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on reported studies, as well  as a  review of  unpublished and ongoing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^CMr—i — 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 x—s +-> > — (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.:  1973—746-757-3807 Region No. 4

-------

-------

-------

-------