estimated for each of the
target years of the present analysis (i.e., 1970,1975,
1980, 1985, and 1990)." Source category-specific
activity level indicators were then established and
linked to changes in corresponding activity indica-
tors provided by the J/W macroeconomic modeling
results. Activity levels were estimated for each source
category, for each of the target years, and for each of
the two scenarios.
Finally, source category/pollutant combination
incidence levels for both the control and no-control
scenarios were developed. These incidence levels were
developed based on the baseline incidence levels, the
activity indicators, and the control levels for each year.
Both of these latter two factors varied between the
control and no-control scenarios. The activity levels
differed based on the specific levels of related sector
economic activity predicted by the J/W model for the
control and no-control scenario. The control levels
prevailing in each of the target years were used for
the control scenario, and the 1970 control level was
applied throughout the 1970 to 1990 period for the
no-control scenario.12 The formula used for these cal-
culations was as follows:13
lby'
(i)
where:
I =
A =
P =
C =
ty =
by =
Findings
cancer incidence for a source category-
pollutant combination
activity level for a source category
population
control level for a source category-pol-
lutant combination
target year (1970 ... 1990)
base year
The PES analysis concluded that substantial re-
ductions in HAP-related cancer cases were achieved
during the reference period of the present study. The
vast majority of these estimated reductions were at-
tributable to reduced exposures to asbestos, particu-
larly from manufacturing and fabricating sources.14
In fact, roughly 75 percent of the total reduction in
cancer cases averaged over the 1970 to 1990 period
were attributed to asbestos control.15 Figure H-l sum-
marizes the PES study overall cancer incidence re-
ductions and the relative contribution of asbestos-re-
lated reductions over the study period.
The Project Team had several concerns about the
PES results. First and foremost, the reductions in as-
bestos-related cancer cases appeared to be substan-
tially higher than expected, particularly in the earlier
target years. Second, the control scenario activity level
indicators for several sources with which Project Team
members were familiar did not appear to be even re-
motely consistent with actual historical activity pat-
terns.16 Finally, the level of documentation of the ana-
lytical methodologies, assumptions, and results was
insufficient to ascertain the validity and reliability of
9 For example, six discrete sources for vinyl chloride were identified in the Six-Month Study Update. Point estimate incidences
for each of these source categories came from separate references with databases corresponding to different years. (See PES, "retro-
spective analysis for section 812(a) Benefits Study," September 30, 1992, p. 8.)
10 See PES, March 22, 1993 memorandum, p. 3.
11 Control level estimates were based on one of the following: control efficiencies for related criteria pollutants defined in the
criteria pollutant analysis, reference documents such as Control Technology Guidelines (CTGs) or Background Information Docu-
ments (BIDs), preambles for related regulations, or EPA experts. (See PES, March 22, 1993 memorandum, p. 3.)
n More detailed descriptions of the methodology and associated uncertainties are provided in "Retrospective Analysis for section
812(a) Benefits Study," a September 30, 1992 memorandum from Ken Meardon, PES to Vasu Kilaru, US EPA.
" See PES, March 22,1993 memorandum, p. 4.
14 PES, "Cancer Risk Estimates from Stationary Sources," memorandum from Ken Meardon, PES to Vasu Kilaru, US EPA,
March 5,1993.
15ICF, "Direct Inhalation Incidence Benefits," Draft Report, November 11, 1994, p. 10.
16 For example, the activity indicators for Municipal Waste Combustors (MWCs) incorporated in the PES analysis decline
dramatically throughout the 1975 to 1990 period. (See PES, March 5,1993 memorandum to Vasu Kilaru, p. 10). In reality, overall
MWC capacity and throughput increased significantly over this period. .
_ _ _
-------
Appendix H: Air Toxics
Figure Hn. PES Estimated Reductions in HAP-Related
Cancer Cases.
a>
"
1975 1980 1985 1990
Year
the results. Ultimately, the Project Team determined
that it was necessary to conduct a formal review and
re-analysis of the cancer incidence reductions associ-
ated with non-utility stationary source HAP controls.
The results of the PES analysis remain a relevant part
of the record of the present study, however, since they
provided a substantial basis for the subsequent re-
analysis by ICF Incorporated.
ICF Re-analysis
Methodology
The purposes of the ICF Re-analysis were to ex-
amine the methodology and results of the PES study,
particularly to address the aforementioned concerns
of the Project Team, and to develop a revised set of
estimates. Due to significant constraints on the re-
sources remaining for HAP analysis in the section 812
study, however, only a few key HAPs could be inves-
tigated in depth and many important issues could not
be addressed.17 Furthermore, the effects of two early
and potentially important HAP standards -the Beryl-
lium and Mercury NESHAPs- could not be evalu-
ated. Nevertheless, the ICF Re-analysis clarified some
potential sources of uncertainty in the PES re-
sults and provided revised cancer incidence re-
duction estimates for several HAPs.
A key uncertainty in the PES results was
associated with the use of a "back-calculation"
technique to estimate incidence reductions for
some HAPs. The back-calculation technique
estimates uncontrolled incidence by dividing
residual incidence by the assumed control effi-
ciency. This approach means uncontrolled inci-
dence, and therefore incidence reductions, are
highly sensitive to small changes in assumed
control efficiency.18 In some cases, the PES
analysis may have used control efficiencies
which were too high, resulting in overestima-
tion of uncontrolled incidence and therefore in-
cidence reductions attributable to the CAA.19
The vinyl chloride incidence reduction estimates ap-
pear to be significantly influenced by the use of this
back-calculation technique. Another important source
of uncertainty identified by ICF involved the poten-
tial overestimation of incidence totals when source
apportionment is based on measured ambient concen-
trations.20 ICF was unable, however, to perform an
extensive evaluation of the activity level indicators
used in the PES study.21
The first step undertaken in the re-analysis was to
conduct a screening test to identify the HAPs which
accounted for the most significant estimated incidence
reductions. Based on this screening analysis, ICF
eliminated 1,3-butadiene, carbon tetrachloride, chlo-
roform, gasoline vapors, chromium, formaldehyde,
and PICs from the detailed re-analysis effort.
Detailed reviews were then conducted for the re-
maining HAPs: vinyl chloride, dioxins, ethylene
dibromide (EDB), ethylene dichloride (EDC), ben-
zene, asbestos, and arsenic. In the re-analysis of these
HAPs, ICF determined whether a forward- or back-
calculation technique was used for the relevant source
categories of a given HAP, reviewed the regulatory
1 For example, the Project Team sought to develop and apply a methodology for estimating a central tendency estimate for the
total carcinogenic risk imposed by all the HAPs examined. The intent was to address concerns about potential overestimation of
aggregate risk measures when combining upper bound risk estimates of multiple HAPs. Unfortunately, resources were insufficient to
continue development of this methodology.
18 An example of this back-calculation technique illustrating the sensitivity to the assumed control efficiency is presented on page
12 of the draft ICF report.
19 See ICF Draft Report, p. 12.
20 See ICF Draft Report, p. 9.
21 See ICF Draft Report, p. 13.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
history of the relevant source categories to re-evalu-
ate the assumed control efficiencies, and reviewed
the upper-bound unit risk factor for each HAP.
Revised total incidence reduction estimates for each
HAP and for each target year were then calculated
using the same basic calculation procedure used
by PES. Finally, ICF identified a number of residual
deficiencies in the analysis which could only be
addressed through additional research and analy-
22
Figure H-2. ICF Estimated Reductions in Total HAP-
Related Cancer Cases Using Upper Bound Asbestos
Incidence and Lower Bound Non-Asbestos HAP Inci-
dence.
sis.
Findings
The ICF Re-analysis largely affirmed the origi-
nal results obtained by PES; primarily because the
PES analysis itself served as the basis for the re-
analysis and only minor adjustments were adopted
for many critical variables. In particular, most
Project Team concerns regarding the PES method-
ology could not be resolved, including uncertain-
ties associated with activity levels, assumed con-
trol efficiencies, and the unexpectedly high esti-
mated incidence reductions associated with asbes-
tos. In fact, the ICF Re-analysis produced a revised
upper bound estimate for vinyl chloride-related in-
cidence reductions which were even higher than
the asbestos benefits.
Several sets of results were developed by ICF
and presented in either the November 1994 draft
report or in briefing materials prepared for the Sci-
ence Advisory Board Clean Air Act Compliance
Analysis Council Physical Effects Subcommittee
(SAB/ACCACAPERS) in May 1995. The first set
of results is based on the assumption of 100 per-
cent source compliance with HAP control require-
ments. An alternative set of results was developed as-
suming an 80 percent compliance rate with applicable
standards. Given the linear effect of changes in com-
pliance rates, these results were precisely 20 percent
lower than the first set of estimates. At the May 1995
ACCACAPERS briefing, estimates based on the 100
percent compliance estimates were presented. For as-
bestos, the revised incidence reductions were pre-
sented and characterized as upper bound. The asbes-
tos estimates were then combined with upper and
lower bound estimates for vinyl chloride and for "all
other compounds." Figure H-2 presents the total can-
cer incidence reductions derived from the ICF Re-
analysis, using the asbestos estimates combined with
the lower bound estimates for non-asbestos HAPs.
dOtherHAPsj
(•Asbestos
1975 1980 1985 1990
Year
Figure H-3. ICF Estimated Reduction in Total HAP-
Related Cancer Cases Using Upper Bound Incidence for
All HAPs.
CjOtherHAPs
«Asbestos
1975 1980 1985 1990
Year
Figure H-3 presents a comparable compilation reflect-
ing the upper bound estimates for all HAPs.
The Project Team remains concerned about these
incidence reduction estimates, particularly given the
doubts raised by the SAB/ACCACAPERS at the May
1995 presentation of these results. For instance, sev-
eral critical assumptions are needed to make this analy-
sis valid when applied to EPA's NESHAPs. The flaws
in these assumptions are described below.
(1) The risk estimates described in the 1990 Can-
cer Risk study, which served as the baseline for deter-
mining risk reductions, were accepted without ques-
tion. There are myriad uncertainties in these estimates
a Additional details of the ICF Re-analysis methodology can be found in ICF, "Direct Inhalation Incidence Benefits," Draft
Report, November 11,1994.
H-8
-------
Appendix H: Air Toxics
that must be recognized, as the study was designed
only to generate rough order-of-magnitude estimates
of the extent of the air toxics cancer problem.
(2) The percent control efficiency for emission
reductions, which is calculated in each standard, would
have to apply across every facility. Typically, the
emissions reductions vary between facilities; using a
single average reduction could skew the results.
(3) There is a direct correlation between the num-
ber of tons of emissions reduced and incidence re-
duced by a specific regulation. Given the assumption
of a linear, non-threshold dose-response curve (as is
typically done for cancer), this is theoretically cor-
rect.
(4) Finally, the back calculation approach assumes
that there is 100 percent compliance with the regula-
tion.
EPA staff reviewed the "back calculation" ap-
proach for one of the more controversial aspects of
the vinyl chloride (VC) NESHAP. The PES study es-
timates benefits at 426 cases reduced in 1990. The
ICF Re-analysis resulted in an even higher estimate,
between 1,000 and 7,000 cases annually. An analysis
by EPA staff indicated that these vinyl chloride risk
estimates are highly suspect given historical cancer
incidence data for hepatic angiocarcinoma, a specific
cancer that has been linked to vinyl chloride (Koppikar
and Fegley, 1995). The following analysis demon-
strates the inadequacies of the assumptions in the 1993
study.
(1) In the actual standard, no control technology
was required for emissions from oxychlorination vents
at ethylene dichloride (EDQ/VC plants. Applying
"back calculation" for these emissions is inappropri-
ate.
(2) In 1985, there were an estimated 8,000 fabri-
cation plants which processed resins produced by PVC
plants, thus resulting in VC emissions, which were
exempt from the VC NESHAP. They emit very small
quantities of VC and back calculation is not appropri-
ate.
(3) The 1993 study uses a baseline estimate of 18
residual cases from the NESHAP/ATERIS data base.
There is no evidence that these cases resulted only
from emissions from PVC and EDC/VC plants.
(4) The risk analysis performed for the October
21, 1976 final VC regulation projected an incidence
reduction of 11 cases per year.
In contrast, the PES study, using the "back calcu-
lation" method derived the following annual incidence
reductions:
1980-250 cases
1985-360 cases
1990 - 430 cases
The subsequent back calculation conducted in the
ICF Re-analysis resulted in incidence reductions as
much as an order of magnitude higher than these.
Even considering the slightly different industrial
output assumptions imposed by macroeconomic mod-
eling, such a stark contrast is difficult to explain ex-
cept for a critically flawed approach. Growth in ac-
tivity and population nor other factors explain the dif-
ference in these two estimates. Given that the same
general methodology was used for all of the ah- toxic
pollutant assessments as was used for the VC
NESHAP evaluation, there is reason to believe that
cancer incidence results for the other ah- toxic pollut-
ants are also flawed.
Mobile Source HAP Exposure
Reductions
EPA's Cancer Risk report estimated that approxi-
mately 60 percent of the total carcinogenic risk posed
by HAPs was attributed to mobile sources, with sta-
tionary sources contributing 15 percent and area
sources contributing the remaining 25 percent.23 The
relative importance of mobile sources to total HAP
exposure was a significant motivation behind EPA's
subsequent effort to examine exposures and risks from
mobile source HAPs.24 Although available analytical
resources were severely limited, the Project Team
nevertheless decided it was necessary to perform at
least an initial screening analysis to estimate the dif-
ferences in mobile source HAP exposures between
the control and no-control scenarios configured for
the present study.
23 Cancer Risk report, Page ES-12.
24 See US EPA/OAR/OMS, "Motor Vehicle-Related Air Toxics Study," EPA 420-R-93-005. April 1993.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Methodology
The approach used by ICF/SAI in conducting the
mobile source HAP analysis closely followed the ap-
proach used in the EPA Motor Vehicle-Related Air
Toxics Study (MVATS).25 Recognizing the dearth of
HAP ambient concentration and exposure data, both
studies use carbon monoxide (CO) concentrations as
the basis for estimating mobile source HAP concen-
trations and exposures. An important difference be-
tween the two studies, however, is that the ICF/SAI
study adjusted the estimated change hi ambient CO
concentrations to take account of background26 and
non-mobile source27 CO emissions. The HAP expo-
sure function used in the ICF/SAI analysis is summa-
rized by the following equation:
(2)
where:
E
C
B
S
M
voc
HAP
CO
exposure to motor vehicle-emitted
HAP
annual ambient CO concentration to
annual CO exposure concentration
conversion factor
county-level annual average ambient
CO concentration
background CO concentration
no-control to control scenario CO
concentration adjustment factor
(equals 1 for the control scenario)
total CO exposure to mobile source
CO exposure conversion factor
VOC emissions by year, county, and
scenario
VOC speciation factor by mobile
source HAP
CO emissions by year, county, and
scenario
Details of the derivation of each of the variables
applied in the above equation are provided in the ICF/
SAI report. However, in essence, the calculation in-
volves the following basic steps.
First, annual average county-level CO ambient
monitoring data are compiled from the EPA
Aerometric Information Retrieval System (AIRS)
database. After adjusting for background and non-
mobile source contributions, these annual average
ambient CO concentrations are converted to annual
average CO exposure concentrations. As in the EPA
MVATS, this conversion is made based on the Haz-
ardous Air Pollutant Exposure Model - Mobile Sources
(HAPEM-MS) population exposure model, which
takes account of time spent in five indoor and out-
door microenvironments: indoors at home, other in-
door, in-vehicle, outdoors near roadway, and other
outdoor.28 After adjusting for CO exposures attribut-
able to non-mobile sources of CO, the CO exposures
are converted to exposures for each of the mobile
source HAPs based on available VOC speciation data
and the ratio of co-located VOC and CO emissions.29
These calculations are repeated for the no-control sce-
nario after adjusting for differences in CO ambient
concentrations for each target year and for differences
in fuel composition.
Results
By 1990, CAA controls resulted in significant
reductions in exposure to motor vehicle HAPs. Fig-
ure H-4 summarizes the nationwide annual average
exposure levels, in micrograms per cubic meter, for
each of the five HAPs analyzed under the control and
no-control scenarios. Additional detailed results, in-
cluding breakdown by urban versus rural environ-
ments and comparisons with the EPA MVATS esti-
mates, are provided in the ICF/SAI report.
Analytical resources to carry forward these expo-
sure estimates to derive estimates of the changes in
motor vehicle HAP-related adverse effects attribut-
able to historical CAA programs were not available.
25 ICF/SAI, "Retrospective Analysis of Inhalation Exposure to Hazardous Air Pollutants from Motor Vehicles," October 1995, p. 4.
86 Background CO is produced by the oxidation of biogenic hydrocarbons. See ICF/SAI, p. 7.
57 The EPA MVATS attributed all measured CO to motor vehicles, resulting in an overestimation of motor-vehicle contributions to
CO concentration changes. See ICF/SAI, p. 8. The MVATS assumption would also lead to a subsequent overestimation of changes in
HAP exposures.
» See ICF/SAI, p. 3.
29 The same HAP emission fractions used in the EPA MVATS were used herein, except for diesel PM which is not proportional to
VOC emissions. Instead, diesel PM emission factors were developed using year-specific PART5 diesel PM emission factors and VMT
estimates for diesel-powered vehicles.
" H-10
-------
Appendix H; Air Toxics
Figure H-4. National Annual Average Motor Vehicle
HAP Exposures (ug/m3).
O Control
•No-Control
Benzene AcetaHehyde. Diesel PM
Formaldehyde 13-Butadbne
Non-Cancer Health Effects
Broad gaps exist in the current state of knowl-
edge about the quantifiable effects of air toxics expo-
sure. This is particularly true for a wide range of health
effects such as tumors, abnormal development, birth
defects, neurological impairment, or reproductive
impairment, etc. For example, the EPA's Non-Can-
cer Study30 found that ambient concentrations for a
substantial number of monitored and modeled HAPs
exceeded one or more health benchmarks.31 However
no accepted methodology exists to quantify the ef-
fects of such exceedences. More data on health ef-
fects is needed for a broad range of chemicals.
Ecological Effects
Through the 1970s and 1980s, the adverse effects
of toxic pollution on the Great Lakes became clear
and undeniable. Over the same time period, scientists
began collecting a convincing body of evidence that
toxic chemicals released to the air can travel long dis-
tances and be deposited on land or water far from the
original sources. An example of this evidence is the
presence of such contaminants as PCBs, toxaphene,
and other pesticides in fish in Lake Siskiwit, a lake on
an island on upper Lake Superior, which has no water-
borne sources of pollution. Toxaphene, a pesti-
cide used primarily in the southeastern U.S. cot-
ton belt, has been found as far away as the Arc-
tic, with a decreasing ah" concentration gradient
from the southeast toward the Great Lakes and
the north Atlantic regions.
Similarly, a growing body of evidence
showed that pollutants that were persistent (do
not easily break down) and bioaccumulating (not
significantly eliminated from the body) were
magnifying up the food chain, such that top
predator fish contained levels up to millions of
times greater than the harmless levels in the
water. As such, those who ate those large fish,
such as humans, eagles, mink, and beluga whales
could receive very high exposures to the pollut-
ants. Wildlife were beginning to show adverse
effects in the wild, that could be duplicated in the lab.
In the Great Lakes, such chemicals as PCBs, mercury,
dieldrin, hexachlorobenzene, Lindane, lead com-
pounds, cadmium compounds, DDT/DDE, and oth-
ers are of significant concern. In other places in the
country, similar effects are being experienced, espe-
cially with mercury, which is transported primarily
by air, but exposure to which is primarily through con-
taminated fish. It was this kind of information about
DDT and toxaphene that led to their being banned in
the U.S. under FIFRA.
While ecological and economical sciences are not
yet sufficiently advanced to support the kind of com-
prehensive, quantitative evaluation of benefits needed
for the present study, selected local and regional scale
adverse ecological effects of HAPs, and their adverse
consequences for human health and welfare, can and
have been surveyed. In May 1994, the EPA issued its
first "Report to Congress on Deposition of Air Pollut-
ants to the Great Waters."32 The Great Waters Report
examined the pollutants contributing to adverse eco-
logical effects, the potential significance of the con-
tribution to pollutant loadings from deposition of air-
borne pollutants, and the potential adverse effects as-
sociated with these pollutant loadings. Key HAPs iden-
tified in the Great Waters Report include PCBs, mer-
cury, dioxins, and other heavy metals and toxic or-
ganics.
30 U.S. Environmental Protection Agency, "Toxic Air Pollutants and Noncancer Risks: Screening Studies " External Review
Draft, September, 1990.
31 Relevant benchmarks include Acceptable Daily Intake (ADI), the estimate of daily exposure at which adverse health effects are
unlikely; and Lowest Observed Actual Effect Level (LOAEL), which is the lowest exposure level at which significant adverse health
effects are observed.
1994.
32 USEPA/OAR/OAQPS, "Deposition of Air Pollutants to the Great Waters, First Report to Congress," EPA-453/R-93-055, May
H-ll
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Of particular relevance to the present assessment,
the Great Waters Report demonstrated the significance
of transport and transformation of HAPs through food
webs, leading to increased toxicity and
biomagnification. A prime example of adverse trans-
port and transformation is mercury. Transformation
from inorganic to methylated forms significantly in-
creases the toxic effects of mercury in ecosystems. A
prime example of biomagnification is PCBs. As noted
in the Great Waters Report:
"Pollutants of concern [such as PCBs]
accumulate in body tissues and magnify up
the food web, with each level accumulating
the toxics from its diet and passing the burden
along to the animal in the next level of the
food web. Top consumers in the food web,
usually consumers of large fish, may
accumulate chemical concentrations many
millions of times greater than the
concentrations present in the water—High risk
groups...include breast-feeding mothers
because breast-fed babies continue to
accumulate [pollutants] from their mothers
after birth. For example, they can have PCB
levels four times higher than their mothers
after six to nine months of breastfeeding."33
Because of the risk of significant exposure to in-
fants and other high-risk groups, such as "sport an-
glers, Native Americans, and the urban poor,"34 a sub-
stantial number offish consumption advisories have
been issued in recent years. Current fish advisories
for the Great Lakes alone include widespread adviso-
ries for PCB's, chlordane, mercury and others, cau-
tioning that nursing mothers, pregnant women, women
who anticipate bearing children, female children of
any age and male children age 15 and under not eat
certain high-food chain fish species. It should be noted
as well that 40 states have issued mercury advisories
in some freshwater bodies, and nine states have is-
sued mercury advisories for every freshwater
waterbody in the state (these states are Maine, New
Hampshire, Vermont, Massachusetts, New York, New
Jersey, Missouri, Michigan, and Florida).
There is little evidence indicating that the CAA
had much beneficial effect on air toxic deposition to
water bodies. Since the early NESHAPs were based
on direct inhalation, primarily cancer effects close to
33 EPA-453/R-93-055, May 1994, p. ix.
34 EPA-453/R-93-055, May 1994, p. x.
: H-12
a plant, they did not address the issue of cumulative
effects of persistent pollutants far from the source. It
was for this reason that section 112(m) was included
in the 1990 CAA Amendments, with requirements to
study and document the atmospheric contribution of
water pollutants, the adverse human health and envi-
ronmental effects resulting and the sources that should
be controlled to prevent adverse effects, and addition-
ally, to promulgate regulations to prevent adverse ef-
fects.
Conclusions — Research Needs
As has been demonstrated, there are broad gaps
in the current state of knowledge about the quantifi-
able effects of air toxics exposure for a wide range of
both human health and environmental effects. The
following discussion outlines areas in which further
research is needed in order to adequately quantify the
benefits of air toxics control.
Health Effects
• Develop health effects data on pollutants for
which limited or no data currently exists. Such
studies should be focused on pollutants with
a relatively high probability of exposure and/
or potential adverse health effects.
• Understand mechanism of action of pollut-
ants, for example through pharmacokinetic
modeling. This will allow for a more accu-
rate assessment of the effects of these pollut-
ants on humans.
• Conduct research on factors that affect varia-
tions in susceptibility of human populations
and determine the distribution of these fac-
tors in the U.S.
• Conduct research to better understand inter-
active effects of multiple pollutant exposures.
• Develop methodologies to derive alternative
estimates of human cancer risk from existing
upper-bound methods.
• Acquire data and develop dose-response re-
lationships for critical noncancer effects such
as developmental, neurotoxic, mutagenic, res-
-------
Appendix H: Air Toxics
piratory and other effects. In particular, de-
sign methodology to quantify effects of ex-
posures above health benchmarks.
Acquire data and develop methods to estimate
effects from acute exposure.
Exposure Assessment
• Expand data collection efforts: pre- and post-
control emissions; HAP speciation; facilities
location; facility parameters (stack heights,
distances from stacks to fencelines, etc.).
• Develop more comprehensive exposure mod-
els which incorporate activity patterns, indi-
rect exposures, total body burden, ratios of
time spent indoors to outdoors.
• Continue to refine uncertainty analysis meth-
ods.
Ecosystem Effects
• Reliable estimates/measures of the levels of
persistent bioaccumulating toxics in different
media (air, water column, soils and sediments)
• Work to correlate levels of persistent
bioaccumulating toxics with exposures, biota
concentrations/accumulation, and adverse
effects, especially subtle effects such as wast-
ing, behavioral effects, and developmental
effects.
Criteria for effects, such as a wildlife corre-
late to a RfD or dose-response curve. This
work should be done to complement the mass
balance efforts now being completed, which
will model source emissions to water column
concentrations, then design research to pre-
dict effects on living resources given those
predicted levels.
• Work to determine the effects of mixtures of
persistent bioaccumulating toxic pollutants,
and to determine cause-effect relationships of
exposures over long periods of time.
• Studies to evaluate toxic effects in less well
understood terrestrial systems such as: soil
organisms/invertebrates, food web effects,
amphibian effects, effects on endangered spe-
cies and phytotoxic effects.
• Work to improve understanding of effects of
toxic air pollutants on wetland species and
wetland functions.
Economic Valuation
Develop valuation estimates for endpoints for
which inadequate estimates currently exist.
These valuation estimates must be consistent
with the kinds of damages expected.
• Initiate broad-scope economic valuation of air
toxics program using survey techniques.
H-13
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Air Toxics References
Hunt, W.R, R.B. Faoro, and T.C. Curran, "Estima-
tion of Cancer Incidence Cases and Rates for
Selected Toxic Air pollutants Using Ambi-
ent Air Pollution Data, 1970 vs. 1980," U.S.
Environmental Protection Agency, April
1985.
ICF Kaiser, "Direct Inhalation Incidence Benefits,"
Draft Report, November 11,1994.
ICF Kaiser and Systems Applications International,
"Retrospective Analysis of Inhalation Expo-
sure to Hazardous Air Pollutants From Mo-
tor Vehicles," October 1995.
Koppikar, Aparna and Robert Fegley. 1995. "Analy-
sis of 'Reasonableness' of Cancer Risk As-
sessment Findings for Asbestos and Vinyl
Chloride in section 812 Retrospective Cost-
Benefit Analysis," Memorandum to Jim
DeMocker, Office of Policy'Analysis and
Review, Office of Air and Radiation, U.S.
Environmental Protection Agency. Novem-
ber^ 1995.
Pacific Environmental Services, "Cancer Risk Esti-
mates From Stationary Services," Memoran-
dum to Vasu Kilaru, U.S. EPA, March 5,
1993.
Pacific Environmental Services, "Draft Summary of
Methodology Used For Cancer From Station-
ary Services," Memorandum to Vasu Kilaru,
U.S. EPA, March 22,1993.
Pacific Environmental Services, "Retrospective
Analysis for Section 812 (a) Benefits Study,"
September 30,1992.
f
U.S. Environmental Protection Agency, The Air Toxics
Problem in the United States: An Analysis of
Cancer Risks for Selected Pollutants, Office
of Air Quality Planning and Standards, EPA-
450/1-85-001, May 1985.
U.S. Environmental Protection Agency, Cancer Risk
From Outdoor Exposure to Air Toxics, Of-
fice of Air Quality Planning and Standards,
EPA-450/l-90-004a, September 1990.
U.S. Environmental Protection Agency, Deposition
of Air Pollutants to the Great Waters, First
Report to Congress, Office of Air Quality
Planning and Standards, EPA-453/R-93-055,
May 1994.
U.S. Environmental Protection Agency, Motor Ve-
hicle-Related Air Toxics Study, Office of
Mobile Sources, EPA-420/12-93-005, April
1993.
U.S. Environmental protection Agency, "Toxic Air
Pollutants and Noncancer Risks: Screening
Studies," External Review Draft, September
1990.
U.S. Environmental protection Agency, Unfinished
Business: A Comparative Assessment of En-
vironmental Problems, Office of Policy, Plan-
ning, and Evaluation, February 1987.
H-14
-------
Appendix I: Valuation of Human Health and
Welfare Effects of Criteria Pollutants
This appendix describes the derivations of the
economic valuations for health and welfare endpoints
considered in the benefits analysis. Valuation esti-
mates were obtained from the literature and reported
in dollars per case avoided for health effects, and dol-
lars per unit of avoided damage for welfare effects.
This appendix first introduces the method for mon-
etizing improvements in health and welfare, followed
by a summary of dollar estimates used to value ben-
efits and detailed descriptions of the derivation of each
estimate. These economic valuations are given both
in terms of a central (point) estimate as well as a prob-
ability distribution which characterizes the uncertainty
about the central estimate. All dollar values are
rounded and are in 1990 dollars. Next, results of the
economic benefits analysis are presented. Finally, un-
certainties in valuing the benefits attributable to the
CAA are explored.
Methods Used to Value Health
and Welfare Effects
Willingness to pay (WTP) and willingness to ac-
cept (WTA) are the two measures commonly used to
quantify the value an individual places on something,
whether it is something that can be purchased in a
market or not. Both WTP and WTA are measures of
the amount of money such that the individual would
be indifferent between having the good (or service)
and having the money. Whether WTP or WTA is the
appropriate measure depends largely on whether an
increase or a decrease of the good is at issue. WTP is
the amount of money an individual would be willing
to pay to have a good (or a specific increase in the
amount of the good) — i.e., the amount such that the
individual would be indifferent between having the
money and having the good (or having the specific
increase in the good). WTA is the amount of money
the individual would have to be compensated in order
to be indifferent to the loss of the good (or a specific
decrease in the amount of the good). WTP is the ap-
propriate measure if the baseline case is that the indi-
vidual does not have the good or when an increase in
the amount of the good is at issue; WTA is the appro-
priate measure if the baseline case is that the indi-
vidual has the good or when a decrease in the amount
of the good is at issue. An important difference be-
tween WTP and WTA is that, in theory, WTP is lim-
ited by the individual's budget, whereas WTA is not.
Nevertheless, while the underlying economic valua-
tion literature is based on studies which elicited ex-
pressions of WTP and/or WTA, the remainder of this
report refers to all valuation coefficients as WTP esti-
mates. In some cases (e.g., stroke-related hospital
admissions), neither WTA nor WTP estimates are
available and WTP is approximated by cost of illness
(COI) estimates, a clear underestimate of the true
welfare change since important value components
(e.g., pain and suffering associated with the stroke)
are not reflected in the out-of-pocket costs for the
hospital stay.
For both market and non-market goods, WTP re-
flects individuals' preferences. Because preferences
are likely to vary from one individual to another, WTP
for both market and non-market goods (e.g., health-
related improvements in environmental quality) is
likely to vary from one individual to another. In con-
trast to market goods, however, non-market goods
such as environmental quality improvements are pub-
lic goods, whose benefits are shared by many indi-
viduals. The individuals who benefit from the envi-
ronmental quality improvement may have different
WTPs for this non-market good. The total social value
of the good is the sum of the WTPs of all individuals
who "consume" (i.e., benefit from) the good.
In the case of health improvements related to pol-
lution reduction, it is not certain specifically who will
receive particular benefits of reduced pollution. For
example, the analysis may predict 100 days of cough
avoided in a given year resulting from CAA reduc-
tions of pollutant concentrations, but the analysis does
not estimate which individuals will be spared those
days of coughing. The health benefits conferred on
individuals by a reduction in pollution concentrations
are, then, actually reductions in the probabilities of
having to endure certain health problems. These ben-
efits (reductions in probabilities) may not be the same
for all individuals (and could be zero for some indi-
viduals). Likewise, the WTP for a given benefit is
likely to vary from one individual to another. In theory,
the total social value associated with the decrease in
risk of a given health problem resulting from a given
1-1
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990 •
reduction in pollution concentrations is
N
. (ID
where B, is the benefit (i.e., the reduction in prob-
ability of having to endure the health problem) con-
ferred on the ith individual (out of a total of N) by the
reduction in pollution concentrations, and WTPj(B.)
is the ith individual's WTP for that benefit. If a re-
duction in pollution concentrations affects the risks
of several health endpoints, the total health-related
social value of the reduction in pollution concentra-
tions is
(2)
M
where B.. is the benefit related to the jth health
endpoint (i.e., the reduction in probability of having
to endure the jth health problem) conferred on the ith
individual by the reduction in pollution concentrations,
and WTP^By) is the ith individual's WTP for that
benefit.
The reduction in probability of each health prob-
lem for each individual is not known, nor is each
individual's WTP for each possible benefit he or she
might receive known. Therefore, in practice, benefits
analysis estimates the value of as tatistical health prob-
lem avoided. For example, although a reduction in
pollutant concentrations may save actual lives (i.e.,
avoid premature mortality), whose lives will be saved
cannot be known ex ante. What is known is that the
reduction in air pollutant concentrations results in a
reduction in mortality risk. It is this reduction in mor-
tality risk that is valued in a monetized benefit analy-
sis. Individual WTPs for small reductions in mortal-
ity risk are summed over enough individuals to infer
the value of a statistical life saved. This is different
from the value of a particular, identified life saved.
Rather than "WTP to avoid a death," then, it is more
accurate to use the term "WTP to avoid a statistical
death," or, equivalently, "the value of a statistical life."
Suppose, for example, that a given reduction in
PM concentrations results in a decrease in mortality
risk of 1/10,000. Then for every 10,000 individuals,
one individual would be expected to die in the ab-
sence of the reduction in PM concentrations (who
would not die in the presence of the reduction in PM
concentrations). If WTP for this 1/10,000 decrease in
mortality risk is $500 (assuming, for now, that all in-
dividuals' WTPs are the same), then the value of a
statistical life is 10,000 x $500, or $5 million.
A given reduction in PM concentrations is un-
likely, however, to confer the same risk reduction (e.g.,
mortality risk reduction) on all exposed individuals
in the population. (In terms of the expressions above,
Bj is not necessarily equal to B., for i j). In addition,
different individuals may not be willing to pay the
same amount for the same risk reduction. The above
expression for the total social value associated with
the decrease in risk of a given health problem result-
ing from a given reduction in pollution concentrations
may be rewritten to more accurately convey this. Us-
ing mortality risk as an example, for a given unit risk
reduction (e.g., 1/1,000,000), the total mortality-re-
lated benefit of a given pollution reduction can be
written as
::'"'N ,.!./>*" *"' "**>' '7.,,'" ;""
, • j_ •-|» '' '.> < ••
*\,2\(numberofimiis0frii;K'reduction)f „ , „,,,,,
v'>-.ja.T ^ K (WTP.per unit risk r,eduction)i (3)
where (number of units of risk reduction); is the
number of units of risk reduction conferred on the ith
exposed individual as a result of the pollution reduc-
tion, (WTP per unit risk reduction); is the ith
individual's willingness to pay for a unit risk reduc-
tion, and N is the number of exposed individuals.
If different subgroups of the population have sub-
stantially different WTPs for a unit risk reduction and
substantially different numbers of units of risk reduc-
tion conferred on them, then estimating the total so-
cial benefit by multiplying the population mean WTP
to save a statistical life times the predicted number of
statistical lives saved could yield a biased result. Sup-
pose, for example, that older individuals' WTP per
unit risk reduction is less than that of younger indi-
viduals (e.g., because they have fewer years of ex-
pected life to lose). Then the total benefit will be less
than it would be if everyone's WTP were the same. In
addition, if each older individual has a larger number
of units of risk reduction conferred on him (because a
given pollution reduction results in a greater absolute
reduction in risk for older individuals than for younger
individuals), this, in combination with smaller WTPs
of older individuals, would further reduce the total
benefit.
While the estimation of WTP for a market good
(i.e., the estimation of a demand schedule) is not a
simple matter, the estimation of WTP for a non-mar-
ket good, such as a decrease in the risk of having a
particular health problem, is substantially more diffi-
cult. Estimation of WTP for decreases in very spe-
cific health risks (e.g., WTP to decrease the risk of a
day of coughing or WTP to decrease the risk of ad-
mission to the hospital for respiratory illness) is fur-
ther limited by a paucity of information. Derivation
of the dollar value estimates discussed below was of-
ten limited by available information.
1-2
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
Valuation of Specific Health Endpoints
Valuation of Premature Mortality Avoided
As noted above, it is actually reductions in mor-
tality risk that are valued in a monetized benefit
analysis. Individual WTPs for small reductions in
mortality risk are summed over enough individuals
to infer the value of a statistical life saved. This is
different from the value of a particular, identified
life saved. The "value of a premature death
avoided," then, should be understood as shorthand
for "the value of a statistical premature death
avoided."
The value of a premature death avoided is based
on an analysis of 26 policy-relevant value-of-life
studies (see Table 1-1). Five of the 26 studies are
contingent valuation (CV) studies, which directly
solicit WTP information from subjects; the rest are
wage-risk studies, which base WTP estimates on
estimates of the additional compensation demanded
in the labor market for riskier jobs. Each of the 26
studies provided an estimate of the mean WTP to
avoid a statistical premature death. Several plau-
sible standard distributions were fit to the 26 esti-
mates of mean WTP. A Weibull distribution, with
a mean of $4.8 million and standard deviation of
$3.24 million, provided the best fit to the 26 esti-
mates. The central tendency estimate of the WTP
to avoid a statistical premature death is the mean of
this distribution, $4.8 million. The considerable un-
certainty associated with this approach is discussed
in detail below, in the subsection titled "The Eco-
nomic Benefits Associated with Mortality," within
the section titled "Uncertainties."
Life-years lost is a possible alternative measure
of the mortality-related effect of pollution, as dis-
cussed in Appendix D. If life-years lost is the mea-
sure used, then the value of a statistical life-year lost,
rather than the value of a statistical life lost would be
needed. Moore and Viscusi (1988) suggest one ap-
proach for determining the value of a statistical life-
year lost. They assume that the willingness to pay to
save a statistical life is the value of a single year of
life times the expected number of years of life remain-
ing for an individual. They suggest that a typical re-
spondent in a mortal risk study may have a life ex-
pectancy of an additional 35 years. Using a mean es-
timate of $4.8 million to save a statistical life, their
approach would yield an estimate of $ 137,000 per life-
year lost or saved. If an individual discounts future
additional years using a standard discounting proce-
dure, the value of each life-year lost must be greater
than the value assuming no discounting. Using a 35
year life expectancy, a $4.8 million value of a statisti-
cal life, and a 5 percent discount rate, the implied value
,1990 dollars).
:
''>
fj
5
fff
*
_,
••'•• :•••"*. •:"'•'.•''•
tSjeiSnetaad'ieeCa'^i^trs') * ~;
SmitU ahd.,G}lbeit-'(;l 984),,,; ~; ; , - , ,
IXl!ingham''(l985y,r.' - ' v "- ,.
,Butier (1.9831,-,- ;„ ;,,„ „" - ,
Mtfier'and Guria (,S991) - '''"'- •
Mooie an8'-Vjscusi (1,9888),;- '""• >:
, Viscasi, Magat, and Huber (199$ bj" '
Gegax,etaL'(1985), '/' ">'„' '
-Maria5ans '", ;,r
Coasineauj Lacroix.'and Girard'- ,-;
(1988},^ ' ,- 1, '•'"".'•• -' N-
Jooes-Lee (1989),' ">„ '*, -^ ;,,,
DiKingham (1985) ( ,
Viscusi (19:78^,979) ', , ' '. »",£„
•R,S, Smith {1976)" \""" •';, , --'-
y.-K. Smith (l'976h,, ' , " ; , """••
C«SOn'(19gl)"s,', =: \ -,„„','•' *>• ,.
Viscusi (1981) t; '•' ,,, -
Rv$.,Snutli,.(1^74J! ? ^,,,,--'' - ,' ;
WfooTs-and1 Viscusr^(I988a) •- ------
Kneisnerand Leeth'(I991) ,'< ,„ " - -, ,', - ',"
somfi& y-w^vvi - --;^^
Estimafe' s
Labor Market ''
Labor Market "
Labor Market "
'Labor Market!
;Cont. Value '•"
Labor Market - '
Cent. Value- -
'Cont.'Vatue '-
'•Larjor.'Marfck /
LaborMat^,
Cont.,,Va!ae ; ,
Labor Market
'Com.' Value *-
Labor I^arfcet ',/
Labor,Mirket<
Labor Market -i
-Labor Market ',,
-Labof MMet ?"'
-CabosMarlcet
Labftf Market "
'LabofcMarket / •
Labor. Market -,-
Labof Market •'£
La,bor, Market-' ,
Labor Market
-La'bor Market'' >,-
"*<-;, ', ~ y~, • ""-
^Valualtlon'-,
(mUlions
", a- ,
',2'8--:'
,;:: 3:3',,,
„' , 3.4 '
„ •-',/???>
'•',, 3.8>,/
319"
, -', '-'4,1 '-,-
',, 4.6
•„ "f^j"~-
':r,-5,2-''
' - "62T'
„''''* ,7,2-^
";, , ;73'
,', .„ '7.6V',
'/'"'X'ftl1--,
~ , ' 9.7" ''
, 1O4-,
1 f3.5
;?;£«k-
of each life-year lost is $293,000. The Moore and
Viscusi procedure is identical to this approach, but
uses a zero discount rate.
Using the value of a life-year lost and the expected
number of years remaining (obtained from life ex-
pectancy tables), and assuming a given discount rate,
values of age-specific premature mortality can be de-
rived. Examples of valuations of pollution-related
mortality using the life-years lost approach are given
below, in the subsection titled "The Economic Ben-
efits Associated with Mortality," within the section
titled "Uncertainties."
Valuation of Hospital Admissions Avoided
In the case of hospital admissions, cost-of-illness
(COI), or "costs avoided," estimates were used in lieu
of WTP because of the lack of other information re-
1-3
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
garding willingness to pay to avoid illnesses that ne-
cessitate hospital admissions. For those hospital ad-
missions which were specified to be the initial hospi-
tal admission (in particular, hospital admissions for
coronary heart disease (CHD) events and stroke), COI
estimates include, where possible, all costs of the ill-
ness, including the present discounted value of the
stream of medical expenditures related to the illness,
as well as the present discounted value of the stream
of lost earnings related to the illness. (While an esti-
mate of present discounted value of both medical ex-
penditures and lost earnings was available for stroke,
the best available estimate for CHD did not include
lost earnings. The derivations of the COI estimates
for CHD' and stroke, both lead-induced effects, are
discussed in Appendix G.)
In those cases for which it is unspecified whether
the hospital admission is the initial one or not (that is,
for all hospital admissions endpoints other than CHD
and stroke), it is unclear what portion of medical ex-
penditures and lost earnings after hospital discharge
can reasonably be attributed to pollution exposure and
what portion might have resulted from an individual' s
pre-existing condition even in the absence of a par-
ticular pollution-related hospital admission. In such
cases, the COI estimates include only those costs as-
sociated with the hospital stay, including the hospital
charge, the associated physician charge, and the lost
earnings while in the hospital. The derivations of these
costs are discussed in Abt Associates Inc., 1996.
These COI estimates are likely to substantially
understate total WTP to avoid an illness that began
with a hospital admission or to avoid a particular hos-
pital admission itself. First, most of the COI estimates
fall short of being full COI estimates either because
of insufficient information or because of ambiguities
concerning what portion of post-hospital costs should
be attributed to pollution exposure. Even full COI es-
timates will understate total WTP, however, because
they do not include the value of avoiding the pain and
suffering associated with the illness for which the in-
dividual entered the hospital.
Valuation of Chronic Bronchitis Avoided
Although the severity of cases of chronic bron-
chitis valued in some studies approaches that of
chronic obstructive pulmonary disease, to maintain
consistency with the existing literature we do not treat
those cases separately for the purposes of this analy-
sis. Chronic bronchitis is one of the only morbidity
endpoints that may be expected to last from the initial
onset of the illness throughout the rest of the
individual's life. WTP to avoid chronic bronchitis
would therefore be expected to incorporate the present
discounted value of a potentially long stream of costs
(e.g., medical expenditures and lost earnings) associ-
ated with the illness. Two studies, Viscusi et al. (1991)
and Krupnick and Cropper (1992) provide estimates
of WTP to avoid a case of chronic bronchitis. The
study by Viscusi et al., however, uses a sample that is
larger and more representative of the general popula-
tion than the study by Krupnick and Cropper (which
selects people who have a relative with the disease).
The valuation of chronic bronchitis in this analysis is
therefore based on the distribution of WTP responses
from Viscusi et al. (1991).
Both Viscusi et al. (1991) and Krupnick and Crop-
per (1992), however, defined a case of severe chronic
bronchitis. It is unclear what proportion of the cases
of chronic bronchitis predicted to be associated with
exposure to pollution would turn out to be severe cases.
The incidence of pollution-related chronic bronchitis
was based on Abbey et al. (1993), which considered
only new cases of the illness.1 While a new case may
not start out being severe, chronic bronchitis is a
chronic illness which may progress in severity from
onset throughout the rest of the individual's life. It is
the chronic illness which is being valued, rather than
the illness at onset.
The WTP to avoid a case of pollution-related
chronic bronchitis (CB) is derived by starting with
the WTP to avoid a severe case of chronic bronchitis,
as described by Viscusi et al. (1991), and adjusting it
downward to reflect (1) the decrease in severity of a
case of pollution-related CB relative to the severe case
described in the Viscusi study, and (2) the elasticity
of WTP with respect to severity. Because elasticity is
a marginal concept and because it is a function of se-
verity (as estimated from Krupnick and Cropper,
1992), WTP adjustments were made incrementally,
in one percent steps. At each step, given a WTP to
avoid a case of CB of severity level sev, the WTP to
avoid a case of severity level Q.99*sev was derived.
This procedure was iterated until the desired severity
level was reached and the corresponding WTP was
derived. Because the elasticity of WTP with respect
to severity is a function of severity, this elasticity
changes at each iteration. If, for example, it is believed
that a pollution-related case of CB is of average se-
' It is important that only new chronic bronchitis be considered in this analysis because WTP estimates reflect lifetime expendi-
tures and/or losses associated with this chronic illness, and incidences are predicted separately for each year during the period 1970-
1990. If the total prevalence of chronic bronchitis, rather than the incidence of only new chronic bronchitis were predicted each year,
valuation estimates reflecting lifetime expenditures could be repeatedly applied to the same individual for many years, resulting in a
severe overestimation of the value of avoiding pollution-related chronic bronchitis.
1-4
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
verity, that is, a 50 percent reduction in severity from
the case described in the Viscusi study, then the itera-
tive procedure would proceed until the severity level
was half of what it started out to be.
The derivation of the WTP to avoid a case of pol-
lution-related chronic bronchitis is based on three com-
ponents, each of which is uncertain: (1) the WTP to
avoid a case of severe CB, as described in the Viscusi
study, (2) the severity level of an average pollution-
related case of CB (relative to that of the case de-
scribed by Viscusi), and (3) the elasticity of WTP with
respect to severity of the illness. Because of these three
sources of uncertainty, the WTP is uncertain. Based
on assumptions about the distributions of each of the
three uncertain components, a distribution of WTP to
avoid a pollution-related case of CB was derived by
Monte Carlo methods. The mean of this distribution,
which was about $260,000, is taken as the central ten-
dency estimate of WTP to avoid a pollution-related
case of CB. Each of the three underlying distributions
is described briefly below.
The distribution of WTP to avoid a severe case of
CB was based on the distribution of WTP responses
in the Viscusi study. Viscusi et al. derived respon-
dents' implicit WTP to avoid a statistical case of
chronic bronchitis from their WTP for a specified re-
duction in risk. The mean response implied a WTP of
about $1,000,000 (1990 dollars)2; the median response
implied a WTP of about $530,000 (1990 dollars).
However, the extreme tails of distributions of WTP
responses are usually considered unreliable. Because
the mean is much more sensitive to extreme values,
the median of WTP responses is often used rather than
the mean. Viscusi et al. report not only the mean and
median of their distribution of WTP responses, how-
ever, but the decile points as well. The distribution of
reliable WTP responses from the Viscusi study could
therefore be approximated by a discrete uniform dis-
tribution giving a probability of one ninth to each of
the first nine decile points. This omits the first five
and the last five percent of the responses (the extreme
tails, considered unreliable). This trimmed distribu-
tion of WTP responses from the Viscusi study was
assumed to be the distribution of WTPs to avoid a
severe case of CB. The mean of this distribution is
about $720,000 (1990 dollars).
The distribution of the severity level of an aver-
age case of pollution-related CB was modeled as a
triangular distribution centered at 6.5, with endpoints
2 There is an indication in the Viscusi paper that the dollar values in the paper are in 1987 dollars. Under this assumption the
dollar values were converted to 1990 dollars.
at 1.0 and 12.0. These severity levels are based on the
severity levels used in Krupnick and Cropper, 1992,
which estimated with relationship between In(WTP)
and severity level, from which the elasticity is derived.
The most severe case of CB in that study is assigned a
severity level of 13. The mean of the triangular distri-
bution is 6.5. This represents a 50 percent reduction
in severity from a severe case.
The elasticity of WTP to avoid a case of CB with
respect to the severity of that case of CB is a constant
times the severity level. This constant was estimated
by Krupnick and Cropper, 1992, in the regression of
In(WTP) on severity, discussed above. This estimated
constant (regression coefficient) is normally distrib-
uted with mean = 0.18 and standard deviation = 0.0669
(obtained from Krupnick and Cropper, 1992).
The distribution of WTP to avoid a case of pollu-
tion-related CB was generated by Monte Carlo meth-
ods, drawing from the three distributions described
above. On each of 16,000 iterations (1) a value was
selected from each distribution, and (2) a value for
WTP was generated by the iterative procedure de-
scribed above, in which the severity level was de-
creased by one percent on each iteration, and the cor-
responding WTP was derived. The mean of the re-
sulting distribution of WTP to avoid a case of pollu-
tion-related CB was $260,000.
This WTP estimate is reasonably consistent with
full COI estimates derived for chronic bronchitis, us-
ing average annual lost earnings and average annual
medical expenditures reported by Cropper and
Krupnick, 1990. Using a 5 percent discount rate and
assuming that (1) lost earnings continue until age 65,
(2) medical expenditures are incurred until death, and
(3) life expectancy is unchanged by chronic bronchi-
tis, the present discounted value of the stream of medi-
cal expenditures and lost earnings associated with an
average case of chronic bronchitis is estimated to be
about $77,000 for a 30 year old, about $58,000 for a
40 year old, about $60,000 for a 50 year old, and about
$41,000 for a 60 year old. A WTP estimate would be
expected to be greater than a full COI estimate, re-
flecting the willingness to pay to avoid the pain and
suffering associated with the illness. The WTP esti-
mate of $260,000 is from 3.4 times the full COI esti-
mate (for 30 year olds) to 6.3 times the full COI esti-
mate (for 60 year olds).
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Valuation of Other Morbidity Endpoints Avoided
WTP to avoid a day of specific morbidity end-
points, such as coughing or shortness of breath, has
been estimated by only a small number of studies (two
or three studies, for some endpoints; only one study
for other endpoints). The estimates for health end-
points involving these morbidity endpoints are there-
fore similarly based on only a few studies. However,
it is worth noting that the total benefit associated with
any reduction in pollutant concentrations is determined
largely by the benefit associated with the correspond-
ing reduction in mortality risk because the dollar value
associated with mortality is significantly greater than
any other valuation estimate. More detailed explana-
tions for valuation of specific morbidity endpoints is
given in Table 1-2.
Estimates of WTP may be understated for a couple
of reasons. First, if exposure to pollution has any cu-
mulative or lagged effects, then a given reduction in
pollution concentrations in one year may confer ben-
efits not only in that year but in future years as well.
Benefits achieved in later years are not included. Sec-
ond, the possible effects of altruism are not consid-
ered in any of the economic value derivations. Indi-
viduals' WTP for reductions in health risks for others
are implicitly assumed to be zero.
Table 1-2 summarizes the derivations of the eco-
nomic values used in the analysis. More detailed de-
scriptions of the derivations of lead-related endpoints
(hospital admissions for CHD and stroke, Lost IQ
points, IQ below 70, and hypertension) are discussed
in Appendix G.
Valuation of Welfare Effects
With the exception of agricultural benefits, eco-
nomic valuations for welfare effects quantified in the
analysis (i.e., household soiling damage, visibility and
worker productivity) are documented in Table 1-2. For
agricultural benefits, estimated changes in crop yields
were evaluated with an agricultural sector model,
AGSIM. This model incorporates agricultural price,
farm policy, and other data for each year. Based on
expected yields, the model estimates the production
levels for each crop, the economic benefits to con-
sumers, and the economic benefits to producers asso-
ciated with these production levels. To the extent that
alternative exposure-response relationships were
available, a range of potential benefits was calculated.
Appendix F documents the derivation of the monetary
benefits associated with improved agricultural pro-
duction. The derivation of the residential visibility
valuation estimate is discussed further below.
Visibility Valuation
Residential visibility has historically been valued
through the use of contingent valuation studies, which
employ surveys and questionnaires to determine the
economic value respondents place on specified
changes in visibility. A number of such studies have
been published in the economics literature since the
late 1970s, and have reported a wide range of result-
ing values for visibility, expressed as household will-
ingness to pay (WTP) for a hypothesized improve-
ment in visibility. Those studies were carefully re-
viewed for their applicability to the retrospective
analysis.
One limitation of many existing contingent valu-
ation studies of visibility is that they are local or re-
gional in scope, soliciting values for visibility from
residents of only one or two cities in a single region
of the country. Studies of visibility values from west-
ern cities, the most recent of which was published in
1981, have reported somewhat lower values than those
from eastern cities, raising the question of whether
eastern and western visibility are different commodi-
ties and should be valued differently in this analysis.
While the different visibility values reported in
the literature may appear to imply that visibility is
not valued equally by survey respondents in the east-
ern and western U.S., other evidence suggests that
eastern and western visibility are not fundamentally
different commodities, and that the retrospective ben-
efits calculations should not be based on separate east-
ern and western visibility values. For example,
NAPAP data indicate that California's South Coast
Air Basin, which encompasses Los Angeles and ex-
tends northward to the vicinity of San Francisco, has
median baseline visibility more characteristic of the
eastern U.S. than of other areas of the west (NAPAP
1991; lEc 1992, 1993a), reflecting the influence of
the higher humidity typical of coastal areas. While
inland areas of the west will tend to have lower hu-
midity, and hence greater baseline visibility than ei-
ther the eastern region or the western coastal zone,
such baseline visibility differences are accounted for
in the conversion from the visual range metric to
DeciView.
1-6
-------
Appendix I: Valuation of Human Health and Welfare Effects ofCriieria Pollutants
Perhaps the most compelling rationale for employ-
ing a single nationwide visibility valuation strategy
in the retrospective benefits analysis, however, relates
to the air quality modeling output used to calculate
the control and no-control scenario visibility profiles,
and its implications for the valuation of visibility as a
commodity. The RADM model and linear scaling
technique used for the retrospective analysis model
visibility improvements nationwide as changes in re-
gional atmospheric haze. In other words, although the
magnitude of visibility effects may vary between re-
gions, the model output does not distinguish between
a change in eastern visibility and a change in western
visibility as distinct phenomena. Thus, there is no clear
reason to value those same visibility changes differ-
ently in calculating the benefits of visibility improve-
ments. Consequently, a single, consistent valuation
basis has been applied to residential visibility improve-
ments nationwide for this analysis.
In light of advances in the state of the art of con-
tingent valuation over the last decade, the age of many
of the existing studies raised questions regarding their
suitability to serve as the primary basis for the vis-
ibility benefits estimates. A review of the survey and
data analysis methods used in the available studies
indicated that a study conducted for EPA by
McClelland et al. (1991) addressed many of the meth-
odological flaws of earlier studies, employing survey
methods and analytical techniques designed to mini-
mize potential biases (lEc 1992). Although this study
is unpublished, given its methodological improve-
ments over earlier studies it was chosen as the basis
for the central tendency of the visibility benefits esti-
mate, yielding an estimated value of $14 per unit im-
provement in DeciView as the annual household WTP
for visibility improvements (lEc 1997), as specified
in Table 1-2.
The difficulty of accurately defining the expected
statistical distribution of WTP values for visibility
improvements on the basis of published studies of
uneven reliability, along with the considerable varia-
tion in reported visibility values, led to the selection
of a hypothesized triangular distribution of values to
characterize the uncertainty in the visibility benefits
estimate. Reliance on any single study to estimate the
uncertainty range would be unlikely to adequately
characterize variations in visibility values that might
exist across cities, and in any case would fail to cap-
ture the full variability of visibility values reported in
the literature. Therefore, to ensure that the retrospec-
tive study characterizes the full range of uncertainty
in visibility values nationwide, the upper and lower
bounds of the triangular distribution were derived by
combining results from appropriate eastern and west-
ern residential visibility valuation studies.
Most of the existing residential visibility valua-
tion studies were found to suffer from part-whole bias,
which results from the failure to differentiate values
for visibility from those for other air quality ameni-
ties, such as reductions in adverse health effects. Of
the studies reviewed for this analysis, only the
McClelland study and Brookshire et al. (1979) have
attempted to obtain bids explicitly for visibility im-
provements (ffic 1992). Since part-whole bias will tend
to produce overstated values for visibility, reported
values from all studies that do not correct for part-
whole bias were adjusted prior to calculating the lower
bound of the uncertainty range. The upper bound of
the uncertainty range was calculated using the unad-
justed values from all studies, which is equivalent to
assuming that the entire value of respondents' stated
WTP for improved air quality can be attributed to in-
creased visibility.
The uncertainty range specified in Table 1-2 cal-
culated using a consensus function derived from a
regression analysis, incorporates a 25 percent adjust-
ment for part-whole bias (i.e., reported values were
multiplied by 0.25) in calculating the lower bound.
This represents an approximate midpoint of the range
defined by the McClelland study's finding that respon-
dents allocated, on average, 18.6 percent of their total
WTP to improvements in visibility, and Chestnut and
Rowe's (1989) conclusion that visibility improvement
accounted for 34 percent of the total WTP reported in
the Brookshire et al. study. Similarly, the "Denver
Brown Cloud" study results indicate that respondents
allocated 27.2 percent of their total WTP to visibility
improvements (Irwin et al. 1990). Therefore, the ap-
plication of a 25 percent adjustment for part-whole
bias to all but the McClelland and Brookshire values
would appear to be supported by the recent literature,
with the resulting consensus value representing a plau-
sible lower bound for the uncertainty range of visibil-
ity values. The consensus function approach, incor-
porating the part-whole bias adjustment, yields esti-
mated upper and lower bound values of $21 and $8,
respectively, for annual household WTP per unit im-
provement in DeciView.
1-7
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
c
t—<
•*
I
ates
Derivation of E
•8
1-S
& 3
u o.
o •»
stati
stati
: -M
•sij?
sifjif
g g g P -| ••§
S, g £> «j rg rg
^l|^l|
«.§ s ^3 --a -s
•a * S .a « a
jj go ••§ J .rS ^
2 « ••§ a S "»
SJ I !<§£
3 *= es * Z2 ««
8 S !.«
Jh^t|s
Ull^.
?. « Is .g,
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
tribution,
see text for f
of WTP wit
18 and stan
992). See te:
2.
to have a triangul
points at 1.0 and
onstant in the ela
ribution with
pnick and
study)
level 6.5
on); and
is n
= 0.
II
o
ft
g£
I
a
i
I
•i
£
ts
C
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
d
I
A
B
S
t
2
1«
P- 'o
O 13
Sw
S
W
Est
I
g
§
o
I
o
g,
g
vo
d
e
I
tu
ffi
o
S
o
§ S
oH
la
,Q T3
2 '3
^ &««§
0 "g S I £ I
rf * •= -a
o
o
1-10
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
1-11
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
£
s
¥
I
a
I
1
S
»
w
^
C!
'•5
Is
"S
I
I"
t£ -g
&•§•
1s
W
s estimate,
and the cos
e spent in
et al, 1993
1992.
s
ted
300.
alu
or a
ate
the
te
e
hosp
phys
Sour
phys
Applie
nent t
ents th
conge
|
CQ f^
es,
ser
Inc.
Abt
Applied the
onent to the
sents the vas
r ischemic h
E
C
spital c
ysician
urce of
mp
re
CO
cT
•" 2
II
•a u
§ S
o
ffi
a8
I
I
1
i
Ce
o\
1-12
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
1-13
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
£
s?
cidence
e Per
a
I
W
I
a>
III!
a
o
SO
I
.
de
Ce
1-14
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
ncertainty
IS,
o -o
5 CE]
«
tS
ffl S -o 22.
S
I
be
int
J5 6O A K « K R
lilllll
9 £ -a « «i S S
158.8
lasticity n
study of
PA, 1994)
gaged in
ed
tral
con
and
dail
U
S
£
WLDs
1-15
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Results of Valuation of Health
and Welfare Effects
Table 1-3 presents the results of combining the
economic valuations described in this Appendix with
the health and welfare effects results presented in
Appendix D. As noted in Appendix D, there are alter-
native estimates for some health and welfare impacts,
which form the basis of several alternative benefit
estimates. Each of the health effects estimates also
has quantified statistical uncertainty. The range of
estimated health and welfare effects, along with the
uncertain economic unit valuations, were combined
to estimate a range of possible results. The combin-
ing of the health and economic information used the
Monte Carlo method presented in Chapter 7. Table I-
3 shows the mean estimate results, as well as the mea-
sured credible range (upper and lower five percen-
tiles of the results distribution), of economic benefits
for each of the quantified health and welfare catego-
ries.
The results for aggregate monetized benefits were
also calculated using a Monte Carlo method. The re-
sults of the Monte Carlo simulations for the economic
values for each of the major endpoint categories are
presented in Table 1-4. Note that for the upper and
lower fifth percentiles the sum of the estimated ben-
efits from the individual endpoints does not equal the
estimated total. The Monte Carlo method used in the
analysis assumes that each health and welfare end-
point is independent of the others. There is a very low
probability that the aggregate benefits will equal the
sum of the fifth percentile benefits from each of the
ten endpoints.
Table 1-5 shows the estimated total benefits ranges
for the four modeled target years of this study: 1975,
1980,1985, and 1990. The results of the Monte Carlo
simulations of the aggregate economic benefits for
these four target years are depicted in Figure 1-1.
Table 1-6 examines the impact of limiting the
scope of the analysis to locations with more certain
air quality estimates. The main analysis (as shown in
Tables 1-3 through 1-5) covers almost the entire popu-
lation of the 48 States.3 However, the air quality in-
formation is less certain for locations far from a moni-
tor. Table 1-6 presents the results of limiting the analy-
sis to people living within 50 km of an ozone, NO2,
SO2, or CO monitor, or in counties with a PM moni-
tor. The availability of monitors changes over time.
Hence the proportion of the population included in
this analysis changes over time as well. Table 1-6 in-
dicates that approximately a quarter of the total ben-
efits estimated in the main analysis comes from areas
with less certain air quality information.
The results of the "all U.S. population" analysis
provides a more accurate depiction of the pattern of
economic benefits across years. The accuracy of the
scale of incidence is less certain. These results pro-
vide a better characterization of the total direct ben-
efits of the Clean Air Act in the lower 48 states than
do the "monitored area only" results because the lat-
ter completely omits historical air quality improve-
ments for about 25 percent of the population. How-
ever, the "all U.S. population" results rely on uncer-
tain extrapolations of pollution concentrations, and
subsequent exposures, from distant monitoring sites
to provide coverage for the 25 percent or so of the
population living far from air quality monitors. Thus,
the main results presented in Tables 1-3 through 1-5
include important uncertainties.
Uncertainties
The uncertainty ranges for the results on the
present value of the aggregate measured monetary
benefits reported in Table 1-3 reflect two important
sources of measured uncertainty:
• uncertainty about the avoided incidence of
health and welfare effects deriving from the
concentration-response functions, including
both selection of scientific studies and statis-
tical uncertainty from the original studies; and
• uncertainty about the economic value of each
quantified health and welfare effect.
These aggregate uncertainty results incorporate many
decisions about analytical procedures and specific
assumptions discussed in the Appendices to this re-
port.
In order to provide a more complete understand-
ing of the economic benefit results in Table 1-3, sen-
sitivity analyses examine several additional important
aspects of the main analysis. First, this section ex-
3 Except for lead, two to five percent (depending on pollutant) of the population who live in sparsely populated areas are
excluded from the main analysis to maximize computer efficiency. All of the population of the 48 states is included in the lead
analysis.
1-16
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
jTable 1-3, Critetia f ollutants Healli,and -Welfare Benefits ".E^tfapoMte^tVE
' Population Pfeseat,,Yalae.(in 1990 using 5% discount rate) of ..Benefits from 1970 - 1990 (in billions of
.1990 dojjars). K _ \, '';--- > >*' ' ~ , ' \' -' " -X/'T' ' ">> V "~*> "
EadJ^nuBlt ,,' "f
^Mortality „ ' "" ' s ~r '
Mortality
~ -/KJ<70 •, f - „<'"*'
Hypertension ^ */, " y>
' Coronary Heart Diseased , -
Atherothrombotie brak infarction *
Initial eerebio vascular accident A ^
Hospital Admissions „ , ^ '- ,
*AH Respiratory " ^ ^
-^ v *COED+ Pneumonia >'T » - ',/'
/ ^ ^ V
Jsclieraic HeartDisease '' " ' '
Congestive Heait Failure "J '
Other Respiratory- Related" Ailments, -*>
_ _ iShOrtne's^ of Syeath, days' ' v , „ ,,
" 'IJAeuteBroncJiitiS;^ ' £' ^ <
**I^>per& Lejwer Respiratory Symptoms
**-xiAdyJt8v ' '/ " ' * '''*'•',
< "Anyofl9Acute§ymptomss'; , ~, t
'-•; ,m ,„ *" '' -*<.'
"C,""^ Asthma A (tacks < v, v "' , '"','•
Th'crease inRfSpirltoryJlIness \"
* Any Symptom *<* " f ^ ;/
Re$Uict|d Acfivity and Work i^ssBays' *
j ,^« WRAD -- ~ ,-/ ,., - ; -^
( "VVork Loss^ys {WLD) t >^ / ' - -T
Human Wel^ie « ",' /^ ' ' '!„,„
' ;H-''
Visibaity-Eastern^lS. ' ~* ""' \ *
;,>f Jfecreased Worker Proftictivity ^ ( ^!
^ ' J
Peffibtraitfs)
4 '
PM
^Lead "
'PM - ;;
Lead
-"y
Lead - < '
'Eead, * / *
LeaS" /
Lead"'"1-
•Lead
"" <
PM&03 A '*
PM&03 < "
PM
PM&CO
t 4' - '
PM] -/%
PM "t
PM '' '
W > ' ••
PM^03. »m
i #*
PM'&O3 ,„
NQ2 "'
SO2
•f. f
EM&O3"
PM , ^ ^
",.,'- "' '\y
?M""
particuiates '
3
O3
MooiciUKfHi^flrnM
',y Sth.%Be ' ' '' Mzaa.*-
' " Sv >
> ' ,' $2,369 * > " , '$16,632
: ' $m ' ^' '$U3? v
$409 ^ S3t3I3
•> $ 24&' > x ^77"
v^ ^*i»^i~rw > ^ w ^ -*J/J!
^ . $15 , " -~- '$22
$77--. , > v $98,
*< $0 ' ' $13^ '
$1, ' - » 510
-' * $&• / ^"$16
s ~~ ' ' "
" ' ? ! "'"' s ,
$8 $9
~$S _ ' ' $9^
'/ v * $1 *"-^ ,**'',
, $3 ' ' . ,' $5
# ' ' / ' x *' *
-' '' ' *", $0 - '' ' -$6>
7 *i - L/ ,$7" "•
'$1 << * > ^$2
/ * ^ Vyj /.
^ ^4 ^ / *fe4A
j' '/y ,' ' * s - " " ° f>, *
' $o ^ so""
$1 ' , ' '> , $2
', /so ";w '
v s> ^b*vft < * CO<
' 5^" Ipo- J ^
f ,""$30 J\^ , $34 '
{ ^. ''
"' ' ' '$6 " $74
I $38,., "' '' ^4 "
<- _, '$3, ' - „ ' ' i$3
^^ t ^ ^ C'^'S -Iv
^ *i-I :, ' JW5J *
[I \
^»St%a:^
< ' -' ^
"$40,597
^ $3^10,
#? - ''
^, ^XO
•^ " "-' '$29
$120
''$40
! , r $30
" ,$45
^ J
$11
$10'
" " ,$?*
'* ,^ " ,$J7
$18
" -: !> $4
$117
' « ^
' $1
-, s $4
1 ', $o
, $123
$3>
*
'; ' < ${92
, ' '" $71
$3
J ^- »./x ''^ ' ^ ^x> , y™,^
:vfoUowingendpoin.ts,weiE treated as alternatives: >1
*HQSpitalaamissioAs for COPDcorabiueS wcfth those fotpastimonla are teated asln equalty-\veighte(J alternative to hospital
* ^admissions for all respiratory illnesses. ( ™^ •> , ' ' •> ,-, ,t '
*«Rje 'deMfions^of ac^VbinncWtis ajid_opj>er a«d lower respiratory 81ness oveiiap,- both studies comt trouble breathing, '
£ dry c^ugh, and wheezing-in their estimates. T|ese twostudies aretreated as alternatives, wfeich reflects the variability of
1-17
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Table 1-4.
Population
Endpoint Pollutants) ,' ,
Mortality PM, ",„„,, '',-/'' J
vlortality Pb t \, •* -
Chronic Obstructive Pulmonary Disease PMU" /,; ,,' -, ,'.'•>„
IQ (Lost IQ Pts. + Children w/ IQ<70) " Pb - ' '.^, .
Hypertension P,b "•-, > .'* „,
Hospital Admissions PM,,O3 ? Pb,&, CO'
Respiratory-Related Symptoms.RestrictedPMVOS.^p^'&^SO^
Activity, & Decreased Productivity , •" - -<*- •'"'?-/''"',,
Soiling Damage PM , , ,,s • , - a;<
Visibility particulates- ' > -;, ,„
Agriculture (Net Surplus) 03 r ' \, ' -
. 5^h %^
fv , $2,369';
,"'"'- »/; $121;-'
% f'f
'•<.','. ' T^38;
*; - "'•-
I'esenT^aiw;
- ,„ Mean - ;
*f '' ''•>$•
''*;,' ''f':$j'4'
*/ ; _, Vi^ '
'95ttf>'fle"
^'^- ~'$j'2p~
"'.,».>
;''^-',
^" ?
Table 1-5. Monte Carlo Simulation ModelResult^forT^getYearsi'Plu&Pjeserit Valu
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
Figure 1-1. Monte Carlo Simulation Model Results for Target Years (in billions of 1990 dollars).
$3,000 -
$2,500 -
CO
| $2,000 -
S
^,
| $1,500 -
a
3 $1,000 -
o
$500 -
$0 -
•wwwm
•MMMB
r — rH 95th%
^ 95th%
^ Mean
^j 5th%
^^•^
'" 'k'-
'%!,'
H'
' *; ^
• ,„-,,
^£
•^ 95th%
•^ Mean
•^j 5th%
maatmm
'>;
X ',
, ^
'^
%
";::-
^ 95th%
^ Me an
•^ 5th%
~& •
iky
'-""- ,..<
>**&£
'''-!"
•MWMMW
V
r ^ >!
^ Mean
^ 5th%
1975 1980 1985 1990
_ I '*-£<.< r'~ s^ftf^-'^,, ?',/'- ,/, ' ,. - ,„ ,„' "> / ' '' f~ ,
fTabje 1-6. CQrapariso»,of 1990 (Single Year):Mofletrzed -Benefits by £ndpoint for 48~Sj;aie ',/'
-'Population and'Monltored Areas (in'millionsoif 1990dollars). *,"* -''"«•• ' ,
y "" >A •; VS'A ff * / ^ ^ ^ ?>-'
Mean Estimate of Monetized
" " < Benefits ,^' .,,
_-4miflloBsoll990-doilars) >
'48 State Pop,
Monitored
s Area's*"1-,
Mortality' *,,, _ ^ ^x'** " '"- 'PM ' 'x^
M'oriaSit'y „ "s '" > ^ , » >-"""' * " Pb - ' ' -~
Chronic Brojijcjtitis - /s ."' " \s-, >PM ^ 's,
IQ (L'ostlQ Points + Cbiidrea with" IQ|s,70) Pb '" ; V" - '
HypertensioK- / * /" . „ s, >pg ^ '->
Hospital A4m5ssions x ^'J /x' PM, O3, Pb, &'CO »
R expiratory-R elated Symptoms, R estricted PM,,O3, NO2, & SO2
Activity, & Decreased^Productivity / ,
Soifing Damage f '^ ' }; * PM ; , ^ ',,_ '>
VjsibHity ^ -<- s ^- ' - participates,, "''-'"'
Agricultu|e 04et Surplus) ^ " *„ ' OS ' '•- „ " *
$892,390
, $580^299
$179,755,,tx^ , $120^)$3s
"$32,381 ^ '"$32,381'
$8,584 " ,,$8,584,
$3^94
$10,2^9 ' \ „ ,$7,089
$3,964
$3,382
$986
$2,709
» * "> »
$3,382
,/ $986
TOTAL <$MiIlions>
$1,247,713
"* Monitored areas «fe^ those within 50 km of anO3,NOZ, SO2,or CO monitor or^aPM-monitored county,;
The ""48 State Population" modeling estimate capture's benefits for^popuSatiQnsJn onmOnitOred^aTeas. Air
pollution concentrations tat thesse areas are assigned based-on eoncentrations'measured at the- c to sest
jaonitor, for O3> ^fO2, SO2, and'CO. PM coacentrations m anmoaltored counties are derived by ^
;extrapolatingthoseIn,mOB!|ored,countIes.» ' > 'Z ^''""' '" ' < ^ " ''^
* /
1-19
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
plores the effect of selecting alternative discount rates
on the aggregate present value benefits estimation.
Second, this section examines the sources of the mea-
sured aggregate uncertainty, identifying which of the
measured uncertainty components of incidence and
valuation for individual health effects categories drive
the overall uncertainty results. Third, this section ex-
amines several issues involving the estimated eco-
nomic benefits of mortality.
The Effect of Discount Rates
The main analysis reflected in present value re-
sults shown in Table 1-3 uses a five percent discount
rate. The discount rate primarily enters the calcula-
tions when compounding the economic benefits esti-
mates from individual years between 1970 and 1990
to estimate the present value of the benefits in 1990.
The discount rate also directly enters in the calcula-
tions of the economic values of an IQ point and an
initial case of coronary heart disease.4 There is con-
siderable controversy in the economics and policy lit-
erature about the appropriate discount rate to use in
different settings. Major alternatives recommended by
various authors include a discount rate based on the
social discount rate (typical estimates are in the 2 to 3
percent range), and a discount rate based on the risk-
free rate of return on capital (typically in the 7 to 10
percent range). Table 1-7 presents the aggregate un-
certainty results using three different discount rates:
3 percent, 5 percent and 7 percent. While the aggre-
gate benefits estimates are. sensitive to the discount
rate, selecting one of these alternative discount rates
affects the aggregate benefits estimates by only about
15 percent.
The Relative Importance of Different
Components of Uncertainty
The estimated uncertainty ranges in Table 1-3 re-
flect the measured uncertainty associated with both
avoided incidence and economic valuation. A better
understanding of the relative influence of individual
uncertain variables on the overall uncertainty in the
analysis can be gained by isolating the individual ef-
fects of important variables on the range of estimated
benefits. This can be accomplished by holding all the
inputs to the Monte Carlo uncertainty analysis con-
stant (at their mean values), and allowing only one
variable — for example, the economic valuation of
mortality —to vary across the range of that variable's
uncertainty. The sensitivity analysis then isolates how
this single source of variability contributes to the varia-
tion in estimated total benefits. The results are sum-
marized in Figure 1-2. The nine individual uncertainty
factors that contribute the most to the overall uncer-
tainty are shown in Figure 1-2, ordered by the relative
significance of their contribution to overall uncer-
tainty. Each of the additional sources of quantified
uncertainty in the overall analysis not shown contrib-
ute a smaller amount of uncertainty to the estimates
of monetized benefits than the sources that are shown.
Table 1-7. Effect of Alternative Discount J^tes Otf-Present-Value of Total Monetized Benefits for '/
1970 to 1990(in trillions of 1990^d ' "' " " ' '" '
Present Value in 1990 of Total Benefits/;
(Trillions of 1990 Dollars)
5th percentile _" ' 'f
Mean „ " ^
95th percentile , *v
' " ' " -,
3%
'~ $4.9
^ -,y y ' i/^ ^
'^$42.1- -
'V , , 'X
, 5%'-, ' 7%
„ ;, $5/6 y ^ ;*J $6:5
^p^^*^- •'' i|JjiJ*Q
" " $49.4 $57,5
Present value reflects compounding of benefits ftoaM97-l to 1990.
t A. ''£.'•>•'•
4 The estimated economic value of lost IQ points due to lead exposure is based on the present value of the impact on lifetime
earnings. A discount rate is required to calculate that present value. The impact on income primarily occurs during adulthood, which
is 20 to 70 years after the initial lead exposure. This significant lag results in the discount rate having a significant impact on the
estimated economic benefits of the IQ loss. Similarly, the cost of illness estimate for an initial case of CHD includes the present value
of the annual stream of medical costs incurred after the event, the calculation of which requires an estimate of the discount rate.
1-20
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
Figure 1-2. Uncertainty Ranges Deriving From Individual Uncertainty Factors.
$50-
C*
J $45-
£ $40-
1""am^5m"°-Jc '"^ > 2-5
3 2 c ' ^ ^' £ £ ^ "g "ggo
2 — x: .c — -r--— —
< Q. o o -t- = -55
CO _3
Because of the multiple uncertainties in the ben-
efits estimation, the total estimated present value of
the monetary benefits of the 1970 to 1990 Clean Air
Act range from a low of about $5.6 trillion to a high
of about $49.4 trillion (in 1990 dollars, discounted at
five percent). Most of the uncertainty in the total esti-
mated benefit levels comes from uncertainty in the
estimate of the economic valuation of mortality, fol-
lowed by the uncertainty in the incidence of mortality
from PM (as a surrogate for all non-lead air pollu-
tion). The incidence of lead-induced mortality also
has a significant influence on the overall uncertainty.
The importance of mortality is not surprising, because
the benefits associated with reduced mortality are such
a large share of overall monetized benefits.
The uncertainty in both the incidence and valua-
tion of chronic bronchitis are the two other signifi-
cant factors driving the overall uncertainty range. The
modeled uncertainty in the other remaining health and
welfare endpoints in the analysis contribute relatively
small amounts to the overall uncertainty in the esti-
mate of total monetary benefits of the Clean Air Act.
Most of these other endpoints account for a relatively
small proportion of the overall benefits estimates,
making it unlikely that they could contribute signifi-
cantly to the overall uncertainty. Estimates of either
the mean values or standard errors of these variables
are generally very small relative to estimated total
monetary benefits.
Economic Benefits Associated with
Reducing Premature Mortality
Because the economic benefits associated with
premature mortality are the largest source of mon-
etized benefits in the analysis, and because the uncer-
tainties in both the incidence and value of premature
mortality are the most important sources of uncertainty
in the overall analysis, it is useful to examine the
mortality benefits estimation in greater detail.
The analytical procedure used in the main analy-
sis to estimate the monetary benefits of avoided pre-
mature mortality assumes that the appropriate eco-
nomic value for each incidence is a value from the
currently accepted range of the value of a statistical
life. As discussed above, the estimated value per pre-
dicted incidence of excess premature mortality is
modeled as a Weibull distribution, with a mean value
of $4.8 million and a standard deviation of $3.2 mil-
lion. This estimate is based on 26 studies of the value
of mortal risks.
1-21
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
There is considerable uncertainty as to whether
the 26 studies on the value of a statistical life provide
adequate estimates of the value of a statistical life
saved by air pollution reduction. Although there is
considerable variation in the analytical designs and
data used in the 26 underlying studies, the majority of
the studies involve the value of risks to a middle-aged
working population. Most of the studies examine dif-
ferences in wages of risky occupations, using a wage-
hedonic approach. Certain characteristics of both the
population affected and the mortality risk facing that
population are believed to affect the average willing-
ness to pay (WTP) to reduce the risk. The appropri-
ateness of a distribution of WTP estimates from the
26 studies for valuing the mortality-related benefits
of reductions in air pollution concentrations therefore
depends not only on the quality of the studies (i.e.,
how well they measure what they are trying to mea-
sure), but also on (1) the extent to which the subjects
in the studies are similar to the population affected by
changes in pollution concentrations, and (2) the ex-
tent to which the risks being valued are similar. As
discussed below, there are possible sources of both
upward and downward bias in the estimates provided
by the 26 studies when applied to the population and
risk being considered in this analysis.
If the individuals who die prematurely from air
pollution are consistently older than the population in
the valuation studies, the mortality valuations based
on middle-aged people may provide a biased estimate
of the willingness to pay of older individuals to re-
duce mortal risk. There is some evidence to suggest
that the people who die prematurely from exposure to
ambient particulate matter tend to be older than the
populations in the valuation studies. In the general
U.S. population far more older people die than younger
people; 88 percent of the deaths are among people
over 64 years old. It is difficult to establish the pro-
portion of the pollution-related deaths that are among
the older population because it is impossible to iso-
late individual cases where one can say with even rea-
sonable certainty that a specific individual died be-
cause of air pollution.
There is considerable uncertainty whether older
people will have a greater willingness to pay to avoid
risks than younger people. There is reason to believe
that those over 65 are, in general, more risk averse
than the general population, while workers in
wage-risk studies are likely to be less risk averse than
the general population. More risk averse people would
have a greater willingness to pay to avoid risk than
less risk averse people. Although the list of recom-
mended studies excludes studies that consider only
much-higher-than- average occupational risks, there
is nevertheless likely to be some selection bias in the
remaining studies — that is, these studies are likely to
be based on samples of workers who are, on average,
more risk-loving than the general population. In con-
trast, older people as a group exhibit more risk averse
behavior.
In addition, it might be argued that because the
elderly have greater average wealth than those
younger, the affected population is also wealthier, on
average, than wage-risk study subjects, who tend to
be blue collar workers. It is possible, however, that
among the elderly it is largely the poor elderly who
are most vulnerable to pollution-related mortality risk
(e.g., because of generally poorer health care). If this
is the case, the average wealth of those affected by a
pollution reduction relative to that of subjects in
wage-risk studies is uncertain. In addition, the work-
ers in the wage-risk studies will have potentially more
years remaining in which to acquire streams of in-
come from future earnings.
Although there may be several ways in which job-
related mortality risks differ from air pollution-related
mortality risks, the most important difference may be
that job-related risks are incurred voluntarily whereas
air pollution-related risks are incurred involuntarily.
There is some evidence (see, for example, Violette
and Chestnut, 1983) that people will pay more to re-
duce involuntarily incurred risks than risks incurred
voluntarily. If this is the case, WTP estimates based
on wage-risk studies may be downward biased esti-
mates of WTP to reduce involuntarily incurred air
pollution-related mortality risks.
Finally, another possible difference related to the
nature of the risk may be that some workplace mor-
tality risks tend to involve sudden, catastrophic events
(e.g., workplace accidents), whereas air pollution-re-
lated risks tend to involve longer periods of disease
and suffering prior to death. Some evidence suggests
that WTP to avoid a risk of a protracted death involv-
ing prolonged suffering and loss of dignity and per-
sonal control is greater than the WTP to avoid a risk
(of identical magnitude) of sudden death. Some work-
place risks, such as risks from exposure to toxic chemi-
cals, may be more similar to pollution-related risks. It
is not clear, however, what proportion of the work-
place risks in the wage-risk studies were related to
workplace accidents and what proportion were risks
1-22
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria PoHutants
from exposure to toxic chemicals. To the extent that
the mortality risks addressed in this assessment are
associated with longer periods of illness or greater pain
and suffering than are the risks addressed in the valu-
ation literature, the WTP measurements employed in
the present analysis would reflect a downward bias.
The direction of bias resulting from the age dif-
ference is unclear, particularly because age is con-
founded by risk aversion (relative to the general popu-
lation). It could be argued that, because an older per-
son has fewer expected years left to lose, his WTP to
reduce mortality risk would be less than that of a
younger person. This hypothesis is supported by one
empirical study, Jones-Lee et al. (1985), that found
the value of a statistical life at age 65 to be about 90
percent of what it is at age 40. Citing the evidence
provided by Jones-Lee et al. (1985), a recent sulfate-
related health benefits study conducted for EPA (U.S.
EPA, 1995) assumes that the value of a statistical life
for those 65 and over is 75 percent of what it is for
those under 65.
There is substantial evidence that the income elas-
ticity of WTP for health risk reductions is positive
(see, for example, Alberini et al., 1994; Mitchell and
Carson, 1986; Loehman and Vo Hu De, 1982; Gerking
et al., 1988; and Jones-Lee et al., 1985), although there
is uncertainty about the exact value of this elasticity.
Individuals with higher incomes (or greater wealth)
should, then, be willing to pay more to reduce risk, all
else equal, than individuals with lower incomes or
wealth. Whether the average income or level of wealth
of the population affected by PM reductions is likely
to be significantly different from that of subjects in
wage-risk studies, however, is unclear, as discussed
above.
The need to adjust wage-risk-based WTP esti-
mates downward because of the likely upward bias
introduced by the age discrepancy has received sig-
nificant attention (see Chestnut, 1995; lEc, 1992). If
the age difference were the only difference between
the population affected by pollution changes and the
subjects in the wage-risk studies, there might be some
justification for trying to adjust the point estimate of
$4.8 million downward. Even in this case, however,
the degree of the adjustment would be unclear. There
is good reason to suspect, however, that there are bi-
ases in both directions. Because in each case the ex-
tent of the bias is unknown, the overall direction of
bias in the mortality values is similarly unknown.
Adjusting the estimate upward or downward to com-
pensate for any one source of bias could therefore in-
crease the degree of bias. Therefore, the range of val-
ues from the 26 studies is used in the primary analy-
sis without adjustment.
Examining the sensitivity of the overall results to
the mortality values can help illuminate the potential
impacts of alternative mortality valuations. As men-
tioned above, a contractor study performed for EPA
used one approach to evaluate the economic value of
sulfate-related human health improvements resulting
from 1990 Clean Air Act Amendments Title IV acid
rain controls. That study assumed that 85 percent of
the people dying from sulfates (an important compo-
nent of paniculate matter) were over 65, and that
people over 65 have a willingness to pay to avoid a
mortal risk that is 75 percent of the values that middle-
aged people have. Using this approach, the value of
an average statistical life (using a weighted average)
is reduced to 79 percent of the previous value.
If statistical life-years lost are used as the unit of
measure, rather than statistical lives lost, the benefit
attributed to avoiding a premature death depends di-
rectly on how premature it is. One way to estimate
the value of a statistical life-year assumes that the value
of a statistical life is directly related to remaining life
expectancy and a constant value for each life-year.
Such an approach results in smaller values of a statis-
tical life for older people, who have shorter life ex-
pectancies, and larger values for younger people. For
example, if the $4.8 million mean value of avoiding
death for people with a 35 year life expectancy is as-
sumed to be the discounted present value of 35 equal-
valued statistical life-years, the implied value of each
statistical life-year is $293,000 (using a 5% discount
rate). The average number of life-years lost by indi-
viduals dying prematurely from exposure to PM is 14
years. This average is obtained by multiplying the
predicted number of PM-related premature deaths in
each age category by the life expectancy for that age
category and dividing by the total number of PM-re-
lated premature deaths.) Using $293,000 per life-year,
the discounted present value of a statistical life for a
person with 14 years of expected life remaining (e.g.,
a 70 year old) is $2.9 million). If statistical life-years
lost are used to value fatal risks, however, other
sources of uncertainty are introduced in the valuation
process.
1-23
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
If statistical life-years lost is the unit of measure,
the value of a statistical life lost depends on (1) how
many years of expected life are lost, (2) the
individual's discount rate, and (3) whether the value
of an undiscounted statistical life-year is the same no
matter which life-year it is (e.g., the undiscounted
value of the seventy-fifth year of life is the same as
the undiscounted value of the fortieth year of life).
Each of these is uncertain. The uncertainty surround-
ing the expected years of life lost by an individual
involves the uncertainty about whether individuals
who die from exposure to air pollution are average
individuals in the demographic (e.g., age-gender-race)
classification to which they belong. The uncertainty
surrounding individuals' discount rates is well docu-
mented. Finally, even if it is assumed that all life-years
are valued the same (apart from differences due to
discounting), the value of a statistical life-year is de-
rived from the value of a statistical life (of a 40 year
old) and the discount rate, each of which is uncertain.
Using life-years lost as the unit of measure means
that, rather than estimating a single value of a statisti-
cal life lost (applicable to all ages), the analysis would
instead estimate age-specific values of statistical lives
lost. It is unclear whether the variability of estimates
of age-specific values of statistical lives lost (in par-
ticular, for ages greater than the average age of work-
ers in the wage-risk studies) would be less than or
greater than the variability of the original estimate of
the value of a statistical life lost from which they would
be derived. If there is an age-related upward bias in
the central tendency value of a statistical life that is
larger than any downward bias, then valuing life-years
rather than lives lost may decrease the bias. Even this,
however, is uncertain.
In spite of the substantial uncertainties and pau-
city of available information, this section presents an
example of a preliminary estimate of the present value
of avoided premature mortality using the life-years
lost approach. The basic approach is to (1) estimate
the number of pollution-related premature deaths in
each age category, (2) estimate the average number
of life-years lost by an individual in a given age cat-
egory dying prematurely, and (3) using the value of a
statistical life-year of $293,000, described above (as-
suming that the undiscounted value of a life-year is
the same no matter when in an individual's life it is)
and assuming a five percent discount rate, calculate
the value of a statistical life lost in each age category.
To obtain estimates of the number of air pollu-
tion-related deaths in each age cohort, it is preferable
to have age-specific relative risks. Many of the epide-
miological studies, however, do not provide any esti-
mate of such age-specific risks. In this case, the age-
specific relative risks must be assumed to be identi-
cal.
Some epidemiology studies on PM do provide
some estimates of relative risks specific to certain age
categories. The limited information that is available
• suggests that relative risks of mortality associated with
exposure to PM are greater for older people. Most of
the available information comes from short-term ex-
posure studies. There is considerable uncertainty in
applying the evidence from short-term exposure stud-
ies to results from long-term (chronic exposure) stud-
ies. However, using the available information on the
relative magnitudes of the relative risks, it is possible
to form a preliminary assessment of the relative risks
by different age classes.
The analysis presented below uses two alterna-
tive assumptions about age-specific risks: (1) there is
a constant relative risk (obtained directly from the
health literature) that is applicable to all age cohorts,
and (2) the relative risks differ by age, as estimated
from the available literature. Estimates of age-spe-
cific PM coefficients (and, from these, age-specific
relative risks) were derived from the few age-specific
PM coefficients reported in the epidemiological lit-
erature. These estimates in the literature were used to
estimate the ratio of each age-specific coefficient to a
coefficient for "all ages" in such a way that consis-
tency among the age-specific coefficients is preserved
— that is, that the sum of the health effects incidences
in the separate, non-overlapping age categories equals
the health effects incidence for "all ages." These ra-
tios were then applied to the coefficient from Pope et
al. (1995). Details of this approach are provided in
Post and Deck (1996). Because Pope et al. considered
only individuals age 30 and older (instead of all ages),
the resulting age-specific PM coefficients may be
slightly different from what they would have been if
the ratios had been applied to an "all ages" coeffi-
cient. The differences, however, are likely to be mini-
mal and well within the error bounds of this exercise.
The age-specific relative risks used in the example
below assume that the relative risks for people under
65 are only 16 percent of the population-wide aver-
age relative risk, the risks for people from 65 to 74
are 83 percent of the population-wide risk, and people
1-24
-------
Appendix I; Valuation of Human Health and Welfare Effects of Criteria Pollutants
75 and older have a relative risk 55 percent greater
than the population average. Details.of this approach
are provided in Post and Deck (1996).
The life-years lost approach also requires an esti-
mate of the number of life-years lost by a person dy-
ing prematurely at each given age. The average num-
ber of life-years lost will depend not only on whether
relative risks are age-specific or uniform across all
age groups, but also on the distribution of ages in the
population in a location. As noted above, using the
same relative risk for all age categories, the average
number of life-years lost in PM-related premature
deaths in the United States was estimated to be 14
years. Using the age-specific relative risk estimates
developed for this analysis, the average number of
life-years lost becomes 9.8 years. In a location with a
population that is younger than average in the United
States, the same age-specific relative risks will pro-
duce a larger estimated average number of life-years
lost. For example, using the same age-specific rela-
tive risks, the average number of life-years lost hi PM-
related premature deaths in Los Angeles County,
which has a younger population, is estimated to be
15.6 years.
The present value benefits estimates for PM-re-
iated mortality using the alternative approaches dis-
cussed above are shown in Table 1-8. Table 1-8 is based
on a single health study: Pope et al., 1995. Alterna-
tive studies, or the uncertainty approach used in the
primary analysis, would result in a similar pattern of
the relationship between valuation approaches. The
pattern of monetized mortality benefits across the dif-
ferent valuation procedures shown in Table 1-8 is es-
sentially invariant to the particular relative risk and
the particular dollar value used.
As noted above, the life-years lost approach used
here assumes that people who die from air pollution
are typical of people in their age group. The estimated
value of the quantity of life lost assumes that the people
who die from exposure to air pollution had an aver-
age life expectancy. However, it is possible that the
people who die from air pollution are already in ill
health, and that then: life expectancy is less than a
typical person of then- age. If this is true, then the num-
ber of life years lost per PM-related death would be
lower than calculated here, and the economic value
would be smaller.
The extent to which adverse effects of particulate
matter exposure are differentially imposed on people
of advanced age and/or poor health is one of the most
important current uncertainties in air pollution-related
health studies. There is limited information, prima-
rily from the short-term exposure studies, which sug-
gests that at least some of the estimated premature
mortality is imposed disproportionately on people who
are elderly and/or of poor health. The Criteria Docu-
ment for Particulate Matter (U.S. EPA, 1996) identi-
fies only two studies which attempt to evaluate this
disproportionality. Spix et al. (1994) suggests that a
small portion of the PM-associated mortality occurs
in individuals who would have died in a short time
anyway. Cifuentes and Lave (1996) found that 37 to
87 percent of the deaths from short-term exposure
could have been premature by only a few days, al-
though their evidence is inconclusive.
'M^l996^m&nomjof'l9^>^
"/•y^t ''/'A 'i d !• '"'•L i* •> ' "*'"? ' S V/ f''s
ValuafionProceclife ;
PritnawAtuaysK^tW ' '''''
'',, , X :;>-., "> ;-~^~;;, -V^t,; '•'-'- *"; -' -.'ft'/. - ;• s.A^v;, ,t/-v ' ' ' : '
" ' "*'" " ' * '"
J;5,% dlSCO,untmg,, ; ?^-v ''" ^'--~'«:
ApptDxifriat^'age-sfaecific relative risk.- valuatioiii usiri^ 5%' di scountirt g -
y .- t
-!'^Mortality Benefits"
'$16.6
<.
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Prematurity of death on the order of only a few
days is likely to occur largely among individuals with
pre-existing illnesses. Such individuals might be par-
ticularly susceptible to a high PM day. To the extent
that the pre-existing illness is itself caused by or ex-
acerbated by chronic exposure to elevated levels of
PM, however, it would be misleading to define the
prematurity of death as only a few days. In the ab-
sence of chronic exposure to elevated levels of PM,
the illness would either not exist (if it was caused by
the chronic exposure to elevated PM) or might be at a
less advanced stage of development (if it was not
caused by but exacerbated by elevated PM levels).
The prematurity of death should be calculated as the
difference between when the individual died in the
"elevated PM" scenario and when he would have died
in the "low PM" scenario. If the pre-existing illness
was entirely unconnected with chronic exposure to
PM in the "elevated PM" scenario, and if the indi-
vidual who dies prematurely because of a peak PM
day would have lived only a few more days, then the
prematurity of that PM-related death is only those few
days. If, however, in the absence of chronic exposure
to elevated levels of PM, the individual's illness would
have progressed more slowly, so that, hi the absence
of a particular peak PM day the individual would have
lived several years longer, the prematurity of that PM-
related death would be those several years.
Long-term studies provide evidence that a por-
tion of the loss of life associated with long-term ex-
posure is independent of the death from short-term
exposures, and that the loss of life-years measured in
the long-term studies could be on the order of years.
If much of the premature mortality associated with
PM represents short term prematurity of death im-
posed on people who are elderly and/or of ill health,
the estimates of the monetary benefits of avoided
mortality may overestimate society's total willingness
to pay to avoid particulate matter-related premature
mortality. On the other hand, if the premature mortal-
ity measured in the chronic exposure studies is de-
tecting excess premature deaths which are largely in-
dependent of the deaths predicted from the short term
studies, and the disproportionate effect on the elderly
and/or sick is modest, the benefits measured in this
report could be underestimates of the total value. At
this time there is insufficient information from both
the medical and economic sciences to satisfactorily
resolve these issues from a theoretical/analytical stand-
point. Until there is evidence from the physical and
social sciences which is sufficiently compelling to
encourage broad support of age-specific values for
reducing premature mortality, EPA will continue to
use for its primary analyses a range of values for mor-
tality risk reduction which assumes society values re-
ductions in pollution-related premature mortality
equally regardless of who receives the benefit of such
protection.
1-26
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
Economic Valuation References
Abbey, D.E., F. Petersen, P. K. Mills, and W. L. Beeson.
1993. "Long-Term Ambient Concentrations of
Total Suspended Participates, Ozone, and Sul-
fur Dioxide and Respiratory Symptoms in a
Nonsmoking Population." Archives of Environ-
mental Health 48(1): 33-46.
Abt Associates, Inc. 1992. The Medical Costs of Five
Illnesses Related to Exposure to Pollutants.
Prepared for U.S. EPA, Office of Pollution Pre-
vention and Toxics, Washington, DC.
Abt Associates, Inc. 1996. Section 812 Retrospective
Analysis: Quantifying Health and Welfare Ben-
efits. Draft. Prepared for U.S. EPA, Office of
Policy Planning and Evaluation, Washington
DC. May.
Alberini, A., A. Krupnick, M. Cropper, and W.
Harrington. 1994. "Air Quality and the Value
of Health in Taiwan." Paper presented at the
annual meeting of the Eastern Economics As-
sociation, Boston, Massachusetts, March.
Brookshire, David S., Ralph C. d'Arge, William D.
Schulze and Mark A. Thayer. 1979. Methods
Development for Assessing Air Pollution Con-
trol Benefits, Vol. II: Experiments in Valuing
Non-Market Goods: A Case Study of Alterna-
tive Benefit Measures of Air Pollution Control
in the South Coast Air Basin of Southern Cali-
fornia. Prepared for the U.S. Environmental
Protection Agency, Office of Research and De-
velopment.
Chestnut, Lauraine G. 1995. Dollars and Cents: The
Economic and Health Benefits of Potential
Paniculate Matter Reductions in the United
States. Prepared for the American Lung Asso-
ciation.
Chestnut, Lauraine G. and Robert D. Rowe. 1989. "Eco-
nomic Valuation of Changes in Visibility: A
State of the Science Assessment for NAPAP,"
as cited in National Acid Precipitation Assess-
ment Program, Methods for Valuing Acidic
Deposition and Air Pollution Effects. NAPAP
State of Science and State of Technology Re-
port No. 27, Part B. December.
Cifuentes, L. and L.B. Lave. 1996. "Association of
Daily Mortality and Air Pollution in Phila-
delphia, 1983-1988." J. Air Waste Manage.
Assoc.'. in press. ,
Crocker T. D. and R. L. Horst, Jr. 1981. "Hours of
Work, Labor Productivity, and Environmen-
tal Conditions: a Case Study." The Review of
Economics and Statistics 63:361-368.
Cropper, M.L. and AJ. Krupnick. 1990. "The Social
Costs of Chronic Heart and Lung Disease,"
Resources for the Future Discussion Paper QE
89-16-REV.
Dickie, M. et al. 1991. Reconciling Averting Behav-
ior and Contingent Valuation Benefit Esti-
mates of Reducing Symptoms of Ozone Ex-
posure (draft), as cited in Neumann, I.E.,
Dickie, M.T., and R.E. Unsworth. 1994. In-
dustrial Economics, Incorporated. Memoran-
dum to Jim DeMocker, U.S. EPA, Office of
Air and Radiation. March 31.
Elixhauser, A., R.M. Andrews, and S. Fox. 1993.
Clinical Classifications for Health Policy
Research: Discharge Statistics by Principal
Diagnosis and Procedure. Agency for Health
Care Policy and Research (AHCPR), Center
for General Health Services Intramural Re-
search, U.S. Department of Health and Hu-
man Services.
Empire State Electric Energy Research Corporation
(ESEERCO). 1994. New York State Environ-
mental Externalities Cost Study. Report 2:
Methodology. Prepared by: RCG/Hagler,
Bailly, Inc., November.
Gerking, S., M. DeHaan, and W. Schulze. 1988. "The
Marginal Value of Job Safety: A Contingent
Valuation Study." Journal of Risk and Un-
certainty 1: 185-199.
Industrial Economics, Incorporated (lEc). 1992. Ap-
proaches to Environmental Benefits Assess-
ment to Support the Clean Air Act Section 812
Analysis. Prepared by Robert E. Unsworth,
James E. Neumann, and W. Eric Browne, for
Jim DeMocker, Office of Policy Analysis and
Review, Office of Air and Radiation, U.S.
Environmental Protection Agency. 6 Novem-
ber.
1-27
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
Industrial Economics, Incorporated (lEc). 1993a.
"Analysis of Visibility Valuation Issues for
the Section 812 Study," Memorandum to Jim
DeMocker, Office of Policy Analysis and
Review, Office of Air and Radiation, U.S.
Environmental Protection Agency, prepared
by Jim Neumann, Lisa Robinson, and Bob
Unsworth. September 30.
Industrial Economics, Incorporated (ffic). 1997. "Vis-
ibility Valuation for the CAA Section 812
Retrospective Analysis," Memorandum to
Jim DeMocker, Office of Policy Analysis and
Review, Office of Air and Radiation, U.S.
Environmental Protection Agency, prepared
by Michael H. Hester and James E. Neumann.
18 February.
Irwin, Julie, William Schulze, Gary McClelland,
Donald Waldman, David Schenk, Thomas
Stewart, Leland Deck, Paul Slovic, Sarah
Lictenstein, and Mark Thayer. 1990. Valuing
Visibility: A Field Test of the Contingent Valu-
ation Method. Prepared for Office of Policy,
Planning and Evaluation, U.S. Environmen-
tal Protection Agency. March.
Jones-Lee, M.W., et al. 1985. "The Value of Safety:
Result of a National Sample Survey." Eco-
nomic Journal 95(March): 49-72.
Krupnick, A.J. and M.L. Cropper. 1992. "The Effect
of Information on Health Risk Valuations,"
Journal of Risk and Uncertainty 5(2): 29-48.
Loehman, E.T., S.V. Berg, -A.A. Arroyo, R.A.
Hedinger, J.M. Schwartz, M:E. Shaw, R.W.
Fahien, V.H. De, R.P. Fishe, D.E. Rio, W.F.
Rossley, and A.E.S. Green. 1979. "Distribu-
tional Analysis of Regional Benefits and Cost
of Air Quality Control." Journal of Environ-
mental Economics and Management 6: 222-
243.
Loehman, E.T. and Vo Hu De. 1982. "Application of
Stochastic Choice Modeling to Policy Analy-
sis of Public Goods: A Case Study of Air
Quality Improvements." The Review of Eco-
nomics and Statistics 64(3): 474-480.
Manuel, E.H., R.L. Horst, K.M. Brennan, W.N. Lanen,
M.C. Duff, and J.K Tapiero. 1982. Benefits
Analysis of Alternative Secondary National
Ambient Air Quality Standards for Sulfur Di-
oxide and Total Suspended Particulates, Vol-
umes I-IV. Prepared for U.S. Environmental
Protection Agency, Office of Air Quality
Planning and Standards, Research Triangle
Park, NC.
McClelland, Gary, William Schulze, Donald
Waldman, Julie Irwin, David Schenk, Tho-
mas Stewart, Leland Deck and Mark Thayer.
1991. Valuing Eastern Visibility: A Field Test
of the Contingent Valuation Method. Prepared
for Office of Policy, Planning and Evaluation,
U.S. Environmental Protection Agency. June.
Mitchell, R.C. and R.T. Carson. 1986. "Valuing Drink-
ing Water Risk Reductions Using the Con-
tingent Valuation Methods: A Methodologi-
cal Study of Risks from THM and Giardia."
Paper prepared for Resources for the Future,
Washington, DC.
Moore, MJ. and W.K. Viscusi. 1988. "The Quantity-
'Adjusted Value of Life". Economic Inquiry
26(3): 369-388.
National Acid Precipitation Assessment Program
(NAPAP). 1991. Acidic Deposition: State of
Science and Technology (Summary Report).
(Washington, DC: NAPAP). September.
Neumann, J.E., M. T. Dickie, and R.E. Unsworth.
1994. Industrial Economics, Incorporated.
Memorandum to Jim DeMocker, U.S. EPA,
Office of Air and Radiation. Linkage Between
Health Effects Estimation and Morbidity
Valuation in the Section 812 Analysis —
Draft Valuation Document. March 31.
Ostro, B.D., MJ. Lipsett, J.K. Mann, H. Braxton-
Owens, and M.C. White. 1995. "Air Pollu-
tion and Asthma Exacerbations Among Afri-
can American Children in Los Angeles." In-
halation Toxicology.
Pope, C.A., III, MJ. Thun, M.M. Namboodiri, D.W.
Dockery, J.S. Evans, F.E. Speizer, and C.W.
Heath, Jr. 1995. "Paniculate Air Pollution as
a Predictor of Mortality in a Prospective Study
of U.S. Adults." Am. J. Respir. Crit. Care
Med. 151: 669-674.
1-28
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
Post, Ellen and L. Deck. 1996. Abt Associates Inc.
Memorandum to Tom Gillis, U.S. EPA, Of-
fice of Office of Policy Planning and Evalua-
tion. September 20.
Rowe, R.D. and L.G. Chestnut. 1986. "Oxidants and
Asthmatics in Los Angeles: A Benefits Analy-
sis—Executive Summary." Prepared by En-
ergy and Resource Consultants, Inc. Report
to the U.S. EPA, Office of Policy Analysis.
EPA-230-09-86-018. Washington, D.C.
March.
Salkever, D.S. 1995. "Updated Estimates of Earnings
Benefits from Reduced Exposure of Children
to Environmental Lead." Environmental Re-
search 70: 1-6.
Schwartz, J. 1994. "Societal Benefits of Reducing
Lead Exposure." Environmental Research 66:
105-124.
Spix, C., J. Heinrich, D. Dockery, J. Schwartz, G.
Volksch, K. Schwinkowski, C. Collen, and
H.E. Wichmann. 1994. Summary of the
Analysis and Reanalysis Corresponding to the
Publication Air Pollution and Daily Mortal-
ity in Erfurt, East Germany 1980-1989. Sum-
mary report for: Critical Evaluation Work-
shop on Particulate Matter—Mortality Epi-
demiology Studies; November; Raleigh, NC.
Wuppertal, Germany: Bergische Universitat-
Gesamthochschule Wuppertal.
Taylor, T.N., P.H. Davis, J.C. Tomer, J. Holmes, J.W.
Meyer, and M. F. Jacobson. 1996. "Lifetime
Cost of Stroke in the United States." Stroke
27(9): 1459-1466.
Tolley, G.S. et al. 1986. Valuation of Reductions in
Human Health Symptoms and Risks. Univer-
sity of Chicago. Final Report for the U.S.
Environmental Protection Agency. January.
U.S. Department of Commerce, Economics and Sta-
tistics Administration. 1992. Statistical Ab-
stract of the United States, 1992: The National
Data Book. 112th Edition, Washington, D.C.
U.S. Environmental Protection Agency (U.S. EPA).
1994. Documentation for Oz-One Computer
Model (Version 2.0). Office of Air Quality
Planning and Standards. Prepared by:
Mathtech, Inc., under EPA Contract No.
68D30030, WA 1-29. August.
U.S. Environmental Protection Agency (U.S. EPA).
1995. Human Health Benefits From Sulfate
Reductions Under Title IV of the 1990 Clean
Air Act Amendments. Prepared by Hagler
Bailly Consulting, Inc. for U.S. EPA, Office
of Air and Radiation, Office of Atmospheric
Programs. November 10.
U.S. Environmental Protection Agency (U.S. EPA).
1996. Air Quality Criteria for Particulate
Matter, Volume III of III. Office of Research
and Development, Washington DC. EPA/600/
P-95/OOlcF
Violette, D.M. and L.G. Chestnut. 1983. Valuing Re-
duction in Risks: A Review of the Empirical
Estimates. Report prepared for the U.S. En-
vironmental Protection Agency, Washington,
D.C. EPA-23.0-05-83-002.
Viscusi, W.K. 1992. Fatal Tradeoffs: Public and Pri-
vate Responsibilities for Risk. (New York:
Oxford University Press).
Viscusi, W.K., W. A. Magat, and J. Huber. 1991.
"Pricing Environmental Health Risks: Survey
Assessments of Risk-Risk and Risk-dollar
Tradeoffs." Journal of Environmental Eco-
nomics and Management 201: 32-57.
Wittels, E.H., J.W. Hay, and A.M. Gotto, Jr. 1990.
"Medical Costs of Coronary Artery Disease
in the United States." The American Journal
of Cardiology 65: 432-440.
World Health Organization (WHO). 1996. Final Con-
sultation on Updating and Revision of the Air
Quality Guidelines for Europe. Bilthoven, The
Netherlands 28-31 October, 1996 ICP EHH
018VD962.il.
1-29
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
1-30
-------
Appendix J: Future Directions
Research Implications
In virtually any benefit analysis of environmen-
tal issues, the state of scientific information limits the
degree of coverage possible and the confidence in
benefit estimation. For most benefit categories, fur-
ther scientific research would allow for a better quanti-
fication of benefits. One of the major outcomes of the
retrospective analysis is a clear delineation of the
major limitations in the scientific and economics lit-
erature in carrying out an analysis of this scope. Of-
ten, a list of research needs is generated in studies
such as this, but there is no clear internal mechanism
to address these needs. With this study (and the ongo-
ing section 812 program), a process has been initiated
where identified research needs are to be integrated
into EPA's overall extramural research grants pro-
gram, administered by the Office of Research and De-
velopment. It is hoped that the research projects that
flow from this process will enable future analyses to
be less uncertain and more comprehensive.
Certain of the limitations in the retrospective
analysis are directly related to the historical nature of
the analysis, such as sparse information about air qual-
ity in the early 1970's in many areas in the country.
Other important limitations are related to the effects
of elevated airborne lead concentrations, which has
been virtually eliminated by the removal of lead from
gasoline. A better understanding of these relationships
would improve our understanding of the historical
impact of the Clean Air Act, but would only indirectly
contribute to developing future air pollution policy.
However, most of the research that will reduce the
major gaps and uncertainties needed to improve the
section 812 analyses will be directly relevant to EPA's
primary ongoing mission of developing and imple-
menting sound environmental policies to meet the
national goals established in the Clean Air Act and
other legislation.
There are a number of biological, physical and
economic research areas which the EPA Project Team
identified as particularly important for improving fu-
ture section 812 analyses. These research topics can
be divided into two principal categories: (1) those
which might reduce uncertainties in cost and benefit
estimates with significant potential for influencing
estimated net benefits of the Clean Air Act, and (2)
those which might improve the comprehensiveness
of section 812 assessments by facilitating quantifica-
tion and/or monetization of currently excluded cost
or benefit endpoints. The following subsections pro-
vide examples of research topics which, if pursued,
might improve the certainty and/or comprehensive-
ness of future section 812 studies.
Research Topics to Reduce Uncertainty
Scientific information about the effects of long-
term exposure to air pollutants is just beginning to
emerge, but continues to be the subject of intense sci-
entific inquiry. The relationship between chronic PM
exposure and excess premature mortality included in
the quantified results of the present analysis is one
example of such research. However, many other po-
tential chronic effects that are both biologically plau-
sible and suggested by existing research are not in-
cluded. Research to identify the relationship linking
certain known or hypothesized physical effects (e.g.,
ozone's effects on lung function or fibrosis) with the
development of serious health effects (e.g., cardiop-
ulmonary diseases or premature mortality), and the
appropriate economic valuation of the willingness to
pay to avoid the risks of such diseases, would reduce
the uncertainty caused by a major category of excluded
health effects which could have a significant impact
on the aggregate benefits estimates.
As described in Chapter 7 and Appendix I, pre-
mature mortality is both the largest source of benefits
and the major source of quantified uncertainty in the
J-l
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
retrospective analysis. In addition to the quantified
uncertainty, there is considerable additional
unquantified uncertainty about premature mortality
associated with air pollution. Much of the informa-
tion base about these relationships is relatively new,
more is coming out virtually daily, and there is sub-
stantial disagreement in the scientific community
about many of the key issues. EPA's Research Strat-
egy and Research Needs document for particulate
matter, currently under development, will address
many of these scientific issues as they relate to PM.
The following selection of highly uncertain issues
could have a significant impact on both the aggregate
mortality benefits estimates and the measured uncer-
tainty range:
• the relationship of specific pollutants in the
overall premature mortality effect, including
the individual or interactive relationships be-
tween specific chemicals (e.g., ozone, sul-
fates, nitrates, and acid aerosols), and particle
sizes (i.e., coarse, fine and ultra-fine particles);
• the degree of overlap (if any) between the
measured relationships between effects asso-
ciated with short term exposures and effects
from long term exposure;
• the confounding effect of changes in historic
air pollution, including changes over time in
both pollution levels and the composition of
the pollutant mix;
• the extent to which life spans are shortened
by exposure to the pollutants, and the distri-
bution of ages at the time of death;
• the willingness to pay to avoid the risks of
shortened life spans; and
• the extent to which total PM25 exposure in-
crementally augments the variability of out-
door PM and increases the dose that would
cause excess morbidity or mortality.
After premature mortality, chronic bronchitis is
the next largest health effect benefit category included
in the retrospective analysis. There is considerable
measured uncertainty about both the incidence esti-
mation and the economic valuation. Additional re-
search could reduce uncertainties about the level of
the pollutants associated with the observed effects,
the baseline incidence used to model the changes in
the number of new cases, and the correspondence be-
tween the definition of chronic bronchitis used in the
health effects studies and the economic valuation stud-
ies.
Another area of potentially useful research would
be further examination of the effects of criteria pol-
lutants on cardiovascular disease incidence and mor-
tality. Considering available epidemiological evidence
and the potential economic cost of cardiovascular dis-
ease, the value of avoiding these outcomes may sig-
nificantly influence the overall benefit estimates gen-
erated in future assessments.
Further research on the willingness to pay to avoid
the risk of hospital admissions for specific conditions
would reduce a potentially significant source of non-
measured uncertainty. The Project Team used
"avoided costs" for the value of an avoided hospital
admission, based on the avoided direct medical cost
of hospitalization (including lost wages for the em-
ployed portion of the hospitalized population).
Avoided costs are likely to be a substantial underesti-
mate of the appropriate willingness to pay, especially
for such serious health effects as hospitalization for
strokes and congestive heart failure, particularly be-
cause they omit the value of avoided pain, suffering,
and inconvenience. Furthermore, in addition to hos-
pitalization, there is evidence that some people seek
medical assistance as outpatients. It is also likely that
there are additional people adversely affected by short-
term air pollution levels who seek physician services
(but stop short of hospital admissions). Revised esti-
mates of the appropriate economic value of avoided
hospitalization and other primary care medical ser-
vices could increase the total economic benefits of
this cluster of health effects sufficiently that it could
be a much larger portion of the aggregate benefit to-
tal.
Finally, one of the challenges in preparing the
retrospective analysis was modeling the integrated
relationships between emissions of many different
chemicals, the subsequent mixture of pollutants in the
ambient air, and the resulting health and welfare ef-
fects of simultaneous exposure to multiple pollutants.
One element of the uncertainty in the analysis derives
from the limited current understanding of any inter-
active (synergistic or antagonistic) effects of multiple
pollutants. The need to better understand these com-
plex issues is not a limited scientific question to im-
prove section 812 analyses, but is the primary focus
of EPA's current activities, organized under the Fed-
J-2
-------
Appendix J: Future Directions
eral Advisory Council Act (FACA) process, to de-
velop an integrated set of attainment policies dealing
with ozone, particulate matter, sulfur and nitrogen
oxides, and visibility. Further research on multi-pol-
lutant issues may both (a) reduce a source of unmea-
sured uncertainty in the section 812 analyses and (b)
allow for effective apportionment of endpoint reduc-
tion benefits to specific pollutants or pollutant mixes.
Research Topics to Improve
Comprehensiveness
Even though research efforts falling in this cat-
egory may not result in significant changes in net mon-
etary benefit estimates, one of the goals of the section
812 studies is to provide comprehensive information
about Clean Air Act programs. For example, programs
to control hazardous air pollutants (HAPs) tend to
impose costs and yield benefits which are relatively
small compared to programs of pervasive national
applicability such as those aimed at meeting National
Ambient Air Quality Standards. Nevertheless, there
are significant social, political, financial, individual
human health, and specific ecosystem effects associ-
ated with emissions of HAPs and the programs imple-
mented to control them. Under these circumstances,
continued efforts to understand these consequences
and evaluate their significance in relation to other pro-
grammatic and research investment opportunities
might be considered reasonable, particularly in the
context of comprehensive program assessments such
as the present study.
Some cost and benefit effects could not be fully
assessed and incorporated in the net monetary benefit
estimate developed for the present study for a variety
of reasons. Various effects were excluded due to (a)
inadequate historical data (e.g., lack of data on his-
torical ambient concentrations of HAPs), (b) inad-
equate scientific knowledge (e.g., lack of concentra-
tion-response information for ecological effects of
criteria and hazardous air pollutants), or (c) resource-
intensity or limited availability of analytical tools
needed to assess specific endpoints (e.g., indirect ef-
fects resulting from deposition and subsequent expo-
sure to HAPs). Other specific examples of presently
omitted or underrepresented effect categories include
health effects of hazardous air pollutants, ecosystem
effects, any long-term impact of displaced capital on
productivity slowdown, and redirected technological
innovation.
Although the primary focus of 1970 to 1990 CAA
programs was reduction of criteria pollutants to
achieve attainment of national ambient air quality stan-
dards, emissions of air toxics were also substantially
reduced. Some air toxics were deliberately controlled
because of their known or suspected carcinogenicity,
while other toxic emissions were reduced indirectly
due to control procedures aimed at other pollutants,
particularly ozone and particulate matter. The current
analysis was able to present only limited information
on the effects of changes in air toxic emissions. These
knowledge gaps may be more serious for future sec-
tion 812 analyses, however, since the upcoming pro-
spective study will include evaluation of the effects
of an expanded air toxic program under the CAA Title
III. Existing knowledge gaps that prevented a more
complete consideration of toxics in the present study
include (a) methods to estimate changes in acute and
chronic ambient exposure conditions nationwide, (b)
concentration-response relationships linking exposure
and health or ecological outcomes, (c) economic valu-
ation methods for a broad array of potential serious
health effects such as renal damage, reproductive ef-
fects and non-fatal cancers, and (d) potential ecologi-
cal effects of air toxics.
In addition to research to improve the understand-
ing of the consequences of changes in air pollution on
human health and well-being, further research on non-
health effects could further improve the comprehen-
siveness of future assessments. Improvements in air
quality have likely resulted in improvements in the
health of aquatic and terrestrial ecosystems and the
myriad of ecological services they provide, but knowl-
edge gaps prevented them from being included in the
current analysis. Additional research in both scien-
tific understanding and appropriate modeling proce-
dures could facilitate inclusion of additional benefits
such as improvements in water quality stemming from
a reduction in acid deposition-related air pollutants.
Water quality improvements would benefit human
welfare through enhancements in certain consump-
tive services such as commercial and recreational fish-
ing, in addition to non-consumptive services such as
wildlife viewing, maintenance of biodiversity, and
nutrient cycling. Similarly, increased growth, produc-
tivity and overall health of U.S. forests could occur
from reducing ozone, resulting in benefits from in-
creased timber production, greater opportunities for
recreational services such as hunting, camping, wild-
life observation, and nonuse benefits such as nutrient
cycling, temporary CO2 sequestration, and existence
J-3
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
value. Finally, additional research using a watershed
approach to examine the potential for ecological ser-
vice benefits which emerge only at the watershed scale
might be useful and appropriate given the broad geo-
graphic scale of the section 812 assessments.
While there are insufficient data and/or analyti-
cal resources to adequately model the short-run eco-
logical and ecosystem effects of air pollution reduc-
tion, even less is known about the long-run effects of
prolonged exposure. Permanent species displacement
or altered forest composition are examples of poten-
tial ecosystem effects that are not reflected in the cur-
rent monetized benefit analysis, and could be a source
of additional benefits. In addition to these ecological
research needs, an equally large, or larger, gap in the
benefit-cost analysis is the lack of adequate tools to
monetize the benefits of such ecosystem services.
Future Section 812 Analyses
This retrospective study of the benefits and costs
of the Clean Air Act was developed pursuant to sec-
tion 812 of the 1990 Clean Air Act Amendments. Sec-
tion 812 also requires EPA to generate an ongoing
series of prospective studies of the benefits and costs
of the Act, to be delivered as Reports to Congress every
two years.
Design of the first section 812 prospective study
commenced in 1993. The EPA Project Team devel-
oped a list of key analytical design issues and a
"strawman" analytical design reflecting notional de-
cisions with respect to each of these design issues.1
The analytical issues list and strawman design were
presented to the Science Advisory Board Advisory
Council on Clean Air Compliance Analysis (Coun-
cil), the same SAB review group which has provided
review of the retrospective study. Subsequently, the
EPA Project Team developed a preliminary design
for the first prospective study. Due to resource limita-
tions, however, full-scale efforts to implement the first
prospective study did not begin until 1995 when ex-
penditures for retrospective study work began to de-
cline as major components of that study were com-
pleted.
As for the retrospective, the first prospective study
is designed to contrast two alternative scenarios; how-
ever, in the prospective study the comparison will be
between a scenario which reflects full implementa-
tion of the CAAA90 and a scenario which reflects
continued implementation only of those air pollution
control programs and standards which were in place.
as of passage of the CAAA90. This means that the
first prospective study will provide an estimate of the
incremental benefits and costs of the CAAA90.
The first prospective study is being implemented
in two phases. The first phase involves development
of a screening study, and the second phase will in-
volve a more detailed and refined analysis which will
culminate in the first prospective study Report to Con-
gress. The screening study compiles currently avail-
able information on the costs and benefits of the imple-
mentation of CAAA90 programs, and is intended to
assist the Project Team in the design of the more de-
tailed analysis by providing insights regarding the
quality of available data sources and analytical mod-
els, and the relative importance of specific program
areas; emitting sectors; pollutants; health, welfare, and
ecological endpoints; and other important factors and
variables.
In developing and implementing the retrospective
study, the Project Team developed a number of im-
portant modeling systems, analytical resources, and
techniques which will be directly applicable and use-
ful for the ongoing series of section 812 Prospective
Studies. Principal among these are the Criteria Air
Pollutant Modeling System (CAPMS) model devel-
oped to translate air quality profile data into quantita-
tive measures of physical outcomes; and the economic
valuation models, coefficients, and approaches devel-
oped to translate those physical outcomes to economic
terms.
The Project Team also learned valuable lessons
regarding analytical approaches or methods which
were not as productive or useful. In particular, the
Project Team plans not to perform macroeconomic
modeling as an integral part of the first prospective
analysis. In fact, there are currently no plans to con-
duct a macroeconomic analysis at all. Essentially, the
Project Team concluded, with confirmation by the
SAB Council, that the substantial investment of time
and resources necessary to perform macroeconomic
modeling would be better invested in developing high
quality data on the likely effects of the CAA on key
emitting sectors, such as utilities, on-highway vehicles,
refineries, etc. While the intended products of a mac-
1 Copies of the prospective study planning briefing materials are available from EPA.
-------
Appendix J: Future Directions
roeconomic modeling exercise - such as overall ef-
fects on productivity, aggregate employment effects,
indirect economic effects- are of theoretical interest,
the practical results of such exercises in the context
of evaluating environmental programs may be disap-
pointing for several reasons.
First, the CAA has certainly had a significant ef-
fect on several industrial sectors. However, the coarse
structure of a model geared toward simulating effects
across the entire economy requires crude and poten-
tially inaccurate matching of these polluting sectors
to macroeconomic model sectors. For example, the J/
W model used for the retrospective study has only 35
sectors, with electric utilities comprising a single sec-
tor. In reality, a well-structured analysis of the broader
economic effects of the CAA would provide for sepa-
rate and distinct treatment of coal-fired utility plants,
oil-fired plants, and so on. Furthermore, the outputs
of the macroeconomic model are too aggregated to
provide useful and accurate input information for the
sector-specific emission models used to project the
emissions consequences of CAA programs. Again, the
critical flaw is the inability to project important de-
tails about differential effects on utilities burning al-
ternative fuels.
The second critical problem with organizing a
comprehensive analysis of the CAA around a macro-
economic modeling approach is that the effect infor-
mation produced by the macroeconomic model is rela-
tively unimportant with respect to answering the fun-
damental, target variable: "How do the overall health,
welfare, ecological, and economic benefits of Clean
Air Act programs compare to the costs of these pro-
grams?" The Project Team believes that any adverse
effect, no matter how small in a global context, should
not be deemed "insignificant" if even one individual
is seriously harmed. However, the retrospective study
results themselves have shown that, when analytical
resources are limited, the resources invested in the
macroeconomic modeling would have been better
spent to provide a more complete and less uncertain
assessment of the benefit side of the equation. Even
on the cost side of the equation, it is far more impor-
tant to invest in developing accurate and reliable esti-
mates of sector-specific compliance strategies and the
direct cost implications of those strategies. This will
be even more true in the prospective study context
when the Project Team will be faced with forecasting
compliance strategies and costs rather than simply
compiling survey liata on actual, historical compli-
ance expenditures.
The third and most important limitation of mac-
roeconomic modeling analysis of environmental pro-
grams is that, unlike the economic costs of protection
programs, the economic benefits are not allowed to
propagate through the economy. For example, while
productivity losses associated with reduced capital
investment due to environmental regulation are
counted, the productivity gains resulting from reduced
pollution-related illness and absenteeism of workers
are not counted. The resulting imbalance in the treat-
ment of regulatory consequences raises serious con-
cerns about the value of the macroeconomic model-
ing evaluation of environmental programs. In the fu-
ture, macroeconomic models which address this and
other concerns may be developed; however, until such
time EPA is likely to have limited confidence in the
value of macroeconomic analysis of even broad-scale
environmental protection programs.
Based on these findings and other factors, the de-
sign of the first prospective study differs in important
ways from the retrospective study design. First, rather
than relying on broad-scale, hypothetical, macroeco-
nomic model-based scenario development and analy-
sis, the first prospective study will make greater use
of existing information from EPA and other analyses
which assess compliance strategies and costs, and the
emission and air quality effects of those strategies.
After developing as comprehensive a data set as pos-
sible of regulatory requirements, compliance strate-
gies, compliance costs, and emissions consequences,
the data set will be reviewed, refined, and extended
as feasible and appropriate. In particular, a number of
in-depth sector studies will be conducted to develop
up-to-date, detailed projections of the effects of new
CAA requirements on key emitting sectors. Candi-
date sectors for in-depth review include, among oth-
ers, utilities, refineries, and on-highway vehicles.
The first prospective study will also differ from
the retrospective study in that analytical resources will
be directed toward development of a more complete
assessment of benefits. Efforts will be made to ad-
dress the deficiencies which prevailed in the retro-
spective study relating to assessment of the benefits
of air toxics control. In addition, the Project Team
will endeavor to provide a more complete and effec-
tive assessment of the ecological effects of air pollu-
tion control.
J-5
-------
The Benefits and Costs of the Clean Air Act, 1970 to 1990
"US. GOVERNMENT PRINTING OFFICE: 1999-450-329-10175
J-6
------- |