EPA-AP-85
COST OF AIR POLLUTION
DAMAGE:
status report
, 11 -! '<%'"'
U. S. ENVIRONM
AGENCY
-------
-------
COST OF AIR POLLUTION DAMAGE:
a status report
Larry B. Barrett
and
Thomas E. Wadded
National Environmental Research Center
ENVIRONMENTAL PROTECTION AGENCY
National Environmental Research Center
Research Triangle Park, North Carolina
February 1973
-------
The AP series of reports is issued by the Environmental Protection
Agency to report the results of scientific and engineering studies,
and information of general interest in the field of air pollution. Infor-
mation reported in this series includes coverage of intramural activities
and of cooperative studies conducted in conjunction with state and local
agencies, research institutes, and industrial organizations. Copies of
AP reports are available free of charge to Federal employees, current
contractors and grantees, and nonprofit organizations - as supplies
permit - from the Air Pollution Technical Information Center, Environ-
mental Protection Agency, Research Triangle Park, North Carolina 27711,
or from the Superintendent of Documents.
Publication No. AP-85
-------
PREFACE
This report was conceived and undertaken in order to: (1) survey
estimates made to date of the cost of air-pollution-induced damage; and
(2) develop, on the basis of (1), an estimate of the national annual
cost of damage from air pollution.
Fulfillment of these purposes entailed a survey of the literature
on reported studies, as well as a review of unpublished and ongoing
studies. The survey presents for the first time an opportunity to review
critically and to compare the many fragmented studies on the cost of air
pollution.
Those studies that survived critical review were amalgamated and a
national estimate of the cost of damage was prepared. A by-product of this
process was the identification of information shortages. Although the
suggestion of further research topics was not the purpose of this
report, such needs may now be more readily apparent.
The authors acknowledge that the scarcity of work in several
effects areas necessitated assumptions that may have weakened the con-
clusions. These qualifications have been evaluated in each sectior- of
the report. While recognizing these limitations, the authors believe
that the damage estimates presented are reasonable. The reader is
cautioned against accepting the values as conclusive; they are reasoned,
national estimates.
ACKNOWLEDGMENTS
The authors wish to acknowledge the continual encouragement provided
by Paul H. Gerhardt and Paul A. Kenline over the course of the project.
To a large measure, the impetus for this study came from Paul Gerhardt.
A special thanks also goes to Brian W. Peckham for his counsel.
-------
LIST OF TABLES
Table
1 Effects of Pollutants by Type of Receptor 5
2 Resource Costs of Diseases Associated with Air Pollution ... 9
3 Annual Health Costs of Air Pollution 11
4 Economic Value of Materials Exposed to Air Pollution 18
5 Summary and Rankings of Damage Factors 20
6 Estimated Costs of Air-Pollution-Resistant Materials 22
7 Costs of Shortened Service Life of Rubber Products 23
8 Estimated Costs of Dye Fading in Textiles 25
9 Summary of Data from Mellon Institute Study on Soiling .... 34
10 Summary of Data from Beaver Report on Soiling 35
11 Relationship of Cleaning and Maintenance Operations
to Air Particulate Levels 41
12 Cost of Cleaning Up after Boiler Malfunction 46
13 Amount of Money Respondents Would Pay Annually to Reduce
Air Pollution in Morgantown, West Virginia 47
14 National Costs of Pollution Damage, by Pollutants, 1968. ... 59
15 National Costs of Pollution Damage, by Source and
Effect, 1968 60
16 Estimates of Nationwide emissions, 1968 61
LIST OF FIGURES
Figure
1 Total Benefits and Costs of Pollution Abatement 3
2 Marginal Benefits and Costs of Pollution Abatement 3
IV
-------
CONTENTS
ABSTRACT vii
SUMMARY AND CONCLUSIONS viii
CHAPTER 1. INTRODUCTION 1
PURPOSE 1
BACKGROUND 1
Economics of Air Pollution 1
Functions for Cost of Pollution 3
SCOPE 4
METHOD 6
CHAPTER 2. HUMAN HEALTH EFFECTS 7
INDIVIDUAL STUDIES 7
Ridker - Morbidity and Mortality, Respiratory Diseases ... 7
Lave-Seskin - Heart Disease and Cancer of Stomach,
Esophagus, and Bladder 8
Riggan - Human Health and Motor Vehicle Pollution 11
NATIONAL ESTIMATE 12
CHAPTER 3. EFFECTS ON MATERIALS 13
INDIVIDUAL STUDIES 13
Uhlig - Corrosion of Metals 13
Rust-Oleum Corporation - Corrosion of Metals 13
Hudson Painting and Decorating Company - Painting 14
Stanford Research Institute - Electrical Contacts 15
Midwest Research Institute - General Materials 16
Haynie - Corrosion of Galvanized Materials 21
Battelle-Columbus Laboratories - Rubber Products 21
Salvin - Textiles 23
NATIONAL ESTIMATE 25
CHAPTER 4. EFFECTS ON VEGETATION 27
INDIVIDUAL STUDIES 27
Middleton and Paul us - Crop Survey 27
Lacasse, Weidensaul, and Carrol - Pennsylvania
Plant Survey 27
Stanford Research Institute - Nationwide Survey 29
NATIONAL ESTIMATE 30
CHAPTER 5. SOILING 33
INDIVIDUAL STUDIES 33
-------
Mellon Institute - Pittsburgh Smoke Nuisance 33
Beaver Report - London Smog Episode 34
Michelson and Tourin - Household Costs 35
Ridker - Urban Soiling 37
Ando - Costs of Air Pollution in Japan 38
Booz-Allen and Hamilton - Philadelphia Survey 39
CONCLUSIONS 42
CHAPTER 6. ANIMAL HEALTH EFFECTS 43
CHAPTER 7. EFFECTS ON AESTHETIC PROPERTIES 45
INDIVIDUAL STUDIES 45
Ridker - Soot Incident Survey 45
Lawyer - Morgantown Study 46
Williams and Bunyard - St. Louis Opinion Survey 47
Medalia and Finkner - Clarkston, Washington, Study 47
Williams and Edmisten - Nashville Perception Interviews. . . 48
Current Efforts 48
CONCLUSIONS 49
CHAPTER 8. EFFECTS ON RESIDENTIAL PROPERTY VALUES 51
INDIVIDUAL STUDIES 51
Ridker and Henning - Housing Market Indicators 51
Crocker and Anderson - Three-City Comparison 52
Zerbe - Toronto Study 53
Peckham - Delaware Valley Study 53
NATIONAL ESTIMATES 53
CHAPTER 9. LITIGATION 57
CHAPTER 10. COST OF AIR POLLUTION BY CATEGORIES 59
SOURCE EMISSIONS 59
ASSIGNMENT OF COSTS 61
CHAPTER 11. DISCUSSION 63
DAMAGE ESTIMATES 63
INFORMATION GAPS AND DATA LIMITATIONS 64
REFERENCES 69
VI
-------
ABSTRACT
Published and unpublished estimates of the cost of air pollution
damage are reviewed and national total estimates are given. In addition,
cost estimates are developed for air pollution effects on human health,
vegetation, materials, and residential property values. Estimates of
the cost of air pollution damage to animals and to aesthetic aspects of
the environment are not derived. Although discussed at length, the cost
of soiling from air pollution is not estimated.
The total national cost in 1968 of damage resulting from air
pollution was $16.1 billion, which includes $5.2 billion for residential
property, $4.7 billion for materials, $6.1 billion for health, and $0.1
billion for vegetation. The cost for each effect is distributed among
the several pollutants considered responsible for that effect according
to their relative emissions. The same cost is distributed among the
sources by their relative emissions. No cost function relations could
be developed for various levels of pollution. As a consequence, the
pivotal functions of marginal benefit are precluded. The $16.1 billion
figure is considered to be a reasonable, conservative estimate.
vii
-------
SUMMARY AND CONCLUSIONS
Relatively few studies on the costs of air pollution damage have
been conducted, and, of the approximately 36 studies extant, about one-
third have not yet been published. In the studies conducted to date,
air pollution damage to human health, vegetation, materials, and resi-
dential property values has been examined and costs for this damage have
been derived.
The national cost of air pollution damage for 1968 totaled $16.1
billion, which includes $5.2 billion for residential property, $4.7
billion for materials, $6.1 billion for health, and $0.1 billion for
vegetation. The cost of damage from soiling was excluded from this
national estimate to avoid counting twice the costs of damage to pro-
perty and materials.
The cost of each effect was distributed among the several pollutants
considered responsible for that effect in proportion to the relative
emissions of those pollutants from all sources. The same cost was dis-
tributed among air pollution sources according to their relative
emissions.
The national total cost of air pollution for 1968 may be considered
a function of the pollution levels in that year. No cost relations,
however, could be developed for various levels of pollution. As a
consequence, pivotal functions that are of marginal benefit are pre-
cluded.
Considering all factors, a $16.1 billion national cost of air
pollution in 1968 is believed to be a reasonable, conservative estimate.
The subject of assessing the costs of air pollution would benefit
from further study. The scope of estimates should be broadened to
include more pollutants and more effects. Furthermore, costs should be
related to specific pollutants or pollutant synergisms, rather than to
broad pollutant categories, and to specific sources or source cate-
gories. Functions for the total cost of pollution require the applica-
tion of more sophisticated estimation procedures than those used to
date.
vn i
-------
COST OF AIR POLLUTION DAMAGE:
a status report
CHAPTER 1. INTRODUCTION
PURPOSE
The study reported here was designed to survey the literature on
estimation of costs associated with pollution damages and to provide for
the first time a comparison of the many fragmented studies on air pollu-
tion costs. In addition, the study was designed to review work still in
progress and those cost estimates completed but as yet unpublished.
Another purpose of the study was the development of estimates of the
national total annual cost of ai> pollution. Cost-of-pollution studies
that survived critical review were used in estimating these national
costs. A by-product of this process has been the identification of
deficiencies in data. Although it was not the purpose of this report to
suggest further research topics, such needs have become readily apparent
as a result of this survey.
BACKGROUND
1 p
Economics of Air Pollution
Air pollution has a multitude of effects on man and his behavior,
on his property, and on his environment. Normally, these effects have
some economic value or cost, such as increased human morbidity, reduced
property values, and diminished agricultural output.
The effects of air pollution typically occur external to the emitter.
If these effects cost society but not the emitters, the tendency will
exist to produce more pollution than is economically efficient. It may
be shown by well-established economic principles that pollution should
be reduced to the level at which marginal savings effected by pollution
abatement equal the marginal cost of that abatement. In other words,
-------
savings in the cost of pollution are equivalent to the benefits that
accrue from pollution reduction. Accordingly, additional or marginal
savings in the cost of pollution are marginal benefits.
Consider that the total cost of abatement rises as collection of
pollutants increases, as shown by the total cost of abatement (TCA)
curve in Figure 1. Consider also that the total cost of abatement rises
at an increasing rate for equal incremental reductions in pollution. It
can be said, then, that the marginal cost of pollution abatement rises
as pollution falls. This is shown by the marginal cost of abatement
(MCA) curve in Figure 2. This curve is the slope, or first derivative,
of TCA for each level of pollutant collection.
Total savings in the cost of pollution, or total benefits, are
considered to rise, but at a decreasing rate as abatement or collection
increases. Figure 1 shows the total benefits of pollution abatement
(TBA) and Figure 2 depicts the marginal benefits (MBA). If society
were collecting an amount, OB, of its pollutants (Figure 1), an incen-
tive to increase collections would exist, because, as may be seen from
Figure 2, the additional benefit from collections, ?, exceeds the addi-
tional cost, P-,. The most efficient level of collections is at OA
(Figure 1), where the slopes of TBP and TCA are equal, or, in terms of
Figure 2, where MBA equals MCA.
At the optimum level of pollution control, the total benefits
exceed the total costs. One might wish to infer from this simple obser-
vation that there ought to be additional collections. It must be
realized, however, that beyond a certain point the marginal costs of
additional collection will exceed the marginal benefits and thus result
in a reduction of net benefits.
The foregoing demonstration supports the need for two types of
functional relations in analyzing the economics of pollution control. One
function would seek to identify the costs of abatement for each level
of pollution. A second function would relate the benefits derived from
pollution reduction, or their parent, the cost of pollution, to each
level of pollution. Total costs of abatement are addressed, at least
3 4
partially, in annual reports to Congress. ' Only the second, the cost-
of-pollution function, is considered in this report.
COST OF AIR POLLUTION DAMAGE
-------
POLLUTION COLLECTION, tons
Figure 1. Total benefits and costs of pollution abatement.
POLLUTION COLLECTION, tons
Figure 2. Marginal benefits and costs of pollution abatement.
Functions for Cost of Pollution
Ideally, cost-of-pollution functions should be constructed for
each economic effect, for each pollutant, and for each emitter. Deter-
mination of economic effects should include direct and indirect ef-
fects. Direct economic effects, such as vegetation damage, are the
immediate result of pollution. Pollution costs are readily measured for
many direct effects as the product of the number of units damaged and the
price per unit.
Introduction
-------
Indirect effects of pollution, which are induced by the direct
effects, include economic loss both to individuals and to market values.
For example, a person is affected indirectly by pollution when he accepts
a different job with a lower income in a cleaner city, an occurrence that
is not uncommon. Van Arsdol,6 for example, reports that Los Angeles popu-
lations are shifting away from sites seriously affected by air pollution.
Indirect market effects result from the accumulation of individual
effects. As will be seen later, residential property value studies
build on the market-price differentials associated with demands by the
population for relocation away from pollution.
Calculation of the costs related to each type of pollution effect
could result in some duplication in counting. It is conceivable, for
example, that the costs of pollution measured by residential property
price differentials might include some of the direct pollution costs of
materials damage and soiling. As pointed out by Ridker, however, no a_
priori way exists for discriminating among the various economic measure-
ments of the cost of air pollution; therefore, both direct and indirect
effects should be studied.
All pollutants or groups of pollutants that produce economic con-
sequences should be included in the derivation of functional cost-of-
pollution relationships for each type of pollutant effect. Table 1
lists effects of specific pollutants according to categories of recep-
tors affected. Cost-of-pollution functions should be derived for
each emitter in order to represent better the true balance of marginal
costs and benefits of pollution control. Certainly, one of the costs of
control is the expense of determining the cost-of-pollution functions;
since the cost of obtaining information usually exceeds the gains asso-
ciated with knowing these functions, emitters are seldom the smallest
optimum air pollution control unit. Instead, aggregations by industry,
region, or nation are the rule.
SCOPE
Eight measures of air pollution effects are discussed in this report;
however, only four were used to estimate pollution costs: human health,
materials, vegetation, and residential property. The others, which in-
clude soiling, animal health, aesthetic values, and litigations, were
excluded because available data were inadequate for a cost determination.
COST OF AIR POLLUTION DAMAGE
-------
Table 1. EFFECTS OF POLLUTANTS BY TYPE OF RECEPTOR
Pollutant
Participate matter
Sulfur oxides
Oxidants
Carbon monoxide
Hydrocarbons
Nitrogen oxides
Fluorides
Lead
Polycyclic organic
matter
Odors (including
hydrogen sulfide)
Asbestos
Beryllium
Hydrogen chloride
Chlorine
Arsenic
Cadmium
Vanadium
Nickel
Manganese
Zinc
Copper
Barium
Boron
Mercury
Selenium
Chromium
Pesticides
Radioactive
substances
Aeroallergens
Receptor
Human
health
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Materials
xa
X
X
-
-
X
-
-
-
X
_
-
X
X
-
_
_
-
-
-
_
-
-
X
-
X
-
-
-
Aesthetic
properties
Xa
X
X
_
X
X
-
-
-
X
.
-
-
-
-
_
-
_
-
-
_
-
-
-
-
_
_
-
-
Vegetation
X
X
X
_
X
X
X
X
-
-
_
X
X
X
X
-
.
_
-
-
_
-
-
X
-
_
X
-
-
Animals
X
X
_
_
-
_
X
X
X
-
_.
X
-
X
X
X
_
_
-
X
_
_
_
X
X
_
X
X
-
Includes soiling by particulate matter.
Six of the eight criteria can be considered indices of direct
effects of air pollution: human health, deterioration of materials,
soiling, vegetation damage, animal health, and aesthetic properties.
The cost of direct effects is determined by multiplying the extent
of effects in a particular category times the economic value of that
category. Two measures by which indirect effects may be estimated are
differential residential property values and litigation. Residential
property value studies concentrate on market dislocations, and litigation
studies examine individual reparations.
Introduction
-------
Air pollution is believed to have pervasive effects - which may
transcend other effects combined - in the biosphere and on geophysical
processes. These effects of air pollution are not without some economic
value, but until ecological relationships can be more clearly identi-
fied, economic analysis is premature.
Because data on the cost of damage resulting from specific pollu-
tants are limited, presentation is restricted to the major pollutants:
sulfur oxides, particulate matter, oxidants, carbon monoxide, and
nitrogen oxides. Furthermore, since specific cost-of-pollution func-
tions are not available for any emitters, only the broad aggregations
are considered; namely, fuel combustion in stationary sources, trans-
portation, industrial processes, solid-waste disposal, and miscellaneous
sources.
METHOD
The study reported here proceeded in two general steps. First,
what should be known about the costs of pollution was defined.
Second, a survey of the literature was made to ascertain what i_s known.
As a result, areas that have not been investigated can be identified,
although such an identification is not explicitly made here.
What should be determined with respect to costs arising from air
pollution damage has been described under "Background." What is already
known is given in the remaining chapters, which present results of the
literature search for each of the principal measures or economic effects
of air pollution. Limitations of the report are described in the pre-
ceding section, "Scope."
Studies on the costs of air pollution are reviewed critically for
the utility of their objectives, fulfillment of these objectives, and
the analytical methods used. Where necessary, and where feasible, the
results are adjusted to allow annual cost comparisons for 1968. The
eight chapters in which measures of economic effects are discussed have
been made as mutually exclusive as possible in order to minimize counting
the same cost twice. Some overlapping is inevitable, however. Mitigat-
ing this are omissions, which weigh in favor of conservative damage
estimates.
COST OF AIR POLLUTION DAMAGE
-------
CHAPTER 2. HUMAN HEALTH EFFECTS
INDIVIDUAL STUDIES
It is not uncommon to see references that cite the health effects
of air pollution separate from the economic effects. One inference
from this division is that the effects of air pollution on health trans-
cend economic values. In fact, however, economic values can be attached
to health effects even though they are not easily measured. The sepa-
ration may have been influenced largely by the absence of cost estimates
for health relative to other types of damages.
As of 1970, only two studies estimating the health costs resulting
from air pollution had been published.8>9 in addition to these studies,
there is a report internal to the National Environmental Research Center
of the Environmental Protection Agency that deals with health effects
from motor vehicle pollution.10 The first two studies attempt to
measure the cost of air pollution in terms of costs of health for the
nation. The third study is restricted to an assessment of Federal
revenues and expenditures growing out of morbidity and mortality related
to automobile pollution.
All the studies follow the same general method: (1) an estimate is
made of the total dollar value associated with health losses; (2) a co-
efficient is determined for the share of this value attributable to air
pollution; and (3) total health losses are multiplied by the coefficient
to determine the value of health losses associated with air pollution.
The proportion of health losses associated with air pollution is
determined to be a constant in all the studies. Thus the cost-of-pol-
lution function for health may be taken to be linear.
Ridker - Morbidity and Mortality, Respiratory Diseases
The earliest published attempt at estimating the cost of damages to
o
health from air pollution is that of Ridker, who estimated the total
national cost of morbidity and mortality for diseases associated with
the respiratory system. The diseases considered, chosen partly on the
basis of crude empirical determinations and partly on reasonable hypoth-
eses, were cancer of the respiratory system, chronic and acute bronchitis,
pneumonia, emphysema, asthma, and the common cold.
-------
The cost estimates for each disease included - where possible - the
costs of treatment, absenteeism, premature death, and premature burial.
Treatment costs were estimated for each disease using per capita values
of drug manufacturers' shipments. Absenteeism costs were the product of
days lost for each disease times the average annual earnings of those
suffering from that disease. The costs of premature death were the 1958
value of future earnings lost. Discount rates of 5 and 10 percent were
applied to expected lifetime earnings from full-time employment. Costs
of premature burial represented the difference between the present cost
of burial and the present value of future expected burial costs discount-
ed at rates of 5 and 10 percent. The total costs for 1958 yielded a
total economic loss from these diseases of $1.990 billion at a discount
rate of 5 percent. Table 2 presents the results of this study.
Ridker estimated that 18 to 20 percent of the approximately $2
billion in national health costs results from air pollution. His coeffi-
cients of 0.18 and 0.20 are taken from two studies relating, in one
case, respiratory mortality rates and, in the other, lung cancer mor-
tality rates to urban and rural areas. The coefficients are corrected
for age, sex, race, smoking habits, and the proportion of population in
the United States that is urban. Thus the damage to health from air
pollution in 1958 had an economic value of $360 to $400 million.
Ridker considered the estimate to be quite conservative. For one
thing, he recognized the absence of expenditures made to avoid air
pollution, such as moving costs and possible reduced earnings of those
who migrate to areas of lower pollution because of their health. He did
not take into account the full value of housewives' services lost because
of death or illness related to air pollution. No attempt was made to
include the psychic costs associated with death or illness. Data limita-
tions prevented estimation of the full costs of some of the diseases
considered and also prevented consideration of certain diseases.
Ridker was unable to relate morbidity and mortality to specific
types of pollution, and therefore did not attempt to construct individual
damage functions.
Lave-Seskin - Heart Disease and Cancer of Stomach, Esophagus, and Bladder
Q
The Lave-Seskin study expanded the number of diseases covered by
Ridker. In particular it included air pollution damages to health in
COST OF AIR POLLUTION DAMAGE
-------
CO
0
I
_J
1
o
a.
C£
e^
nr
, ,
3
LU
1
o
o
t/1
CO
LU
CO
UJ
CO
Q
U.
0
CO
h-
0
LU
0
ZD
O
CO
Cd
CM
QJ
_Q
ns
1
'rrj
,jj_>
O
f-
g
-E
*
LO <1J
O 01
r- >>
Q-
l/J1 UJ
0-
r^ E
0) O T3
c/1 E
E o
_E 0 U
>-> O
i
3
01
"a -t
01. -t-1
-M a) -i-
(O 4-» _E
U U E
0 < 0
01 S-
01 -Q
(0 .
to t/)
01 O -I-
01 E -i-
O O -d
O S- U
-E E
O O
i.
-°
1
>>
O
S- S- -t->
IE -f- 01
JtO M- 0- >,
C-J O 01 t/l
2
f-
o
+J
(!) 01
a. o
>i O
o
C\J
01
Ol
01
LO
10
Ol
CM
ja
to
o
UD
o
CO
CO
LO
a)
S-
13 -E
to to
E 0)
a) -a
S-
a.
0^
OJ
CvJ
CM
CO
»
fti
c
CM
O
l*-^
O
LO
,
tu
i- 1
3 IO
E ;3
S-
CL,
o
LO
CO
LO
CO
CO
'
E
CO
^
o
CD
CM
to
E
O
cn
CO
LO
CO
4_>
E
Q)
10
0)
t-
1
0
O
CO
*
0
LO
c
LO
^
(
CO
1
to
c
o
CM
LO
CM
(0
u
s-
01
Q.
LO
"»-
(U
(O
s-
1 *
c~
O 0)
o .
i- to
,I_
to to
cn to
E
01 o
to .Q
Human Health Effects
-------
the form of heart disease and of cancer of the stomach, esophagus, and
bladder.
Health costs from all causes have been estimated by Rice, who
included the costs of premature death, treatment, and absenteeism.
These costs were broken down by major disease category except for
costs of some types of treatments and of expenditures for items such as
drugs, eyeglasses, and school health services. Lave and Seskin con-
sidered only those costs given by disease category. The cost of pre-
mature death was calculated as foregone earnings discounted at 6 per-
cent. All costs were computed on the basis of 1963 data.
Lave and Seskin consider the evidence relating air pollution and
health to be quite good in the case of bronchitis and lung cancer, and,
although evidence correlating air pollution with heart disease and non-
respiratory cancers is not as good, they believe that a consideration of
all factors suggests causality for these diseases as well.
Approximately half the foregone income and current medical expenses
associated with bronchitis morbidity and mortality are ascribed to air
pollution. The coefficient for lung cancer that is attributable to air
pollution is 0.25. In the other categories, air pollution is responsi-
ble for 15 percent of the damages associated with nonrespiratory lung
cancers, 25 percent of all respiratory diseases other than bronchitis,
and 10 percent of the cardiovascular disease. These coefficients were
estimated by regressions that were run on data from published epidemio-
logical studies in which morbidity rates were expressed as functions of
air pollution and selected socioeconomic variables.
The total annual cost for increases in human morbidity and mortality
caused by air pollution is $2.08 billion for 1963; stated inversely,
a 50 percent reduction in air pollution would result in an annual savings
of $2.08 billion. The index that Lave and Seskin prefer is that air
pollution damage amounts to 4.5 percent of all the economic damage
associated with morbidity and mortality. Table 3 summarizes their
results.
Lave and Seskin consider their estimate to be conservative on
several grounds. They argue that, for conceptual meaning, the willing-
ness of an individual to pay for improved health or longevity, given a
certain level of income and wealth, is the true measure of health costs
10 COST OF AIR POLLUTION DAMAGE
-------
Table 3. ANNUAL HEALTH COSTS OF AIR POLLUTION9
Disease
Respiratory
Bronchitis
Other
Subtotal
Cancer
Lung
Other
Subtotal
Cardiovascular
Total
$ billion
0.500
0.722
1.222
0.033
0.357
0.390
0.468
2.080
attributable to air pollution. They consider the sum of foregone income
and current expenditures that result from morbidity and mortality from
air pollution to be a gross underestimate of willingness to pay. Some
health costs may have been overlooked, so that the estimate may be even
more conservative. Finally, the exclusion of some treatment costs
results in an underestimate of true damage costs.
If one accepts with Ridker the inclusion of premature burial costs
in damage estimates, then their absence from the Lave-Seskin estimate
further corroborates the conservative nature of that estimate. Yet
Lave and Seskin suffer from the same difficulties as Ridker in deriving
meaningful estimates. They assume linear relationships between air
pollution and health even though it is not possible to relate health
costs to levels of pollution or to sources of pollution. The result
of their efforts, then, is another gross estimate that can only be
suggestive of the direction air quality management programs should take,
but not of the level of effort required.
Riggan - Human Health and Motor Vehicle Pollution
The objective of the two previous studies was to quantify the effects
of air pollution on human health and to estimate the economic benefits
that would result from abatement. Riggan set a more limited objective
of investigating the effects of motor vehicle pollution on human
health. He made a study of the effects of automobile pollutants and
their derivatives on lung cancer, bronchitis, heart disease, and motor
vehicle accidents. Although some estimates were generated, they are
too preliminary for reporting here.
Human Health Effects 11
-------
NATIONAL ESTIMATE
Lave and Seskin calculated the national cost of air pollution
damage to health as $2.08 billion for 1963. This represents a savings
that would result from a 50 percent abatement of air pollution. Under
their assumption that the functional relationship of pollution and
mortality is linear, a 100 percent reduction in air pollution would re-
sult in a savings of $4.16 billion in that year. This amount includes
the discounted 1963 value of future earnings lost because of mortality
and the costs of treatment, prevention, and morbidity. Because an
approach based on discounted, future earnings lost places value only
on those people in the working force, homemakers who do not work for an
income, as well as retired individuals over 65, are not valued. Thus,
the estimate of $4.16 billion surely is an understatement of the true
cost of deaths and illnesses caused by air pollution.
The authors believe that some caveats should be mentioned. Lave
and Seskin seem to have a stronger faith in the magnitude, sign, and
statistical significance of their regression coefficients than their
analysis would seem to support. Their many statements about the causes
of these "effects" are not as justified as the authors seem to conclude.
Yet in fairness, despite the authors' questionable data, analysis, and
extended discussion of their results, their final cost estimate is
believed to be reasonable.
A national economic estimate of health losses from air pollution
may be calculated for 1968. The 1963 loss of $4.16 billion amounted to
0.70 percent of the 1963 Gross National Product (GNP) of the United
12
States, which was $590.5 billion. Assuming that the economic loss
was the same percentage of the 1968 GNP, which was $865.7 billion,
the cost of air pollution effects on health in 1968 was $6.06 billion.
12 COST OF AIR POLLUTION DAMAGE
-------
CHAPTER 3. EFFECTS ON MATERIALS
INDIVIDUAL STUDIES
Air pollution has a variety of effects on materials, including cor-
rosion of metals, deterioration of materials and paints, and fading of
dyes. Several attempts have been made to estimate the economic losses
attributable to the detrimental effects of air pollution on materials.
The resulting reports are listed and discussed in the following para-
graphs.
Uhlig - Corrosion of Metals
The first serious attempt at estimating economic losses resulting
14
from corrosion of metals was made by Uhlig in 1950. He distinguished
between two types of losses: (1) direct losses resulting from the cost
of protecting or replacing corroded equipment or parts; and (2) indirect
losses resulting from shutdown, over-design, loss of product, ineffi-
ciency, explosion, and contamination. The total corrosion bill was
estimated to be $5.4 billion. In Uhlig's study, air pollution was
merely implicated as a causal agent of corrosion.
Rust-Oleum Corporation - Corrosion of Metals
The Rust-Oleum Corporation updated the Uhlig information and
estimated the rust bill for the United States as $7.5 billion in 1958.
According to this study, rusting is not so great a problem in rural as
in urban atmospheres because industrial fumes and acids are not present
in rural areas to any appreciable extent. Corrosion in industrial
atmospheres was associated mainly with the sulfur-containing gases,
which become dissolved in water to form acid mists that act as electro-
lytes. Chlorides were also blamed as a corrosive agent in many
industrial environments.
The data employed in constructing the estimates were acquired from
exposed metal plates placed in cities that have had a population of
10,000 people or more for the past 25 years. Some of the costs making
up the $7.5 billion estimate included, among others, $500 million
annually for corrosion to railroads and $1.25 billion to homeowners
13
-------
for corrosion damage. The estimate encompassed all types of corrosion,
and to determine the portion attributed to SOx in particular would
necessitate a number of weak assumptions. The study dealt with gross
estimates only and gave no justification or basis for these estimates.
Hudson Painting and Decorating Company - Painting
The Hudson Painting and Decorating Company estimated, in 1967, the
increased costs of painting resulting from air pollution damage to paint
in New York.16 Using as a starting point the sum grossed through the
sale of paint and allied materials manufactured in New York and New
Jersey in 1963, they arrived at an amount attributable to air pollution
effects through the assumptions and calculated sums given below:
$ billion
1. Sum grossed through the sale of paint and allied 0.238
materials manufactured in New York and New Jersey
in 1963.
2. Assumptions and resultant sums:
a. That two-thirds of this dollar volume was 0.158
for metropolitan New York.
b. That 1967 prices had increased 10 percent 0.209
over 1963 prices and that materials were
sold to the contractor at a 20 percent
markup.
c. That the cost of labor was three times the 1.254
material cost and that the common method of
pricing a paint job was to double the labor
cost.
d. That contractors do one-third of the 0.418
painting done.
e. That 15 percent is deducted for any over- 0.350
statement.
f. That about one-third of the cost of painting 0.150
is assumed to be attributable to air pollu-
tion damage.
The assumptions as presented here are speculative, and the use of data
such as these yields results of questionable value.
14 COST OF AIR POLLUTION DAMAGE
-------
Stanford Research Institute - Electrical Contacts
The objective of the Stanford Research Institute study was to
determine whether air pollution creates problems of economic signifi-
cance in the operation of electrical contacts; and, if it does, to study
the cause-and-effect relationships. Information was compiled by search-
ing the literature as well as by consulting manufacturers through
letters, telephone calls, and on-site visits. The major costs that were
investigated were: (1) the direct cost of plating contacts with pre-
cious metals and (2) the indirect cost associated with air-conditioning
and purification.
The types of contacts usually plated are switches, relays, connec-
tors, potentiometers, and commutators that are used mainly in the
electronics and communications industries. More money is spent combat-
ting the effects of S02 and H2S air pollution on low-voltage electrical
contacts than is spent combatting all other air pollution effects on
electrical devices combined. Organic gases form "frictional" polymers
on sliding contacts, whereas particulate pollutants are a problem
because particulate matter is an excellent adsorber of H^O and corrosive
agents. The effects of organic gases and particulates are of less
importance, however, than the effects of S02 and H2S.
It was estimated in this study that $20 million is spent on plating
contacts with precious metals to prevent air pollution corrosion and
$25 million is spent annually on air-conditioning and purification.
Another $4 million is spent annually for washing insulators; $5 million
is spent for research and development by firms whose products might be
affected by air pollution; and an estimated $10 million is spent because
of equipment failures. The total annual expenditure is approximately
$65 million.
The most important conclusion drawn from this study is that the
damage to electrical contacts from air pollution is not as serious as
originally thought. In addition, the estimated annual cost of $65 mil-
lion was concluded to be unnecessarily high because two or more individ-
ually effective countermeasures are being applied simultaneously to
minimize losses. Furthermore, losses will decrease as cheaper materials
that are more resistant to air pollution are used in electrical contacts.
A critical review of the study has shown that the assumptions on
which estimates of expenditures for air-conditioning and purification
Effects on Materials 15
-------
were based were somewhat questionable, but they were perhaps the best
that could be made.
Midwest Research Institute - General Materials
The most comprehensive survey of the economic effects of air pol-
lution on materials was undertaken by Salmon of Midwest Research
Institute.'^ The objectives of the study were: (1) to identify the
materials, air pollutants, and environmental parameters that should be
studied in order to assess the economic value of air pollution damage to
materials; (2) to analyze systematically the physical and chemical
interactions among the variables identified in (1) above for the purpose
of determining cause-effect relationships; (3) to determine, where
possible, the pollutant-dosage/material-response relationship for
materials that are significant because of their relative economic value,
and to indicate how this may be done where such relationships are pre-
sently defined; and (4) to translate the pollutant-dosage/material-
response relationship into a pollutant-dosage/monetary-loss function.
Information was gathered through literature searches, and by per-
sonal, mail, and telephone interviews. Economic losses from damage to
materials were attributed either to damaged properties or to impaired
serviceability. The basic problem was to determine the extent of
economic damage associated with a given level of physical or functional
damage. This problem was approached in different ways, depending on the
material and its application. In some cases a "percentage condition"
approach was adequate, with economic damage considered to have been
incurred at the same rate as the physical damage. The replacement of
damaged materials was also included in this category, with replacement
presumably occurring at the 100 percent damage (or zero percent
condition) level. In other cases, a cost-of-prevention or cost-of-
restoration basis proved more suitable.
The economic value of material exposed to air pollution was calcu-
lated as the product of the annual production volume in dollars, times
a weighted average economic life of the material (based on usage), times
a weighted average factor for the percentage of the material that is
exposed to air pollution. The in-place or as-used value of the
material was estimated by inclusion of a labor factor. The rate of
economic loss was calculated as the product of the economic value of
material exposed to air pollution times a value of interaction; the
16 COST OF AIR POLLUTION DAMAGE
-------
latter was calculated by estimating the difference between the rate of
material deterioration in a polluted environment and that in an unpol-
luted environment. The interaction value is expressed as dollars lost
per year. The results of the operations described are presented in
Table 4. The total value of materials exposed to air pollution and
values of interaction between the various materials and pollutants have
been combined to produce a single figure - a damage factor - that repre-
sents the extent of economic damage attributable to air pollutants.
These damage factors are summarized and ranked in Table 5.
To interpret the results in Table 5, it is imperative that one
realize that the individual material loss estimates were made to
determine relative importance rather than actual value. The sum, how-
ever, of the economic losses - $3.8 billion in 1968 - appears to be a
reasonable estimate. If it is assumed that this list of materials
represents only 40 percent of the total value of materials exposed to
air pollution, and that damage functions for the other 60 percent are
similar, then the total loss resulting from the chemical attack of
materials by air pollution is estimated at $9.5 billion.
The pollutants, in decreasing order of economic importance, and the
materials they damage are as follows:
1. SOX: metals, cotton, finishes, coatings, building stone,
paints, paper, and leather.
2. 63: rubber, dyes, and paints.
3. NOX: dyes and paints.
4. C02: building stone.
5. Particulate matter: stone, clay, and glass.
The study concluded that organic pollutants (hydrocarbons and
aldehydes) are not damaging to materials, except for elastomers, which
they damage to some extent. Much information was available on the
effects of air pollutants on metals and rubber. Some information was
available on fibers such as cotton and nylon, but little was available
on paints, paper, and leather. Virtually no information existed on
plastics, wool, wood, and concrete. Direct quantitative correlations
of damage with specific pollutants have been made (dose-response) only
Effects on Materials 17
-------
CO
z.
o
o
to
o
1
, 'o c
<0 O- O
si S- r^
O) !- (
+-> (0 !-
(O -O
E O
(U
C S-
0 0
(J (-*
o
S-
U S-
r- >>
C
O
u
LU
C
o
u c
3 O
"O *r
O-*r~
3
C
C
(O
s-
0)
IE
-g
(O
r COODCMCnOOLOOLOOi Ot£>
CMCOO» tocOi cn ^* 01 10 o r-«.
uD^t-^-cocoi£>^-ooojcor*.^±o
i LOtnCOCMCMCMCMCMi i i (
ro
^-orocococorocTii cooco
OJCMCMCMCMCMCMrOCSJOJ^-f CM
OO CM
LOLDmvDr^t£)r^p^.h^<^>or--.o
OOOOOOCDOOOOOO
OCM^£)OOU)0^-OlOOCOCO
^f CM r ^f CO CO "^ t ^ CO r
r r CO CD CM CD LO r*1"- VO CD LO CD CD
^OCMCOOOOOCMi i O^£)CO
c/>
S 5
?! 3
J3 S- t/1 S- OJ
*o E d nj en o v> *^ t/)
3 *o i c: ecu +*
4->S-C CD -i- -i i OJC
E Q- 3 n3 (OOCS-
^fO
r-CM^^-LO^r-OOOO-CMfr,
OLTlCO OCOLDkOOC30r^CX3COCX3OO«^-
*^-l£)LnLOCOLOi OOCO^CMri O^CT^LO
c-JPSXPXI^LnLnLr)crlrocNJcsJCSJCX,r_l ,_
^_
COCOOCOOi OOCTtCf>O«a-OOPOO
CMCMCMCMCMCMi CMi i i CMCMi CMi
kD CO CM
Lnr-^O«=J-|^>«3-CMOi CMCMCDCOCMCMCM
oooooooooooooooo
^-orocMoirotnoto^or^^-Lnr^uD^
i LD i COr CM CMCM CM i
^cncMc?c\Jo$^cocScM§cMSoo
ooooooLnooooi CDCMOOCDO
CM
' ^ E ro t- o
C. L. fO r^ Cl) r-
Scu- a. o ^:
_Q J_ QJ r (J
V) - CU E OJ >,
CT 3 o * -i- -a +J oo>>
i - in cu -r- c: -^-- to i o -i-
0)-OS-> °?5'
t_)*r O-CHi C fO fO-*-* O QJi fO CUM i
r-3
-------
CO
z.
o
00
O
O-
O
o
3-
O)
-Q
ra
1
i c:
ra O O
r Q.T
S- r
O "
It) T) C
I O
CX t/) !-
1 O -I-1
c: o_ zi
i i X r~
a>
+-> S-
c o
0 +->
o o
<0
S- 4-
o
_Q
ra
S- ra
3 X
l/> 0)
O "D
a. E
X 'I-
UJ
U 4-
1- >>
E
o
o
UJ
c:
o
+J
o c
rj o
a -r-
O r
i. J
O_'i
JD
,_
ro
i
S-
O)
4->
ro
s:
c
r^
^j-r^r^rjOQLf-jLQCsj,^-,
^- ro i o O CT1 CO CO ^£) t£>
r i r r OOOOO
COOO^OCJiOCTiOO
C\]i i C\Ji i i i i
i CMCM^-OOi CMi CMCM
OOOOOOOOOO
ir> ro r- o
oocoLnncocDoi^oooo
oour^., r*.rOLr>i or^.
i CNJCTli CNJCOCOr^^OO
-i- M- Wl T3
r- to xs +-> > (O ro
40
t/) i ! fd tO D- 3
fO O- *-> i S- -J2
i * if) O. Q) t
CL fO- 4-> 0) CU
^^ cu i i- tn c i
C O.CUCUCUCUT--
>*-> -i-
Q-QJ^EtnM-i-trt Q---- -M
I-4J.C 3 U ' >j O O Col
Q.oi-4-J !-!- -l->i S-On3
CU > >i >> O >j O >» >> >, CL
« oo-c:u>>ocuou
focomc^jcocororororo
(-QP^^., ^Q^ cyicnr^.U3c\j«?j-
OOOOOOOOOOCDOOO
oooaiooo^^cncnoo^ocn
C\JCOr-CXlr r CXJ^ i i
CO LO C\J CSJ
Of^OOOJCVJi CD i i OJi C\Ji
oooooooooooooo
C\J CO ^
to to co co r*--. co co CTI co LO *jO *-o co *o
^j-i r-^cMi cxjojcoi i i i oo
oooooooooooooo
U
+->*-> S-
^-^ V) -i- ^-x- x QJ
i- i Q. S- -* O O) -i- -r-
C S- 4-J O- fQ CU S- » _.O *> T- t/) ^-s
S- -Q ra S-Ol'i-^^l/lM-fd
- JD i in CU 4- M- ro 'r t/>
OJ S- 3 -^ *f- - C - Q. S-- -
i C yi M- ro t/) - CU 4-
o ^ CU t/) O * * O CU 4-* CU 1)
rt3i_n3"Ol/)(/)-4-)i
>:3>iX" p_a>>4->o>>-M:>>
fa-r-raOCLCUfOrOCJO >>OOO
°5~55^^5:«5g^~S
Q)
S-
a>
o
Q-
X
0)
S-
01
OJ
QJ
X
a>
-a
c
Effects on Materials
19
-------
Table 5. SUMMARY AND RANKINGS OF DAMAGE FACTORS^
Rank
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
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Material
Paint
Zinc
Cement and concrete materials
Nickel
Cotton (fiber)
Tin
Synthetic rubber
Aluminum
Copper
Wool (fiber)
Natural rubber
Carbon steel
Nylon (fiber)
Cellulose ester (fiber)
Building brick
Urea and melamine (plastic)
Paper
Leather
Phenolics (plastic)
Mood
Building stone
Polyvinyl chloride (plastic)
Brass and bronze
Polyesters (plastic)
Rayon (fiber)
Magnesium
Polyethylene (plastic)
Acrylics (plastic)
Alloy steel
Polystyrene (plastic)
Acrylics (fiber)
Acetate (fiber)
Polyesters (fiber)
Polypropylene (plastic)
Acryl oni tri le-butadi ene-
styrene (plastic)
Epoxies (plastic)
Cellulosics (plastic)
Bituminous materials
Gray iron
Nylon (plastic)
Polyolefins (fiber)
Stainless steel
Clay pipe
Acetate (plastic)
Malleable iron
Ch romi urn
Silver
Gold
Flat glass
Lead
Molybdenum
Refractory ceramics
Carbon and graphite
Total
Value of
interaction
$/yr
0.50 x 10-"1
0.29 x 10-1
0.10 x ID"2
0.25 x TO"1
0.40 x 10-1
0.26 x ID"1
0.10 x 10°
0.21 x 10-2
0.20 x TO'2
0.40 x 10-1
0.10 x 10°
0.50 x ID'2
0.40 x lO'1
0.40 x 10-'
0.10 x ID"2
0.10 x ID'1
0.30 x ID"2
0.40 x ID'2
0.10 x 10-1
0.10 x TO"2
0.23 x ID"2
0.10 x 10-1
0.42 x ID'3
0.10 x 10-1
0.40 x 10-'
0.20 x NT2
0.10 x 10-1
0.10 x 10-1
0.40 x ID'2
0.10 x 10-1
0.40 x 10-1
0.40 x 10-1
0.40 x ID-'
0.10 x 10-1
0.10 x 10-1
0.10 x 10-1
0.10 x ID"1
0.10 x ID'3
0.50 x ID"3
0.10 x 10-1
0.40 x 10-1
0.85 x ID"4
0.10 x ID"2
0.10 x 10-1
0.16 x ID"2
0.75 x 10-3
0.12 x TO'2
0.10 x TO'3
0.10 x ID'4
0.11 x 10-3
0.25 x ID'3
0.10 x lO'4
0.10 x TO'5
In-place
value of
materials
exposed,
$ billion
23.9
26.83
316.21
10.40
3.80
5.53
14.00
54.08
54.88
2.48
0.54
10.76
0.95
0.82
24.15
2.27
7.53
5.15
1.98
17.61
7.65
1.54
33.12
1.37
0.33
6.50
1.17
1.00
2.18
0.85
0.19
0.19
0.16
0.64
0.61
0.47
0.40
22.45
3.86
0.17
0.04
18.90
1.44
0.12
0.58
1.08
0.57
5.80
28.59
2.18
0.51
1.93
0.30
Economic
loss,
$ million
1195.0
778.0
316.0
260.0
152.0
144.0
140.0
114.0
110.0
99.2
54.0
53.8
38.0
32.8
24.2
22.7
22.6
20.6
19.8
17.6
17.6
15.4
13.9
13.7
13.2
13.0
11.7
10.0
8.7
8.5
7.6
7.6
6.4
6.4
6.1
4.7
4.0
2.2
1.9
1.7
1.6
1.6
1.4
1.2
0.9
0.8
0.7
0.6
0.3
0.2
0.1
0.02
0.00
3800.00
20
COST OF AIR POLLUTION DAMAGE
-------
for zinc and SOg. several varieties of rubber and 03, and for cotton
and S02- In summary, major information shortages exist concerning air
pollution effects on concrete, paints, fibers, and plastics.
This effort is a landmark in summarizing the effects of air pollu-
tion on materials and lays the groundwork for more intensive studies.
Some of the data, such as those used in formulating interaction and
corrosion rates, are of questionable validity; and in a quantitative
analysis, such as that used here, results are only as good as the data
used. Only the direct effects were investigated; the "value" of ser-
vice was not assessed. As Salmon cautioned in his report, deterioration
of materials indicated susceptibility to economic loss, or potential
loss. The results could not be interpreted as economic loss actually
incurred. A primary purpose of the study was the ranking of materials to
indicate relative measures of air-pollution-induced damage.
Haynie - Corrosion of Galvanized Materials
On the basis of work done by Midwest Research Institute on zinc,
19
Haynie has suggested that the annual cost of the corrosion of galvan-
ized steel, including prevention costs, is $4.5 billion. This estimate
is the total of several costs: (1) costs of coil coating for corrosion
prevention, (2) costs of substituting aluminum sheet for steel because
of corrosion, and (3) replacement costs. By taking different fractions
for various product forms, different economic lives, and different
"in-place" to zinc ratios and combining them correctly, the extreme
values were calculated. The minimum annual value is $1.4 billion and
the maximum is $13 billion. Because of the lack of evidence for most
of the assumptions used in estimating the many different fractional
parts, the minimum value of $1.4 billion probably reflects the most
defensible estimate.
Battelle Columbus Laboratories* - Rubber Products
20
The Battelle study, entitled, "Survey and Economic Assessment
of the Effects of Air Pollution on Elastomers," correlates technical
information relating to the effects of air pollution on rubber products
with an estimation of the yearly cost of this pollution damage. Costs
measured are: (1) increased expenditures required at the manufacturers'
*Formerly the Battelle Memorial Institute.
Effects on Materials 21
-------
level to provide products that are resistant to atmospheric pollutants
(these costs are normally passed on to the consumer) and (2) direct
economic loss on the part of the consumer in the form of shortened
useful life of products. The combined costs to the consumer were used
in estimating the total cost of pollution.
The literature was reviewed for technical information and question-
naires were sent to industry to determine the cost of atmospheric pol-
lution at the manufacturing level. Sixty questionnaires were sent to
firms representing the broad spectrum of the rubber industry; 30 were
returned, and many of them had unanswered questions.
In estimating the cost of atmospheric pollution at the rubber pro-
duct manufacturers' level, two independent calculations were made. One
was based on the information supplied from the questionnaires sent to
industry, whereas the second calculation was based on the total of in-
dividual compounding costs.
Analysis of the industry questionnaire and some extrapolation of
results revealed that the yearly cost for rubber products of $50.7
million represents the added yearly cost of compounding new or improved
products. To calculate the added cost at the manufacturers' level, it
was necessary to add individually, where available, the estimated costs
of resistant polymers, antiozonants, waxes, protective finishes, and
wrappings. Estimated costs are given in Table 6. The two figures of
$50.7 million and $56.4 million, which are costs at the manufacturers'
level, compare favorably. Industry estimates suggest that an average
retail price is three times the manufacturing cost. If this is true,
then at the retail level the cost of compounding would be approximately
$150 million to $170 million.
Table 6. ESTIMATED COSTS OF AIR POLLUTION RESISTANT MATERIALS
Resistant polymers
Antiozonants
Waxes (50% of total)
Protective finishes
Wrappings
Research for compounding
Total (approximate)
$ million
$17.31
34.00
5.00
7
7
7
$56.40
22 COST OF AIR POLLUTION DAMAGE
-------
The information in Table 7 summarizes the cost of early replacement
of rubber products (shortened service life). The total annual cost of
air pollution as it affects the rubber industry is estimated to be $150
million to $170 million plus $216.7 million, or approximately $380 million.
The first is the added cost at the manufacturers' level that is passed on
to the consumer in the form of an increased price for the product. The
second cost is for the early replacement of rubber products because of
shortened service life, a cost that must be borne directly by the
consumer.
Table 7. COSTS OF SHORTENED SERVICE LIFE OF RUBBER PRODUCTS
Tires
Innertubes
Footwear
Mechanical goods
Medical goods
Belting
Hoses
Total
$ million
$ 32.2
-
-
27.6
100.5
21.9
34.5
$216.7
Finally, it was determined that almost all damage to rubber is
done by ozone. Very little is known about the effects of other
pollutants on elastomers. Virtually no information is available on the
damage threshold for rubber, so that few or no data are available for
the construction of a meaningful damage function.
This study is deficient because it deals only with gross figures.
Labor costs in connection with the early replacement of rubber products,
not considered in this report, would add substantially to any estimate
of losses. Although an estimated annual cost of almost $400 million
was presented, this figure has no sound basis.
Salvin - Textiles
Victor S. Salvin, of the University of North Carolina, is engaged
in a study of economic losses resulting from air pollution damage to
21
textiles. The objectives of the study are: (1) to conduct a compre-
hensive survey to identify and document known and suspected air-pollu-
tion-induced effects on various textiles and dyes and (2) to survey the
Effects on Materials 23
-------
economic losses resulting from air pollution effects on textile fibers
and dyes.
Aspects of the problem to be discussed with manufacturing and in-
dustrial representatives are: (1) extent and mechanisms of air pollu-
tion damage, (2) preventative measures required, and (3) research
costs. The costs of dyes and dyeing processes (preventative measures)
will be obtained from the suppliers of dyes. Each industry will have
a technology committee that will serve as a clearinghouse for informa-
tion. Production figures for each industry, awareness in each industry
of the aspects of air pollution damage that affect it, and actions by
major manufacturers in offering goods with increased performance poten-
tial will be documented. The costs in this case will be those associ-
ated with research, quality control, the use of expensive dyes and
textiles, and the use of associated costly new procedures. Consumer
awareness of air pollution effects will be assessed through direct
interviews and the use of questionnaires. The annual cost of air pollu-
tion damage to textile fibers and dyes will be estimated. This economic
analysis will not only include estimates of costs resulting from damage
but also the costs of steps taken to mitigate or eliminate damage, such
as (1) substituting more costly materials, (2) using additional protec-
tion, (3) using more expensive production techniques, (4) initiating
closer quality control, (5) conducting relevant research and development,
(6) conducting environmental testing, and (7) providing consumer sales
services.
Preliminary estimates of the costs of the fading of dyes on tex-
tiles caused by air pollutants are given in Table 8. These numbers
are tentative and qualified, inasmuch as the final figures will be based
upon more complete production data and remedial processing costs. This
is the only information of its type available. There are few or no
supporting economic data and none that would contribute to the construc-
22
tion of any damage function. Other preliminary estimates from Salvin
indicate that the total economic loss caused by air pollution damage to
textiles and fibers is estimated to be $2 billion annually. Included
in this figure are: (1) $300 million to $400 million for the shortened
useful life of cotton and nylon resulting from the disintegration of the
fibers, (2) $800 million for the cost of extra laundering and dry clean-
ing (this is considered in the section on soiling), (3) $350 million
for the fading of fabrics, and (4) the remainder for the discoloration
of white fabrics.
24 COST OF AIR POLLUTION DAMAGE
-------
Table 8. ESTIMATED COSTS OF DYE FADING IN TEXTILES
Pollutant
NOX
03
Effect
Fading on acetate and triacetate
Fading on viscose rayon
Fading on cotton
Yellowing of white acetate-nylon-Spandex
Subtotal
Fading on acetate and triacetate
Fading on nylon carpets
Fading on permanent-press garments
Subtotal
Total
$ million
$ 72.800
21.600
22.050
5.650
$122.100
24.985
41.500
17.050
83.535
$206.000
NATIONAL ESTIMATE
The Midwest Research Institute report offers evidence on which a
reasonable national estimate can be based. These data generally will
be used except for those cases in which more intensive analyses have
been made, such as costs of air pollution'effects on galvanized steel
and elastomers. Thus, to prevent any double counting, the material loss
values of zinc, synthetic and natural rubber, and carbon and alloy steel
were omitted from Table 5 in order to obtain an estimate of material
loss for categories other than galvanized steel and rubber.
The $9.5 billion estimate is not used because of the lack of any
data that would suggest a similar response to air pollution of those
materials not included as compared to those included in the $3.8 billion
figure. The assumption is made as well that those materials represented
in Table 5 are the ones most affected physically, and thus economically,
by air pollution. For consistency with other estimates, the conserva-
tive figure of $3.8 billion is considered only in the national estimate.
If the above materials are omitted from the value in Table 5, the
economic loss then becomes $2.766 billion. If Haynie's value of $1.400
billion for the cost of corrosion of galvanized steel is considered,
plus the Battelle estimate of $0.380 billion for effects on elastomers,
plus the Salvin estimate of $0.206 billion lost because of the fading of
Effects on Materials
25
-------
dyes-which was not considered in the Midwest study-then the total cost
of materials is estimated at $4.752 billion annually.
In all probability, the property-value approach includes implicitly
some portion of the material costs estimated here. It is assumed, how-
ever, that those material costs contained in the differential property-
value approach are not a large proportion of the total costs of air
pollution damage to materials.
26 COST OF AIR POLLUTION DAMAGE
-------
CHAPTER 4. EFFECTS ON VEGETATION
Damage to vegetation as a result of air contaminants has been
recorded in the United States since the turn of the century. What was
once a problem associated only with point sources has evolved into a
widespread problem generated by urban expansion. The problem is severe
enough that commercial and non-commercial production of crops and
forests in many areas has been jeopardized and in some cases discontinued.
Several approaches have been used to assess the economic losses
resulting from air pollution damage to plants. One approach has been to
survey air-pollution-induced vegetation losses on a statewide basis by
using existing manpower at the local level, such as county agricultural
agents and commissioners. From these estimates of damage at the local
level, extrapolations can be made that provide an estimate of gross
national damages. Another approach is that of incorporating data on
pollutant emissions, crop statistics, and meteorological parameters into
a predictive model of plant losses; this model is subject to continual
refinement as the true situation at the local level is better defined.
INDIVIDUAL STUDIES
Middleton and Paul us - Crop Survey
Economic assessment by manpower at the local level was first done
in the California survey made in 1949. A somewhat similar survey in
1955, reported by Middleton and Paulus, was designed to show the loca-
tion of injury, the crops injured, and the toxicant responsible for the
damage. Specialists in agriculture throughout the state were trained as
crop survey reporters. The survey covered four categories of crops:
field, flower, fruit, and vegetable.
Lacasse, Weidensaul, and Carrol - Pennsylvania Plant Survey
A program similar to that in California was established in Pennsyl-
vania in 1969, where a training course for county extension agents,
university extension specialists, and state forestry extension personnel
was held to teach them how to identify and evaluate air pollution damage
to plants.24 The objectives for the survey were: (1) to estimate scien-
tifically the total cost of agricultural losses resulting from air
27
-------
pollution in Pennsylvania, (2) to determine the relative importance of the
various pollutants in Pennsylvania, (3) to survey the extent of the air
pollution problem in Pennsylvania, (4) to provide a basis for estimating
the nationwide impact of air pollution on vegetation, and (5) to provide
a basis for guiding research efforts.
A professional plant pathologist was enlisted to coordinate the
field survey. He assisted reporting personnel in the detection and
evaluation of air pollution damage to crops and performed independent
field surveys in areas where sources of pollution were located. Commer-
cial and non-commercial plants were studied. Past air pollution episodes
were investigated for purposes of detecting possible trends. Estimates
of losses were based on crop value and production costs incurred by
harvest time. Direct losses to producers or growers included only pro-
duction costs, whereas indirect losses included profit losses, costs of
reforestation, grower relocation costs, and the cost of substituting
lower-value crops for higher-value crops. Costs associated with destruc-
tion of aesthetic values, erosion and resultant stream silting, damage
to watershed retention capacity, and farm abandonment were not considered.
Of the 92 field investigations made, 60 revealed damage that was
attributable to air pollution. Damage resulting from pollution was
observed in 23 counties, most of which were located in southeastern and
western Pennsylvania. Direct losses uncovered in the survey exceeded
$3.5 million. The air pollutants responsible for the damage, in order of
decreasing importance, were: oxidants, sulfur oxides, lead, hydrogen
chloride, particulates, herbicides, and ethylene. The vegetation most
affected, also in order of decreasing importance, was: vegetables,
fruits, agronomic crops, lawns, shrubs, woody ornamentals, timber, and
commercial flowers. Indirect losses were estimated at $8 million, of
which $7 million reflects profit losses, $0.5 million reflects reforesta-
tion costs, and the remainder reflects costs for grower relocation. In
summary, economic losses in Pennsylvania from air pollution damage amount
to approximately $11 million annually.
The major criticisms of the Pennsylvania effort reflect the state
of the art. Little is known of the extent to which home garden plantings
and flowers are being affected by air pollution and, if they are affected,
the value that should be assigned to these losses. The method used in
assessing losses is somewhat questionable because grower profit losses
28 COST OF AIR POLLUTION DAMAGE
-------
are not included as direct costs. Also, methods of translating physical
injury into economic loss have not been standardized.
The major strengths in such an approach are that: (1) existing
manpower can be utilized for achieving continual coverage over an area;
(2) agents at the local level have rapport with growers in that area;
are familiar with crop peculiarities; and are probably knowledgeable
about local sources of pollution in the area; and (3) a field coordina-
tor supplies expertise to the reporting personnel and provides some
degree of standardization in reporting losses.
Because of the successful first year's attempt in Pennsylvania and
the inherent advantages in such an approach, a similar statewide survey
was initiated in California in the summer of 1970. The added data from
other states will make possible more realistic estimates of losses.
Stanford Research Institute - Nationwide Survey
Another major survey of plant damage, "Economic Impact of Air Pollu-
25
tants on Plants," has been undertaken by H. M. Benedict of Stanford
Research Institute(SRI). This study, initiated in 1969, will develop
an estimate of the annual economic losses to agriculture, in all regions
of the United States, that result from damage to vegetation by air
pollutants. Special emphasis will be placed on those losses ascribable
to automotive emissions or to pollutants arising as a result of auto-
motive emissions.
SRI will: (1) develop a method for assessing agricultural losses
caused by air pollutants, based on established evaluation methods now
used to assess agricultural losses from natural causes such as plant
pathogens or insects; (2) determine the type and general extent of air
pollution injury to agricultural crops of major importance and the
symptoms of such injury; (3) locate the principal sources or areas of
pollution and estimate the economic value of plant damages, placing
special emphasis on the economic loss attributable to specific automotive
emissions; (4) consider the possible influence of meteorological condi-
tions affecting the exposure of vegetation to air pollutants in various
geographical regions; and (5) estimate the yearly dollar losses to
agriculture that result from damage to vegetation by air pollutants.
SRI reviewed the literature to obtain information on the concentra-
tion and duration of fumigations necessary to affect plants and the nature
Effects on Vegetation 29
-------
of the effects produced. Counties to be surveyed were identified by
locating potential sources of air pollution throughout the United
States. The economic value of agricultural production in each of these
counties was estimated. Theoretical consideration was given to possible
models for estimating losses resulting from other factors such as
disease, insects, and weeds. The major pollutants that have been
identified by SRI as causing 90 percent of the damage to plants are
ozone, peroxyacetyl nitrate, ethylene, nitrogen oxides, sulfur oxides,
and fluorides.
Emissions in standard metropolitan statistical areas were estimated
from known fuel consumption. Crop production and emissions were esti-
mated for the 500 most important counties in the country. In the
initial phases of the study, estimates have been confined to losses
caused by the direct effects of major pollutants on yield, quality, and
marketability. Preliminary estimates indicate that these direct effects
amount annually to about $70 million nationwide. Hydrocarbons have been
associated with losses of about $64 million, sulfur oxides with losses
of about $4 million, and fluorides with losses of about $2 million. If
it is true that this reflects 90 percent of plant losses, then the total
direct loss of selected crops approaches $80 million annually.
The major benefit from the SRI study will be the accumulation of
good background data for the development of predictive models for use in
estimating losses. The basic assumptions made of the extent to which
crops are exposed to pollution as well as the general severity of damage
to plants at varying emission levels are not altogether realistic as
given in this study, but perhaps they are the most reasonable possible
at this time. Also, this study considers observable, direct effects
only. Home plantings and flowers, which some experts believe comprise
the largest portion of plant losses caused by air pollution, have again
been largely ignored.
NATIONAL ESTIMATE
The preliminary estimate of direct losses in Pennsylvania by the
SRI survey amounts to about $4.9 million (for oxidants only). This
compares with the estimate of the statewide survey in Pennsylvania of
about $9.0 million (profit losses added). If it is assumed that the
Lacasse-Weidensaul-Carroll estimate for losses in Pennsylvania is more
30 COST OF AIR POLLUTION DAMAGE
-------
defensible than that made by SRI, then by substituting the former esti-
mate into the SRI model, a realistic estimate of direct annual losses
for the nation, based on these two studies, might be approximately $120
million for 1968.
It should be noted that this estimate represents those losses
caused by the visible, direct effects of the major pollutants on
agronomic, field, horticultural field, and forest crops only. Losses
of ornamental plantings, flowers, and gardens are assumed to be implicit
in the property-value estimate of air pollution costs. Also ignored,
because of the lack of meaningful data, are losses resulting from
indirect effects of air pollution on agriculture, such as grower reloca-
tion, the denudation of land and resultant erosion, and lost productivity.
Effects on Vegetation 31
-------
-------
CHAPTER 5. SOILING
Individuals, households, and commercial establishments are affect-
ed by air pollution in many ways, only a few of which are obvious.
When dust particles fall, the need to dust window sills and furniture
is distressingly obvious. The effects of air pollution in most cases,
however, are gradual enough to go unnoticed. Yet dealing with these
effects may involve considerable extra expense, of which the household
is usually unaware. Some families in urban areas spend very little as
a result of air pollution, but many spend hundreds of dollars more each
year than they would need to spend if the air were clean. The costs
associated with soiling of materials are discussed under the following
reviews of studies of soiling from air pollution.
INDIVIDUAL STUDIES
Mellon Institute - Pittsburgh Smoke Nuisance
Surely the best known of the early studies of economic losses re-
sulting from air pollution is the Mellon Institute Study of the
oc
Pittsburgh smoke nuisance in 1913. The purpose of the study was to
assess the economic cost of the smoke nuisance to the populace of the
city of Pittsburgh. The cost estimates were based upon literature
searches, observations, and informal surveys. The estimated annual cost
(1913), shown in Table 9, was more than $9.9 million, or $20 per capita.
The damage estimates obviously included some direct costs as well
as some adjustment costs. The costs involve losses resulting from soil-
ing by particulate matter as well as corrosion, depending on the damage
category, and the obstruction of sunlight by particulate matter.
No attempt was made to measure the extra cost of cleaning the inte-
rior of public buildings and of cleaning and maintaining the exteriors of
buildings, nor costs of health, agricultural, and aesthetic properties.
Questionable statistical techniques were used in averaging damage costs,
in estimating the number of units affected (for example, stores), and
in arriving at the percentage damage cause by air pollution. As Kneese
has said, "Perhaps one can be forgiven for suspecting that this informa-
tion is somewhat dated."27
33
-------
Table 9. SUMMARY OF DATA FROM MELLON INSTITUTE STUDY ON SOILING
26
Cost
Cost to the smoke emitter resulting from imperfect
combustion
Cost to the individual
Laundry
Dry cleaning
Cost to the household
Exterior painting
Sheet metal work
Cleaning and renewing wallpaper
Cleaning and renewing lace curtains
Artificial lighting
Cost to wholesale retail stores
Damage to merchandise
Extra precautions
Inside maintenance
Artificial lighting
Department stores
Cost to quasi-public buildings (inside maintenance
and artificial lighting)
Office buildings
Hotels
Hospitals
Total
$/yr
$1,520,740
1,500,000
750,000
330,000
1,008,000
550,000
360,000
84,000
1,650,000
450,000
750,000
650,000
175,000
90,000
22,000
55,000
$9,944,740a
a$20 per capita in 1913.
Beaver Report - London Smog Episode
The next major attempt to estimate soiling costs was an outgrowth
of the Mellon Study. In 1953, as a result of the London smog episode of
1952, a committee was appointed to examine the nature, causes, and
effects of air pollution and the efficacy of preventive measures. The
28
report of this committee to Parliament was released in 1954.
Many of the data were secured through literature searches and in-
formal surveys. The actual method used to make the estimates was
similar to that used in the Mellon Study. In the Beaver Report,28 how-
ever, "black" areas were compared with "clean" areas, whereas in the
34
COST OF AIR POLLUTION DAMAGE
-------
Mellon Study, Pittsburgh was compared with different cities. Costs
were assessed by estimating the proportion of the total expenditure for
specific items damaged by air pollution. The necessary proportional
estimates were obtained from additional estimates of the amount and
frequency of expenditures in polluted as compared with nonpolluted
areas, as determined by interviews with local authorities. The pollu-
ted areas used in the study were assumed to account for half of all
other items, such as painting of buildings.
The net results of the study are given in Table 10.
Table 10. SUMMARY OF DATA FROM BEAVER REPORT ON SOILING26
Cost
Direct
$ million/yr
Laundering "'"'
Painting and decorating
Soiling and depreciation of buildings
(other than houses)
Corrosion of metals
Damage to textiles and other goods
Indirect - loss of efficiency
Total
Cost/person
Nonpolluted areas
Polluted areas
56
70
147
280
707
$/yr
14
28
It is evident that this method of estimation resulted in extremely
crude sums that do not reflect variation with pollutant level or type
of pollutant. Where little or no information was available, the authors
did not hesitate to use pure guesswork. They recognized, however, that
their results did little more than suggest broad orders of magnitude.
Michelson and Tourin - Household Costs
29 31
In recent years, several attempts have been made to identify
the costs of soiling from air pollution. For the most part, these
studies have dealt with the household as the primary unit of investiga-
tion in an attempt to measure pollution-related cleaning and maintenance
costs in certain localities. The review by Jones-'2 is particularly
helpful in evaluating the Michelson-Tourin study.
Soiling
35
-------
In the evaluation of household costs resulting from soiling caused
by air pollution, the work of Michelson and Tourin has received the most
attention. Their method of study is based upon the theory that if air
pollution causes significant soiling, the extent of damage by soiling
may be reflected in shortened time intervals between successive cleaning
and maintenance operations in areas with higher levels of pollution.
If this relationship could be established, then by using the cost of
each operation studied and the damage functions relating particulate
level and costs, the costs of soiling in a polluted environment could
be calculated.
To test this theory, Michelson and Tourin conducted a survey by
mailed questionnaire in the towns of Steubenville and Uniontown in the
Upper Ohio River Valley.29 These towns have annual average particulate
matter concentrations of 235 ug/m3 and 115 yg/m^, respectively. A good
rate of response to the questionnaire was achieved through a large
publicity campaign. A positive relationship was found to exist between
the frequency of cleaning of the house and personal care items and
levels of particulate pollution.
Cost comparisons were made between two income groups (less than
$8,000 and more than $8,000), and total costs were calculated on the
basis of the number of families and persons in each income group in
each city. The differences in frequency of cleaning were calculated and
then converted into dollar differences by applying local market prices
for the various household services covered by the survey. The resulting
figures showed that the economic loss associated with air pollution in
Steubenville was $3.1 million, or $84 per capita, more than in Uniontown.
In an attempt to validate this study, a subsequent survey was con-
ducted in three suburban cities of the Washington, D. C., area.
The Washington area was chosen for the validation study because it was
thought to offer a severe test to the methodology. First, the absolute
levels of suspended particulate matter in the D. C. area were much
lower than those of Steubenville and Uniontown. Second, the difference
in the levels of suspended particulate matter was much smaller in the
paired cities in the Washington area than in the paired cities in the
Upper Ohio River Valley. Finally, the two areas had very different
industrial and population characteristics.
36 COST OF AIR POLLUTION DAMAGE
-------
Michelson and Tourin again found a positive relationship between
the frequency of performance of cleaning and maintenance operations and
the level of suspended particulate matter in the ambient air. Although
the findings of the second study would seem to support the findings of
the first, major differences between the two studies and inherent prob-
lems within each throw a great deal of doubt upon this conclusion. For
example, although income level was the only controlling factor in the
analysis, only the responses of the above-average income group were
analyzed in the Washington study. Once the relationship was found to
exist in that income group, it was assumed to exist for the below-
average income group. Also, within each study there are considerable
problems with the sample survey design, the data, and the analysis.
The chief problem, however, is the lack of statistically reliable tech-
niques that would allow a certain degree of confidence in the results.
Since these two major studies, Michelson and Tourin have applied
their methodology in other sections of the country. In 1968, they com-
pleted a study of the total extra household costs resulting from air
32
pollution in Connecticut. In these studies, no household survey was
performed to measure the frequency of cleaning and maintenance opera-
tions. Instead, the frequencies found in the Upper Ohio River Valley
and Washington area studies were used. Because these frequencies were
not alike, some kind of averaging must have been done. The local costs
of the operation were investigated, and the demographic figures from
census materials were used to come up with a total damage estimate for
the state of Connecticut. Such a use of the methodology without ade-
quate verification is highly questionable.
Ridker - Urban Soiling
Ridker conducted research to identify costs associated with soil-
33
ing caused by air pollution. In 1965, he conducted a study in high-,
medium-, and low-pollution zones of Philadelphia to determine whether
family behavior and expenditures were affected by air pollution. Des-
pite an apparently adequate collection of data, the results of the
analysis were inconclusive. Although many detailed problems and errors
were apparent in the analysis, the principal problem involved the use
of time expended in routine household cleaning as a basis for cost
estimates. The relative frequency with which these tasks are performed
may be a more appropriate measure than the time spent performing them.
Soiling 37
-------
Ridker also conducted a time-series analysis of a pollution episode
in Syracuse.34 A questionnaire was developed and administered through
personal interviews. Although the results of this household survey
were much better than the cross-sectional analysis made in Philadelphia,
the approach was obviously limited to the episode-type situation and
could not be put to widespread use.
The Ridker study, along with the Michelson and Tourin studies, call-
ed attention to major problem areas with regard to evaluating household
expenditures resulting from soiling caused by air pollution:
1. Isolation of economic losses resulting from air pollution
from those resulting from other causes.
2. Sample selection and bias.
3. Development of a survey technique that will provide
reliable answers.
4. Inclusion of all household tasks or services whose costs are
influenced by soiling damage from air pollution.
Ando - Costs of Air Pollution in Japan
Itaru Ando attempted in his report, "Social Expenditures for Public
Hazards and Their Apportionment,"35 to study the causes of public haz-
ards, to review the present status of social expenditures, and to inves-
tigate the shares of expenses allotted to reduce hazards produced by air
pollution.
Ando took the following steps in isolating the effects of air
pollution: (1) examined the fact that public hazards have increased
rapidly for the last 20 years; (2) reviewed the growth of economic
complexes; (3) discussed the direct losses caused by public hazards and
the problem of representing these in economic units; (4) used dustfall
and S02 as representative indices of air pollution; and (5) patterned
cost estimates after the Mellon Study of 1913.
Ando estimated that direct losses to Kawasaki City from air pollu-
tion damages were nearly $5 million, or approximately $19 per household.
Health expenditures accounted for $1.1 million; damage to residential
property and household goods accounted for $2 million; damage to mer-
chandise accounted for $0.05 million; and other damage accounted for
$1.66 million. Average household costs attributable to air pollution
were $38.89 in Osaka and $28 in Sapporo City.
38 COST OF AIR POLLUTION DAMAGE
-------
The scope of this study goes beyond that of other studies because
the author relates public hazards to industrial development. Commenting
that it is impossible to enrich the national life without industrial
development, he states that the residents and businessmen in an area
must bear expenses in proportion to their restored profits. A weak
point of the study is that the author patterned his cost estimates
after those of the Mellon Study.
Booz-Allen and Hamilton - Philadelphia Survey
Q/-
The current Booz-Allen and Hamilton, Inc., study in Philadelphia
is expected to improve upon and extend the methodologies already
developed. The objectives of the Philadelphia community study are:
(1) to determine the impact of various air quality standards and regu-
lation schemes on the region by using a benefit-cost approach and
(2) to extend existing knowledge of how to assess damages resulting
from air pollution.
A questionnaire consisting of two parts was developed to determine
the frequency of cleaning. The first section included questions regard-
ing cleaning operations and the second consisted of a set of self-
referent statements designed to determine cleaning attitudes. A total
of 1,800 personal interviews were conducted in the Philadelphia Air
Quality Control Region.
An attempt was made on a smaller scale to determine the costs of
soiling borne by commercial establishments because of particulate pollu-
tion. A sample survey of 138 stores was conducted, and various cleaning
operations were investigated. Because of sample size, the results have
proved inconclusive thus far.
Rigorous statistical survey research techniques were employed from
the beginning of the project because of the initial belief that many,
perhaps more, dominant, non-pollution-related variables explain differ-
ences in residential cleaning frequencies and maintenance operations to
a far greater degree than the variations in the annual air particulate
level measurements in the Philadelphia area. Therefore, the survey
techniques included: (1) a probability sample within several zones of
the Philadelphia area; (2) group interviews leading to preliminary
estimates of attitudes toward cleaning and the best ways of phrasing
survey questions; (3) personally administered questionnaires, rather
Soiling 39
-------
than mail or telephone surveys; (4) a factor analysis of the question-
naire respondents to separate the population into attitude groups in
order to explain better why people clean; (5) collection of demographic
data on each respondent and his residence; and (6) the use of qualified
interviewers to perform coding and key-punch operations.
The study of residential household soiling costs made an attempt to
discern between cleaning necessitated by pollution and cleaning performed
out of habit or for other reasons. Before any relationship could be
established between the frequency of these cleaning operations and the
level of particulate pollution, other socioeconomic variables that may
contribute to the frequency were identified and the degree of their
interaction was established through a factor analysis.
From the study of 27 cleaning and maintenance operations, prelimi-
nary results indicate that the range of annual air particulate levels
experienced in the Philadelphia area (approximately 50 to 150 micrograms
per cubic meter) had no statistically significant differential effect
on the residential cleaning and maintenance costs for over 1,500,000
households in the area. These operations included painting, cleaning,
and washing. Of the 27 operations studied, 11 were determined to be
somewhat sensitive to air particulate levels. These are shown in Table
11. Each of the sensitive operations is a low-cost, do-it-yourself
item, and many are associated with being able to see out of the houses -
washing windows, cleaning screens, and cleaning Venetian blinds. It
must be pointed out that these do-it-yourself operations were considered
to be free of labor cost and the material costs were considered only
when these were non-trivial, such as for painting. Theories on various
methods of imputing labor costs to household member operations were not
applied partly because these costs are not direct costs.
The preliminary conclusion of the study is that some low-cost
cleaning and maintenance operations appear to be sensitive to air par-
ticulate levels, but, more important, that the high-cost operations are
unaffected by variations in air particulate levels in the Philadelphia
area. Another finding of interest is that a higher proportion of resi-
dents of high pollution areas believed their neighborhoods were dirtier
than residents of low pollution areas believed theirs to be.
The Booz-Allen report has been criticized on several grounds. First,
statements concerning the statistical significance of operations have
40 COST OF AIR POLLUTION DAMAGE
-------
Table 11. RELATIONSHIP OF CLEANING AND MAINTENANCE
OPERATIONS TO AIR PARTICULATE LEVELS36
Relationship
Sensitive
Insensitive
Inside
Clean and oil air conditioners
Clean furnace
Clean Venetian blinds and shades
Dry-clean carpeting
Dry-clean draperies
Paint walls and ceilings
Replace air conditioner filter
Replace furnace filter
Shampoo carpeting
Shampoo furniture
Wai 1 paper walls
Wash floor surfaces
Wash walls
Wash windows (inside)
Wax floor surfaces
Outside
Clean and repair awnings
Clean and repair screens
Clean and repair storm windows
Clean gutters
Clean outdoor furniture
Maintain driveways and walks
Maintain shrubs, flowers, etc.
Paint outside trim
Paint outside walls
Wash automobiles
Wash windows
Wax automobiles
not been adequately justified in many instances. Second, the report
does not fully explain the sensitivity or insensitivity of the cleaning
and maintenance operations. Third, accepted economic principles justify
including with the cost of materials some imputed value for homemakers'
time spent in cleaning and maintenance operations. This study should
Soiling
41
-------
provide guidelines for improving the methodology used in estimating costs
and applying such methodology to other regions,
CONCLUSIONS
In conclusion, the Michelson and Tourin and Ridker studies dealt
mainly with the estimation of household cleaning and maintenance costs.
Except for the Michelson and Tourin study, however, the evidence to date
indicates that air pollution does not have significant economic effects
in terms of household maintenance and cleaning operations. Yet, there
are other costs associated with pollution soiling that deserve attention
if not estimation. Some of these costs are: commercial cleaning and
maintenance costs; individual adjustments such as laundering, dry clean-
ing and hair and facial care; car washing; and costs to quasi-public
properties, which might include cleaning and maintenance costs of buil-
dings and monuments and washing of street-lighting luminaries. The
magnitude of soiling costs associated with specific effects undoubted!/
runs into the millions of dollars annually, but because of the lack of
data, these soiling costs are not estimated in ..his report.
42 COST OF AIR POLLUTION DAMAGE
-------
CHAPTER 6. ANIMAL HEALTH EFFECTS
Air pollution has many effects on animals. Some of the oldest
documented cases of the deleterious effects of air pollution have been
associated with the Meuse Valley disasters, notably those of 1897, 1902,
and 1911. It is evident from reports of those disasters that vegetation
was damaged and that cattle suffered from a malady, caused by the adverse
atmospheric conditions, known locally as "fog disease."^ In actuality,
op
the cattle had been stricken with asthma and emphysema.
Pollution of the air and damage to plants used as animal feed -
and the resulting diseases of animals - are frequent in regions where
39
metallurgical plants are located. Fluorine by-products, lead, and
molybdenum are the major offending constituents in industrial air
pollution. Cement dust, sulfur dioxide, and arsenic are lesser problems.
Pollution of agricultural, as opposed to industrial, origin is essen-
40
tially linked to the misuse of pesticides.
Fluoride poisoning of cattle grazing in the vicinities of aluminum
reduction and phosphate fertilizer plants has received much attention
in the literature. Fluorosis, a disease common to cattle, occurs when
fluorine compounds are ingested for long periods of time. The animals
eat contaminated fodder, grass, and hay, and also inhale quantities of
fluorine. No adverse effects occur in dairy cattle when the fluorine
content of the daily ration is less than 30 ppm, but 40 to 60 ppm will
result in moderately adverse effects. Acute fluorosis results from in-
gestion of 250 ppm fluoride per day, a level that often results in stiff-
41
ness, anorexia, weakness, convulsions, and cardiac failure. Chronic
fluorosis is typified by the ingestion of 60 to 100 ppm fluoride per dav
and causes severe dental malformations and bone lesions. Cattle often
produce less milk and conceive poorly as a result of fluorosis.
Losses to livestock have been known to occur in the vicinity of
lead refineries. Cattle raised in the area were estimated to have eaten
26 milligrams of lead in 100 grams of herbage. Near the plant they ate
0.07 to 0.16 percent lead by weight, which resulted in lead poisoning.
Molybdenum dust, scattered from the chimney of the neighboring molyb-
denum-smelting factory, also allegedly caused damage to livestock.
43
-------
The cattle experienced diarrhea, symptoms of malnutrition, decreased
milk production, and decreased rate of conception.
Animals also have been damaged by the effluents of a copper smelter.
The high-copper- and arsenic-containing effluents that are deposited on
plants and grass cause numerous cases of chronic poisoning and even
45
death of domestic animals such as cattle, horses, sheep, and poultry.
A number of general conclusions have been drawn regarding effects of
air pollution on wildlife. From field investigations, the economic
poisons (insecticides, herbicides, etc.), such as chlorinated hydrocar-
bons and organic phosphates, appear to outweigh by far all other types of
air pollutants as hazards to wildlife health. The hazard to wildlife
occurs chiefly from ingestion of the "fallout" of the air pollutant.
Relative species susceptibility to specific air pollutants is far from
clear, but it would appear that mammals are considerably more suscep-
46
tible than birds. Nevertheless, air pollution has been implicated as
the causal agent of primary lung cancers in birds in the Philadelphia
zoo. Waterfowl that were kept outdoors the year round were the animals
47
most affected, which led Snyder to suggest the possibility that the
levels of carcinogens in the atmosphere are increasing.
No studies of the economic impact of air pollution on animals have
been made. The damage to animals caused by air pollution has been
localized and its economic consequences have been probably relatively
unimportant; yet the social consequences of this pollution damage are
potentially more severe. Though indirect, the risk to the food cycle,
especially when pesticides are implicated, is serious, and it may indeed
be true that the economic importance of heavy metal air pollutants lies
in their effects on animals.
44 COST OF AIR POLLUTION DAMAGE
-------
CHAPTER 7. EFFECTS ON AESTHETIC PROPERTIES
One possible effect of air pollution is the deterioration of
materials, with historic artistic significance, such as paintings,
statuary, and rare books. In addition, air pollution can reduce visi-
bility and obscure vistas and thus can have a depressing psychological
effect on individuals. Noxious odors as well are considered here to
have aesthetic effects.
The adverse aesthetic effects of air pollution belong in the cal-
culus of air pollution damages because of values that can be attached
to their prevention or avoidance. For example, the New York City Public
Library spent $900,000 between 1952 and 1967 to microfilm books that
were in an advanced state of deterioration largely as the result of air
48
pollution. Part of this expenditure represents what the library was
willing to pay to avoid loss of its books. Presumably other examples
could be found that indicate a willingness-to-pay for avoiding the
adverse aesthetic effects caused by air pollution.
Studies of the costs of air pollution in terms of damage to aesthe-
tic properties are discussed below.
INDIVIDUAL STUDIES
Ridker - Soot Incident Survey
49
Ridker surveyed the residents near a power plant to determine
the cost of cleaning up after malfunctioning boilers fumigated the
neighborhood with unusually high amounts of soot. The 1965 survey
not only determined costs of cleaning, but also attempted to estimate
the willingness-to-pay costs. Table 12 records measured and willing-
ness-to-pay costs provided by 10 respondents. The difference between
respective measured and willingness-to-pay costs is termed the "psychic"
cost, which refers to the loss beyond the direct, measured costs.
These results indicate that the households generally were willing to
pay at least the measured cost of pollution cleanup in order to prevent
the need for cleanup. Usually they were willing to pay more than the
measured cost. The results indicate that the residents were willing to
45
-------
Table 12. COST OF CLEANING UP AFTER BOILER MALFUNCTION
Respondent
1
2
3
4
5
6
7
8
9
10
Measured
costs,
$
9.45
43.75
45.61
19.78
15.67
25.32
13.59
12.95
4.25
8.79
Willingness-
to-pay costs,
$
10.00
50.00
45.61
25.00
25.00
30.00
25.00
20.00
4.25
8.79
Psychic
costs,
$
0.55
6.25
0.00
5.22
9.33
4.68
11.41
7.05
-
-
pay, on the average, 27 percent above the measured cost of cleanup in
order to avoid the occurrence of soiling from pollution.
These results must be qualified on several accounts. First, since
the survey was constructed in only a few days, one must allow for the
possibility of an incomplete questionnaire and an unreliable sample.
Second, too few respondents provided willingness-to-pay costs for them
to be representative of the psychic costs; only 122 residents of the
nearly 3,500 residential units in the population were interviewed.
Lawyer - Morgantown Study'
Another survey on willingness-to-pay was conducted in 1965 by
50
Lawyer among 362 of the 6,424 families in Morgantown, West Virginia.
Table 13 shows the percentage respondents who would pay the indicated
amount each year . . . "if all air pollution were reduced below the
point where it was noticeable (or harmful)."
The average amount respondents were willing to pay was calculated
to be $16.46. The "zero or no response" category is not considered in
the average because it is doubtful that people would be unwilling to
pay anything at all and it is probable that most of the numbers reported
46
COST OF AIR POLLUTION DAMAGE
-------
Table 13. AMOUNT OF MONEY RESPONDENTS WOULD
PAY ANNUALLY TO REDUCE AIR POLLUTION
IN MORGANTOWN, WEST VIRGINIA51
Amount, $
Zero or no response
1 to 5
6 to 10
11 to 15
16 to 20
21 to 25
26 to 30
31 to 35
36 to 40
40
Total
Respondents, %
38.4
23.9
9.4
2.5
4.7
6.3
0.3
0.6
0.3
13.5
99. 9a
Error resulted from rounding of figures.
indicate no response. The mark of the $1 to $5 class is taken to be
$3.50 because it is assumed the class really extends to $5.99. The
mark of each subsequent class is $5 higher than the previous mark. The
highest class is assumed to have a mark of $40.
Williams and Bunyard - St. Louis Opinion Survey
An average payment of $16.46 per year per respondent is much
higher than the willingness-to-pay costs determined in a study by
52
Williams and Bunyard. They reported that 66 percent of those inter-
viewed in a 1963 survey of the St. Louis area were willing to pay $5
per year in higher living costs for clean air, and that 85 percent of
those interviewed would pay $1 per year in higher taxes.
Medalia and Finkner - Clarkston, Washington, Study
One aspect of evaluating aesthetic damages from air pollution con-
sists of examining individual and community perception of pollution in
general and the perception of visibility and odors in particular. In
Effects on Aesthetic Properties
47
-------
an investigation of the effects of smoke and odor from a pulp mill in
53
Clarkston, Washington, Medalia and Finkner interviewed 104 local resi-
dents. Of the persons queried, 91 percent perceived air pollution in the
community as an odor problem, 74 percent perceived it as a visibility
problem, and 62 percent saw it as a problem in nose-throat irritation.
Concern with air pollution was found to be unrelated to the location of
the respondent's residence, probably because of the pervasive nature of
the pollution. Concern with air pollution was found to vary directly
with social status and with factors such as civic pride, desire to elim-
inate the problem, length of residence, and occupational prestige of the
household head.
Williams and Edmisten - Nashville Perception Interviews
54
A broader study by Williams and Edmisten for Nashville, Tennessee,
included an examination of individual perceptions of pollution. People
in 3,032 dwelling units were interviewed to test the hypothesis that
perception and concern for air pollution are directly related to neigh-
borhood pollution levels. The hypothesis was found to be true. Further-
more, the higher the socio-economic status was, the greater the correla-
tion between degree of concern and air pollution levels. They also
found that citizens' perceptions of air pollution were influenced more
by the frequency of high daily levels of pollution than by high monthly,
seasonal, or average levels.
Current Efforts
Several studies are in progress that relate to the aesthetic effects
of air pollution. These deal with either the problems of visibility or
55
odors. A group at Oregon State University is examining social and
economic questions relating to the open burning of agricultural waste,
which generates particulates that create a visibility problem. A series
of interviews among residents in two cities of Oregon's Willamette Valley
will provide some basis for evaluating the social aspects of the visi-
bility problem. A team of economists is focusing on: (1) the estimation
of a pollution production function for the Willamette Valley; (2) the
estimation of the value of the right to burn grass seed fields in the
Willamette Valley; and (3) the economic effects of air pollution (visi-
bility reduction) on Oregon's tourist-related industries.
48 COST OF AIR POLLUTION DAMAGE
-------
In another study, the Copley International Corporation is examin-
ing social and economic factors surrounding the problem of odors. Some
economic assessment of odor perception may result from this effort.
CONCLUSIONS
These studies have examined the factors influencing the perception
of pollution. Once these factors are known, the economic value to the
individual or community of reducing pollution can be estimated. One way
to measure this value is to determine the individual's willingness to pay
to reduce pollution. This has been done in two studies and may be
repeated in two current projects. An alternative method of measuring
the value is to estimate the expenditure required to prevent damages to
aesthetic objects such as paintings, tapestries, and statuary.
Effects on Aesthetic Properties 49
-------
-------
CHAPTER 8.
EFFECTS ON RESIDENTIAL PROPERTY VALUES*
The value of residential property is contingent upon many factors.
One factor believed to influence the value of property is the quality of
air in a neighborhood. All other things considered, the value of
property in areas with more air pollution can be expected to be lower
than the value of property in neighborhoods with less air pollution.
Measures of this relationship, obtained through standard multiple regres-
sion analysis, should yield rough estimates of air pollution damages to
property.
INDIVIDUAL STUDIES
Ridker and Henning - Housing Market Indicators
Ridker and Henning made the first serious use of the housing market
estimator as an index of the effect of air pollution on property
values. Using 1960 census data and pollution readings from the 1963-
1964 interstate air pollution study, they explained over 90 percent of
the variation in the median property values of the St. Louis Standard
Metropolitan Statistical Area (SMSA) census tracts. The variables in
their regressions were as follows:
1. MPV = median property value for census tract, St. Louis,
Missouri, SMSA.
2. SLJL = measure of annual geometric mean sulfation levels for
February 20, 1963, to February 10, 1964, denominated in
units of 0.25 mg SOs/lOO cm2-day.
3. MNR = median number of rooms in tract.
4. PER = percentage of homes recently built.
5. HPM = houses per mile.
6. TIZ = bus travel time to the St. Louis central business district.
7. HWA = accessibility of census tract to highways.
*This chapter was prepared in conjunction with Brian W. Peckham, formerly
with U. S. DHEW, PHS, National Air Pollution Control Administration,
Raleigh, N. C.
51
-------
8. SCHI = school quality within census tract.
9. OCR = occupation ratio (ratio of craftsmen, foremen, opera-
tives, and laborers to total work force).
10. PPU = population density in residential areas.
11. PNW = percentage non-white residents in a census tract.
12. RILL = dummy variable indicating whether census tract is in
Illinois or Missouri, orthogonal to sulfation.
13. RMFI = median family income, orthogonal to MNR, HPM, and OCR.
Assuming that the observed sample was generated by a classical
normal linear regression model, the construction of 95 percent asymptotic
confidence intervals for the sample coefficient of SUL yields an interval
estimate of 73 to 417 for the corresponding population coefficient.
This is to say that there is a 0.95 probability that the slope of the
population regression surface in the MPV-SUL plane lies in the closed
interval [73,417] or that the mean of the marginal conditional probabil-
ity distribution of the dependent variable MPV changes by an amount
within [73,417] for every 0.25 mg S03/100 cm2-day change in SUL. When a
reduction in sulfation of this amount is experienced evenly over a city,
other things being equal, the amount of increase in MPV will vary from
census tract to census tract; it is a random variable with a probability
distribution. The coefficient of SUL estimates the mean of this distri-
bution and thereby serves to estimate also the minimum sum needed to
persuade receptors to endure various levels of whatever pollution a
sulfation index measures.
Crocker and Anderson - Three-City Comparison
At least three other studies have been made that are similar in
method and results to that of Ridker and Henning. Crocker and Anderson
studied the covariation of sulfation, suspended particulates, and census-
tract median property values in St. Louis, Washington, D.C., and Kansas
City.58 They obtained the highest precision by isolating these variables:
1. AMS = arithmetic mean sulfation rate.
2. SPT = annual arithmetic mean suspended particulates.
3. MFI - median family income.
4. DLP = percentage of dilapidated homes.
52 COST OF AIR POLLUTION DAMAGE
-------
5. OLD = percentage of homes more than 20 years old in 1959.
6. DIS = distance to central city.
From their equations, Crocker and Anderson concluded that the pollution
(AMS and SPT combined) elasticity of MPV lies between 0.1 and 0.2. For
every change of 0.1 mg S03/100 cm^-day in sulfation and of 10 mg/m^-day
in suspended particulates over a particular census tract, the best
estimate of the change in that tract's MPV lies in the interval $300 to
$700.
Zerbe - Toronto Study
Zerbe, using similar methods, reported for Toronto "that, other
things being equal, for each increase in 1 mg SQ^/lOO cm^-day property
values will fall by an amount between $800 and $1800 for each single-
family detached dwelling, and that the best estimate is that the values
will fall by about $966."59
Peckham - Delaware Valley Study
Finally, Peckham,of the Division of Economic Effects Research,* used
the extensive sulfation data collected by the Division of Abatementt in
the Delaware Valley to estimate property value differentials. Although
this equation has not been purged of multicollinearity, non-normality
of residuals, and the thousand other shocks that the regression is heir
to, it does have the advantage of utilizing quite accurate sulfation data
gathered from lead dioxide plates from 400 stations during January 1969.
It suggests that the mean of the marginal conditional probability dis-
tribution of MPV probably changes for every 0.5 mg SOs/100 crn2-day by
some $663.
NATIONAL ESTIMATES
From these four studies have come four similar estimates of the
marginal capitalized damage caused to residential property by that pol-
lution that is measured by sulfation. From Ridker's study, it can be
taken that the mean change in MPV per 0.25 mg S03/100 cm2-day change in
sulfation lies within [73,417] with a 0.95 probability, and that the best
linear unbiased estimate of the mean change is $245, or about $100 per
*Now the Division of Ecological Research, National Environmental
Research Center, North Carolina,
the Applied Technology Division, Office of Air and Water Programs, EPA.
Effects on Residential Property Values 53
-------
0.1 tng change. From Crocker and Anderson comes an estimate of $300 to
$700 per 0.1 mg SOs and 10 mg/m3-day change in suspended participates.
Note that because it derives from a curvilinear functional form this
coefficient of damage applies only for marginal changes about existing
pollution levels; it cannot be used to indicate the benefits of cumula-
tive abatement. Both Zerbe's estimate of $966 per 1 mg SOs change and
the $663 per 0.5 mg SOs estimate of Peckham seem close to that of Ridker
and Henning, although somewhat below that of Crocker and Anderson. All
four studies, however, agree that sulfation is inversely related to MPV
and that the magnitude of the marginal capitalized sulfation damage for
residential structures, for a marginal decrease of 0.1 mg S03/100 cm^-
day, probably lies in the range $100 to $300. This remarkable uniformity
of results, for five major metropolitan areas, warrants some confidence
in the worth of the housing market estimator for national pollution
damages.
Given a marginal capitalized damage coefficient and the assumption
that sulfation changes are always evenly distributed among census tracts
(i.e., a 10 percent drop in the annual average sulfation rate for a city
implies a 10 percent drop in the corresponding rates for each tract),
estimates of sulfation damage to entire communities can be calculated by
the following equation: Damage = (SUL coefficient) (number of marginal
changes needed to reduce arithmetic annual mean sulfation rate for the
SMSA to desired background) (number of housing units).
The damage given by this relation is total capitalized pollution
damage, or the decrease in real property wealth caused by whatever pol-
lution is measured by a sulfation index. Total annual damage, or the
decrease in real property income as the result of pollution, is obtained
by multiplying total capitalized damage by a discount rate reflecting the
average return on capital. Clearly the total capitalized pollution
damage depends on the choice of SUL coefficient, desired background sul-
fation rate, arithmetic annual mean sulfation rate of the SMSA, housing
units over which aggregation occurs, and discount rate.
In this case these choices were made as follows:
1. The four studies taken together show that the magnitude of the
marginal capitalized sulfation damage for residential structures,
for 0.1 mg SOs/lOO cm2-day, probably lies in the range of $100
to $300. A middle estimate of $200 is used here.
54 COST OF AIR POLLUTION DAMAGE
-------
2. Selection of the desired background level at 0.1 mg S03/100 cm2-
day was guided by the annual sulfation averages that obtain in
suburban or rural regions (Pensacola, Florida: 0.1 mg S0,/100
cm2-day;61 Columbus, Georgia: 0.0666 mg S03/100 cm -day;61 and
Las Vegas, Nevada: 0.1410 mg S03/100 cm2-day).62
3. Annual arithmetic mean sulfation rates were established for all
SMSA's by either averaging the annual averages for all indi-
vidual monitoring stations (in the case of Washington, D. C.,
Kansas City, and St. Louis ' ); or by averaging the monthly
averages for a single SMSA ' station; or by estimating, in the
case of 164 SMSA's for which sulfation data were lacking, from
a regression of annual SOX emission data (supplied by the Divi-
sion of Air Quality and Emissions Data, BCS, NAPCA)* on sulfa-
tion annual averages. Variations in emissions of SOX were
found to explain more than 40 percent of the variance of sulfa-
tion about its mean.
4. All estimates were calculated for the number of households in
the SMSA as of January 1, 1969.66
5. As a reasonable approximation to an average return on all real
property wealth in the economy, a 0.10 rate was uniformly used
in all calculations.
The charge can reasonably be made that sulfation readings from
single stations in large SMSA's do not give an accurate picture of pol-
lution conditions throughout the entire SMSA. To take account of this
source of bias, annual average sulfation data for the ten dirtiest of
the larger SMSA's (Philadelphia; St. Louis; Detroit; Chicago; Newark;
Cincinnati; Buffalo; Hammond-Gary-East Chicago, Indiana; Boston; and
New York) were adjusted downward in all of the estimates by 28 percent,
which is the average percentage difference between annual average sul-
fation for the SMSA and for six center-city stations in Washington,
St. Louis, Philadelphia, and Kansas City.
Objections can also be raised to the lack of synchronism between
pollution and the other variables in the original regression analyses
of the relationship of sulfation to MPV. Given certain assumptions,
*Now part of Division of Atmospheric Surveillance and of Applied Technol-
ogy Division, EPA.
Effects on Residential Property Values 55
-------
this condition will probably make the ordinary least squares estimator
of the coefficient for SUL or AMS asymptotically biased downward.
As a result of the calculations described above, and of assuming a
marginal capitalized damage coefficient of $200 for each reduction of
n
0.1 mg S03/100 cm-day, the national annual estimate conies to $5.2
billion. This estimate, to repeat, spans all housing units within all
SMSA's and assumes that pollution changes are spread evenly over all
census tracts and that there is a negative linear relationship between
sulfation and MPV. It indicates the approximate amount that residents
of American cities would demand, under emitter liability, to forego
asserting their rights to have pollution abated so that arithmetic
mean sulfation rates in all SMSA's would be 0.1 mg SOs/100 cm2-day or
lower.
56 COST OF AIR POLLUTION DAMAGE
-------
CHAPTER 9. LITIGATION*
By 1969, after many attempts at estimating pollution costs, the
realization came that personal opinion polls often did not yield truth-
ful responses; that surveys of the technical coefficients of household
production functions failed to pick up all the myriad actions taken in
adjusting for pollution loadings; and that property-value studies were
only as good as the data used in them, which were often not very good at
all. The hope was that some new technique could be developed to circum-
vent the difficulties of the traditional estimators; specifically, that
legal scholars could suggest some way of deriving information on air
pollution damages from the decisions of the judicial system in adjudi-
cating conflicts of interest over air resources.
As originally drawn up, a litigation survey conducted by Havighurst
and his staff had two major objectives. First, find and report in
sufficient detail all litigation, at the original or appellate levels,
that might bear on the problem of finding out how much air pollution
damage costs and on the problem of finding,out the extent to which the
people of Philadelphia have returned to the courts for redress of these
injuries. Second, using the information gathered from the study, eval-
uate the properties of judicial data as an estimator of damage functions.
Many hours were spent talking to lawyers, court clerks, state and
local control officials, and anyone who might have knowledge of past or
pending litigation relevant to the study. In all, three useful cases
in Philadelphia were found. Havighurst concluded that citizens of urban
areas are much less inclined to control pollution through private legal
action than are citizens of rural areas. City dwellers apparently
become conditioned to air pollution - over time they adapt to dirty win-
dows and brown skies - and, in a dense industrial city, there is some
difficulty in knowing just what sources are primarily responsible for
the pollution. Because of the paucity of interesting cases in the
*This chapter was prepared in conjunction with Brian W. Peckham, formerly
with U. S. DHEW, PHS, National Air Pollution Control Administration,
Raleigh, N. C.
57
-------
Philadelphia area the project was broadened to include the Berks County-
Bethlehem region. Except for a few cases that turned up, this effort,
too, proved unavailing.
It was obvious that no damage functions, or even many useful data,
would emerge from the records of the few cases located. The most
important thing, then, was to evaluate carefully the feasibility of this
litigation technique as a means of measuring pollution damage.
To proceed on such an appraisal required a careful comparison of
the type of damage information desired by economic analysts and the type
yielded by the courts. Noting that most courts, in practice, make
nuisance awards on the basis of the estimated decline in the market
value of the injured property and on the basis of the court's allowance
for special "discomfort and annoyance," Havighurst concluded that the
economic usefulness of such awards depended on the similarity between
the preferences of the market (as often estimated by real-estate
appraisers in litigation of this type) and the preferences of those
actually injured. The more these preferences coincided, whether result-
ing from the absence of consumer surplus enjoyed by the owner of injured
property or from wide knowledge of the effects of the pollution nuisance,
the stronger the case for disregarding the court's special annoyance
allowance in deriving estimates of economic damage.
The final product of the project was a recommendation that litiga-
tion surveys of this type be continued. Despite the lack of success in
the Philadelphia area, it was thought that a national survey, perhaps
of cases involving odors, would turn up enough damage awards that some
tentative functions might be drawn. Havighurst suggested, however, that
legal records as they now stand are frequently unsatisfactory for this
purpose because of a failure to itemize pollution injuries and to
specify the ambient air quality involved in the nuisance conditions. He
urged, therefore, that the help of state and local control agencies be
enlisted to maintain a reporting network for air pollution litigation
and to provide air quality data pertinent to pending cases.
58 COST OF AIR POLLUTION DAMAGE
-------
CHAPTER 10.
COST OF AIR POLLUTION BY CATEGORIES
Past and current studies have been examined to determine the cost
of air pollution to the people of the United States in terms of addition-
al costs of health care and impairment of human resources, reduction in
residential property values, degradation of materials, and damage to
vegetation and agricultural productivity. A national cost estimate was
not calculated for soiling and aesthetic effects in order to avoid
counting twice those damages affecting property values. Estimates were
not determined for other effects because of insufficient data.
The cost of air pollution for each category of economic loss is
identified with the specific pollutants considered most responsible for
that loss in Table 14 and in the following discussion. An attempt is
also made to distribute the cost of the effects of air pollution among
specific sources (Table 15).
Table 14. NATIONAL COSTS OF POLLUTION DAMAGE,
BY POLLUTANTS, 1968
($ billion)
Effects (loss category)
Residential property
Materials
Health
Vegetation
Total
SOX
2.808
2.202
3.272
0.013
8.295
Part.
2.392
0.691
2.788
0.007
5.878
Oxidant
-
1.127
-
0.060
1.187
NOX
-
0.732
-
0.040
0.772
Total
5.200
4.752
6.060
0.120
16.132
SOURCE EMISSIONS
The Office of Air Programs, EPA, estimated national emissions of
CO
principal pollutants by major source category for 1968. Table 16
presents their estimates. The principal pollutants are carbon monoxide
(CO), hydrocarbons (HC), nitrogen oxides (NOX), particulates (Part.),
and sulfur oxides (SOx)- National emissions of these pollutants amounted
to about 214 million tons in 1968. Forty-two percent of all national
59
-------
CO
UD
cn
1
o
LU
U.
LU
Q
s
LU
O
O
01
CO
UJ ^-~
c ri r-
<: o
<=£ r^
a i
2: -Q
O
_J
o
CL.
LL,
O
01
1
01
0
0
=c
CD
£
*
in
cu
_Q
fO
h-
-D
O
to
!
ro
i
s-
to
ZJ
-o
E
a>
u
s-
o
CO
re
o
h-
CO
3
o
ai
c
ro
^
cu
a
CO
M
CO
S
CO
cu
CO
CO
-M
i_
O
CL
CO
ro
h-
c:
o
CO
Z3
£"§
(O
O
ro
OO
O
O
,
CD
CO
0
OJ
14-
LU
0
O
CM
LO
^>
co
CO
o
0
o
CO
CM
,
LO
LO
O
CM
O
CO
CM
4-)
S-
0)
CL
O
s-
CL
i
CO
c
CU
.,
CO
cu
CM
LO
^
LO
LO
co
o
CO
o
00
o
CO
o
CO
cn
o
LO
CO
p
CO
ro
si
0)
(O
3^
0
10
0
LO
LO
O
0
cn
o
CO
LD
p^
p^
cn
o
CO
CM
CO
c~
P
ro
CU
3:
o
CM
CD
P_
CM
0
0
O
O
O
o
CM
O
O
co
CM
O
O
CD
O
0
'4^3
ro
(U
cn
cu
:>
CM
CO
r
LO
LO.
ID
I.
CM
O
CO
o
^*
CO
LT5
n
^}-
f^^
CO
CO
cn
r-^
rO
O
1
60
COST OF AIR POLLUTION DAMAGE
-------
Table 16. ESTIMATES OF NATIONWIDE EMISSIONS, 1968°
(106 tons/year)
Source category
Transportation
Fuel combustion in
stationary sources
Industrial processes
Sol id waste disposal
Miscellaneous
Total
CO
63.8
1.9
9.7
7.8
16.9
100.1
Part.
1.2
8.9
7.5
1.1
9.6
28.3
SOX
0.8
24.4
7.3
0.1
0.6
33.2
HC
16.6
0.7
4.6
1.6
8.5
32.0
NOX
8.1
10.0
0.2
0.6
1.7
20.6
emissions comes from transportation sources including automobiles, trucks,
buses, trains, and aircraft. Fuel combustion in stationary sources
such as public utility and industrial power plants, commercial boilers,
and residential furnaces accounts for 22 percent of national emissions.
Pollutants from industrial sources other than fuel combustion - namely,
industrial processes - make up 14 percent of national emissions. Dumps
and incinerators and related solid waste disposal practices generate
5 percent of the national emissions. The remaining 17 percent derives
from a variety of sources including forest fires, agricultural burning,
and coal refuse smoldering.
ASSIGNMENT OF COSTS
The national health costs of pollution in 1968 were estimated to be
$6,060 million. Studies by Ridker and by Lave and Seskin related health
effects to particulate and sulfur oxide pollutants. Until better
evidence is forthcoming, it is assumed that the health costs of air
pollution stem from participates and sulfur oxides. Sources of these
two pollutants are shown in Table 15 according to their relative emission
levels. For example, 54 percent, or $3.272 billion of the $6.060 billion
in health losses, is attributed to sulfur oxide pollution inasmuch as
this pollutant amounts to about 54 percent of the national emissions of
sulfur oxides and particulates combined.
The cost of air pollution damage reflected in residential property
values amounted to $5.200 billion in 1968. Onus for this cost is
attached principally to sulfur oxides and particulates, acting individ-
ually and synergistically. The costs of pollution damage to residential
property are allocated in direct proportion to the emissions of these
Cost of Air Pollution by Categories
61
-------
two pollutants. The liability for these costs among sources is assigned
according to each source's contribution of particulate plus sulfur oxide
emissions to the national total of these two pollutants. For example,
24 percent, or $1.248 billion of the $5.200 billion in residential
property losses, is assigned to industrial process sources because they
account for 24 percent of the nation's sulfur oxides and particulate
emissions.
In the case of materials, the Midwest Research Institute cost esti-
mate of $2.766 billion is allocated in proportion to the pollutants emit-
ted except for CO, which is omitted because there is no evidence that CO
is damaging to materials. The corrosion of galvanized steel is attri-
buted solely to SOX- The fading of dyes and effects of pollution on
elastomers are attributed to oxidants and nitrogen oxides. Because the
formation of oxidants is directly related to the emission of hydrocarbons,
the costs of fading and of elastomer deterioration are therefore allocat-
ed to oxidants and nitrogen oxides according to the relative emissions
of hydrocarbons and nitrogen oxides respectively.
The two vegetation studies yielded basically the same results,
which show that oxidants account for about 90 percent of observable
direct crop losses. If this is true, then the portion of the estimated
total loss of $120 million attributable to oxidants is approximately
$100 million. This value is distributed according to relative emissions
of hydrocarbons and nitrogen oxides, inasmuch as these pollutants are
the raw materials from which the plant-damaging oxidants, PAN and 03,
are formed. Work by Waddel1 indicates that SOX accounts for direct
damage assessed at approximately $13 million. The remainder of the
pollution cost is allocated to particulates, which include dusts of
fluoride, lead, and other pollutants that, if deposited on crops, would
make them unmarketable. Carbon monoxide was not considered because
evidence indicates that only extremely high levels - which are not
found in ambient air - will noticeably damage plants.
62 COST OF AIR POLLUTION DAMAGE
-------
CHAPTER 11. DISCUSSION
DAMAGE ESTIMATES
The national total cost of damage from air pollution in the United
States for 1968 is estimated to have been $16.1 billion. The cost was
principally at the expense of residential property, materials, and
health. Transportation, fuel combustion in stationary sources, and
industrial processes were the most significant sources.
The magnitude of this estimate provides some indication of the
possible benefits that could accrue from pollution reduction. Use of
this figure, however, should be tempered by several qualifications. If
optimum resource allocation is to be achieved, pollution control should
proceed to the point at which marginal benefits equal marginal costs of
control. The estimate given in this report of the total annual cost of
pollution is for 1 year, 1968. Based on the methods used to determine
total annual costs of pollution, it is possible to suggest the marginal
benefits of abatement. Total annual costS'Of pollution were based on
assumed and empirically determined proportional relationships between
the cost and level of pollution. Therefore, it may be claimed, for
example, that a 90 percent reduction in pollution would result in an
equivalent reduction in cost. If the relationship between pollution and
its cost is not proportional, then greater care should be exercised in
interpreting and using the results.
The temptation will exist to use the $16.1 billion estimate of the
total cost of pollution as the measure of total benefits to be received
from controlling pollution. Yet, in fact, some of the pollution costs
associated with the miscellaneous source category are not likely to
become benefits instead of liabilities as a result of general pollution
reduction, mainly because emissions from structural and forest fires,
which account for about 50 percent of the miscellaneous pollution and
about 8 percent of all pollution, are not normally controlled under
traditional air quality management programs. If there were complete con-
trol of all pollutants considered here from all sources except structural
and forest fires, and if the cost-of-pollution function is linear, then
the total national benefit would be $14.8 billion.
63
-------
The $16.1 billion estimate can be compared with those developed by
others. A commonly cited figure for the cost of pollution damage is $11
billion for 1959, based on a $60 per capita cost for that year. This
amount was extrapolated from results of the 1913 Mellon Institute Study
on the basis of commodity price index and population. More recently,
Gerhardt estimated the cost of pollution for 1968 to be within a range
of $6.8 to $15.2 billion, with $8.1 billion the probable cost. Ridker72
has suggested a total cost of pollution damage in 1970 of $7.3 to $8.9
billion. The basic procedure of the latter two efforts involved: (1)
the identification of categories of air pollution damage, (2) an estima-
tion of the total value of category regardless of the air pollution
effects, (3) the assumption of an air pollution damage factor, (4) the
application of this damage factor to the total value of the category, and
(5) the summation of the estimates across all damage categories.
The principal difference between these national estimates is the
determination of damage factors. The factors applied for national cost-
of-pollution estimates for this study are thought to have been deter-
mined by more reliable and objective procedures than in the previous
studies.
No comparative analysis of pollutants in terms of their effects has
been made. We do not know, for example, whether a ton of SOX pollution
causes greater or lesser damage to vegetation than a ton of hydrocarbon
emissions. Neither has there been any comprehensive regional analysis
to examine location-related differences in pollutant effects within and
between regions.
INFORMATION GAPS AND DATA LIMITATIONS
The objectives in identifying information deficiencies are to
qualify the $16.1 billion estimate and to isolate those areas in which a
greater research effort might be desirable. It is the authors' opinion
that the estimate of $16.1 billion is realistically conservative and
defensible. Many pollution effects were not costed simply because of
data limitations. Other estimates were not possible because existing
evidence is meager. Conservative figures were used generally throughout
the estimation procedure.
With respect to health, $6.06 billion can be considered a conserva-
tive estimate for several reasons. First, Lave and Seskin did not
64 COST OF AIR POLLUTION DAMAGE
-------
consider the indirect costs of some medical services because these could
not be assigned readily to disease categories. Second, Lave and Seskin
stated that people generally would be willing to pay more than the
expenses of medical care and lost productivity that they incur. Finally,
in all probability, health expenditures have increased proportionally
more than the gross national product has from 1963 to 1968.
It is doubtful that the assumption of a straight-line functional
relationship of mortality and pollution is accurate, but perhaps it is
the best that can be derived at this time. Another problem in estimat-
ing health costs is determining what costs are appropriate measures of
the economic impact of air pollution; e.g., should premature burial
costs be considered as a cost of pollution? Furthermore, the functional
relationship of health morbidity to air pollution levels and selected
socioeconomic parameters might be somewhat spurious in that a more
sophisticated statistical consideration should be given to this problem.
Also, as in most economic assessments of pollution effects, great leaps
are made in assigning the technical coefficients or damage factors. Lave
and Seskin's work takes the first reasonable step in looking at the
economic assessment of air-pollution-related health costs.
The property-value approach, with its sophisticated econometric
handling of data, has provided the soundest basis for estimating pol-
lution costs. Even though the assumption is made that some soiling,
aesthetic, and plant effects are implicit in this approach, grave uncer-
tainty exists as to what effects are actually being measured. One
serious limitation has been the reliability of air quality data and the
assumptions that have to be made in using them. What pollutants are
actually being measured is uncertain. Nevertheless, the $5.2 billion
estimate of losses in property value appears to be quite reasonable,
even though all of its ramifications are not entirely understood.
The large number of variables that must be considered present a
serious obstacle to the isolation of those parameters that are signifi-
cant. Also, the application of different discount rates in the
determination of total costs of pollution could change the relative
cost estimates. The 10 percent rate of interest used in the residential
property value to estimate results is an understatement of costs
relative to health costs, in which a 6 percent interest rate was applied.
Discussion 65
-------
The $4.8 billion estimate of damage to materials represents a con-
servative loss figure despite the use of weak data. It is clear that
little is known about air pollution effects on paints, concrete, and
certain fibers. Although an argument can be made that the property-
value approach implicitly contains some material deterioration costs,
it is assumed here that these are insignificant with respect to the
total impact of air pollution on materials.
Although vegetation losses resulting from air pollution are
believed to be somewhat greater in magnitude than the suggested $0.1
billion, few data could be found to justify such an assumption. This
figure is conservative in that costs associated with ornamental plant-
ings are not included (these are considered part of the property-value
estimate), and the effects that have been measured represent the
observable direct effects only. Much remains to be learned about
subtle, chronic, low-level-pollution growth effects. Also, no attempt
has been made to quantify the effects of air pollution on the nutritional
content of edible crops. Until some of these areas are investigated
further, the vegetation-loss estimate must be used with an understand-
ing of its deficiencies.
Conclusive evidence on the soiling costs attributable to air pollu-
tion is still lacking. As seen in the Booz-Allen study, the participate
pollution levels in Philadelphia have no statistically significant effect
on residential cleaning and maintenance costs. Even if the household
soiling costs are contained in the property-value estimates, there remain
many soiling costs that cannot be totally ignored, as mentioned earlier
in this report. In particular, little is understood of air-pollution-
related costs of building maintenance and of losses associated with the
soiling of artistic statuary and historical monuments.
In the area of aesthetics, little is known about what people would
be willing to pay to preserve or achieve a given cleanliness of air.
Such an approach is relevant in that air pollution perception is
usually linked with visibility or odor problems. Yet, it is still
largely unknown as to what is actually being measured in willingness-to-
pay schemes. The lack of information suggests that virtually nothing
is understood about the "psychic" costs people suffer as a result of the
deterioration of artwork.
66 COST OF AIR POLLUTION DAMAGE
-------
As suggested earlier, no known attempt has been made to investigate
the economic effects of air pollution on animals, domestic and wild;
however, the effects of the chlorinated hydrocarbons, in particular,
pose a threat to the balance of animal and related populations.
Discussion 67
-------
-------
REFERENCES
1. Kneese, Allen V. Air Pollution - General Background and Some
Economic Aspects. In: The Economics of Air Pollution, Wolozin, H.
(ed.). New York, W. W. Norton and Co., Inc., 1966. p. 23-29.
2. Kneese, Allen V. The Economics of Regional Water Quality Manage-
ment. Baltimore, The Johns Hopkins University Press, 1964.
3. The Cost of Clean Air. Report of the Secretary of Health, Educa-
tion, and Welfare to the Congress of the United States. Government
Printing Office. Washington, D. C. First Report. June 1969.
4. The Cost of Clean Air. Report of the Secretary of Health, Educa-
tion, and Welfare to the Congress of the United States. Government
Printing Office. Washington, D. C. Second Report. March 1970.
5. Ridker, Ronald G. Economic Costs of Air Pollution. New York,
Frederick A. Praeger, 1967. p. 12-29.
6. Van Arsdol, M. D., Jr. Social Organization and Air Pollution.
Presented at 60th Annual Meeting, Air Pollution Control Assoc.,
Cleveland, Ohio. June 12-16, 1967.
7. Ridker, 1967. p. 12.
8. Ridker, 1967. p. 30-56.
9. Lave, L. B. and E. P. Seskin. Air Pollution and Human Health.
Science 169_(3947): 723-733, August 21, 1970.
10. Riggan, Wilson B. Cost to the Federal Government of Health Effects
Attributed to Pollution from Motor Vehicles. U.S.EPA, National
Environmental Research Center. Research Triangle Park, N. C.
Unpublished report.
11. Rice, Dorothy P. Estimating the Cost of Illness. U.S. Dept. of
Health, Education, and Welfare. Washington, D. C. Health Economics
Series, No. 6. 1966.
12. Statistical Abstract of the United States: 1968, 89th Ed. U.S.
Dept of Commerce, Bureau of the Census. Washington, D. C. 1968.
p. 312.
13. U.S. Dept. of Commerce. Survey of Current Business. 5Jhl, January
1970.
14. Uhlig, H. H. The Cost of Corrosion in the United States. Corrosion
510) = 29-33, i960.
15. The Rust Index and What it Means. Rust-Oleum Corporation. Evanston,
Illinois. 1964.
69
-------
16. Private communication between Lee Brodsky, Hudson Painting and
Decorating Company, New York, N. Y., and J. J. Schueneman, U.S.
DREW, PHS, National Air Pollution Control Administration, Durham,
N. C. October 23, 1967.
17. Stanford Research Institute. Inquiry into the Economic Effects of
Air Pollution on Electrical Contacts. U.S. DHEW, PHS, National Air
Pollution Control Administration. Final Report, Contract No. PH-
22-68-35. April 1970.
18. Salmon, R. Midwest Research Institute. Systems Analysis of the
Effects of Air Pollution on Materials. U.S. DHEW, PHS, National
Air Pollution Control Administration. Final Report, Contract No.
CPA-22-69-113. January 1970.
19. Haynie, F. H. Estimation of Cost of Air Pollution as the Result of
Corrosion of Galvanized Steel. U.S. EPA, National Environmental
Research Center. Research Triangle Park, N. C. Unpublished
Report.
20. BatteHe-Columbus Laboratories. A Survey and Economic Assessment
of the Effects of Air Pollution on Elastomers. U.S. DHEW, PHS,
National Air Pollution Control Administration. Durham, N. C.
Final Report, Contract No. CPA-22-69-146. June 1970.
21. Private communication between V. S. Salvin, University of North
Carolina, Greensboro, N. C., and Thomas E. Waddell, U.S. DHEW, PHS,
National Air Pollution Control Administration, Durham, N. C.
Subject: Contract No., CPA-22-68-2, A Survey and Economic Assess-
ment of the Effects of Air Pollution on Textile Fibers and Dyes.
May 1970.
22. Salvin, W. S. Textile Pollution Loss is in Billions. Raleigh News
and Observer, March 29, 1970, Section 4, p. 10.
23. Middleton, J. T. and A. 0. Paulus. The Identification and Distribu-
tion of Air Pollution through Plant Response. AMA Arch. Ind.
Health. 14.; 526-532, December 1956.
24. Lacasse, N. L., T. C. Weidensaul, and J. W. Carroll. Statewide
Survey of Air Pollution Damage to Vegetation - 1969. Center for
Air Environment Studies (CAES). State College, Penn. CAES Publica-
tion No. 148-70. January 1970.
25. Stanford Research Institute. Economic Impact of Air Pollutants on
Plants. CRC-APRAC Project No. CAPCA-2-68. Palo Alto, Calif.
Annual Report. August 1970.
26. O'Connor, J. J. The Economic Cost of the Smoke Nuisance to Pitts-
burgh. Mellon Institute. Pittsburgh, Pa. Smoke Investigation
Bulletin No. 4. 1913.
27. Kneese, A. V. How Much is Air Pollution Costing Us in the United
States. Proceedings of the Third National Conference on Air Pollu-
tion, Washington, D.C., December 12-14, 1966. p. 529.
28. Beaver, H. Committee on Air Pollution Report. Her Majesty's
Stationery Office, London. 1954.
70 COST OF AIR POLLUTION DAMAGE
-------
29. Michelson, I. and B. Tourin. Comparative Method for Studying Costs
of Air Pollution. Public Health Reports. 81_(6) :505-511, June 1966.
30. Micnelson, I. and B. Tourin. Report on the Validity of Extension
of Economic Effects of Air Pollution Data from Upper Ohio River
Valley to Washington, D. C., Area. Environmental Health and Safety
Research Assoc., Hartford, Conn. August 1967.
31. Michelson, I. and B. Tourin. The Household Costs of Air Pollution
in Conrecticut. The Conn. State Dept. of Health and Environmental
Health and Safety Research Assoc., Hartford, Conn. October 1968.
32. Jones, A. C. Studies to Determine the Costs of Soiling Due to Air
Pollution: An Evaluation. In: Economics of Air and Water Pollu-
tion. Virginia Polytechnic Institute, Water Resources Center,
Blacksburg, Va. April 1969,
33. Ridker, 1967. p. 73-89.
34. Ridker, 1967. p. 90-110.
35. Ando, Itara. Social Expenditures for Public Hazards and Its [sic]
Apportionment. Economic Planning Agency. Tokyo, Japan. September
1967.
36. Private Communication between Booz-Allen and Hamilton, Inc.,
Washington, D.C., and Thomas E. Waddell, U.S. DHEW, PHS, National
Air Pollution Control Association, Durham, N. C. Subject: Contract
No. CPA-22-69-103, Survey to Determine Residential Soiling Costs
of Particulate Air Pollution.
37. Rubay, M. About the Fog Observed in the Meuse Valley in December
1930 and Its Noxious Effects on Animals. Ann. Med. Veterinaire
77:97-110, March 1932. (Abstract)
38. Firket, J. Comparative Pathology and Air Pollution. Seventh
Lausanne Congr. Intern. Pathol. Com. 7_(2):57-80, 1955. (Abstract)
39. Miessner, H. Damage to Animals Caused by Industry and Technology.
Deut. Tieraerztl. Wochschr. 39_:340-345, 1931. (Abstract)
40. Ferrando, R. and G. Milhad. The Biological Effects of Air Pollu-
tion on Animals. Rev. Med. Soc. J7_(3):295-306, 1969. (Abstract)
41. Shupe, J. L. Levels of Toxicity to Animals Provide Sound Basis for
Fluoride Standards. Environ. Sci. Technol. 3(8):721-726, August
1969.
42. Bohne, H. Industrial Smoke Damage from Fluoride. M. H. Deut.
Landwirtsch. Ges. 77J17) :575-578, 1962. (Abstract)
43. losif, C. Acute and Chronic Lead Poisoning in Cattle. Rec. Med.
Vet. Ecole Alfort. ]42_(2) :95-106, February 1966. (Abstract)
44. Ogura, Y. Molybdenum Poisoning in Cattle Due to Air and Soil
Contamination as an Industrial Hazard. Tokyo National Institute
Animal Health Bulletin. 50:24-29, 1965.
References 71
-------
45. Bischoff, 0. Poisoning of Domestic Animals through Copper and
Arsenic Containing Fly Dust. Deut. Tieraerztl. Wochschr. 47:442-
447, 1939.
46. Stokinger, H. E. Effect of Air Pollutants on Wildlife. Conn. Med.
27(8):487-492, August 1963.
47. Snyder, R. L. and H. L. Ratsliffe. Primary Lung Cancers in Birds
and Mammals of the Philadelphia Zoo. Cancer Res. 26:514-518,
March 1966.
48. Library Battles Decay of Rare Books. New York Times, February 20,
1967, p. 37. ~~
49. Ridker, 1967. p. 90-114.
50. Lawyer, Robert E. An Air Pollution Public Opinion Survey for the
City of Morgantown, West Virginia. Morgantown, Univ. of West
Virginia. Unpublished Master's Thesis. 1966.
51. Lawyer, 1966. p. 24, 41.
52. Williams, J. D. and F. L. Bunyard. Interstate Air Pollution Study:
Opinion Surveys and Air Quality Statistical Relationships. U.S.
DHEW, PHS, National Center for Air Pollution Control. Cincinnati,
Ohio. 1966.
53. Medalia, Nahum Z. and A. L. Finkner. Community Perception of Air
Quality: An Opinion Survey in Clarkston, Washington. U.S. DHEW,
Public Health Service. Cincinnati, Ohio. Publication No. 999-AP-
10. June 1965.
54. Williams, J. D. and Norman 6. Edmisten. An Air Resource Management
Plan for the Nashville Metropolitan Area. U.S. DHEW, Public Health
Service. Cincinnati, Ohio. Publication No. 999-AP-18. September
1965.
55. Private Communication between Robert Alexander, Oregon State Uni-
versity, Corvallis, Oregon, and Thomas E. Waddell, U.S. DHEW, PHS,
.National Air Pollution Control Administration, Durham, N. C. May
1969. Subject: Contract No. CPA-70-109, Social and Economic
Effects of Changes in Air Quality.
56. Private Communication with David Flesh, Copley International Cor-
poration, La Jolla, California. May 1970. Subject: Contract No.
CPA-70-116, Social and Economic Impacts of Odors.
57. Ridker, Ronald G. and John Henning. The Determinants of Residential
Property Values with Special Reference to Air Pollution. Rev. Econ.
and Statistics. 4^:246-257, May 1967.
58. Anderson, R. J., Jr., and T. D. Crocker. Air Pollution and Housing:
Some Findings. Paper No. 264 . Presented at Institute for Research
in the Behavioral, Economic, and Management Sciences, Krannert
Graduate School of Industrial Administration, Purdue University,
Lafayette, Indiana, December 1969.
72
COST OF AIR POLLUTION DAMAGE
-------
59. Zerbe, R. 0., Jr. The Economics of Air Pollution: A Cost-Benefit
Approach. Report to the Ontario Department of Public Health.
Toronto, Ontario, Canada. 1969. p. 54.
60. Huey, N., M. Wallar, and C. Robson. Field Evaluation of an Improved
Sulfation Measurement System. Paper No. 69-133. Presented at 62nd
Annual Meeting, Air Poll. Control Assoc., New York, N. Y., June 1969.
61. Interstate Effects Surveillance Project. U. S. DHEW, PHS, National
Air Po'lution Control Administration, Division of Abatement. Durham,
N. C. Unpublished Report. 1968.
62. Air Quality Data from NASN and Contributing State and Local Networks,
1966. U.S. DHEW, PHS, National Air Pollution Control Administration.
Durham, N. C. Publication No. APTD-68-9. 1968. p. 155.
63. Washington, D. C., Metropolitan Area Air Pollution Abatement
Activity. U.S. DHEW, PHS, National Center for Air Pollution Control.
Cincinnati, Ohio. November 1967. p. 37.
64. Interstate Air Pollution Study: Phase II, Air Quality Measurements.
U.S. DHEW, PHS, National Center for Air Pollution Control. Cincin-
nati, Ohio. 1966.
65. Kansas City, Kansas-Missouri, Air Pollution Abatement Activity:
Phase II, Pre-Conference Investigations. U.S. DHEW, PHS, National
Center for Air Pollution Control. Cincinnati, Ohio. 1967.
66. Commercial Atlas and Marketing Guide, 100th Ed. Chicago, Rand
McNally, 1969.
67. Havighurst, C. C. A Survey of Air Pollution Litigation in the
Philadelphia Area. U.S. DHEW, PHS, National Air Pollution Control
Administration. Durham, N. C. Final Report, Contract No. CPA-
22-69-112. December 1969.
68. Nationwide Inventory of Air Pollutant Emissions, 1968. U.S. DHEW,
PHS, EHS, National Air Pollution Control Administration. Raleigh,
N. C. Publication No. AP-73. August 1970.
69. Waddell, Thomas E. Economic Effects of Sulfur Oxide Air Pollution
from Point Sources on Vegetation and Environment. U.S. DHEW, PHS,
EHS, National Air Pollution Control Administration. Raleigh, N.C.
Unpublished Report. June 1970.
70. Air Quality Criteria for Carbon Monoxide. U.S. DHEW, PHS, EHS,
National Air Pollution Control Administration, Washington, D.C.
Publication No. AP-62. March 1970. p. 7-2.
71. Gerhardt, Paul H. An Approach to the Estimation of Economic Losses
Due to Air Pollution. U.S. DHEW, PHS, CPEHS, National Air Pollution
Control Administration. Raleigh, N. C. Unpublished Report.
June 1969.
72. Ridker, Ronald G. The Problem of Estimating Total Costs of Air
Pollution: Discussion and Illustration. Report to U.S. DHEW,
PHS, National Center for Air Pollution Control. Cincinnati, Ohio.
July 1966. p. 21.
References 73
&U.S. G.P.O.: 1973746-757-3807 Region No. 4
-------
-------
-------
------- |