PB81-218968
Socioeconomic Analysis  of Hazardous
Waste Management Alternatives
Methodology and Demonstration
Denver Research  Inst.,  CO
Prepared  for

Municipal  Environmental  Research Lab
Cincinnati, OH
Jul 81
                     U.S. DEPARTMENT OF COMMERCE
                  National Technical Information Service
                                                 •

-------
                                           EPA-600/5-81-001
                                           July 1981

                                              PB81-218968
SOCIOECONOMIC ANALYSIS OF HAZARDOUS WASTE  MANAGEMENT

    ALTERNATIVES:   METHODOLOGY AND DEMONSTRATION
                         by
                  Graham C.  Taylor
           Industrial  Economics  Division
      University of Denver Research  Institute
              Denver,  Colorado  80208
                 Grant No.  R804661
                  Project Officer
                 Oscar W.  Albrecht
    Solid and Hazardous Waste Research  Division
    Municipal Environmental  Research  Laboratory
               Cincinnati, Ohio  45268
    MUNICIPAL ENVIRONMENTAL RESEARCH  LABORATORY
         OFFICE OF RESEARCH AND DEVELOPMENT
        U.S.  ENVIRONMENTAL PROTECTION AGENCY
                CINCINNATI. OHIO  45268

-------
                 NOTICE





TEIS DOCUMENT  HAS BEEN REPRODUCED



FROM TEE BEST  COPY FURNISEED US  BY



TEE  SPONSORING AGENCY.  ALTHOUGH  IT



IS RECOGNIZED TEAT  CERTAIN PORTIONS



ARE  ILLEGIBLE,  IT IS BEING  RELEASED



IN TEE  INTEREST 0?  MAKING  AVAILABLE



AS MUCH  INFORMATION AS POSSIBLE.

-------
                                   TECHNICAL REPORT DATA
1. REPORT NO.
 EPA-600/S-81-OQ1
                                                           3. RECIPIENT'S ACCESSION NO.
ORD Report
218966
4. TITLE AND SUBTITLE
 Socioeconomic  Analysis Of Hazardous Waste Management
 Alternatives:  Methodology And Demonstration
                         6 REPORT DATE
                          July 1981
                         6. PERFORMING ORGANIZATION CODE
17. AUTHOR(S)
 Graham C. Taylor
                                                           8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Industrial Economics Division
 University of Denver Research Institute
 Denver, Colorado   80208
                         10. PROGRAM ELEMENT-NO.

                            73D1C	
                         II. CONTRACT/GRANT NO.

                            Grant « R804661
12. SPONSORING AGENCY NAME AND ADDRESS
 Municipal Environmental  Research Laboratory—Cin., OH
 Office of Research and Development
 U.  S. Environmental  Protection Agency
 Cincinnati, Ohio  45268
                         13 TYPE OF REPORT AND PERIOD COVERED
                           Final	
                         14. SPONSORING AGENCY CODE

                            EPA/600/14
15 SUPPLEMENTARY NOTES

 Project Officer: Oscar W.  Albrecht
          S13/684-4319
16 ABSTRACT
           A methodology for  analyzing economic and social  effects  of alternatives in
 hazardous waste management is  presented and demonstrated.  The  approach includes the
 use of environmental threat  scenarios and evaluation of effects on and responses
 by parties-at-interest.  The methodology is demonstrated by  a case study of alterna-
 tive approaches to hazardous waste management for Oregon.
      The report organizes, summarizes, and provides extensive information on hazardous
 waste management techniques.   Research on risk preferences and  its relationship to
 decision-making is reviewed, and appendices on methods for valuing various effects
 (such as environmental impacts).public attitudes toward environmental  issues, and
 intergenerational discounting  are included.  The report constitutes a  useful manual
 for hazardous waste management personnel and policymakers  involved in environmental
 quality management.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                                       c. COSATI Field/Gioup
 G DISTRIBUTION STATEMENT

    Not restricted
            19 SECURITY CLASS (This Report I
               Unclassified
     21 NO. OF PAGES
      275
                                              20 SECURITY CLASS (Thispage)
                                                 Unclassified
                                                                        22 PRICE
EPA Font 2220-1 (R.v. 4-77)

-------
                                  DISCLAIMER


     This  report  has been  reviewed by  the  Municipal  Environmental Research
Laboratory, U.S. Environmental  Protection  Agency, and  approved for publica-
tion.  Approval  does not  signify that  the  contents  necessarily reflect the
views  and policies  of  the  U.S.  Environmental  Protection Agency,  nor does
mention  of  trade   names or  commercial  products  constitute  endorsement  or
recommendation for use.
                                      ii

-------
                                   FOREWORD
     The  U.S. Environmental   Protection  Agency  was  created   because  of
increasing public  and government concern  about the dangers  of  pollution to
the health and welfare  of the American people.   Noxious air, foul water, and
spoiled  land  are  tragic testimonies  to  the  deterioration  of   otr  natural
environment.   The complexity of the environment and the interplay between its
components  require  a concentrated  and  integrated  attack  on the  problem.

     Research and  development is the  first necessary  step, in problem -solu-
tion; it  involves  defining  the problem, measuring its  impact, and searching
for solutions.  The  Municipal  Environmental  Research Laboratory  develops new'
and improved  technology  and systems  to prevent, treat, and manage wastewater
and the  solid and hazardous  waste pollutant  discharges  from municipal  and
community sources, to preserve and  treat public drinking-water supplies, and
to minimize  the  adverse  economic,  social, health  and aesthetic  effects of
pollution.  This  publication  is one  of the products  of that  research and
provides  a  vital  communications link  between the  researcher and the  user
community.

     The report describes and demonstrates a methodology for the analysis of
the complex  socioeconomic problems associated  with hazardous waste  manage-
ment.   It should  be  of  wide value to both researchers and decisionmakers who
are concerned with hazardous wastes.
                                      Francis T.  Mayo,-DIRECTOR
                                      Municipal Environmental Research
                                        Laboratory
                                    iii

-------
                                  ABSTRACT
     This  report  develops' and  presents  a  methodology  for  analyzing  the
economic  and social  effects  of  alternative  approaches to  hazardous waste
management.  The techniques of economic analysis that have been developed for
conventional pollutants are not  always appropriate or feasible for hazardous
wastes, and thus a new approach is desirable.  The approach proposed involves
(1) the generation  of a series of environmental  threat  scenarios that might
arise  from  the  use  of various hazardous waste management techniques, and (2)
identification of parties-at-interest  to these techniques.  By examining how
the parties-at-interest are affected by  alternative  approaches to hazardous
waste  management,  it is  possible  to make decisions  based  on economics that
recognize sociological factors.

     This  approach  is  applied in  a generalized manner  to  tne various tech-
niques  that can be  used  to manage hazardous wastes, and  the methodology is
demonstrated  in two  management  decision situations.   One  example  analyzes
alternative  techniques  that could  be  applied to a single  waste  stream,  the
second  is  a case study of alternative  approaches  to  hazardous waste -nanage-
ment  for  Oregon.    These  cases  demonstrate  that  though   the  methodology
simplifies  the   decisionmaker's  task,  the ultimate  decision  depends  on  the
degree of risk aversion favored and may involve subjective elements.

     The  report  organizes, summarizes, and  provides  references to extensive
data on various hazardous  waste management  techniques  and their associated
risks.   It  includes  appendices  on the methods  that may  be used to value the
various  effects (such  as environmental  impacts)  of waste  management tech-
niques, on public attitudes  toward environmental  issues  and  on  methods of
handling  effects that  extend  over a  long  period of time.   This includes a
pragmatic solution to the problem of intergenerational discounting.  Research
on risk-taking  and  its  relationship to decisionmaking is also  reviewed.  The
report  should constitute  a useful  handbook for hazardous  waste management
personnel  and  for many  others  who  are involved in  environmental  decision-
making.

     This work .was supported in.part by Grant No. R804661 awarded by the U.S.
Environmental  Protection  Agency  to the  Colorado  School  of Mines,  Golden,
Colorado.   This report  was  submitted  in  fulfillment   of  this  grant which
extended  from  November  1976 to  April  1979.   Work was  completed   as  of
September 1979.
                                      iv

-------
                                   CONTENTS
Foreword	    111
Abstract	     iv
Figures	   viii
Tables	     ix
Acknowledgements 	     xi


Section 1.   Introduction .  .  .•	,	    -  1

     Research Objectives	'.  •	      1
     Scope	      2
     Organization of the Report	      3

Section 2.   Conclusions and Recommendations	      4

     Summary of Findings ...'	•	      4
     Applications of the Methodology 	      6
     Recommendations for Further Research	      6

Section 3.   Overview of Hazardous Waste Management 	      8

     Unique Aspects of Hazardous Waste Management	      8
     The Nature of Hazardous  Wastes	     10
     Techniques for the Management of Hazardous Wastes 	     18
     Legislative Background of Hazardous  Waste Regulation
       in the United States	     18
     Previous Research on Hazardous Waste Management 	     22

Section 4.   Economic and Social  Aspects of
              Hazardous Waste Management		     27

     Cost-Benefit and Risk-Benefit Analysis for
       Environmental Problems	     27
     Threats That May Arise From Hazardous Wastes	     31
     Economic and Social Effects of Hazardous
       Waste Management Techniques 	     33
     The Parties-at-interest  in Hazardous Waste Management 	     38
     The Socioeconomic Interaction Process 	     43

-------
Section 5.   General Methodology for Analysis of Hazardous
              Waste Management Alternatives	     52

     Introduction	     52
     Theoretical Considerations	     53
     Use of Threat Scenarios	     58
     Prerequisite Information for Analysis 	     51
     The Analytical Framework	     67
     Aids to Decisiormaking	     74

Section 6.   Application of the Methodology to a
              Single Waste Stream	     78

     Introduction	     78
     The Problem	     73
     Threat Scenarios	     79
     Analysis of the Alternatives	     80
     Oecisionmaking	     82

Section 7.   Case Study of Hazardous Waste Management in Oregon ....     36

     Introduction	     86
     Prerequisite Information and Decisions	     90
     Application of Analytical Framework 	     98
     Decisionmaking	    121
     Appendix:Development of Certain Threat Scenarios	    128

References	    134

Appendix A.  Hazardous Waste Management Techniques 	    148

     Techniques Involving Waste Stream Changes 	    148
     Resource Recovery 	    149
     Waste Treatment	    149
     Storage and Disposal Techniques 	    152
     References	    159

Appendix B.  Environmental Threats Associated with
               Hazardous Waste Management Techniques 	    164

     The Nature of Hazardous Waste Threats 	    164
     Probabilities That Threats Will Occur . . .  	    179
     References.	    185

Appendix C.  Valuation of the Effects of Hazardous
               Waste Management Techniques 	    190

     Control Costs 	    190
     Valuation of Environmental Impacts	    194
     References	    208
                                     vi

-------
Appendix D.   Public Attitudes Toward.Environmental Issues	    214

     Introduction	    214
     Public Concern for the Environment	    215
     Attitudes and Behavior of Specified Parties 	    222
     The Economics of Pollution Abatement	    230
     References	    233

Appendix E.   Factors Influencing Discount Rates for
               Environmental Projects	    240

     Introduction	    240
     The Social Discount Rate	    240
     Adjustment of the Discount Rate for Risk	    2<*1
     Intergenerational Effects 	    242
     Objections to Discounting	"	    243
     Application to Hazardous Waste Management 	    244
     References	    247

Appendix F.   Risk and Decisionmaking 	    249

     Aspects of Risk	    249
     Decisionmaking Under Uncertainty, and Risk Aversion 	    251
     Risk and Society	    253
     References	    258
                                     vn

-------
                                   FIGURES
Number                                                                  Page
 1       Routes by which hazardous wastes can affect the environment .     11
 2       The Battelle hazardous waste decision model 	     15
 3       Conventional economic approach to pollution control 	     29
 4       Morphological map of environmental threats	     34
 5       Interaction model for hazardous waste management	     44
 6       Model for pollution toleration	     50
 7       Process for assessing damages associated with
           hazardous waste management	     59
 8       The geography of Oregon	     87
 9       Approaches for hazardous waste management in Oregon 	     99
B-l      Process for estimating damages caused by spills
           (sulfuric acid) into water	    169
B-2      Contamination of groundwater by waste disposal practices. .  .    175
B-3      Frequency of property damage resulting from natural
           and man-caused events 	    181
B-4      Frequency of fatalities due to man-caused events	    182
B-5      Frequency of fatalities due'to natural events	    182
F-l      Utility of wealth for a risk-averse individual	    251
F-2      Determinants of acceptable risks as indicated by
           revealed and expressed preferences	    257
                                    viii

-------
                                   TABLES
Number                                                                  Page
 1       Hazardous Materials Classification Criteria 	     13
 2       Techniques for the Management of Hazardous Wastes 	     19
 3       Potentially Hazardous Wastes Generated in the United States .     24
 4       Environmental  Threats in Hazardous Waste Management 	     32
 5       Major Threats  That May Occur in Hazardous Waste Management.  .     35
 6       Environmental  Threats, Costs, and Parties-at-interest for
           Hazardous Waste Management Techniques  	     41
 7       Matrix of Effects on the Parties-at-interest	     71
 8       Summary of the Methodology	     77
 9       Analysis of Five Alternative Waste Disposal Plans 	     81
10       Comparison of  Three Selected Waste Disposal Plans 	     33
11       Employment by  Industry in Oregon and the United States.  ...     88
12       Generation and Disposal of Potentially Hazardous Wastes
           in Oregon	     91
13       Oregon Firms That May Generate Hazardous Wastes 	     94
14       Projected Data on Future Hazardous Waste Generation in
           Oregon Under the Status-quo Approach	    101
15       Threats Associated With Three Approaches to Hazardous
           Waste Management	    108
16       Descriptions of Principal Threats 	    109
17       Summary of Approaches 	    Ill
18       Generator's Costs for Alternative Approaches	    112
                                     ix

-------
19       Impacts of Different Hazardous Waste Management
           Approaches on the Parties-at-interest 	    115
20       Comparison of Five Hazardous Waste Management Approaches.  .  .    119
21       Comparison of Two Hazardous Waste Management Approaches .  .  .    125
22       Theoretical Average Concentrations of Ions From
           Heavy Metals Wastes at Portland, and Relevant Standards .  .    129
A-l      Currently Available Hazardous Waste Treatment and
           Disposal Processes	    150
B-l      Selected Transport Accident Statistics	    156
B-2      Summary of Pollution-caused Fish Kills for the United
           Sates, 1960-1975	    170
B-3      Extent of Fish Kill Damage	    171
B-4      Analysis of 1970 Fish Kill Data	    171
B-5      Mechanisms Involved in Damage Incidents by Disposal Metnod.  .    173
B-6      Summary of Data on 42 Municipal and 18 Industrial
           Landfill Contamination Cases	    177
B-7      Sample Components of Release Sequence Probabilities
           for Geologic Disposal 	    184
C-l      Estimated Cots of Some Disposal Techniques	    191

-------
                               ACKNOWLEDGMENTS
     The author  gratefully acknowledges the encouragement  and  assistance of
Oscar Albrecht,  U.S.  Environmental  Protection  Agency project  officer.   His
support during  the research program  and  his interest in the results  of the
study were particularly gratifying.

     It is  not  feasible  to list all  of the  persons who provided -nformation.
or assistance  for this  study,  but the author wishes to  thank  the following
persons who made  important contributions'.•  -Or:  Roger-'Bo 1 ton,-  professor of-
economics,  Williams  College,  Massachusetts, was  involved in  this  research
from its inception, provided extensive advice,  and reviewed most of the draft
material.    Dr.  Fred  Bromfeld, supervisor,  Hazardous Waste 'Secfon,  Oregon
Department  of  Environmental Quality,  gave  considerable  assistance  with the
case study data, reviewed Section 7, and helped me obtain an understanding of
the practicalities of hazardous waste management.   Dr. Robert Anderson of the
Environmental Law  Institute, Washington,  D.C.,  reviewed Sections 4 and 5 and
the material  in Appendices C  and  E.  -Dr.  Paul Slovic of Decision Research,
Eugene, Oregon,  discussed •his  research-and reviewed Appendix  F.   Dr.  Stan
Albrecht,   associate  professor  of  sociology,   Brigham  Young  University,
reviewed Appendix D.

     Members of the  staff of both the Office of  Research Services,  Colorado
School of Mines,  and  the Industrial Economics  Division,  University of Denver-
Research   Institute,   provided   valuable    support   services.     Ms.   Nanci
Avitable functioned  efficiently and cheerfully as my research  assistant and
coauthored  Appendix  0,  and  Ms.  Jeanne Vannoy helped edit  the  final  report.
                                     xi

-------
                                   SECTION 1

                                 INTRODUCTION
     For years, the Love Canal in upstate New York had been used as a dumping
ground  for  chemical  wastes.   This  had  been  done legally,  accorcing to the
practices of   the  1940's  and  1950's.   Recently,  serious  pollution problems
became  apparent when  trees  died, pools-of bubbling  liquids  appealed in low-
lying backyards, and fumes permeated basements:   Evidence of miscarriages and
birth  defects  was  reported,' the State  -Health  Commissioner  urged immediate
evacuation of pregnant women and young children, and homes in the area became
virtually unsaleable  (Anon.,  1978a).    The  problem  attracted.international
attention when it became the -subject of  an article  in .The- Economist after
President Carter  declared  the  Love  Canal  a disaster  area in  August' 1978
(Anon., 1978C).

     The incident described  above is not an isolated one.  Although estimates
vary widely, one  authoritative source (Anon., 1978a)  reports that more than
1,000  disposal  sites  comparable  to  the Love Canal  may  exist  in the United
States.  The  management of  hazardous  wastes has been of growing concern to
environmental   agencies  for  some time.   In  December 1977,  the  Council  on
Environmental  Quality stated:

          The  problem  of  hazardous waste  has grown  to  serious propor-
     tions  in  recent  years  for  several reasons:   as a  nation,  we are-
     increasing our  consumption  of  all materials,  including hazardous
     materials; several toxic substances have been banned from use, and
     existing  stocks are  "thrown away";.and as air  and  water pollution
     controls  increase, hazardous waste residues result.
     (Council  on Environmental Quality, 1977:45)

     Incidents  such  as that of   the  Love  Canal  and  the Keypore  tragedy at
Hopewell, Virginia,  in 1976 (Council  on  Environmental  Quality,  1977) are,
however, likely, to  intensify the pressure for public involvement in environ-
mental  management  decisions,'1  and   for  careful  consideration  of- public
attitudes in   these  decisions.   Indeed,- when  discussing the  nuclear power
issue,  a  recent engineering  journal  editorial  stated that  "The  need to pay
constant  attention  to  public attitudes may well  become  a faature  of all
engineers'  lives."  (Anon.,  1978b:333)

RESEARCH OBJECTIVES

     The  objective  of  the  research  was  to   develop  and  demonstrate  a
methodology for the analysis of  hazardous waste  management  problems  that is

-------
based on economics but is cognizant of sociological factors.   The .nethodology
was  expected  to  take the  form  of  an  analytical framework  that  could  be
adapted by  decisionmakers to  suit a particular  problem or  situation.   The
research was expected  to throw some light on the attitudes of the public and
special interest  groups  toward hazardous waste management  alternatives, and
toward  taking  environmental  risks.   It  was also expected to  provide  some
generalized analysis  of  the  costs  and  risks  associated  with  the  various
techniques (i.e.,  technical options) for hazardous waste management.

     This report  is timely because  it  provides  environmental  decisionmakers
with  a methodology  for  the  analysis  of hazardous waste problems  that can
respond to  societal  concerns and  reflect public  attitudes.   The methodology
is designed to  permit the identification of costs and other effects associ-
ated   with   alternative   approaches  to  hazardous   waste   management.    It
encourages  a  decisionmaker to  consider attitudes and  equity  as na assesses
trade-offs among  alternatives.   As exemplified  by incidents such as tne Love
Canal disaster, the effects that may arise from some hazardous waste disposal
practices  are  often  ill-defined  or  virtually   unknown.   The  methodology
encourages a decisionmaker  to identify  possible environmental threats and to
evaluate the costs of different degrees  of risk aversion.

SCOPE

     This report  deals  only  with wastes and their management.   It  does not
address the  broader question  of curtailing the  economic  activity that gen-
erates  such waste.    The  research  was oriented  toward  industrial  process
wastes as opposed to  special wastes such as hospital  wastes, pesticide con-
tainers, U.S.  Department of Defense wastes,  sewage sludge, and mine tailings.
(The  latter two wastes are not normally considered hazardous.)  The research
specifically excludes radioactive wastes,  although research  on radioactive
waste  disposal  was reviewed for  its applicability to nonradioactive wastes.
Note,  however, that   the  evaluation  methodology  could be  applied  to any
category of waste.

     The  methodology  is   intended   to  be  used  to  evaluate  alternative
approaches to hazardous waste management on a local or  regional basis, but it
could  also  be  used to evaluate  specific  problems  on  a national basis (e.g.,
the  disposal of polychlorinated biphenyls [PCB's]).  The role of the method-
ology  is that of an aid  to decisionmaking and value judgments are required to
choose between alternatives.

     The  potential  utility of this  report goes  far  beyond the applications
mentioned above.   The methodology itself could be -adapted to decisionmaking
for  a wide  range of  environmental  problems.   The compilations of  data in
Appendices A and B should be useful to those concerned with the management of
hazardous  wastes  at  many  levels, and  Appendices C  through  F  should  be of
value  to a variety of persons  concerned with environmental issues.

     One  important caveat  is  in order.  Much of  the  work described herein,
and  especially that relating to the case study in  Section 7, was performed in
1977.   Since  then  the  U.S.  Environmental  Protection  Agency  (EPA)  has
developed  hazardous  waste  regulations  under the  1976 Resource Conservation

-------
and Recovery Act  CRCRA).   Some of the  details .in.thjs report do ridt reflect
the impact of  these regulations, dnd some of the waste management techniques
discussed may  not be  permissible  under the new law.   However,  the  scope of
this study is  not restricted to situations or substances regulated under the
RCRA.    It is  intended to  assist  hazardous waste  management decisionmaking
under a variety of circumstances,  including those  that might not be covered
by regulations (e.g.,  wastes that are not defined as hazardous under present
Federal regulations).   The methodology  could  also  be  applied to evaluating
possible alternatives to existing regulations.

ORGANIZATION OF THE REPORT

     The  report  has two  main  aspects.   First, Sections  1  through  5 analyze
the special  features of  hazardous  waste  management,  describe the decision-
making  methodology  and  discuss  applications.   Second, 'Sections   6  and  7
demonstrate  the   methodology   in   two   different  situations.   Appendix _A_
describes hazardous waste management techniques, and Appendix°B discusses the'
environmental threats that may'be1 associated'vith-those techniques.   Appendix
C describes  the methods that may be used to value environmental effects, and
Appendix 0 reviews surveys of attitudes towards the environment.  Appendix E
discusses the  discounting of environmental effects, and  Appendix F  presents-
material on the role of- risk1 in -decisionmaking.

-------
                                  SECTION 2

                       CONCLUSIONS AND RECOMMENDATIONS
SUMMARY OF FINDINGS
     The management  of hazardous wastes  has certain  features  that  frcm tne
economist's viewpoint  differentiate it  from that of  other wastes  or pollu-
tants.    A  basic characteristic  of  a  hazardous  waste is  that it posas far
stronger ' threats  to   man  or  the  environment  than   a  nonhazardous  waste.
Because  of  the strength of  these threats,  waste management  techniques that
may be  acceptable  for  nonhazardous  wastes, such  as   using the assimilative
properties of  the  environment,  are  not suitable  for hazardous  wastes, and
techniques that are  intended  to  minimize the exposure of these wastes to the
environment must-generally be used.   Consequently, when  analyzing  potential
damages  from  hazardous  wastes,   the  economist  or  decisionmaker is  largely
concerned with threats  or  risks  (e.g., from the  failure  of waste management
techniques) rather than with predictable environmental impact's.

     Many 'hazardous  wastes are  nondegradable  or persistent  in the  environ-
ment.    Environmental   effects  may  thus  be irreversible,  and  it  could  be
necessary to  consider management techniques that provide  for the  perpetual
care of these wastes.

     Some  hazardous  wastes  are  biologically  magnified  or  have  cumulative
effects  on  organisms.   Waste  stream compositions are  subject to substantial
variation, and when  the wastes contain multiple components, antagonistic and
synergistic effects  can occur.   Although  most  of these  characteristics may
also be  found in  nonhazardous wastes, they are  particularly significant in
hazardous waste management analysis,  as they make it  difficult to  determine
the precise nature of the threats that are posed by hazardous wastes.

     Because of  the  special  characteristics of hazardous wastes, traditional
approaches-to  the  economic analysis of pollution cqntro'l  will o.ften be  inap-
propriate,  and  comprehensive  cost-benefit  or • risk-benefit  studies  may  be
neither  feasible nor warranted for many, hazardous .waste  problems.   Instead,
the author proposes a methodology for the analysis of  hazardous waste manage-
ment alternatives  that is  comparatively simple to apply  and  that has modest
data requirements.  At the same time, the methodology encourages a decision-
maker to examine the  sociological aspects of a situation and to evaluate the
cost and effects of  varying degrees of risk aversion.  Since the methodology
builds  on  a cost-benefit  foundation,  it could also be used  to supplement a
cost-benefit study to  take into account those effects that are difficult to
quantify.

-------
     Determining  control  costs  for hazardous  waste management  presents  no
special  problems;  the major analytical difficulty  lies  in  the uncertainties
associated  with damage  functions.   Conventional  analysis  of environmental
damages  starts  by determining  pollutant  emissions, evaluates  exposures and
consequent effects on organisms, and then attempts to place a dollar value on
these  effects.   Instead,  a  central  feature of the  methodology presented in
this report is the use of environmental threat scenarios.  Each scenario is a
hypothetical chain of events which  results from  some  initiating event (such
as  an  accident or exceptional  weather conditions)  and leads  to an adverse
environmental impact  (such  as contaminated drinking water).   Thes.e scenarios
could  be  derived from  modeling studies, but  they  can  also  be  based  on
previous  experience,   public fears,  or  assumptions of  the worst  possible
consequences.   Some of the  effects  of these threat  scenarios  may readily be
valued using well  established techniques, but others may prove difficult to
translate  into  dollars.   However,  the mere description'of  plausible threat
scenarios  is  valuable because it helps to  identify the parties-at-interest.
which are  groups,  firms,  etc.,  that are  affected  in a common manner oy some
hazardous waste management alternative.

     Identification  of parties-at-interest  is another key  feature  of  the
methodology, as it helps  a decisionmaker to recognize differing attitudes and
viewpoints on hazardous waste management.   It also encourages him to consider
equity,  since   it  highlights the distribution  of favorable and unfavorable
effects.   The methodology uses a simple conceptual model of the 'socioeconomic
interaction process that  focuses  on the effects that hazardous waste manage-
ment techniques will  have on the parties-at-interest and their responses to
these  effects.   These responses  will  in  turn influence the set of outcomes
that result  from a  specified approach to  hazardous waste  management.   The
report includes some  broad  indications of the likely  attitudes and behavior
of  the  parties-at-interest, derived from  hazardous  waste  management practi-
tioners  and  the literature.  A decisionmaker should  be able  to supplement
these  data  with   his perceptions  of  any specific  situations.  The local
viewpoint  may  be  important, because responses of  the  parties-at-interest to
environmental  threats will  depend  on  their  perceptions  of  those  threats,
irrespective of the true  probabilities and magnitudes.

     Because hazardous waste management decisions involve value judgments, a
decisionmaker  must   usually make   the  final  choice  among  alternatives,
examining  them against  the  agency's  objectives   and  deciding  on preferred
trade-offs.  However,  some  alternatives  can  be  eliminated  because  they are
dominated  by others,  such  as  where  both quantifiable and nonquantifiable
costs are higher for one  alternative than another, and the nature and distri-
bution of  the  costs  is similar for  both.   Ultimately,  the critical aspect of
decisionmaking  will  usually be to  decide on  an  appropriate  degree of risk
aversion.   Research  on risk  evaluation,  i.e., determination  of the accept-
ability of  risks  to  society, has provided some useful  background information
on  the  public's   perceptions  of  risk.   However,   this  research cannot  at
present  provide the  specific  guidance  that a decisionmaker  would  need  to
choose  between  hazardous waste  management  alternatives.   Deciding on  an
appropriate  degree  of risk aversion  remains  his  most difficult  problem.

-------
 APPLICATIONS OF THE METHODOLOGY

     The  research described  here  is believed  to  meet the needs of  analysts
 and  decisionmakers for  a simple methodology  that can  analyze a variety  of
 hazardous  waste   management  problems.   The   methodology   is   adaptable   to
 specific  situations,  is  firmly  based  on  economic principles,  and recognizes
 the  sociological   factors involved.   When  necessary,  it  can   be  used with
 comparatively  limited information,  but  it can exploit.more  sophisticated data
 when  these are available.  Ultimately, however,  it requires a  decisionmaker
 to choose  between  screened alternatives.

     The   methodology  has  been demonstrated  by  application  to  two  widely
 differing  situations.   In a hypothetical case (Section 6),  a decisionmaker  is
 required   to choose  between  alternative  techniques  for  the  disposal  of  a
 single  high-volume waste stream.   In  this  example, analysis concentrates  on
 the costs  and  threats associated with  the various  disposal  techniques and the
 attitudes  of the parties-at-interest to these techniques.

     The  second  demonstration (Section 7)  involves a  case  study of  hazardous
 waste  management  in  Oregon.   Though   this  study would  not be  sufficiently
 detailed  or comprehensive to provide  definitive  planning guidance to  Oregon
 environmental  agencies, the usefulness of  the  methodology should be evident
 from  this  wide-ranging analysis.    In  contrast to the  first  demonstration,
 simplification of  the data to facilitate analysis  is an  important part of the
 procedure, and the  use  of  scoring  to weight the divergent   interests and
 attitudes  of the parties-at-interest is illustrated.

     Both  demonstrations  show how the  decisionmaker's  task  can  be simplified.
 In  the first  case some  options  can be eliminated because  they are  dominated
 by  others; in the Oregon case  study,  certain waste management  approaches are
 eliminated through a series of  paired  comparisons.  However, in neither case
 is  the  final  decision  clear-cut.   In  both cases,   the  decisionmaker must
 decide  on an  appropriate degree of risk aversion,  and in  the Oregon  case,  he
 must also  subjectively  balance  several  different  considerations before he can
 come to a  conclusion.

     These two demonstrations illustrate both  the width of possible  applica-
 tions  and  the  potential  value of the methodology  as an aid  to decisionmaking.
'Other   potential   applications   include  analysis   of  alternatives   for  the
 treatment  or  disposal  of a  particular type of  waste  (e.g.,  PCB's)  at the
 regional   or' national   level,  and  as  an aid   to  comparing the  nontechnical
 aspects of promising  new disposal   techniques  with existing ones.    Parts  of
 the  methodology  could be of  assistance in  conducting  a  technology assessment
 on  a new  treatment or disposal  technique.   Furthermore,  the methodology could
 be  extended to  assist  in a  wide variety of decisionmaking  situations where
 costs   and  benefits  cannot  readily   be   compared and where  sociological
 considerations are important.

 RECOMMENDATIONS FOR  FURTHER  RESEARCH

     The   methodology  has  been   demonstrated  usi-ng  comparatively  simple
 examples.   In practice,  it should  be capable of dealing with more complex

-------
situations; thus  It  should be tested and, if necessary, developed to fulfill
a decisionmaker's needs under more complex circumstances.   In particular, two
of the links in the socioeconomic interaction model require further analysis.
One  linkage is  between a policymaker's objectives and  the  approaches (i.e.,
strategies) that  may be  used  to control  hazardous wastes, and  the  other is
between  the  approaches  and  the  physical  techniques  that  are  actually
employed.  The  latter  is  particularly important, since the use of nonregula-
tory  policy  elements such  as  incentives, subsidies, and penalties  has been
extensively analyzed  in  the  literature,  but their  application  to practical
hazardous waste management situations needs further investigation.

     The  EPA's  documentation  and analysis of  hazardous  waste  incidents is
helpful  in developing  the threat scenarios used by  this  methodology, and it
is recommended  that  this  work be continued.   (The threat scenarios developed
in the  demonstrations  presented  in  this  study  benefited  significantly from
data  on  damage  incidents  collected  and published by EPA..)   In  addition, it*
would  be  valuable   if  modeling  studies  (e.g.,  of  leachate movenent  from
landfills) included typical results for commonly encountered situations as an
aid to the generation of threat scenarios.

     Research  on  risk evaluation,  and  in  particular  on  the  psychometric
expressed preference method  (see Appendix F), is promising and should now be
developed  to   provide   more  specific  guidance  for  common  environmental
decisions.

-------
                                  SECTION 3

                   OVERVIEW OF HAZARDOUS WASTE MANAGEMENT


UNIQUE ASPECTS OF HAZARDOUS WASTE MANAGEMENT

     As  a  first  step  toward  developing  a methodology  for  analysis  of
hazardous waste management  alternatives,  an economist must determne -hat is
special about hazardous wastes (i.e., how does the management  of tnese wastes
differ from that  of  other wastes or pollutants).   A clear-cut answer to this
question is difficult to find, but the definition of hazardous waste provides
some  indications  of  its  special   characteristics.    The  hazardous  wasta
definition used for this study classifies any waste as hazardous if it

     .  .  .  [poses] a  substantial  present  or potential  hazard  to  human
     health or living organisms because such wastes are nonaegradaole or
     persistent in nature or because they can be  biologically  macnvied,
     or oecause they  can  be lethal, or because they may otnerwise  cajse
     or tend to cause detrimental  cumulative effects.   [Emphasis added.]
     (U.S.  Environmental Protection Agency, 1974b:3)

     This definition  is based on  the proposed Hazardous Waste Management Act
of 1973.  This  statute was not enacted and was replaced by Subtitle C of the
1976 RCRA.  Though the definition used in  that Act is basically similar, it
is  slightly narrower  and  less  illuminating  for  the purposes  of  analysis.

     Several  distinguishing characteristics  relevant to  economic  analysis,
and in particular  to  the  potential  damages associated with hazardous wastes,
emerge from this definition.  First, a hazardous  waste can pose a substantial
or strong threat to man or the environment.Thisstatementsuggeststnat
hazardous wastes  need more  careful  management than  inert  wastes  or conven-
tional pollutants, such as  an organic waste that creates a biological oxygen
demand (BOO).   For example, when an organic waste is discharged into a river,
the level of dissolved oxygen below the discharge point will fall as bacteria
degrade  the waste.   However,  little harm  is  done to the  river,  wnich will
essentially return  to normal  downstream  provided the BOD  of  the  waste does
not cause oxygen  to  fall  to a  level  where fish  are threatened, and arovidea-
nutrients  released  from  the  waste  do not  cause eutrophication  (Freeman,
Haveman, and  Kneese,  1973).   This example illustrates a management technique
that may be acceptable for conventional pollutants (i.e.,  using the natural
environment,  particularly  air and  surface waters, to  assimilate  the waste,
while  accepting some  local  degradation).   In contrast, for a  hazardous waste
the  potential  for  damage  may  be so  great or the  environment may  have so
little assimilative  capacity that uncontrolled discharge  to  the environment
may be unacceptable.

                                     8

-------
     Second, hazardous  waste management  will  be more concerned with threats
or risks  as opposed  to  readily  anticipated  environmental   impacts"!177
for example, a  paper  mill  discharges an organic waste  to  a river,  the  level
of  dissolved  oxygen  in the water  will fall  as previously described.   The
effect  of  discharging  this waste  to  the  river  can be  comparatively  well
predicted,  and  waste  management decisions  will  be based,   in part,  on  these
effects.  On the other hand, if a heavy metal  waste is injected  into a saline
aquifer as a means  of disposal, the  intention is  that  the waste wll remain
in  the aquifer  and  thereby  cause  no  harm to  the environment.   One  must,
however, be  concerned  about threats  that may  arise from  the use  of  this
disposal method.  For example,  the  saline aquifer may be interconnected with
an aquifer used  as  a  source of  freshwater, which  could  become  contaminated.
Unlike the comparatively predictable effects  from nonhazardous  waste manage-
ment techniques, it is  usually  difficult to estimate the  probability that a
hazardous waste  threat  will  materialize,  or to predict the magnitude or cost
of the potential damage.

     Third, hazardous  wastes   are  often  persistent  or  nondegradabla.   This
quality is significant  because  (1)  effects may be irreversible, and (2) time
scales  of  *'nterest can span more  than one  generation.   Injecting  a  heavy
metal   waste  into  an  aquifer  may  be  essentially  irreversible.  Should  the
aquifer later  be needed as  a  water or mineral  resource,  or be  found  to be
interconnected  with  a  freshwater  source,  decontamination mighc  not  be
feasible.   (In  practice there  are  degrees of reversibility and  irreversibil-
ity, and  some  irreversible  changes are more  significant  than  others.   See
Fisher  and Krutilla, 1974.)    Irreversibility  can  also   introduce  "option
value" and associated concepts  that relate to the benefits of avoiding  fore-
closure of future  courses  of  action,  or  options.   Furthermore,  when  the
decisions of today affect future generations,  it raises the difficult problem
of whether or  not  it  is appropriate to discount to reflect a reduced present
value  of  costs  that  occur in  the  future.   These  two  topics  are  discussed
later.

     Fourth, hazardous  wastes   include  those that  are biologically magnified
or  have  cumulative effects.   Biological magnification  (or bioconcsntration)
is  the ability of  organisms to accumulate  chemical  contaminants  to levels
that are  higher than  those in  their  food sources.  Such magnification can
occur  at  several stages in  the food chain, with the possible  result that a
contaminant that is present in  insignificant concentrations at  the lower end
of the  food chain  could concentrate to toxic  or lethal  levels  higher in the
chain  (Van Hook,  1978).   An  effect  is  cumulative when  the   impact on  the
organism depends on the entire history of exposure to the contaminant, rather
than depending  only  on the immediate  exposure.   The difficulties  of  esti-
mating  damages  due  to  exposure  are  compounded   when  bioconcentration  or
cumulative  effects  are  present.   However,  this  problem  is not  unique  to
hazardous wastes, as  pollutants that are not  normally regarded as  hazardous
may also  have  cumulative  effects   (such as that of  urban air  pollution  on
human health [Lave and Seskin,  1970]).

     A  final  characteristic of  hazardous  wastes (one  that is  not apparent
from  the  definition)  is   that  the composition of wastes can  vary substan-
tially, not only because of different sources,  but also from day to day  for a
given source.   This factor complicates treatment to reduce the hazard and can

-------
inhibit  resource  recovery  activities.   Furthermore,  it may  mean  that  the
degree  of  hazard  posed  by  the waste  is  not  well  defined.   Again,  this
characteristic is not unique to hazardous wastes.

     To  summarize,  hazardous  wastes are  characterized by  strong potential
adverse effects, and  their  management may involve irreversible decisions and
intergenerational time scales.   The threats that these wastes pose to man and
the environment are  often  difficult to specify because insufficient informa-
tion is available.   Waste  composition variability, biological  magnification,
and cumulative effects compound the problem.

THE NATURE OF HAZARDOUS WASTES

     The  two basic  aspects of  a  hazardous waste  are  (1) that  it  is  an
unwanted material,  and (2)  that the material  has hazardous properties.

Wastes and the Environment

     As Ayres  and  Kneese.  (1969) have pointed  out,  except  for increases in
inventory,  all   materials   that  enter  the  economy  end  up as  wastes  (or
residuals,  as  they are commonly termed in  the literature  of  environmental
economics).   The  inputs  to the economic  system  are  fuels, foods,  and  raw
materials, and these inputs are partly converted to final products and partly
discarded  as process  residuals.   After  the  final  products  have fulfilled
their  role,  they  too  are  discarded.   Whereas  final  consumption  products
provide  services  to  man,  wastes usually  provide disservices.   They either
consume resources  to achieve disposal without  environmental degradation, or
they  may  cause pollution   that results  in  such  effects  as  fish  kills,
increased  difficulty  in water treatment, reduced  public health,  etc.  (Ayres
and Kneese, 1969).

     Figure  1  illustrates  the  routes  by  which  wastes  can  affect  the
environment.  It also  shows that there are several stages or points at which
wastes can be controlled,  as follows:

     1.   The waste stream from the manufacturing process can be changed
          to reduce the generation of hazardous wastes;

     2.   Some waste streams can be treated to reduce the hazard;

     3.   The initial  disposition  of the wastes  in  the environment can
          be controlled;

     4.   Subsequent  interchange   among   environmental   media  can  be
          •restricted;

     5.   Interaction  between  the  wastes  and  living  receptors  (i.e.,
          organisms  that   may   be   affected  by  the  wastes)  can  be
          controlled;

     6.   If  the  wastes  have  reached some   living  receptors,  waste
          migration   to  other  receptors,   especially  man,   can   be
          controlled.

                                     10

-------
      WASTE GENERATION
        AND DISCHARGE
WASTE DISPOSITION AND  I     RECEPTORS AND
   TRANSPORT IN THE     [ENVIRONMENTAl EFFECTS
    ENVIRONMENT       I
MAN

©
'NON-HUMAN
] LIFE
 RESOURCE RECOVERY
  Figure 1.   Routes  by which  hazardous  wastes  can  affect the  environment.
             (The numbers  refer to points at which control  can  be exercised.)

     Different waste disposal  techniques  provide varying degrees of control.
For example, ocean dumping provides no control  of marine  life exposure  to  the
wastes (i.e., no  control  at  point 5 in Figure  1).  On occasions  such dumping
has lead to  the  need to ban fishing  and  shellfish harvesting  and to impound
fish because of  the  danger to  health posed by  human  consumption  of contamin-
ated  marine life  (Council  on  Environmental Quality,  1970).   These actions
constitute  control  at  the last possible  stage (i.e.,  point 6 of  Figure  1).
In contrast, disposal of wastes in deep mines should  give rise  to very  little
waste migration  to air  and water (i.e.,  good  control  at point 4).   [f, how-
ever, groundwater were  to  become  contaminated  as a  result  of  mine disposal,
it  is  unlikely  that the  groundwater would  be used  for irrigation or as  a
drinking water supply because  of  the depth of  the aquifer.  Thus in addition
to control  at point 4,  mine disposal provides control at  point  5.
                                     11

-------
     Whereas  society's  primary  concern  is  to control  the extent  to  which
living  organisms  are exposed  to hazardous wastes,  in practice  this  can be
achieved indirectly by control ling at points 1 through 4 as well as by direct
control at point 5.  However, there may be some wastes, especially those that
are  generated  on  an  irregular basis  (such  as off-specification  batches  of
products,  cleanup  wastes,  discarded  laboratory  chemicals, etc.)  for  wnich
control  at  point  1,  and  in  some  cases at  point 2,  is  not  feasible.   This
underscores the  importance  of controlling waste  disposition  in the environ-
ment  (point  3),  even  though the  strategy  of control at  earlier stages nas
been deemed to be  more desirable by the U.S.  Environmental Protection Agency
(1976a).

Hazardous Attributes

     According to  Kohan  (1975),  hazardous substances generally fall into one
or more of the following categories:  Toxic to human and/or lower life forms,
radioactive,  flammable,  reactive,  explosive,  oxidizing,  irritating, geneti-
cally  active,  strongly  sensitizing,   and/or subject to  bioconcentration.
Though all classification systems  reviewed by Kohan  include  toxicity  as one
criterion  for designating  a  material  as  hazardous,  there  is considerable
variation  among  different   systems  with  respect  to  the choice   of  other
attributes  that  can  render  a material  hazardous.   This  variation may,  in
part, stem from the differing use orientations of the classification systems,
and  also  from  overlap  between  attributes.   For  example,   flammable  and
explosive wastes are  generally toxic,  and many radioactive and some biolog-
ical wastes are  also toxic  (U.S. Environmental Protection Agency, 1974b).  A
comparison of  some of  these classification systems  is provided  in Taole 1.

     The  problem  of  deciding what  makes  a  material  hazardous is not,  of
course,  restricted to  the  attributes to be considered,  but also encompasses
the  potential magnitude  or  severity of the effect.  While some materials are
universally 'regarded  as  hazardous,  there is a gray area in which authorities
will  disagree as   to  whether  or  not a  material  should  be  classified  as
hazardous.  In many cases, the problem is compounded by inadequate data about
the  potential effects of the material on man and the environment.

     The  term  hazardous may  be  regarded as  having  two  connotations,  one of
which  relates to  the intrinsic  properties of the  waste  itself,  i.e.,  the
amount of  damage that it is  capable of  rendering to man or the environment.
The  second  relates to extrinsic factors,  such as the degree  of exposure to
the  hazard,  e.g.,  the quantities and  circumstances  surrounding the exposure
(Battelle  Memorial Institute, Pacific  Northwest  Laboratories, 1974, Vol.1).
•This  discussion  focuses on  the  intrinsic  properties,  but it  will  be noted
that the quantity  of waste  (an  extrinsic factor) is  included  in several  of
the  systems' summarized in Table 1.

     Approaches  for identifying  hazardous materials  fall into  two general
categories:

     1.   Specific  rules  or decision models  for  determining  whether or
          not a material is hazardous;
                                     12

-------
            TABLE 1.   HAZARDOUS MATERIALS CLASSIFICATION CRITERIA*

Criteria














System
Title 15, U.S. Code, Sec. 1261
CPSC-Title 16, CFR, Part 1500
Food, Drug, and Cosmetic Act
OOT-Title 49, CFR, Parts 100-199
Pesticides-Title 40, CFR, Part 162
Ocean Dumping-Title 40, CFR, Part 227
NOISH-Toxic Substances List
Drinking Water Standards
FWPCA Sec. 304 (a)(l)
Sec. 307 (a)
Sec. 311 (b)(2)(A)
Clean Air Act-Sec. 112
California State List
National Academy of Sciences
TRW Systems Group
Battell e Memorial Institute [N.W.]
Booz-Allen Applied Research, Inc.
Oept. of the Army
Dept. of the Navy
National Cancer Institute

^—
Oi
u
1^*
0)
o

'o
u
•r™
X
o
1—

X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
ra
£-
0)
+J
>> ro
<-> =
*^~ ^">
i— 01 O> *J C1
•— > > — c
.3 ••-•!- > —

! 2 o IJ •£
S i— S- U TS
ro C. i- ro —
•— x o o x
U. LLI U Of O

X X
X X

XXX X
X







XXX
X X
XXX
XX XX
XX X

X X X X X

s- c
01 0
N •!—
•f™ ^J
«J 03
01 -i- I-
> ^ -»J
•i- 01 C
4-> J-J Ul 1>
u c u
TO ^ ^n cz
o -J c o
•r- •>- O U
•a s- t. o
m s- -J •—
Q£ I— i 1/1 C3

XXX
XXX
X
X X
X
X

X
X
X

X
X X

X X
XX X


X
X
re iiiugcnic
(0
»_
^
ijO

X
X
X
X

X



X
X



X

X


X

* Source:   Kohan, 1975:2.





                                     13

-------
     2.   Listings  of  materials  deemed  to  be  hazardous  (the  pure-
          compound approach).

     There are  difficulties associated with both approaches.   The decision-
model approach  involves  specifying levels  of  hazardous attributes,  sucn as
flash point,  degree of toxicity, etc.   A major problem is that it is not easy
to decide upon  the criteria to be used.   As  Flinn,  Thomas and Bishop (1974)
point  out,  most  classification systems  in use are  deficient in  that they
focus  on  acute effects  to the  neglect of chronic  effects,  have  a  limited
domain of  concern, and  are not designed to handle  degradation  products and
synergistic effects when one material  mixes with another.

     The resources  required for testing waste streams could be considerable,
especially with respect  to toxicity  and genetic effects.  Estimates of costs
ranging up to  £750,000  for animal  toxicity testing of a single chemical have
been  suggested  (Portney, 1978).  Furthermore, variations  in  the composition
of  a  given  hazardous  waste  stream  could make  it difficult  to obtain  a
representative sample.

     Enormous  resources   could also   be  required  to  evaluate  hazardous
materials using the pure-compound  approach.   Flinn,  Thomas anc! Bishop (1974)
indicate that a quarter of a million  chemical entities may be identified each
year, and that  several  hundred are introduced into  commercial  use annually.
The  pure-compound  approach also raises the problem  of  defining  the  concen-
tration at which  the presence of a  given substance renders  a  waste stream
hazardous.    Furthermore,  extrapolation  from  laboratory  animal   studies  to
effects on man is  fraught with difficulties (Rail,  1975).  These problems are
compounded by the  observation that low concentrations of some elements may be
essential to  certain  forms of life,  whereas  higher concentrations  may be
toxic or lethal (Venugopal  and Luckey,  1975).  In addition, the approach does
not  allow  for synergistic  or  antagonistic effects that may occur when mix-
tures of chemicals are present in a single waste stream.

     Williamson (1975)  includes a  detailed discussion of  the  advantages and
disadvantages  of  a number  of  versions  of  these two  general  approaches, and
Battelle  Memorial   Institute,   Pacific  Northwest Laboratories  (1974, Vol.1)
comprehensively discusses  the  attributes that could  make  a  waste hazardous.

     Of various hazardous waste decision models,  the Battelle model (Battelle
Memorial  Institute, Pacific Northwest  Laboratories,  1974,  Vol.1) is probably
the   best  known.   This  model  was   included  in  the  Report to Congress:
Disposal  of Hazardous  Wastes  (U.'S.  Environmental  Protection Agency,1974b)
and  is illustrated  in Figure  2.  While  the usual  purpose  of decision models
of this  type is  to provide a yes/no answer,  some  designate  two degrees of
hazard—for  example,  "dangerous" and  "extremely hazardous"  (Mehlhaff,  Cook
and Knudson,  1977).  In addition, some models have  been designed to produce a
hazard rating  or  ranking.  Klee  (1976)  has  reviewed three  such models and
points  out   that   because  each  represents  a  different evaluator's  utility
function, the  correlation  between  results obtained from each  model  is poor.
                                     14

-------
                              WASTE
                             STREAM
                       DOES WASTE CONTAIN
                    RADIOACTIVE CONSTITUENTS
                          > virc LEVELS?
                                  NO
                        IS WASTE SUBJECT TO
                        BIOCONCENTHATIONI
                       IS WASTE FLAMMABILITV
                        IN MFPA CATEGORY a
                                  NO
                       IS WASTE REACTIVITY
                       IN NFPA CATEGORY 47
                     DOES WASTE HAVE AN ORAL
                         LDIU < SOmg/kg'
                   IS WASTE INHALATION TOXICITV
                      200 oprn AS GAS OR WIST'
                      LC,. < 2 mg/hur AS OUST?
                                , ,  NO
                   IS WASTE DERMAL PENETRATION
                     TOXICITV LO,n < 200ftig/kg?
                                  NO
                    IS WASTE DERMAL IRRITATION
                       REACTION < GRADE 8'
                                , ,  NO
                     DOES WASTE HAVE AQUATIC
                      96 hr TLm < I 000 
                                  NO
                        DOES WASTE CAUSE
                        GENETIC CHANGES'
                                  NO
                          OTHER WASTES
                                                'ES
                                               YES
                                               Y.ES.
                                               YES
YES
                                               »ES
                                                                 HAZARDOUS WASTES
               Figure 2.   The  Battelle hazardous waste  decision  model.

Source:   U.S.  Environmental  Protection  Agency,  1974b:57.
                                             15

-------
Categories of Hazardous Waste

     There  are  numerous  ways  of  classifying  wastes  into generic  groups.
Bases include:

     1.    Hazardous attributes;

     2.    Material  or   chemical   classifications   (metals,   organics,
          inorganics,  etc.);

     3.    State of matter (gas,  liquid, sludge/slurry, solid);

     4.    Geographic occurrence;

     5.    Industrial origin  (e.g.,  either  Standard  Industrial  Classifi-
          cation [SIC] or process origin);

     6.    Amenability to different forms of resource recovery,  treatment
          and/or disposal;

     7.    Control by existing and/or proposed regulation;

     8.    The  nature  or significance  of 'the threats that  they  pose  to
          the environment.

     Although  similar  to  classification  by  hazardous  attributes, the  last
approach (8) takes  into  account the waste's interaction with the environment
(e.g., the ease  with  which it can be biodegraded, transported  and/or  changed
to  other  forms  by  natural  processes).  It might also  take account of  the
background level of the material in the environment.

     With some  classification  schemes, there may- be  difficulties  in  alloca-
ting wastes  to  specific  categories because of overlap problems—for example,
if  classified   by hazardous -attributes,  a. waste might- be both  'toxic  and
flammable, or  if classified  by  chemical composition, a waste  might be  basi-
cally organic but with low concentrations  of heavy metals that  cause it  to be
hazardous.

     In practice, hybrid  classification systems  are often used;  for example,
Berkowitz, March, and Home (1975) classify all  industrial wastes (including,
but  not  limited to,  hazardous  wastes), into  29  general  waste  streams  based
largely on  the  states of  matter  and. the  materials present  or  chemical .com-
position.    These same  authors also• propose a•  hierarchical  classification
system that  has 21 dimensions,  including   such  items as geographic  and  SIC
origins,  onsite  treatment  received,  mode  of transport to disposal, hazardous
attributes,   and  properties   related  to  recyclability  and  decomposibility
(Berkowitz,  March and Hprne,  1975).   In contrast, Perna  (1977)  suggests  the
use  of  only eight  basic  categories,  one  of which is  "hazardous wastes."

     In  dealing  with  hazardous  wastes,   it may   sometimes  be  useful   to
distinguish between wastes from different  sources, as follows:

     1.    Industrial process  wastes;


                                     IS

-------
     2.    Radioactive wastes;

     3.    Hospital  wastes (pathological);

     4.    Chemical  laboratory  wastes;

     5.    Surplus pesticides and pesticide containers;

     6.    Obsolete explosives;

     7.    Chemical  and biological warfare  wastes;

     8.    Other special wastes.

     These distinctions are made on the grounds of the  aoproaches that nay-be
used to control  the  disposition of the wastes.   Each  type  of waste may call
for a different  control  method.   For example, radioactive wastes are already
subject  to different  regulations  than  chemical  wastes,  wh'ile  regulations'
developed to deal with large quantities of industrial process wastes might be
very cumbersome  if applied to small quantities of laboratory wastes or.empty
pesticide containers.

     Although post-consumer wastes undoubtedly  contain  hazardous components
on occasions, they are not usually regarded  as  a source of hazardous waste.
In contrast, mill  tailings can  have concentrations  of heavy metals that.are
in excess  of background  levels  and could  therefore pose  a hazard (Midwest
Research  Institute,   1975).   However,  because  of  the  large quantities  of
materials  involved,  there are few  feasible management alternatives for such
wastes and hence they were not specifically considered in this study.

Treatable and Nontreatable Wastes

     Although much general  analysis can  apply to all hazardous wastes, there
is  one  hazardous  waste   classification  that  can  usefully  be   employed  in
general policy or planning studies—whether a waste  is "treatable" or not.  A
treatable waste  can  readily be  detoxified or  rendered harmless  by physical,
chemical, or biological means; whereas a nontreatable waste cannot readily be
detoxified.  For example,  waste sulfuric  acid and phenol-contaminated waste-
water may  qualify  as hazardous wastes under  various classification schemes.
However, sulfuric acid may be neutralized by reaction with a low-cost, widely
available  alkali  such as  lime,  and phenol  is  readily  biodegradable at low
concentrations  (Rosfjord, Trattner,  and  Cheremisinoff,  1976).  •  Thus these
"treatable" wastes may usually be simply and inexpensively rendered harmless.

     In  contrast,  the toxic  properties  of a  heavy  metal  are fundamental to
that element, and  the waste cannot be treated to render it harmless.  Hence
if a waste  contains  heavy metals  in significant quantities, it  is necessary
to find  ways  to prevent the  release of these elements if the environment is
to be  protected.   (This  could,  for example, be achieved by physical contain-
ment  or by  insuring  that the  waste  remains  in a highly  insoluble form.)
However, a  "nontreatable"  waste of this type  remains a permanent threat, and
                                     17

-------
must  be  considered a  candidate  for perpetual  care,   requiring  continuing
monitoring  to insure  that  its  disposal  has  not affected  the environment.
This is an important distinction from a "treatable" waste.

     Inevitably,  there  is a  zone  of uncertainty or overlap  between  the two
categories.   For example, PCB's exhibit serious chronic toxicity, are subject
to bioaccumulation, and  are  highly .persistent'in the environment (U.S.  Envi-
ronmental   Protection   Agency,  1976b).    However,   as   syntnetic   organic
compounds,  they  are capable  of  degradation to a harmless  form by incinera-
tion.  Unfortunately, the  required incineration parameters (high temperature
and  long dwell  time)  are such  that effective  thermal  degradation  is  very
costly,  especially compared  with   simple  treatments of the  type  discussed
above.   Hence waste  PCB's  do not  readily  fit  into  either the "nontreatable"
or  "treatable"  category,  as  they are  technically  treatable,  yet  economic
considerations may result in  their not being treated.    For  this  waste,  the
most viable alternative might be some form of perpetual  care.

TECHNIQUES FOR THE MANAGEMENT OF HAZARDOUS WASTES

     The term "technique"  is used  -in this report .to denote a technical  means
of  changing,   treating,  or  disposing of  a  hazardous  waste.   The   term  is
restricted  to  direct physical activities  and does  not  imply anything  about
its  economic   or   social  effects -  or  the  policies  that  might  encourage  or
discourage the use of that technique.

     The  techniques  that are  available  fall  into  four  groups, as  follows:

     I.   Techniques that  change  the composition  or  magnitude of  the
          waste stream itself;

     2.   Techniques that  recover  values  (materials  or  energy)  from a
          given waste stream;

     3.   Techniques that  treat  the waste stream in order  to render it
          less harmful;

     4.   Techniques that store or dispose of the waste.

     A waste  may  be sequentially  subjected to more than  one technique;  for
example, a disposal technique may be preceded by some form of treatment.   The
term "disposal" (or sometimes "ultimate disposal") is commonly used to denote
removal' qf  the waste  from the immediate •location, of generation to some other
location where, it  is  put into"permanent 'storage' (as  in  a landfill), diluted
or dispersed  (as may  occur in ocean  dumping).   The  available techniques are
listed in Table 2 and are described in Appendix A.

LEGISLATIVE BACKGROUND OF HAZARDOUS WASTE REGULATION IN THE
UNITED STATES

     In the past decade,  significant advances have been  made in legislative
controls of most of the major sources of environmental  degradation.
                                     18

-------
         TABLE 2.   TECHNIQUES FOR THE MANAGEMENT OF HAZARDOUS WASTES*
          Technique

Change in waste streams:

  Process change


  Source reduction


  Waste separation


Resource recovery:

  Materials recovery
  Energy recovery

Treatment to reduce hazard:

  Physical treatment
  Chemical treatment
  Biological treatment
  Thermal treatment
  Encapsulation

Storage or disposal:

  Land application

  Landfill ing
  Mine disposal

  Lagooning
  Deep well injection
  Ocean dumping
  Engineered storage

  Space disposal
                      Comment
              To generate wastes that
                are less hazardous

              To generate less hazardous
                waste

              To separate hazardous *aste
                from nonhazaraous waste
Location!



Onsite


Ons He
Usually
  onsite
On/offsite
On/offsite
On/offsite ^
On/offsite  I  Variety of processes
On/offsite  \    available (see Table A.I.)
On/offsite J
              To immobilize wastes
Usually
  offsite
On/offsite
Usually
  offsite
On/offsite
On/offsite
Offsite
Usually
  offsite
Offsite
              For storage, liquid volume
                reduction by evaporation;
                may also involve treatment

              Liquids only
* Source:   Adapted  from U.S.  Environmental  Protection  Agency,  1974b;
  Kovalick, 1975, Appendix A; and other sources.

T Location  at which technique is  used,  relative to  site of  waste generation.
                                     19

-------
Amendments to  the  Clean  Air Act in 1970  (PL 91-604) established a system of
air quality and  emissions  standards that has resulted in marked reduction of
pollutants discharged into the air.  The Federal Water Pollution Control Act,
passed  in 1972  (PL 92-500),  instituted  a  similar  system  for  controlling
effluents  passing   into  the Nation's watercourses.   Both of  these statutes
include provisions designed to prevent hazardous pollutants from oeing dumped
into  the  respective environmental  media  with  which  they   are  concerned.
Hazardous air pollutants are controlled by establishing stringent limitations
for stationary  source  emissions of designated pollutants  (Arbuckle  et al.,
1976).  Discharges  of hazardous  or toxic pollutants into surface waters from
point sources  are  controlled  in  several different ways  in  the Federal Water
Pollution Control  Act, but the basic concept is  to  establish limitations on
the  amounts   and  types  of these  effluents.   In some  cases, standards  of
performance are  set  for  certain  industries,  while effluent  standards  for
hazardous pollutants are set in others (U.S.  Environmental Protection Agency,
1974b).

     Ocean  dumping, commonly  used  in  the  past,   has   now  been  virtually
eliminated as  a technique for the disposal of hazardous wastes (U.S. Environ-
mental  Protection  Agency,  1977a).   Under the Marine  Protection Research  and
Sanctuaries Act  of  1972 (PL 92-532,  as amended), the dumping  of  any radio-
logical   waste  is  prohibited; and a  permit  is  required  before   any  other
material  can  be  dumped.   It is  the policy  of the Act to  regulate  all  ocean
dumping and to prevent  or  strictly limit the ocean  dumping  of any material
that  would  adversely  affect the marine   environment  (U.S.  Environmental
Protection Agency,  1977a).   Also,   in  1974,  the  United  States ratified  the
1972  International  Convention  on  the  Prevention of  Marine Pollution by  the
Dumping of Wastes and Other Matter.  This convention prohibits the deliberate
ocean disposal  of  certain waste  materials (which  are  additional to  those
considered above)  and requires that special  care be taken in  issuing permits
for other specified  materials  (U.S.  Environmental Protection  Agency,  1977e).

     In  addition  to  the  legislation  mentioned  above,  there  are  13  other
Federal  statutes that have some  bearing on the treatment, storage,  transpor-
tation,   and   handling  of  hazardous  wastes  (U.S.   Environmental  Protection
Agency,   1974b).   Much  of   this  legislation  applies to  specific   wastes  or
categories of  wastes (explosives, for example).

     At the beginning of this  decade, it was  realized  that (a) increasingly
stringent control  of air and  surface water pollution was diverting wastes to
various forms  of  land disposal that were largely uncontrolled,  and that (b)
although  some aspects  of   hazardous  wastes 'were addressed  piecemeal  by  a
variety of Federal  and State statutes, some comprehensive control  was  needed.
These two forces  prompted Congress  to  enact  Section  212 of the Resource
Recovery Act  of  1970 (PL 91-512),  which directed EPA to  undertake a study to
better  identify  the nature and  scope of  the  hazardous  waste problem,  with
special  orientation  toward establishing a system of  national  disposal  sites
for hazardous  wastes.   The resulting  report,  delivered to Congress  in June
1973  (U.S.  Environmental Protection Agency, 1974b),  strongly  emphasized  the
need to regulate hazardous  wastes in a comprehensive manner.
                                     20

-------
     The Initial legislation proposed  subsequent  to this report provided for
identification of hazardous wastes,  establishment of standards for treatment
and disposal of  such  wastes,  and establishment of  guidelines  for State pro-
grams for implementing such standards.   It did not, however,  propose a system
of  Federally  controlled  national  disposal  sites.  After  several years  of
deliberation,   similar  legislation  was  finally  passed  in  October  1976  as
Subtitle C of  the  RCRA (PL 94-580).   The  objectives  of  this Subtitle are as
follows:

          The  basic  thrust of the hazardous waste title, is  to identify
     what  wastes are  hazardous and  in what  quantities, qualities, and
     concentrations  and the  methods  of disposal  which  may make such
     wastes hazardous.  The title requires that the Administrator prom-
     ulgate  regulations  applicable  to  generators.   Such  regulations
     include recordkeeping, informing those that transport or dispose of
     such hazardous waste  of  the characteristics  of such waste and the
     initiating of a  manifest  system so that the  waste generated can be
     traced to the site of ultimate  disposal. .  .  .

          Regulations  are  imposed on  transporters of hazardous  waste.
     Most important is  the initiation  of  a  manifest  system  so that the
     hazardous waste can be traced from the generator to a facility that
     has an approved permit. .  . .

          Other  regulations  required to be  promulgated  relate to those
     who treat,  store or dispose of hazardous  waste.   Such  regulations
     are  to consist  of compliance  with  the manifest  system, record-
     keeping requirements and inspections.

          The  Administrator  is also  empowered to recommend methods of
     treatment,  storage  or   disposal  of  hazardous  waste,andthe
     operation  of  such  facilities,  to  assist  the operators  in safely
     handling such hazardous waste.

          Finally, those who store,  treat, or dispose of hazardous waste
     are required  to  receive  a permit either from the Administrator or
     from  the  appropriate state  agency  authorized by the Administrator
     to grant such a  permit. .  .  .  [Emphasis added.]
     (U.S. House of Representatives,  1976:6,7)

     Prior  to  the passage  of  the RCRA,  only a limited  number of States had
comprehensive  hazardous waste  legislation  (Lehman,  1976), although at least
25  had  legislation or regulations that  provided  some  control   (U.S.  Environ-
mental  Protection  Agency,  1974b).  EPA  is currently promulgating regulations
that  will   insure  adequate management  throughout the United  States.  These
regulations  are  designed  to  provide  cradle-to-grave  control  of hazardous
wastes.   The  basic management  philosophy  is to  control the  disposition of
hazardous  wastes,  and  to  provide adequate  environmental  protection without
specifying  the technologies  that must  be used.   One of  the most difficult
problems  has  been to define hazardous  waste.  The regulations specify that
certain materials  are hazardous, whereas  tests are utilized to determine the
status  of some waste  streams.


                                     21

-------
PREVIOUS RESEARCH ON HAZARDOUS WASTE MANAGEMENT

     Prior to 1970, when Congress passed Section 212 of the Resource Recovery
Act,  little  attention  had been  paid to  the. problems of  hazardous  wastes,
although the risks and procedures -Involved In the transportation of hazardous
materials  had  been  studied  (Smith,  1976).   Following  the  1970  Resource
Recovery  Act,   EPA  commissioned a  number  of wide-ranging  research  studies.

     In an initial  study,  Booz-Allen Applied Research Inc. (1972) compiled a
candidate list  of  hazardous materials and attempted to assess the nature and
magnitude  of the  problem.   This  was followed  by  an in-depth  study,  which
included  developing profile  reports  (summarizing  quantities  generated and
hazardous  properties)  on  over  500  potentially  hazardous  materials  and
analyses of a variety of treatment and disposal techniques that might be used
in hazardous waste  management  (Ottinger et al., 1973).  All of this work was
predisposed  toward  the concept  of a  system  of national  disposal  sites for
hazardous  waste,   and   another  EPA-sppnsored  study  examined  alternative
approaches to  the  use  of  such  disposal' sites.  Three basic  approaches were
considered:   Onsite waste  processing Cive.,  treatment and/or disposal of the
waste  at  the  location where it is  generated),  offsite processing  at some
regional  hazardous  waste  facility, and a  combination  of  onsite pretreatment
and  offsite  treatment  and  disposal.   This  study  examined  the  process
economics and  other considerations associated with these  alternatives  for a
number of common  waste  streams  that were  regarded  as  strongly hazardous and
concluded that most should be processed at national  disposal sites (Arthur D.
Little, Inc.,  1973).   Meanwhile,  in Program for the Management of Hazardous
Wastes. Battelle  Memorial   Institute,  Pacific Northwest  Laboratories  (1974)
estimated the  quantities of hazardous waste generated in  the United States
and made  a detailed examination  of the feasibility of a  system of  national
disposal sites, including conceptual designs, etc.

     The results of this series  of studies were integrated into the Report to
Congress:   Disposal of Hazardous Wastes    (U.S.    Environmental   Protection
Agency,  1974b).Tn  this  report  EPA  concluded   that  for  the most  part,
hazardous wastes were disposed of using low-cost methods, that did not provide
adequate'environmental protection.   The technology  to adequately manage most
hazardous wastes was  found  to  be available,  but adequate management is often
costly.  Hence the waste generators frequently have an economic incentive for
inadequate management.

     The  Report to Congress:   Disposal of Hazardous Wastes   indicated  that
about  9  million tonnes  (10 million short tons) of nonradioactive  hazardous
wastes were being generated annually in the United States, increasing at 5 to
10 percent per  year.   About 90  percent of these wastes  were in liquid form,
the remainder  being solids, sludges, and  slurries.   (Emissions of  hazardous
materials to the  air  were  not considered  in  these  studies as such emissions
were already controlled  under Section 112 of the Clean Air Act as amended in
1970.)  About 60 percent of the  wastes were organic  materials.  Virtually all
of these wastes were  industrial process wastes, and  practically all  of them
were toxic (U.S. Environmental  Protection Agency, 1974b).
                                     22

-------
     Subsequently,  EPA  commissioned  detailed studies  on the  14  industries
that were  believed to  contribute  the bulk  of all  hazardous  process  wastes
(Abrams, Guinan,  and  Derkics,  1976;  Arthur D. Little,  Inc.,  1976;  Battelle-
Columbus  Laboratories,   1976;   Foster  D.   Snell,   Inc.,  1976a;  Gruber  and
Ghassemi, 1975; Jacobs  Engineering  Co.,  Inc., 1976; McCandless at al., 1975;
Leonard et al., 1975;  SCS Engineers, Inc., 1976;  Shaver et al., 1975;  Swain,
1976; WAPORA, Inc., 1976,  1977a, 1977b).   Each of  these studies characterized
the structure of  the  industry,  estimated the  total  process  wastes  generated
(both at the  time of  the  study and in the future), identified the portion of
the waste  that was considered  to be  potentially  hazardous,  determined  the
disposal methods  currently in  use,  and estimated  direct control  costs  for
various  levels  of treatment and  disposal  technology.  The  results of these
studies, which are summarized in Table 3, showed that these industries gener-
ated nearly  29  million  tonnes  (32 million short tons)  of hazardous waste in
1974.   This  is expected  to  increase to  38 million  tonnes  (42 million short
tons)  by 1983,  largely because  of  residues  from  additional  air  and water
pollution controls.   A  pervasive  problem with this research has been that of
defining a hazardous  waste.   In each of the  14  industries  studied, the. con-
tractor  chose  the definition employed, which  was  not necessarily consistent
with  that  used  for  the  Report to Congress:  Disposal  of Hazardous Wastes
(U.S.  Environmental Protection  Agency, 1974b).The  estimates  in the  latter
report  of  9  million  tonnes  (10 million  short  tons)  of  wastes  considered
potentially  hazardous  were  generated  using  the  Battelle hazardous  waste
decision model  (U.S.  Environmental   Protection Agency,  1974b).   However,  the
problem  of  determining  how  much  hazardous  waste  is  generated  is .not
restricted to  the choice  of  criteria.  In addition,  waste  stream magnitudes
and concentrations of constituents are uncertain or subject to variation,  and
the  toxicity  and  genetic  effects  of-  the  components  may be  ill-defined,
especially in the presence of other components.  Consequently, the difference
(a factor of three) between the two totals is not unreasonable.

     For some  industries,  additional studies have been completed en alterna-
tive control  technologies  that  might be employed and on the economic effects
(e.g.,  changes  in product prices  and plant  closings)  of possible hazardous
waste  management   regulations  (e.g.,  Versar,  Inc.,  1977;  Williams  et al.,
1976).  Another study examined  the structure and capacity of the independent
hazardous waste management  industry.   In 1975, this  was  found to consist of
about 95 firms  operating  110 facilities.  Annual  capacity  was determined to
be  about 7.3 million tonnes  (wet basis),  but only  about half that, capacity
was  being  used  (Foster  0.  Snell,  1976b; also  see Farb  and  Ward,  1975).
Comparison of  these data  with those for total generation of hazardous wastes
(Table  3)   supports  EPA  findings  that  most  industries  dispose  of  their
hazardous waste locally to the land and that only a small proportion of this
waste  is handled   in  a  manner  that  EPA regards  as environmentally adequate
(U.S. Environmental Protection Agency, 1977b).

     Other Federal  research  has included the collection and analysis of data
on  hazardous waste incidents (situations  in  which significant environmental
degradation  or  damage to  health or  life has occurred), and numerous support-
ing  studies  have   been  conducted on  hazardous waste treatment  and disposal
technologies.  This research is discussed  in Appendices A and B.
                                     23

-------
             TABLE 3.  POTENTIALLY HAZARDOUS WASTES GENERATED
                       IN THE UNITED STATES*
                                          Quantity of waste generated
                                           (million tonnes per year)
                                           197*
                       1983
 Industry category
Dry
Wet
Dry
'.vet
1.
2.
3.

4.
5.
6.
7.
8.

9.
10.
11.
12.
13.
14.

Batteries
Inorganic chemicals
Organic chemicals, pesticides
and explosives
Electroplating
Paint and allied products
Petroleum refining
Pharmaceuticals
Primary metals smelting
and refining
Textile dyeing and finishing
Leather tanning
Special machinery
Electronic components
Rubber and plastics
Waste oil re-refining
Total
0.005
2.000

2.150
0.909
0.075
0.625
0.062

4.454
0.048
0.045
0.102
0.026
0.205
0.057
10.763
0.010
3.400

6.360
5.276
0.096
1.757
0.065

8.335
1.770
0.146
0.153
0.036
0.785
0.057
28.755
0.105
2.800

3.300
1. 751
0.105
0.811
0.104

5.536
0.179
0.068
0.157
0.050
0.299
0.1-ki
15.909
0.209
4.300

12.565
5.ISC
0.145
1.S88
0.108

10.-il3
0.715
:.:i*
0.209
0.1C8
1.20*
O.l-i-i
38.089

* Source:   U.S.  Environmental Protection Agency, 1977b:14.
     Public input to  hazardous  waste management decisionmaking was solicited
via four public  meetings  on hazardous waste management held in Decemoer 1975
(Corson, 1976).  EPA has prepared a synopsis of these meetings (U.S. Environ-
mental Protection Agency,  1976c), and Edelman et al.  (1976) have analyzed the
issues  that  were raised.    In addition,  meetings were held early  in  1977 in
each of the  10 EPA  regions on implementation of the RCRA (U.S. Environmental
Protection  Agency,   1977d).   An  early  study  attempted  to  gauge  likely
attitudes  of  the public  toward the national disposal  site  concept (Lackey,
Jacobs and Stewart,  1973), and is discussed'later.

Other Research

     EPA has encouraged the States to conduct hazardous waste generation and
disposal surveys  as a  preliminary stage toward formulating  hazardous waste
management plans.   Various EPA  publications  directed toward  these ends are
available  (Porter,  1975,  1976;  U.S.  Environmental  Protection Agency,  1977b).
Although a number of  statewide  and some regional and local surveys have been
initiated,   several  have   met  with  only  limited  success  because of  poor
                                     24

-------
response from the  generating  firms.   This difficulty has  been  compounded by
the absence of  an  accepted and easily applied definition of hazardous waste.
On the other  hand,  California has obtained excellent responses  using a care-
fully designed  and strongly  followed-up  survey technique  on a county basis
(Sanders, 1977).

General  Findings on Previous Hazardous Waste Research

     The author has concluded that the following generalizations may be made
about previous hazardous waste research:

     1.    It has emphasized process  wastes (as opposed to post-consumer
          or  post-industrial  wastes)  and  has  not  to  any significant
          extent  considered  hazardous  emissions  to  the  atmosphere:

     2.    It has emphasized toxic and radioactive wastes (as opposed to
          flammable wastes, for example);

     3.    It  has  been  demonstrated  that  precise  and  comprehensive
          definition  of hazardous  waste  is  difficult  and subject  to
          disagreement among authorities;

     4.    It has had  a technological emphasis and  has  not to  any great
          extent attempted to  consider   public  attitudes  toward waste
          management alternatives;

     5.    It has been oriented toward end-of-pipe solutions, and much of
          it was predisposed  toward  the  national disposal  site concept;

     6.    It  has  largely  been   conducted  on an  industry-by-industry,
          waste-by-waste  basis  and  has  shown  that  virtually  every
          industry  has  a  number of  unique  features  with respect  to
          hazardous waste management;

     7.    It has made evaluations largely in terms of direct control and
          disposal  process economics and has generally not  evaluated any
          indirect  costs  (such   as  environmental  costs)  or  social
          impacts,   nor has it considered  the  implications of  differing
          attitudes toward risk-taking;

     8.    Its philosophy of  equity  has  been, "Let the pollJter-pc.y.";

     9.    The studies  have used generalized analysis and have tended to
          specify what their authors regard as the best solution, rather
          than  dealing with  specific  problems  and  displaying  the pros
          and cons of alternative solutions.

     This summary of previous research on hazardous wastes  is not intended as
criticism, but  is presented to identify gaps in our knowledge and pitfalls to
be avoided.   Again,  it must be emphasized that these are general findings to
which there may be certain exceptions.
                                     25

-------
     Thus  to  date,  economic  analysis  has  not  addressed  the  problem  of
selecting socially optimal  hazardous  waste management policies, except in an
extremely general  way  (Talley and  Albrecht, 1974).   No  significant attempt
has  been made to  apply  the techniques  of  cost-benefit  or  risk-benefit
analysis to  hazardous  waste management.   And though  decisionmakers  have not
been insensitive to public  attitudes, no ways have been found to factor them
systematically into"  the  process  of  hazardous waste  policy  selection.   This
report represents a step  toward rectifying these omissions.
                                     26

-------
                                   SECTION 4

          ECONOMIC AND SOCIAL ASPECTS OF HAZARDOUS WASTE MANAGEMENT
COST-BENEFIT AND RISK-BENEFIT ANALYSIS FOR ENVIRONMENTAL
PROBLEMS

Established Applications

     Techniques of cost-benefit and risk-benefit analysis are well  developed,
and have already  been  applied to several  classes of environmental  problems.*
For example, cost-benefit  analysis  has  been applied to  air  and rfater pollu-
tion control programs  (Peskin and Seskin, 1975), while risk-benefit analyst's
has been  used  to compare  alternative means  of  generating electric  power
(Barrager,  Judd, and North, 1976).

     Cost-benefit  analysis  is  usually  used  to determine whether or  not  -a
project or  an  activity  should be undertaken,  which requires that the total
benefits conferred should exceed the total costs involved.   At the same time,
cost-benefit analysis  frequently involves  determining the  optimum  scale of
activity,  i.e.,  the project  scope  at which the net benefit (total  benefits
less total  costs)  is  maximized.   Analysis of this  type  could be appropriate
to deciding whether or not to create a park or to preserve a natural  environ-
ment as a wilderness area.
*  In  cost-benefit analysis,  all  the costs  of a  proposed  actiort,  including
  social and environmental  costs,  are summed and  compared  with  the benefits
  arising from the  action.   Since  costs and especially  benefits, can involve
  effects (such  as  environmental changes)  for which there  is. no established
  marketplace,   values  for   effects  must  frequently  be  imputed from 'Other
  indicators.   (This  is  discussed  with  reference  to  hazardous wastes  in
  Appendix  C.)   The   distinction  between  cost-benefit  and  risk-benefit
  analysis  is   not   clear-cut.   The  term  "risk-benefit  analysis"   is  often
  applied to a  category of  cost-benefit analysis  in which  risks to life and
  health are an  important component of the costs  (e.g.,  National Academy of
  Engineering,  1972).   It would not  be applied to an analysis  in  which the
  risks  were  purely  economic  (e.g.,  where  there  is  construction  cost
  uncertainty  or where  there is  doubt about  the magnitude af  the project
  benefits).  Some  authors  dealing with risks to  life retain the term "cost-
  benefit  analysis"  (e.g.,   National  Academy  of  Sciences, 1977),  whereas
  others use the term "cost-risk-benefit analysis" to suggest that the costs
  include both  conventional  costs  and risk-related ones  (e.g.,  U.S.  Atomic
  Energy Commission, 1974).

                                     27

-------
     In  a  pollution  control  analysis,  however,   the  analyst's  terms  of
reference do not  usually  permit questioning the desirability of the economic
activity that generates the pollutant,  .In many cases the activity is already
in existence.   Hence analysis considers only costs associated with pollution,
and the  conventional  economic approach to pollution control  becomes  tnat of
determining the optimum level of pollution and devising a policy that results
in that  level.  Conceptually (but not in practice)  this  process is straight-
forward  and involves  controlling  the discharge of the pollutant to the level
that minimizes  the total  cost (i.e.,  Ql  in  Figure 3).   In  this  case  it is
implicity assumed  that overall,  the benefits of  the  activity  outweigh  the
costs.

     Risk-benefit  analysis  can  examine  whether  or  not   an activity  tnat
involves risk  should be  undertaken  or allowed.   It could,  for axamole, oe
used to  decide  whether or not a particular toxic substance should be used in
a given application (Provenzano, 1973).  This involves comparing the benefits
of such use with the risks incurred.   In some risk-benefit analyses, however,
the net  benefit of the activity  is  assumed  to  be positive, and the analysis
is used to compare alternative ways of achieving the objective.   For example,
in  assessing  alternative  means of  generating  electric  power,  the  exoected
numbers  of  accidents in  mining and transportation  have been estimated  and
have been  expressed as a cost (U.S.  Atomic Energy Commission,  197J-).   The
effects  of air  pollution  (e.g., sulfur dioxide) on  health and property could
also be  included in the same way,  although in practice there are najor diffi-
culties  involved   in  predicting  the  effect of  airborne pollution on  nuoian
health  (Sagan,  1972;  Goldstein,  1975)..  The  probability  and results  of
accidents  could  also  be  estimated,  as  in the  Rasmussen  report on  major
accidents in nuclear power plants  (Rasmussen, 1975).

     Most risk-benefit assessments use expected values to describe the risks.
By  assigning  probabilities  and  economic  values to  the risks,  an  expected
value  of total  damage or cost  can be  obtained,  and the least cost means of
achieving a  specified objective  can thereby be  determined.   Alternatively,
the expected value approach  can allow the optimum  level  of exposure to risk
(e.g.,   radiation   exposure  from  mammography  [National  Academy  of  Sciences,
1977])  to be determined  in a manner that  is  analogous  to Figure 3.   This is
comparatively simple where  large  numbers of individuals  are exposed to risks
that are statistically well  defined,  but  the  approach does  not  allow  for a
decisionmaker's aversion to risk-taking which could  be particularly important
where  there are low probability risks with major consequences (e.g.,  nuclear
power plant disasters).  (One solution to this difficulty would be to specify
a decisionmaker's utility function to build in a degree of risk aversion—see
Appendix F.)

     Tihansky and  Kibby  (1974) make a conceptual  extension  of  the expected
value  approach by  introducing confidence intervals  and comment  that  a  risk-
averse decisionmaker  might base  his decisions on  values  that  are displaced
from the means.   For example, if  asked to approve the manufacture of a toxic
substance,  he  might choose  the upper  decile for damage costs and  the  lower
decile  for  benefits.   Of course, the difficulty with  this  approach  is to
determine all  the necessary data.
                                     28

-------
                                INCREASING POLLUTION
                        (DECREASING ENVIRONMENTAL Q'UALITY)
        Figure   3.   Conventional  economic  approach  to  pollution control.


     Though the basic approach in cost-benefit and risk-benefit analyses is
to  express  all effects  in terms  of a common measuring  rod (the dollar),
studies rarely  succeed  in  placing  dollar  values  on .a]J the. environmental
effects.   Some  authors  have  endeavored to make up for omissions  by  listing
predictable  impacts,  such  as annual  quantities  of  effluents  and wastes
requiring  disposal.   The difficulties- that may be  encountered in  attempting
to  perform  a  comprehensive cost-benefit  or risk-benefit  analysis are dis-
cussed by  Fischhoff  (1977)  in  an excellent critique of  the  techniques.
                                    29

-------
Application to Hazardous Waste Management

     From an  analytical  viewpoint,  there are significant differences between
the types of  pollutants  and risks discussed above, and those associated with
most hazardous waste  problems.   The conventional economic approach to coT'u-
tion  control  (Figure  3)  arrives at  a  least cost  solution by  changing  the
control  measures  used  so  that  the  quantity  of  pollutant released  to  trie
environment varies.   This  approach  assumes  a single pollutant  (or inaex of
pollution, such as  BOO)  as the independent  variable  (the  abscissa in Figure
3).  Except in cases such as waste treatment for a single industry, hazarccus
waste  management  techniques (landfilling, for example)  usually  involve many
different wastes  that may have1 different  hazardous  attributes.   Henca,  in
many cases,  a variable representing the magnitude of a pollution threat would
be difficult to generate.   In the absence of a suitable index,  analysis would
have to  be  undertaken on  a waste-by-waste  basis and would  need to consider
interactions  between  wastes.  This  would require  numerous data  that would
rarely be available,  and the problem would be complicated by the variaoi"ity
in composition of many hazardous  wastes.

     Further  difficulties   arise  with the  damage function.   It may  not  be
desirable to  use  an expected value or similar measure of damage cost.  There
are two arguments against use of  an expected value; risk aversion has already
been mentioned, and the second is that the public perception of a risk rcay be
more important  than its  true  probability and magnitude.   These  issues  ars
discussed later.

     However,   even  if  expected  values  are  used,   compare  what  would  ce
involved  in  making  a hazardous  waste  risk-benefit  analysis  with  a  risk-
benefit  analysis  of  alternative  means of  generating electric  power (e.g.,
Barrager, Judd, and North,  1976;  U.S. Atomic Energy Commission,  1974).  Most
generation of electric power uses  a limited number of  comparatively uniform
technologies  (e.g., coal,  oil, gas  and certain  nuclear  fuel cycles).   Like-
wise,  the  fuel  extraction  technologies  are of  limited diversity,  and  most
have been established for a sufficient length of time to provide good data on
risks  such   as   occupational   injuries  and  adequate  data   on  potential
environmental   damages  such  as acid  mine drainage.   The  uniformity  of  the
technologies  and  the  magnitude  of  the  resources that might be  committed to
them,  together with the  availability of comparatively good data, can warrant
expenditure of considerable effort to make comparative risk-benefit analyses.
For example,  the  Rasmussen  (1975)  study of  accident risks in  two types  of
nuclear power plants cost some $4 million.

     In contrast to the electric  power situation, hazardous wastes are highly
diverse,  and  while  some  of the treatment and disposal  technologies are com-
paratively  uniform, the  environmental  conditions  associated  with disposal
techniques (such  as precipitation,  soil  and aquifer properties)  are highly
variable.  Hence,   the threats that  hazardous   wastes  pose to  man  and  the
environment can vary considerably with the circumstances.   It will frequently
be difficult  to predict  the probability of  their occurrence, and it may also
be hard  to project  the damage if a  threat  does  materialize.   This diversity
implies that  even  if  feasible  to comprehensively and accurately characterize
the  threats  that might  arise from  hazardous  waste  management  alternatives
would often take considerable resources.

                                     30

-------
     Though detailed  analysis  (such as is used  in  conventional  cost-benefit
and risk-benefit analyses) may  be possible and  justified  for  some hazardous
waste  problems,  there  will  be  many for  which  it is  not.   This  may  occur
because the necessary  data cannot realistically be generated,  cr because the
effort that would  be  required  would be out of proportion to the scope of the
problem  (as  measured  by the  worst  case  damage potential, or the  cost  of
achieving  a  high  level  of  control).  What  is  needed  is  a me:hodology  for
evaluating  hazardous  waste   management  alternatives  that  recognizes  the
principles of risk-benefit analysis  but  that is simple to apply and does not
require extensive  data.   This  study attempts to provide  a  suitable method-
ology.  A  key component  of  the  approach  proposed is  the identification  of
threats  and  the  use of  threat scenarios to  describe the  possible adverse
consequences of hazardous waste management techniques.

THREATS THAT MAY ARISE FROM HAZARDOUS WASTES

     In the analysis  of hazardous waste management alternatives,  the damage
or risk component can be regarded as a series of different threats of adverse
events that may  arise from  the use  of various  control  techniques.  The term
"threat"  is  used  because,  in  most  cases,  there  is some  probability (not
necessarily known) of  the  specified  event occurring.   It may  occur  soon,
later, or  never;  and  if the threat does materialize, the  magnitude of the
effect may also  be uncertain.   For  example, there  is a threat that a lagoon
containing hazardous waste may overflow because of exceptional  rainfall.  The
timing and the  size of the spill (and hence its effects) cannot be forecast,
although in this case they  could quite readily be expressed in probaoilistic
terms.

Identification of Threats

     A threat is always present, but for the threat to materialize, some sort
of  initiating event  is  required.   Many  initiating  events  are  well-defined
specific  incidents such as an  industrial accident  or an  unlikely environ-
mental occurrence  (e.g.,  an  earthquake)  that can trigger a threat mechanism.
However,  a threat may  also  arise  from  inadequate design  or  poor practice,
such  as  the operation of a landfill  in  such a way as to provide no environ-
mental protection  from hazardous leachates.  A broad  (as opposed.to detailed)
list  of initiating  events  is  given  in   Table  4, which  also  lists threat
mechanisms and outcomes.

     Threat mechanisms  define  the nature  of the  threat  itself, and may  be
sequential or  hierarchical  (for  example,  fire  could lead to  explosion,  or
vice  versa).   Outcomes  are  the  end results of  threats.   A threat  may  not
always  carry  through to  result in  an observable  or  measurable  adverse out-
come,  such as  poisoning of some  life forms.   However, even where there is no
outcome  of this type,  there  may well  be a loss  of  some  potential options.
For example,  a  toxic leachate could contaminate an aquifer that is used as a
drinking  water  source.   In  this  case,  the  outcome  would  probably  be
poisoning, and  alternate water  supplies  would  be needed.   If,  however,  the
aquifer  (or  at  least  the part  that was contaminated) was  not  used for any
sort  of  water  supply, no poisoning would  result, but the option o~ using the
aquifer  as a  freshwater source  would be lost  unless  decontamination were


                                     31

-------
        TABLE 4.  ENVIRONMENTAL THREATS IN HAZARDOUS V/ASTE MANAGEMENT
Initiating events:

  Geologic event (e.g., earthquake, erosion, change in aquifer,
    meteorite impact)
  Climatic event (e.g., unusual storm, flood, lightning, etc.)
  In-plant accident
  Transport accident
  Sabotage
  Operational failure or error (man caused)
  Inadequate design or poor practice (including corrosion)

Threat mechanisms:

  Spillage
  Overflow
  Containment failure
  Leaching
  Unintentional  or unwanted mixing
  Unintentional  or unwanted contact
  Fire
  Explosion
  Ground movement or shock waves
  Unintentional  or unwanted emissions (to air)
  Odor
  Vector
  Bioconcentration

Outcomes:

  Destruction of life (man, fauna, flora)
  Destruction of real property
  Poisoning
  Modification of an ecosystem (by changing balance of species)
  Olfactory insult

  Loss of option(s)

Note:  Aesthetic degradation has not been listed as a specific outcome as it
       is an aspect of other outcomes
possible.   Similarly,  an area of  barren ground  could  be contaminated  by  a
spill  without  causing   destruction  of  life,   but  the  contamination  could
preclude various future uses.

     By combining initiating events,  threat mechanisms and outcomes,  a series
of  possible   threats  can  be developed.   For  example,  inadequate  landfill
design results  in  a  leachate  containment failure, which causes contamination


                                     32

-------
of  groundwater  (by mixing),  which leads 'to 1-ivestock poisoning.   Numerous
such threats could  be  identified,  depending on the degree  of  comprehensive-
ness required.   Thus in  the example above, the leachate damage could lead to
chronic or acute poisoning of flora, fauna, man,  etc.

     Figure 4 presents  a morphological  map that demonstrates in a general way
how a threat evolves from some initiating event to some physical outcome.  In
Figure  4  the  threats  mapped  do  not  extend  beyond  initial  outcomes;  thus
subsequent  outcomes such  as bioconcentration  and spread  of  poisoning via
vector  or  the  food chain  are not  included.   Although the  map presented in
Figure 4 is  not  fully  detailed,  it can  be seen  that  there are numerous ways
in  which  threats  could  develop.   Thus  for practical  purposes it will  be
necessary to limit the  number of  threats that are considered.

Common Threats  in Hazardous Waste Management

     Table 5 identifies  some of  the more important categories  of 'threat that
may occur  in hazardous  waste  management and indicates the  waste management
techniques that are likely to pose these threats.   Threat mechanisns are more
fully discussed in  Appendix B, which also includes data on the probabilities
of some initiating events.

ECONOMIC AND SOCIAL EFFECTS OF HAZARDOUS WASTE MANAGEMENT TECHNIQUES

Classification  of Effects

     Use of various hazardous waste management techniques results in economic
and social  effects.  The effects may occur  either  as  a direct result of the
waste  management  techniques  used,  or  they may  occur  via  environmental
impacts.  For hazardous wastes,  many environmental impacts will taka the form
of  threats  as  opposed  to  readily predictable  impacts.   Analysis  of  waste
management  problems usually identifies two major categories of economic and
social  effects—control   costs   and  damages.  This is  consistent with the
cost-benefit approach  to pollution control discussed  earlier.   Though actual
benefits (as opposed to  damages  averted) may sometimes arise from the use of
waste  management  techniques  (as  with  resource  recovery),   these  can  be
accounted for by credits against  control costs.

     The term "damage  costs" has been replaced by  "environmental  costs" and
"social impacts"  for   two  reasons.   First, the term "damage costs"  does not
take  account of differing  viewpoints.   If, for  example,  a  waste management
scheme  involves  the construction  of a dam on a  river, moving water recrea-
tionists (e.g., kayakers)  will perceive this as a damaging effect, but still
water  recreationists   (e.g.,  water-skiers)  will  perceive  it   as  a benefit.
Second, although most  authorities  agree that it  is conceptually possible to
attribute a  dollar  value to any effect  (see  Appendix  C),  the practical dif-
ficulties  can  be  overwhelming   (Organisation  for Economic  Co-operation and
Development, 1974).  Hence some  effects may  be  more  appropriately described
than given a dollar value.

     In this report, the term "environmental costs" is  used to describe the
direct  results of  the  environmental threats already discussed, translated as


                                     33

-------
   INITIATING EVENTSh-
                                          - THREAT MECHANISMS •
                                                                                          -| OUTCOMES
                                                                                             DESTRUCTION
                                                                                             OF PROPERTY
                   UNINTEN
                   TIONAL
                   MIXING
                                                  GROUND
                                                 MOVEMENT,
                                                  OR SHOCK
                                                   WAVES
                                                                                            DESTRUCTION
OPERATIONAL
                                                              UNINTENTIONAL
                                                               OR UNWANTED
                                                                 CONTACT
                                                                                             MODIFICATION
 GEOLOGIC
   EVENT
                                                                                           OF ECOSYSTEM
                                                                NINTENTIONAL
                                                               OR UNWANTED
                                                                 MIXING
CLIMATIC
 EVENT
INADEQUATE
 DESIGN, OR
   POOR
  PRACTICE
ODOR


OLFACTORY
INSULT
                           Figure 4.  Morphological  map of environmental

-------
                       TABLE 5.  MAJOR THREATS THAT MAY OCCUR IN HAZARDOUS WASTE MANAGEMENT
                          Threat category
                                                                Hazardous  waste  management technique
                                                               	to  which threat may  apply	
CJ
in
1.  In-plant accident (spill,  fire,  explosion)  leading
    to acute poisoning,  destruction  of property,  etc.

2.  Systematic in-plant  exposure resulting in chronic
    poisoning of workers

3.  Waste handling accident (spill,  fire,  explosion)
    during waste collection or disposal  operations,
    leading to acute poisoning, destruction of  property,
    etc.
4.  Transportation accident leading  to acute poisoning,
    destruction of property, water pollution, etc.
5.  Failure of treatment process resulting in waste
    that retains hazardous properties or acquires new
    hazardous attributes

6.  Groundwater contamination  via leachate
       7.  Surface or groundwater contamination by failure
           of containment
       8.  Surface water contamination via runoff or
           indirectly from seepage of polluted groundwater
       9.  Modification of marine ecosystem by presence of
           waste
      10.  Direct groundwater contamination by presence
           of waste
                                                                            All


                                                                            All

                                                                            All
                                                                            Wherever waste is transported
                                                                            from site of generation

                                                                            Any treatment process
Lagoon ing, land application,
landfilling, mine disposal,
possibly engineered storage

Lagooning, landfilling, mine
disposal, engineered
storage
Lagooning, land application,
landfilling
Ocean (lumping

Deep well injection

-------
far as practicable Into economic terms.  "Social Impacts" Is used to describe
all remaining  effects, including  those that are  essentially psychological.
Thus in general,  environmental  costs arise from the threat of physical envi-
ronmental   degradation  leading   to  some  specific  potential  economic  loss.
Social impacts, on  the other hand,  need not involve any physical damage, are
more  likely  to involve differing viewpoints, and will  more  frequently defy
quantification in dollar  terms.   This terminology was chosen for convenience
and may not  coincide with that  used by  other sources.   Furthermore, in some
cases,  allocation  to  an  environmental  cost  or  a  social  impact  may  be
arbitrary.   Nevertheless,  maintaining this division may help to ensure that
all possible effects are considered.

Exclusion of Secondary Effects

     This  study  does  not  generally address secondary effects,  v/hicn  n tne
economic  context are  multiplier effects  that  reflect the  fact  that  one
person's  expenditures constitute  another's  income.   Secondary  benefits  are
usually disregarded  in cost-benefit and  cost-effectiveness  analysis because
under  conditions  of  full  employment of  resources,  the  resources used  in
secondary  activities  would be employed elsewhere  in equivalently productive
activities  (Herfindahl and  Kneese,  1974).   Clearly, this  argument  is  not
always valid.  Regional economists who are faced with a contracting economy
and immobile  factors  of production  (or conversely with a boom-town situation
and shortages) may  be vitally interested in economic multiplier effects; and
such effects could be  important  in some hazardous waste management decisions.
However,  the author  supports  the view expressed  by Maass  (1966)  that only
where  secondary  effects  (such  as income redistribution) are made  a part of
project  objectives   (and   thereby cease  to  be secondary),  should they  be
included  in  evaluations.    Hence in this  analysis  the costs considered  are
restricted  to those  conventionally  used in assessing  economic efficiency,
except that environmental  externalities (sometimes called "spillovers") will,
of course, be included.

Types of Control Costs

     Control costs associated with any approach to hazardous waste management
may include the following:

     1.   Generator's  costs,  or  those  incurred  by  the  firm that gen-
          erates or may generate the hazardous waste.  These include the
          costs  of  treatment,   transport,  and  disposal  of  the waste,
          which  may   be   performed  by  the  firm  itself  or  by  its
          contractors.   They  'also   include  the  generator's  relevant
          administrative,  legal,  and  research  and  development costs.

     2.   Administrative costs,  or administrative  and enforcement costs
           incurred  by governments or by any other body that oversees or
          monitors hazardous waste management.
                                      36

-------
     3.   Social control costs,  or costs  that reflect  the  differences
          betweenthedollar  costs  actually  incurred by  the  waste
          generator  for various  services,  and the true  cost  to society
          of these services.  Such costs might arise, for example, where
          a government  provides  a waste disposal facility free  or at a
          subsidized charge.   (These  costs  do not include an accounting
          of external environmental costs,  which are covered elsewhere.)

     In general, all  control  costs can quite  readily  be expressed in dollar
terms,  although for  some  there  can  be  problems  in  deciding  on  the most
appropriate  method of  valuation.  For  example, should a  landfill  site  be
valued at historic cost, at replacement cost, or at opportunity cost with the
land  used  for another  purpose?   If  the  latter,  what use  (e.g.,  fanning  or
housing) should  be the  basis  of valuation?  Control costs are  discussed  in
more detail in Appendix C.

Types of Environmental Costs and Social Impacts

     Several sources provide broad discussions of the  evaluation of environ-
mental  damage,   including  both  environmental  costs and  social  impacts, as
defined  in this  study  (Maler  and Wyzga,  1976;  Organisation  for  Economic
Cooperation  and Development,  1974;  Saunders,  1976;  Bishop  and  Cicchetti,
1975).  In  general,  empirical  studies that involve pervasive pollutants (air
pollution and radiation) appear to be the most advanced,  possibly because the
sources  and  receptors   are comparatively  easy  to  identify.   During  this
research,  no  studies were  found  that addressed  the empirical  valuation  of
damages associated with solid or liquid hazardous wastes  in a cost-benefit or
risk-benefit  framework,  and  only one was  found that did  so  for hazardous
emissions to air (Moll et al.,  1975).   Talley and Albrecht (1974) have made a
preliminary analysis of the economics of hazardous waste  control in wiich the
emphasis was  theoretical,  and at  the other end of  the  scale,  individual
waste-related  damage  incidents   have  been  investigated  (see  Appendix  8).
Although  a  systematic   analysis  of  the  various environmental   and  social
impacts of hazardous waste control techniques has not been made, the types  of
damage that may  occur  and many of the effects are common to other aspects  of
environmental and  cost-benefit,analyses.   Hence methods of valuation  may  be
derived from these sources.

     The environmental  impacts that  most  frequently arise  from  the  -use  of
hazardous  waste  management techniques can conveniently  be divided into five
categories for valuation purposes:

     1.   Destruction of,, or.damage to man-made structures;

     2.   Damage to human life and health;

     3.   Destruction  of,  or  damage  to  animals, vegetation and land
          ecosystems,  and  kills   of   fish  and  other  aquatic   life  in
          surface  waters  (including  the   impacts   of   ocean  dumping);
                                     37

-------
     4.   Changes in property values;

     5.   Aesthetic factors and option values.

     The  first  three categories  (which  mainly  lead  to  environmental  costs)
relate to the physical impacts that can arise from the threat mechanisms.  An
additional physical  impact,  the  modification of climate, can arise frcm many
sources  (Saunders,  1976),  but  is  unlikely  to  be significant  in hazardous
waste management.

     Changes in  property  values  may reflect actual damages (usually noise or
air  pollution),  or   they  may   be  essentially  psychological  in  origin,
reflecting aesthetic factors.  Thus changes in property values can constitute
environmental cost; and/or social impacts, as defined here.
     The  final  category is  intended  to cover  all  remaining  social
(i.e.,  those  that do  not  involve direct  economic  costs).   It  includes the
aesthetic value of  the environment,  the existence value  that stems from the
knowledge  that something  exists or  is being  conserved, and option  values
associated with risk aversion and uncertainty.
     Note that  a  single  environmental  impact can give  rise  to ^lore than one
of  the  above categories  of  effect.   For example, persistent  po". ". ution of a
river  could cause  fish   kills,  changes  in  adjacent  property values,  and a
reduction  in tne  aesthetic   appeal  of  the river  to bystanders.   Thus the
distinctions drawn  above among  the  five categories  of  effects  stem largely
from the  methods, described  in  Appendix C, that can be  used for valuation.
Since there  can be  overlaps  between the categories  (e.g.,  between damage to
property, property  value, and  aesthetic value) care must be  taken  to avoid
double counting when evaluating costs and impacts.

THE PARTIES-AT- INTEREST IN HAZARDOUS WASTE MANAGEMENT

The Concept of  Parti es-at- Interest

     In analyzing the effects of using a hazardous waste management technique
or  an  approach  to  hazardous  waste  control,  the  viewpoint of  man  is  of
ultimate  concern; i.e.,   What effects  does  the use  of  a given  approach or
technique have  on mankind?   As has been shown,  these effects can range from
those  that  are direct  and  straightforward, such  as dollar  costs  actually
incurred in waste disposal, to those that are highly indirect, as for example
the  value  that  many  individuals -place  on  the preservation  of a  natural
undisturbed  ecosystem,  which  causes  them  to  associate  a  cost  with  the
modification of that  system  by the introduction  of  waste materials.   Corre-
spondingly, the distribution of these effects can vary from those that affect
a single firm to  those that impinge upon the general public.

     This  research  is oriented  toward the  needs  of decisionmakers  who may
have  multiple  criteria,  including  social  and  political   goals,   for  the
acceptability of  a hazardous  waste  management  plan.   Hence  in  this study,
recognition  of  different viewpoints  and a means  of  taking  account of social
interactions is desirable.  This can be achieved by grouping individuals into


                                     38

-------
parties-at-interest.   Each party-at-interest  constitutes  a class or group of
individuals or enterprises that  can reasonably be expected  to  have a common
interest  and  viewpoint  on  the  outcome  of  any  particular  plan  or  policy
alternative  (Gilmore  et  al.,  1971;  also see  Royston and  Perkowski,  1S75).

     For manageability,  the parties-at-interest must be limited to those that
are  quite  strongly   affected  by  one  or  more  of   the  alternatives  being
evaluated.   Clearly,  the  grouping  of  numerous   individual  viewpoints  and
attitudes into a  limited  number  of parties-at-interest constitutes something
of a blunt  instrument.   However, it does permit  the  sociological aspects of
alternative plans or  policies  to be considered, and by being able to observe
the distribution of effects  among  different parties-at-interest, it provides
a useful foundation for  considering equity.

Identification of Parties-at-interest in Hazardous Waste Management

     The parties-at-interest in  hazardous waste management may be determined
from  the  effects  that  might arise  from  the use  of  the various  control
techniques.   These effects  fall  into two general categories:   Effects Csuch.
as  control  costs) that  arise  directly  from the use  of a  hazardous waste
management technique,  and effects that arise via environmental impacts.- • Note
that  an environmental impact  need not  actually  occur to  be real  from  the
viewpoint  of  the  analysis.   Both  very  unlikely   threats  and  completely
imaginary effects may be  of  importance.   For  example,  residents near  a pro-
posed landfill may fear  that their property values would be depressed by the
proximity  of  the  landfill.    Even if  this  fear  ultimately  proves  to  be
incorrect, it  is  a real  fear to  the residents, who will therefore respond to
it in some way, and hence it is an  effect to be considered.

     Some economic effects,  and  hence parties-at-interest, arise indirectly.
For example,  in  considering the effects of  using different  hazardous waste
management  techniques  within  a  given  industry,  all  the  firms that use  a
particular hazardous waste generating  process could constitute one party-at-
interest;  firms  using another  process  that  does  not generate  a  hazardous
waste could constitute  a different  party-at-interest.   In  this  case,  the
effect  would  occur through  changes  in the competitive situation within  the
industry  reflecting different  hazardous waste disposal  costs  to the  firms,
and leading to changes in production costs.

     The parties-at-interest in  a  given  hazardous waste management situation
may be deduced by examining the-effects of every technique that,miglv: be used
to manage the  wastes  (including  no change from present practice),  'olitical
officials and  administrators in  environmentally oriented government agencies
are  also included, as  they will  want  to  minimize  dissension  among their
constituencies and facilitate effective administration.

     Examination of the  approaches  that  can be used  to place a value  on the
effects   (Appendix C)  may  be  helpful   in  identifying  parties-at-interest.
Gilmore   et  al.  (1971) list  four groups of parties-at-interest,  as follows:

     1.    Parties internal to the affected industry (e.g.  owners, stock-
          holders, management,  employees and their unions);


                                     39

-------
     2.    Suppliers and customers of an affected Industry (e.g.,  venders
          of materials and services, including financing, insurance,  and
          intermediate and final  consumers);

     1.    Government  (at  different  levels   and  in  different  roles,
          including   legislators,   executors,   adjudicators,   taxers,
          regulators,  and  those  concerned   with  economic  stability,
          social welfare, and national  security);

     4.    Affected  bystanders  (e.g.,  those   concerned with  resources,
          wildlife,  recreation   potential,  and  aesthetic  effects,  and
          investors, employees,  residents,  and  other  property  owners).

     Gilmore  et al. included parties-at-interest associated with  secondary
economic  effects  (suppliers  and  customers   of  an  affected industry).   As
already  indicated,  the  methodology  proposed  in  this  study  follows  the
commonly accepted  approach of not  evaluating secondary effects.   However, in
cases where  secondary  economic  effects  are   particularly  significant,  these
could  lead  to  some important parties-at-interest.   Consideration  of  these
parties-at-interest may help in  understanding the sociology of the situation,
even if the  associated  secondary economic effects are  not  directly  included
in any cost-benefit calculations.

Costs and Parties-at-Interest for Hazardous Waste Management Techniques

     Table 6  summarizes  the  most important environmental threats,  costs (and
impacts), and parties-at-interest  associated  with hazardous waste management
techniques.    In  specific situations,  additional threats, costs  and  parties-
at-interest may  be important.    Table  6  also includes disposal  to  the  sewer
and to surface waterways.  Though these would not normally be options legally
available to  the  hazardous waste generator,  they  are  of interest  since they
might be  used illicitly  or  for  wastes that  unexpectedly  remained hazardous
after some form of treatment.

     Certain  costs  and  parties-at-interest  are  present for  every available
technique.  For example, the firms generating the waste will always have some
disposal  costs,  and their managers and workers  will  always be parties-at-
interest.   This does not imply that the costs or the posture of the firm will
be  the  same  for  each  technique.   Similarly,  it  can  be assumed  that  local
government and  environmental .agencies  will  be interested  in  every technique
that might be used.   The term "local," in this  context, means having juris-
diction over the location at which the technique is employed, hence different
local officials may be  parties-at-interest  to different techniques.   Similar
considerations  apply every  time   the  term   "local"  is used   (e.g.,  local
residents, local  property owners,   etc.).   These  parties-at-interest are the
individuals  that are  close  enough to  the  location  of the  technique  under
consideration to potentially be  affected by  it.

     The  potential  for  freshwater contamination  occurs  with  many of  the
techniques, giving rise to associated costs.   Where water contamination costs
are  listed  in Table 6,  these costs  can  include  the  contamination  of  water
supplies (from surface or groundwater sources) for human or agricultural use,


                                     40

-------
TABLE 6   ENVIRONMENTAL THREATS.  COSTS.  AND PARTIES-AT-INTEREST FOR HAZARDOUS WASTE MANAGEMENT TECHNIQUES
Technique
All techniques
All techniques
Involving
offslte
activities
Change In
wiste
streams*
Resource
recovery
Treatment to
reduce
hazard
-Physical
-Chemical
•Biological
-Thenul
-Encapsulation
Land application
EnvlronMnlal threats
Suireplillous dumping or
accldentdl discharge to
land, sewer, or waterway
lii-plant accidents mvolvfng
fire or explosion
Acute and chronic poisoning
Suirepttlious dumping
Accidental spillage
(to land or water)


IrealKnt may fall
leaving waste still
hazardous (most
likely with biological
treatment)


Air pollution

Crop take-up of toxic
sle«nts
Soil sterilization
teaching/ nip-off
Odor
Pailies-at-interetl
Waste generators-
managemcnl.
Waste genorators-
workers.
Competing firms using
different process,
Waste disposal Industry,
local political officials.
Local environmental
officials,
Water supply authorities,
Environmentalist*
Waste transporters,
Residents along transport
corridors
Secondary materials
Industry >
Secondary materials
industry.
Virgin materials
suppliers
Fishermen (where effluent
is discharged)
Chemical suppliers

Lotal lesldenls/woikeis
local property owners

Farmers
local water supply users
Residenls/woiLeis/
properly owners
adjacent to land
F I shermen
Major costs, etc
Capital and operating
costs for waste
disposal (internal
costs to generator)
lianspoit cost


Additional capital and
operating costs for
necctsaiy processes,
less value of any
resources recovered




Aesthetics
Poisoning via food
chain
Water contamination
Conmenls


Hay facilitate
lesourco recovery
Resource recovery
may have adverse
effect on energy
usage

Increased energy
usage likely

Possibility ul
energy lecovciy

Any null I ems. etc ,
present way enhance
crop growth
 The lhre«u,
•ml co&u lUtnUMeU turn lM» point on are In aUdilluu lu ihu^e
                                                                       *bu«ef

-------
                                                                            TABLE  6 (Continued)
ro
Technique
Laodfllling
Nine disposal
Environmental threats
Leaching/ run-off
Odor
Vectors?
Leakage to grounowaler
Part les-at- interest
Local water supply use is
Local residents/workers/
property owners
Fishermen
Holders of adjacent
•inera) rights
Hajoi coils, etc
Water contamination
Aesthetics
Gi ounduater conldiin
                       Ldgouning           teaching
                         (for storage      Overflow
                         or evaporation)   Odor and air pollution
                       Deep well
                         injection
                       Ocean duping
                       Engineered
                         storage


                       Space disposal
                       Discharge to
                         sewei
                       Discharge to
                         waterway
Groundwater contamination
Land movement and
  consequent property dapage
Modification of marine
  ecosystem
                                           Containment  failure
Contamination of space
Launch tdiluie consequences
Waste uy destroy biological
  treatment colonies
Waste may cause sewage
  iludye lo become ha/ardous
Slugs of waste may overload
  system
Waste may destroy
  aquatic life
Residents/workers/
  property owners
  adjacent lo site
Local water supply users
Fowl hunters
Farmers?

Authorities concerned
  with resource usage
Local residents/properly
  owners
Scientists

Ocean fishermen
Ocean recreationallsts
Coastal iecreation-
  related industry
Local residents/workers/
  property owners
Scientists
Other nations, national
  politicians

Local residents
local water supply users/
Scientists
Other nations, national
  politicians

Sewer authorities
I isheiiwn
Other water useis
Local lecredliun-ieldlcd
  industry
Re!* Idtinls/hur km s/|n uperty
  uuners adjacent lu
  waterway
local walei  supply u..eis/
Silinlltts?
                                                                Wattir contdmlndtiun
                                                                Aesthetics
Groundudtur contamination
Property damage
Reduced or Inedible
  fish catch
Reduced tourist income
Aesthetics
Possible International
  political strife
Maintenance after
  disposal is completed
Water contamination

International political
  strife
Unknown scientific costs

Malfunction of sewage
  woiks and consequent
  damage
Aesthetics
Reduced n*cioatloiul
  o|>poi tunnies
                                                                                                                                             Comments

                                                                                                                                        Need to consider
                                                                                                                                          opportunity cost
                                                                                                                                          of land
                              Need  lo  consider
                                opportunity cost
                                of  mine

                              Need  to  consider
                                opporlunily cost
                                of  land
fhere may be an
  opportunity cost to
  using the aquifer
                                                                                              Requites purpeliidl
                                                                                                muniloring  and
                                                                                                maintenance
These two actions die
  nut normally leyiti-
  mdle hdtaidous wdste
  mandgemenl techni-
  ques, but die in-
  cluded here becdtise
  a treatment piocess
  ••iy fall lesullnuj  in
  a hazardous el fluent,
  01  they may be used
  coveilly

-------
and the  pollution of other  surface waters.   The cost, involved  in 'tha first
case can  include  installing  replacement sources of drinking and agricultural
water, corrosion  and  materials  damage costs (to pipes and appliances), fore-
going  the use  of water  or  continuing to  use the  contaminated  water  and
accepting  lower crop yields and values  (for agricultural  irrigation),  and
aesthetic  costs.   (Persons  who  use  their  own well  water may  consider it
aesthetically more satisfying than piped water, and hence suffer an aesthetic
cost  if  they are  forced to  obtain water from  public systems.)   The costs
associated with  the pollution  of surface sources not  used  for  water supply
can  include  fish  kills,  devalued  recreational  opportunities,  environmental
aesthetic costs, and loss of future options.

     Cleanup or mitigation costs could be involved where there is any form of
contamination.    These costs  can be  particularly expensive  for groundwater
sources which  cannot feasibly  be treated after withdrawal.   In this avent,
normal procedure  is to  counterpump wells drilled to  intersect  the plume of
contaminated water, and  then to- dispose of  this water.   Even this procedure
is  not always  entirely  successful  (U.S. -Environmental   Projection  Agency,
1977c; Miller,  1974).

THE SOCIOECONOMIC INTERACTION PROCESS

The Interaction Model

     This  section presents  a  simple conceptual  model  of  the  interactions
between the  technical (i.e.,  physical) aspects and  the  economic and social
aspects  of  hazardous waste  management.   The purpose  of the  model  is to
enhance understanding of the relationships between hazardous waste management
policy, what physically  happens to the wastes, and the effects that this has
on society.  The  model  forms the basis of the evaluation procedure described
in  Section  5,  and  is  inevitably  a  compromise  between  the  desire  for
simplicity  in  analysis,  and  the complexities  that may  be  encountered in
real-life situations.

     The  model  is  illustrated  in  Figure  5.   It  is  divided  into  three
sections,  or  levels:  The  policy level, the technical  level,  and the socio-
economic  level.   The policy  level  is concerned with the philosopny  of  how
hazardous  wastes  are to  be  managed.   Decisions  at  the  policy  level  are
largely  responsible for  determining what  goes  on  at  the technical level.
which  deals  with what physically happens to  the wastes  and  to  the environ-
raent.   In  turn,  actions  at  the   technical  level  have  effects  at  the
socioeconomic level  (i.e.,  on  society).   There is  feedback  from the socio-
economic  level  to the policy level  through the technical level.

     The  elements  in  the model  and the  linkages  between the  elements---are
briefly described below.   Most  of these elements are discussed in more.detail
elsewhere.
Policy Objectives—
     Policy objecti
considered to be an exogenous input to the model (see Section 5).
Policy objectives,  dealing  with  normative  (judgmental)  issues,  are
         b<
                                     43

-------
POLICY
LEVEL
      POLICY  \
    OBJECTIVES/
\

APPROACHES
to hazardous
waste management
          TECHNIQUES
       for the management
       of hazardous waste
                   OUTCOMES
                 (what physically
                    happens)
PHYSICAL
Oft TECHNICAL
LEVEL
                                                            WASTE
                                                         DISPOSITION
ENVIRONMENTAL
    IMPACTS
   (threats, etc.)
        ECONOMIC AND
        SOCIAL EFFECTS
          of techniques
                    RESPONSES
                   of the parties-
                     at - interest
5OCJOECONOM/C
LEVEL
                            PARTIES-
                          AT-INTEREST
                         (those affected) /
                   COSTS & IMPACTS

                    - control costs
                    - environmental costs
                    - social impacts
      Figure 5.   Interaction model  for  hazardous waste management.
                                   44

-------
Approaches to Hazardous Waste Management—
     The  approaches  to hazardous  waste management  represent  strategies for
the  control  of hazardous  waste that  are  consistent with  the policy ob'jiec-
tives.  Approaches may favor the use  of certain  techniques (see Section 5).

Techniques for the Control of Hazardous Waste--
     Techniques are  the physical  methods (e.g., treatment, landfilling) that
may  be  used  to manage or control hazardous waste (see Section 3 and Appendix
A).   They may include environmentally  unacceptable  techniques,  such as sur-
reptitious dumping.  The use of a given technique results directly in control
costs  (one  of the economic and  social  effects)  and also  causes  or has the
potential to cause environmental impacts.

Environmental Impacts--
     Environmental impacts  are  the physical  effects, or  potential  effects,
that  could  arise  from  the use of  various- hazardous waste- management tech-
niques.   They  occur  largely in the form of threats  (discussed early in this
section  and  in Appendix  B).   In addition, pervasive effects--that rebate-to•
resource  use may  be  of  interest.   Thus  although  the economic  aspects  of
energy  or materials  consumption  attributable to  the use  of a technique are
accounted  for within  the  control  costs (including  any  credits  for resource
recovery), these  topics may be of specific interest and call  for individual
treatment.  The same argument applies to land use, which is another aspect of
resource  use.   Note  that  if the  emphasis  is on  conservation of resources,
some base case will be needed for comparison.

Economic and Social Effects of the Technique—
     The economic and social effects are the effects that the techniques have
on man.   These  effects give rise to  costs and impacts  (i.e.,  control costs,
environmental costs, and  social  impacts, discussed earlier in this section).
In addition,  man  may have a special  interest in  certain aspects of resource
use, as discussed above.

The Parties-at-Interest—
     The  economic  and  social  effects •will  affect,  different  groups  of
individuals or enterprises  in  different ways.  Each group that.Is relatively
homogeneous in terms of its interests and attitudes and in the'way that it> i-s.•
affected  by  the economic, and social  effects  of the  techniques constitutes, a
party-at-interest (as discussed earlier in this section).

Responses of the Parti es-at-Interest—
     The  various  parties-at-interest  will  respond to the economic and social
effects  in  ways that  will  be determined  by  their  interests  and attitudes.
Their  responses  could  include opposing  or  supporting  a scheme  or policy,
choosing  a waste  disposal  technique,  changing business  activities, moving to
another  location  to  avoid an adverse effect,  etc.   (Responses are discussed
later in this section.)

Outcomes—
     The  outcomes represent what physically  happens  in terms  of hazardous
waste management, allowing  for  all  the interactions  and  linkages  that exist
in practice.   Thus prediction of outcomes involves determining what is likely


                                     45

-------
to happen to the various hazardous wastes. I.e., how much will be disoosed of
by each  technique,  and the  extent  to which waste generation  will  change as
the result  of using  any  particular approach to  hazardous  waste management.
The  outcomes  also  include  the  environmental   effects  that occur  or  are
threatened.

     There is feedback from the outcomes to the  approaches to hazardous .vasta
management.    When  a  decisionmaker  predicts  or  observes the outcomes  that
result from  a given  approach  to  hazardous waste management,  he  may  wisn to
modify that approach to change the outcomes and  the associated economic costs
and  social   impacts.    (Outcomes  and  the complete  interaction   process  are
discussed in Section 5.)

Attitudes Toward Hazardous Wastes and Their Management

     To be able  to  predict responses of the parties-at-interest and ^.anca to
determine  likely outcomes  of  approaches  to  hazardous waste  management,  it
would be  desirable  to  know something  about  the attitudes  that individuals
have toward hazardous  wastes and the environment.

     Only one study has specifically addressed  public  attitudes toward haz-
ardous waste management  facilities.   This extensive  study surveyed  2;th  a
random sample  and selected influential  respondents (e.g., newspaper editors)
in  10 U.S.  counties   that were  considered  feasible locations  for national
disposal  sites  (Lackey, Jacobs,  and Stewart,  1973).   The study reported gen-
erally favorable attitudes  toward  the national  disposal   sits  conceot  and
location of  such a  site in all counties surveyed.  These favorable attitudes
appeared  to  relate   to the  beliefs  that a disposal  site  would conserve
material   resources  and  result   in  a  strong local  economy  (i.e.,  provide
employment) (Lackey, Jacobs, and Stewart, 1973).  However, during the testing
of  the  questionnaire,  it was  found  that the  term "disposal"  had negative
connotations  for the  respondents,  and accordingly national  disposal  sites
were  described as  "regional processing  facilities"  in  the  survey  instrument
(Lackey,  Jacobs, and  Stewart,  1973).   This situation does raise the question
as to whether or not  the  respondents  really  understood what they were being
asked.  If  not,  the attitudes reported  by Lackey,  Jacobs,  and Stewart might
change when the  issue  becomes clearer.

     It was  initially  hoped  that a study of attitudes toward nuclear power
(and  nuclear wastes)  and  toward  hazardous waste generating industries (such
as  the  chemical industry)  would  provide useful  information  on  attitudes
toward hazardous wastes.  The intention was to use the  chemical industry as a
proxy for a hazardous  waste management  facility and to  draw parallels oetween
nuclear  power,  with  its   radioactive waste  disposal problem, and  nonradio-
a'ctive  hazardous waste  disposal.   Other  than  for  nuclear power,  little
relevant material on  attitudes toward  waste  generating industries  was found
during this research.

     Though  there  have been  many  attitudinal   studies  relating  to  nuclear
power (several of which have included the question of nuclear wastes), it was
concluded that  attitudes  toward nuclear power would be of very limited value
in  predicting  attitudes   toward  nonradioactive  hazardous  wastes  and  their


                                     46

-------
management.  This  is  because  several  studies have  suggested  that the public
associates the possible  discharge  of  radioactivity from nuclear power plants
and wastes with the effect of nuclear weapons (Pahner, 1976; Louis Harris and
Associates, Inc.,  1976;  Rappeport  and Labaw, 1975; Slovic  and  Fischhoff,  in
press).   Though  hazardous  wastes  comprising  obsolete chemical  warfare and
ordnance materials could  to some extent pose an analogous type of thre.it,  in
general, reactions to nuclear facilities appear too extreme to be appropriate
for  nonnuclear  hazardous wastes.  However,  one  finding  about  nuclear power
plants  may be  applicable  to  some  hazardous  waste   management  facilities:
Familiarity can  evidently  breed  acceptance.   Those  living  near  a nuclear
reactor perceived  it  as  safer than those living farther away (Maderthaner et
al., 1975; also see Klein, 1978).  (However, it is possible that this finding
may  reflect an element  of self selection, with those who are concerned about
the risks tending to move away from nuclear power plant sites.)

     As a  result  of  the paucity of data that could provide specific guidance
on attitudes toward hazardous wastes,  a broad survey of research on attitudes
to  the  environment  was  undertaken.   This  survey   provided  some  general
guidance  on  the  attitudes  and  priorities  of  the  general public  and  some
special  interest  groups  (i.e.,  parties-at-interest)  toward pollution  pro-
blems,  and it  is summarized  in Appendix D.   Based  on these  findings,  the
EPA-sponsored meetings on hazardous waste management (see Corson, 1976;  U.S.
Environmental   Protection Agency,  1976c;  Edelman  et  al.,  1976),   and  the
material discussed  above,  the  following list  presents  some  generalizations
about  the  likely  attitudes  and  behavior  of various  parties-at-interest.

      1.  Firms  desire  to  minimize  their internal  costs,  including
          management costs.

      2.  Wastes  are  a  nuisance to  manufacturing  firms, which in most
          cases will  not devote much effort  to  their  disposal  unless
          disposal  represents a  significant  cost  to  them,  or  unless
          there is a  significant risk of public  opposition to  the firm
          or  its   products  because  of  its waste  disposal  practices.

      3.  In selecting  a waste  disposal  technique,  firms  will  tend to
          favor those that  enable  them  to dispose of the responsibility
          for the waste along with the waste.

      4.  Large  firms   are   the  most   likely  to  be  environmentally
          responsible,  as  they  have  high  public  visibility.   SmaTer
          firms are more  variable  in  their concern for the environment.

      5.  Workers are concerned  with  their own physical safety and with
          security of employment.  Often,  however,  the latter outweighs
          the former in determining their actions.

      6.  Local government  and environmental officials  prefer  to adopt
          policies that  minimize the risk of  adverse  incidents  (i.e.,
          they are strongly risk averse).
                                     47

-------
      7.   Wastes are politically  negative,  and local politicians prefer
          them to go elsewhere.

      8.   Residents are concerned with  property values.  They fear that
          nearby  waste  processing   or  disposal   sites  will  depress
          property values.

      9.   Residents  are  generally  uneasy  about  wastes.   They  often
          object  strenuously  to  wastes   from  another  jurisdiction,
          especially another State.

     10.   Residents have  some interest  in  local employment,  tax base,
          etc.; but the strength of this interest tends to depend on the
          employment  history  in  the   area.    Local  politicians  and
          businessmen often have strong interests in these areas.

     11.   Environmentalists wish to minimize all environmental  risks ana
          tend to  resist  change, with  only limited  concern  for costs.

     12.   Environmentalists   exhibit   high   existence   values   (see
          Appendix C),  and may claim that no compensation would be great
          enough to justify some adverse environmental impacts.

     13.   The  public   has   become   cautious  about  new  technologies,
          especially those  that  they  do  not  understand.   Establisned
          technologies  are more acceptable (hence,  the chemical industry
          is  less  threatening  than   nuclear  power).   Public  credulity
          toward scientific expertise is declining.

     14.   In some cases,  those  close to a facility that is perceived to
          be hazardous  are  less  concerned about it  than  those that are
          somewhat farther away.

     15.   The  public  favors conservation and  recycling.   Most (but not
          all) accept the  need  to dispose of some wastes.  However, few
          individuals  are  prepared  to  go  to  great  lengths  to promota
          their ideals.

Responses of the Parties-at-Interest

     The responses of the parties-at-interest are very closely related to the
economic and  social  effects that arise from the use of the various hazardous
waste control techniques.   Although human behavior is not always predictable,
it may  be possible and useful  to identify  likely responses  of the parties-
at-interest to  the  various  economic  and social effects that they experience.
The  views  expressed  in this  section  are based  on the  material  discussed
above,  the  author's  experiences  with  manufacturing  industries,  and from
discussions with individuals who are involved with hazardous waste management
in the western and northwestern United States.
                                     48

-------
Responses to Positive Effects—
     Many  of   the   impacts  that  would  be  regarded  as  positive  (i.e.,
beneficial)  by  the  parties-at-interest  are  comparatively  weak  and  are
unlikely to elicit  an active response.  For  example,  assume that it is pro-
posed that one  class of firm be  required  to  use a costly process to dispose
of  its   hazardous   waste   (a  strong  negative impact  on  these  generators).
Competing firms that use a different process that does not generate hazardous
waste will perceive a mild positive  impact (since  the competitive situation
will be  changed in  their favor).  However, these  competing firms are likely
to  be  passive  in taking  advantage of this situation  and  probably would not
publicly support the proposal,  partly out of fear that they might be next to
be  so  regulated.   On  the other  hand,  the  hazardous waste  generators are
likely  to  protest  the  proposal (on  the grounds  that it will  weaken   their
competitive position) and probably will exert whatever pressure that they can
to  get  it  modified.  This type of behavior was  recently demonstrated in the
U.S. steel industry.   Several  plants had difficulty in meeting air and water
effluent standards  and asked  that variances  be permitted  on  the grounds of
economic  hardship.    On  the  other  hand* -firms  whose  plants   had  little
difficulty  in  meeting  these•• standards  remained  comparatively quiet on the
subject.

     Perhaps the only category of positive impact that is likely to trigger a
strong response  is  that of an impact on an environmentalist.  Environmental-
ists tend to support actions that they perceive as positive by interaction at
public meetings-and through the political process.

Responses to Changes in Generator's Costs--
     Any change in  generator's costs for hazardous waste disposal must either
be  absorbed  by the firm or  reflected in product  prices,  or both.   If the
change  is  an increase,  internal  absorption will  reduce the firm's earnings
and weaken  its position  in the capital  market to  the ultimate  detriment of
the  shareholders;   whereas  an  increase  in product price  will   reduce  unit
sales, depending on the price elasticity of demand.  If the long-run elastic-
ity  is   known  and  the  firm is  assumed  to  be  a profit  maximizer,  it is
theoretically possible  to  determine  the extent to  which the firm can pass a
cost  increase  on  to  its  customers.   However,   some  manufacturing   firms
(primarily those  that  make  nondifferentiated products)  have  little control
over the prices that they receive for their products, and even where this is
not the  case,  pricing  decisions are  not  always made on  the  basis of maxi-
mizing,  profit   (Kotler,   1967;  Backman,   1965).   Indeed,  some  marketing
authorities  claim  that, manufacturing  costs  are  one -of  the  last  factors
considered when selecting a consumer product price (Oxenfeldt, 1960).  Hence,
prediction  of-  a  firm's .response  to a  change  in costs-may-not  be  an,easy'
matter.

     Fortunately,   it  appears  that  the  costs of  adequate  hazardous  waste
management are  often a small  proportion  (about  1 percent)  of  the value of
shipments  (U.S.  Environmental.  Protection Agency,  1974b).   Hence  in  many
cases,  it  may  be possible to neglect the effect,  even  though small changes
probably  have  an   effect   on  pricing  in  the  long run.   Where  changes in
hazardous waste management costs could  represent a significant proportion of
the product  price,   an  examination of the pricing  structure of  the revel ant
industry would  be desirable to establish likely behavior.

                                     49

-------
Response to Negative Impacts--
     The  category of impact that  appears to be  the most  likely to evoke  an
active  response  is  that  of a  negative  impact  via an  environmental  threat.
Responses  are likely  to  include local  opposition to the siting  and  operation
of  most hazardous waste  management  facilities,  but avoidance actions  (e.g.,
moving  to  another house)  are only  likely when the  threats become realities.
Nizard  and Tournon  (1974)  have developed  a  simple  model  (Figure 5)  th?t
predicts the  circumstances under which  an individual  will tolerate  pollution
and  those  under  which he  will  be predisposed  to respond  against it.   Local
residents,  etc.  frequently object to  chemical  landfills.   This  can  extend  to
lagoons  and some hazardous waste storage facilities (see Appendix C).   Under
some  circumstances   resource   recovery  facilities  could  benefit  from  the
positive  connotations  associated with  the concept of  resource  recovery,  but
this  could easily  be degraded by  negative  experiences.   Alternatively,  a
resource  recovery operation might  merely be  regarded  as  a regular manufac-
turing  plant.
                           • Lessening of perception,-*—
                            which not only is no longer
                            useful for the individual
                            but could be disturbing
                                 YES-
Neutral perception
of the phenomenon

Putting a high value
on the activity   —"•
    I
  NO

Awareness of risk
or disamenity    ——

  YES
    I
Feeling of helplessness— YES

  NO
    I
Negative effects
of possible action ~~
   TOLERATION OF
-t» THE POLLUTION
                                 NO
                                 YES
             NO

              I
           PSYCHOLOGICAL CONDITIONS PROMOTING
           ACTION BY THE INDIVIDUAL TO PROTECT
           HIS ENVIRONMENT
                    Figure  6.   Model  for pollution toleration.
                                       50

-------
     Environmentalists  usually  respond vigorously  to perceived  threats  and
are likely  to  be  joined by other groups  strongly  affected by threats (e.g.,
fishermen, where there is a threat to surface water quality).   In addition to
political actions, these  parties-at-interest may use the judicial process to
delay  or halt projects.   Responses  of local  administrative  officials,  etc.
are likely  to  be  more subdued and could  include such tactics as negation by
delay.    There  have,  however,  been  a  number of  recent moves  by  political
officials   to  prevent   the   disposal   of  particularly  tnreatening  wastes
(obsolete nerve gas  and radioactive  wastes, for example) within their juris-
dictions and even to prevent the movement of such wastes across them.

     Again, when  and if  the  threats materialize,  fishermen,  tourists,  etc.
will avoid  polluted  areas,  and  even if there is no official pronouncement on
the matter, the public may reduce their consumption of suspect fish and game.
(Some  pollutants  such as  phenols  and  certain  heavy metals give  fish a bad
taste or smell  [Cannon,  1974].)
                                     51

-------
                                   SECTION 5

                GENERAL METHODOLOGY FOR ANALYSIS OF HAZARDOUS
                        WASTE MANAGEMENT ALTERNATIVES
     This section outlines  a  methodology that may be used by a decisionmaker
to  evaluate  the  effects  of  alternative  approaches  to managing  hazardous
wastes.  The methodology involves three phases, as follows:

     1.   Obtaining prerequisite information;
     2.   Applying the analytical framework;
     3.   Decisionmaking.

     Before  the procedures  and techniques for conducting these  three phases
are described, this  section explains the objectives and  some  of the philos-
ophy  behind the  methodology,  and  it  introduces some  aspects  of  economic
theory that may be relevant to the analysis.

INTRODUCTION

     The primary  objective  of the methodology is to provide a framework for
the analysis of hazardous  waste problems that is based on economics and that
is cognizant of social  factors.  The methodology is intended (1) to assist a
decisionmaker to systematically examine the various alternative approaches to
controlling  hazardous  wastes,  (2)  to  determine  the  nature,  and  as  far as
possible, the  magnitude of  the various effects  that  can  occur,  and (3) to
thereby  make  informed  and  balanced hazardous  waste  management  decisions.

     The methodology is  referred to as a "framework for analysis"  because it
provides structure and  methodology  for analysis, but  it  does  not  attempt to
determine  an  optimum   solution.   Choice  between  alternatives  remains  the
prerogative  of the decisionmaker,  who can make his own trade-offs  and intro-
duce whatever  degree of risk aversion  he  favors.   Indeed,  the conceal of an
optimum solution (in the mathematical sense) is of limited value in hazardous
waste  decisionmaking, since where there are both economic and social consid-
erations,  decisions  are  considered  "normative," i.e.,  they involve  value
judgments.    Such  judgments  are necessary  because  the various  impacts  fall
upon different parties-at-interest,  thereby introducing questions  of equity.
There  is also  the question  of risk.   Different persons will  have various
attitudes  toward  risk-taking  and hence will  require  different benefits to
offset a given risk.   Though an optimum solution could be determined if rules
for decisionmaking were  specified,  in this methodology, the decisionmaker is
encouraged  to  develop  his own criteria  on a case-by-case  basis.   Another
feature  of  this  approach  is that  although it  is  possible to  place dollar


                                     52

-------
values  on changes  in  environmental features,  the available  techniques  and
data  do  not generally  permit  these valuations  to be made with  much  confi-
dence, and hence once again, judgments  are likely to be necessary.

     Thus  situations  encountered  in  hazardous waste  management  call   for  a
systematic analysis of  the  various possibilities in such a way as to provide
a  decisionmaker with  information  about the  trade-offs  between  the  various
alternatives.  The  decisionmaker can then  use his own norms  or critana, or
those he  believes  are  representative  of agency policy, to select an approach
that  satisfies  whatever  policy  objectives   may  exist.   This  methodology
attempts  to  provide a  decisionmaker with  the information that is  needed to
make decisions of the type outlined above.   Though it was developed as  an aid
to making  decisions  about  the  management of hazardous industrial  wastes on a
regional or  local  basis,  it could quite readily be  adapted  to special cate-
gories of  hazardous waste  or to specific  industries,  possibly on a national
basis.

     The  methodology  is  not appropriate  to  determining  whether  or   not  a
particular substance  should be  manufactured  or to what extent it  should be
used  (e.g.,  the use  of PCB's).   The  latter  problem  has  been addressed by
other  studies  (e.g.,  Kennedy   et  al.,  1976;  National  Academy of  Sciences,
1977; Moll et  al.,  1975).   The distinction is that the cost-benefit or risk-
benefit  studies  mentioned  above  examine  the  cradle-to-grave  costs  and
benefits  of  using  a  particular  material,  whereas  this  study  essentially
addresses  only the  problem 'of dealing with hazardous wastes.  once  those
wastes have been created.

THEORETICAL CONSIDERATIONS

     Some concepts from economic theory are difficult to apply in practice or
are subject  to debate,  but they provide useful  insights  for  hazardous waste
management  decisionmakers,  and  for this  reason  are discussed here.   These
concepts  are  Pareto  optimality and  market  failures,  and  the  handling of
intertemporal effects.

Pareto Optimality and Market Failures

The Pareto Criteria—
     The  concept  of  Pareto optimality  constitutes  the  apogee  of planning
goals  in  welfare  economics.*   Although  it is  unrealistic to  expect  a real
economic system to  be  Pareto optimal,  the concept  is  worth  examining, as it
provides useful guidance for decisionmaking.
* Welfare  economics has  been  described as  ".  .  .  the theory of  how  and by
  what  criteria economists  and policy-makers  make or  ought to  make their
  choices between alternative policies and between good and bad institutions"
  (Arrow  and Scitovsky,  1969:1).   Consequently,  hazardous  waste decision-
  making by public officials falls within its purview.


                                     53

-------
     The  basis  of  Pareto  optimality  is  that a  situation  is  optimal  (or
"efficient") when  no one can  be made better off without  at least one person
being made worse off!Herfindahland  Kneese provide a useful discussion  of
Pareto optimality  and show that it implies the following:

     1.    Efficiency in  production:   It  is impossible in an  optimum  to
          increase  the  production  of  one  good without decreasing  the
          production of at least one other good;

     2.    Efficiency in distribution:   It is impossible in an optimum  to
          redistribute the  goods  among the consumers, so that one con-
          sumer is  better  off while  no other  consumer  is worse off;

     3.    Allocation of  resources in  accordance  with consumer  prefer-
          ences.  (Herfindahl  and Kneese, 1974:41,42)

     It can  be  shown that  there  is  no unique Pareto optimum, as  an  optimum
depends  on  the  initial  conditions (Herfindahl  and  Kneese,  1974).    Since
arrival  at an optimum  may be  too much to expect in practice,  projects can be
examined for Pareto  improvements,  which  occur if some economic  change makes
one or more  persons  better off without making any worse off (Mishan,  1971a).

     Because of the  difficulties  inherent in arriving at a Pareto optimum or
making a Pareto improvement within a real economic system,  the  concepts of a
potential Pareto optimum and a potential  Pareto improvement are  often  substi-
tuted as  planning goals  or tools.A situation is  said  to be  potentially
Pareto optimum if  it can be transformed to Pareto optimality merely by making
economic transfers between individuals—that is, that gainers be  able  to more
than  compensate  losers   (Mishan,  1971a).   The  acceptability of any  given
distribution of costs,  benefits,  income, etc. is considered  separately,  and
this aspect is termed "equity" in this methodology.

The Problem of Market Failures--
     It can  be  shown  that perfect competition  and its associated  marginal
cost  pricing leads  to  Pareto  optimal  conditions.   However, there  may  be
circumstances in which major divergences from this situation occur, and these
are termed "market failures."   The failure that comes readily to  mind  is that
of monopoly, as it  is  well known that where a monopoly exists,  it is in the
monopolist's  interest  to  price goods  at  a level   that  is higher than  the
marginal cost of production.  Thus where both monopoly and competition exist,
to arrive  at an  optimal  solution,  it  would be necessary to  replace  monopo-
listic prices by those based on marginal  costs.

     A particularly  difficult problem arises  where the monopoly  is  what is
termed a  "natural monopoly,"  in  which long-run average costs decrease with
increased output.    In  this case,  the monopolist could not  use  marginal cost
pricing unless  he were  given  a subsidy,  as  the  marginal  cost would be below
the average  cost, and  the monopoly would be  incapable  of  recovering  all  its
costs.   This  situation occurs  where  there  are  economies  of scale over the
entire range of output  that is of potential interest.   Many public services,
such  as  sewage  disposal  and  electricity supply, can be  natural  monopolies,
                                     54

-------
and  monopoly  situations (both  natural  and otherwise)  could  well  arise with
some techniques  for  hazardous  waste disposal.  For example, high temperature
incineration  facilities  suitable  for  hazardous  waste might  fall  in this
category, as  they  are comparatively capital intensive, and unless there were
particularly large volumes of waste generated in one location, a high temper-
ature  incinerator  would be  likely  to  establish a monopoly within  a zone of
influence determined by transport costs.

     Landfills  constitute  a  rather interesting  situation, because  to some
extent, their operation resembles the  mining of a mineral  deposit.  Because
of the  finite capacity of a landfill,  and  because the land after filling is
worth less than the land before filling, the variable costs should include an
element of  depletion  to allow  for  the consumption of  a  resource  (i.e., the
space for disposal).   Even so,  there seems to be no reason why marginal costs
should rise with the volume of waste handled for most chemical landfills.  As
a result, both  their pricing and behavior  are  likely  to reflect elements of
monopoly.   Since fixed costs'(e.g., licensing, 'environmental  monitoring, and
security) can be comparatively  high,  marginal and average costs could differ
significantly.

     The  second important  area  of market  failure is  where prices  do. not
reflect  externalities.   Free  use  of  environmental resources,  such  a's the
pollution assimilation capabilities of a river, provides a classic exanple of
an externality  (or "spillover  effect," as  it is  sometimes called).   If the
waste generator does  not pay for the use of the environment (which is a cost
to society,  since  some  individuals are damaged by  a  degraded environment),
his production  costs  (known  as.his "internal costs")  will  be lower than the
true  costs,   and   consequently  he  will produce  more   of the  good  4:ha*n  is
societally efficient  (i.e.,  Pareto optimal).  Furthermore, under  these cir-
cumstances,  there is no incentive for efficient use of  the environment; since
it is  free to the  generator, he  will  theoretically use whatever quantity it
takes  to  minimize  his  unit  production cost.   (For a  detailed analysis, see
Barnett  and  Morse,  1963.)  Since  the environment  generally has  a limited
restorative or  treatment capacity,  this  can lead to  a  "commons  situation"
(Hardin,  1968),  in which  such  capacity,is overloaded  because no individual-
user has sufficient incentive to reduce use.

     A  key  part of  the methodology described in  this  report  is  to identify.
and,  if possible,  evaluate such externalities.   Furthermore,  -some approaches
to hazardous  waste  management may invo.lve  the   use .of .economic  incentive.
means, such as  user  charges' and effluent fees,.-that internalize'these'e'xter-
nalities.    (Externalities  can  also be internalized  by specific  pollution
control laws  and the general laws of liability.   See Anderson et al.,.1977,)
However, the  question  that is  pertinent here is:   What adjustments, if any,
should be made  to  data where there are uncompensated market failures?  There
is some  debate  about  the  extent  to which "shadow prices"  should be used in
cost-benefit  analysis,  where such  prices are  nonmarket prices--for example,
prices based  on marginal  costs  and benefits that fully reflect externalities
(McKean, 1968).   The differences between shadow and market prices will affect
quantities of products  and  wastes produced  (Freeman, Haveman,  anc  Kneese,
1973).  While  shadow  prices may be  necessary  to  evaluate   Pareto  optimal
conditions,  the  problem is  that  once  one  adjustment  is made  to  one price,


                                     55

-------
then  output,  consumption  of other  products, other  prices,  etc. will  also
change.   In short,  something of a chain reaction will  be set off.   Arriving
at  an optimum  solution under  these conditions  is  usually  a  complex  task
(Lipsey  and  Lancaster,  1956;  Herfindahl  and  Kneese,  1974;  Mishan,  1971a).

     The analyst  or  decisionmaker has  two  interests  when he  identifies  a
market failure:   First, what should he do if he has  control  over pricing, and
second,  how should  he  conduct his evaluations if he  does not have that con-
trol?   The  pragmatic  answer to  the first  question  is  that  where  he can
control  prices  (through  some policy  means),  it is probably more efficient to
move towards marginal cost pricing, even though this is not employed in other
sectors of  the  economy  (see Price,  1977;  Bohm,  1973).   In the  second case,
the  analyst  should be  cautious  about substituting shadow  prices  for market
prices in evaluations, unless the market failure  is  clearly a  najor one.   The
literature provides  some  support  for this  -viewpoint (see McKean,  1963), and
in  addition  a  project is more  likely to  be accepted by  the  various parties
concerned if the  analysis  is straightforward and readily comprehensible.  Of
course,  external  costs (such as  environmental costs) should be included in
the  evaluation,  which  will  partly correct the  deviations   from  optimal
conditions,   and  the  analyst  can  also seek strategies  that  endeavor  to
eliminate market failures.

     Ultimately,  however,  one should not  lose sight  of  the fact  that most
analyses involve  making changes  from  an existing  situation.   Provided that
the  new situation  represents a  potential  Pareto  improvement,  society has
gained and hence the analyst need not be too inhibited by the  comolexitiss of
trying to find  an optimum solution.   (For more detailed  treatments  of these
topics,   see  Mishan,  1971a,b; Scitovsky,  1951;   Arrow  and Scitovsky,  1969;
Price, 1977;"Bohm, 1973; Chase,  1968).

Intertemporal  Considerations

     It is very widely accepted that the discounted cash flow (DCF) approach
is an appropriate technique for evaluating  the economics of business projects
where income and expenditures do not coincide in  time.   The approach is based
on  the  concept  that future income is less valuable than  present income and,
correspondingly, that future costs are less onerous  than current costs, since
in  the  intervening period  the  capital  could be  invested in  some other way.
Thus  the  discount  rate  chosen  for any  evaluation  should  be related to the
opportunity cost of capital.  This argument does  not presuppose  the existence
of  inflation, and is therefore  applicable  to costs  and revenues expressed in
constant dol.l.ars.   In the event that under  conditions of inflation with costs
expressed in current  dollars,'the discount rate would  need -to  be increased.
(For  further information  on this topic, see Taylor, 1964; Merrett and Sykes,
1963; Stermole,  1974.)

The Social  Discount Rate—
     The basic  concept  of  discounting  can  be  applied  to  cost-benefit  or
cost-effectiveness  studies  of public investments  but some difficulties can
arise over the  choice of an appropriate discount rate.   There  are two major
aspects  to  these  difficulties:    (1)  The  question of the use of  a social
                                     56

-------
discount rate, which  is  lower than the business rate, and (2) whether or not
the  discount  rate  should  be raised  to  reflect  risk.   Arguments  for  and
against the  use  of  a social discount rate are summarized in Appendix E.   The
key  argument favoring the  use  of business  discount rates is  that if lower
rates are used for public projects, capital will incorrectly be diverted from
the  private  to  the  public  sector,  reducing  overall  economic  efficiency.
There  are   several  countervailing  arguments,  but  they  all  amount  to  the
assertion that a  low  public discount rate is  necessary to insure the appro-
priate mix  of public  and  private sector projects.   A  business project  that
carries a strong risk of failure must usually offer a higher expected rate of
return than  one  that  is  essentially risk free.  However,  when a project that
involves risk  or uncertainty is  proposed for the public  sector,  there  are
some arguments against raising the discount rate to allow for risk.

     Analysis of the  literature  suggests that the appropriateness of using a
social  discount  rate  depends on  the  nature  of  the project  involved,  and
though arguments for using a risk-free discount rate predominate, there could
be circumstances where a risk premium is appropriate for public projects.   To
a  large  extent,  the  solution.depends.on the  degree  of competition with  the
private  sector.   For  example,  evaluation  of  the perpetual  care   costs  of
storing long-lived wastes might  use a low social discount rate, while evalua-
tion of a project  to install an  incinerator as  an alternative to a landfill
might  use a high-risk commercial  rate, as  this has the  nature  of a normal
business decision.

Intergenerational Effects—
     Up to  this  point,  the -discussion  has  been appropriate 'to  actions  and
projects with  a  time span  of  a  decade or two,  since they  revolve -around
decisions  from the viewpoint of the present generation.   However,   where  a
longer time  span is involved, the viewpoint of future generations should also
be considered.   This  is particularly  relevant  for  projects  that  '-esult  in
irreversible changes—which  may arise  from any project that involves modifi-
cation of the natural environment,  the use  or disposition  of norrenewable
resources,  or even construction.

     When irtergenerational effects are"present, the arguments for the use of
a  low  social discount rate become stronger,  and  some  arguments  against  any
discounting   have  been advanced.    The  main objection to  discounting is  that
events (such as  genetic  changes)  that may happen  far into the future are so
heavily  discounted  that  they   are virtually,  disregarded  in  conventional
analysis.   Thus society  may be  making Failstian  bargains  that provide-short-
term advantages but  that could  be bitterly  regretted by'future generations.
This subject is discussed in detail in Appendix E,  where a pragmatic solution
to the problem is suggested.

     This study proposes discounting  over a normal project  life  of approxi-
mately  one   generation-,  but- -no  further  discounting  of  effects  that  occur
beyond that  period  (i.e.,  the discount  factor  is  held  constant fcr any year
that  is  beyond  one  generation  from project  inception).   This  solution  is
attractive  because  it  simulates  conventional  business  decisionmaking  pro-
cedures over normal project lives (which rarely exceed two or three decades,
i.e.,  one   generation),  but it  avoids  excessive discounting  of   the  very


                                     57

-------
distant  future.    Some  justifications  for  this  approach  are  advanced  in
Appendix E.  Note,  however,  that with this procedure, it is only possible to
use lump sum valuations  of threats (e.g., a  one-shot cleanup cost, raolaca-
ment cost,  etc.)   if the  period  of evaluation is  infinite.   Any annual  cost
continuing  for  an infinite  period (such as  maintenance cost for  perpetual
care)  will  have  an  infinite  present  value  if  it  is  not  continuously
discounted.  Thus where  no  lump  sum equivalent can  be  found to  real ace  a
continuing  cost,   the  planning  horizon  must  be  limited  or  conventional
discounting must be employed.

The Optimum Timing of Projects--
     A  substantial  body   of  literature  deals  with  the  optimum  timing  of
projects.   For example,  it can be shown that  the  time profile of the strsam
of  benefits or income  can be such  that net  present value  is  maximized ay
delaying project inception (see Herfindahl  and Kneese, 1974).  This situation
could apply to  some  pollution abatement projects because the annual benefits
(which are  damages averted)  increase rapidly or abruptly  during the project
life.  For example,  it might pay to  delay  the construction  of an affluent
treatment  plant  because   the  quantity  of  effluent  or  its  potential   for
environmental  damage is expected to increase in the future.

     Another situation  in which  delay  can be  beneficial  *r  whers ;nere is
uncertainty about  the  magnitude  of future benefits,  and where delay sennits
additional   data to be  obtained.   This benefit of reducing uncertainty can be
regarded as an  option value  (see  Appendix  C),  and could  b9  particular:.-
significant in  hazardous  waste management where the  irreversible  impacts of
the  use  of a particular  technique are  not well defined.   In  this  event, it
could be an advantage  to society to delay taking an essentially  irreversible
action (such  as  deep well  injection of a waste)  until  further  researcn and
analysis of the  impacts  of this action can be completed.   During the inter-
vening period,  it would  be necessary to use  a management technique (such as
engineered  storage)  that  has a  low potential  for  adverse impact  (both in
terms of.1ow .expected  impact and  limited uncertainty about the  impacts) but
is  unsuitable  as   a   long-term solution because of  high  cost or  some other
reason (such as limited capacity).

USE OF THREAT SCENARIOS

     The most  difficult  part  of any economic  analysis  of pollution control
problems,  is almost  invariably  that  of  determining  damages.   According to
Fisher  and  Peterson (1976),'  there   are  four  stages in  the assessment of
damages  from  conventional pollution  sources,  as shown  in  the upper part of
Figure 7. -Starting with a specified emission or waste discharge, the amoient
conditions  and the  physical  effects  must be  determined before  the dollar
damage costs can  be estimated.   Again, note  that  there  will be  effects  that
are difficult to quantify in dollar terms.   To extend the Fisher  and Peterson
model  to  hazardous  wastes,   it  is necessary  to add  one  preliminary stage:
Identification  of the possible  threat  mechanisms.   This  stage  is necessary
because  the nature of  most hazardous waste management problems  is such  that
the management  techniques used present a variety of threats of adverse envi-
ronmental  impacts, whereas  conventional  pollution  control  analysis usually
centers  on the effects  of  a  known  waste stream  discharged  to  a specified
environment.

                                     58

-------
     APPROACH FOR CONVENTIONAL POLLUTION PROBLEMS
EMISSIONS
(e.g. , m of
row sewage to river)


AMBIENT
CONDITIONS
(ppm of dissolved oxygen)


EFFECTS
(thousands of dead
fish, etc.)


DOLLAR
COSTS
(value of dead fish, etc.)
en
u»
              THREAT
            MECHANISM

       (e.g., injected wastes infiltrate
            freshwater aquifer)
     THREAT
    SCENARIO

(thousands of wells
 contaminated  over
   specified area)


       COSTS  &
       IMPACTS

   (cost of replacement
water supply, health damage)
     ADAPTATION TO HAZARDOUS WASTE PROBLEMS
              Figure 7.   Process  for assessing damages associated wiUrhazardous waste  management.

-------
     In principle,  it  is  possible to model environmental  impacts  and hence
arrive at dollar  values for the damages attributable to the use of any tech-
nique, using willingness-to-pay where necessary.   In practice, however, this
can be a  far  from simple procedure and could require considerable resourcas.
For each  technique,  it might be necessary to consider several threat mecnan-
isms, while  each mechanism  could probably  cause  impacts with  a variety of
magnitudes,   depending,  for  example,  on ambient  conditions,   toaste  stream
variability could further compound  the  number of cases  to  be considered IP
the  physical  modeling.   Hence  assuming adequate data were  availaole,  the
dollar damages would  ideally be expressed, not as a point estimate, but as a
probability distribution  of  costs.   The ways in which these costs would fall
on  the  various  parties-at-interest  could  also  vary  from  situation  to
situation.  Furthermore, as shown in Appendix F, the way in which individuals
perceive  a  risk  or   threat may  be  more  important  in  determining  tneir
responses than  the actual  magnitudes and dollar  sums associated  witr: that
threat.

     The  sheer complexity and the fact  that the  probabilistic approach out-
lined above  fails to  recognize perceptions  of threats are  good  reasons why
this  approach may not  be  an  appropriate  tool  for  many  hazardous  waste
decisions.  The  problem  of  perception could be partly overcome  by modeling
the physical effects  and then asking  the parties-at-interest  how they xould
respond  to  these threats.   However,  this  approach  would not  overcome  the
difficulty  (so  apparent  with  nuclear power)  that  the public  may  not trust
experts'   assessments  of threats.  Furthermore, under most circumstances nany
of the necessary  data  are unavailable and those that are available are fuzzy.
To generate the  missing data and  refine  those  that  are available could be a
major task, requiring a level of effort that is simply not available or that
is beyond that justified either by the  nature  of  the decision to be made or
by the  sophistication  of the available  techniques for modeling  and valuing
effects.   Even where  this probabilistic approach is conscientiously followed
through  in  detail,  there are  numerous  possible sources  of  error  and bias
(Slovic and Fischhoff,  in press).

     An objective of  this research was  to develop an analytical  methodology
that  does not  involve  detailed  analysis  and that does not require extensive
data.  A  central  concept of the proposed methodology is 'to replace the first
three stages of the conventional damage model (emissions,  ambient conditions,
and effects) by a threat  scenario, as  illustrated in the lower part of Figure
7.  The  scenario  describes what might typically happen in the event that any
specified threat  should become a reality.  Where appropriate,  more than one
scenario  could be  used  to  cover different"'threat mechanisms  or  different
outcomes  arising  from  a given mechanism.

     Judgment will  be  necessary to limit the  number of  scenarios that are
considered.    As   shown in  Section 4,  there are  usually  numerous  possible
threats.   The analyst should  pick  those  that appear  to  be comparatively
likely,  those  that (as far as  is known) could have  particularly disastrous
consequences, and those with which the general public or certain parties-at-
interest  are  especially  concerned.   In practice (as illustrated in Section
6),  reducing  the  number of  scenarios  to manageable proportions may not be as
difficult as it appears at first sight, since in a given situation, there may
be a consensus on the  threats that are significant.

                                     60

-------
     Though  merely  qualitative  descriptions  of  'threats  would be  useful,
typical  quantitative data  should  be suggested  when possible.   These  would
reflect judgments based on the results of modeling studies, actual experience
with that  type of  threat,  or worst  case assumptions.   Where site-suecific
modeling of threats is not feasible, the analyst may be able to adapt some of
the available analyses or case studies of hazardous waste damage incidents to
meet his needs (see Appendix B).

     A  major  difficulty with  the  use of  typical  threats  taken from actual
experience elsewhere  will  be  to choose  a magnitude or  scope  for the impact
that is appropriate to  the  types  and quantities of wastes  concerned and to
the  local  circumstances (e.g.,  geohydrologic conditions).  However, simple
worst case assumptions  could be  useful  to place limits on some impacts.   For
example, in evaluating  the  effect  of landfill leaching, the assumption could
be  made that  after many  years,  a steady  state  is  achieved  in  whi:h  the
leachate contains  the same  quantities  of- nondegradable  toxic "elements  that.
enter  the  landfill  and that  this  leachate  enters  the  local  river system
without attenuation.   If the streamflow is known, the  average concentration'
of toxic elements in the river could be calculated, and the effect on aquatic
life could  be predicted.  Another  approach  to the same,problem would  be to
assume  that  any   leachate  was  normally  highly  attenuated  (e.g.,  by  ion'
exchange)  before  it  left  the vicinity  of  the  landfill,  but  to  estimate  a
cleanup cost  represented by  the  cost of  installing  and pumping a sufficient
number  of  interceptor  wells  to  contain  the leachate  should   it  become
necessary.

     Though admittedly simplistic,  the  threat scenario approach overcomes or
avoids  many  of  the difficulties  associated with  the more detailed proba-
bilistic  approach.    It  can  accommodate whatever data  are  available,  but
perhaps its  most attractive  feature  is that  it recognizes the sociological
dimensions of  a  decision situation.  Threat scenarios  can be  constructed to
reflect or include actual public  perceptions and concerns.  The attitudes and
behavior of  parties-at-interest  can be  predicted,   and  decisions  can  take
these  factors  into account.   In  many  respects,  the  absence  of accurate
qualitative data need  not. be. of  undue concern,  the  decisionmaker is largely
interested  in the  reactions  of  individuals 'to  the threats  from  hazardous
waste management  alternatives.  In  many cases  these  are  likely  to reflect
what has happened  in the past, even if the circumstances are different.- ••For
example, individuals tend  to rely  on recent experience when making judgments
on  the  probability and..maximum  (expected)  magnitude of an  event,  such-as  a-
flood.   (See  Slovic,   Kunreuther,  and White,  1974;  .Slovic,  Fischhoff,  and
Lichtenstein,  1976; and  Slovic  and  Fischhoff, in  press.)   Thus  the  mere
identification of  threats  is-an  important part  of this methodology, even if
the magnitudes  of  the  threat impacts and the probabilities of their occur-
rence are ill-defined.

PREREQUISITE INFORMATION FOR ANALYSIS

     Obtaining prerequisite  information is  the  first  phase in applying  the
methodology.    Four  steps must be  taken to provide the inputs necessary  for
analysis:
                                     61

-------
     (1)  Define the  scope of  the  study in  terms of both  the  type  of
          waste and the geographic area to be considered;

     (2)  Obtain an inventory of the existing hazardous waste situation,
          including both generation and disposal;

     (3)  Determine  how the hazardous  wastes are  currently controlled
          within the study area;

     (4)  Ascertain  policy  objectives  for  hazardous  waste  control.

     These steps are discussed in the following subsections.

Defining the Scope of the Study

     The first step in collecting input for analysis is deciding on the scope
of  the study.   The geographic  scope  is usually  dictated  by the terms  of
reference of the study and is likely to correspond to a political division or
unit such as  a State or a planning region.   If any choice is possible, it is
desirable that the area chosen be geographically isolated.  Otherwise, wastes
crossing the  study area boundaries could complicate  analysis.   For example,
where  two  separate political  units  share a  major metropolitan  araa,  tnere
could  be difficulties   if  the  two  units   adopted   significantly  diffarent
hazardous waste management  policies,  possibly  leading  to  waste  transfers
between  the  units  that  would  be  unlikely to  promote  overall  economic
efficiency.

     Two  aspects  of waste type need  to  be considered.    The  source-related
categories of  waste, and within these categories,  the choice of a definition
of  hazardous  waste.    Although  this  study  is  oriented toward  industrial
(process) wastes,  the   general  methodology  could be adapted to  cover a wide
range  of potentially  hazardous wastes.   Since these wastes  tend  to have
different  characteristics  (largely  in  terms  of  type  of  generator  and
frequency  of  generation,  also  with  regard  to degree   of  exposure   10  the
hazard),  they  may  ideally  require  different  management  policies.   For
example, it  is unlikely that exactly  the same approach  toward waste manage-
ment  would  be  ideal for  large recurrent  quantities of  industrial  process
wastes,  occasional  obsolete   laboratory  chemicals,  and  empty  pesticide
containers.   Because of these  differences,  it may be  desirable  to limit the
scope  of any  study to   insure that the wastes considered exhibit some degree
of  homogeneity,  or  alternatively  to  consider  different  approaches  for
different wastes.  Furthermore, the agency conducting the study may not have,
or may not  wish to exercise control  over certain categories of wastes (e.g.,
Department  of Defense  wastes).  Some categories  of waste  may  already  be
adequately  controlled  (this could be the case  for  radioactive  wastes),  and
therefore additional  study would  be unnecessary.    Hence some limitation of
the scope of the study will probably be necessary.   (The major source-related
categories of potentially hazardous wastes were listed in Section 3.)

     Within the source-related categories of waste, there remains the problem
of deciding which  wastes are hazardous.  Where  a  definition is  not mandated
it  is  practical to  start  off  with  a broad working  definition and narrow it


                                     62

-------
down as  the  study data are analyzed.  However,  certain  marginally hazardous
wastes  could be  eliminated at  the  outset  if  it would  be  impractical  to
regulate  them in  the  same  way  as  the other  hazardous wastes.   Waste  oil
(other than perhaps that from the oil refining and re-refining industries) is
a  good  example   of  such  a waste.    The  use  of  oils  in the  engineering
industries is so widespread that it is difficult to control their disposal in
small quantities.   In view of the comparatively limited threat that waste oil
poses  to the environment,  it  is often omitted  from  hazardous  waste studies
(Stradley, Dawson, and Cone, 1975).

     Of  course,   in  addition to  studies  that  cover  all   hazardous  wastes,
studies  can  be  conducted  on the wastes of a single industry (e.g., pesticide
manufacture)  or  on the  disposal  of  a particular waste  (such as  PC3's)  or
wastes  containing a  particular  element (mecury,  for example).   For  single
industry studies,  the  industry  can be defined by its SIC number (Statistical
Policy  Division,   Executive  Office of the  President, 1974).   This approach
works  well  for some  industries  such as the basic process industries  [e.g.,
copper  smelting), but care  must  be exercised when  dealing with  diverse
industries (such  as  electronics)  that can appear  under many  different  SIC
numbers.

     For  studies   that  deal  with  a  particular  waste  or  element,  the  main
problem  will  be  to define  the concentration  or quantity  that qualifies a?
hazardous.   For  example,  even  in  the absence of  industrial  sources,  sewage
sludge   usually  contains  low  concentrations  of  some  heavy  metals  that
originate from plumbing fixtures.   (Personal  communication, W.L. Ross,  Denver
Research  Institute,   Denver,  Colorado,  September  1977.)   However,  though
sewage  sludge is   not  without  its disposal  problems, this  sludge would  not
usually  be   regarded  as a  hazardous  waste.   The solution  to this type of
problem  is to establish cutoff concentrations and/or quantities.

Taking Inventory of the Existing Hazardous Waste Situation

     Before any economic analysis can be performed, it is necessary to obtain
a general picture of  the existing hazardous waste situation within the study
area.   The   information  required  depends  on the  precise  objectives  of  the
study,  but  in most  cases  it would  be appropriate  to obtain  the following
data:

     1.  Sources of wastes (SIC categories and locations);

     2.  Types of wastes;

     3.  Annual quantities;

     4.  Current dispositions of wastes.

EPA  has published  a   guide  to conducting  hazardous  waste  surveys (Porter,
1975),  but  a more detailed  appreciation  of what might be  involved could be
obtained by  reviewing  one  of the State or industry surveys (depending  on the
orientation of the study).
                                     63

-------
     The critical  aspect of  any  hazardous waste  survey lies  in  the  way in
which  the  assessment  of waste  generation is  approached.   There  are three
principal approaches that can be used:  '

     1.   Attempt to inventory all hazardous waste sources;

     2.   Sample  hazardous  waste  sources  and  extrapolate   to  estimate
          total study  area  waste  on the basis of industrial employment,
          physical output, or value added within SIC categories;

     3.   Use  waste  generation factors  (e.g.,  tonnes/year  per  employee
          for  a  given SIC  category)  obtained  from national  studies,
          together with study area employment, physical  output,  or value
          added  by  SIC   category  to  estimate  total  study area  wasta
          generation.

     Clearly,  the  first  method is the most accurate, but it has rarely been
used because  of  the  high cost.   To  date, it  has  only been  feasible wnere
there were a limited number of firms involved, as in certain industry studies
and some State and regional studies.   For example, this approach was adopted
for some sectors  in  the  study of hazardous waste practices  in the nonferrous
smelting industries  (Leonard  et  al.,  1975), and it was attempted by 33-telle
Memorial Institute,  Pacific Northwest  Laboratories in their study of naiara-
ous waste management in  EPA Region X (Stradley, Dawson, and Cone, 1975).   In
the future, under Section-3002 of PL 94-580, data should be  available for ail
generators of  such wastes as are deemed hazardous under Section 3G01 of that
law.   However,  data would  not be available for wastes  that were  not  deemed
hazardous under the  above law, and hence studies that addressed other wastes
would still need to collect information.   Note that even with a general study
at  the  State  level,  for  example, it would be virtually impossible to inven-
tory every  organization  that might  occasionally  have small  quantities of
hazardous wastes.

     The second approach  (extrapolating  from a sample), has been  by far the
most frequently used to date.  For example, this approach was used in a study
of  hazardous wastes  in Massachusetts  (Fennelly et  al. ,  1976)  and in a study
of  hazardous  waste   management  in the  pharmaceutical  industry  (Arthur D.
Little,  Inc. ,.1976).   A  common method is  to  attempt to obtain data from all
the major  generators  and  then to  assume  that the  smaller ones produce the
same proportion  of  waste  per unit of  physical  output  (process  industries
producing a  single  major  output) per employee or  (less  commonly) per value
added  do.llar.   The  disadvantage  to this method is that it implies  that the
process  technology used  in  the smaller firms is similar to  that in the large
firms.    Since  small  firms are rarely miniature versions of  large firms, this
assumption can lead  to significant error.  A further drawback to this method
is  that unless  there are  additional independent  data, the disposition of
wastes  from the smaller firms will be unknown.

     The third approach  (use of waste generation factors)  is inexpensive but
is  liable  to  be  of  questionable  accuracy, because in  addition  to the defi-
ciencies noted above, it  does not take  account of  regional  differences in
technology or  (for the per employee and value added versions) of differences


                                     64

-------
in labor  productivity,  etc.   A hazardous waste generation study for the Twin
Cities  area,  Minnesota, used  this  approach  in  part (Barr  Engineering Co.,
1973).

     A  fourth approach,  that is  something  of  a  hybrid between  all  three
approaches,  involves  the  development of  a  series  of  model  plants.   These
plants would  be  of  different sizes and would use different processes.   Hence
any particular industry structure  could  be simulated by specifying an expro-
priate mix of model  plants.   For example,  largely  because  of the paucity of
survey  data,  this  approach  was adopted  in  the  study  of  hazardous  waste
management in the electroplating  and metal  finishing  industries (Battelie-
Columbus Laboratories, 1976).

     While collecting data on the quantities and types of wastes generated in
the study  area,   it is  also  convenient to collect data  on  existing disposi-
tion.    However,  to  some  extent,  an  independent  check  on  wastes with",n the
study area  can be  made by  obtaining  data on wastes being  sent  for various
forms  of  disposal   (e.g.,  landfilling  at  licensed  sites')  and  resoi/r'ce
recovery.    These data  will  not  reveal  the extent of uncontrolled  waste
disposal  or   storage  at a manufacturer's  site,  so  this approach  cannot  be
substituted  for   some  sort  of  study  of waste  generation.    However,  waste
disposal  data  can  provide  a  supplementary source of  information  about
hazardous waste  generation (e.g.,  by  identifying hazardous  waste generators
in unexpected SIC categories) and can permit checks  on the quantity estimates
of some firms.

     Many waste  surveys also  include estimates of  future  waste  generation.
This  can  be  particularly  significant  when  new  air   and  water  pollution
controls  are  expected  to   lead  to  additional  wastes  for  disposal  (e.g.,
scrubber and  water  treatment sludges) or where process  technology is  under-
going change.   Estimates  are  available  for the solid wastes  expected  to  be
generated as  a result of the Federal Water  Pollution Control Act Amendments
of 1972, the Marine Protection, Research and Sanctuaries Act of 1972, ,and the
Clean Air  Act of 1970  (Stone et  al.,  1974).  However,  the available  waste
disposal options  (and  their  costs)  may influence  the  quantities  of  wastes
that  are  generated, so prediction  of future  waste generation will  be con-
sidered later in the analysis.

Determining How Hazardous Wastes  Are Currently Controlled

     The existing situation  or status quo of  hazardous  waste generation and
disposition  makes  a useful  base case against which  to  measure changes that
might result from various alternative approaches.   Hence.it is also necessary
to determine how hazardous wastes in the study area  are  controlled.

     In addition to explicit controls, such as mandating ultimate  disposal  of
certain wastes at a chemical landfill or  other approved facility, there may
be indirect controls that must be identified.  For example,  regular landfills
in the  study area might be prohibited or  restricted in accepting industrial
wastes.   Even  if  these  restrictions were not scrupulously adhered to (as may
well be the  case  in practice), the  effect would  be to  divert most hazardous
wastes  to  alternative  forms  of  disposal or to land  disposal  within another


                                     65

-------
jurisdiction that does not have such restrictions.   It is therefore necessary
to  examine  rules  and regulations, licensing  requirements  and  practices  in
order to  seek out  indirect  ways  in  which hazardous wastes  are controlled.
The Clean Air Act Amendments  of 1970 (PL 91-604) and the Federal Water Pollu-
tion Control  Act Amendments  of  1972  (PL 92-500) are ubiquitous examoles  of
such  indirect controls,  which,   although  not  specifically  directed  toward
hazardous waste  (other than  Section  112 of PL 91-604),  nevertheless  'lave a
major impact  on  hazardous  waste  management.   Several other  Federal  laws can
have generally minor  impacts  (see Section 3),  and in addition, there will  be
many State (and sometimes local)  laws  and ordinances that also exert indirect
influence.  An  important feature  of  much  of  this legislation  is  that it  is
not the  actual  statutes,  but the  administrative proscriptions and decisions
that  are  important  (Haskell  and  Price,  1973).    Hence  to establisn  flow
hazardous wastes are  controlled  under  the  status  quo,  it  is  iiioortant  to
examine these administrative  decisions and their enforcement.

Ascertaining Policy Objectives

     The  final prerequisite  is to ascertain the policy  objectives that will
govern  the  approach  that is  adopted to  control  hazardous  wastes.   Policy
objectives generally  deal  with  normative  issues,  and  it is  not infrequent
that optimization  of a  given approach,  or  choice between  approaches,  dill
require trade-offs  between achievement  of different objectives.  The goal  of
economic  efficiency  in  the allocation of resources  (i.e.,  striving  toward a
potential Pareto optimum) is  usually assumed without  question (Haveman ana
Weisbrod,  1975;  Planning  Branch,  Treasury  Board  Secretariat,  1975),  even
though it may not  be achievable   in practice.   Other policy objectives nignt
cover the following topics:

     1.    What is  regarded as equitable, and the extent to which depar-
          tures  from an equitable situation can be tolerated;

     2.    The  extent to  which  policies  should reflect  risk aversion;

     3.    Preferences for the use of taxation and economic incentives as
          policy tools;

     4.    The degree  to  which government should proscribe and regulate,
          as  opposed  to  relying on  market  forces backed  up  by  the
          judicial  process for determining liability questions;
     5.
The degree of autonomy permitted to relevant individual juris-
dictions, agencies, etc.
Some  policy  objectives may  not be  specifically  stated, but  they will con-
stitute  a  tradition of  that agency, or  they may reflect  the  mores of that
society.

     When  policy  objectives  (including  implied  objectives)  are  not suf-
ficiently detailed  or  complete, it is probably best to apply the methodology
to  the  evaluation  of various  feasible  alternative  approaches,  and then
highlight  policy  implications  along  with  other  information  required  for


                                     66

-------
declsionmaking.   It may  also  be  found  that It  is  not possible  to- devise
approaches that  satisfy  all  the policy objectives.   For example,  it may not
be  feasible  to  achieve  economic  efficiency  as  a  result  of  uncorrectable
market failures, or there may be trade-offs between the efficiency and equity
that can  be  attained.   In this case, the most expeditious procedure would be
to consider a variety of approaches that look as though they may come reason-
ably  close to  meeting  the  objectives.   When  the  outcomes of  using  these
approaches have  been  determined,  any shortfalls in meeting objectives can be
identified and brought to the decisionmakers1  attention.

THE ANALYTICAL FRAMEWORK

     Applying the  analytical  framework  is the second phase in the use of the
methodology.   The sequential steps involved in applying the analytical frame-
work are as follows:

     (1)  Develop alternative approaches for hazardous waste management.

      For each approach under consideration:

     (2)  Allocate wastes to techniques;

     (3)  Develop threat scenarios, list other impacts;

     (4)  Determine economic and social  effects;

     (5)  Determine impacts on the parties-at-interest;

     (6)  Project responses of the parties-at-interest;

     (7)  Predict physical outcomes;

     (8)  Enumerate costs and impacts;

     (9)  Reiterate Steps 2 through 8 as required.

Each step  is  discussed below.   Because these steps closely follow the inter-
action model  (Figure 5)  discussed  in Section 4,  there  is  some  overlap with
that  discussion.   However,   in  Section  4 the  orientation was  behavioral,
whereas that which follows is intended to provide practical guidance.

Development of Alternative Approaches to Hazardous Waste Management (Step 1)

     Each  approach represents  an alternative  general philosophy  or actual
strategy  for  managing hazardous  waste  'that  is  broadly consistent  with  the
policy  objectives.    For example,  one  approach  could  be to  reqjire  all
hazardous wastes either  to  be detoxified or  to  be disposed of in a chemical
landfill.   Another  example  could  be  an  incentive  approach  to  encourage
disposal  at  chemical  landfills by  subsidizing their  operation.   'Different
definitions of  hazardous waste and what  constitutes  detoxification,  or dif-
ferent levels of subsidy would be considered as falling within one approach.
Thus an approach is a general strategy rather than a detailed plan.


                                     67

-------
     In  the  author's  terminology,  approaches  may  be either  positive  or
negative and  specific  or  influencing.   A positive approach  directs  actions
toward a solution  or  form of management, whereas a negative approach  directs
action away from something.  In the case of specific approaches,  requirements
are spelled out; thus  a  specific positive approach would  mandate  something,
such as the use  of a particular pollution control  technology.   Conversely a
specific  negative  approach would  ban  something,  such  as  the  use  of  a
particular  means  of  waste   disposal.    On  the  other  hand,   influencing
approaches  attempt to encourage or  discourage  an  action  (e.g.,  by  using
economic incentives) as  opposed  to  mandating something.  This distinction is
important when the linkage between approaches to  hazardous  waste  management
and techniques for the control  of hazardous waste is examined.

     In most  situations,  it will be appropriate to include the status quo as
a base case,  even  though it may prove  difficult  to define the approach that
it  represents.   The principal  advantage of  using  a base  case  is that  the
effects and outcomes  of  the  various  approaches can be  expressed  as  cnanges
with respect  to  this  case,  and it  is  often easier to  determine  changes in
some parameter,  as  opposed to  calculating absolute values.   For  example, it
can  be very  difficult to  determine  the total  magnitude  of the  consumers'
surplus, whereas the  size of  a small  change can  often  be estimated  with
comparative ease.  Aesthetic and existence values must be evaluated in terms
of changes  (see  Appendix C).   The status quo is often a gooa starting point,
as  people  are familiar with it  and are largely  concerned with  changes  from
the  existing  situation.   However,  in  some circumstances, a  base  case other
than the status  quo  might be appropriate.  This  could  occur when some major
new  development  (such  as  a  change  in  the law, a  major new waste-generating
plant, or a new disposal  facility) is already underway.   In these situations,
the base case  would need to reflect such developments.

Allocation of  Wastes to Techniques (Step 2)

     As  a  preliminary  action,  it  is  necessary to  determine  which  waste
management techniques  should be  considered (see Appendix A).   Techniques can
be  ruled out  for a variety of reasons,  including local infeasibility (e.g.,
lagooning  for evaporation  in  wet  climates),  technical  infeasibility (e.g.,
biological  treatment when there are no  biodegradable wastes),  conflict with
policy  or  objectives  (e.g., the use of ocean dumping),  and excessive cost
(e.g., space disposal for most wastes).   Clearly, one has to be careful about
eliminating disposal techniques  on  economic grounds before economic analysis
has  been conducted.   However,  there  may  be some  situations  in which  one
technique has a  very  high control cost  and  appears  to  provide no offsetting
advantages  over  a  technique  that  has  a much  lower  control cost.   If  the
parties-at-interest  are   similar for both,  then  it  would be reasonable to
eliminate the technique with the higher control cost.

     The next action is  to  try  to  predict which techniques  will  be  used to
control what  wastes.   Each  approach will have a  different  influence on the
techniques  that  are used.   In  the  case of a  specific  positive  approach the
linkage will be direct—that is, the technique(s) will be mandated.  However,
in  all  other  cases,  including  those that represent combinations of the types
of  approach,  the linkage is indirect.    For an influencing approach the dis-


                                     68

-------
positions of  wastes are  steered  toward or  away from  certain  control  tech-
niques,  but  all  feasible  techniques  are still  theoretically  availaole.   In
the case of  a specific negative approach, the  available  options ar» reduced
by the  elimination of  one or more  techniques, but  other  factors determine
which  techniques  are  used for what  wastes.   These  factors are  the  nomal
economic forces in which a firm generally minimizes its own (internal) costs,
tempered by  the desire to minimize managerial effort (which is really a cost
to  the  firm).   This   is  equivalent  to minimizing  generator's  cost  (see
Appendix C)  and will  cause firms  to  favor the  use of certain disposal  tech-
niques.  However,  the techniques that are actually used will be influenced by
the actions  of the  various parties-at-interest, and the desires of the firms
to avoid  risk.   Because outcomes have  yet  to be predicted  at  this  stage in
the evaluation process,  only  a tentative allocation  of wastes  to tacnmques
can be made.

     A difficulty  arises  when waste  stream  changes and  treatment techniques
(as opposed  to disposal  techniques)  are being considered.   For  the  common
regional  situation where a wide variety of wastes are produced by many gener-
ators, it will not  be feasible to examine changes  that may occur on most' of
the  generators'   sites.    Changes  in  the  opportunities   for   and  costs  of
disposal   techniques  could cause  generators  to  change  waste  streams  and
treatment methods.  In this  event,  some broad assumptions about such changes
will   have  to be made,  or such changes disregarded (as already not 3d,  waste
disposal   costs  are  usually   only  a  small  portion  of product  value,  which
suggests that onsite operations may not be very sensitive to offsite disposal
costs).  However,  where there  are  major regional  industries  producing sub-
stantial   quantities   of   reasonably   homogeneous  wastes  (e.g.,  petroleum
refining in Texas), further investigation would clearly be warranted.

Development of Threat Scenarios. Etc.  (Step 3)

     The  identification and  description of  threats  has been  discussed in
Section  4  and Appendix B, and the philosophy  of using threat  scenarios  was
expounded  earlier   in  this section.   To proceed with  the analysis,  it is
necessary to  identify  one or more threats for  each  technique  being  consid-
ered.   In  many cases,  it will be  possible   to  establish that,  for  a  given
technique, one  threat  is  of  far  greater  import than  all  others.   In this
event, that  threat scenario  should  be developed  as  fully  as  possible,  and
other  less  significant  threats   can  merely  be  identified.   However,  an
attractive feature  of  the methodology is that  it  provides  a flexible frame-
work  for analysis  that  can  readily  accommodate inputs  from a  variety of
sources.   Waste  management personnel,  for  example,  may  generate the threat
scenarios that  they consider  to  be  the most relevant to a given situation.
If, however,  it  becomes  apparent that the public  is  largely  concerned with
some other threat, an appropriate  scenario can be added without disrupting or
contradicting the previous work.

     The quantitative  data used to describe threat scenarios should probably
be kept  simple.   For  example,  means, modes  and possibly ranges  can  be used
rather than probability distributions for effects.  More detailed data may be
appropriate when the waste management alternatives have been narrowed down to
two or three options.


                                     69

-------
     Environmental  impacts  of  hazardous waste  management  techniques  other
than those that  arise  through threats can also be listed.  These impacts are
all related  to resource  use  (i.e., energy consumption,  materials,  and land
use).   Though  this aspect  of environmental  impact is accounted  for via the
cost mechanisms (for example,  the control cost for land disposal includes the
cost of  land and energy  used),  there  are many who consider  that  the market
prices for some  resources,  such as energy,  may not reflect their true values
to society.  Hence  there  is frequently additional interest  in  resourca use,
and for this  reason, identification of these data is  helpful.

Determination of Economic and Social Effects (Step 4)

     This process,  which leads  to  the  evaluation of some of  the  costs and
impacts  (see  Figure  5)  was  discussed  in  Section  4.    Effects   should  oe
evaluated for  each  of  the techniques involved in  any approacn  being consid-
ered.

Determination of the Impacts on the Parties-at-Interest (Step 5)

     Determination  of  the  economic and social  effects   leads   directly  to
determination of  the parties-at-intereTt!Section 4 provides some generali-
zations about  attitudes  and benavior of the parti es-at-int?r-?st.   Tr,ase data
can  be' used-  to  examine  the  nature  and  degree  of  impact  that  a  vas:a
management technique may have on a party-at-interest.

     Though  predicting  individual  responses  of  the  parties-at-intarast (tne
next step) may  be important,  a general  analysis  of the impacts  of the use of
the various techniques on the parties-at-interest can be  a powerful  tool when
it comes  to  comparing  the  effects  of  the  use  of different  techniques, and
hence, alternative  aproaches.   Table 7  presents  a matrix  of  the  parties-at-
interest for each major technique  and suggests the nature of the effect that
use of  the technique  has on  each party-at-interest.  This  is  a  generalized
matrix and   in  any specific  situation  the effects  could differ from  those
indicated in  Table 7,  and there could be additional parties-at-interest. Note
that where a  reference  level  was  necessary  to  determine the  nature of the
impact, each  technique has been compared with temporary storage  at the gener-
ator's site.    Although  not an acceptable long-term  solution, this  situation
represents a  common starting point.   In other situations, the status quo or a
base case could be used as a reference level.

     It may  be  useful  to examine a  few  of  the entries in Table  7  to under-
stand how judgments about the nature of the effects were  made.  Consider, for
example,  a technique involving chemical  treatment to  reduce hazard potential.
Management of  the firm  generating  the waste  will  have mixed views about the
technique (+-),  as it  is likely to be comparatively costly;  but  treatment
should reduce the risk of an adverse incident.  The firm's workers are likely
to favor treatment (+) because it probably makes  their job safer.   The effect
on the  waste disposal  and  transport industries  will depend on  the process
streams  following treatment  (+-);  they  may have safer wastes  to  dispose of
(+),  or there may be no waste requiring offsite disposal  (-).  Local  officials
are  likely  to  favor  chemical treatment  because of   the  reduced  risk  of  an
environmental  incident  (+),  but water supply  authorities  might be  concerned


                                     70

-------
                           TABLE  7.   MATRIX  OF EFFECTS  ON THE PARTIES-AT-INTEREST
                                                               Party-at-interest
Nature of effects are:
|++|Favorable
r+~| Somewhat favorable
rr—I Mixed, or depends on
L-^-1   circumstances
I-  | Somewhat adverse
|--| Adverse
      t significant
 Hazardous  waste control
        technique
       Other
parties-at-interest
Change in waste stream
Resource' recovery
Treatment - physical
- chemical
- biological
- thermal
- encapsulation
Land application
Landfill - secure
- ordinary
Mine disposal
Lagoon ing
Deep well Injection
Ocean dumping
Engineered storage
Space disposal
* ,
[Discharge to sev/er
Discharge to waterway

+-
+-

.
+-
-
+
+-
•f
-
+-
+
+
-
•—

+
•• +

+-
•»•

+-
+-
+
+
^
+.
+
+-
+
+
+
+

+
+

+-
t-

+-
+
t
+-
++
-
•H

-
-
+
t

-
--

.
+-

-
-
+
+-
+4
+
0

t-
+t
t-
+-

—


+-
+-

.
-
++
+
++
+-
++

+-
+
-
•f

—


++
+

.
-
-
-
_
-
+.

—
-
+-
-

-


+
+

•f-
+-
+
.
+
.
++

-
-
+-
-

-


+
+

+-
+-
++•
+-
+
-
++

-.
--
+-
-

-






.

-
-
—



-
-









.

.

-


-
.





t-
+
+-
+-
-
h
+
-
+
-
•f

.
+
t
44

-








-
•*•-
-


-

+-






+-
+-
-


.

-



--


















-





+
++
+
+
+-
.+-
++
-
+
-
+-f
-.
-
--
+-
--

-













-
--

-


-

.

+



r +-


-
-

-

-

-
--

Virgin material supplier

Chemical suppliers



Farmers


Adjacent mineral intcres
Fowl hunters, farmers

Other nations, politicia

Other nations, pollticia

Sewer authorities
Other water rccreationis

-------
over  the  possible discharge  of  an  undesirable  effluent to  a  river  that
constitutes part  of  a water supply (-).  Though water supply authorities ana
environmentalists  are  likely to have definite views on  most  techniques, tne
perception of  threats  and benefits from chemical  treatment may be remote to
most local residents  (no entry in Taole 7).  Thermal treatment, on the otner
hand,  which  could cause the  deterioration of  local  air quality,  .iiigr.t  ca
viewed negatively  by  many residents and property owners (-),  and a tacr.niaje
such as lagooning  might pose a discernible threat to local water supply users
(-).

     In identifying and evaluating impacts in Table 7,  each party-at-interest
was assumed to represent only one viewpoint.  But  any individual  could fall
into more  than one  category of  party-at-interest;  for example,  one  person
could  be a local   resident,  property owner, worker  and environmental is*.   In
Table  7,  the  attitudes  of  the parties-at-interest  are  "pure;" cor axamole,
business management  is assumed to adopt only  those attitudes characteristic
of  firms  (see  Section 4).   In  the event that the chief  executive  of  a firm
happened to be a strong environmentalist, that particular firm would probably
exhibit  some  mixed behavior.   However,  this  is  not allowed  for  in Taole  7,
where  the  impacts on  waste generators-management  and environmentalists ara
maintained separately.

     To.  apply  the parties-at-interest  matrix  to  a  specific  situafion,  it
might  sometimes   be  more appropriate  to  conduct the  analysis in  tarns  of
approaches  (which  could encompass  more  than  one  technique),  rasper  than
techniques.

Projection of  Responses of the Parties-at-Interest (Step 6)

     Attitudes   that   predispose  the   parties-at-interest   toward  certain
responses  were discussed  in  Section  4,  along  with  some  likely responses.

     Responses include a variety of actions, ranging  from raising the price
of  a  product  to  cover increased  hazardous  waste  management  costs, to public
protest  about  potential  adverse  effects.    Individual  responses can,  to  an
extent,  be predicted  from  a knowledge of  the situation and  the parties-at-
interest.  In  evaluating approaches, it  is useful to note possible responses,
even if these  are  not  certain.  Some responses are in the nature of threats--
for  example,  requirement of costly disposal  techniques  increases  the threat
of  illicit disposal (dumping) of  wastes.

Prediction of  Physical Outcomes (Step 7)

     The  physical outcomes   include  the waste dispositions  and some  of the
responses  of   the  parties-at-interest  (such as  householders  moving to avoid
threats  or  actual  pollution,  or fishermen  avoiding  depleted  fisheries).
Waste  dispositions (including the nondisposal options such as process change
and resource   recovery)  are  largely determined by  the initial  allocation of
wastes to  techniques,  described under Step 2.  If there were no socioeconomic
interaction  (or  policy  level   feedback),  simple  cost  minimization  should
determine  the  ways in  which  the firms choose to distribute their wastes among
the  available  techniques.   However, the responses of  the parties-at-interest


                                     72

-------
may also  affect the  outcomes.   For example,  some  parties-at-interest might
oppose the use  of  certain techniques,  and their actions might thereby render
them  unavailable  to  the  waste  generators  or  cause  them  to become  less
attractive than  others  because  of this opposition.   For these reasons, waste
dispositions other than those based on minimization of a generator's ccst may
be chosen.

     At this stage,  it  is also appropriate to consider how the quantities of
wastes will change  in the future.  Although  data on  the price elasticity of
demand for  industrial waste  disposal  services are rare,  it  can  be expected
that  the  quantities  generated  will  exhibit  some  response  to  price,  as
increased disposal  costs will encourage in-plant treatment, volume reduction,
and resource recovery.   Known plans for new plants or expansions of existing
ones  can  be  factored  in at  this  stage,  but  it  should be  noted  that, waste
generators will  probably use state-of-the-art technology, and  in  some cases
will  replace less  advanced  systems.   Hence even  if economic  activity in the
study area is  expected  to grow, the quantities of  wastes  • requiring disposal
may not increase at the same rate.

     Environmental  threats can  also  be listed as outcomes.   Of course,  only
those that materialize  constitute actual  physical outcomes,  but  it dees ,not
seem  appropriate to  segregate definite (though ill-defined) outcomes such as
those of  ocean dumping  from those such  as  lagoon overflow  that  are proba-
bilistic  in  nature.   All are possible outcomes,  but few  if  any  are clearly
defined.

Enumeration of Costs and Impacts (Step 8)1

     Once the waste  dispositions  are  determined,  it  is  possible  to list all
the costs associated with that particular approach to hazardous waste nanage-
ment.   These include  the generators'  costs that are associated directly with
the disposition of the  wastes,  and the  other costs  of control  (i.e.,  the
administrative and social control  costs).

      In addition  to  these costs,  there may  be some definite environmental
costs or  social impacts  that can be  specified, such as changes, in property
values and noise  insult to  residents along a road leading to a landfill.  As
discussed  in  Section 4,  the dividing  line  between environmental  costs and
social impacts  is  not  clearcut  and is  based largely on  the feasibility of
quantifying costs.

     Many  of  the  environmental   costs  and  social  impacts  arising  "rom1 an"
approach  to   hazardous   waste  management  will  be  associated  with  threats.
These should  be listed  as  part  of  each  threat  scenario, which  should also
include  an estimate  of  the  probability  of  the  threat's occurrence  (if  a
realistic estimate can.be .made).

     Though  all  costs  should  be  specified  in constant   dollars  without
allowing  for future  inflation,  they should-also be discounted using whatever
rate  or  approach is  chosen  (as  discussed  earlier  in  this  section).   Where
threats are concerned,  it will  normally involve choosing a time at which the
threat  is assumed to  materialize.   Where .a  process  such  as leacning  is


                                     73

-------
involved,  it  may  be  possible to  use  engineering judgment  to  decide  the
earliest  likely  time that  a  threat may materialize.  Where  events  are com-
pletely random,  the  analyst will  be forced to  use some arbitrary assumotion
such as halfway through the planning period, or at the end of one generation.
If  the recommended  approach   of  not  discounting beyond  one  generation  is
employed,  threat materialization  at the  end of  one  generation  is  in -nany
respects an attractive choice, since if it occurs afterwards, the actual time
of materialization will not affect the discounted values.

Reiteration of Procedure (Step 9)

     Once the above procedure  has been carried out and the physical outcomes,
costs,  and impacts  associated with  any  approach are  predicted,  thera  is
feedback  to  the policy  level.   An  analyst can examine  the  results  for eacn
approach,  can  test them  against  the policy  objectives, and  can iioai*"y tne
approaches to  improve  the results.   In this way, he can optinn'za the results
within  a  given  approach  by making  one or more  iterations  of the evaluation
procedure  (Steps 2 through 8).   For example,  the analyst  could chance the
number  and location  of  landfills  to  arrive at  a least-cost  land  disoosal
solution,  or he could  change the  levels  of taxes  or  subsidies  to ennanca
effectiveness  or correct the  equity of a  situation.    Once  optimization  of
each approach  is reasonably complete,  the decisionmaker is  in a DOS••v;on to
compare  the  results  of  different approaches.  However,  sometimes  it :iay  ae
necessary to design a new or hybrid approach.

AIDS TO DECISIONMAKING

     Decisionmaking  is  the  third  phase  in  applying  the methodology,  and
involves five steps as follows:

      1.  Array the alternatives;

      2.  Eliminate approaches that are dominated by others;

      3.  Check each approach against policy objectives;

      4.  Examine trade-offs between known costs and threats;

      5.   Select  an  approach,   using  an  appropriate  degree  of risk
          aversion.

The decisionmaking phase is illustrated in Sections 6  and  7.   However, some
aspects of decisionmaking are discussed below.

Arraying the Alternatives

     Once  the  framework  described  above has been applied to  each  approach
being  considered,  the decisionmaker must choose  among  approaches or develop
new ones.  Cost-benefit and risk-benefit analyses usually seek to reduce all
effects  to dollar terms  so that the  alternative that  has  the greatest net
present value  can  be selected.  A simple  refinement would  be to choose only
among those alternatives that also passed certain other  tests, such as equity


                                     74

-------
and-government cost  criteria.   Although the methodology presented-here draws
strongly  on  the  techniques  of  cost-benefit  and  risk-benefit  analysis,
reducing all data  to dollar terms suppresses too much  information for envi-
ronmental planning.

     There have been a  variety of approaches proposed  for  systematizing the
decisionmaking  process  where  there  are  complex  considerations  such  as
multiple objectives.  Some  tend toward the use of  a  utility-based approach,
often  in conjunction  with event  trees  to  cope with  alternative  outcomes
(e.g., Bell, Keeney  and Raiffa, 1977; Wendt and Vlek,  1975;  Fishburn, 1964;
and  Schlaifer,   1967).   This  does,  however,   involve  selecting a  utility
function which then  essentially represents part of the policy  objectives of
the  agency  or  decisionmaker.   The  use of simple  scales (both  ordinal  and
cardinal) for achievement with respect to a number of objectives, and ranking
systems  that  combine such  scales  is  commonly  used in marketing  ard  in cor-
porate planning and  could  be useful here.  Sewell  (1973)  discusses a variety
of evaluation techniques that  have been used for resource-oriented problems.

     The approach  proposed  here is to use a balance  sheet  format which sets
out for each approach the costs and threats,  their effects on the parties-at-
interest, possible  responses of the latter,  and the  physical  outcomes.  'The
decisionmaker is then  in  a position to select  his own trade-offs between the
approaches.   If maintaining  the status quo is  used as  a  base case,  wnatever
approach is chosen should represent an improvement (at least by the decision-
maker's  measuring  rod).   This is  equivalent  to  requiring  that a  project
assessed  by traditional  cost-benefit  techniques  has  a  benefit/cost  ratio
greater than unity.

     Some methods that may be used to handle the data and present them to the
decisionmaker are  illustrated  in  Sections 6 and 7 of this report, which also
discuss  considerations  that may  be involved in making  trade-off decisions.
An  illustration  of  the  selection of trade-offs to achieve  differing objec-
tives (in the context of water supply  planning)  is  included in M'lliken and
Taylor (in press).

Dominant Approaches

     There may  be  some  situations in which one  approach can  be eliminated
from  further consideration because it is  subordinate to  another.  Consider,
for  example,  two projects,  A  and  B,  that are designed  to  achieve  the same
objective (e.g., disposal of wastes).  If the net monetary control costs of A
exceed those of  B  and the environmental costs  of A clearly exceed those of B
(even though the environmental  costs are not quantified), then approach B is
said  to  dominate  approach  A, as A is higher on both types  of  cost.   Hence,
assuming -that  the  only factors that  enter  into the comparison of  the two
projects are the control costs and the environmental costs,  approach A can be
discarded.    Analysis  for   dominance  can  be  a  useful   way  of  eliminating
approaches without  needing to  fully .evaluate  some of the costs  (Fisher-and
Peterson, 1976).
                                     75

-------
Risk Aversion

     The  role  of  risk  in  decisionmaking  is  discussed  in  Appendix  F.
Virtually everyone—firms,  the  public,  decisionmakers, politicians, etc.—is
risk  averse to some  degree.   The  decisionmaker  needs to  reflect an appro-
priate  degree  of  risk aversion in  his  choice  among alternatives.   In making
this  choice,  he will  generally have to trade-off added costs against reduced
probabilities  that environmental  threats will  materialize.   The added costs
will  usually  be known with comparative certainty, whereas the threats .-nay be
quite  ill-defined.   An additional  complication is  that  known  costs  nay be
borne by one party-at-interest, while the risks may fall on another.

     A  decisionmaker   should  remember  that  if individuals  feel threatened,
even  if the threat does not materialize, then their welfare is reduced (i.e.,
feeling threatened is a cost).  On  the other hand, it  is probably not reason-
able—even  if  feasible—to  achieve a situation that  is  virtually  risk free,
since  it  is likely that  in many  cases, the marginal  cost  of risk reduction
'increases as the level of risk is reduced (Tihansky and Kibby, 1974).

     The  role  of  risk  aversion and the nature of  risk-related decisions in
waste  management  are  illustrated  in  Sections  6 and 7.   Though  Appendix F
contains  useful  background  information  on  public  attitudes  toward  risk-
taking,  it does  not  offer much  specific  guidance  on  the  degree  of  risk
aversion  that  should  be  incorporated  into environmental  decisions.   Hence
with  the  present  state  of knowledge, selection of  an appropriate degree of
risk  aversion  remains the responsibility  of  the  decisionmaker,  wno  must
subjectively incorporate public feelings into his judgment.

Equity

      Equity is  a normative  facet of economics.   For example, one viewpoint on
equity  is that  potential beneficiaries should pay to obtain that benefit.  An
alternative approach  is  that cleanup costs should be  borne directly by those
who  cause the  environmental  degradation.   Yet another  view  is  that unless
there  are  exceptionally  powerful  arguments  in  favor of  the  action  it is
unreasonable  to   drastically  change  business  competition  by banning  (or
rendering highly costly) a  particular industrial process, for example.  These
three  viewpoints  of  equity could easily be in conflict, calling for judgment
on  the  equitability  of   the  various  outcomes  arising  from  alternative
approaches  to waste management.

      The  identification of  the parties-at-interest is particularly useful in
this  respect,  as  it is comparatively easy to compare  the effects of alterna-
tive  approaches on each of  the parties-at-interest.   By examining the ways in
which  costs and impacts fall on different parties-at-interest, the decision-
maker  can  evaluate  the  acceptability  of  the  results.   He can  also devise
strategies  to render  a  given approach  equitable  by  finding  a way to shift
some  of the  costs and impacts from one party-at-interest  to another.   For
example,  examination  of  the  alternatives  for  the  disposal of  a  particular
waste might lead to the conclusion  that economic efficiency would be achieved
by discharging  this waste to a landfill, but that this could render the water
unsafe  to drink in a limited number of wells.  To make this solution
                                     76

-------
equitable, the waste  generator could be made to pay for the cost of .in-stall-
ing  and operating  an  alternative  water  supply,  possibly together  with an
additional  payment to  compensate the  well  owners  for  a loss  of aesthetic
value  caused  by changing  from their well  water to  the alternative supply.

Summary of the Methodology

     The  complete  procedure  involved  in  defining  the  scope of  the study,
applying  the  analytical  framework,  and  deciding  among  alternatives  is
summarized in Table 8.

	TABLE 8.   SUMMARY OF THE METHODOLOGY	


PHASE I.       OBTAIN PREREQUISITE INFORMATION

               1.   Define scope of study;
                    a.  Geographic area;
                    b.  Types of wastes;'
               2.   Inventory existing waste situation;
               3.   Determine how wastes are currently controlled;
               4.   Ascertain policy objectives.

PHASE II.  APPLY ANALYTICAL FRAMEWORK

               1.   Develop alternative approaches for hazardous waste
                    management.  (Consider the status-quo as a base case.)

               For each approach under consideration:

               2.   Allocate wastes to techniques;
               3.   Develop threat scenarios, list other
                    impacts (resource use);
               4.   Determine economic and social effects;
               5.   Determine impacts on the parties-at-interest;
               6.   Project responses of the parties-at-interest;
               7.   Predict physical  outcomes, including future wastes;  .
               8.   Enumerate costs and impacts (discount as appropriate);
               9.   Reiterate Steps 2 through 8 until results for each
                    approach have been optimized.

PHASE III.      MAKE DECISIONS

               1.   Array alternatives;
               2.   Eliminate approaches that are dominated by others;
               3.   Check approaches  against policy objectives (e.g., for
                    equity);
               4.   Examine trade-offs between known costs and threats;
               5.   Select an approach,  using an appropriate level of
                    risk aversion.
                                     77

-------
                                  SECTION 6

            APPLICATION OF THE METHODOLOGY TO A SINGLE WASTE STREAM
INTRODUCTION.

     This section  provides  a  simple  example of  the use  of  the methodology
described in Section  5.   This example considers only a  single waste stream,
and hence much of Phase I of the methodology (obtaining  prerequisite informa-
tion for  analysis) is  inapplicable,  as  Phase I is  oriented  toward ccmoiete
studies of hazardous  waste  management within a specified  area.   The example
concentrates on  applying the  analytical  framework  (Phase II)  and  on illus-
trating  the decisionmaking  process  (Phase  III)-    Though   the examola  is
hypothetical,  it was  inspired by the situation in Oregon analyzed in Sscvion
7.    Familiarity  with  Section 7  is  not  necessary  to  follow  this  section.

THE PROBLEM

     An agency responsible  for hazardous  waste management receives an apoli-
cation from a  firm that wishes to  dispose  of a hazardous waste oy deep well
injection.  The agency has no policy or regulations that specifically ban the
use  of  deep  well  injection, but  any technique  used  for   hazardous  waste
disposal requires agency approval.

     The waste will come from a new process that is assumed to have a 20-year
technological   life.   The  process  will generate  250,000  m3  per year  of  an
aqueous-waste .containing 50  parts  per million (ppm) of nondegraaable toxic
elements (e.g., heavy metals).  The firm proposes to dispose of this waste by
injecting  it  into a  saline aquifer  some  600 m below  their  premises.   They
estimate  that this   will   cost  them  $50,000  per   year   (including  capital
charges) over the 20-year life span.

     The  first step  is  to  investigate the technically feasible alternatives.
These are found to be as follows!

     1,   The waste stream  can be  reduced to 25,000 m3  per  year with a
          corresponding  increase in the  concentration of toxic elements
          at a cost of $20,000 per year to the firm.

     2.   The waste stream  can be treated to provide an effluent that is
          acceptable  to  the municipal sewer.  Treatment  results in 250
          m3 per  year of  a toxic  sludge.   The cost of treatment plus
          effluent charges would be $115,000 per year.
                                     78

-------
     3.   There are  two landfills  that  could accept  either the liquid
          waste  from alternative  1 or  the  sludge from  alternative 2.

          The  local   landfill   is   a  public  sanitary  landfill  located
     immediately adjacent  to a  river 50  km  from the  generating  fir-n.
     Rainfall  in  this   region   is  much  greater than  either open  can
     evaporation or potential evapotranspiration.  This landfill  charges
     $3.00 per m3 for any waste.

          The  secure  landfill  is  a chemical  landfill  located  in  a dry
     zone  (rainfall  is  much less  than  evaporation  or evapotranspira-
     tion), 360 km from the generating firm.   The gate fee is $30.00 per
     m3 for the sludge and $20.00 per m3  for the liquid.

          Transportation to either  landfill would be by truck, at a cost
     for either sludge  or  liquid of $7.00 per m3 to the local landfill,
     and $22.00 per m3 to the secure landfill.

     4.   Other techniques for  dealing with  the waste (such as resource
          recovery or ocean dumping) are  not feasible.

     Hence there  are five technically  feasible disposal  plans,  as follows:

     A.   Deep well injection on the firm's premises.
     B.   Sludge sent to the local landfill.
     C.   Concentrated liquid waste sent to the local  landfill.
     D.   Sludge sent to the secure landfill:
     E.   Concentrated liquid waste sent to the secure landfill.

THREAT SCENARIOS

     The next  step is  to  develop likely threat scenarios,  at leas" one for
each disposal  plan.   In-plant  accidents  under any plan  are  expected to have
approximately  similar  impacts  and  hence do  not  have to  be evaluated.  The
following scenarios represent the major threats identified,  •

Threat Scenario I:   Water Contamination From Deep Well Injection'

     This  scenario  applies  only   to  Disposal  Plan  A.   Drinking  water  is
obtained from numerous wells that penetrate a shallow aquifer in the vicinity
of the generating  firm.   This  aquifer may become contaminated as a result of
some unanticipated interconnection  with  the  deep saline aquifer.  The proba-
bility  of  this cannot  be determined.   If contamination  occurs,  corrective
action  could be  taken  by providing  temporary water  supplies  to  the  local
residents  and  by drilling several  additional wells into  the saline aquifer
and counterpumping to reverse  the migration of the waste.  The total cost of
this cleanup operation,  including some hospitalization costs, is estimated to
be some $2.4 million, which  would be considerably less costly than providing
a permanent new water supply to the local residents.
                                     79

-------
Threat-Scenario II:  Leaching From the Local Landfill

     This  scenario applies  only  to  Disposal Plans B  and C.   If the concen-
trated  liquid  waste  were  discharged  to the  local   landfill,  it  could be
assumed that  the  entire waste would quickly infiltrate the river, because of
the, wet  conditions and absence of leachate barriers at the landfill.   If the
sludge  were  deposited at  this   landfill,  an  appreciable  proportion  of the
toxic  elements  would probably be retained,  especially in the earlier  years;
but the  above assumption could be used  as the worst  case.   Either  sludge or
liquid^waste  contributes  12.5 tonnes of  toxic  elements  per year.   The river
has a'mean flow  of  200 m3/s, implying a  toxic  element  concentration  of two
parts  per  billion (ppb) if  the waste were uniformly diluted.  However,  local
concentrations  are expected  to be higher.

     The  river supports  an  important  salmon  fishery,  and  experts value  a
typical  year's  fishing  at  $800,000,  excluding  indirect  effects  such as
tourist dollars brought into the region  by  the fishery.   The experts  expect
the onset  of  high fish mortality  to  occur at  toxic element concentrations
somewhere  between 20 and 100 ppb, but  they are reluctant  to say what  impact
two ppb would  have  on the  salmon because of the effect variability with
duration of  exposure,  alkalinity, and the  presence  of other elements.  They
point  out, however, that  there  is  some  evidence that  fish 3void  sublethal
concentrations  of toxic elements; hence the waste could  conceivably ruin tne
fishery by discouraging the  salmon from returning to spawn.

Threat Scenario III:   Transport Accidents

     This  scenario is  applicable to Disposal  Plans  B  through E.  Statistics
indicate that 50  accidents  involving waste spills can  be  expected per billion
kilometers traveled  by truck in the region.  The cleanup  cost associated with
a  typical  accident is estimated  to be  $10,000.   Serious injuries and  deaths
directly  attributable  to  the  properties of  the waste  are expected   to be
negligible.

Threat Scenario IV:  Flash  Flood  at the Secure  Landfill

     This  scenario  is  applicable only  to Disposal Plans  D and E.   The most
likely  threat  from  the secure landfill  is  contaminated  runoff from a flash
flood.   Such  a flood  is expected  to  occur less than  once  per hundred  years,
.and damage along  the  flood  path  directly attributable to the toxic elements
is  expected  to be minimal.   Leaching problems are highly unlikely because of
the dry. climate and  the-extreme depth to  usable  aquifers.

ANALYSIS OF THE ALTERNATIVES

     Table 9  presents  a comparison of the  alternative  plans.  Only the  gener-
ator's cost portions of control costs have been  evaluated,  as other  costs are
not expected to  differ significantly between  alternatives.   The net present
value  of  the control  costs  has  been  calculated by discounting  at 10 percent
per year.   (The  Office of  Management  and  Budget  recommends the  use   of 10
                                      80

-------
                             TABLE  9.   ANALYSIS  OF  FIVE  ALTERNATIVE WASTE  DISPOSAL PLANS


Control costs:
Generator's annual costs:
Ons ite operations
Transport
Gate fees
Deep well
injection
Plan A

$ 50,000
Local
Plan B
(sludge)

$ 115,000
1,750
750
landfill
Plan C
(liquid)

$ 20,000
175,000
75,000
Secure
Plan D
(sludge)

$ 115,000
5,500
7,500
landfill
Plan E
(liquid)

$ 20,000
550,000
500,000
               Total
                              $   50,000     $  117,500   $  270,000    $  128,000     $1,070,000
00
Present value of generator's
  total cost (20 years'  cost
  disco.unted at 10 percent1
 4>er year)                   $  426,000
                                                      $1,000,300    $2,298,000     $1,089,700     $9,109,500
         Other control  costs
           (Administration and monitoring)      ND*

       Environmental  impacts

         Resource use                          ND
         In-plant accidents                    NO
                                                     ND
                                                     ND
                                                     ND
 ND
'NO
•NO
ND
ND
ND
ND
ND
ND
       * Not significantly different.

-------
 percent per year as the opportunity cost of capital [Executive Office of the
President, 1972].)

     As soon as  the control costs are evaluated, it is possible to ernr'nate
Plans C  and E  because  these  plans  are dominated by 8  and  D,  rescecfively.
Consider Plan B versus Plan C.  The control costs for B are 5117,500 ser year
versus $270,000  per year for C, and detailed  evaluation is  not necassary to
show that the  potential  environmental  damages from B are also less than from
C.   There is  less  possibility for release  of  toxic elements from the sludge
(Plan B) than from the concentrated liquid (Plan C); and Plan B requires less
transportation  than Plan C,  which should  result  in fewer  accidents.   Thus
Plan B  is clearly  preferable to  Plan  C as  both  the quantified  costs (the
generator's costs)  and  the  nonquantified costs (the potential  for environ-
mental  damage)  are  lower for Plan B than Plan C.  Similar arguments aooly to
Plan 0 versus Plan E.  This approach cannot, however, be used to compare PUn
A  with  any  other  plan,  as  the  environmental  threat from  Plan A  is  quite
different from those of all other plans.  This reduces the number of plans to
be evaluated to three (A, B, and D).

     The next step  is to examine the threats associated with each plan and to
determine the  nature of  the effects of each plan on the parties-at-intsrest.
Table 10  compares  the major effects for the three plans.  From tr,-'s analysis
the expected attitudes of the parties-at-interest can be projected, ana these
are also summarized  in Table 10.

     The  parties-at-interest  most strongly affected  in  this example include
the water supply authority, and to a  lesser  extent the water users near the
plant,  who  would  be concerned about  the threat  from deep well  injection
(Threat Scenario I). Fish experts would oppose Plan B, although fishermen and
fish-related  industry might perceive only a weak  threat.   Fishing interests
might,  however, have  an unexpected  ally.   If  Plan  A were  prohibited,  the
generating firm  itself could well prefer Plan 0 over Plan B.   Though the firm
will strongly favor  Plan A because of its low cost, the annual cost of Plan D
is only  $10,500 greater  than Plan B,  and  if  the firm opted  for  B,  it would
receive adverse  publicity  if the fish threat (Scenario II) materialized.   In
contrast, the  firm  might  enhance its  reputation  as a  responsible  environ-
mental citizen  if  it sent  its waste to the secure landfill under Plan D.   In
this case a great  deal  is  at stake  (an  $800,000 per  year fishery and the
firm's  image)   for  only a  small  net  benefit  ($10,500  per  year  in reduced
costs).    Although   the  probability  of  the  fish  threat   materializing  is
unknown,  the  firm  would  not have to be very  risk averse to prefer Plan D to
Plan B.   -Note  that all  parties perceive negative, impacts for  Plan  B,  which
confirms the general unattractiveness of this plan.

DECISIONMAKING

     If Plan B  is dropped from further consideration, the remaining choice is
between  Plans  A and 0 and  involves  reduced control  costs and greater damage
potential if  A is  preferred to D.   If Plan A is selected, the present value
of  the  control  costs  discounted over  the  20-year project  is  $663,700 less
than for  Plan  0 (i.e.,  $426,000 versus $1,089,700).  On the other hand, Plan
A  poses  the  threat of water contamination  (Threat Scenario  I) with  its


                                     82

-------
                            TABLE  10.   COMPARISON OF THREE SELECTED WASTE DISPOSAL PLANS
                                        Plan A                   Plan B                      Plan D
                                 Deep well  Injection    Sludge to local landfill   Sludge to secure landfill
85
      Present value  of  gener-
      atr's  total  cost:

      Major  threat scenarios:   I:
    $ 426,000

Water supplies con-
  taminated
Mitigation cost:
  $2,400,000, plus
  cost of changing
  plan (There is some
  risk that the miti-
  gation measures
  might be un-
  successful)
Probability:   not
  estimated
      Other economic  effects
      Effect  on  parties-at-interest*

        Generating  firm's management
        Local  water supply  users
        Water supply authority
        Fish  experts
        Fishermen,  fish-related
          industry
        Environmentalists
      $ 1,000,300

II:   Salmon fishery at
        risk
      Direct value:
        $800,000 per year
      Indirect effects
        locally important
      Probabi1i ty:   not
        estimated
 III.  Transport accident
      Cleanup cost:
        $10,000
      Probability:
        2x10 Vyear

      Product prices
        slightly higher
        than with Plan A
           -1+
       $ 1,089,700

III:   Transport accident
      Cleanup cost:
        $10,000
      Probability:
        1.4xlO"4/year
 IV.   Contamination via
        flash flood
      Damage potential
        minimal
      Probability:
        10 Vyear
                                                      Product prices
                                                        slightly higher
                                                        than with Plan A
      *  Key to  attitudes:  ",  favorable;
        not significant.
          somewhat favorable; -, somewhat adverse; --, adverse; (blank),

-------
cleanup costs  and the  need to find an alternative waste  disposal  method if
deep  well  injection  does  contaminate  the water  supply.   The  threats from
transport accidents  (Scenario  III) and from flash floods at the secure land-
fill  (Scenario  IV)   appear to  t>e  so  minor   that  they  can be  neglected.
Nevertheless,  it  was  important to recognize them and  demonstrate  (or obtain
consensus judgment) that they could be disregarded.

     If Plan A were adopted, and problems with the deep well scheme developed
after  five  years,   the additional  costs  of  taking  the  corrective  action
described  under  Threat Scenario  I and  switching  to Plan  0  for the  next 15
years would  have  a  present value of $1,895,000 when discounted at 10 percent
per year.  (Note  that it was  necessary  to  select  a time at which the threat
is assumed to  materialize in order to be able to calculate a present value.)
Thus, assuming that  the  mitigating measures  prove  successful  and  that  the
data  above  completely and accurately  represent  the  choice,  the  economic
question becomes:   Is it worth risking an unknown probability of future costs
that  have  a  present  value of $1,895,000  to  save certain  future  costs that
have a present value of $663,700?  If the decisionmaker disregards  equity and
is  not  risk  averse,  he will  favor  Plan A when its expected value  is lower
than that  of Plan 0.   This will occur if the probability of water contamina-
tion  (Threat  Scenario  I)  is  less  than  35  percent  (i.e.,   S663.700  -r
$1,895,000).

     A 35 percent probability that contamination will occur seems quite high.
There is no  known reason why contamination should occur, so on this basis, a
decisionmaker  who is  not  unduly risk averse  would probably  favor  the deep
well  injection plan.    However,  he must consider  some other  factors  before
making a final judgment.

     First,  there is  a slight possiblity that if water contamination occurs,
the  mitigating measure  of drilling  additional  wells  and  counterpumping to
reverse waste  migration might be unsuccessful.   In this event, a  new water
supply would have to be piped in  at  a cost of tens of  millions  of dollars.
Although  this  possibility  is  not  formally  analyzed,   its  existence  will
encourage the  decisionmaker to be risk averse.

     Second,  the  decisionmaker should consider equity.  If he  favors  Plan A
over  Plan  D., the  waste-generating firm will gain economically,  but the local
residents  will be  at  risk.   However,  should the  threat of  contamination
materialize,   the  water  supply   authority  could  bring  a  suit for  damages
against the  generating  firm.  Hence although at first sight Plan A is inequi-
table  because benefits  and  risks accrue  to  different  parties-at-interest,
there is a"mechanism  that might be capable  of  redressing this inequity.  Note
that  benefits to the  firm will ultimately be  returned to  society  through
lower prices or  higher net income, so a decisionmaker who  takes the societal
view  will  not necessarily  oppose a plan that benefits  a  firm  while putting
the public at  risk.

     Finally,  the decisionmaker must consider  the less tangible factors.  Are
the  local  residents  and  environmentalists highly disturbed  about  the waste
injection  proposal?   If so, they will be  subjected  to psychological  damages
not accounted  for in the dollar costs discussed above.  How important is the


                                     84

-------
option value associated with not contaminating the saline aquifer?  Are there
any other  factors that  have  not been considered?  Public  hearings  could be
used to gauge the strength of local  feelings and concerns.

     Even if a  decisionmaker  does  not consider the personal risk of making a
choice that  is  later  perceived  to  be  a poor  one,  the issues  are  ccmolex.
Excessive risk  aversion  will  reduce society's welfare just as excessive risk
proneness will.   Each  decisionmaker must formulate his  own trade-offs among
the various  factors.   However,  it  is hoped that by systematically laying out
and  analyzing   the  principal  features  involved  in   the   alternatives,  the
decisionmaker1 s task  can be  made easier.   He must, however,  still  make the
decision.
                                     85

-------
                                   SECTION 7

               CASE STUDY OF HAZARDOUS WASTE MANAGEMENT IN OREGON
INTRODUCTION

Rationale Behind the Choice of Oregon

     The methodology described earlier could be applied to several categories
of problems.  Those considered for a case study included:

     1.    A study of a complete local or regional situation;

     2.    Analysis oriented toward a specific industry;

     3.    A  study  of  a  specific  material  that  may   be  pr-esant  in
          hazardous wastes;

     4.    A study  of a  specific  means of  disposal  and  altarr.atv/es to
          this.

     The  topic  chosen  was  a  study  of  alternatives for  hazardous  waste
management in the State of Oregon.  There were two principal reasons for this
choice.    First,  EPA  suggested  that  the  methodology   be  demonstrated  by
application to a State, as this would be most useful when the RCRA was imple-
mented.   Second, Oregon had good information on hazardous waste generated and
disposed  of  in the  State,  which  provided  the necessary  data  bass.   In
addition,  the  hazardous waste  situation in Oregon  is not  so  complex  as to
present an  intractable  example,  but the presence of some regional wastes and
climatic differences within the State provide an interesting analysis.

Some Characteristics of Oregon

     Oregon has a total area of 251,000 km2, and divides naturally into three
major parts separated  by  two mountain ranges.   The coastal area is separated
from the Willamette Valley and southwestern valleys (of the Rogue and Klamath
Rivers)  by the  Coast  Range,  while  the High  Plateau of eastern  Oregon is
separated  from the  Willamette Valley  and  the  southwestern valleys by the
Cascade  Mountains  (see  Figure 8).   The  coastal  area  and Coast  Range are
comparatively wet (e.g., 150 to 255 cm of precipitation per year), while much
of the  Willamette  Valley  and southwestern valleys receive 102 cm per year or
less.   Beyond  the  Cascades,  most of eastern Oregon  is dry, with large areas
averaging less than 30 cm per year (Sternes, 1967).
                                     86

-------
S3

                                                                              __L
                     (JALIt'OiiNlA
                                   Figure 8.  The geography of Oregon

-------
     In  1975,  Oregon's  population was  approximately 2.23  million  persons.
Much of  the  State's  population lives in the Willamette Valley, including the
Standard  Metropolitan Statistical  Areas  (SMSA's) of  Eugene-Springfield ana
Salem," together  with the  Portland metropolitan  area,  which is  st  tr.e con-
fluence  of  the  Willamette  and  Columbia  Rivers.   The  1975 copulation  of
Oregon's  portion of  the Portland SMSA was  0.94  million,  and the State tcta'i
for  the  three SMSA's  was  1.38 million (U.S. Department  of  Commerce,  1977).
Most  of  the  remainder  of the  population  lives  along  the coast, along -.he
Columbia  River,  and  in the southwestern valleys.  Industrial (manufacturing)
activity largely correlates with population, being particularly strong i-n the
Willamette Valley.

     The  main  communication routes  are  east-west along  the Columbia River.
which forms  much of  the State's northern boundary, ana north-south along the
Willamette River  and southwestern  valleys.   The  distriogtion  of Doou'.at-.on
and  the  transport  corridors  results in the economic isolation of Oregon from
the  adjacent  States  of Idaho and Nevada, and  in part from California,  riow-
ever, there  is  substantial  interaction with the  State  of Washington,  as the
Willamette  Valley  transport  corridor  extends  north in  Washington  to  the
Seattle area, and Vancouver (Washington) is the major city immediately across
the Columbia River from Portland.

     The  1970  census  of  population  revealed  that  civilian employment  in
Oregon totaled  788,500, or  37.7 percent of  the population,  distrioutad  by
industry as shown in Table 11.

            TABLE 11.  EMPLOYMENT BY INDUSTRY IN OREGON
	AND THE UNITED STATES*	

                                          Percent of total employment
Industry
Agriculture, forestry & fisheries
Mining
Construction
Manufacturing
Transportation, communications &
other public utilities
Wholesale & retail trade
Various services
Industry not reported

Oregon
5.3
0.2
5.4
20.4

6.8
20.9
36.1
4.9
100.0
United States
3.5
0.8
5.5
24.4

6.4
18.9
34.4
6.1
100.0

* Source:   U.S.  Department  of  Commerce, Bureau of  the Census,  1973.  (Data
  relate to employed persons aged 14 years and over.)


If employment for Oregon  is  compared to  that for the entire United States, it
will  be  seen  that Oregon has a  lower proportion of its work force engaged in
manufacturing  and mining, a significantly  higher  proportion  in agriculture,
forestry, and fisheries,  and somewhat more in trade and services.

                                     88

-------
      The  methodology  described  in  Section  5 will  now be  applied  to  a  study of
 hazardous waste management alternatives in Oregon,  insofar as  possible  fol-
 lowing  the  procedure  outlined in Table 8.

 Hazardous Wastes  in Oregon

      Three  major  sources  of  data  about hazardous  wastes  in  Oregon  .vere
 available:   A  survey conducted  by  the  Oregon  Department of  Environmental
 Quality  (DEQ)  in  1972  and  1973,  a study  conducted  by  Battelle  Memorial
 Institute,  Pacific Northwest  Laboratories (Battelle)  in  1975,  and  data  on
 wastes  actually shipped  to  a chemical waste disposal  facility  near  Arlington,
 Oregon.

      The  survey  conducted by the  Oregon OEQ  (State of  Oregon,  1974)  identi-
 fied  major Oregon  producers  of wastes  that  were  subjectively  judged -:o  be
 potentially hazardous.   Total  hazardous  waste  generation  in Oregon was  in
 most  cases  estimated by extrapolation  on the  basis of  employment  in-  each.
 industrial  category.

      The  study  conducted  by  Battelle  (Stradley,  Dawson, and  Cone,..1375)
'endeavored  to  identify all  significant generators of  hazardous  was.tei-wTthi.n.
 EPA  Region X  (Oregon, Washington,  Idaho, and Alaska).   As far as possible,
 this  study  used the Battelle hazardous waste  decision  model  (-see Figure 2)  or
 the  inclusion  of  a waste component on  the EPA proposed  list  of nonremovable
 hazardous substances   (U.S.  Environmental  Protection Agency, 1974a) as  its
 criteria  for identifying hazardous wastes.

      Data were available on  wastes  actually shipped to the  secure  disposal
 facility  operated  by Chem  Nuclear  Systems,  Inc.,  near Arlington,  Oregon.
 (Arlington  is  located in Gilliam  County,  adjacent to  the Columbia  River—see
 Figure  8.)  This  facility, which includes  a chemical  landfill  and "neutraliza-
 tion/evaporation  lagoons,  commenced regular  operation   in May  1976 and  is
 currently the  only chemical landfill in Oregon.   Its geographic location  is
 such  that it is likely to receive  almost all  the  Oregon-generated wastes  that
 are disposed of in a  secure disposal  facility.  For  the  period covered by the.
 data  used  in  this study  (1975-77),  the  only categories  of nonradicractive
 hazardous waste  that  were  legally required  to be  disposed of at a  chemical
 landfill  were  pesticide wastes  and PCB's.*  Hence  most  Oregon-originated
 wastes  that were sent  to  Arlington were -not   specifically  required  to  be
 disposed  of in that way.   However,  all.wastes that  are  sent to  Arlington are-
 subject to  Oregon DEQ  permit..
* The  Oregon  statutes that  related  to  "environmentally  hazardous wastes"
  (Oregon  Revised  Statutes  [ORS]  459.410-459.690,  first  enacted  in 1971)-
  required  such  wastes, to be designated  by  the Oregon DEQ.  Until  May 1979,.
  only waste pesticides, pesticide manufacturing residues,  and PCB's had been
  so  designated.   (Personal  and  telephone.interviews,  F.  Bromfeld,  Oregon'
  DEQ,  Portland,  Oregon,  March 9, 1977; April 17,  1979.)   Subsequently,   in
  May 1979,  the  Oregon Environmental  Quality Commission  adopted a comprehen-
  sive   set  of   criteria  for  establishing  wastes  as   hazardous  (Oregon
  Administrative Rules [OAR]  63-100 to 63-135).

                                     89

-------
PREREQUISITE INFORMATION AND DECISIONS

Scope of the Study

Geographic Area—
     The geographic area  of the case study was taken as the State of >3gon.
This choice  is  convenient because the State is a complete political unit, and
because its  industrialized  area (the Willamette Valley) is isolated from the
industrialized areas of all adjacent States except Washington.

Types of Wastes-
     Industrial  process  wastes  were  the  primary  category considered,  but
waste  laboratory  chemicals  and unwanted pesticides were  also  included,   ~he
case study  excluded  radioactive wastes,  mining wastes, pesticide conta-i.-ars.
and hospital wastes.   Apart from the possibility of some reactive (axa'icsi ,-e;
wastes  at  the  Umatilla  Army Depot  (these  wastes would  not fall  Linear CEQ
control) no special wastes were known to exist in the area.

Inventory of Existing Waste Situation

Quantities and Types of Wastes Generated—
     Table  12  presents   data  derived  from  the  Battelle  study  ui.-
Dawson, and Cone, 1975).   The data, which were collected in  mia-1375,
believed by the authors  of that study to amount virtually to a census of a"
potential'  hazardous  waste  generators  in  Oregon.    (Personal  intarviaj,
G. Dawson  and  J,  McNeese,   Battelle  Memorial  Institute,  Pacific  Northwest
Laboratories,  Richland,  Washington, April  25, 1977.)   The data  took account
of  the earlier survey by the  Oregon DEQ (State of Oregon, 1974).   However,
the' data  in  Table 12 cannot  be compared  directly  with  those  published by
Stradley,  Dawson, and  Cone (1975)  because  of slightly  differing  coverages
(e.g.,  wastes  recycled within  a plant were excluded  from Taole 12).   Since
then,  in  1978-79, the  Oregon  DEQ has conducted another  survey, but the
results of  that  survey  were not  available  in time to be  used  in  this  case
study.

     In Table  12,  hazardous wastes have been divided by the author into five
categories:   Pesticides,  heavy metals,  solvents and oils, phenols,  and other
chemicals.   In practice, a given waste often fell into more than one of tnese
categories.   When  this   occurred,   the  waste  was attributed  to  the  first
category that was  represented by more than trace quantities.  The sequence is
intended'to be crudely  hierarchial  in terms of the  threat or potential for
damage  that the waste category is capable of causing.  Clearly, even if only
intrinsic  factors (i.e., the  hazardous  properties of  the waste itself, not
the  potential  magnitude  of effects)  are considered,  a  rigorous  hierachy
cannot  be  established,  as this would depend on the actual materials involved
and  their  concentrations.   This   is  particularly  true  for  the  catch-all
category of "other chemicals,"  although the bulk of  this material  does not
appear  to  be very hazardous.  Much  of this  material  arises from the primary
aluminum manufacturing  industry and is spent potliners (carbon) contaminated
with  fluoride  and cyanide,  and a  calcium  fluoride  sludge.  Most of the rest
is  lime sludge.
                                     90

-------
TABLE 12. GENERATION AND DISPOSAL OF POTENTIALLY HAZARDOUS WASTES IN OREGON*?

                                  Amount of waste (m3 per year)


                                Heavy   Solvents            Other
       Item        Pesticides   metals  and oils  Phenols  chemicals  Total
Waste generated by industrial
Pesticides and
wood treating 995
Resin & glue
Paints & inks
Other chemicals
Batteries
(lead-acid)
Electroplating
Electronics, etc.
Metals manufacture
Miscellaneous 	 -
Total 995
category:
-
-
-
-

109
74
1,723
517
155
2,578

-
26
113
'474

-
-
2
-
143
758

-
1,112
-
-

-
-
-
-
-
1,112

-
-
42
1,3-12

-
-
-
10,796
	 4
12,154

995
1,128
155
1,786

109
74
1,725
11,313
302
17', 597
Waste disposed by method:
Onsite
Landfill
Storage
Lagoon
Sewer
Local landfill
Secure disposal
(chemical landfill)
Resource recovery
Hauler/unknown
Total

-
23
.492
-
-

441
-
39
995

1,310
464
22
7
78

-
622
75
2,578

-
24
-
-
20

-
669
45
758

34
358
10
-
-

20
690
-
1,112

1
2,307
4,149
2
33

-
5,552
10
12,154

1,345
3,176
4,673
9
131

461
7,633
169
17,597

* Source:   Adapted   from  Stradley,  Dawson,  and   Cone,   1975  (including
  unpublished data sheets prepared in conjunction with the report).

t These  data  were developed  using  criteria  for hazard  potential  different
  from those employed  by the  Oregon DEQ.   Only wastes generated on a regular
  basis are included.
                                     91

-------
     Industries have been grouped into nine categories in Table 12, partly on
the basis of the types of wastes generated, and partly on the basis of indus-
trial similarity.   The  first two categories, pesticide  manufacture  and wood
treating, and  resin  and  glue manufacture, were chosen to include all regular
generators of waste pesticides and phenols, respectively.  In addition to the
categories of  wastes listed  above,  small quantities  of PC3's  require  dis-
posal, but  as  they are  not  generated  on a regular basis they were  excluded
from Table 12.   Oregon regulations require that small  quantities of PCS's and
contained PCB's be disposed of in a chemical  landfill, i.e., at the Arlington
disposal  facility, whereas bulk PCB's  are currently being  stored  (for even-
tual incineration) or sent out of state.  (Telephone  interview, F. Blomfeld,
Oregon 0£Q,  Portland, Oregon, August 23, 1979.)

     Table 12  also  shows the 1975 disposition of  these  wastes.   Wastes that
were recycled within a plant  have been excluded,  as well  as  regular effluents
legitimately discharged  under  the  National  Pollutant Discharge Elimination
System (NPDES)  or  the  sewer  regulations in effect  at  the time of the study.
There was no deep well  injection or ocean dumping.   Note that a comparatively
high proportion of the  wastes  (43 percent)  was being subjected to  resource
recovery.  Two main factors account for this.   First,  over half this  material
was spent potlinings  sent from an Oregon .aluminum  plant to a plant  in v/ash-
ington (operated  by the  same company) for cryolite recovery.   Even if tm's
major waste  stream is subtracted,  20  percent of all Oregon  Hazardous waste
(originating from  32  firms) was  still  being  sent  for resource  recovery.

     The  relatively high proportion  of resource-recovered wastes (although
not  strictly  comparable, the  U.S.  proportion is 1.7 percent  [U.S.  Environ-
mental Protection  Agency, 1977b]) appears to  relate to the  presence  of  a
resource  recovery  infrastructure  in  the  Pacific  Northwest.   This  in  part
probably  reflects  the  ecological  consciousness  of  the area.   In  1975-77,
Oregon had an oil re-refiner in Portland and three solvent reclaimers (two in
Portland, and  one in Eugene),  plus a number of wax recyclers  in Portland.
Although there  was  no  longer a secondary lead smelter in Oregon (the nearest
was in Seattle, Washington),  there was one Portland firm specializing in lead
recovery,. primarily serving  the  lead-acid  battery  manufacturers.   Perhaps
more  significantly, however,  there were two  specialist  resource  recovery
firms in  Seattle,  Washington,  that accepted a wide variety of materials for
processing (Stradley, Oawson, and Cone, 1975).  In addition  to materials that
are  frequently  reclaimed (such  as solvents),  these firms have found uses for
wastes that  generally  are not  considered  attractive  candidates  for  resource
recovery,  such  as  certain  metal  finishing  wastes  (see  Battelie-Columbus
Laboratories, 1976; Grumpier, 1977).

     Table 12  shows  that only 461 m3  per  year  of wastes were being sent for
secure disposal.   At  the time that these  data were collected, there were no
chemical   landfills operating  in either  Oregon  or Washington.   The nearest
secure disposal  site  was the Wes Con, Inc.,  facility  at Grand View in south-
western  Idaho, some  740  km from  Portland.   This  site provides  permanent
storage  of  wastes  in abandoned missile  silos  (see Ghassemi  and  Quinlivan,
1975).   Since   that  time, Chem Nuclear  Systems,  Inc.,  opened  their secure
disposal  facility near Arlington, Oregon, about 225 km by road from Portland.
Consequently,  it is not  surprising that by 1977 greater quantities were being
sent  to  Arlington.   The permit data  have  enabled  one-shot shipments  to

                                     92

-------
Arlington to  be separated  from  recurring shipments.   Permit  data show that
recurring wastes from  Oregon  are expected to total 6,450 m3 per year, broken
down as follows:

               Pesticides	   1,047
               Heavy metals	   1,148
               Solvents and oils	     422
               Phenols	   1,679
               Other chemicals	   2,154

The  actual  total  for  September  1975  through August 1977 was  3,687  m3 (one-
shot shipments  for this  period  were  285 m3).  The  Arlington  site commenced
regular operation  in  May 1976,  and the  period  September 1976  through August
1977  was  the  latest  full year  for  which  data were  available wnen  this
analysis was  conducted.   Elimination  of data  from  the first  few months of
operation should have  reduced the impact of wastes being taken from storage-.

     One feature  of particular  interest 'is that the  recurring wastes to be
shipped  to Arlington are  produced  by  38  firms,  only   15  of   which  were
identified  by Battelle  as  producing  hazardous wastes.  (These 15 firms do,
however, generate  the  bulk of the recurring wastes that are being shipped'to
Arlington;   permit  data show that they expect  to  produce 4,474 m3 per year.)
This apparent discrepancy  may be in part attributed to hazardous waste clas-
sification  criteria.   In some cases a waste  that is being sent to Arlington
may  have  been classified .as  nonhazardous by Battelle.  On the  other hand,
several  of  the firms  involved were  from SIC  categories  that might  not be
expected to produce hazardous wastes--(e.g.,  a manufacturer of fireplaces).

     Any comparison of the permit data with  the  shipments  data must be made
with caution, because  of the  comparatively short period  involved (one year)
and the fact that some of the permits were issued during the period for which
the data were collected.   Analysis of data on a permit-by-permit basis showed
that though some of the permit quantity projections were accurately realized,
in  several  cases  there  were  discrepancies as  large  as- a- factor of  two,in
either  direction,  while  in a few cases  the  discrepancies  were  much larger.
On  the  other hand,  the  permit  quantities  tended to  be larger  (often -by a
factor  of two to  four) "than the Battelle data  for the same waste stream.  No
shipments had been  received from nine firms  expecting to  produce 919 m3 per
year.   In   the  case  of  one major  stream of waste  acid,   this was,  probably-
because  of  the  use  of  an alternative  method of. disposal;  in some  of the
others  it. may have been  that the  quantities  of wastes that had ac:rued were
comparatively small and had yet to be shipped.

Statistics  on Firms Generating Hazardous Wastes--
     Table  13  provides   some  data  on  the  number  of  firms  identified  by
Battelle as -generating hazardous or probably hazardous wastes.  The industry
categories  are  the same  as those  used  in  Table  12,  and Table 13 lists the
SIC's  of  the firms  identified  by  Battelle.   Of  the  94  firms,  six electro-
platers did not generate a hazardous waste as such but had hazardous elements
in their effluents.  Increasingly stringent sewer regulations are expected to
force  these  firms  and  some  others to  treat their  effluents, resulting in
additional   heavy  metals  sludges.   The quantities in  Table 12 only reflect


                                     93

-------
          TABLE 13.   OREGON FIRMS THAT MAY GENERATE HAZARDOUS WASTES

Industrial category

Pesticides and wood
treating
Resin and glue
manufacturers
Paints and inks
Other chemicals

Batteries
Electroplating
Electronics, etc.

Metals manufacturers



Miscellaneous







Total
Total , excluding
SIC's
included3


2491, 2879
2821, 2865
2869, 2891
2851, 2893
2813, 2818
2892, 2992.
3691
3471
3661, 3679
3825.
3312, 3324
3334, 3339
3341, 3361
3398
2231, 2253
2711, 3111
3231, 3412
3425, 3429
3433, 3479
3482, 3612
3629, 3743
9711

miscellaneous
No. of firms
identified by
Battell e
studyb


9

8
14

6
9
18d

4



8







18
94
76
No. of firms
in Dirsc'orv
of Orsson
Manufacturers0


16

17
23

5
3
27

^ •*
J. /



27







197
337
140

  SIC's of  firms  included  in  Table 12  that generate potentially  hazardous
.  wastes.
  Based  on  Stradley,  Oawson,  and Cone,  1975  (including unpublished  data
  sheets prepared in conjunction with  the report).
  Number of  firms having  listed SIC(s) as primary SIC in Directory of Oregon
. Manufacturers (State of  Oregon,  Department of Economic Development, 1976).
  Twelve firms  generate  hazardous  wastes,  but an additional  six  firms  have
  hazardous elements in their effluents (see text).
                                     94

-------
data  from  68  firms.   Twenty additional firms generated hazardous wastes, but
the  amounts were  not determined.   However,  in most  cases the  quantity is
thought to be small, and hence inclusion of these missing data would probably
not cause major changes in Table 12.  The magnitudes of the wastes (excluding
effluents)  from the  68 firms form a typical Pareto distribution, as fellows:
The five  largest  generators produce 75 percent  of  the waste,  the 10 largest
generators  produce  89 percent  of the waste,  and the  20  largest generators
produce 97  percent  of the waste.   (Note  that  a firm may  have  more  than one
waste  stream.)   Of  course  these  quantities give  little   indication  of the
damage potential  of these  wastes,  but the  data do  suggest that  it  may be
reasonable to simplify analysis by concentrating on the larger waste streams.

     Table  13  shows that  if the numbers  of firms  identified  by Battene as
generating potentially hazardous  wastes  are compared to the total numoers of
firms  in  the same  SIC designations,  in  most  cases  there are significantly
more  firms  in Oregon than those identified by  Battelle.   In many cases this
may reflect differences in manufacturing processes, but it could also be that
the smaller  firms are more reluctant to admit  that  they  generate hazardous
wastes.  In the case of  the "miscellaneous" category, the huge difference, in
figures  is  largely accounted  for  by  SIC 2711   (Newspapers:   Publisning,
Publishing  and  Printing),   as  Battelle  only  identified   one  generator of
hazardous waste from a total of 123 firms.

     It is  interesting to  note  that of the 14  broad  categories of industry
identified  byyEPA as  generating most of the potentially  hazardous  waste in
the United States  (U.S. Environmental Protection Agency, 1977b, see Taole 3),
only  petroleum  refining   (SIC  2911)  is  absent from  Table  13.   Naturally
however,   the  relative"  proportions  of  these  industries   do  not  mirror the
aggregate of the  United  States and not all the subcategories of the 14 major
categories are represented.

     The  geographic  distribution of  the  firms  listed  in Table  .3  is  as
follows:
              *•
     Portland SMSA (Oregon counties only)	66
     Remaining Willamette Valley 	   19
     Coos County (coast)  	    1
     Jackson and Josephine Counties (south). ...    5
     Eastern Oregon	    3
          Total	94

These  data  confirm the  importance  of  the greater  Portland  area and the
Willamette Valley as  the  industrial areas of Oregon.   They also simplify the
analysis,   since  there is  only  one large-scale  generator  of hazardous waste
(an aluminum plant in Wasco County) outside the Willamette Valley.

Future Waste Generation—
     At this  stage, a preliminary estimate of  future changes  in hazardous
waste  generation  can be  made (this estimate may  require  modification after
the approaches  to  hazardous waste  management  have  been  selected).   Heavy
metals wastes are expected  to  increase from  2,578 m3 to  about  3,600  m3 per
year,   partly because  of  increasingly stringent sewer regulations.  (Estimate


                                     95

-------
by L.W. .Rass,  Denver  Research Institute, Denver, Colorado, based on analysis
of effluent data  and  chemicals used by  the waste  generating firms.)  Other-
wise,  substantial  increases  do  not seem  particularly likely.   The largest
generator of  hazardous wastes" in  Oregon,  the primary  aluminum industry,  is
unlikely to  expand its  activities  in the  region  because  its economics are
being unfavorably  affected  by the  loss of low-cost  power.   Oregon generally
tends   to   attract  clean  industries   that   generate   comparatively  small
quantities  of  hazardous wastes  (e.g.,  electronics).   In  cases where  new
plants  have  replaced  older  ones,  waste generation  has  frequently  been
reduced.  The  finding that  for  several  industrial  sectors,  hazardous waste
generation  in  Oregon  appeared to  have decreased  between  studies in 1972/73
(State  of Oregon,  1974)  and 1975 (Stradley,  Dawson,  and Cone, 1975) tends to
support the  view  that waste  generation is  unlikely to  exhibit substantial
growth  in the future.

     This conclusion  is  reinforced  by examining the  national  projections of
hazardous waste generation  for the  14 major industrial groups shown in Table
3, only three of which are projected to have aggregate increases  in hazardous
waste generation  that exceed  100  percent  between 1974 and  1983.  The three
industrial groups with large increases are  battery manufacture (2,000 percent
increase  in hazardous  wastes),  textile  dyeing and finishing  (373 percent
increase) and waste  oil  re-refining  (253  percent increase).   Increases  in
hazardous wastes from the battery industry  appear to stem from ennanced water
pollution controls.   (Note  that  the consultants'  figures  on hazardous waste
generation  from battery  manufacture  [McCandless  et  al.,  1975]  and textile
dyeing  and  finishing  [Abrams, Guinan,  and Derkics,  1976] are  higher  than
those of  EPA  used in  Table 3  [U.S.  Environmental  Protection Agency, 1977b],
possibly  because   the  Agency  may  have  reelassifled some  waste  streams  as
nonhazardous.)  For Oregon an increase  in hazardous waste  from the battery
industry  has  been incorporated  in  the estimate  of increased  heavy metals
wastes.  (This  estimate  assumes  that there is no  major change in the struc-
ture of the  industry.)  In the case of  firms  involved in textile dyeing and
finishing, the  Battalia  data  show only very small quantities of potentially
hazardous wastes   being  generated,  so  a  significant  increase  (in absolute
terms)  in the wastes from this Oregon industry seems unlikely.  For waste oil
re-refining,  the  increase is  expected to stem largely from increased produc-
tion,  with  newer plants  replacing  many  of  the old  ones  (Swain,  1976).
Oregon's oil  re-refiner  produces  comparatively small quantities  of hazardous
wastes, and so  speculation  as to the future of this industry  in  Oregon would
not be  very important.

     On the other hand,  the  Oregon  DEQ  took a subset  of  the industry cate-
gories  investigated in 1972 and  reevaluated their waste  generation in 1978.
For  these  industries,  wastes  that might be hazardous  increased 105 percent,
from  approximately 11,300 to 23,200 m3.   (Telephone  interview, F. Bromfeld,
Oregon  DEQ, Portland,  Oregon, April 17, 1979.)  These data may be misleading,
however.  At the time of  the first OEQ survey (and also the Battelle survey),
there  was  no  chemical  landfill  in Oregon, and  only  pesticide  wastes  were
required  by law  to be disposed of  (in Oregon)  in  a  secure disposal site.
Out-of-state  disposal  alternatives were unattractive  (see Stradley, Dawson,
and  Cone,  1975)  and  hence  the generators  had an incentive to minimize the
quantities  of  wastes  actually generated,, or at least the quantities to which


                                     96

-------
they  admitted.   In  contrast,  there  is  an incentive  to the* generators  to
overstate  the  quantities on the DEQ permit  applications, since  they would
need to  reapply  if the quantity was  found  to  have been significantly under-
stated.  This effect  may have  carried over to the DEQ survey in 1978, giving
rise  to  the apoarent  increase.   A second  factor that may  have resulted in
increased  quantity estimates is  the  more extensive  use  of  offsite disposal
(e.g., sending a waste to Arlington, as opposed to storing or disposing of it
onsite).    Where  disposal took place  on a firm's premises, the  firm may not
have  been  particularly  well  aware  of  the  quantities  of  wastes  actually
generated,  whereas when  a waste is  shipped  offsite, the firm would become
much more conscious of the quantities involved.

Control of Hazardous Wastes in  Oregon

     Oregon first  enacted  a  statute to control  hazardous wastes-in 1971 (ORS
459.410 through 459.690).  The  effect of the statute was  to designate waste
pesticides and pesticide  manufacturing residues as "environmentally hazardous
wastes,"  and to make provision  for industrial  process wastes  to be classified
as hazardous by  the Environmental Quality Commission after notice and public
hearings   (ORS  459.410).   In 1977,  PCB's were also  designated  as hazardous,
but no other process  wastes were classified as  hazardous under this scheme
until  1979.   Oregon law  prohibits  the  disposal  of  a hazardous  waste within
the State,  except  at  a licensed hazardous waste disposal  site (ORS 459.510).
The Arlington chemical landfill  is  the only licensed site in Oregon, and as
already  mentioned,  it  is  believed  to  receive virtually  all   of  Oregon's
hazardous wastes that are sent  for secure disposal.  Very small  quantities of
Oregon wastes are  disposed of  in Wes Con's facility in Idaho,  and Arlington
receives   substantial   quantities  of  wastes  from  Washington and occasional
wastes from British Columbia.   (Personal  interview,  F.  Bromfeld, Oregon DEQ,
Portland, Oregon, March 9, 1977.)  Officials have no knowledge of any regular
hazardous waste movements  between Oregon and California (Personal communica-
tion,   0.  Storm,  California  Department of  Health,  Berkeley,  California,
September 19, 1977),  and because of the  location  of  the Oregon hazardous
waste generators there is no  reason  to expect Oregon wastes to be disposed of
in Nevada.

     Under the pre-1979  rules,  all  wastes not designated or  officially clas-
sified as  hazardous were controlled by indirect methods.  Industrial wastes
are not  officially allowed  to  be disposed of in  sanitary landfills, without
DEQ permission.    The  DEQ normally  refused permission  to dispose  of heavy
metals and  highly  oily wastes  in sanitary landfills.  However,  some.marginal
industrial wastes  from the greater Portland area were'Channeled to the city's
St. Johns  landfill adjacent to  the Columbia Slough  in  Portland.  (Personal
interview,  C.  Gray,  Oregon  OEQ,  Portland, Oregon, March 11,  1977.)   Some
industrial wastes  originating outside the Portland area went to other public
landfills.   (Personal  interview,  W.  Dana,  Oregon  DEQ,  Portland,  Oregon,
April  17, 1977.)   It was  generally felt that industry had cooperated well by,
for example, storing  materials while awaiting a  suitable  method of resource
recovery,  treatment,  or  disposal,  and  by sending some wastes  that  were not
really hazardous to the Arlington secure disposal facility.
                                     97

-------
Policy Objectives

     The Oregon  DEQ  has  established objectives (State  of  Oregon,  Department
of Environmental Quality,  1978)  for its solid waste program  (which includes
hazardous waste management),  as follows:

     1.    To  minimize  the  generation  of  and  maximize  recovery  of
          resources from solid waste;

     2.    To   provide   for    environmentally   acceptable   collection,
          transport,   storage,  treatment, and  disposal  of  solid  wasta;

     3.    To develop  positive public attitudes,  awareness,  and actions
          that maximize  citizen  involvement in solid waste management;

     4.    To develop  and  implement  a  continuing process for solid waste
          management planning;

     5.    To   maximize   the    effectiveness   of  agency   resources  to
          accomplish solid waste  management programs;

     6.    To achieve  a  reasonable  balance  between  environment, energy,
          natural resources,  and  economic considerations.

These general  objectives  are  supplemented by specific goals,  such as one for
1979, which was  to increase  the  annual  amount  of industrial  hazaraous waste
recycled or recovered by 100  percent over that of 1977.   None of the specific
goals addresses  objective 6,  which calls  for  the  most  difficult decisions
from the planning viewpoint.

APPLICATION OF ANALYTICAL FRAMEWORK

Development of Alternative Approaches for Hazardous Waste Management

     Before discussing  possible  approaches to  hazardous  waste  management in
Oregon,  a  major  caveat  is  necessary.    The  discussion  of  hazaraous  waste
management alternatives  in this  chapter must not be  taken to imply that the
alternatives would be legal  under either Oregon law or the provisions of the
RCRA  (PL 94-580).   The  analysis is presented both  to  illustrate  the metho-
dology .and to demonstrate what might be involved if different approaches were
employed.  Discussion of an  approach  does not constitute an endorsement of
that  approach—as  already stressed, this is the  prerogative of  the appro-
priate  decisi comakers.   Furthermore,  a far more  detailed  analysis than that
presented  here would be  necessary to  develop a  complete  hazardous  waste
management policy for Oregon.   "

     Figure 9  provides  an overview of four possible  approaches to hazardous
waste management in Oregon and links these approaches to the waste management
techniques that  may  be  involved.  The four approaches are arranged, from top
to  bottom,  in increasing comprehensiveness  and  complexity of  regulation.
Depending  on  precisely  how   the approaches are  specified,  this  could also
correspond  to  an  increasing  degree  of  risk  aversion  on  a  decisionmaker1s
part, or to increasing anticipated environmental quality.

                                     98

-------
  APPROACHES
  TECHNIQUES
 MARKET FORCES
  STATUS QUO
 REGULATION BY
   'PATHWAYS'
 REGULATION BY
 INDUSTRY (S.I.C.)
                                    ADDITIONAL
                                    TREATMENT *
                                                                             OCEAN
                                                                            DUMPING
                                                                           DEEP WELL
                                                                           INJECTION
                                                                             LAND
                                                                            DISPOSAL
                                                                          ADDITIONAL
                                                                       RESOURCE RECOVERY
WASTE STREAM
  CHANGES
* Treatment Includes  Incineration.
                i
                 Figure 9.  Approaches for hazardous waste management in Oregon.

-------
      The  market-forces approach would  essentially allow waste generators  to
 decide what  to  do  with their wastes.   This would  open  up the possibilities  of
 ocean dumping  and  deep well  injection and  might cause  changes  in  *asta
 streams  and  resource  recovery  (the  latter  relationships  are shown dashed  in
 Figure 9).

      The  status-quo approach, i.e., continuation of the pre-1979  accrcac.'  ts
 hazardous  waste management:  in  Oregon,  is used as a base case.  The emonasis
 in-the status-quo  approach  is on  responsible land disposal.

      The  status-quo  describes the  situation   in Oregon  in  1977,  and  uses
 1975-77  data.   At  that  time, hazardous waste  management in  Oregon  was  in
 transition  from a form of  market-forces  approach toward regulation by  path-
 ways.  When  the May  1979  rules are'fully implemented, Oregon will  be  js^'ng a
 regulation-bypathways approach,  but  this approach  will  not be identical  to
 the regulation-by-pathways  approach  considered  in this study.

      The  approach  of  regulation-by-pathways is  intended to correspond  to the
 approach  that  EPA is  adopting  under tne RCRA.   The  concept  is  to  control
 rigorously the  paths that  the wastes can take,  but not to mandate  any  soecial
 forms of  treatment  technology,  etc.  The  general  effect  of  this accroach
 would be  to  enhance  the security  of  land disposal with respect  to  tne  status-
 quo.   Economic  forces  could  tend  to cause  changes  in  wasta  straams and
 increase  resource  recovery.

      If wastes  were  subject to regulation-by-industry  (SIC), it could  involve
 specifying  the  control or  treatment technology,  and/or the means  of disaosal
 for  each  industrial  category.   This approach could  be arranged  to  give  a
 higher  or lower expected  level  of environmental  quality than  regulation-oy-
 pathways,  depending  on  the  levels  of  treatment  required  and  the   wastes
 encompassed.  However, as will be  shown, it  is the most complex  in terms  of
 the degree of regulation.

      In the  analysis that  follows  (i.e. , Steps  2  through 8 in the  Application
.Of the  Analytical  Framework phase  of the methodology outlined  in  Section
 5--see Table 8),  the  market-forces  and the regulation-by-pathways  aporoaches
 will  be discussed together  and  compared with the status-quo approach.  The
 regulation-by-industry (SIC) approach  is quite different and will  be  treated
 separately following the other cases.

 Allocation of Wastes to Techniques

 Status-Quo Approach—
      Although Table  12 provides one  set of data-on hazardous waste  generation
 in  Oregon,   as  already  indicated   there  are  other  conflicting  data   on
 quantities  and  disposal methods.   In  addition,  even  under  the  status-quo
 approach,  future waste streams will  differ, because of  increasing  stringency
 of sewer  and NPDES discharge regulations, opening of new firms  and  closing  of
 old  ones,  and   process changes.   For  these  reasons,  projected  data  on the
 status-quo  approach,  given  in Table 14, will  be used  for  the  evaluation.
 These data  in  part  reflect the  larger quantities (compared to the Battelle
 data) permitted for  disposal  at  Arlington.  However,  allowance has been made


                                      100

-------
                    TABLE 14.   PROJECTED DATA ON FUTURE HAZARDOUS WASTE GENERATION
                               IN OREGON UNDER THE STATUS-QUO APPROACH*

Cubic Meters Per Year
Disposition
Onsite landfill
Ons ite lagoon
Local landfill
Secure disposal!
Resource recovery
Total
Heavy
Pesticides metals
1,800
500
200
600 1,000
	 - 600
1,100 3,600
Solvents
and oils
-
•
100
300
700
. ; 1,100
Phenols
300
200
1
1,500
700
2,000
Other
chemicals
2,300
200
100
1,500
5.600
9,700
Total
4,400
900
400
4,900
6.900
17,500
•
* Source:   Author's estimates,  based on all  available data.   Data are applicable to the eary 1980's,
  see text.   (Note that criteria for designating waste as hazardous are not the same as those that
  would be applied under 1979 Oregon law.)

t At Arlington waste disposal site.

-------
for overstatement  of  quantities on the permit applications,  and  it has also
been  assumed that  some of  these materials would  not qualify as  hazardous
under the  Battelle hazardous  waste  decision model  and should therefore  be
excluded.   The data  reflect the author's views on  the likely disposition of
the  anticipated  increase  in  heavy  metals  wastes,  and  they allow  for  the
planned elimination of  the large quantity of fluoride-bearing sludge that is
presently lagooned.  In addition, the data have been simplified by disregard-
ing  minor  combinations  of  waste  type  and disposition.   Since  potentially
hazardous waste volumes  are expected to remain more or less static once the
effect of more stringent sewer (and NPOES) regulations have taken effect,  no
specific year  has  been  associated with Table  14.   The data  should  oe most
realistic in the early 1980's.

     The simplification  of  data  required  to project Table  14  should not  oe
regarded as  a drastic  step.   Policy development and  analysis  shoula not  be
allowed to be  unduly  sensitive to basic assumptions, unless these are excep-
tionally well  established.  Hence  it is perfectly  appropriate to  won* with
approximate  or  simplified  data  and  to  test   the  sensitivity  of  policy
implications to these data assumptions.

Market-Forces Approach--
     It will  be assumed that pesticide wastes continue  to be cijposed of at a
chemical landfill, except  for some very dilute wastes  (wash waters) that are
lagooned.   Battelle only identified  one waste stream  of  significant  size  in
this  category.   The  waste  is so  dilute  that it  is not considered  to con-
stitute a  significant  hazard,  and for  this reason the  Oregon OEQ  has  not
required that it  be sent to  Arlington.   (Telephone   interview,  E.  Chiong,
Oregon  DEQ,   Portland,  Oregon,  June  5,  1979.)   (For  pesticide wastes  this
represents no change from the status-quo approach.)

     Possibilities for ocean disposal — In terms of cost, barge disposal might
be an attractive proposition for high-volume wastes from the greater Portland
area  or along the  Columbia  River.    Mixing of  lower  volume wastes  from
different sources  to  provide full loads could be  dangerous  ana has not been
considered.    The   smallest  sized  barge  used  for  bulk  ocean dumping  has  a
capacity of  about  900 tonnes (Smith and Brown, 1971).   There appear to be no
more  than four nonsolid hazardous waste streams that exceed this  quantity in
a  year  and  although  they  all  originate from the greater Portland  area,  the
total annual  volume   (3,800  m3 or approximately  4,800 tonnes) would almost
undoubtedly  be too small (even if a few other waste generators joined in)  to
support the operation of a specially constructed disposal barge during even a
part of the year.   Furthermore, three of these wastes are comparatively dense
sludges that could make  disposal  from  tank barges difficult  if  not impos-
sible.   It  would  probably be  some  200 km  to  the  nearest  feasible  dumping
ground  in  the Pacific Ocean,  which  would  likely place the  cost  at the high
end  of  the usual  range  (see Table C-l).   Under the status-quo,  most of  the
3,800  m3  of  waste is  being disposed of  at low cost to the  generator  (by
resource recovery, and in one case by neutralization with a waste  stream from
a  nearby  firm),  and hence even  in   the  absence of  restrictions   on ocean
dumping, this  solution  would  have very limited  appeal to  firms  in Oregon.
                                     102

-------
   •  The only  other  high-volume wastes are solid  (spent  potliners)  and have
good resource  recovery  potential.   The high cost of ocean disposal for solid
materials would render this option uneconomic for these wastes.

     Possibilities for deep well injection—Since  deep  wells  are  compara-
tivelycapitalintensive,they  are  only suitable  for  high-volume  liquid
wastes.   For  the United  States,  the  average  volume  of  industrial  waste
injected per well  is in excess of 200,000 m3 per year (Reeder et al., 1977).
Wells  that  take  less than 10,000 m3 per year are uncommon (Donaldson, 1972).
However, the  entire  volume  of hazardous  waste  generated in  Oregon  is only
17,500  m3   per year  (Table  14),  and most  of  it  is  solid or  sludges.   The
largest  single  liquid waste  stream that  is  a potential  candidate  for deep
well injection is only about 500 m3 per year.

     In  the Willamette  Valley there  are  permeable  beds  containing saline*
water which are  separated  by generally low-permeability marine deposits from
the alluvial aquifers  that  are used extensively for potable'water throughout
the valley.  Any  waste  injection  well  in the Willamette Valley Would nedd to
be a minimum of  600  m deep,  depending  on location.   (Personal interview, W.
Bartholomew, Oregon  Water  Resources Department,  Salem,  Oregon,  October 7,
1977.)   Naturally,   the  costs would depend on the  location -and  waste type
(which  affect  pretreatment and construction  materials).   However,  -the 1977
capital cost of an 800-m deep, .low-injection-volume well would probably be at
least $170,000.  Thus if amortized at 10 percent per year, the capital.-charge-
would  be  at least $17,000  per' year.   In  addition, operating  costs  would be
about $2,000 per year, for a total of at-least $19,000 per year.  This figure
would barely change if the waste quantity were increased from 500 m3  per year
to 5,000  m3 per  year.   (Based on  data from L.  Reeder,  Louis  R.  Reeder and
Associates, Tulsa, Oklahoma, August 24, 1979.)

     Five hundred m3  of waste could be trucked from the Willamette Valley and
disposed of.at Arlington for about $20,000 per year.  • (The typical  gate fee
at Arlington  is  $22 to $27  per  tonne,  depending  on'-the "waste.   [Personal
interview,  P.   Wicks,   .Chem   Nuclear   .Systems,    Incv,'   'Portland,   'Oregon.;
March 9, 1977.]   The charge  fo'r transporting* a hazardous  waste to Arlington
in a  ful'ly loaded truck costs  about $16 per tonne from.Portland, more from
Willamette  Valley.   [Personal  interview,  B.  Larsen,  Widing  Transportation,
Inc.,  Portland,  Oregon,  April 28, 1977.]   The  cost  estimate  is based  on a
specific gravity  of  1.0  for an injectable  waste.)  Disposal  at Arlington is.
one of  the  lowest risk and highest cost  disposal  alternatives..  Less costly
disposal methods  might  be  feasible for many of the wastes,  but in any event
deep well  injection  appears  unattractive, since its  cost  .is  similar to that
of disposal at Arlington.

     The potential for deep well injection of waste streams from new firms or
processes is less  clear.   Faced with a high-volume waste, there might be the
alternative of  treating the  waste  and  discharging it to  a  sewer  or surface
waters, or  using  deep  well  injection.   An  example  of this type was  analyzed
in Section  6,  and for the hypothetical data used,  there was a clear genera-
tor's  cost  advantage to deep well -injection.  There are presently  no deep
                                    •103

-------
waste disposal wells  in  Oregon.   The practice is against State policy and it
would be  opposed by  the Oregon  Water Resources Department,  although  it nas
not been totally outlawed.  (Personal interview, W.  Bartholomew, Oregon Water
Resources Department,  Salem,  Oregon, October 7, 1977.)   It  does  not appear
that any  firm has  seriously  pursued the possibility, although  at least one
has considered it.   Hence deep  well injection  from  new  processes  would need
to be analyzed on a case-by-case basis.

     Possibilities for land disposal--Land  disposal   is  already  extensively
used in  the status-quo  approach,  and the  main possibility  under a market-
forces approach would be the use of less  costly land disposal.  The principal
option would  be  to use  local  sanitary   landfills  instead of  the Arlington
disposal  facility.

     Since under  the  market-forces  approach,  disposal costs may be decraasea
with  respect  to  the  status-quo  approach,  the  question of  increased waste
generation arises.  Data on the price elasticity of demand for waste disposal
services  are,  not surprisingly,  rare,  and it  would require  an analysis of
each significant individual firm's operations  to make any useful predictions.
However,   it can  be  expected that even in the long-run,  demand would be com-
paratively inelastic.  Data on  reduction of BOD in response to sewer cnarges
suggest a price  elasticity of less than  unity  (Elliott  and Sjagraves, 1972;
also  see  Elliott,  1973;  Seagraves,  1973)  and  it  is comparatively  easy to
reduce BOD  loadings  in effluents.   By contrast, changes  in tne .nagnituae of
some waste  streams  could be very difficult to  effect,  and as alreacy noted,
the cost of waste disposal is  frequently  a small proportion of product value,
which  tends to  lead  to  inelastic  demand.   Since  data on  increased waste
generation are not available,  this aspect has  been neglected.

     Resource recovery—The potential  for less costly land disposal  than in
the status-quo approach could reduce resource  recovery.   However, the largest
waste stream  going  for resource  recovery is spent potliners from an aluminum
manufacturer,  and it  is  to  the  company's  advantage  to  practice  cryolite
recovery  on this  material, even  if the disposal  cost were  zero.   Another
significant waste stream,  lead  metal  and oxide  wastes  (largely  from the
battery manufacturing  industry),  falls  into  a  similar  position,  as  is the
case with solvents  that  are reclaimed and a calcium hydroxide sludge that is
used by  another  industry.  Examination  of the  Battelle  data suggested that
the  only  likely significant  change would  be  land disposal  of a  phenol-
contaminated  calcium  carbonate  sludge that  was sent  out  of  State  for lime
recovery.   It was  found  that  this change had already occurred, as the sludge
was being shipped to Arlington.

Regulation-by-Pathways Approach—
     The  basis  or   the   regulation-by-pathways  approach is  to require all
hazardous wastes  actually needing  disposal  to be disposed  of in an environ-
mentally  secure  manner  broadly  consistent  with the  provisions  of the RCRA.
The only  feasible  option available for waste  disposal  (as  opposed to treat-
ment  or  resource   recovery)   is  secure  land  disposal  (incineration  is
considered to be a treatment process) .
                                     104

-------
     Land disposal--Three variations  of the  regulation-by-pathways  approach
will be  considered.   In  Scheme  I, all  hazardous wastes  requiring  disposal
would  go  to the  Arlington facility.   In  Scheme  II,  generators would  be
allowed to construct secure landfills (and lagoons)  for disposal of their own
wastes.   Scheme  III  includes  a  public chemical  landfill  in  the  Willamette
Valley,  as  well  as  private secure landfills.  There are  no  abandoned mines
suitable for mine disposal in Oregon,  so this possibility was not considered.

     The  feasibility  of the various  options  is  strongly influenced  by the
Oregon climate.  Most of  the Willamette Valley receives  an  average  o'r about
100 cm of  precipitation per year, largely during the winter (Sternes, 1967).
Although  there  is a  deficit in the  summer,  potential  evapotranspiration is
less than this (Oregon State University, 1963).  Data on open pan evaporation
in  Oregon are  sparse,  but  those  that are  available (National Oceanic and
Atmospheric Administration,  1976;  Sternes, "1967)  suggest that an evaporation
lagoon could not reliably be used in the-Wi11araette  Valley.•-Hence if lagoons
are lined  with an impervious  barrier to protect the groundwater,  they will
not constitute  a form  of disposal—only  short  term storage.'  In contrast,
Arlington  averages 23 on of annual precipitation (Sternes,  1967),  far less
than potential evapotranspiration  or  open  pan evaporation.  Since it  is not
possible to get much  closer to the Willamette Valley without leaving Oregon's
dry zone, there seems little alternative to disposal  of dilute aqueous wastes
from the Valley at Arlington.

     If pesticide wastes, solvents and oils, and wastes  that  are  subject to
resource recovery are excluded,  sludges and solid wastes from the Willamette
Valley total some 6,000 m3  per year.    Under  Scheme  III,  the latter could be
disposed  of  in a chemical  landfill  within the Valley.   To  provide  adequate
safeguards against water contamination, any such landfill would  need to be
lined  and to  have  a  leachate  collection system.    Residues   from  leachate
treatment might have  to be sent to Arlington.

     The waste  generators  in  southwest Oregon would  probably have  to send
their wastes to  the  Willamette Valley chemical landfill  proposed above or to
Arlington.  Waste generators in  eastern Oregon would probably use Arlington,
but the  cost of  this would probably cause  the  aluminum producer  in Wasco
County (the  only large-volume waste  generator in eastern .Oregon)  either to
subject  its  spent potliners to  resource  recovery,  or (under  Schemes  II and
III) to find a local  disposal site that could meet the requirements for water
protection.   (The  existing  storage/disposal.. site  is   very   close  to  'the
Columbia  River,  and  would  be  difficult to  upgrade  adequately.)  This firm
already  has  plans to eliminate  its  flouride-bearing  sludge  by a change in
scrubber technology.

     Under Scheme II, it appears unlikely that any other firm would construct
its own  secure disposal  facility.   Not only  are  there  no really large waste
streams  requiring disposal  (the  largest  is  1,200 m3  per  year),   but  in
addition,  all  the candidate waste streams originate  from wet  areas.   This
would  preclude  local  disposal   of  liquids  and  would make  construction  of
secure landfills for  solids  and sludges a costly project.
                                     105

-------
     Waste  stream changes,  additional  treatment  and resource  recovery-
The  increased  costs  of disposal  under the  regulation-by-pathways  approach
would prSbably  cause some  process  changes (to  reduce waste)  and  Increased
resource  recovery.   However,  the  extent  of this  is  difficult  to  predict.
Some wastes might  be capable of treatment to render them nonhazaraous.   For
example,  phenols  are  a  significant  regional  waste.   The  Battelle  data
indicate  that  there  are  eight producers  of  potentially hazardous  phenolic
wastes,   and permit  data  show that  seven firms  regularly  expect  to  ship
phenolic wastes to Arlington.  (Only three of these firms are coincident with
the  Battelle  list,  suggesting that  there may  be  quite a large number of
generators  of  phenolic  wastes.)  Though  there  are  few  options  for  handling
concentrated phenolic wastes except land  disposal, biological  treatment is
feasible  at  comparatively  low  cost  when  the concentration of  phenol  is low
(Rosfjord,  Trattner,  and Cheremisinoff,  1976).   The  Battalle  and  Arlington
permit  data do  not  indicate  that  there  are any  regular  high-volume,  low-
concentration  phenolic  waste   streams,  but  if there are,  there  might  be
possibilities for  treatment of such  wastes in conjunction with hardboard and
particle board plants (especially in southwest Oregon,  where these industries
are concentrated).

     Another  possibility  would  be  to  incinerate  certain  wastes  (e.g.,
solvents, oily wastes and other combustible materials,  and  also  some  wastes
that are  hazardous because  of  molecular structure  as opposed to the presence
of  hazardous  elements).    This  category  could encompass  pesticice  wastes
presently sent to  Arlington, and it could include  PCB's (at higher cost) if
the  incinerator  had  appropriate characteristics.   Table  14  shows that there
is only  a minor  quantity (400 m3 per year) of solvents and  oils that are not
already  subject  to resource recovery,  and even the total volume of pesticide
wastes  (which  largely originate in  the Willamette  Valley)  is comparatively
small (1,100 m3  per  year)  in  relation  to  common  incinerator capacities (see
Powers,   1976).    Few other wastes  would be  appropriate  for  incineration.
Costs for incineration  vary extremely  widely, depending  on  the waste  stream
quantities  and  characteristics.  For  example,  Scurlock  et  al.  (1975) quote
$0.25 to $65 per  m3 for liquid  injection incineration  and S22 to  S44 per
tonne for fluidized  bed incineration of  sludges.   Although not specifically
stated,   these  data  tend  to assume  that  the waste  stream  has a significant
calorific value.   Hence  the dilute aqueous pesticide  wastes  would  be  at the
top  end  of  the cost range.   The upper end of this  range is  comparable to the
cost of  transporting a  waste from the  Willamette  Valley and disposing of it
at Arlington (approximately $48 per tonne, or $48 to S76 per m3).

     Thus largely because of the small  volume of appropriate waste,  incinera-
tion would not appear to be an economically attractive proposition in Oregon.
This  tentative  finding  is  confirmed by a report  that Chem Nuclear Systems,
Inc., performed a more detailed investigation and concluded that installation
of an appropriate incinerator  would not be profitable in the Pacific North-
west at  that  time.   (Telephone interview, F.  Bromfeld, Oregon DEQ, Portland,
Oregon, April 17, 1979.)
                                     106

-------
Development of Threat1 Scenarios

     To  proceed  with  the  application  of  the analytical  framework,  threat
scenarios are  needed for  all  treatment,  transport,  and disposal tecnm'cues
that might  vary  from  one approach  to  another.   Although  there  may be some
variation in the threats  associated with  onsite  waste  treatment  (as opposed
to disposal),  lack  of  data precludes analysis of this possibility.  However,
disregarding treatment  threats,  it is still appropriate to develop scenarios
for threats  under  the  status-quo approach since some of these threats may be
eliminated or  reduced  with alternative approaches to hazardous waste manage-
ment.

     Table 15  identifies  some  of the most important  threats associated with
the various techniques that may be used under the three approacnes to hazard-
ous waste management  being considered.   (Refer to Figure 9 for the approach-.
technique relationships.).. The threats specifically -associated with deep well
injection, ocean  dumping  and 'incineration (Threats  H.,  I,  and 1)-are  not o-f
immediate interest  because the  techniques with which they are associated are
unlikely  to  be used under any  approach.   Table  16 provides  more details of
the relevant threats.   Tnese  are -.till,  however,  only threat descriptions:
quantitative data for a given threat generally depends on the approach, since
quantities of  wastes,  disposal  locations, etc., will vary with the approach.
Within the constraints of.this .research, it was not feasible to develop fully
all the possible threat scenarios.  However, at the end of this section, some
have  been  developed   in   sufficient  detail  to  illustrate  possible  method-
ologies, and limited details on others are included.

     It  is  worth examining  the status-quo  to see if  any  threats appear to
have  materialized.   No   reliable  data   could  be  obtained  on  industrial
accidents  involving  hazardous   wastes  in Oregon,  and  no  serious  adverse
effects on water  quality from land disposal have been detected.   There have,
however, been a number of  fish kills arising from various causes.  The Oregon
Department of  Fish  and Wildlife is aware  of seven pollution-caused incidents
in 1975,  and 19  in 1976.  Total numbers of fish  killed are unknown, but they
were in  excess of  12,000  in 1975 and were probably of the order  of 54,000 in
1976.   (Data supplied  by G. MacLeod, Oregon Department of Fish and Wildlife,
Portland, Oregon, October  5, 1977.)  Most  of these kills-were caused acciden-
tally and probably  did not involve hazardous wastes (as opposed  to hazardous
materials).   There  was  one apparent case  of illicit hazardous waste dumping.
These  data  suggest  that few  threats - involving hazardous  wastes  actually
materialize  under the status-quo-approach.-    -

     Now  that  the  wastes have been  allocated  to  disposal -techniques and the
threat scenarios have  been developed, the  key implications of each approach
are summarized in  Table 17 to  assist  the reader in following the subsequent
discussion.

Determination  of Economic  and Social Effects

     The  most  significant economic and social effects are  the control costs,
and the  effects  of  the  threats  arising  from  different  disposal  methods.
                                     107

-------
                  TABLE  15.   THREATS  ASSOCIATED  WITH  THREE  APPROACHES  TO  HAZARDOUS WASTE MANAGEMENT
       Approach and technique
                                                Threat
         Comments
o
00
       All  approaches
       Status-quo
         Local  landfill ing and
           lagooning (not secure)
         Secure land disposal
           (at Arlington)

         Resource recovery
Market-forces
Threats A through G plus:
  Deep well injection
         Ocean dumping
       Regulation-by-pathways
       Threats  A,B,D,E,F,G plus:
         Secure local  landfill ing
           and  lagooning
         Waste  stream  changes

         Incineration
                                         A.   Transport accidents
                                         B.   Illicit dumping
                                         C.   Leaching
                                         D.   Overflow
                                         E.   Odor

                                         F.   Washout
                                         G.   Not identified
                                                H.   Water resource
                                                    contamination
Changes in resource recovery would
  probably change in-plant accidents
Technique unlikely to be used
                                         I.   Reduction of ocean-  Technique unlikely to be used
                                             related recreation
                                             and industries
                                         J.   Liner failure

                                         K.   Not identified

                                         L.   Air polluLion
Would probably affect in-plcint
  accidents
lochniquo unlikely to be used

-------
                              TABLE 16.   DESCRIPTIONS OF PRINCIPAL THREATS
          Threat description
     Applicability
              Comments
A.  Transport accidents
    Mechanism:  Accidents to truck
      involving loss of cargo.
    Outcomes:  Cleanup requirements,
      possible damage to life
      and health.

B.  Illicit dumping
    Mechanism:  Dumping on land or to
      surface waters.
    Outcomes:  Destruction of vegeta-
      tion, poisoning of livestock,
      and possibly man (via crops?).
      Aesthetic damages.
All approaches
All approaches
Virtually all Oregon hazardous waste is
  transported by truck.  Very little is
  flammable/explosive.   Event probability
  is proportional to distance (see
  Table B-l).

The incentive for illicit dumping is
  least under the market-forces approach,
  and greatest under the more costly reg-.
  t.l at ion-by-pathways approach.   However,
  opportunities should be diminished
  under regulation-by-pathways.
C.  Leaching from local'landfills and   Status-quo approach  Th'is threat.is difficult to quantify
      lagoons
    Mechanism:   Leaching from unlined
      landfills and lagoons.
    Outcomes:  Surface and groundwater
      contamination leading to fish
      kills, unsafe drinking water,
      and aesthetic damages.

D.  Overflow from lagoons      '] '
    Mechanism:   Exceptional precipita-
      tion or flood causes lagoon to
      overflow.
    Outcomes:  Damage to adjacent
      crops, structures, etc.   Cleanup
      costs.
Market-forces
  approach
  because the large numbers of sites
  involved, each with different wastes
  and hydrologic conditions.
All approaches
  (but few lagoons
  under regulation-
  by-pat hways ,
  since only E.
  Oregon is
  suitable)
This threat is difficult to quantify
  because of the variety of climatic and
  hydrologic conditions, as well as the
  variety of wastes that might be
  involved
                                                                                           (Continued)

-------
                                         TABLE 16 (Continued)
       Threat description
Applicability
       Comments
E.  Odor from lagoons and landfills
    Mechanism:  Perceptible odor from
      lagoon or open landfill operation.
    Outcomes:  Aesthetic degradation,
      possible reduction in property
      values.

F.  Washout of Arlington secure land
      disposal site
    Mechanism:  Exceptional precipita-
      tion causes washout or excessive
      leaching of trench and lagoon
      contents at Arlington waste dis-
      posal site.
    Outcomes:  Contamination of some
      local rangeland and possibly a
      few wells.

J.  Liner failure at secure local
      landfill
    Mechanism:  Liner at local landfill
      in wet region of state fails, and
      leachate is not contained.
    Outcomes:  Local waste contamination
      leading to fish kills and unsafe
      drinking water.
All approaches
Problem largely arises with organic
  wastes (e.g., pesticides).
All approaches
Location of site makes washout very
  unlikely.   Leaching problems are
  even less probable.
Regulation-by-      Extent of threat is less than that of
  pathways approach   C, as failure should be detected by
  (Scheme III only)   monitoring wells before much damage
                      is done.

-------
                     TABLE 17.  SUMMARY OF APPROACHES
     Approach
Waste dispositions
     Principal threats
Status-quo
Market-forces
Regu1a ti on-by-pathways:
  Scheme I
  Scheme II
  Scheme III
L-and disposal
Greater use of local
  landfills than under
  status-quo
All wastes to Arlington
  secure disposal site

Wastes to Arlington,
  or generator's
  secure landfill

Same as Scheme II,
  but western Oregon
• solids and sludges
  sent to chemical
  landfill in-the'
  Willamette Valley
Leaching from local land-
  fills and overflow from
  lagoons leading to fish
  kills, etc.

As above, but more severe
Washout of disposal site
  (unl'ikely)

As above .  .
Same as above, plus
  Willamette Valley
  chemical landfill liner
  failure, leading to
  limited fish kills, etc.'
Since  (with  the minor  exception of transport accidents)  expected values of
damages from  threats  will  not.be used,  the  threats  will  be considered later
with regard to  their  impacts on the parties-at-interest.   Control costs are
discussed below.

Generators' Costs--                                   "        • •       ...
     It is not  possible to determine accurately the costs of hazardous waste
disposal under the status-quo approach, as onsite disposal costs are unknown.
In many cases,  these  are likely to be  minimal,  and a cost  of  $1 per m3 has
arbitrarily been assumed  for all  onsite  disposal.   Another  problem arises
with  the  transport  cost  and value  of  materials  being  sent for  resource
recovery.   However, since this is not expected to change significantly, these
costs have not been evaluated.  If the cost of local landfill disposal, is $13
per m3  and the  cost..of .disposal at Arlington  is $53-per m3, then the annual
generators' cost for  the status-quo  approach would be $270,200.   Details.of
the calculations  are  provided -in  Table 18, which compares  the  five schemes
and approaches.

     Under the market-forces approach,  the annual cost would drop to 598,200.
This assumes  that  only  pesticide wastes continue to be sent to Arlington for
                                     111

-------
               TABLE  18.   GENERATOR'S  COSTS  FOR ALTERNATIVE APPROACHES'

Disposition
Approach
Market- forces
Quantity m3
Annual cost $
Onsite
disposal

5,300h
5,300°
Local
landfill

4,700
61,100C
Secure
disposal
at Arlington

600.
31,300°
Total

10,500
93.200
Status-quo

   Quantity  m3
   Annual  cost  $
   5,300b
   5,300°
     400r
   5,200C
     4,900.
   259,700°
 ::,5co
270.200
Regulation-by-
   pathways

   Scheme I
     Quant i-ty-m3
     Annual  cost $

   Scheme II
     Quantity m3
     Annual  cost $
Secure  land-
fill  in  the
Willamette
   Valley
Local  secure
   landfill
    (spent
 potliners)
                            0
                            0
                       2,300f
                      34,500r
Secure disposal
 at Arlinctin
                        10,600.
                       536,500t
                          8,300.
                       439,900°
                          13,600
                        525,500
                          10,500
                        47d.400
Scheme III
Quantity m3
Annual cost $
6,000n
150.0009
2,300f
34,500r
2,300,
121,900a
10,600
205. -100


                                                                     Srecon
1 Quantities fro* Table 14.  data excludt wastes snipped for resource recovery

  Assuring II per »J  Insufficient data were available for an accurate estimate

  Based on S5 per in3 gate  fee.  plus S8 per n3  for transport in  average of SO vn. assuming a iptcif'c  anv'.t  3f 1 '
  Sources   Derived from .1975 average U S  sanitary landfilling cost of $3 59 oer tonne (Booz-Al'en arc Hum tan. 1975)
  Transport tariff data supplied Dy B  Larsen, Widinj Transportation. Inc , Portland, Oregon. April 23, 1977

  Based on J30 per it3  gate  fee plus J23 per  m3  for  transport  an average  of 300 lot,  assuring  a  5oecif'C
  of  1.2   Sources,  Disposal  fee  from  personal  interview,  P  Wicks,  Chen  Nuclear  SyUtM.  Inc .  »ort!and.
  March 9. 1977  Transport  tariff data as In c

* Based on 8.300 m3 at  $53 per of  (as  in d), plus 2,300 m3  of spent not liner waste at S30 per ir1 gate fee olus J12
  for transport 80 km, assuming a specific gravity of 1.5   Sources   As in d

  SIS per m3 used for illustration   Insufficient data were available to estimate this cost

' Based on $13 per «J disposal  fee  and S12 per  m3 for transport an average distance of 120  km, assuming 3  specific
  gravity  of  1.2   Sources    Transport tariff data as in c   Disposal cost based on  the cost  of two  Hectare  'anafill
  of  this  type constructed  in Cowhu County, Washington, for use  with  snredded winicipal refuse   Conditions  it this
  site should be  similar  to those in the  Portland area   The 1976 capital cost was J6 36 per m3 of capacity (Teltpnone
  interview,  M  Kennedy.  CH,M Hill.  Portland.  Oregon.  October 19.  1977)   The average  1975  landfill  ooerating cost
  of S3 15 per r1  (Booz-Allen and Hamilton, 1976) has been suostantially increased to cover additional costs or Handling
  hazardous wastes and leachate treatment
                                                  112

-------
disposal  and  that all  other chemically landfilled wastes are  sent  tc local
landfills.

     I.f,,.under  Scheme  I  of the  regulation-by-pathways  approach,  all  wastes
that are  not  resource-recovered go to Arlington, then the annual generators'
cost would  rise  to $536,500.  This figure  is  based  on 8,300 m3 of wastes at
$53, and  2,300  m3 of spent potliner waste at $42 per m3.  However, this cast
for  spent  potliner   disposal  would  almost undoubtedly  prompt the  company
concerned  to  negotiate  a  special  contract for  disposal at Arlington,  or,
under  Scheme  II  or III, to  either  find a local site  that could satisfy the
various safeguards, or  recycle the material.  If secure local disposal could
be accomplished for $15 per m3, then the generators'  cost would be recuced to
$474,400 for Scheme II of the regulation-by-pathways approach.

     Recycling  of the  spent  potliner  waste,  might  occur  under any  of  the
regulation-by-pathways schemes.  In the past, part of this waste was recycled
at  another  aluminum  producer's  facility  in  Washington.    However,  this
producer is no longer able to take the waste for recycling, and under status-
quo conditions,  the   company has  not  found it economic  to construct  its own
recycling   facility.     The   possibility  of  building  a  regional  'Dotliner
recycling and cryolite  recovery plant to  service  several  aluminum companies
has been  discussed,   but  the anti-trust  regulations  are considered to be a
possible barrier to this action.

     Under  Scheme  III of 'the  regulation-by-pathways  approach,  there  is  the
option of disposing of'G.'OOO m3 per year of sludges and solid wastes from the
Willamette  Valley  in.a. landfill with a-liner  and  leachate'collection  system
located somewhere  in  the Valley.   If disposal cost  is $25 per m3, including
leachate treatment, then  the generators' costs would be reduced by a further
$168,000 to $306,400 per year.

Administrative Costs—
     Under the status-quo approach, the Oregon DEQ employed two professionals
specifically  to  deal with  hazardous  waste issues.   Allowing for  support
services and overheads,  this cost can be  taken  as  $100,000  per year.   Under
the market-forces  approach,  it might  be claimed that no administrative costs
would be  involved, but  the DEQ would probably  retain  one professional as an
adviser  ($50,000 per  year).   Since  the regulation-by-pathways approach  is
broadly  equivalent to  operation  under  RCRA  conditions, the same  projected
staffing, i.e.,  the  equivalent of seven full-time professionals at a cost of
approximately  $300,000  per  year  can  be  assumed.   (Telephone  interview,
F. Bromfeld, Oregon DEQ, Portland, Oregon, June 12, 1979.)

     These are only the more obvious administrative costs; many other oersons
(such as Department of Fish and Wildlife employees) might have minor involve-
ment with hazardous wastes,  but this  cost  cannot  realistically be evaluated
and  is  probably  subject  to  very little variation  between  the  different-
approaches.

Social Control Costs--
     No  social  control  costs   have  been  identified.   Although the fees  at
certain  landfills may  not  reflect  full  cost,  the use  of  an  average rate


                                     113

-------
avoids this  issue.   There  is also the possibility that the Chem Nuclear fees
could ref-lect  monopoly  power.   However, these fees  are  somewhat constrained
by competition with other treatment, resource recovery, and disposal alterna-
tives  (such  as  the Wes  Con disposal  facility  in  Idaho).   In view  of  the
extensive  front-end costs  associated  with  establishment  of  the  Arlington
site, the present fees do not appear to be unreasonably high.   Indeed, one of
the  Oregon  DEQ's concerns has been  that  if  the volume of  wastes  shipped to
Arlington fell  off too much  after  the initial  backlog of  stored  wastes  had
been accepted, Chem Nuclear might not find it economic to continue to operate
the  disposal  site.   (Personal interview, F. Bromfeld, Oregon OEQ,  Portland,
Oregon,  March 9, 1977.)  This would be a real possibility  under the market-
forces approach, and  ideally the costs of secure disposal should be adjusted
to reflect economies of scale.

Other Effects--
     Under the three approaches considered,  no significant change in resource
recovery  is  anticipated  unless  cryolite  is recovered  from  the  spent pot-
liners.   (Such recovery would be particularly desirable,  as the United States
currently imports  82 percent of its fluorine needs  [U.S.  Department of  the
Interior,  Bureau  of  Mines,  1979].)   Because  of  varying   truck  haulage
distances, there are  very  minor  differences in energy consumption  among  the
approaches.    All  approaches  will  require about  the  same  area of  land  for
disposal, but the  opportunity   cost  of  using  this  land  is  higher  for  the
approaches   that   stress   disposal   in  the  fertile  and  partly  urbanized
Willamette Valley,  as opposed to use of rangeland at Arlington or other arid
locations.

Determination of Impacts and Responses of Parties-at-interest

     Although the  impacts  on the parties-at-interest and their responses  are
conceptually separate,  in  practice  they may often conveniently be considered
together.  Determination  of  the  impact involves analyzing  how the party-at-
interest  would be  affected  by  a  waste  management  technique  or  approach,
whereas  the  party-at-interest1s  response relates  to  any  action taken as  a
result of the impact.  However,  at the broad level of analysis employed here,
these  two  steps  have  been  merged  into  one   that  emphasizes  the  likely
attitudes of each  party-at-interest.   This  approach  is useful  in  a planning
or  policy study,  since  it  provides  an  indication  of  where  opposition or
support for  specific plans is likely to be encountered.

     Since  all  approaches  emphasize   land  disposal,  and several  possible
techniques   (ocean  dumping,  deep  well  injection,  and  incineration)  are
unlikely to  be used, it is appropriate in this case to consider the reactions
of   the  parties-at-interest  to   different  approaches   (philosophies   or
strategies)  as  opposed to  different  techniques  (specific  technical  waste
management methods).   In Table  19,  the author's judgment  of the  impacts of
each approach  on  the major parties-at-interest has  been  shown using a seven
point  (-3 through  +3)  scale.   A  weighting factor  (1   through  5)  for  the
significance  of the  parties-at-interest is  also  included,  so  that  it is
possible  to  compare scores  for  each approach.   Since approaches  (not tech-
niques)  are  being  compared,  it  was considered inappropriate to include  the
environmental management agency  (i.e., the Oregon DEQ).  In many respects  the


                                     114

-------
       TABLE 19.  IMPACTS OF DIFFERENT HAZARDOUS WASTE MANAGEMENT
                  APPROACHES ON THE PARTIES-AT-INTEREST

u
o
u
re
Li-

ra
c

*J
^
O)

41
* Party- at- interest
3 Major generators of hazardous
wastes
2 Minor generators of hazardous
wastes
1 Competing firms (different •
process)
1 Waste transport sector
1 Firms generating wastes
deemed to be nonhazardous
2 Operators of sanitary land-
fills
3 Residents/land owners adjacent
to sanitary landfills
2 Operator(s) of secure -disposal
facilities .....
2 Residents/land owners adjacent
to Willamette Valley chemica-1.
landfill
1 Resource recovery interests
2 Water supply officials
2 Anglers
2 Environmentalists
5 General public
Composite score
Waste management approach
U)
>— >> c;
>— i >—i '— *J

01
0)
u

o
<*-
1

o>
.*
• u'
re

' 2

2

-1

-1
1

2

-2

-3

. *


-2
-3
-2
-3
. -1
-22




o

a-
i
in
3
f»
re

«/>
I

0

0

0
0

0

-1

1

*


-i
-i
-i
-2
0
-7

o
^
U U)
 C
in ui o
>» re 4-*
re 3 a

£ »™~ •F*'
4-) r™ i™~
re. re u

-3

-3

1
••
2
2

-1

0 •

3
-
'*


2
3
2
3
1
17
CJ (A
£ O
j= re
U UI U>
to ai o
i t-> a.
u> 01 en
^i (9 'H '
re 3 T3
2
JT r- a

rtj (O O
A. v^ N
-^

-3

1

1 '
2

-1
•
• c

2
,.
»


1
2
2
2
1
15
si
-H .J C
1-1 ai re
HI S •—
re
a> r- i—
f^™ (O
— 0
«• jB -r—
U S
<*o o) w
l = -C
U) •!— U
3*1 ^
re s >»
S •— 
-:1

-1

1

1
2

-1

0 .

2

2


0
0
1
1
1
6

Not applicable
                                   115

-------
impact on  the agency would  be some  composite of  the  impacts on  the other
parties-at-interest,  since  the agency  must  respond  to all  the  parties-at-
interest.
     Waste generators would  favor the market-forces approach  as  it provides
them with  the greatest  flexibility  and minimizes  their costs.   Conversely.
they will  dislike  the  three  regulation-by-pathways approaches because of the
higher costs  involved.   Firms  generating large quantities of wastes (in this
case, an  aluminum producer  was  the only firm  specifically identified) will
find a  pathways approach  that permits them to establish  their  own chemical
landfills  (Scheme  II)  more  attractive  than  one that  does not  (Scneme I),
whereas smaller generators will  be unaffected by this variation.   The effect
on  competing  firms is  weak,  but an interesting situation  arises *ith fims
whose wastes  are  deemed  not to be hazardous under  the  pathways  aporoacr.es.
Instead of being  in  an ill-defined situation under  the  status-quo acoroacn,
they will  be  relieved  by an official  decision on tne status of their wastes.

     While sanitary  landfill  operators would  in many cases prefer to receive
more  wastes   under  the  market-forces  approach,   local   residents  may  feel
threatened  by  these  industrial  wastes.   Conversely,   under  the  ?atnways
approaches, the landfill operators will receive less waste, b-ju the residents
will  feel  more   comfortable.   The  reaction  of   the  chemical   landfill
operator(s) will depend on the quantities of wastes anticipated.

     Despite  the  prediction  that  in  general  there will  be Tittle  change in
resource recovery among approaches, the resource recovery industry will favor
the more strict approaches, since the potential  for resource recovery will be
increased.   Water  supply  officials,   environmentalists,   and  anglers  will
distrust the  market-forces approach  because of its  threats to  water quality
and will prefer the pathways approaches, which reduce these threats.

     The general public has been included as a party-at-interest.   The impact
on  the  public  is only  weak,  but concerned  and  informed citizens  can be
expected to prefer the more  environmentally secure approaches, especially as
the  resulting  higher  costs to generators appear unlikely  to have a signifi-
cant negative impact on local  economies.

     Although  potentially  useful, the  development  of a composite  score, as
illustrated in  Table 19,  is highly subjective, especially when  it comes to
assigning  weights  to  the parties-at-interest.   This will be discussed later.

     Several   possible  parties-at-interest  have been omitted from  Table 19,
either on  the grounds  that impacts on them hardly change between approaches,
or  that  they  are  too weak to  justify inclusion.  For example, residents/land
owners adjacent to the Arlington disposal site have not been included because
the  nearest dwelling  is 2 km  from the disposal  area and because any changes
in  the quantities of wastes received will have little impact on the owners of
the  rangeland adjoining the site.
                                     116

-------
Prediction of Outcomes

Waste Dispositions—
     In this case study, the prediction of waste dispositions is very simple,
and was largely determined when wastes were allocated to techniques.

     Under  the market-forces  approach,  most  nonpesticide  wastes would  be
landfilled onsite or sent to local landfills, although a few generators would
probably  still  prefer  to  send  wastes  to  the  Arlington disposal  site for
reasons of environmental  consciousness.   (Additional  costs arising from this
possibility  have  not been  included  in  the  evaluation of  the  market-forces
approach.)  The latter  might  include generators of only  small  Quantities of
wastes to whom the extra cost was not significant, and major firms that had a
high  public  profile.   Firms   that  were  particularly  anxious to  disoose  of
responsibility  for  a waste might be more  inclined to  use  public landfills
(possibly through waste hauler services) rather  than  onsite disposal.   How-
ever, the concept that a waste generator can rid itself of responsibility for
a waste  by  simply paying another organization  to take it away  is now being
challenged (Taubman, 1979).   Use of disposal techniques such as ocean dumping
and deep well  injection appear unlikely, though  the  latter  might be attrac-
tive to some  new  plants.   There might be some reduction in resource recovery
where high shipping costs (e.g., to Seattle, Washington) are involved.

     Under  the  regulation-by-pathways  schemes,  all   wastes deemed  to  be
hazardous would  be  required to  be  disposed of in  chemical  landfills  or the
Arlington  facility.   This  approach  would  undoubtedly  prompt  the  aluminum
producer  that is  not  practicing resource  recovery  on  spent  potliners  to
reexamine the economics of this, and it could encourage other firms to modify
their waste  streams  and increase resource recovery.  The extent of increased
recovery cannot  be  predicted.   If  a chemical  landfill for  solid wastes and
sludges was  established in the  Willamette Valley,  it would  divert  most of
these wastes  from Arlington,  other  than those generated  in eastern Oregon.
The only significant  solid/sludge waste in eastern Oregon  is the spent pot-
liners  discussed  above.    If  this  waste  was  not   subjected  to  resource
recovery, the  firm  would  probably examine the  feasibility  of establishing a
local  landfill  that  complied  with  the  chemical  landfill  regulations  under
Scheme II.

     Because of higher  disposal  costs,  an increased level of illicit dumping
would  be  expected  under  the  regulation-by-pathways  schemes than  under the
status-quo approach.  However,  the  registration of generators and the use of
a manifest system should reduce the opportunities for illicit dumping.   Under
the  market-forces  approach,  dumping might  decrease  from  the  existing low
level.

Opposition to Approaches--
     Si nee there appear to  be few hazardous waste-related problems under the
status-quo approach (such as threats materializing), it seems unlikely that a
change   in  approach  would  produce  very strong  reactions,  except from any
organization very adversely affected.  A change to the market-forces approach
could  be put  into  effect  by  an   administrative  decision  to  allow  local
landfills (primarily  in  the   Willamette  Valley) to  accept  more  industrial


                                     117

-------
wastes,  and  this  change of policy could be  implemented  in  stages providing an
opportunity  to  gauge public and  interest group  reactions.

      Those most likely to be concerned about  the adverse effects  that could
arise from  the  market-forces  approach would  be  the water supply  officials,
anglers,  and environmentalists;  but residents  adjacent to landfills  and  even
the  general  public could become concerned  if the change  in approach  received
"scare"  publicity.  Though the water supply officials would probably  indicate
their concern by administrative actions, the other  parties-at-interest might
mount organized public protests.  Although  adversely affected,  it  is  unlikely
that waste  haulers and the resource recovery industry  would  voice  opposition
to the market-forces approach, but  the impact on the Arlington waste  disposal
operation would be so strong that  the operator would doubtless  lobby against
the   approach.    While  waste  generators  would  prefer  the   market-forces
approach,  they  are more likely'to  address  the  disposal of specific wastas of
interest on   a  case-by-case basis  than to come out  strongly  in favor of the
approach per se.

      By  contrast, the main opposition to the regulation-by-pathways  approach
would come  from the generators  of  hazardous  wastes, who  would  object to the
increased costs  involved.  The weight of these  objections  would  be  -educed as
requirements were made less costly  (e.g., Scheme  III as oppr-bed  to  Scneme I).
While virtually  all  the other  parties-at-interest will  favor the  pathways
schemes  over the  status-quo  approach, only  environmentalists are likely fo
voice their views  publicly.   There  would   probably  be  opposition  to   a
Willamette   Valley chemical  landfill  (Scheme  III)  from  local   residents/
property owners,   and  less  support from many  of  the parties-at-interest for
this scheme  than  for Schemes I or II.

Enumeration  of  Costs and Impacts

      The final  step  in  applying the analytical  framework is to collect and
summarize the data already generated.  In some  cases, it  would  be  appropriate
to  prepare  a synopsis for each  approach  and then to compare approaches—the
first step  of  the  decisionmaking  phase of  the  methodology.   In this  case
study,  however,  these two  steps  have  been  combined  to produce Table 20.

      As  soon as  the  comparison  is attempted,  a  difficulty  arises  with the
threat scenarios.  Unlike the case presented  in  Section  6,  where  each  tech-
nique has its  own scenario,  most  of the threat  scenarios apply to  all  five
.schemes  or approaches.  Clearly,  the intensity  of  the threat  (in terms of the
magnitude  of the effect  or the  probability  of  occurrence) varies  between
approaches.   But except  for   transport   accidents,   the descriptions  and
quantification  of the  threats  presented at the  end of this  section  are not
sufficiently detailed  to  be   able  to  differentiate  between  approaches.
Fortunately, it is comparatively easy to rank the  intensity of  a given threat
as   a  function  of the approach,  and this ranking  has  been included  in  Table
20.   This ranking is  possible  because  threat intensity  generally increases
with quantities of waste.   Ranking the  different threats as they  apply  to  a
given approach  might also be a  useful  adjunct to decisionmaking,  but it has
not   been  attempted,  as  (in the   absence  of  detailed data and  the use of
                                      118

-------
                 TABLE  20.   COMPARISON OF FIVE HAZARDOUS WASTE MANAGEMENT APPROACHES




Control Costs (dollars per year):
Generators' costs3
Administrative costs
Total Control Costs
Threat Scenarios:
A. • Transport accidents
B.. Illicit dumping
C. Leaching from local land-
fills and lagoons
0. Lagoon overflow
E. Odor problems
F. Arlington site washout
J. Liner failure at Willamette
Valley chemical landfill
Other Effects:
Transportation distance
(thousands of km)
Ranking in terms of incentive
for resource recovery
Parti es-at- i nterest
composite impact score

Market-forces
"

$ 98,200
50,000
$ 148,200 ;

5
5

1
1
1 .
5
•
*


40

5

-22

Status-Quo


$ 270,200
100,000
• $ 370,200

4
4

2
2
2
4
"B
*


120

4
;
-7
Approach


Regulation by pathways
Scheme III

$ 306,400
300,000
$ 606,400

3
3

*
3
3
3

1


116

3

6
Scheme II

$ 474,400 $
300.000
$ 774,400 $

2
1

A
3
4
2

*


203

2

15
Scheme I

536,500
300,000
836,500

1
1

*
3
4
1

*


214

1

17

* Not applicable. '
Cv/»l nHi nn malAkvi&lc caitt* ff\v+ v*aes\nv»r*a v*e*f*f\\m**\t / I a L' <*n ac im i 4 nr»ni si** v*/\c c ^iM\v*nu/*hac ^
L. LAI. i uu i »iy ma ic i i a i 3 3d* y  |ui  P6SOUrC6 TCCOVery \ laitcu a a un i i ui in o
  Ranked by significance across approaches;  1 = most significant, 5

-------
expected  values)  It  would  Involve   subjective  valuations  because  of trie
differing natures of  the  threats.

     Section  5  suggests that some  iteration may be desirable to optimiza tna
results within an approach.   For  example, under Scheme III of the regula^ion-
by-pathways   approach,  the   possibility  of  using  more  than  one  cr.enical
landfill  in  the  Willamette Valley  could be examined; under the market-forces
approach  different  wastes and alternative dump sites might be considered *or
ocean  dumping.   However,  the data presented  in  Table  20  and  the precading
sections  do  reflect  some  informal  iteration  during data  development, and
further iterations are  not justified  for this case study.

Regulation by Industry  (SIC)

     It would be conceptually possible to devise an approach to wasta raanage-
ment  in  Oregon  that  would establish separate  regulations  for  eacn inc'jstry
(SIC  classification)  producing  hazardous  wastes.   This could be  a similar
approach  to  that used  for controlling discharges  to  water under tha raderal
Water  Pollution  Control Act  (PL  92-500), in which the technology to be used
for  effluent treatment is specified for  each industrial  sector.   Hov/evar,
even at the national  level, this  has  proved to be an immensely complex jncer-
"taking,  and at  tne  State level,  it  would  involve administrative costs that
would  almost certainly be  out of proportion  to  the  benefits  gained  cy tra
added  sophistication  of  the  approach (compared with regulation-Dy-paf.way:
for example)'.

     One  attraction of a regulation-by-industry   (SIC)  approach  lias  in the
possibility  of  specifying a  treatment technology  or  disposal method that is
matched  to  the  characteristics  of  the  waste stream,  ideally  reaching  an
optimum trade-off  between cost and risk (an optimum  trade-off  could only be
determined  if a degree of risk  aversion was  specified).   However,  the dis-
advantage of  the approach is the number of  industrial  categories that would
need  to  be  considered.   As  Table 13 shows,  39  4-digit SIC categories were
identified  by Battelle as  including hazardous waste  generators.   However,
Battelle  identified  only  96  hazardous waste generating  firms from a total of
337   listed   in   the   Directory  of Oregon Manufacturers  (State  of  Oregon,
Department  of Economic Development,  19/6).In addition to  the SIC's  listed
in Table  13,  other SIC's  undoubtedly  generate hazardous wastes.   For example,
many  engineering  firms  have electroplating  operations  that  produce  heavy
metals wastes,  and hence a  wide variety  of  other  SIC's  could be involved.

     One  method  of solving the .SIC problem  would  be to mandate treatment or
disposal  methods for  wastes  from  specified processes  as opposed  to  SIC's.
However,  even if this  were  done, the sheer number of processes  to  be dealt
with  would  make this a major undertaking if  the  approach  were to encompass
all  hazardous wastes.  Some  check would need to  be  made  on virtually every
manufacturing firm (the Directory of  Oregon Manufacturers lists approximately
4,600  establishments)  and  a number of utilities,  service companies,  and
government  agencies  as well.   Furthermore,  this  approach would  only deal
effectively  with wastes  that were generated  on  a regular  basis,  primarily
process  wastes.   As  noted  in  Section 3,  some  hazardous wastes  arise from
accidents,  cleanup  operations,  off-specification  products,   and  discarded


                                      120

-------
laboratory chemicals.  None  of  these fit well into an approach'that involves
control on the basis of the processes used.   Thus without formal  analysis, it
appears that  an  attempt  to regulate all hazardous  wastes  on an industry-by-
industry  basis  at  the  State level  would  not be  feasible (let alone  cost-
effective)  in view of  the  immense  administrative  effort  that  would  be
involved.

     Another  variation  of this  general  approach would  be to regulate  only
selected  industries.   Two  regional  industries  are  obvious  candidates—the
primary aluminum  industry, and  the  timber products  industries  that produce
phenolic  wastes.   Two  waste  streams from the aluminum  industry  (spent  pot-
liners and  a calcium  fluoride  sludge) account  for 58  percent of  the  total
hazardous waste  in Table  12.   The  phenolic  wastes are significant  because
they  are  a  major  category of waste  in the Pacific Northwest, but  they are
uncommon  in other  parts  of the United States, with the result that they have
received  little  attention  in waste management research.   (All  generators of
phenolic wastes  identified in Table  12 are resin and glue  manufacturers, but
it is  known  that other timber products industries,  such as plywood manufac-
turers, also  produce phenolic wastes.   One  hazardous waste  practices  study
[Gruber and  Ghassemi,  1975]  deals with some  of the SIC's  included  in  Table
13, but does  not discuss phenolic waste streams  to any  significant extent.)

     If   either  the   primary  aluminum  industry  or  the  timber  products
industries were  regulated  by mandating treatment or disposal technology, the
question  of  equity would  arise:   Is  it  fair"that there  should be  special
regulations  for  certain  industries, .while  others  are  only subject  to  some
more  general-regulations?   In  practice,  .the  ability to  specially  r&gulate
these  industries  could  be  exercised  under  the  status-quo  approach  by
officially designating the aluminum  industry  wastes and phenols as hazardous
wastes.   This would have the effect of requiring  the  wastes to be sent to a
licensed  secure  disposal  site if they were disposed of in  Oregon.   There is,
however,  an  important  distinction between these two  alternatives.   W:th the
regulation-by-industry approach,  the waste  generator might have littla  or no
latitude in  deciding what to do with his waste (e.g., if a  specific treatment
technology was  required),  whereas under  the  status-quo approach,  the  waste
generator would  only  have  to  use  a  specified technology  if he wished to
dispose of  the  waste.   Based  on the above  discussion,  no version of the
reguTation-by-industry approach appears to be attractive for the situation in
Oregon, and  the approach has therefore been discarded.

DECISIONMAKING                •

Array Alternatives and Eliminate Approaches Dominated by'Others'

     In this  case  study,  none of the  six  schemes  or •approaches is dominated
by any other  (see  Section 5), and hence  none can be eliminated in this  way.
But the regulation-by-industry approach is too cumbersome (or too inequitable
if only a few industries are regulated) to be used in Oregon.   This reduces
the approaches under active consideration to those listed in Table 20.
                                     121

-------
Comparison of Results of Approaches With Policy Objectives

     None  of the  approaches  is  clearly  in conflict  with the  Oregon  DEQ's
policy  objectives (see  Policy Objectives  earlier  in  this  section),  since
these are  largely framed  in  nonspecific terms.   However, the  objective  of
achieving a reasonable balance between environment,  energy, natural  resources
and  economic considerations  clearly requires  a subjective decision,  wnich
could vary considerably between individual decisionmakers.

     The  OEQ objectives   consider  energy  use.   Since  incineration is  not
involved under any of the approaches, the main  variation in  energy use will
be in transportation.  It is estimated that (disregarding the  wastes sent for
resource recovery) the  market-forces approach  would consume 22  m3  of diesel
fuel per year, compared to 116 m3 »under the regulation-by-pathways (Scheme I)
approach.  (Based on an average fuel consumption of 0.54 liters per kilometer
for  heavy trucks  without  return  loads  [telephone interview,  L.  Fisnback,
Widing  Transportation,  Inc.,  Portland,   Oregon,  June 22, 1979].)   The  dif-
ference between  these two extreme energy consumptions would,  if produced as
heating oil,  heat  approximately  32 typical  single family Oregon homes (based
on   data   from   J.  Fand,   Oregon  Department   of  Energy,   Salem,   Oregon,
August 23, 1979).  Of course  the  cost  of the  fuel  has  been  induced in the
generators' costs.

Examination of Trade-offs Between Known Costs and Threats

     It is important to recognize that any comparisons involving threats must
be  subjective since  different individuals  (both decisionmakers  and parties-
at-interest)  are liable  to perceive and compare threats  in  different ways.
Therefore, each  decisionmaker  must draw his own  conclusions  from the trade-
off  analysis, and may disagree with the discussion that follows.

     In general, the  intensities of Threats A,  B, and E (transport accidents,
illicit  dumping, and odor) appear  to be  minor compared  to C  or J  (landfill
leaching  and landfill  liner  failure).   The intensity  of Threat D  (overflow
from lagoons) is probably more site-specific than any other,  and will depend
on the  location  and  construction of the  lagoons.  But in general this threat
appears  to be less  significant  than  C  (leaching) with  which  it is usually
associated.   By  using  these  assumptions to  reduce  the  number of  threats
considered,  a series of comparisons among the  various  approaches can now be
made.

Market-forces versus  Status-quo—
     Changing  from  the  status-quo  to   the  market-forces approach  reduces
control  costs by  $222,000 per  year (see  Table 20).    The major offsetting
damage  potential  is the  increased risk  of water  pollution from   landfill
leachate  (Threat C)  and to a  lesser extent from lagoon overflow (Threat 0).
As  shown at  the end of  this  section the value of fishing at  risk is about
$9.8 million per year.   To convert  this to a  single sum, assume  that the
problem  arises   after ten years  and persists  for  five years.   The present
value  of the economic  loss would be $14.32 million after discounting  at 10
percent per year.  This may be compared with a known saving of $1.36 million,
i.e.,  the present value  of $222,000  discounted at 10  percent  for  the ten
years before  the problem  developed.

                                     122

-------
     One problem with  this  comparison is that Threat C, water pollution from
leachate, could arise  under the status-quo approach.  However, it is clearly
a far less intense threat as few industrial wastes reach nonsecure landfills.
This could  be handled  by assigning a threat intensity  under  the status-quo
of, say, 10 percent of the value under the market-forces approach, and treat-
ing  this as  reducing  the  net  present  value of  the damage  potential  from
$14.32 to $12.89 million.

     Since more wastes  are  sent to Arlington under the status-quo than under
the market-forces approach,  the consequences of Threat F (site washout.) could
be more  severe under  the latter approach.   However, it  is  considered that
this threat  is  unimportant  by comparison to that  of leaching from nonsecure
landfills, and can be disregarded.   Hence the main elements in the comparison
are  a  threat  involving  an  economic  loss  of S12.9  million,  compared  to  a
saving of at  least $1.36 million if  the  market-forces  approach  is preferred
to the status-quo.   A  choice between these two  approaches  will  be dependent
on  a  decisionmaker's   perception   of Threat  C,  and  his  desire  for  risk
aversion.

Regulation-by-pathways Scheme I versus Scheme II--
     Under  Schemes  I   and  II  of  the  regulation-by-pathways approach  the
threats  from  leaching  and  lagoon  overflow  (Threats   C  and D) should  be
eliminated.    Although   Scheme  I provides  slightly  more control, Scheme  II
reduces  generators'  costs by  $62,100 per year  (see Table  20)  and  is  more
equitable since the  generators  are  given some latitude  over  the  disposal  of
their wastes.   Even though  it  appears  that only one generator is  'ikely  to
exercise the option to develop his own secure disposal site, the  existence  of
this choice  should  make the approach appear more  reasonable  to  many parties
who might  be concerned  about  the  monopoly  power that  a  sole disposal  site
could exercise.   The threats from these two schemes appear to be  quite minor.
Both washout of  the Arlington  disposal facility  (Threat F) and  transport
accidents involving  cargo loss  (Threat  A) are  expected to occur  less than
once every  100 years,  and  would probably cause  comparatively minor damage.
The probability  and effects of leaching or  washout of a  disposal  site for
spent potliners (under Scheme II) would depend on the site, but assessment  of
the  risk  is  simplified  by the  fact that  only  two hazardous  components,
cyanide  and  fluorine,   should  be present.  Hence  Scheme II  is preferred  to
Scheme I of the regulation-by-pathways approach.

Regulation-by-Pathways Scheme II versus Status-quo—
     Adoption of Scheme II  of  the regulation-by-pathways approach instead  of
the status-quo  approach would  increase  control  costs  by $404,200  per  year
(see Table  20).   Of this,  $200,000  is  increased  administrative  costs which
include  the operation  of a  manifest system; hence illicit dumping should not
increase.  Offsetting  the increased control  costs is the virtual  elimination
of  Threats   C and  0  (leaching  and  lagoon  overflow).    However, under  the
status-quo approach, these threats are already quite  minor (and do not appear
to have  materialized to date).   For  example, if  as  in  the previous  section,
only 10  percent of the full impact of Threat C is considered to  apply to the
status-quo approach,  then the  present  value of this threat only amounts  to
$1.43 million, based on the same assumptions about timing.   In contrast, the
present  value of the increase  in the control costs is $2.48 million, so if a


                                     123

-------
declslonmaker were  to  accept these dollar values  and  timing assumptions, he
should  logically  prefer  the  status-quo  approach  j_f these  were  the  only
factors  that  he considered.   Risk  aversion  is  not  a  factor  because  the
reduction.in  control  costs is greater than the damage cost should the threat
materialize—and  in  practice it  may not materialize.  Of course, a decision-
maker might wish to make a more comprehensive analysis and take other factors
into account.

Regulation-by-pathways Scheme III versus Scheme II—
     Scheme  III  of the  regulation-by-pathways approach  was  developed partly
because  it was  evident  that Schemes  I  and II  would be  substantially  more
costly than  the  status-quo approach, and that they might not offer commensu-
rate environmental  advantages.   The major threat  from Scheme  III is a liner
failure  at the  secure  landfill  in the Willamette Valley  (Threat J).   It is
shown  (at the  end  of this  chapter) that  should  this  occur,  the  cost of
countermeasures  is  expected  be  about $200,000, plus a small threat (taken as
$100,000  per  year)  to local  fishing.  In  addition,  illicit dumping might be
reduced  in comparison to  Schemes  I or II.  (While  a  manifest system should
virtually  eliminate  illicit dumping  of  regularly  generated  waste streams,
wastes that  do  not occur on a regular basis  are  still liable to be dumped.)
Making the same timing assumptions about threat materialization as above, the
present  value  of the reduction in control costs between Scheme II and Scheme
III  of  the  regulation-by-pathways  approach  is   SI.03  million  whereas  the
present  value of threat materialization would be only $0.22 million.  Changes
in  other threats between these two alternatives should be minor, so again vf
these  data were  considered  accurate, and  no  other factors  were taken into
account,  a logical  decisionmaker  should  prefer Scheme  III  to Scheme II (or
Scheme I)  of the regulation-by-pathways approach.

Regulation-by-pathways Scheme III versus Status-quo—
     Under Scheme  III  of the regulation-by-pathways approach, annual control
costs are  $236,200 greater than under the status-quo approach  (see Table 20),
but  of this  difference,  $200,000 is  increased administrative cost.   If all
threats  other than  D  and  J  (landfill   leaching  and secure  landfill  liner
failure)  are disregarded  as minor, then  using  the data  already developed,
adoption  of  Scheme III of the regulation-by-pathways approach  instead of the
status-quo approach would  raise the present value  of ten years' control costs
by  $1.45 million,  against a reduced  present value of threat materialization
of  $1.21  million.   Although on  the  basis of  these data  alone  a logical
•decisionmaker. should prefer  the  status-quo to Scheme  III  of the regulation-
by-pathways  approach,  the numbers are sufficiently  close  to warrant further
investigation.   This is  particularly  important because two quite different
approaches are  being compared,  as opposed  to  divisions  of a  single approach
(e.g.,  Scheme II vs. Scheme III  of  regulation-by-pathways). .  With a compar-
ison of  schemes, many of  the factors  involved may remain constant or similar
(e.g.,  the nature  of  a threat remains the same  but its intensity changes).
This  implies that  these comparisons  should be  subject to  less uncertainty
than those associated with two fundamentally different approaches.

     In  the  above comparison of the status-quo and  Scheme III of regulation-
by-pathways  approaches,  in addition to the uncertainty surrounding the threat
                                     124

-------
scenarios,  several  factors  were not  considered.   A major  one is  that  the
benefits of spending  an additional  5200,000 per year of administrative ccsts
(which will support a manifest system and enhanced waste classification work)
are only  indirectly  incorporated  into the  foregoing  analysis.   If  part  of
this additional cost  were  disregarded in the  analysis  (on the basis that it
provided  separate  benefits),   Scheme  III  of  the  regulation-by-pathways
approach might  be preferable  to the  status-quo  approach.  Hence  a broader
analysis that considers  additional  factors is needed to choose between these
two approaches.

Selection of an Approach

     Based  on  the foregoing  discussion,  the  critical  decision  in  this case
study appears to  lie  in choosing between  the  status-quo approach  and Scheme
III of  the regulation-by-pathways approach.   However,  others  might disagree
with  this.   For  example,  an  environmentalist  might  argue  that   damage  to
wildlife and aesthetic losses arising from water quality degradation would be
beyond  valuation,  and that  such threats  should  be minimized  regardless  of
cost.    This would lead  him  to favor Scheme  I of the regulation-by-pathways
approach,  or  some  more stringent approach.   In contrast,  an industrialist
might favor the market-forces approach, but the version of this approach used
here is unlikely to attract much additional support.  One way to take account
of these differing  viewpoints is to examine  the  composite impact  score from
the parties-at-interest analysis  in  Table 18.  As  already  notad,  such  an
analysis  is highly  subjective,  but  subjective elements  are  unavoidable  at
this stage  in  applying the methodology.   Table  18  shows that the approaches
that result in  a  greater anticipated level of environmental protection score
higher  on  the parties-at-interest impact  analysis,  because most parties-at-
interest  favor   such  schemes.    (Note  that  the  absolute  score  is  not
particularly important--it is  relative scores that provide insight.)  Scheme
III of  the  regulation-by-pathways approach scores 6, compared to  -7 for the
status-quo  approach.     (However,  Scheme  II  of  the  regulation-by-pathways
approach scores 15, which  might justify its further examination, even though
Scheme  III  appeared  to be clearly preferable when relative control costs and
threats  were  compared.)   The decisionmaker can  gauge the  significance  of
these differences by  comparing them to the entire range of scores, i.e., -22
through 17.

     Table  21 provides a summary of  key aspects  of the two approaches being
compared.   The  present values of 30 years'  control  costs  have been included
as  representing the  costs  involved  over  the entire project  life  (i.e.,  30
years has  been  taken  as planning horizon).  If the threats did not material-
ize  until   the  end of that period  their   present  values  would be  about  15
percent of  the  values shown in  Table  21.   If the recommendation that threat
values  should  not  be  discounted beyond  one  generation  is  adopted,  these
values  would   also  represent the  minumum  present  values that   should  be
assigned to threat  materialization.   Use of  the  30-year data  instead of the
10-year data minimizes the weight given to  the threat scenarios.

     How  the   decisionmaker   weighs  the many  factors  involved in choosing
between  these  alternatives  is  (in  absence  of  any  specific  agency  policy
guidance) up  to him,  and the  decision  should be tested to see how sensitive


                                     125

-------
                        TABLE 21.  COMPARISON OF TWO HAZARDOUS WASTE MANAGEMENT APPROACHES
                     Item
                                                                              Approach
                                             Status-quo
                              Regulation-by-pathways  Scheme  III
NJ
en
tontroi^Costs:
   Generators'  annual  costs
   Annual  administrative  costs
   Total
   Present value  of  total  control
     costs over 10 years  discounted
     at  10% per year
   Present value  of  total  control
     costs over 30 years  discounted
     at  10% per year

 Threat  Scenarios
   Major threat scenario:

     At  risk
          Dollar estimate
          Additional risks
                                                    $ 270,200
                                                      100.000
                                                    $ 370,200
                                                  $ 2,274,700
                                                  $ 3,489,800
Leaching from local land-
  fills and lagoons
Partial loss of fishing
  valued in total at $9.8
  million per year
Aesthetic damages

10% loss of fishing
  ($980,000 per year)
  from years 11 through 15
Present value=$l,432,100
                                       Contaminant concentrations
                                         might locally become high
                                         enough to render v^ll water
                                         unsafe.  It threat occurs,
                                         cost of providing alterna-
                                         tive water supply would
                                         be involved
                                          $ 306,400
                                            300.000
                                          $ 606,400
                                        $ 3,726,100
                                        $ 5,716,500
Liner failure at secure land-
  fill in Willamette Valley
Minor loss of fishing valued
  in total at $9.8 million
  per year
Aesthetic damages

Counterpuming cost of
  $200,000 in year 11
Fishing loss of $100,000 in
  years 11 through 15
Present value=$223,200

Contaminant concentrations
  might locally become high
  enough to render welI water
  unsafe.  Less likely than
  under status-quo approach
                                                                                                 (Continued)

-------
                                        TABLE 21.   (Continued)
               Item
                                                                       Approach
Status-quo
                                                                           Regulation-by-pathways Scheme  III
         Other  threat  scenarios:
           A.   Transport  accidents
           B.   Illicit dumping

           C.   Leaching from  land-
                 fills, etc.
           0.   Lagoon  overflow

           E.   Odor  problems

           F.   Arlington  site washout

       Other  Effects
M .       Resource recovery

         Energy use:   diesel  fuel
           consumption in transportation
         Composite score  for  impact on
           parties-at-interest
                                     Very minor threat
                                     No significant difference
                                       anticipated

                                     See above
                                     Judged a minor threat com-
                                       pared to threat C
                                     Can be controlled to be
                                       a minor threat
                                     Very minor threat
                                     Little difference in
                                       incentive

                                     65 m3

                                     -7
                          Very minor threat
                          No significant difference
                            anticipated

                          Judged a very minor threat
                          Judged a very minor threat

                          Can be controlled to be
                            a minor threat
                          Very minor threat
                          Little difference in
                            incentive

                          63 m3

                           6
.  Excluding wastes shipped for resource recovery.
  Incentive is reduced for Willamette Valley solids and sludges and increased for other wastes.
  Range for five approaches evaluated -22 to +17.

-------
it  is  to  the  various  assumptions,  particularly  those   like  the  threat
scenarios that  are  somewhat arbitrary or based on inadequate data.   However,
it is  hoped  that use of this methodology will assist a decisionmaker to ma KG
better decisions.   In  this  particular case, it has demonstrated that the use
of  a  secure  landfill   in  the  Willamette  Valley should  be worth  further
detailed examination.   Under  the RCRA only regulation-by-pathways approaches
would  be permissible,  but Scheme  III  should be  feasible and  could  prove
attractive in comparison to Schemes I or II.

APPENDIX:  DEVELOPMENT OF CERTAIN THREAT SCENARIOS

Scenario A - Transport Accidents

     As  far  as  is  known, no hazardous waste is transported in Oregon by rail
or barge.  Table B-l quotes a national accident rate of 17 events per billion
kilometers  for  tank  trucks  involving  cargo  loss.   There  is no  reason  to
assume that  this rate  is  not  characteristic  of carriers  in  Oregon,  and  it
will be assumed that it also applies to trucks carrying solid wastes (usually
in the form of 208-liter [55-gallon] drums).

     Though all accidents involve costs (which should be covered by insurance
and  thus reflected  in  the  haulage  rates),  only where there  .s  a  soil]  are
costs  imposed that specifically relate to hazardous wastes.   If it is assumed
that the average load is 25 m3, in the worst case situation (in which all  the
10,600 m3 of nonresource-recovered wastes included in Table 14 are assumed to
be  trucked  an average of 300  km)  the accident frequency will be 2.2  x 10"3
events per  year.  Since  the  variation  among  approaches would  be  less than
this,  and since  in many  cases  the damage would  be largely  restricted  to
cleanup  costs  for  toxic materials (only $25,000  per  accident even  at $1,000
per m3 of cargo), transport accidents can be largely ignored in the analysis.

Scenario C - Leaching from Local Landfills and Lagoons

     The  worst  leaching  would  occur under  the market-forces  approach,  as
(apart from  pesticides) local  disposal  in  unlined landfills  is  expected  to
increase.   The  most  potent  contaminants  to  be  found in  the non-pesticide
wastes are probably  heavy  metals.   Other wastes that might go to these land-
fills  contain  biodegradable  materials  (e.g.,  phenols),  materials  that  are
hazardous by virtue of  their  pH,  and solvents and oils  that  present  a fire
and explosion threat.

     All   known  firms   with wastes  that  contain  heavy metals  are in  the
Willamette Valley.  From the Battelle data,  it is estimated that the 1,956 m3
of  these wastes shown  in Table  12  that  are not subject to resource recovery
contribute approximately 135  tonnes  of  heavy metal  ions.   If  allowance  is
made  for increasing  effluent  standards,  this  is expected  to rise  to  150
tonnes,  considering  only those firms for which effluent data were available.
When  allowance is made  for waste streams  for which  there  were insufficient
data,  the total quantity  of  heavy  metal  ions  is  expected to be  about 205
tonnes.  The approximate proportions  (percent by weight) of the various metal
ions (excluding wastes subjected to resource recovery) is as follows:
                                     128

-------
               Chromium	42.3%
               Copper	28.6
               Zinc.
               Nickel	
               Cadmium 	
               Lead	
               Others (includes barium, silver, gold).
                    Total
 22.8
  3.5
  1.5
  1.0
  0.3
ISO*
This estimate is based on the Battelle data, assuming that the ion content in
the wastes  for  which  insufficient data were available  is  the same as in the
known waste streams.

     The mean annual  discharge  of the Willamette River at its mouth at Port-
land has been estimated to oe 1,077 nrVs, and the corresponding flow at Salem
is 752 mVs.  The mean annual flow of the Columbia River at Portland has been
estimated to  be about  6,958 mVs  (Orem,  1968).   If 205 tonnes  per  year of
metal  ions, composed  as shown  above,  were leached  into  the Willamette or
Columbia Rivers  in  the  vicinity of Portland, the expected average concentra-
tions  of  metal   ions  are shown  in  Table 22.  This  table also  includes the
quality standards for the river waters and for potable water supply.

      TABLE 22.   THEORETICAL AVERAGE CONCENTRATIONS OF IONS FROM HEAVY
	METALS WASTES AT PORTLAND. AND RELEVANT STANDARDS (ppj)	

                                                  Calculated concentrations
                                                  from heavy metals wastes

Drinking
Oregon
water quality water quality
Ion
Chromium
Copper
Zinc
Nickel
Cadmi urn
Lead
Others
Total
standard3
50
1,000
5,000
d
10
50
-
•
standard
50
5
100
d
10
50
-
555 •

Willamette
River
2.55
1.73
1.38
0.21
0.09
0.06
0. 02"
OS

Columbia
River
0.39
0.27
0.22
0.03
0.01
0.01
e
57^3
  Standard  for  domestic water supplies (Source:  U.S.  Environmental  Protec-
  tion Agency, 1976d).

  Standards for  Multnomah  Channel  (Columbia River) and Main  Stem  Willamette
  River (OAR 41-045).

  Based on  205  tonnes  of heavy metal ions  distributed  as  given in the text.
  River flow data are also given in the text.
  Not quoted.

  <0.01 ppb.

                                     129

-------
Even  for  the Willamette River,  the  concentrations are at least  an  order of
magnitude below the  quality  standards established for potable water, so this
risk has not been analyzed.   The possibility of higher concentrations affect-
ing  water supplies  should  be  borne in mind  when considering this threat.

     The  effects  that the metals might have on fish  and  other  aquatic life
vary  widely  with  circumstances and  species.*   Salmonids--the fish  of pre-
eminent  concern  in  Oregon—are among  the  most  sensitive,  although bioac-
cumulation is  not a  problem  as it  is with  shellfish.  The  concentration at
which the presence of metal  ions affects fish depends on the water's hardness
and  alkalinity.   The  effects depend on  the  type  of ion  (e.g.,  hexavalent
versus  trivalent  chromium)  and on  the conjunctive  presence of  ions.   For
example, cadmium and  copper,  and zinc and copper have more severe affects in
conjunction than  individually.   The  effects of concern are not restricted to
the  levels for acute or chronic mortality.  Fish  may sense lower concantra-
tions and  avoid  the  area.   This may preclude anadromous  fish  (such as many
species  of  salmonids)   from  returning  to  their spawning  grounds—thereby
decimating the fishery.

     For  chromium,  fish mortality effects  may occur at 20 ppb,  but they do
not  usually  become  apparent  until  higher concentrations.   Some research with
very  high   concentrations   (5,000   ppb)   has  not  resulted  ir   nortality.
Hexavalent chromium  is  generally more  toxic than  the trivalent  form,  and
invertebrates are  frequently more sensitive than  vertebrates.  With cooper,
sublethal effects  are observed  at 5 ppb, and  fish mortality may occur at 10
to 20 ppb.  Mortality for zinc is not observed below 100 ppb; and for nickel,
chronic effects (via  eggs)  occur at 700 ppb.   Cadmium can be lethal to fish
at as  low as 1 ppb,  but this  rises  to  9 ppb  in  hard water.   Fish mortality
occurs at 300 to 600 ppb for lead.


     If these lethal concentrations  are compared with those given in Table 22
for  the Willamette  River,   it  will  be  seen that the metals that  are most
likely  to cause  problems are chromium  and  copper.   Although  they are close
with  respect to  copper,  the concentrations do  not  exceed  the  Oregon water
quality  standards  for  that  location.   (These  standards  are  presumably
intended to protect both aquatic and human life.)  However, it must be remem-
bered that the concentrations given in Table 22 are additional to preexisting
levels.

     Evaluating  the  actual  risk is  difficult.    In  practice,  most  of  the
metals  enter the  landfill  in a relatively  insoluble form  (e.g.,  hydroxide
sludges), but they could be released, depending on pH.  If they are released,
concentrations  could be attenuated by  ion-exchange  with  the  intervening
soils.  On the other hand,  Elzy et  al.  (1974) (also see Elzy and Linostrom,
* This  section  is  based on personal discussions with H. Lorz, Oregon Depart-
  ment  of  Fish and  Wildlife,  and  G.  Chapman,  EPA, at  the  EPA Western Fish
  Toxicology  Station,  Corvallis, Oregon  on  October 4,  1977.   The discussion
  was specific  to  the  conditions and species of  relevance  to Oregon.  (Also
  see U.S.  Environmental  Protection Agency, 1976d;  Van  Hook,  1978;  National
  Academy of Science, National Academy of Engineering, 1972.)

                                     130

-------
1976)  have  modelled the  release  of leachates  from landfills,  using  a case
study, of a  landfill  on Brown's IsVa'nd adjacent to the Willamette River near
Salem,  Oregon.   They  showed that  the  release of  elements would  occur  in
pulses, which  could reach  quite  high concentrations.  The  pulses  relate  to
precipitation and  water table  height  at the landfill.   Unfortunately it  is
not feasible to  simply extrapolate Elzy et al.'s results to simulate concen-
trations at  Portland,  as  behavior is nonlinear except at low concentrations.
However,  the  complete computer  model  could  be  rerun  to  provide  a  ;nore
accurate prediction of the concentrations that might occur in Oregon's rivers
if  various  wastes  were deposited in specified landfills.   (Personal  inter-
view,   T. Lindstrom,   Oregon    State    University,    Corvallis,    Oregon,
October 3, 1977.)  It  must  also be remembered  that  flow  rates  in the rivers
vary considerably throughout the year and from year to year.

     Without  far  more extensive  analysis,  it  is  difficult   to  determine
whether or  not  the concentrations  of  heavy  metals  that, might leech  from.
landfills  in  the  Willamette  Valley  could  affect  aquatic  life  in  the
Willamette River.  It certainly appears-'possible that concentrations :oulB-be •
high  enough  to  affect parts  of  the  river  system.   Concentrations  in  the
Columbia should be lower,  and the Oregon DEQ's practice of directing marginal
wastes to the  St.  Johns landfill  (which must leach into the Columoia River)
rather  than  to other  landfills further, up the Willamette Valley appears  to
be  wise.   The  city  of Portland  has tested the water near  to  the  St.  Johns
landfill,  but.apart from chlorides  and iron,  could not  detect any elevated
levels  of  contaminants  such  as  heavy  metals.    (Personal  interview,. ..H,..
Edwards,  L.  Brownson and  D.  Steiner,  Office  of'the  City  Engineer,  Portland,
Oregon, April  27,  1977.)   It would also  be., useful  to  compare the  likely
increases  in  heavy  metals  concentrations  in  the  Willamette   and  Columbia
Rivers with existing levels. .This step was not taken, because although there
are numerous water quality data for these rivers, these particular parameters
are not yet regularly monitored (U.S.  Environmental Protection Agency, 1974c;
State  of  Oregon, Department  of Environmental  Quality,  1976a,  1976b).   Most
data  that  do exist  on heavy metals in  these rivers relate to  those  in  the
bottom  sediments,  as  opposed  to  dissolved  heavy  metals that  would  affect
fish.    (Telephone  interview, S. McKenzie,  U.S.  Geological. Survey,  Portland,
Oregon, August  23,  1979.   However,  se'e J. Rinella  and.S.  McKenzia,  1977.)

     By determining  the value of  fishing on the Willamette River system,  one
could  obtain  an indication  of  the cost  of  the potential threat from waste
disposal.    In  one  way, this estimate  might overstate  the  damage,  as  some
anglers would move  to  other rivers.   However,  there would  also be  aesthetic
damages  and. secondary effects  such  as  reductions  in  the populations  of
raptors that are dependent on the fish.   Given sufficient resources, it might
be  possible  to  come  up with dollar estimates for most of these effacts.   But
for the  present purpose,  it is probably satisfactory to concentrate  on  the
direct effect on the fishery.                        . •  •   •            .

     Good data are available on the numbers of anadromous fish that  "run"  the
Columbia,  Willamette,  and  other  Oregon' rivers,  and  estimates  far fishing
effort are available.   In 1977, it was estimated  that  anglers  spent 246,148
angler-days fishing  for salmon  and steelhead on the Willamette River system.
Most  of  this fishing  was  by boat on the  lower portion of the  river  in  the


                                     131'

-------
greater Portland area.   (Telephone interview, D.  Swartz, Oregon Oeoartment of
Fish  and  Wildlife,  Portland,  Oregon,  May 8, 1979.    Data  largely  based on
Lowry,  1978.)   An angler-day  of this  type  of fishing  is  valued  at  330.30
(Oregon Department  of  Fish and  Wildlife,  1977)  for a  total  annual  value of
$7.6  million.   It can  be assumed that  all  this  fishing is at risk, because
the salmonids would  be  deterred  from returning to any part of the Vn'llamette
to  spawn.   In  addition to  salmon   and steel head  fishing,   anotnar 921,197
angler-days  are  expended  on  other  types   of  fish,  predominantly  trout.
Although some of these  fish are anadromous,  for simplicity it will be assumed
that  all  are  resident  and  that a  landfill would  only threaten  the  lower
Willamette  (below Oregon  City).   In  this  event,  only 216,280  angler-days
worth $2.2  million would  be  at risk, for a total  of $9.8 million,  (""he 32.2
million is  based  on  $10.60 for resident trout and  $8.82 for most other game
fish  [Oregon Department of  Fish  and Wildlife, 1977].)  In practice, it 'vouid
be  most  unlikely that  all  the  fish  that  live in or  pass througn  the  lower
Willamette would  be  killed,  so some fishing would continue,  making tne 59.8
million per year  a  limiting  (high)  estimate of the  direct  effect.   However,
as  already  noted,  aesthetic  and  secondary effects have  not  been  included in
this evaluation.

     If landfills  that  accepted  toxic  materials  were  located  higher  LH the
Willamette  Valley (e.g.,  at  Eugene/Springfield), more  fishing  :oulc  be at
risk depending on the distribution of contaminant  concentrations in tne river
system.  Based  on  the  predeeding data,  the  total annual  value of '-'smng on
the entire  Willamette  River  system  is $17.2  million, but fishing over tne
entire system would not be destroyed, since there  would be some resident fish
on  sections unaffected  by landfills.

Scenario 0 - Overflow from Lagoons

     Since  lagooning for  evaporation is feasible  only in eastern Oregon, the
potential   for  lagooning is limited.   In  Table 12,  the  bulk  of the material
that  is  lagooned (3,817 m3 per  year)  is  a  calcium  fluoride  sludge that the
firm already has plans  to eliminate.   Much of the  remainder (492 m3 per year)
is  a  very  dilute  pesticide waste (wash water) from a pesticide femulator in
the  Portland  area.  This probably  poses a leaching  threat as  well  as an
overflow  threat,  because  if  leaching  did  not  occur,  the   lagoon  would be
expected  to  fill  up.   But  as  a   result   of  the  very low concentrations
involved,   the  Oregon DEQ  has  not required  the  waste  to  be  disposed  of at
Arlington  (telephone  interview,  E. Chiong,  Oregon  DEQ,  Portland,  Oregon,
June 5, 1979), and so it can be assumed that the overflow threat is perceived
as  being  relatively  minor.   Other lagoons involve only  small  annual quanti-
ties of wastes.  Hence  without further analysis,  it is possible to argue that
the  magnitude   of  the   overflow  threat  is  far less  than  that of  leaching.
However, without  knowing  more  details of the  lagoons,  it  is  not possible to
assess the threat probability.

Scenario E - Odor from Lagoons and Landfills

     This problem is not, of course, peculiar to hazardous wastes, but it has
been  included  because   it  is  frequently the trigger  for citizen  complaints.
At  the lowest  level,  the problem  is  aesthetic and  no permanent  damage is
done, but it could have an adverse effect on the value of properties adjacent

                                     132

-------
to  the  relevant  disposal  site.   As  far  as   is  known,  none of  Oregon's
hazardous  wastes  give  rise to  extensive air  pollution capable  of  causing
significant  physiological  damage to  living  organisms or  physical  damage to
structures.

Scenario F - Washout of Arlington Secure Land Disposal Site

     A  geologist's report on  the  Arlington site (Newcomb  and  Deacon,  1971)
evaluates the possibility  of  water contamination arising from various causes
(e.g.,  leaching,  major  flood,  volcanic  activity,  landslides,  faulting  and
earthquakes).  The actual  disposal  site includes a natural closed depression
(some 18 m deep) used for the lagooning operations,  but the disposal trenches
are on  higher ground.   The report shows that the depth to usable groundwater
in a confined aquifer of Columbia River basalt group at the Arlington site is
152 m.   The  report concludes  that because of the low rainfall (average of 25
cm per  year)  and high evaporation demand (average of 160 cm per year), water
is  generally not  transmitted downwards,  and what  does escape  is  absorbed
locally and does not threaten the basalt aquifer.  The report states that the
possibility  of   operations  being  affected by  landslides,  faulting,  earth-
quakes, ash  falls  and lava flows from volcanic activities are each less than
one event in 10,000 years.  It also states that the site is so well protected
that a  regional  catastrophe  (such  as the  blocking  of the Columbia drainage)
would  be  necessary to flood th'e site,  which should occur  less  than  once in
10,000 years.

     Though  these  findings  suggest  that  the  threat  potential  from  the
Arlington site is very low, the greatest threat would still appear to be some
sort of precipitation-induced event, such as washout of the disposal trenches
or  excessive leaching  following a  very  exceptional  storm.   There  are  two
wells  within  2  km of the  site,  and these, together with  some  of  the range-
land,  could  conceivably be affected.   If  so,  the use of  this  land  could be
lost for  several years,  and the wells  might  have to be extended  to  a lower
aquifer.  As  a   first  approximation,  the costs  involved are  estimated  to be
$100,000, being  largely well redevelopment.

Scenario 0 - Liner Failure at Secure Local Landfill
     If a secure  landfill  was constructed in the Willamette  Valley,  a liner
would be necessary to protect the groundwater.   If this liner failed,  failure
would  be  detected  by monitoring wells  and  suitable  countermeasures  taken.
These could include covering the landfill surface with an impervious material
to prevent water  ingress,  and1 the drilling and pumping of interceptor wells.
Even so, some of the leachate could escape, but the quantity would be compar-
atively small.  Assuming  some ten interceptor wells were  drilled  to  a depth
of 30  m,  the  cost of the  countermeasures might be $200,000  (based  on data
from   L.  Reeder,    Louis R.  Reeder  .  and   Associates,    Tulsa,   Oklahoma,
August 24, 1979).    The  value of  fishing threatened would  be far  less  than
that of Scenario C—perhaps $100,000 per year for five years.

     A further  risk is  that the contamination might travel  sufficiently far
to affect  some wells,  in  which event  an alternative water  supp'y would be
required.   However,  because  of  the use of monitoring wells, damage to health
should be  avoided.

                                     133

-------
                                  REFERENCES

Abrams, E.F., O.K.  Guinan,  and D. Derkics.  Assessment of Industrial Hazard-
     ous Waste  Practices,  Textiles Industry.   SW-125c,   U.S.  Environmental
     Protection  Agency,   Washington,  O.C.,  1976.    275pp.   [NTIS:PB258953]

Anderson,  F.R.,   A.V.  Kneese,  P.O.  Reed,  R.B.  Stevenson,  and  S.  Taylor.
     Environmental  Improvement  Through  Economic  Incentives.    The  Johns
     Hopkins University Press, Baltimore, Maryland, 1977.   207 pp.

Anon.  "A Nightmare in Niagara."  Time. 112(7):46, 1978a.

Anon.  "Focus:   In  the  Pubic Eye."  Electronics and Power, 24(5):333, 1978b.

Anon.  "Love Turned Sour."  The Economist. 268(7041):30,31, r»78c.

Arbuckle, J.G.,  M.A.  James,  M.L.  Miller, andT.F.P. Sullivan.  Environmental
     Law Handbook.  Government Institutes, Inc., Washington, D.C., 1976.  423
     PP-

Arrow, K.J.,  and T.  Scitovsky, eds.  Readings in Welfare Economics.  Richard
     D. Irwin, Inc., Homewood, Illinois, 1969.  740 pp.

Arthur 0. Little, Inc.  Alternatives to the Management of Hazardous Wastes at
     National Disposal Sites.2 volumes.EPA/530/SW-46c, U.S. Environmental
     Protection  Agency,  Washington,  D.C. , 1973.   [NTIS:PB225164, PB237264]

Arthur D.  Little, Inc.   Pharmaceutical Industry; Hazardous Waste  Generation,
     Treatment, and Disposal.  SW-508,  U.S.  Environmental  Protection Agency,
     1976.  190 pp.

Ayres, R.U.,  and A.V.  Kneese.  "Production, Consumption, and Externalities."
     American Economic Review, 59(3):282-297, 1969.

Backman,  J.   "Pricing."   In:   Science in Marketing,  G.  Schwartz, ed.   John
     Wiley and Sons, New York, New York, 1965.  pp. 250-280.

Barnett,  H.J., and  C.  Morse.  Scarcity and Growth:  The Economics of Natural
     Resource Availability.    The  Johns Hopkins  University Press, Baltimore,
     Maryland, 1963.  304 pp.

Barr  Engineering Co.   Hazardous Waste Generation:  Twin Cities  Metropolitan
     Area.  Barr  Engineering Co., Minneapolis, Minnesota, 1973.
                                     134

-------
Barfager, S.M., B.R.'  Judd,  and D.W. North.  The Economic and Social 'Co-its of
     Coal and Nuclear Electric  Generation;  A FrameworK  for Assessment  and
     Illustrative  Calculations  for  the Coal and Nuclear Fuel Cycles.  U.S.
     Government Printing Office, Washington, D.C., 1975.  138pp.

Battelle-Columbus  Laboratories.   Assessment  of Industrial Hazardous  Waste
     Practices;  Electroplating  and  Metal  Finishing  I nous trie's"!  3Vl-136c,
     U.S.Environmental  Protection Agency,  Washington, O.C.,  1976.   [NTIS:
     PB264394]

Battelle  Memorial  Institute,  Pacific Northwest  Laboratories.   Program  for
     the  Management  of  Hazardous Wastes.   2  volumes.   EPA-530/SW54c, U.S.
     Environmental  Protection   Agency,   Washington,   O.C. ,   1974.   [NTIS:
     PB233630/1]

Bell, D.E., R.L. Keeney,'and H:  Raiffa.   Conflicting Objectives in Decisions.
     John-Wiley and Sons,. New,York, New York, 1977.   452 pp.

Berkowitz, J.B., F. March, and R. Home.   Industrial  Solid Waste Classifica-
     tion Systems.  EPA-670/2-75-024,  U.S!EnvironmentalProtection Agency,
     Cincinnati, Ohio, 1975.  412 pp.

Bishop, J., and C. Cicchetti.  "Some Institutional and Conceptual Thoughts oh'
     the  Measurement  of  Indirect  and Intangible  Benefits and  Costs.'1   In:
     Cost  Benefit  Analysis and Water Pollution Policy, H.M.  Pesfcin and E.P.
     Seskin, eds.The Urban Institute,  Washington,  D.C., 1975.  pp. 105-126.

Bohm, P.  Social  Efficiency:  A Concise  Introduction  to Welfare Economics.
     John Wiley and Sons,  New York, New York, 1973.167 pp.

Booz-Allen  and  Hamilton,  Inc.   Cost  Estimating  Handbook  for  Transfer,
     Shredding  and  Sanitary  Landfill ing of Solid Waste.   SW-124c, U.S. En-
     vironmental  Protection  Agency, Washington, D.C. ,"1976.   83  pp.   [NTIS:
     PB256444]

Booz-Allen  Applied  Research,  Inc.   A Study  of  Hazardous  -Waste.  Materials,
     Hazardous- Effects and Disposal Methods.Volumes 2 4 3.EPA-673/2-73-15
     and  16.  -U.S.  Environmental Protection-Agency,  Cincinnati,- Ohio, 1972.
     [NTIS:PB221466/7]

Cannon, J.S.  Environmental  Steel:   Pollution i'n the Iron and Steel Industry,
     J.M. Halloran, ed.Praeger Publishers, New York, New York,  1974.535
     PP.

Chase,   S»B.... Jr., ed.  Problems in Public  Expenditure  Analysis.   The Brook-
     ings Institution, Washington, 0.C.,  .1968..  283 pp.         ~

Cook, T.D., ed.  Underground Waste Management and Environmental Implications.
     Memoir  18,  The  American  Association  of  Petroleum Geologists,  Tulsa,
     Oklahoma, 1972.
                                     135

-------
Corson, A.S.,  ed.   Proceedings:   1975  Public  Meetings  on Hazardous  Waste
     Management.    2  volumes.  SW-9p,  U.S.  Environmental  Protection Agency,
     1976.   1750 pp.

Council on  Environmental  Quality.   Ocean Dumping:   A National  Policy.   U.S.
     Government Printing Office,  Washington, D.C., 1970.  45 pp.

Council on  Environmental  Quality.   Environmental Quality:  The Eighth Annual
     Report.   U.S.  Government Printing Office,  Washington,  D.C. ,  1977.   475
     pp.

Grumpier,   E.P.,  Jr.   Management  of Metal-Finishing  Sludge.    SW-561,  U.S.
     Environmental Protection Agency, Washington, D.C. , 1977.  65 po.  I'NTIS:
     PB263946]

Donaldson,   E.C.   "Injection Wells  and Operations Today."   In:  Underground
     Waste Management and Environmental Implications,  T.D.  Cook,  acT  Memoir
     18,  The  American Association of  Petroleum  Geologists,  Tulsa, Oklanoma,
     1972.   pp. 24-46.
Edelman, A.,  D.  Huber, A.  Kohan,  J.  Kooyoomjian, C.  Metcalf,  C.
     Shannon,  and  R.  Testani.   Hazardous Waste Management:  I.-.e
     From  Discussions at Four Public Meetings. December 1975.   i.v-st-i. o.i.
     Environmental  Protection Agency, 1976.

Elliott, R.D.  "Economic Study of the Effect of Municipal Sewer Surcharges on
     Industrial Wastes  and Water  Usage."   Water  Resources  Resaarcn, 9(5):
     1121-1131, 1973.

Elliott, R.D. , and J.A. Seagraves.   The  Effects  of   Sewer Surcharges  on the
     Level   of Industrial   Wastes  and  the Use of Water by Industry,   rtaoort
     No"!  7"0~i  North  Carolina  Water  Resources  Research  Institute,  Raleigh,
     North Carolina, 1972.  53pp.- [NTIS:PB213267]

Elzy,  E. ,  and  F.T.  Lindstrom.  "Model  of  the  Movement of  Hazardous Waste
     Chemicals for  Sanitary Landfill  Sites."   In:   Proceedings  of the Con-
     ference  on Environmental Modeling and Simulation,  W.R.  Ott,  ea.  £?A-
     600/9-76-016,   U.S.   Environmental  Protection Agency,  Washington, D.C.,
     1976.   pp. 609-613.

Elzy,  E. ,,  F.T.  Lindstrom, L.  Boersma, R. Sweet, and P. Wicks.  "Model  of the
     Movement of Hazardous Waste Chemicals In and From a Landfill Site."  In:
     Disposal  of Environmentally  Hazardous Wastes, Oregon State University
     Environmental   Health  Sciences  Center  Task  Force  on  Environmentally
     hazardous Wastes, Con/all is, Oregon, 1974.  pp. 111-201.

Executive  Office  of the  President,  Office of Management  and Budget.   "Sub-
     ject:    Discount Rates to  be  Used  in Evaluating  Time-Distributed Costs
     and Benefits."   Circular No.   A-94 (Revised),  Office  of Management and
     Budget, Washington, D.C., 1972.  6 pp.
                                     136

-------
Farb,  0.,  and  S.D.  Ward.   Information  About  Hazardous  Waste  Management
     Facilities.   EPA/5307 SW-145,   U.S.   Environmental   Protection  Agency,
     Cincinnati, Ohio, 1975.   138 pp.

Fennelly,  P.P., M.A.  Chillingworth,  P.O.  Spawn,  and M.I.  Bornsteir.   The
     Generation and Disposal  of Hazardous Wastes in Massachusetts.    GCA-TR-
     76-29-G,GCA  Corporation,GCA/TecJinology Division,  Bedford,  Majsacfiu-
     setts, 1976.  127 pp.

Fischhoff,  B.  "Cost Benefit Analysis and the Art of Motorcycle Maintenance."
     Policy Sciences. 8(2):177-202, 1977.

Fishburn,  P.C.   Decision  and Value Theory.   John Wiley  and  Sons,  Inc., New
     York,  New York, 1964.  469 pp.

Fisher,  A.C.,  and  J.V.  Krutilla.  "Valuing  Long  Run  Ecological Conseauences
     and Irreversibilities."   Journal of Environmental Economics and Manage-
     ment.  1(2):96-108, 1974.                             •  • •

Fisher,  A.C.,  and  F.M.  Peterson.  "The Environment in Economics:  A Survey,"
     Journal  of Economic Literature. 14(1):1-33, 1976.

Flinn, J.E.,  T.J. Thomas, and M.D. Bishop..  Identification Systems TO*- Selec-
     ting Chemicals or Chemical   Classes as.Candidates for  Evaluation.EPA-
     560/1-74-001,   U.S.  -Environmental  Protection Agency.,. .Washington",  D.C,,
     1974.   153 pp.  [NTIS:PB238196]     "                       '    .

Foster D. Snell, Inc.  Assessment  of  Industrial  Hazardous Waste Practices.
     Rubber and Plastics Industry.3 volumes.Contract No. 68-01-31S4, U.S.
     Environmental   Protection  Agency,  Washington,   D.C.,  1976a.   [Draft.]

Foster 0.  Snell,  Inc.   Potential  for  Capacity  Creation  in the  Hazardous
     Waste  Management Service Industry.SW-12/c,  U.S.  Environmental  Protec-
     tion Agency, Washington, D.C., 1976b.

Freeman,  A.M.  Ill,  R.H.  Haveman, and A.V.  Kneese.  The Economics of Environ-
     mental Policy.  John  Wiley and  Sons,  Inc.,  New York, New  York,  1973.
     198 pp.

Ghassemi, M.,  and  S.  Quinlivan.   A Study of  Selected -Landfills' Designed as
     Pesticide  Disposal Sites.    EPA/530/SW-114c,  U.S. -Environmental  Protec-
     tion Agency, Washington,  D.C., 1975.  140pp.  [NTIS:PB250717]

Gilmore,  J.S., W.M. Beaney, P.I.  Bortz, M.K.  Duff, and T.D. Nevens.   Environ-
     mental Policy  Analysis;   Public  Po-licy  Intervention in Inter-Industry
     Flows   of Goods  and  Services  to  .Reduce  Pollution.  NSF Grant GI-11,
     Denver Research Institute,  University of Denver,  Denver, Colorado, 1971.
     150 pp.
                                     137

-------
Goldstein,  B.D.    "Assessment  of  the  Health  Effects  of  Stationary  Source
     Fossil Fuel  Combustion Products."   In:   Risk  Benefit  Methodology and
     Application:   Some Papers  Presented at the Engineering Foundation 'vork-
     shop,  September 22-26. 1975,   Asilomar,  California,   0.   Okrent,   ea.
     UCLA-ENG-7598, School  of  Engineering and Applied Science, University of
     California, Los Angeles, California, 1975.  pp. 103-116.

Gruber, G.I., and  M.  Ghassemi.   Assessment  of  Industrial  Hazardous  Waste
     Practices,   Organic  Chemicals,  Pesticides  and  Explosives Industries.
     SW-118c,  U.S.  Environmental  Protection Agency,  Washington,  D.C., 1975.

Hardin,  G.   "The  Tragedy  of  the  Commons."  Science.  162(3859):1243-1248,
     1968.

Haskell,  E.H.,  and V.S.  Price.   State Environmental Management:  Case Studies
     of Nine States.  Praeger Publishers,  New  York, New York, 1973.  310 pp.

Haveman,   R.H.,  and  B.A.  Weisbrod.   "The Concept of Benefits in Cost-Benefit
     Analysis:   With  Emphasis  on Water Pollution  Control  Activities."   In:
     Cost  Benefit  Analysis and Water Pollution Policy, H.M. Peskin and E.P.
     Seskin, eds.   The  Urban Institute, Washington,  D.C.,  1975.   pr>.  37-66.

Herfindahl, O.C.,  and A.V. Kneese.   Economic  Theory  of  Natural  Resources.
     Charles E. Merrill   Publishing  Company,  Columous,  Ohio,  197*1.   .115 po

Jacobs  Engineering  Co.,   Inc.   Assessment  of  Industrial  Hazardous  Waste
     Practices,   Petroleum  Refining.SW-129c, U.S. Environmental Protection
     Agency, Washington, D.C.,  1976.  [NTIS:PB259097]

Kennedy,   R., R. Lowrey,  A. Bernstein, and F. Rueter.  A Benefit-Cost  System
     for Chemical  Pesticides.  EPA  540/9-76/001,  U.S.  Environmental  Protec-
     tion Agency,  Washington, D.C. ,  1976,  322 pp.   [NTIS:PB250988]

Klee,  A.J.   "Models for  Evaluation of  Hazardous  Wastes."  Journal  of  the
     Environmental Engineering Division,  Proceedings  of  the American Society
     of Civil  Engineers, 102(EE1) 111-125, February 1976.

Klein, F.C.  "Friendly  Reaction:   Nuclear Power Plant, Clean and Quiet, Wins
     Approval  of  Neighbors"   Wall  St.  Journal.  Jan.  16th, 1978,  pp.  1,27.

Kohan,  A.M.  A Summary  of Hazardous  Substance Classification  Systems.   EPA/
     530/SW-171,  U.S.  Environmental  Protection   Agency,   Cincinnati,  Ohio,
     1975.  59 pp.

Kotler,  P.   Marketing Management;  Analysis,  Planning and Control.  Prentice-
     Hall,  Inc., Englewood Cliffs,  New Jersey, 1967.  640 pp.

Kovalick,  W.W.  (Chairman)  Report  of  the Non-Sewage  Sludge/Residuals Work
     Group.   2  volumes.   U.S.  EnvironmentalProtection  Agency,  Washington,
     D.C.,  1975.  [Draft]
                                     138

-------
Lackey,  L.L.,  T.O:  Jacobs,   and  S.R.  Stewart. ••Public .Attitudes  Toward
     Hazardous  Waste Disposal Facilities.    EPA-670/2-73-086,U.S.E"nviron-
     mental  Protection  Agency,  Cincinnati,  Ohio,  1973.   195  pp.   [NFIS:PB
     223638]

Lave,  L.B.,  and E.P. Seskin.   "Air  Pollution  and Human  Health."   Science,
     169(3947):723-733,  1970.

Lehman,  J.P.   "Status  of  Legislation, Grants  and Proposed  Hazardous Waste
     Regulations."    Journal  of Environmental Health.   39(2):83-86,   1976.

Leonard, R.P.,  R.C.  Ziegler,  W.R.  Brown,  J.Y.  Yag,  and  H.G.  Reif.   Assess-
     ment of Industrial  Hazardous Waste  Practices in the Metal  Smelt".nq and
     Refining  Industry!3 volumes.EPA/530/SW-145, U.S. Environmental Pro-
istry.
', 1975.
     tection Agency, 1975.  [NTIS:PB276172]

Lipsey,  R.G.,  and K.  Lancaster.   "The General' Theory  of  Second Best."  The
     Review of Economic Studies. 24:11-32, 1956.

Louis Harris and  Associates,  Inc.   A Second Survey of Public and  Leadership
     Attitudes Toward Nuclear Power Development In tne United States,  toasco
     Services Incorporated, New York, New York, November 1976.

Lowryj H.   1977 Oregon Anglers  Survey.  . Survey Research Center, Oregcn State
     University, Con/all is, Oregon, September 1978.

Maass, A.   "Benefit-Co'sf Analysis:   Its Relevance to Public-Investment Deci-
     sions."  Quarterly Journal of Economics. 88(2):208-226, 1966.

Maderthaner,  R.,   P.   Pahner,  G.  Guttmann,  and  H.J.   Otway.   Perception of
     Technological  Risks:  The Effect of  Confrontation.   RM-76-53,  Inter-
     national  Institute   for  Applied  Systems  Analysis,  Laxenburg,  Austria,
     1976.  24 pp.

MSler, K.G., and  R.E.  Wyzga.   Economic Measurement of Environmental  Damage;
     A Technical Handbook.  Organisation  for Economic  Cooperation and Devel-
     opment, Paris, France, 1976.  151 pp.

McCandless, L.C.,  R.  Wetzel/Ji Casaria, and K. Slimak. . Assessment cf Indus- '
     trial  Hazardous   Waste  Practices,   Storage  and  Primary   Batteries
     IndustrierEPS530/SW-102C,   U.S.EnvironmentalProtection  Agency,
     Washington, D.C., 1975..  258pp.  GNTIS:PB241204]

McKean,  R.N.  "The Use of Shadow Prices."  In:  Problems in Public  Expendi-
     ture Analysis.   S.B.   Chase,   Jr.,   ed.    The  Brookings  Institution,
     Washington, D.C.,.1968.  pp; 33-77.

Mehlhaff, L.C., T.  Cook,  and  J. Knudson.  "A'Quantitative  Approach to Clas-
     sification of Hazardous  Waste."   In:   Proceedings "df  a  National--Con-
     ference about Hazardous Waste Management, E.B. Workman,  ed.vector and
     Waste  Management  Section,  California State Department.of. Health, Sacra-
     mento, California, 1978.   pp.  267-272.


                                   '  139

-------
 Merrett,  A.J.,  and  A.  Sykes.   The  Finance and Analysis of Capital  Projects.
      Longmans Green  and Co.  Ltd.,  London,  England,  1963.   564 pp.

 Midwest  Research  Institute.   A Study  of   Waste  Generation.  Treatment  and
      Disposal  in the Metals  Mining Industry.   Volume 1.   SW-132c,  Li.S.  En-
      vironmental  Protection Agency,  Washington,  D.C.,  1975.  [NTIS-PB251052]

 Miller,  D.W.   "Decontaminate  Water  Before  It Gets Into  the Ground."   'tetar
      and  Sewage Works.  121(6):51-53, 1974.

 Milliken,  J.G.,  and G.C.   Taylor.   Metropolitan  Water  Management,   'i/atar
      Resources  Monograph Series  #6, American Geophysical  Union,  Washington,
      D.C.,  in press.

 Mishan,  E.J.   Cost-Benefit  Analysis:  An  Introduction.   Praeger  Puolisners,
      New  York,  New York,  1971a.   374 pp.

 Mishan,  E.J.   "The  Postwar Literature  on Externalities:   An Interpretative
      Essay."  Journal of Economic  Literature,  9(1):1-28, 1971b.

•Moll,  K.D.,  S.  Baum,  E. Capener,   F.W. Dresch,  and R.M.  Wright,   "azardous
      Wastes:   A  Risk-Benefit  Framework  Applied  to  Cadmi :„ and  Ass'^stos.
      EPA-600/5-77-002,   U.S.   Environmental   Protection  Agency,  '.vasnington,
      D.C.,1977.   263pp.   [NTIS:PB257951]
 National  Academy  of  Engineering.   Perspectives  on  Benefit-Risk  Decis^'an
         ing.
          PP-
Making.   The  National  Academy  of  Engineering, Washington,  D.C., 1372.
165 pp.
 National  Academy of Sciences.   Considerations of Health Benefit-Cost Analysis
      for   Activities  Involving  Ionizing Radiation Exposure and Alternatives.
      ETA520/4-77-003,   U.S.   Environmental  Protection  Agency,  Washi ngton,
      D.C.,  1977.   199 pp.

 National  Academy of Sciences,  National  Academy of Engineering.   Water Quality
      Criteria  1972,  A   Report  of  the Committee on Water Quality Criteria.
      National Academy of  Sciences,  Washington, D.C.,  1972.  594  pp.[NTIS:
      PB236199]

 National  Oceanic and Atmospheric Administration.   Climatologies! Data, Annual
      Summary:   Oregon.   1976.    Volume  82,  Number  13.   Environmental  Data
      Service, National  Climatic Center, Asheville, N.C., 1976.

 Newcomb,  R.C., and  R.J.  Deacon.   Geological   and Subsurface  Investigations:
      Proposed  Arlington Disposal Site, Gilliam County, Oregon.    Shannon  and
      Wilson,  Inc., Portland, Oregon, 1971.

 Nizard,  L. ,  and J.  Tournon.   "Social   Perception  and Social Demands in  En-
      vironmental Matters."  In:  Environmental  Damage Costs, Organisation  for
      Economic  Co-operation  and  Development,   Paris,  France,  1974.   pp.
      305-330.
                                      140

-------
Oregon Department of  Fish and Wildlife.  Manual for Fish Management.  Oregon
     Department  of  Fish and Wildlife,  Portland, Oregon,  1977.   pp.  157-192.

Oregon  State University.   Temperature  and  the  Water  Balance  for Oregon
     Weather  Stations.    SpecialReport150,AgriculturalExperimental
     Station, Oregon State University, Corvallis, Oregon, 1963.

Orem, H.M.   Discharge in the Lower Columbia River  Basin, 1928-1965.   Geolog-
     ical  Survey Circular 550,  U.S.  Geological  Survey, Washington,  D.C.,
     1968.  128 pp.

Organisation  for Economic  Co-operation  and Development,  ed.   Environmental
     Damage Costs.   Organisation  for Economic  Co-operation  and Development,
     Pans, France,  1974.  332 pp.

Ott,  W.R.,  ed.   Proceedings of the EPA Conference on Environmental  Modeling
     and Si'mulatiofTEPA 600/9-76-015, U.S. Environmental Protection Agency,
     Washington, D.C., 1976.  861 pp.

Ottinger, R.S.,  J.L.  Blumenthal,  D.F.  DalPorto, G.I.  Gruber, M.J.  Santy, and
     C.C.  Shih.   Recommended Methods of Reduction. Neutralization. Recovery,
     or  Disposal  of  Hazardous Waste.16 volumes.EPA/670/2-73-053,  O.S.
     EnvironmentalProtectionAgency,  Cincinnati,   Ohio,   1973.    [NTIS:
     PB224579 - Set]

Oxenfeldt, A.R.   "Multi-Stage Approach to Pricing."  Harvard Business Review,
     38(4):125-133,  1960.

Panner,  P.O.   A Psychological Perspective of the Nuclear Energy  Controversy.
     RM-75-67,  InternationalInstitute for Applied  Systems  Analysis,  Laxen-
     burg, Austria,  1976.  27 pp.

Perna,  A.J.    "Industrial   Solid  and  Liquid  Waste  Handling and  Disposal.11
     Pollution Engineering. 9(1):148-151, 1977.                           ....

Peskin,- H.M., and E.P.  Seskin,  eds. -Cost-Benefit Analysis.and  Water Pollu-
     tion Policy.   The  Urban  Institute, Washington,  Q.C.,  1975.  384  pp~.

Planning  Branch, Treasury Board  Secretariat.   Benefit-Cost  Analysis Guide.
     Information Canada, Ottawa, Canada, 1976, 80 pp.

Porter,   C.H.   State Program  Implementation Guide:  Hazardous Waste Surveys.
     EPA/530/SW-160, U.S.  Environmental  Protection Agency, Cincinnati..Ohio,
     1975.  42 pp.

Porter,   C.H.   State  Program  Implementation  Guide:   Hazardous Waste Trans-
     portation  Control.    EPA/530/SW-512.U.S.Environmental3rotection
     Agency,  Cincinnati, Ohio, 1976.  39 pp.

Portney,  P.R.   "Toxic  Substance Policy and the  Protection of Human  Health."
     In:  Current Issues in U.S. Environmental Policy. P.R. Portney,  ed.   The
     Johns  Hopkins  University  Press,  Baltimore,  Maryland,  1978.  222  pp.


                                     141

-------
Powers,  P.W.   How to Dispose  of  Toxic  Substances  and Industrial  Wastes.
     Noyes Data Corporation, Park Ridge, New Jersey, 1976.   508 pp.

Price, C.M.   Welfare Economics in Theory and Practice.   The  Maonillan  Press
     Ltd., London, England,  1977.   183 pp.

Provenzano, G.   "Risk-Benefit  Analysis  and  the  Economics  of  Heavy Metals
     Control."   In:   Proceedings  of the Conference  on  Heavy Metals in the
     Aquatic Environment.   Vanderbilt University, Nashville,  Tennessee,  1973.
     pp.  339-346.

Rail,  D.P.   "Extrapolating  Environmental Toxicology,  Costs  and  Benefits of
     Being Right and  Wrong."   A paper presented at  the  American Association
     for  the  Advancement  of  Science  Meeting in New York, New York,  January
     26-31, 1975.  [Draft]

Rappeport, M., and  P.  Labaw.   Public Attitudes and Behavior  Regarding Energy
     Conservation:   Detailed  Tabulations  by  U.S. Population and Population
     Segments.Waves 28  and  29.FEA/D-76/252,  Federal  Energy  Aaimnistra-
     tion, Washington, D.C., 1975.  42 pp.  [NTIS:PB254592]

Rasmussen, N.C.,  director.  Reactor  Safety Study:   An Assessn-'nt of Accident
     Risks in U.S.  Commercial  Nuclear Power Plants.  WASH-1400,  U.S.  Nuclear
     Regulatory Commission,  Washington, O.C., 1975.  207 pp.   [NTIS:PB248201]

Reeder,  L.R.,  J.H.  Cobbs,  J.W.  Field,  Jr.,  W.O.  Finley, S.C.  Vokurka, and
     B.N. Rolfe.   Review  and  Assessment of Deep-Well Injection of Hazardous
     Waste.   VolumeIEPA-600/2-77-029a,U.S.EnvironmentalProtection
     Agency, Cincinnati, Ohio, 1977.   215 pp.

Rinella, J.,  and S.  McKenzie.   Elutriation  Study of Willamette River Bottom
     Material   and Willamette-Columbia River Water.  Open  File Report 78-28,
     U.S. Geological Survey, Portland, Oregon, 1977.

Rosfjord,  R.E.,  R.B.  Trattner,  and  P.N. Cheremisinoff.   "Phenols:   A  Water
     Pollution  Control Assessment."    Water  and Sewage Works,  123(3):96-99,
     1976.

Royston,  M.G.,  and  J.C.  Perkowski.    "Determination of  the  Priorities  of
     'Actors'   in the Framework of Environmental  Management."  Environmental
     Conservation. 2(2):137-144, 1975.

Sagan,  L.A.   "Human  Costs  of Nuclear  Power."  Science.  177(4048):487-493,
     1972.

Sanders,  G.R.   "California's  Hazardous Waste Management  Program:    VIII.
     Survey of  Hazardous  Waste  Production."   In:   Proceedings of a National
     Conference About Hazardous Waste Management,  E.B.  Workman,  ea.   Sector
     and  Waste  Management  Section,California State Department  of  Health,
     Sacramento,  California, 1978.  pp. 84-85.
                                     142

-------
Saunders,  P.J.W.   The Estimation of Pollution Damage.   Manchester University
     Press, Manchester, England, 1976.136 pp.

Schlaifer, R.  Analysis of Decisions Under Uncertainty.  Preliminary edition,
     2 volumes.McGraw-Hill Book Company, New York, New York, 1967.

Scitovsky,   T.    Welfare and Competition:  The Economics of a Fully Enroloyed
     Economy.   Ricnard D.  Irwin,  Inc.,  Chicago,  Illinois,   1951.   473 pp.

SCS  Engineers,  Inc.   Assessment of Industrial  Hazardous  Waste  Practices:
     Leather  Tanning  and Finishing Industry.SW-131c,U.S.Environmental
     Protection  Agency,  Washington,  D.C.,   1976.   367  pp.    [NTIS:PB251018]

Scurlock, A.C.,  A.W.  Lindsey,  T. Fields, Jr.,  and D.R.  Huber.  Incineration
     in  Hazardous  Waste  Management.   EPA/530/SW-141,  U.S.  Environmental
     Protection Agency, Washington, D.C., 1975.  108pp..

Seagraves,  J.A.   "Industrial .Waste  Charges."'  ftmen'can  Society,  of   Civil
     Engineers.  Journal of-the Environmental Engineering  Division. 99(££6-):
     873-881, 1973.                                     . •   •••••

Sewell,  W.R.D.   "Broadening the Approach to  Evaluation  in-Resources Manage-
     ment Decision-Making."  Journal  of .-Environmental  Management. 1:33-60,
     1973.

Shaver,  R.G.,  L.C.  Parker, .E^F,  Rissman,. Ol... Slimak,.-and,-.R.C.'  Smith.   • As--•
     sessment of Industrial Hazardous -Waste- -Practices;'Inorganic  Chemicals'
     Industry.EPA/530/SW-104c,U.S.EnvironmentalProtection  Agency,
     Washington, D.C., 1975.  [NTIS:PB244832].-

Slovic,  P.,  and B.  Fischhoff.   "How Safe  is Safe  Enough?   Determinants of
     Perceived and Acceptable  Risk."   In:   The Management of Nuclear Wastes,
     L. Gould and C.A. Walker, eds.  Yale University Press, in press.

Slovic,  P.,  B.  Fischhoff, and  S,  Lichtenstein.   "Cognitive  Processes and
     Societal Risk Taking."  In:   Cognition and Social Behavior, J.S. Carroll
     and  J.Wi  Payne,   eds.   Lawrence  Erlbaum Associates,  -Potomac, Maryland,
     1976.  pp. 165-184.              -                 ...

Slovic, P., H. Kunreuther, and G.F. White.  "Decision 'Processes, Rationality,
     and  Adjustment  to  Natural  Hazards." .  In:   Natural Hazards:  Local,
     National and Global. G.F. White, ed.  Oxford University Press, New  York,
     New York, 1974.   pp. 187-205.

Smith, D.D.,  and R.P.  Brown.   Ocean  Disposal of Barge-Delivered Liquid'and
     Solid Wastes from U.S. Coastal Cities.SW-19c,  U.S.  Environmental  Pro-
     tection Agency,  Washington, D.C..,. 1971.  128 pp.

Smith, M.F., ed.  Hazardous Materials Transportation:  • Part 1. General.'Studies
     (Bibliography^. ..NTIS/PS-76/0331,NationalTechnical   Information.
     Service, Springfield,. Virginia, 1976.
                                     143

-------
State  of  Oregon.   Hazardous  Waste  Management   Planning:   1972-73.   Oregon
     Department of Environmental Quality, Portland, Oregon, 1974.

State  of  Oregon, Department of  Economic Development.   Directory of Manufac-
     turers.  1976-1977.    Department  of   Economic   Development,   Port", and,
     Oregon, 1976.  324 pp.
State of Oregon, Department of Environmental Quality.   Proposed Watar Qu
     Management  Plan for North Coast-Lower Columbia River Basin.  Cecartment
     of Environmental Quality, Portland, Oregon, 1976a.

State of Oregon, Department of Environmental. Quality.   Proposed Watar Cua'Jty
     Management Plan for Willamette River Basin.  Department of Envivcnmenta1
     Quality, Portland, Oregon, 1976b.

State of Oregon,"  Department  of Environmental Quality.   Goals and Qtracv'vas.
     Department of Environmental Quality, Portland, Oregon, 1373.

Statistical  Policy  Division,  Executive Office of  the  President,  Offica  of
     Management and Budget.   Standard Industrial Classification Marua'. 1972.
     U.S.  Government Printing Office, Washington,  D.C.,  1974.

Stermole,  F.J.   Economic Evaluation and Investment Decision '-isthods.  Invest-
     ment Evaluations Corporation, Golden, Colorado, 1974.

Sternes, G.L.   Climates  of   the   States:   Oregon.  U.S.  Government 'renting
     Office, Washington, D.C. , 1967.   27 pp.

Stone, R. ,  and  D.E.  Brown.   Forecasts of the Effects  of Air and Water Pollu-
     tion  Controls  on  Solid  Waste  Generation.   EPA/670/2- 74-035o,  u.S.
     Environmental  Protection  Agency,  Cincinnati ,  Ohio,  1974.   330  pp.
     [NTIS:PB238219]

Stradley,  M.W. , G.W. Dawson,  and B.W. Cone.  An  Evaluation of the Status  of
     Hazardous  Waste  Management in Region X"  EPA 910/9-76-024, u.s.  ;nvi-
     ronmental  Protection Agency, Seattle, Washington,  1975.  202 pp.

Swain, J.W. , Jr.  Assessment of Industrial  Hazardous  Waste  Management ?"ac-
     tices:  Petroleum  Rerefining Industry Study (SIC 2992).   SW-144c,  U.S.
     Environmental  Protection  Agency,  Washington,  D.C.,  1976.   112  pp.
     [NTIS:PB272267]

Szego,  G.D.    The  U. S.   Energy   Problem.    InterTechnology  Corporation,
     Warrenton, Virginia, 1971.

Talley, R.J., and O.W.  Albrecht.  "Preliminary Economic Analysis of Hazardous
     Waste Control."  In-house research project, Program Element IDA312, ROAP
     OIADC.  U.S.  Environmental  Protection Agency, Cincinnati,  Ohio,  1974.
     79 pp.

Taubman,  P.   "U.S.  Prepares  to  Sue  Hooker Corp. on  Dumping  of  Hazardous
     Wastes."  New York Times. May 21, 1979, 1, B-3.


                                     144

-------
Taylor,  G.A.   Managerial and Engineering Economy:  Economic Decision-Making.
     D.  Van  Nostrand Company,  Inc.,  Princeton,  New  Jersey, 1964.   504 pp.

The Science  and Public  Policy  Program.   Energy Alternatives:  A Comcarative
     Analysis.    Chapter  13  "Energy Consumption."University  of Oklanoma,
     Norman, Oklahoma, 1975.

Tihansky,  D.P., and H.V.  Kibby.    "A  Cost-Risk-Benefit  Analysis  of  Toxic
     Substances."    Journal  of  Environmental  Systems.  4(2):117-134,   1974.

U.S.  Atomic Energy  Commission.  Comparative Risk-Cost-Benefit Study of Alter-
     native  Sources  of  Electrical  Energy.WASH-1224,U.S.   Government
     Printing Office, Washington, D.C.,  1974.

U.S.  Department of  Commerce.   Current  Population  Reports, Population Esti'-
     Mates.  Bureau  of the  Census  Series P-26, No.  76-37, U.S.  Department of
     Commerce,  Washington, O.C., 1977.

U.S.  Department of  Commerce, Bureau of  the  Census.   1970 Census  of Popula-
     tion.   Volume  1:  Characteristics of the  Population.   Parts  1 (Sec. 2)
     and 39.   U.S.  Department of Commerce, Washington, D.C., 1973.

U.S.  Department  of  the  Interior, Bureau of Mines.  "U.S.  Net Import Reliance
     of  Selected Minerals and  Metals  as a Percent of  Consumption in 1973."
     Minerals and Materials/A Monthly Survey.  May 1979, p. 9.            '

U.S.   Environmental   Protection   Agency.   "Designation  and  Determination -of
     Removability  of  Hazardous  Substances   from Water:    Proposed  Rules."
     Federal  Register. 39(164):30466-30471, 1974a.

U.S.   Environmental   Protection   Agency.'  Report  to  Congress:   Disposal  of
     Hazardous  Wastes.  SW-115,  U.S.  Environmental  Protection Agency, 1974b.
     120 pp.                            .   ..

U.S.  Environmental   Protection  'Agency.'   National   Water  Quality  Inventory.
     EPA-440/9-74-001,  U.S.   Environmental  Protection  Agency,  Washington,
     D.C.,  1974C.  315 pp.

U.S.  Environmental   Protection Agency.   "Effective Hazardous Waste Management
     (Non-Radioactive):    A  Position Statement."  Federal  Register.  41(161):
     35050-35051, 1976a.

U.S.  Environmental  Protection  Agency.   "Polychlorinated  Biphenyl-Containing
     Wastes:    Disposal   Procedures."   Federal  Register,  41(64):14134-14136,
     1976b.

U.S.   Environmental   Protection  Agency.    Synopsis   of  Public  Meetings  on
     Hazardous   Waste  Management.    U.S.  EnvironmentalProtectionAgency,
     Washington, D.C., 1976c.55 pp.

U.S.   Environmental   Protection   Agency.   Quality  Criteria  for Water.   U.S.
     Environmental  Protection Agency, Washington,  D.C., 1976d. 270 pp.


                                     145

-------
U.S.   Environmental  Protection  Agency.   Ocean Dumping in the United States--
     1977.    U.S.  Environmental Protection  Agency,  Washington,  D.C.,  1977a.

U.S.   Environmental   Protection  Agency.    State  Decision-Makers  Guide  for
     Hazardous  Waste  Management.    SW-612,  U.S.  Environmental  Protection
     Agency, 1977b.   109 pp.

U.S.   Environmental   Protection  Agency.    The  Report  to  Congress:   Waste
     Disposal Practices and Their Effects on Ground Water.  EPA/570/9-77-001,
     U.S.  Environmental Protection  Agency,  Washington,  D.C., 1977c.   531 pp.

U.S.   Environmental  Protection  Agency.   Transcript:   Regional Public Meetings
     on the Resource Conservation and Recovery Act of 1976.SW-20p (series),
     U.S.  Environmental Protection Agency, 1977d.

U.S.   Environmental  Protection  Agency.   Proposed  Revisions to Ocean Dumping
     Criteria:  Final  Environmental  Impact Statement.    Volume  1.   U.S.  En-
     vironmental Protection Agency, Washington, O.C., 1977e.

U.S.   House  of Representatives.   Report  to the  Committee  on Interstate and
     Foreign  Commerce, U.S.  House of Representatives  on H.R.  1-U96.  Report
     No.  94-1491,  U.S. Government  Printing Office, Wasningtor,  O.C.,  1976.

Van Hook,  R.I.   Transportation  and  Transformation  Pathways  of  Hazardous
     Chemicals  From  Solid Waste Disposal.    Publication No.  1131,  Environ-
     mentalSciencesDivision,Oak  Ridge  National  Laboratory, Oak  Ridge,
     Tennessee, 1978.

Venugopal,  B.,  and  T.D.  Luckey.  "Toxicology of Non-Radioactive Heavy Metals
     and Their Salts."  In:  Heavy Metal  Toxicity, Safety and Hormoloqy, T.D.
     Luckey,  B. Venugopal, and 0.  Hutcheson, eds.  Academic Press, New York,
     New York, 1975.  pp.  4-73.

Versar,  Inc.   Alternatives for Hazardous Waste  Management in the  Inorganic
     Chemicals  Industry.    SW-149c,  U.S.   Environmental  Protection  Agency,
     Washington, D.C., 1977.   288pp.  [NTIS:PB274565]

WAPORA,  Inc.   Assessment  of Industrial   Hazardous  Waste  Practices:   Paint
     and  Allied  Products  Industry  Contract Solvent Reclaiming Operations,
     and Factory  Application of Coatings.   SW-119c,  U.S. Environmental Pro-
     tection Agency, 1976.310 pp.[NTIS:PB251669]

WAPORA, Inc.  Assessment of Industrial Hazardous Waste Practices:  Electronic
     Components Manufacturing Industry.SW-140c,  U.S.  Environmental Protec-
     tion Agency, Wasnington, D.C., 1977a.  207 pp.  [NTIS:PB265532]

WAPORA, Inc.  Assessment of Industrial  Hazardous  Waste  Practices:   Special
     Machinery  Manufacturing   IndustrieT!SW-141c,  U.S.Environmental Pro-
     tection  agency, Washington, D.C. 1977b.  328 pp.  [NTIS:PB265981]

Wendt,  D.,   and  C.  Vlek,  eds.   Utility,  Probability,   and  Human  Decision
     Making.  D. Reidel Publishing Company, Boston, Massachusetts, 1975.426
     PP-

                                     146

-------
Williams,  R.,   R.  Shame 1,  K.  Hallock,  B.  Stangle.^'and S.  Blair. '••'EgionciinJc
     Assessment of  Potential  Hazardous  Waste  Control  Guidelines  for the
     Inorganic  Chemicals   Industry^EPA/530/SW-134c,U.S.Environmental
     Protection Agency, Washington,  D.C.  1976.   320 pp.

Williamson,  S.J.   A Comprehensive  Survey  of -Hazardous". Waste  Geherafon in
     Oklahoma.   Ph7D~!Dissertation,UniversityofMichigan,AnnArbor,
     Michigan, 1975.  349 pp.
                                     .147

-------
                                 APPENDIX A

                    HAZARDOUS WASTE MANAGEMENT TECHNIQUES


     This appendix  describes the  techniques  (listed in Table 2  of  the ?ain
text)  that  may  be  used  for the  control  (including disposal)  of  "acarpous
wastes.

TECHNIQUES INVOLVING WASTE STREAM CHANGES

Process Change

     In 'process  change,  the  industrial  process that generates  the  waste is
changed.  Substitution of a different process will  normally "2suU -'n genera-
tion of  a different waste.   The new waste could be inherently less nazsrdoLS
or nonhazardous, or it  could be generated in smaller quantities than aer'3'"'
An example  of process change is  the  replacement  of the  mercury call  oy a
diaphragm-cell for chlorine production.  It appears that this changeover (aT
new capacity  is  expected to use diaphragm cells) has been caused entirely by
the  problems  associated  with wastes  and  emissions  from  the mercury  cell
(Saxton and Kramer,  1974).

     It  is  not  necessary  to substitute  a new  process  to change the  waste
streams.  In some cases,  process modifications such as changing the operating
conditions  or adding  process  steps  (including  pollution control  devices)
could  cause the  composition of the wastes  to  change,  or could simoly change
the  volume  or concentration  of  the waste  stream  (Saxton  and Kramer,  1974).

Source Reduction

     With source reduction,  the  basic composition of a  waste stream remains
unchanged (except perhaps  for .concentration),  but the quantity  of  the waste
is reduced.   This may be achieved by process modification (including the more
efficient  use  of   materials),  by  changes in  the quality  of  the  material
inputs,  or by  improving  procedures  to  reduce  production   spoilage,  etc.
(Saxton  and  Kramer,  1974;  U.S.  Environmental  Protection  Agency,  1976b).

Waste  Separation

     Waste  separation involves  segregating waste streams to  isolate  those
wastes  that are hazardous  from  those  that are not, or  to  keep  apart wastes
with different hazardous properties.   In the former case, the objective is to
reduce  the  quality  of hazardous waste to be handled.  In the latter case, it
is presupposed that the  mixed waste is  more  difficult  or costly to treat or


                                     148.

-------
dispose of  than the same  total  volume of waste made  up  of several streams,
each of which contains a smaller number of constituents.  This supposition is
not  universally  valid,  as  there could be  antagonistic reaction between two
waste streams (such  as  neutralization), or economies of scale could outweigh
any added complexities of treating or disposing of mixed wastes.

RESOURCE RECOVERY

     In resource  recovery,  the magnitude and composition .of the waste stream
is unchanged, but some  of the materials  or the energy content of  tne stream
is recovered and put to beneficial use.

Materials Recovery

     Recovery of  materials  is often carried out  in  conjunction with various
treatment processes  (U.S.  Environmental  Protection  Agency,  1974).   It does
not  necessarily  achieve  total, recovery of all materials present in the waste
stream.   In many cases,  only the more valuable or readily  isolated constitu-
ents are recovered.

Energy Recovery

     As an  alternative to  materials recovery, where  the waste stream has a
significant calorific value,  energy recovery may be practiced.  This usually
involves burning  the waste'in some type of incinerator that is equipped with
a heat exchanger to enable the heat to be captured and  put  to use.

WASTE TREATMENT           •"

     There are numerous treatment processes that may. be used to render wastes
less hazardous.   Table A-l lists some of the processes  that have been identi-
fied as being appropriate to the treatment of hazardous wastes.   Details of
these and  other  treatment  processes are provided in  Ottinger.et al.  (1973,
Vols. 3 and  4).-  Many treatment processes  are specific to  a limited range of
waste types, and  for this reason they are not di-scussed here. '           ' '

     Treatment processes do not eliminate the waste  stream, but by  separating
out  harmless  components  from those that  are  hazardous, some  processes may
significantly  reduce  the  quantities   of   hazardous  wastes  that  ultimately
require disposal.   Volume reduction by the  evaporation of  water or the pre-
cipitation of a  hazardous sludge leaving a  nonhazardous effluent are examples
of  such  treatments.  Some  wastes can  be  rendered nonhazardous- by treatment
(e.g.,  neutralization of  sulfuric  acid with  lime),  whereas  in other cases,
the  treatment  may  be  a preliminary step toward  disposal  (e.g.,  a change of
chemical  form  to  reduce  the waste's  mobility or toxicity).    Encapsulation,
described below, is  invariably  followed by  a storage or  disposal process.

Encapsulation

     Where a waste is not readily amenable to a detoxification treatment, it
may  be desirable to  immobilize • it  in  some  way  so  that control  can more
readily be maintained over it.   Encapsulation  is often  used to  prevent (or at


                                     149

-------
                 TABLE  A-l.   CURRENTLY AVAILABLE  HAZARDOUS WASTE  TREATMENT  AND  DISPOSAL  PROCESSES*
in
o

Process
Physical treatment:
Carbon sorption
Dialysis
Electrodialysis
Evaporation
Filtration
Flocculati on/settling
Reverse osmosis
Ammonia stripping
Chemical treatment:
Calcination
Ion exchange
Neutralization
Oxidation
Precipitation
Reduction
Thermal treatment:
Pyrolysis
Incineration
Detonation
Biological treatment:
Activated sludges
Aerated lagoons
Waste stabilization ponds
Trickling filters
Soil application
Functions
performed

VR,
VR,
VR,
VR,
VR,
VR,
VR,
VR,

VR
VR,
De
De
VR,
De

VR,
De,
Di

De
De
De
De
De

Se
Se
Se
Se
Se
Se
Se
Se


Se, De


Se


De
Di







Types

1,
1,
1,
1,
1,
1,
1,
1,

1,
1,
1,
1,
1,
1,

3,
3,
6,

3
3
3
3
3

2,
2,
2,
2,
2,
2,
2,
2,

2,
2,
2,
2,
2,
2

4.
5,
8






of

4,
3,
3,
5
3,
3,
4,
3,

5
3,
3,
3,
3,


6
6,







waste

5
4
4, 6

4, 5
4, 5
6
4


4, 5
4
4
4, 5



7, 8







Forms
of waste

L, G
L
L
L
L, G
L
L
L

L
L
L
L
L
L

S, L. G
S, L, G
S, L, G

L
L
L
L
L, S
Resource recovery
capability

Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes


Yes
Yes

Yes


Yes
Yes
No

No
No
No
No
No

       * Source:   Adapted from U.S.  Environmental  Protection Agency,  19741).
         Functions:   VR,  volume reduction;  Se,  separation;  De,  detoxification;  and Di,  disposal.
         Waste types:   1, inorganic  chemical  without heavy  metals;  2,  inonjanic chemical  with heavy metals;
         3,  organic  chemical  without heavy  metals; 4,  organic cheinkal  with  heavy me-Lais; 5,  radiological;
         6,  biological; 7, flammable; and 8,  explosive.
       c Waste forms for which process is feasible:   S,  solid;  1,  liquid;  and G,  yas.

-------
le'as't severely  retard)  leaching  and consequent contamination of groimdwater.
The technique  is  commonly applied to low-level radioactive wastes* (Ottinger
et al.,  1973,  Vol.4).   Hazardous wastes may  be encapsulated  by mixing the
waste with  concrete,  asphalt  and various plastics  (such  as  polyethylene or
polyurethane)  (Fields  and Lindsey,  1975).   Encapsulation  in  glass  has been
proposed  for  high-level   nuclear wastes  (Energy  Research and  Development
Administration, 1977).   Often, for convenience,  the  waste and encapsulating
medium are  solidified in a steel  drum,  and it is  sometimes possible to use
off-specification resins  as the  encapsulating  medium.  The resulting mixture
is typically  60 percent  (by  weight) of  waste when mixed  with  a resin, but
only 25  percent waste  when  encapsulated in cement (Ottinger  et al., 1973,
Vol.4).

     One encapsulation technique, recently developed specifically for hazard-
ous chemical wastes, agglomerates the waste in  polybutadiene and then jackets
the agglomerate in  a thin layer of polyethylene.   The attraction  of this
technique is that  the waste  can constitute '94  'to -96 percent ofthe agglomer-
ate;  nevertheless,  the process  is  still costly, (Wiles  and Lubowitz, 1976). .
                                                                          •'.«..
Incineration

     There are  three  hazardous-waste management techniques that  fall  on the
borderline  between treatment  and  disposal. "These  are  i'ntineration,  land
application,  and  lagooning.   .(The  latter  two  are  discussed  later under
Storage and Disposal Techniques.)         ••••-.          -  .

     Incineration has wide potential  application to hazardous.wastes.  I-t is
a  controlled  process that  uses   combustion to convert  the waste to  a less
bulky,   less  toxic,  or  less  noxious material.   The  principal  products..of
incineration are  carbon  dioxide, water,  and  ash;  but products  of  primary
concern  (because  of  th'eir  deleterious  effects)  are  compounds  containing
sulfur,  nitrogen, and halogens.  Where the combustion products from an incin-
eration process  contain  undesirable  compounds,  secondary  treatment  such as
afterburning, scrubbing, or filtration is re-quired to lower concentrations to
acceptable  levels  for  atmospheric  release (Ottinger  et al.,  1973,  Vol.3).
Thus  incineration  largely  converts the  waste  to  a harmless  gaseous form,
usually  leaving  only •comparatively  small  quantities • 6f  ash and  scrubber
sludge  that.require disposal.                           ....
     There  are  many  di f f erent' types" 6'f " i nci nerators  that  can be  usred on
industrial  wastes,  and  different types  of  incinerators  can  handle solid,
liquid, or  gaseous wastes.   Ottinger et al. -(-1973,  Vol.3)  and Powers (1976)
* In radioactive waste management, the term "high-level" is applied to wastes
  in which  there  is significant  heat  generation  arising  from radyoactive
  decay; low-level wastes  are-those in which this effect is not significant.
  It is  interesting  to  note that Federal  regulations  require  the conversion
  of commercial high-level  liquid  wastes to a stable  solid form preparatory
  to terminal  storage (Energy Resources Council,  1976).
                                     151

-------
provide  detailed  descriptions  of the  various  types,  and  Scurlock et  al.
(1975)  specifically  discuss  incineration  for  hazardous  waste  management.

     There are four  technical characteristics  that affect waste incineration
(Ottinger et al.,  1973, Vol.3).   The first is combustibility, i.e., a measure
of the ease with  which a waste can  be  oxidized in a combustion environment.
The next  two  are dwell  or residence  time  in  the  combustor, and  the  flame
temperature.   These  parameters  affect the degree of  combustion.   The fourth
is the turbulence present  in  the reaction zone  of  the incinerator, which is
required to-insure sufficient mixing of the  air and the waste fuel.

     Turbulence and  dwell  time  are  determined by the incinerator design,  but
for a given  incinerator,  flame  temperature  can  be  varied within  certain
limits.   It  thus  follows  that  different incinerators  will  be  more  or lass
appropriate to  the   treatment of  different  wastes.   Though  high temperature
capability incinerators  with  long dwell  times would  usually be  capable  of
adequately treating  wastes that  need only  low temperatures  and short  dwell
times, such incinerators are  more costly to build  and  operate  than low tem-
perature incinerators.  Hence economic as well  as technical factors influence
the appropriateness   of  a  given  incinerator  design  for  treating different
wastes.

     Clearly  it  is   a  prerequisite  of incineration that the total naterials
stream entering the  incinerator has  sufficient calorific value to ac.nieve the
desired  dwell time   and  temperature.   If   the waste  cannot   fulfill   this
requirement,'  it "can  be  supplemented with a fuel.   However,  where the  calo-
rific  value of the  waste stream  is  itself  sufficiently high, it is possible
to  recover energy from  the  waste via  a heat  exchanger and hence generate
power, process  steam, or  use the surplus  energy  in some other useful way.

     Among the most attractive candidates for incineration are organic wastes
(including many  pesticides)  that are hazardous because of  the  structure of
the molecule  (for example, synthetic organics such  as RGB's),  rather than
those  that are.hazardous  because of some elements  contained in the molecule
(e.g., wastes containing  heavy  metals).   Incineration  may be attractive  for
the disposal  of many ordnance wastes and for some inorganic wastes (Ottinger
et al., 1973,  Vol.1).  Powers (1976) and Scurlock et al. (1975)  provide  lists
of materials  that may  be  suitable  for  incineration,  both  largely based on
Ottinger et al.

     A very specialized  form  of incineration is that  of  incineration at sea
using  purpose-designed  ships  (see  subsequent  section  on   ocean  dumping).

STORAGE AND DISPOSAL TECHNIQUES

Land Application

     Land  application involves  spreading or  spraying  of  wastes  over  large
areas  of  land.  This technique  is often used for certain nonhazardous wastes
such  as  wastewater   (Stewart,  1973; U.S.  Environmental   Protection  Agency,
1975),  sewage  sludge,  animal   and  food  processing  wastes,   and  certain
industrial  wastes  where   the  waste  contains  materials  (nutrients  or soil


                                     152

-------
conditioners)  that, should  enhance crop growth  (Loehr,  1977;  U.S.'  Environ-
mental  Protection Agency,  1977b).   Land  application  is sometimes  used  for
some  biodegradable  hazardous wastes, primarily  oil-related wastes*  (Snyder,
Rice, and Skujins, 1976; Tinnan, 1978; Lofy, 1977).  The EPA is investigating
the   effectiveness   of  land   application  for   other   industrial   sludges
(Schomaker,  1976).   Naturally,  land application  should  be used  only  wnere
there is  careful  control  of access to the land and of its future use, and it
is inappropriate  for any  waste that contains appreciable quantities of non-
biodegradable hazardous components.

Landfill ing

     The  term  landfill ing will  be used to denote  any  type  of land buna! of
wastes close to the surface (as opposed to engineered storage, mine disposal,
and deep  well  injection).The technique is commonly used-to dispose of jnany
types of  solid  wastes  and sludges, but it is also used to dispose of liquids.
(which  are  usually  poured  onto  the more solid  components).   There  is an
extensive body  of literature  on landfill  disposal of hazardous wastes (e.g-.,
Fields and Lindsey, 1975;  Fuller, 1976;  Ghassemi  and Quinlivan, 1975), and on
the  physical  effects  that  may  arise from  landfilling  (e.g.,  Geyer,  1972;
Fungaroli, 1971;  Genetelli  and Cirello,  1976; Hill and  Zipp,  1974;  Banerji,
1977; Garland and Mosher,  1975;  Pavoni,  Hagerty, and Lee, 1972; and Schultz,
1978).  The following material is based on these and other sources.

Types of Landfills—
     Open dumps—The least  sophisticated  form of land disposal  is  fie open
dump, in  which  a waste is simply deposited on the ground and left.  Clearly,
the open dump is an inappropriate means for disposing of any hazardous waste;
it would  only  be acceptable (aesthetics  aside)  for inert wastes such as some
demolition debris.

     Sanitary 11andfi11s—The  sanitary landfill  (e.g.,  California  Class  II
landfill [California State Water Resources Control Board, 1976]) provides for
some environmental protection  from the wastes.   In a  sanitary  landfill,  the
wastes are compacted to the smallest practicable volume and covered, usually
daily, with  earth.   These  procedures  minimize   problems with  blowing litter
and with  disease transmitting vectors (animals  and  insects):   Waste compac-
tion  and  a cellular construction  of the  s-anitary landfill  also  reduce  the
possibility of  fire and  of its  spread should one  occur  (U.S.  Environmental.
Protection Agency, 1976a).  Microbial  decomposition of-wastes.results in the
generation  of  gases  (principally  methane  and  carbon  dioxide)   that  are
generally regarded as  a problem,  but there have been some successful methane
recovery projects (U.S. Environmental Protection Agency,  1976a).

     The  hydrologic  conditions  at  a  landfill  are  of  great  importance.
Groundwater  or  infiltrating  surface water  moving through  solid waste  can
produce  "leachate,"  a  solution  containing  dissolved  and  finely  suspended
solid matter and  microbial  waste products.  The composition  of  the  leachate
  Waste petroleum  oil  has been regarded as  hazardous  by some authors and as
  nonhazardous by others.   (See Jacobs Engineering Co., 1976.)


                                   '  153

-------
naturally  depends  on  the  waste  composition  and  also  on  the  physical,
chemical, and biological activities  within  the fill.   Leaching can  be  mini-
mized by  landfill  designs that  restrict the  ingress  of surface watar,  but
some authorities hold that generation of some leachate  is inevitable (Brunner
and Keller,  1972).

     Where annual precipitation is low compared with  potential  evapotranspir-
ation,  over  the course  of  a year  actual  evapotranspiration  may  balancs
infiltration, resulting in zero  net  percolation and  hence negligible longrun
leachate  production  (Fenn,  Hanley,  and DeGeare,  1975).   This situation  is
characteristic of Southern California (Fenn, Hanley and DeGeare,  1975),  where
large  quantities of  liquid  wastes   are  routinely  injected  into  landfills
apparently without producing  leachate (Tinnan,  1978).   Nevertheless, even in
arid  areas  where  leaching  is  expected to be  negligible,  a gooa  landfill
design will  attempt to restrict the potential environmental  damage that  could
be  caused  by a  leachate.    In  areas  with less  favorable  climates  (e.g.,
Cincinnati,   Ohio,  or Orlando,  Florida) (Fenn,  Hanley,  and DeGeare,  1975),
where  leachate  production  is inevitable  (unless of  course  the surface  or
near-surface of the landfill  is rendered impervious),  some means  of  isolating
the leachate  from groundwater is  essential  to provide  complete environmental
protection.

     Chemical landfills—The  chemical  landfill   (e.g.,  California  Class  I
landfill  [California State Water  Resources  Control Board,  1975])  is designed
to accept industrial  wastes that may include hazardous  wastes.   In a chemical
landfill  particular  attention  is  paid  to  minimizing  the  potential  for
leachate contamination of water sources.  Thus, a chemical  landfill  snould be
designed  so  that any surface  water  runoff  is collected and treated, and so
that  there   is  virtually   no  chance  that   leachate will  percolate  into  any
aquifer.

Isolation of Landfill Contents from the Environment--
     Three  principal  means  are  available  to  minimize the probability that
landfill  leachate can contaminate groundwater:   Geologic isolation, landfill
liners, and  leachate  collection  systems.   Note that  in this report, the term
"leachate" will  be used to denote any  aqueous-based liquid  that  may emanate
from  a  landfill or  lagoon.   Thus,  leachate  may be generated either by  the
interaction  of  environmental water (precipitation, surface water or ground-
water) with  an  essentially solid waste as  described above,  or it may be  the
aqueous  component of  a liquid  or  semi-liquid  waste  that is  sufficiently
mobile  to be able to leave  the  landfill  or lagoon.   With  regard to ground-
water contamination,  the  primary  concern  is  protection of  usable  aquifers,
not  of  small   lenses  of   perched groundwater  that  are  not  significant  as
potential water supply sources.

     Geologic isolation—Geologic  isolation involves  selecting the  landfill
(or  lagoon)  site such  that  there is no natural  hydrologic  interconnection
between the  fill  and aquifer.  This condition may be fulfilled if the perme-
ability of  the  soil  or rock  that  separates  the  landfill  from any aquifer is
sufficiently  low  (i.e., essentially  zero).   This approach is  favored by  the
California Class I landfill regulations, which specify geologic isolation  for
vertical water movement, but which permit liners to control  lateral  movement.


                                     154

-------
A  modification  of  .this  approach  is  found  in  arid  areas,  where  natural
leachate generation  (i.e.,  that resulting from the  infiltration  of external
sources of water) plus any liquid emanating from the waste, is  expected to be
slight.   In  this  case,  provided  the  vertical  distance  to groundwater  is
sufficiently large,  there is less concern over the permeability of the inter-
vening strata on the assumption that the total quantity of leachate generated
will be insufficiently great to percolate down to the groundwater.

     Many  soils  have the capacity  to attenuate leachates  that  pass  through
them  or  to  render  these   leachates  less   hazardous  (see  Roulier,  1977;
Farquhar, 1977; Schultz,  1978).  This can be regarded as a form of treatment,
but  if  the  mechanism  is  that of  ion  exchange  (as  opposed to  microoial
action),  the  treatment  capacity,  although  often  very large,  will   not  be
unlimited, since there is no regeneration mechanism.

     Landfill  liners—Liners are usually composed of either a layer of ifflper-
vious  soil (such  as  clay),  asphalt,  of a  polymeric membrane (Haxo, Haxo and
White,  1977;  Geswein, 1975).   They  may  be used  to replace 6r supplement
geologic isolation  to provide  separation  fronTgroundwater.   For example,•• a
liner  could be  used  to  provide a seal  over  a faulted or fissurec zone'of an
otherwise  impervious stratum,  or  it  could  be used  to control   the  lateral
movement of  leachate where  geologic  isolation is  effective in  the vertical
direction.   Liner materials  could also be used to cap a completed landfill to
prevent the ingress  of surface and nearrsurface waters.  There is one impor-
tant difference between,  complete geologic isolation and the use  of a liner;
because of the  comparative  thinness  of a liner, in most cases  it is probably
only a  matter  of  time (in the context of perpetual care) before  the leachate
penetrates the liner.

     Leachate collection  systems—Where  significant  leachate  production  is
expected,  something  must be  done  with  the leachate.   Leachate collection
systems can be  used  to  divert the leachate  into  treatment or holding tanks.
Even if a  collection system does not  collect  all  the leachate,  the quantity
that is available to threaten aquifers is reduced,  leading to a potentially
more  secure  operation.    Collection  systems  often are .made  up  of a  porous
medium  (e.g.,  loam  or  gravel)  that  permits  the  leachate to migrate into
headers for collection and  treatment..  The porous medium is placed on top of
a liner or other impervious  layer (Fields and Lindsey, 1975).  Of course, the
collected leachate will  usually constitute an additional hazardous waste that
must • be  appropriately  managed  (e.g.,  by  precipitating  -a  slucge that  is
returned to some form of  land disposal);

Mine Disposal

     Disposal  of  hazardous  wastes  in underground mines has been  proposed for
both radioactive  and nonradioactive  wastes.   The  attraction of  the approach
lies in the high  degree  of environmental  protection  that  can  be provided by
such storage  because of  the impermeability and geological  stability  of the
candidate formations. ..The material  of greatest interest is salt  (.both bedded
and domed  formations), followed by  gypsum and potash.  Shale,  limestone, and
granite formations have  also-been  considered.(Stone et al., 1975).  The most
widely  accepted  concept  .is  to place the  solidified  containerized wastes in


                                     155

-------
disused  room and  pillar salt  mines  (Kown  et  al.,  1977).   This  means  of
disposal  has  substantially  greater  direct  economic costs  than landfill ing,
especially  if  it is  necessary  to construct  a mine for  this  purpose rather
than to  adapt  an abandoned  mine.   A special  feature of this approacn is that
provision can be made for future retrieval of the wastes.

Lagooning

     Lagooning involves  placing liquid wastes in open ponds and may incorpo-
rate some of  the chemical  or biological  treatment  processes  listed in Table
A-l, such  as biological oxidation  by means  of aerated lagoons or oxidation
ponds (Ottinger  et  al.,  1973,  Vol.4).  Where  the evaporation  rate is suffi-
ciently   high"  (i.e.,  where  open-pan   evaporation   significantly  exceeds
precipitation),  lagooning  really  constitutes a treatment process  for volume
reduction.   In  other cases,  it  is  more appropriately regarded  as a storage
technique.   In  either  case,  quantities  of  sludges  and/or  solids  will
ultimately require disposal.

     Since the contents  of  a lagoon are  at  least part liquid, protection of
the  groundwater  below and  adjacent  to the lagoon  is  of  particular signifi-
cance.    The  techniques  that  can  be  used  to achieve this  have already been
discussed under  landfill ing.  It  is  also important  to insure tr.it the iagoon
does not  overflow  and thereby contaminate surface waters, and ^nat birds are
protected by being discouraged from landing on the surface.

Deep Well Injection

     Deep well injection involves  disposing  of liquid wastes by pumping them
into deep wells, whereby the wastes become contained within  the interstices
of  the  rock.   This  procedure  has been  used for decades  to  dispose  of oil
field brines, and it is an  accepted means of  disposal  for  such wastes (Reeder
et  al.,  1977).   WAPORA  has catalogued all  the  wells used  for injection  of
industrial wastes  (excluding  oil  field brines) in the United States (WAPORA,
Inc. ,  1974).   Over  half   the  wells  have  been constructed   in  Texas  and
Louisiana, and 68  percent  serve SIC 28 and  29 (chemical  and allied products
and petroleum refining industries).

     The  term "deep  well"  may  be  something  of a  misnomer,  as  the concept
merely  involves  disposing   of the  waste in formations that  are below usable
aquifers.  Nevertheless, 90 percent  of the 278 U.S. wells used for injecting
industrial and municipal wastes  identified by WAPORA, Inc.  (1974) were more
than 305.m .deep.

     In  virtually all deep  wells  used for industrial waste  disposal,  wastes
are  injected into  sands,   sandstones,  or carbonates  (WAPORA,  Inc.,  1974).
Suitable  strata  almost  invariably  contain saline groundwater,  hence injected
wastes  will  displace and/or  mix  with this  groundwater.   Because  any  solid
matter will  plug the host  rock pores, it  follows that only filtered liquids
that are  compatible  with the host fluid (i.e., do not form precipitates) can
be  injected.   Rock  fracturing  to  increase   permeability  is  feasible  under
certain  circumstances (Reeder et  al., 1977).   There is an  extensive body of
literature on deep well  waste injection (Rima, Chase,  and  Meyers, 1971;  Cook,
1972),  and  the  topic has been  well  summarized by  Warner and  Orcutt (1973).

                                     156

-------
Ocean Dumping

     Ocean dumping  has  been  a common means of disposal of unwanted naterials
for  centuries  (Miller,  1973).   Although  the  quantities involved  have  been
dominated  by  dredge spoils,  significant  quantities  of  sewage sludge  and
industrial wastes were dumped in the 1960's, together with smaller Quantities
of  construction  and demolition  debris,  solid waste,  explosives,  and radio-
active wastes (Council on Environmental Quality,  1970).

     The  intent  of  the  Marine  Protection, Research  and Sanctuaries  Act  of
1972  is   to  strictly limit  the  ocean dumping  of any  material that could
adversely  affect  the  marine  environment  (U.S.   Environmental  Protection
Agency, 1977a); and  in recent years, the quantities of industrial wastes that
have been dumped off the U.S. coast have declined significantly.

     The three basic techniques for the ocean disposal  of wastes are:

     1.   Bulk   disposal   of  liquid  or   sludge/slurry  wastes  ising
          specially constructed barges;

     2.   Scuttling  vessels   filled  with  wastes   (usually  obsolete
          munitions);

     3.   The  sinking  at sea of containerized wastes (e.g.,  in 0.2 m3
          [55-gal]  drums)  that  are carried as  deck cargo  on merchant
          vessels.

     Depending on  the details  of the  techniques and the waste involved, the
ocean  may be used as a  reacting  or neutralizing medium, as  a diluent,  as a
cushioning medium  (for  detonated  explosives),  or for  protective  isolation
(Ottinger  et  al.,   1973,  Vol.3).   In  general,  ocean  dumping sites  avoid
estuarine locations, and though  some barged disposal takes, pi ace in compara-
tively shallow water,  containerized  wastes and vessels  are  scutt'ed in the
deep sea (Smith and Brown, 1971).     .                          "

     The  type  of waste  strongly affects its physical disposition.   Hazardous
wastes that  would  float are  clearly unacceptable for ocean disposal.  Wastes
that are  considerably denser than seawater fall to  the  ocean floor, and any
dispersion that occurs will be through ocean bottom processes.  Most aqueous-
based wastes,  on the other hand, have densities similar to  seawater and can
diffuse widely (Ocean Disposal Study Steering Committee,  1976).  Clark et al.
(1971) review  these physical diffusion  processes  and also  discuss disposal
economics.

     A variety of  data  on the practice  and impacts  of ocean disposal may be
found  in  the  literature  (Council  on  Environmental  Quality,  1970;  Ocean
Disposal   Study Steering  Committee,  1976; Smith and Brown, 1971;  Reed, 1975;
Interstate Electronics Corporation, 1973).

     Incinerator ships  can be  considered as a special  case of ocean dumping.
While  the objective of  incineration is to thermally  degrade  the  hazardous
material, it is  almost  inevitable that- a  small  proportion of the waste will


                                     157

-------
escape degradation.   Furthermore,  the  products of degradation may themselves
be hazardous and require some form of disposal.  Thus by conducting incinera-
tion  at  sea,  the  wastes  emitted from the  incinerator  are widely dispersed,
which may be  more  environmentally acceptable than using similar incineration
equipment on  land.   The  technique  has  been  in  use  in  Europe  since 1963
(Powers, 1976).

     Incineration  of U.S.  organochlorine  wastes  has  been conducted  on  an
experimental basis in the Gulf of Mexico using the Dutch incinerator ship M/T
Vulcanus.  The incinerators achieved upwards of 99.9 percent oxidation of the
wastes,  and the  resulting emissions (which  included  hydrogen  chloride) were
discharged to the atmosphere without scrubbing (Wastler et al., 1975).  Had a
similar operation been performed on land, scrubbing would doubtless have been
required to provide environmental protection from the hydrogen chloride (see,
Maritime Administration, no date).

Engineered Storage

     As already indicated, most forms of disposal amount to storage, as it is
not possible  to  eliminate matter.  However, the tern "engineered storage11 is
usually  reserved for placement of wastes into manmade structures (as opcosed
to burial in the ground, for example).

     This technique  has  been  advocated largely for  high-level  raaioactive
wastes  and  for other wastes for which  no  satisfactory alternative means or
disposal  exists.   The  intention  is  that the wastes  are  stored  under very
carefully controlled  conditions until  a safe means of disposal can be found;
for this reason, provision for  easy retrieval must be an integral part of the
design (U.S. Environmental Protection Agency, 1974).

Other Techniques

     Disposal into space has been proposed  for certain  radioactive wastes and
represents  a  unique concept.  As far as the author is  aware,  it has not been
considered  for nonnuclear  wastes, doubtless  because  of  its  extremely high
direct  cost,  which  is  in excess  of  $2,000 per  kg of  waste  including con-
tainers  (Battelle  Memorial  Institute,  Pacific Northwest Laboratories, IS76).

     All other  techniques that have been proposed  appear  to constitute sub-
categories  or  special cases of those discussed  above.   For example, thermal
treatment  can be  split  into   four  principal  subcategories,  which  are  (in
descending  order  of control  maintained  over the  waste  and its decomposition
products):  Pyrolysis,  incineration,  open burning, and detonation.  At least
some  of these  subcategories  are capable  of further  division,  as  already
illustrated for  incineration.   Even ice sheet disposal, which has been pro-
posed  for  some  radioactive  wastes  (Battelle  Memorial  Institute,  Pacific
Northwest  Laboratories,  1976), can be  regarded as  a  special case  of land
disposal, since many  of its  features and potential threats  to  the environment
are   similar  to   the more  conventional   forms  of  land  disposal  already
discussed.
                                     158

-------
REFERENCES

Banerji, S.K., ed.  Management  of  Gas  and  Leachate in Landfills. EPA-600/
     9-77-026, U.S.  Environmental  Protection Agency, Cincinnati, Ohio, :.977.
     288 pp.

Battelle  Memorial  Institute,  Pacific Northwest  Laboratories.   Alta-iatives
     for  Managing  Wastes From  Reactors and Post-Fission  Operations in the
     LWR  Fuel  Cycle.   Volume 4:   Alternatives  for  Waste  Isolation  and
     Disposal.   ERDA-76-43,  Energy Research  and  Development Administration,
     1976.

Brunner,  D.R.  and D.J.  Keller.   Sanitary  Landfill  Design  and  Operation.
     SW-65ts,  U.S.  Environmental  Protection  Agency,  Washington, D.C., 1972.
     59 pp.

California  State  Water Resources  Control  Board.   Waste .Discharge  Require-
     ments for  Nonsewerable  Waste  Disposal to Land: . Disposal Site  Design
     and  Operation  Information.    California State Water  Resources Control
     Board,  Sacramento, California, 1976.  60 pp.          '     • •       • ""

Clark,  B.D. ,  W.F.  Rittal,  D.J.  Baumgartner,  and  K.V.  Byram.   The  Barged
     Ocean Disposal of Wastes:  A.Review of Current  Practice and Methods  of
     Evaluation"!    U.S.  Environmental  Protection  Agency, Corvallis, 0regon,
     1971.  119 pp.  [NTIS:   PB204868]'

Cook, T.D.,  ed.  Underground Waste Management and Environmental  Implications..
     Memoir  18,  The  American  Association  of  Petroleum Geologists,Tulsa,
     Oklahoma, 1972.                    ••       •

Council on Environmental Quality.   Ocean  Dumping:  A National  Policy.  U.S.
     Government Printing Office, Washington, D.C., 1970.45 pp.

Energy Research and  Development Administration..  Alternatives for   Lonq-Term
     Management of  Defense  High-Level - Radioactive  Waste,  Savannan  River
     Plant,  Aiken. South  Carolina.Volume 1.ERDA 77-42/1, Energy Research
     and Development Administration, Washington, D.C., 1977.

Energy  Resources  Council.   Management  of  Commercial  Radioactive  Nuclear
     Wastes:  A  Status  ReporTFEA/A-76/295,   FederalEnergy Administra-
     tion, Washington, D.C., 1976.   18 pp.

Farquhar, G.J.   "Leachate Treatment  by  Soil Methods."   In:   Management  of
     Gas  and  Leachate in  Landfills. S.K.   Banerji,  ed.   EPA-600/9-77-026,
     U.S.EnvironmentalProtection   Agency,   Cincinnati,    Ohio,   1977.
     pp. 187-207.

Fenn,  D.G.,  K.J.  Hanley,  and  T.V.   DeGeare.   Use  of  the  Water  Ba.lance
     Method, for  Predicting  Leachate  Generation  from Solid Waste Disposal
     Sites.  EPA/530/SW-168.  U.S. Environmental Protect!on.Agency, Cincinnati,
     Ohio, 1975.   40 pp.
                                     159

-------
Fields., T. ,  Jr., and A.W. Lindsey.  Landfill  Disposal  of Hazardous   Wastes:
     .A  Review  of  Literature  and  Known Approaches.   EPA/530/SW-165,  U.S.
     Environmental Protection Agency, Cincinnati, Ohio, 1975.   36  pp.

Fuller, W.H.,  ed.   Residual  Management  by " Land' Disposal':--  Proceedings  of
     the Hazardous Waste Research Symposium.EPA-600/9-76-015,  U.S.  Environ-
     mental  Protection Agency, Cincinnati, Ohio,  1976.  280pp.  ,.....-.-•

FungaroTi   A. A.   Pollution  of  Subsurface  Water   by   Sanitary   Landfills.
     3 volumes.   SW-12RG,   U.S.  Environmental  Protection Agency,  RocKviila,
     Maryland, 1971.

Garland^ -G.-A. , •• and   D.C.   Mosher.    "Leachate 'Effects of  Improper  Land
     Disposal."  Waste Age, 5(ll):42-48.  19.75.

Genetelli,  E.'JT,' and  J.Cirello,  eds.   Gas ..and  Leachate  From Landfills:   For-
     mation,  Collection  and Treatment.   EPA/600/9-76-004., U.S.  Environmental
     Protection Agency,  Cincinnati,.Ohio, ,1976.   196  pp.

Geswein.,  A.J.   Liners  For  Land Disposal   Sites:  An Assessment.   EPA/530/
      .W--137,.  .U.S.  Environmental  Protection  Agency,  1975.   66 pp.

Geyer,  \J; A.:  - Landfill  Decomposition Gases.   An Annotated  Bibliography.  SW-
     72-T-i;"-U.S.  Environmental  Protection  Agency,  Cincinnati,  Ohio,  1972.
      36 pip,:  [NTIS:   PB213487]

Ghassemi, M. , and S.   Quinlivan.   A  Study   of   Selected • Landfills  Designed
     as  Pesticide Disposal Sites.EPA/530/SW-114c,  U.S.  Environmental Pro-
     tection  Agency,  Washington,  D.C.,  1975.   140  pp.   [NTIS:   PB250717]

Haxo,  H.E.,  Jr.,  R.S.  Haxo,  and R.M. White.   Liner  Materials  Exposed  to
     Hazardous  and Toxic Sludges:   First  Interim Report.     EPA-oOO/2-77-081,
      U.S.  Environmental Protection Agency,  Cincinnati,  Ohio,  1977.   73 pp.

Hill  D.M,,  and R. Zipp.  Ground  Water  Quality  and Solid Waste Management:
     'A Selective  .Bibliography.   State of  California,  The  Resources Agency,
      Department of Water Resources, Central District, 1974.   59 pp.

 Interstate  Electronics  Corporation.   Bibliography  on  Ocean Waste Disposal.
      Contract-   No,    68i-01-0796,. . .U.S.   Environmental    Protection   Agency,
      Was;hington, D.C. ,1973..

 Jacobs  Engineering  Co. ,   Inc.' ' Assessment  of  Industrial'::''Hazardous   Waste
      Practices-. Petroleum  Refining.   SW-129c,   U.S.  Environmental Protection
      Agency,  Washington, D.C., 19/6.   [NTIS:  PB259097]

 Kown, B.T.,  R.A.  Stenzel,  J.A.  Hepper, J.D.  Ruby,  and R.T.   Milligan. Cost
      Assessment   for  the  Emplacement of Hazardous  Materials  in  a Salt Mine.
      £f>A-600/2-77-21b,U.S.  EnvironmentalProtection  Agency,Cincinnati,
      Ohio,  1977.  128 pp.
                                       160

-------
Loehr,  R.C.,  ed.   Land  as  a  Waste  Management  Alternative:   Proceedings
     of the 1976 Cornell Agricultural Waste Management Conference.Ann Aroor
     Science Publishers Inc.,  Ann Arbor, Michigan, 1977.810 pp.

Lofy,  R.J.   "Petroleum  Refinery  Solid  Waste  Disposal  Practices."   In:
     Proceedings of a National Conference about  Hazardous Waste  Management,
     E.B.  Workman, ed.   Vector and Waste Management Section, California State
     Department of Health, Sacramento, California, 1978.   pp. 161-165.

Maritime  Administration.   Draft  Environmental  Impact  Statement - Chemical
     Waste  Incinerator Ship Project (Volume 1  of 2 - Environmental Analysis
     and  Appendices  I,  II and III).  MA-EIS-7302-76-08D,  U.S.  Department of
     Commerce, Washington, D.C., no date.  236 pp.  [NTIS:  PB246727]

Miller, H.C.   "Ocean Dumping—Prelude  and Fugue."  Journal of '^ar;tims  Law
     and Commerce. 5(1):51-75, 1973.

Ocean Disposal Study Steering Committee.  Disposal in the Marine Environment:
     An Oceanographic Assessment.  National Academy  of Sciences,  Washington,
     D.C., 1976.  85 pp.

Ottinger,   R.S., J.L. Blumenthal,  D.F.  DalPorto, G.I. Gruber, M.J.  Santy, and
     C.C.   Shih.   Recommended Methods of Reduction. Neutralization. Recovery,
     or  Disposal  of  Hazardous Waste.   16 volumes.   EPA-670-2-73-353,  U.S.
     EnvironmentalProtectionAgency,   Cincinnati,  Ohio,   1973.   [NTIS:
     PB224579-Set]

Pavoni, J.L.,  D.J.  Hagerty, and R.E. Lee.  "Environmental  Impact Evaluation
     of Hazardous  Waste Disposal  in  Land."   Water Resources Bulletin. 8(6):
     1091-1107, 1972.

Powers, P.W.  How to Dispose of Toxic Substances and Industrial Wastes. Noyes
     Data Corporation,  Park Ridge, New Jersey, 1976.508 pp.

Reed,  A.W.  Ocean  Waste  Disposal   Practices.   Noyes Data Corporation, Park
     Ridge, New Jersey, 1975.   346 pp.

Reeder, L.R.,  J.H.   Cobbs,  J.W.  Field,   Jr.,  W.D. Finley,  S.C.  vokurka, and
     B.N.   Rolfe.    Review and Assessment of Deep-Well Injection of Hazardous
     Waste.   Volume^TEPA-500/2-77-029a,U.S.EnvironmentalProtection
     Agency, Cincinnati, Ohio, 1977.  215 pp.

Rima,  D.R.,  E.B.  Chase,  and  B.M.  Myers.  Subsurface Waste Disposal by^Means
     of Wells—A Selective Annotated Bibliography.   U.S.  Geological  Survey
     Water  Supply  Paper 2020,  U.S.  Government  Printing  Office,  Washington,
     D.C., 1971.  309 pp.

Roulier, M.H.   "Attenuation of Leachate Pollutants by Soils."  In:  Management
     of  Gas  and  Leachate in Landfills. S.K. Banerji, ed. EPA-600/9-7/-026,
     U.S.EnvironmentalProtection  Agency,   Cincinnati,  Ohio,   1977.   pp.
     127-138.
                                     161

-------
Saxton, J.C. ,  and M.  Kramer.   Industrial Chemicals  Solid  Waste  Generation:
     The  Significance  of  Process  Change,  Resource Recovery, and Improved
     Disposal.EPA-670/2-74-078,U.S.   EnvironmentalProtectionAgency,
     Cincinnati, Ohio, 1974.   154pp.

Schomaker, N.B.   "Current Research on Land Disposal of Hazardous Wastes." In:
     Residual   Management  by  Land  Disposal:   Proceedings of the Hazardous
     Waste  Research  Symposium,  W.H.   Fuller,  ed.   EPA-600/9-76-015,  U.S.
     Environmental   Protection  Agency,  Cincinnati,  Ohio,  1976.    pp.  1-13.

Schultz, D.W.,  ed.   Land  Disposal  of  Hazardous Wastes:  Proceedings of the
     Fourth Annual  Research Symposium.EPA-600/9-78-016,U.S.  Environmental
     Protection Agency, Cincinnati, Ohio,  1978.

Scurlock, A.C., A.W.  Lindsey,  T.  Fields,  Jr., and  O.R.  Huber.   Incineration
     in  Hazardous   Waste  Management.    EPA/530/SW-141,  U.S.   Environmental
     Protection Agency, 1975.108 pp.

Smith, D.D.,  and R.P.  Brown.   Ocean  Disposal  of Barge-Delivered Liquid  and
     Solid  Wastes   from  U.S.  Coastal  Cities.    SW-19c,  U.S.  Environmental
     Protection Agency, 1971.  128 pp.

Snyder, H.J.,  Jr., G.B.  Rice, and J.J. Skujins.   "Disposal  of Waste Oil Re-
     refining   Residues by  Land Farming."   In:   Residual Management by Land
     Disposal:   Proceedings  of  the Hazardous Waste Researcn Symposium, W.n."
     Fuller,ecTEPA-600/9-76-015,U.S.Environmental  Protection  Agency.
     Cincinnati, Ohio, 1976.   pp.  195-205.

Stewart, J.M.,  ed.   Proceedings:  Workshop  on  Land Disposal of Wastewaters.
     Water  ResourcesResearchInstitute,Report No.  91,UNC-WRRI-73-91,
     University of North  Carolina,  Water  Resources Research Institute, 1973.
     118 pp.

Stone,  R.B.,   P.L.   Aamodt,   M.R.  Engler,   and   P.  Madden.   Evaluation  of
     Hazardous Wastes Emplacement in Mined Openings.   EPA-600/2-75-040,  U.S.
     Environmental  Protection Agency, Cincinnati, Ohio, 1975.  568 pp. [NTIS:
     PB250701]

Tinnan,  L.M.   "Methods  Presently Used  to  Treat and  Dispose of  Hazardous
     Wastes  in   Southern  California:    IV.    Methods   Used   in  Southern
     California."     In:      Proceedings   of  a  National  Conference  about
     Hazardous Waste Management, E.B. Workman,  ed.   Vector and Waste Manage-
     ment  Section,   California  State  Department  of  Health,  Sacramento,
     California, 1978.  pp. 45-46.

U.S.  Environmental   Protection  Agency.   Report  to  Congress:   Disposal  of
     Hazardous Wastes.  SW-115,  U.S.  Environmental  Protection  Agency, 1974.
     120 pp.

U.S.   Environmental   Protection  Agency.    Evaluation  of  Land  Application
     Systems.    EPA-430/9-75-001,   U.S.   Environmental   Protection  Agency,
     Washington, D.C.  , 1975.   204 pp.


                                     162

-------
U.S. Environmental Protection Agency..  Decision-Makers  Guide  in Solid Waste
     Management.   SW-500,   U.S.   Environmental   Protection  Agency,  1976a.
     158 pp.

U.S. Environmental Protection  Agency.   "Effective Hazardous Waste Managenent
     (Non-Radioactive):   A  Position Statement."   Federal  Register,  41(161):
     35050-35051, 1976b.

U.S. Environmental Protection Agency.  Ocean  Dumping  in the United Statas—
     1977.   U.S.  Environmental  Protection  Agency, Washington,  D.C.,  lS77a.

U.S. Environmental Protection  Agency.   The  Report  to Congress:  Waste Dis-
     posal  Practices and Their Effects on Ground Water.£PA-570/9-77-001,
     U.S.  Environmental  Protection Agency, Wasnington, D.C.,  lS77b.  531 op.

WAPORA,  Inc.   Compilation  of  Industrial' and  Municipal  -Injection Wells in
     the United States.   Volume  I.    EPA-520/9-74-020,   O7  Environmental
     Protection Agency, 1974.

Warner,  O.L.,  and  D.H,  Orcutt.   "Industrial  Wastewater-Injection  Veils  in
     United  States—Status  of  Use and Regulation, 1973."   In: . Preprints  of
     papers  presented  at. the  Second International Symposium  on Undargrotmd
     Waste  Management  and Artificial  Recharge.   Volume 2. J., Braunstein,  ed.
     The George Banta  Company,  Inc., Menasha, Wisconsin,  1973.  pp.  637-697.

Wastler, T.A., C.K. Offutt,  C.K.  Fitzsimmons, and P.E. Oes Rosiers.  Disposal
     of Organochlorine Wastes by Incineration"at Sea.   EPA-430/9-75-014, U.S.
     Environmental Protection Agency, Washington, O.C., 1975.   238 pp.

Wiles,   C.C.,  and  H.R.  Labowitz.   "A  Polymeric  Cementing  and Encapsulating
     Process  for  Managing  Hazardous  Waste."   In:    Residual Management by
     Land Disposal:  Proceedings  of the Hazardous Waste Research  Symposium.
     VOT  Ful 1 er,  icT    EPA-600/9-76-015,   Ol   Environmental  Protecti on
     Agency, Cincinnati, Ohio,  1976.  pp.  139-150.
                                     163

-------
                                 APPENDIX B

               ENVIRONMENTAL THREATS ASSOCIATED WITH HAZARDOUS
                         WASTE MANAGEMENT TECHNIQUES


     This appendix  discusses some  of  the more important threats  to  ian. and
the environment associated with hazardous waste management techniques.

THE NATURE OF HAZARDOUS WASTE THREATS

     This  section  characterizes  environmental  threats  posed by  hazardous
waste management  techniques,  commencing with threats that  are applicaole to
many  techniques,  followed by  the  threats that are more  specific  in  nature.
In  some  cases, models  or numerical data  are available  arc  can be  used to
provide  estimates of  the magnitude of the  impacts, while in  others,  only
descriptive scenarios are feasible.   (Ott [1976] provides mucn information c:
environmental  modeling techniques.)   Probabilities  that threats will  occ^r
are largely  discussed  later  in this appendix,  and  the  economic implications
of the threats are discussed in Appendix C.

In-pi ant Accidents and Other Events

     Within  the  manufacturing plant that generates the  waste, a  threat of
some  sort of accident  or operational  failure exists regardless of the waste
management  technique  employed, unless  all  hazardous wastes  are eliminated.
However, the  probability  of  occurrence and potential consequences may change
with  different  control   techniques.   In  an' assessment  of hazardous  waste
management  alternatives   (as  opposed,   for  example, to examining  the  total
costs  and  benefits of   using  a  particular  substance  or manufacturing  a
particular  product),  it  is  only   differential  effects  between  alternative
approaches  that  are  of interest.   Thus where a given waste (type, form, and
quantity) is generated and  shipped out  of  the plant  without treatment for
disposal or  resource  recovery, >the probability and consequences of in-plant,
waste-related  accidents  should be  constant.   However,  where the waste stream
itself  changes—arising  from  process   change or in-plant  resource recovery
operations—the probability, nature and consequence of in-plant accidents may
vary.   Likewise,  the  probability   of  in-plant accidents  may change  if the
waste is subjected to an onsite treatment process before disposal.

     The location (on  or off the manufacturing site) of final disposal need
not  affect  the  frequency  of occurrence of  in-plant accidents,  as accidents
and other events  associated  with onsite disposal can be regarded as disposal
threats  rather than  in-plant accidents.  However,  wastes are  often stored
before disposal—sometimes for  long periods while waiting for an appropriate


                                     164

-------
disposal method  to  be  developed—and accidents arising from such storage can
conveniently be included with in-plant accidents.

     Though it  is  not  difficult to generate scenarios for in-plant accidents
associated with  hazardous wastes,  statistical  data on such accidents are not
generally  available.   The  problem is that official  industrial  accident data
(e.g., data on  occupational  illness and injury at manufacturing plants IU.S.
Department of Labor, 1976]) do not usually differentiate beween those involv-
ing  wastes and those  involving  other  materials.   Data  are available  on
accidents  and  illnesses arising  in SIC 495,  Sanitary  Services,  which rould
encompass  various   disposal   techniques  (U.S.   Department  of  Labor,  1976).
However, these  data  are likely to be dominated by  accidents  in sewage works
and in regular  refuse  collection.   Buckley and Wiener  (1978)  have collected
numerous  data  from insurance  loss  records,  newspaper  reports,  and  other
sources that provide a  useful indication of the causes and size distribution
of hazardous material  spills;  but in many cases, it is not known whether"or
not the material was a  waste.

     Possible in-plant  accident scenarios that can  directly  affect  man will
largely be related  to  spills  and emission of  materials  that  can cause acute
human poisoning  by absorption  through  the skin or  inhalation  (Munn,  1975).
Poor  plant practices  could  also  result in systematic exposure  leading  to
chronic poisoning.   Chronic poisoning from industrial chemicals is considered
by some  authors to be  a far  greater threat to workers  than  acute poisoning
(Munn, 1975),  but  statistical data  are  sparse, as  in many cases  it  is dif-
ficult to  relate chronic  illnesses to industrial exposure.   (Some clear-cut
exceptions—such  as  black   lung   disease   and  asbestosis—arise  where  a
substantial  industrial  sector   is   exposed  to   a  single  disease-causing
material.)

     Environmental  damage and  indirect  threats to man could arise from fail-
ure or overflow  of  storage  tanks, operating lagoons,  containment  dikes,  and
sumps, which  could  result  in  destruction  of vegetation  and a  variety  of
surface and  groundwater pollution  problems.   Where wastes are flammable  or
highly reactive, fires  and  explosions could occur.   Methodologies for calcu-
lating the physical  effects  and probabilities  of  many of these incidents for
specific materials  are  presented by Arthur D. Little, Inc.  (1975).

Transportation Accidents

     Comparatively  good data  are  available  on the  nature and frequency  of
transportation  accidents.    Many  of these  specifically  address  hazardous
materials  (Booz,  Allen, and  Hamilton,  1970;  National  Academy of Sciences,
1976; Arthur D.  Little,  Inc., 1974; Jones et  a].,  1973).   Typical hazardous
material   spill  frequency  and size  data  are  given  in Table B-l.   Unfortu-
nately,  the   accident  probability  Statistics   show considerable  diversity,
depending on  source and coverage.

     Some  authors  emphasize the  likely  outcome of  transportation accidents
(Arthur D.  Little,  Inc.,  1975;  Jones  et  al.,  1973;  Angell  and Kalelkar,
1974), sometimes indirectly via studies  of the control of hazardous  material
spills (Dawson,  Shuckrow,  and  Swift,  1970;  Anon.,  1974).   The emphasis  in


                                     165

-------
             TABLE B-l.   SELECTED TRANSPORT ACCIDENT STATISTICS
                                        Events per billion kilometers
Source and type of cargo         Truck

Arthur D.  Little. Inc.a
  Hazardous chemicals
  Flammable liquids
  Corrosive liquids
  Tank trucks (involving              .
    cargo loss)                    17

Booz, Allen, and Hamilton6
  Hazardous materials
    (projections to 1980)       1,119
               Rail
               144^
               186C
                20
         Water (barge)


             261b
              0.68/106
               tonne-km
Jones et al.
  Hazardous materials
  Autos, all accidents

Arthur D. Little, Inc.^
  l,057g
2,140-12,500
     22'
8-ir
 12'
a Arthur D. Little, Inc., 1975:29,30.

  Typical  large  capacities  are  up  to  1,816  tonnes  (2,000   short  tons).
  Average spill size is approximately 48,450 liters.

c For 1965-70.
   For  1968-72.  Typical  tank capacities range  from  22,700  to  37,850 liters
  (6,000  to  10,000  gal).  The  average spill  size is  about  11,350 liters.

6 Booz, Allen, and Hamilton, 1970:15.

f Jones et al., 1973:99-102.
9 Based on FHA data for large carriers.
  For 1960-68; data may be conservative.

1 Based  on a variety of  sources.   (It is  claimed  that  vehicle  size does not
  affect accident frequency.)

j Arthur D. Little, Inc., 1973, Vol.11:111.
I/
  For tank trucks.

  For railroad cars, based on data from Association of American  Railroads and
  Federal  Railroad Administration.
                                     166

-------
most of these  studies  is on damage to human life and to property, 'especially
through the mechanisms of fire and explosion.

     Very  few  authors  attempt  to  analyze   the  impact  of  transportation
accidents on flora or fauna (nonhuman).   In a paper that addresses choices in
risk situations via the determination of a decisionmaker's utility functions,
Kalelkar,  Partridge, and  Brooks  (1974)  use a  13-point  scale  to indicate the
severity of  land-based  environmental  impacts  that are expected to arise from
typical accidents during the transportation of specified hazardous materials.
This  scale,  which  is   used  in  conjunction with  the  area  affected,  is  as
follows:

      1.   No effect;

      2.   Residual surface accumulation of harmless material  such as sjgar or
          grain;

      3.   Aesthetic pollution (odor-vapors);

      4.   Residual  surface  accumulation  of removable  material such  as  oil
          (requires more costly measures of abatement);

      5.   Persistent  leaf  damage  (spotting,  • discoloration),  but  foliage
          remains edible for wildlife;

      6.   Persistent leaf damage (loss of foliage), but new growth in follow-
          ing year;

      7.   Foliage remains poisonous to animals  (indirect cause of some deaths
          upon ingestion);

      8.   Animals  become more  susceptible to predators  because o":  direct
          exposure to chemicals and a resulting physical debilitation;

      9.   Death to most smaller animals (consumers);

     10.   Short term (one  season)  loss  of producers  (fol-iage) with nigration
          of specific  consumers  (those  who eat the  specific  producer),  but
          eventual reforestation;

     11.   Death to producer (vegetation) and migration of consumer (animals);

     12.   Death to consumers and producers;

     13.   Sterilization   of   total   environment  (decomposers,   consumers,
          producers) with  no potential  for reforestation or  immigration  of
          species.  (Kalelkar, Partridge-, -and Brooks, 1974:340)

Spills Into Water

     Dawson and co-workers have  addressed the  problem of spills of hazardous
materials (from  both transportation  accidents and  stationary  sources) that
                                     167

-------
reach surface watercourses (Dawson, Stradley, and Shuckrow, 1975, Vols.II and
IV; Dawson and  Stradley,  1975).  Their approach determines  the  proportion of
annual production of  each material of interest that  is  spilled, and follows
this  through  to provide  an  estimate of  spills that  result  in "substantial
damage"  to  aquatic  systems.   The  estimates  of damage  are based  on  simple
modeling of  the dilution  and  transport processes,  and  they employ toxico-
logical data  for  typical  fish  receptors (Dawson and Stradley,  1975).  Figure
B-l illustrates  this  process  for  sulfuric acid; however,  the  proportion of
production  spilled   (0.0025  percent)  is  believed  to  be typical  of  many
hazardous materials  (Dawson  and Stradley,  1975), and  the  distribution among
the four categories of  spills  and the proportions of these spills that reach
the water  are in fact  based on aggregate data (Dawson  and Stradley,  1375).
However,  the  proportion   of  the  material  spilled  that  causes  substantial
damage  is  expected  to depend  on  the  material.   Note that  the  data shown in
Figure  B-l  appear to mask the  importance  of stationary  sources  in causing
aquatic damage.   Tables B-2,  B-3,  and B-4  present fish  kill  data from the
past  decade.    Of kills  attributed  to  industrial   releases   and  transport
accidents,  only 8 percent of  the  game  fish  and 28 percent of  all  fisn were
killed as a result of transport accidents (Table B-4).

Treatment Process Failures

     One class  of threat  arises  from  the  possibility  that  a   */aste  may be
subjected  to  some treatment process  (e.g.,  detoxification,  neutralization,
immobilization),  but  that the  treatment  process does not work as  iitsnded.
The result  could  be  a  waste that was more  hazardous  than anticipated.   The
variability  in  composition of  waste  streams makes this a real possibility,
although  it  does  not  appear  to  be  a  problem  that  has  attracted  much
attention.

     Some  treatment  processes   are  comparatively straightforward and  easily
checked.   For  example,   simple  acid/base  neutralization  is   controlled by
checking pH,  so only a measurement  error  is likely to lead to an environ-
mental  threat.   On  the  other hand,  biological  processes are  sensitive to
waste  composition (Battelle Memorial  Institute,  Pacific  Northwest Labora-
tories,  1974) and  where   used,  they might  be  rendered  inoperative  by the
presence  of  undetected  materials  in  the  waste  stream.   Worse,  inorganic
mercury, which  is comparatively  nontoxic  in monovalent form  (Venugopal and
Luckey, 1975),  may  be converted to the highly  toxic  methyl mercury (organic
form)  by  anae.robic  bacterial  action  (U.S.   Environmental  Protection Agency,
1974);  hence  it  is  not  inconceivable that under  adverse conditions, a waste
containing  traces of mercury  could  become  more hazardous  after treatment.

     A more typical  treatment threat scenario than that discussed above would
be  the failure of  an incinerator  to  achieve the dwell time  or temperature
necessary to  fully degrade the  waste.   This might arise because the incinera-
tion parameters necessary to degrade the waste of interest are not adequately
known.  The  scrubber  system  could also fail.   Though  it is possible to make
measurements  to detect  virtually any of the  failures  that can  be envisaged,
in  practice,  such measurements  may not be made, or  defective  equipment may
not be shut down.
                                     168

-------




1
Stationary
Sources
43%
1
.001075%
1
Sprlls Reaching
Water
59%
1
.000634%
1





















L



i
Trucking
19%
.000475%
I
Spills Reaching
Water
42%
I
.000200%
I

I
Spills into'
Lakes 18%
1
.000204%
1
Substantial .
Damage
100%
"I
.000204%
I
PRODUCTION
1
Spills
1
_ .0025% _





1
.001136%
1





J



1
Rail
26%
1
.000650%
1
Spills Reaching
Water
15%
1
.000098%
1

1
Spills into
Rivers 82%
1
.000932%
1
Substantial
Damage
. 58%
1
.000540%
J




I
Barging
12%
I
.000300%
I
Spills Reaching
Water
68%
I
.000204%
|






                                     .000744%
                                                                Total
            Figure B-l.  Process  for  estimating damages caused by
                         spills (sulfuric  acid) into water.
Source:   Dawspn  and  Stradley,  1975:25.   Reproduced by permission  of the
         American Institute of Chemical  Engineers.
                                      169

-------
         TABLE B-2.   SUMMARY OF POLLUTION-CAUSED FISH KILLS FOR THE UNITED STATES. 1960-1975*

I860-
NunM ol Sum
(•ipoiicfeng
NuntUf of fflpolk
Rtp»ll mtail IIM IWIiMi
oiitnUM
low ituoiwd iiuiil*» ol
blHUM
AMKgtlutOlkll'
Ldrgtu k« wpulM
Nun** ol lipnud
noOMUkloi 4«ch
pouoiltuo oinriHVi
Inouuul
MunCipH*
Tiantporlaiian
dm*
UIIU1UWII
lOUl ItpOflt


NunM ol inaun ami bsn
KrilKl Dy »u« gfM^MIg
1 OOOUUOoi mum
100 OUO 10 1000 000
10 000 10 100000
1000 10 10 000
oniouo
No nil ripurud klf ACNjafil
Avtitgo aurwan ol KM tfl
Itayl .







38
289

151

6105 000
2925
5 OUO 000










No
ft
von*
3
15
64
68
138




79
103
24
0
a
so
289
No
bin
IIM
K»»|
50
053
031
Oil
002


295
1901







45
413

2G5

I49IOOOU
6535
5387000










NO

"""
4
5
45
107
104
148




74
169
52
0
58
60
413
No
bin

bofu)
065
IDS
034
003


464
1962

37
421

246

44U01 000
5710
J 180 000


51
209
33
1
47
80
421
No
NO tin
"l iro

I'! 69
38 101
89 OX
108 003
175

259
19bJ

38
442

304

b9J/OUI
7775
4000000


64
199
60
17
27
55
442
No
No blh
IB (rfU
lOHh feunkj
I 20
12 $68
54 182
134 041
103 003
138

318
ltk.4

40
590

4/0

n 914 an
5490
/ 1187 000


131
193
120
26
17
103
590
No
No bhll
ie (ml
UUflh ian»|
.1 U
59 165
167 049
424 007
120

444
IWd

44
625

540

12 140000
4310
J 000 000


114
244
125
27
23
92
625
No
No bin
ie |nu
pate wn»l
,* 4.4
63 142
402 059
235 007
105

257
I9ob

4b
532

45J

9614000
5620


88
195
87
27
30
97
534
Nu
No bin
i> |nu
K»I> •"-'
23 546
58 153
185 055
185 005
79

2/1
I9b/

4U
454

364

11491000
6460
b 549 000


U/

91
21
35
11
454
Nil
NO bgi
10 |nu
U0i» bum
1 65
7 2Cu
49 IS«
143 046
164 005
90

1J4
.*»

44
54.'

469

15UI5UUO
6015
4049009


77
1/7
122
39

104

No
Nu L&h
II! (IMlt
Kin. '.on,,
J b 1
JO 744
64 1 79
153 048
2V) 006
73

49D
1969

4;>
594

494

41 IbbOOO
5860
45547000


ll/
19!)
84
32
tl
12!)
5U4
Nu
No bui
l« IIM
,0,1 kuni
4 151
9 3 15
01 206
165 052
23} 006
104

Jll
ID/0

45
635

56J

42 490 OUO
6412
3240000


108
213
120
28
28
138
6J5
No
NU bin
ie (nu
H*H iulll
5114
26 744
91 273
198 062
443 007
72

J45
19/1

40
060

75>J

7J 6/0 OUO
6154
5500000


114
231

52
64

060
No
No Mi
III (Mill
|»l> Ml.,)
2b bJO
26 637
124 333
2U Olio
315 010
101

JJ5
19/2

50
760

637

I/ 71/000
46J9
2942000


113
189
167
56
72
163
760
No
NO tm

IXMI» fcona!
47 "Si
81 260
416 062
3b7 009
63

340
197J

50
749

70J

37114000
S527
10 0110 000


161
196
146
65
56
125
74U
NO
NO hkh
it (mil
poib bonil
b 2953
19 465
88 2 74
251 061
339 009
46

274
19/4

50
721

we

19054 000
6532
47112000


145
168
169
40
74
I2S
721
No
No bin
n (IM
pals lent)
b 107 b
20 7 1
I1O 36
167 06
325 01
73

J58
1975

50
624

543

Ib III 210
3679
10000000


118
122
90
47
78
169
624
No
No bill
i, (ml
IJUH! KMH)
3 1200
9 187
86 1 62
173 057
293 007
II

218



























• Ckxiwl MM fcLiana i«pun» ol 100 000 blk
of m» n bung U»«»UIUIIM

' ffccabng tytum « wka to hiu w mount ol

1960
1 Minup* apMjiMnt nckijf tutuc puwti gwwdng IUIKVII
* Source:   U.S.  Environmental Protection Agency, 1977a:4,5.

-------
                                  TABLE  B-3.   EXTENT OF  FISH  KILL  DAMAGE*
      Item      1960     1961     1962     1963     1964     1965     1966     1967     1968    1969    1970

 RIVER
  Number of
   reports      189
  Miles of
   stream     1,204
 LAKES AND
 RESERVOIRS
  Number of
   reports       25
  Acres
   affected   1,407
 •240     259     271     339     292     ?51     219     264     356     487

1,686   1,448   2,203   1,440   1,300     989   1,039   1,565   2,358   1,865



   50      25      49      57      38      46      33      37      98     111

5,907   2,581   5,644 '12,637   4,630  21,504   1,996   2,400   6,068  33,168
 * Source: . Dawson, Stradley, and Shuckrow,  1975,  Vol.IV:92.   (Data are only for reports where extent
   of damage was  included.)

                            TABLE  B-4.   ANALYSIS  OF 1970  FISH  KILL  DATA*
           Source
      Game fish killed
%      • Non-game fish killed
.Agricultural  pesticides.
   and fertilizers                264,391

'Industrial                      3,250,252

 Transportation                   284,782

 Other                            177.482

   Total                         3,976,907
                              7

                             82

                              7
                           100%
              1,149,472

              6; 569,241

              3,616,348

                307.75'J

             11,642,820
  10

  56

  31

   3
100%
   Source:   Dawson,  Stradley,  and Shuckrow,  1975,  Vol.IV:93.

-------
 Threats  from Land-based Disposal

      Because waste disposal  to land  (both  legitimate  and covert) is a  very
 diverse  and  common  disposal  approach,  there  are more  data  on  the  threats  thai
 can  arise from this  approach  than from other techniques.   EPA has collected
 and  analyzed data on  damage  incidents that  have  arisen  from  land disposal  of
 industrial  wastes  (Lazar,  1975/76; Lazar,  1975;  Lazar, Testani, and Giles,
 1976).   Detailed  descriptions of some  of  these  incidents  have  also   been
 published (Ghassemi,   1976;  Shuster,  1976a,  1976b,  1976c;  U.S. Environmental
 Protection Agency,  1977b;  U.S.  Environmental  Protection  Agency, 1975a, 1975o,
 1976;  Carter et al.,  1975).

      Lazar  (1975/76)   has  found six major routes of environmental transport
 through  which  the  land  disposal  (including  lagooning,  land  application,
 landfill ing,  and  dumping)  of  hazardous waste  can  result  in   damage,  as
 follows:

       1.  Groundwater contamination via leachate;

       2.  Surface water contamination  via runoff;

       3.  Air pollution via  open  burning, evaporation,  sub1-'nation ana  wind
          erosion;

       4.  Poisoning through  direct contact;

       5.  Poisoning via the  food chain;

       6.  Fire  and  explosion.

      Table   B-5 provides  a  matrix of  transport mechanism  versus disposal
 method  for  421 damage cases  arising from  the  land  disposal  of hazardous
 wastes.   It  will  be   noted  that  groundwater and surface water contamination
 are  the  two most prevalent  transport mechanisms  and that most  incidents  of
 direct contact  poisoning arise from the category  "other  land  disposal,"  i.e.,
 "haphazard   disposal  on vacant properties,  on farm  land,  spray  irrigation,
 etc." (Lazar, Testani, and Giles,  1976).   the mechanism  of  "poisoning  through
 the   food  chain"  is   not  represented   in  Table  B-5,  largely  because  of  the
 difficulty  of tracing  some  such  incidents back to  a  specific  source and  also
 because  one  of the  other  identified   transport mechanisms will  normally  be
 involved, in the  contamination of the  lower  order(s)  in the  food chain.
 However, some  such  incidents  and potential incidents  have  been  identified
-(Lazar,  -19.75/76).

 Mechanisms  of Water Contamination—
      The mechanisms by  which  surface  waters and  groundwater can become  con-
 taminated from land  disposal  are  of great interest  in  view  of the frequency
 of such  pollution  events.

      Surface waters—As noted earlier  in this appendix,  Dawson  and  co-workers
 have developed simple  models  for  predicting the  impact of spills  that  reach
 surface   waters.    In   addition  to  spills  from transportation  and  in-plant


                                      172

-------
                  TABLE B-5.  MECHANISMS INVOLVED IN DAMAGE INCIDENTS BY DISPOSAL METHOD'
to

Damage mechanism
Number of cases
Groundwater,
259 cases
Surface water,
170 cases
Air,
17 cases
Fires, explosions,
14 cases
Direct contact poisoning,
52 cases
Wells affected,0
140 cases

Surface
Impoundments
89 cases
57
42
3
0
1
32
Disposal
Landfills,
Dumps
99 cases
64
49
5
11
6
28
method and number of cases
Other land Storage
disposal of wastes
203 cases 15 cases
117 10
71 0
9 0
3 0
40 5
74 4

Smel tings,
slag, mine
tailings
15 cases
11
8
0
0
0
2

        Source:   Lazar,  Testani,  and Giles, 1976:4.  The data refer to 421 cases associated with land dis-
        posal  of  industrial  wastes.   The numbers  in  the  matrix  add up to more  than 421 because several
        damage incidents involved more than one damage mechanism.

        Haphazard disposal on vacant properties, on farm land, spray irrigation,  etc.

      c Not included as a damage mechanism.

-------
accidents,  surface  waters  can  become  contaminated  by  the  overflowing  of
lagoons as a  result  of exceptional climatic conditions  (e.g.,  flash  floods,
100-year floods, etc.)  and  the failure of dams, etc.,  due  to poor design or
earthquake damage.

     It  is  a  feature  of  all  surface water  pollution mechanisms that  the
contamination occurs rapidly,  and that in most cases,  dilution and dispersion
mechanisms limit  the extent  of  the damage.   This  is borne out  by the  fish
kill data presented  in  Tables 8-2 and B-3, which  show  that the average  kill
duration  is  about  three days, and that  on  the average, only a  few miles of
river  are  affected.   Lakes which  have comparatively small inflows and  out-
flows  constitute  an exception  to  these  remarks;  the duration  of  polluting
incidents  may  be  long,  as  the  dilution mechanisms  are  slow  (Dawson  and
Stradley, 1975).   Where pollutants reach estuaries,  their  physical behavior
is  complex,   and  their  fate  is  highly variable   (Dawson,  Stradley,   and
Shuckrow, 1975, Vol.11).

     Groundwater—Figure B-2  illustrates the  processes  by  which groundwater
can  become  contaminated.   Percolation  or  leakage of   leachates  (including
direct  seepage of  lagoon  contents)  into water-table  (unconfined)  aquifers
constitutes the major  threat.   In contrast to surface  water  pollution,  this
process may  take years  or decades to  become  apparent.   First,  true leacning
(of  soluble  and  suspended materials  derived from  the  solid  portion of  a
waste)  is  a  slow  and often  (depending  on climatic conditions) intermittent
process.  Second, groundwater moves comparatively slowly, flow rates for  most
aquifers range  from  meters  per day to meters per year (a few feet per day to
a few  feet per year [U.S.  Environmental  Protection Agency,  1977b]),  and the
average  residence  time of  groundwater in an  aquifer  is of the  order  of  200
years  (U.S. Environmental Protection Agency,  1977b).

     These long time-constants are attested to both by theoretical (modeling)
studies  (Fenn,  Hanley,  and OeGeare,  1975; Konikow,  1977; Schultz, 1978;  Elzy
et  al., 1974  [also  see  Elzy  and Lindstrom,  1976])  and  by  numerous  case
studies  of  incidents  involving  leachates   (U.S.  Environmental  Protection
Agency,  1977b, 1976,  1975a,  1975b; Shuster,  1976a,  1976b,  1976c;  Konikow,
1977;  Walker,  1973).   These studies also show that leachate contamination of
groundwater is usually  fairly restricted  in its area!  extent at any one time*
but  that the  leachate  frequently  forms  a plume  or slug  that travels  in  the
direction  of   groundwater   flow,   often  with  little  mixing.   Naturally,
individual details  vary considerably,  depending  on the  quantity  and  type of
wastes  involved,  the climatic conditions at the disposal site,  and the char-
acteristics  of the aquifer, including its ability to attenuate the waste by
reacting with  it.
*  One  of  the  larger recorded  areas of  contamination—approximately  3,000
  hectares  of  severe  pollution  and 8,000  hectares  over  which traces  of
  pollution  could  be  detected—arose from  discharge  of  several  years  of
  pesticide  wastes   to  unlined   evaporation  ponds at  the  Rocky  Mountain
  Arsenal,  Colorado  (U.S.  Environmental   Protection  Agency,  1975b).    The
  wastes  contained  a  very high  chloride concentration  which  enabled  the
  extent of the contamination to be readily monitored (Konikow, 1977).

                                      174

-------
PRECIPITATION
                           Y      PUMPING WELL
                              , DUMP
                        0* REFUSE PILE
                          DISCHARGE
                          OR
                          INJECTION
                                       rATi* ruiLt  fl


                                      r
-------
     Table B-6 provides  detailed data on 60 cases  of  groundwater contamina-
tion from  landfills  and  dumps in the  northeastern  United  States.   Nota that
the mean annual  potential  evapotranspiration in this area  is  less t.^an near
annual  precipitation  (U.S.  Environmental  Protection Agency, 1977b), a condi-
tion that encourages leaching from landfills.

     Deep well injection—The primary  threat from  deep well injection ''s tne
contamination  of  a   usable  aquifer;  but  contamination  of  other  valuable
resources  and adverse chemical  reactions  are  possible (Warner  and  Crcutt,
1973).   In addition,  there are two known instances where deep well injection
has stimulated earthquakes (Raleigh, 1972).

     There   are    several   mechanisms   by   which   aquifers   could   oecome
contaminated:

      1.   Lateral travel of  injected waste  to  a region  of freshwater in tne
          same aquifer;

      2.   Escape of  waste  into a freshwater aquifer through  some  failure of
          the injection  well  casing or through some nearby deep de~'l  ciiat is
          not adequately cased or plugged;

      3.   Vertical  escape  of the  waste  from  the injection  zone  through
          confining  beds  that   are  inadequate  because  of  high  primary
          permeability,  solution   channels,  joints,    faults,   or  induced
          fractures;

      4.   Indirect  contamination whereby  injection of  the waste displaces
          saline water into  a freshwater aquifer (Warner  and  Orcutt, 1973).

     Note that deep well  injection will inevitably cause some modification of
the  local  groundwater  system.    If  this  technique   is   to  be  used,  the
management objective  should  be  to  insure  that whatever  modification  takes
place does not have  unanticipated effects.   Again,  the question of maintain-
ing options  arises.   Saline  water (usually defined as  water containing 1,000
mg/1  or  more of  dissolved  solids)  has traditionally been  regarded  as  a
nuisance and is  only  used where no other  source is available.   However, as
Nace (1973)  has  pointed out,  advances  in  desalting technology  are changing
saline  water  into  an  extensive  resource, and  potential alternative  uses
should be considered before wastes are injected into a  saline aquifer.

     The  behavior of  wastes  injected into deep  saline  aquifers has  been
modelled by many authors.  Recent publications  on this  topic include a review
by  Reeder  et al.  (1977)  and a model developed  for the  U.S. Geological Survey
(INTERCOM? Research Development and Engineering, Inc.,  1976).

Environmental Impacts  from Ocean Dumping

     Unlike  the  threat  posed  by a possible accident,  it  can  be  stated with
certainty that ocean dumping will have some effect on the marine environment.
There will  inev-itably be  some local contamination because of the mixing of
                                     176

-------
          TABLE B-6.  SUMMARY OF DATA ON 42 MUNICIPAL AND 18
                      INDUSTRIAL LANDFILL CONTAMINATION CASES*
                Findings
   Type of landfill

Municipal     Industrial
Assessment of principal damage
  Contamination of aquifer only                      9
  Water supply well(s) affected                     16
  Contamination of surface water                    17

Principal aquifer affected
  Unconsolidated deposits                           33
  Sedimentary rocks                        •          7
  Crystalline rocks                                  2

Type of pollutant observed
  General contamination                             37
  Toxic substances                                   5

Observed distance traveled by pollutant
  Less than 30 m (100 ft)                            6
  30 to 305 m (100 to 1,000 ft)          .            8
  More than 305 m (1,000 ft)          -              11
  Unknown or unreported                             17

Maximum observed depth penetrated by pollutant
  Less than 9 m (30 ft)                             11
  9 to 30 m (30 to 100 ft)                          11
  More than 30 m (100 ft)                            5
  Unknown or unreported                             15

Action taken regarding source of contamination
  Landfill abandoned                                 5
  Landfill removed             "         '             1
  Containment or treatment of leachate              10
  No known action             •                 •     26

Action taken regarding groundwater resource
  Water supply well(s) abandoned                 •    4
  Groundwater monitoring program established        12
  No known action                                   26

Litigation
  Litigation involved                                8
  No known action taken                             34
                   8
                   9
                   1
                  11
                   3
                   4
                   4
                  14
                   0
                   4
                   2
                  12
                   3
                   3
                   2
                  10
                   6
                   2
                   2
                   8
                   5
                   2
                  11
                   5
                  13
* Source:  Miller, DeLuca, and Tessier, 1974.
                                     177

-------
the waste and  the  surrounding seawater (although if  the  wastes are contain-
erized, the  most serious  contamination  may be  delayed until  corrosion  has
breached  the containers);  hence the  critical  question  becomes:  Will  this
effect be localized  or  dispersed,  and what impact will it have on the marine
ecosystem?

     The dispersion  and  transport  processes that determine the physical  fate
of a  dumped waste are  comparatively well  understood, and  given sufficient
input  data,  they  are in  many  situations  capable  of being  modelled (Ocean
Disposal  Study Steering  Committee,  1976).  However,  even if  the  immediate
physical  fate  of  the dumped  material  can be  predicted,  this  is  only  the
beginning of the waste's  effect on the ecosystem.   Complex  chemical inter-
actions can  occur  between  the  waste and the  seawater (Ocean Disposal  Study
Steering Committee,  1976),  and  most authorities agree that our understanding
of the processes that determine the biological  impact  is very limited.

     In  a  very  broad  study on  ocean pollution,  the  National  Academy  of
Sciences recently stated that:

          The  greatest   uncertainty  in assessing the  impact  of ocean
     pollutants is  the scarcity of data on their chronic toxicity. .  . .

          Even  if  good  experimental data  were  available  on sublethal
     toxic effects  of individual chemicals on representative species, it
     would   be   difficult  to  deduce  effects  on  marine  ecosystems.
     (National Academy of Sciences, 1975:6,7)

     Nor  is definitive  information  likely to  be  forthcoming  in   the  near
future.   Another National  Academy  of  Sciences  study specifically directed
towards waste disposal in the marine environment stated that:

          Once the assimilative capacity of a site and the limits on the
     rate of input  of  wastes are  known,  limiting conditions  on waste
     composition, waste  treatment,  and waste  dispersion  or containment
     can  be derived   and  a variety of designs  to meet those conditions
     can  be proposed.  The  necessary information  for this  is  not  now
     available,  nor  is   it  likely  to be  available in the near  future.
     (Ocean  Disposal  Study Steering Committee,  1976:30)

     In the  absence  of  a detailed understanding of the interaction of wastes
and  the marine  ecosystem,  a case-study  approach could be  used to  provide a
very  general impression  of  the possible impacts.  Though  most of  the ocean
disposal  studies cited  in  Appendix  A  include  descriptions  of  some  of  the
observed environmental impacts of ocean disposal, very few specific data have
been collected.  Smith  and Brown summarize the  results  of nine  case studies
on industrial wastes  (Smith  and Brown, 1971), and the  two National Academy of
Sciences  studies catalog environmental  impacts  that  have  been attributed to
ocean  disposal  (Ocean Disposal  Study Steering Committee, 1976) and the known
effects of some chemicals on ocean life (National Academy of Sciences, 1975).
                                     178

-------
PROBABILITIES THAT THREATS WILL OCCUR

General Discussion

     The  previous  section has described categories of  threat scenarios that
may  arise in hazardous  waste management.   For  transportation accidents and
for spills into water, some data that relate to the probability of occurrence
of such threats were included.  However, the emphasis has been on identifying
the  character  of the  threat, and this  step  alone is  of  value in analyzing
hazardous waste  problems,  as  social  behavior may be influenced by the lature
of the  threat without  much  reference to  its  probability.   Nevertheless,  it
would be useful to know more about the probabilities that threats will accur,
as  these  data  would  be  an  integral  part of  any complete  cost-benefit  or
risk-benefit analysis.

     The  initiating  event  (see  Figure 4 in the main text) is usually the key
to determining  the  probability  that  a threat will occur, although the threat
mechanisms determine the outcomes,  and hence affect the  probability of any
specific outcome.   Some  initiating events, i.e., those associated with ocean
clumping and  deliberate  discharge  to  sewer or waterway, are not probabilistic
in nature, while others such as sabotage and the effects of inadequate design
or poor practice  are  inappropriate  for  a  probabilistic approach.   3n the
other  hand,   geologic  and  climatic  events  are  amenable to  probabilistic
analysis,  as are various forms of accidents.

     The  probabilities  of most geologic and climatic  initiating events will
be site-specific, and  analysis  of the more common climatic events is already
a routine operation.  For example, landfill sites in California are evaluated
with  respect  to the  100-year  flood  level   and the  precipitation  from  a
10-year,  24-hour  storm  (California  State  Water  Resources   Control  Board,
1976).  Note that virtually every initiating event is capable of occurring at
various magnitudes.   Taken  over a  long  time span,  the  larger  or stronger
events  almost  invariably occur  less  frequently than smaller  or weaker ones
(e.g.,  a  500-year flood  level  will  be  higher than a  100-year flood level,
strong  earthquakes  are less  frequent  than weak  ones).   Hence some judgment
will  be necessary  as to the critical magnitude of an initiating event, based
on the  simplifying  assumption  that  initiating events  that are smallar than
the critical  magnitude will not lead to any adverse outcome.

     There has  been some  research on predicting  the  probabilities  and out-
comes of  less  likely initiating events.  Much of  this  work  has stemmed from
analysis  of  the  safety  of  various  aspects  of the nuclear  fuel  cycle
(Rasmussen,  1975), and  in particular from the disposal  of high-level radio-
active  wastes  (Schneider  and  Platt,  1974,   Vol.1;  Energy  Research  and
Development Administration, 1977; Claiborne and Gera,  1974).   While this work
has  generally  attempted to  follow  through  the possible  consequences  of
various initiating events  by  means of event trees, it  was largely concerned
with the  danger  from radioactivity;  so the data are not generally useful for
nonradioactive  hazardous  wastes.   Deonigi  (1974)  provides a  useful  summary
(with the  emphasis  on  safety aspects) of  the general  methodology  that has
been  developed  for  evaluating  radioactive waste  management  concepts,  and  a
related article by  Burkholder et  al.  (1976)  concentrates  on the  fate  of
radioactive  wastes  that are  leached  from geologic storage  (mine disposal).

                                     179

-------
     The  data  on  the  probabilities  of various  uncommon  events  and  their
outcomes  that  have  been  collected  for evaluating  the  safety  of  nuclear-
related  activities  are  available  in  two  forms:    (1)  Estimates  of  the
probabilities  of  initiating events  of sufficient magnitude  to cause  waste
release from a specified  form  of disposal   (usually mine  disposal)  and some-
times at  a specified  location,  and (2) trade-off relationships between  the
frequency  of  occurrence  of these  events   and  the  damages caused  (property
damage  dollars  and  human  fatalities)  for the  entire United  States.   The
second form, which  is  available  for both natural and  man-caused  events (see
Figures B-3, B-4,  and  B-5), cannot be used directly  for  assessing  the proo-
abilities of critical  magnitude  initiating  events,  but it is  a  useful  way to
assess the relative importance  of different events.

Uncommon Natural  Initiating Events

     Many  of  the  threats  to the more secure techniques  for  hazardous  wasta
disposal will be  initiated  by  unlikely climatic and geologic events.   Since
the  time  scale of  interest may  extend  to perpetual  care,  not only  sudden
initiating  events  (such  as earthquakes),  but also  gradual  ones  (sucn  as
erosion) could be  of potential  interest.   Other than meteorite  impact,  wnich
is regarded as a random process (Claiborne  and Gera,  1974), the  probabilities
of  all   natural  initiating  events will  be  site-dependent  to ^  lessar  or
greater extent, varying either  with  the general  location  (e.g., for  earth-
quake) or  with the  specific details of the site  (e.g.,  for flood).   Various
events are discussed below.

     Earthquake—Seismic  activity  varies   considerably  within   the  United
States.    Schneider and  Platt  (1974)  use   a  threshold level  of  IX on  the
Modified  Mercalli   Intensity  Scale,  which  would  result  in  the  following
effects:

          Damage  considerable  in specially  designed  structures;  well
     designed frame structures  thrown out of plumb;  great  in substantial
     buildings with partial collapse.  Buildings shifted off foundations.
     Ground cracked conspicuously.  Underground pipes  broken.   (Coffman
     and  von  Hake,  1972:4-7;  quoted  in  Schneider  and   Platt,  1974).

     Schneider and  Platt  (1974)  calculate  that the probability of  an  earth-
quake of  at least this intensity striking  a  typical  east coast site  in  the
United  States  is  about   2x10 5  per  year,  while  for  a random  point  in
California, the probability would be  some  2xlO"2 per  year.   Clearly,  better
estimates  of probability  could  be derived  for specific sites.   The  risk from
earthquakes  would   probably  most  affect   storage   structures,   treatment
facilities,  and  lagoons.    The   impact  on wastes   in the solid  phase  (in
landfills  or mines) would probably occur via changed water levels  and  paths.

     Faulting and cracking—Faulting  is considered  to  be the  most  likely
aspect of tectonic activity (other than earthquake)  that could interfere with
waste disposal (Schneider  and  Platt,  1974).  Faulting or  cracking would most
likely affect water paths and would be of particular importance  for  deep well
injection,  as  it  might cause  saline  aquifers to become  connected  to  fresh-
water aquifers.  Fault activities vary considerably within the United States,


                                     180

-------
                                  10°
                                          10
                          10°
                                                                10
10'
                                          Property Damage(Dollars)
            Notes:   1- Property damage due to auto accidents is not included because data
                      are not available for low probability events. Auto accidents cause
                      about $15 billion damage each year.

                    2. Approximate uncertainties for nuclear events are estimated to be
                      represented by factors of 1/5 and 2 on consequence magnitudes and
                      by factors of 1/5 and 5 on probabilities.

                    3. For natural and man caused occurrences the uncertainty in prob-
                      ability of largest recorded consequence magnitude is estimated to
                      be represented by factors of 1/20 and 5. Smaller magnitudes have
                      less uncertainty.
     Figure  B-3.


Source:    Rasmussen,  1975:A3.
Frequency  of  property-  damage  resulting  from  natural   and
man-caused  events.
                                                181

-------
          1/100	
         1/1000	
     1/10.000.000*
                        -I—I—+—
                        *        I.     i
                           ^\"«J
        1/10.000-j	r-	,-V^



       i /100.000 -\	>v	\MTS-i-—
                      ^  -mn Nuclear
      1/1.000.000	
100
                          1000  10.000 100.000 1.000.000
                          Fatiilitius
                                                        1/10 000.000
                                                                 10
                    100   1000  10000  100.000  1.000.000

                          Fatalities
       Figure B-4.  Frequency of  fatalities  due
                    to man-caused events.

Note:  Fatalities due to auto accidents  are
  not shown because data are not  available.
  Auto accidents cause about 50,000  fatali-
  ties per year.
       Figure B-5.  Frequency of  fatalities  due
                    to natural events.
Source:  Rasmussen, 1975:A2.
Source:   Rasiiiussen, 1975.

-------
but as  an  example,  one estimate places the probability that a new fault will
intersect  an  8 km2  waste  depository  in  the Delaware Basin,  New Mexico,  as
4x10 li  per year (Claiborne  and Gera,  1974).   On the  other hand,  a  study
relating to Aiken,  South  Carolina,  states that there  is  insufficient  know-
ledge of the  mechanisms  to estimate the probability  that  a small crack will
traverse a stratum  (Energy  Research  and  Development  Administration, 1977).

     Volcanic activity--Several  authors  consider   that  the  probability  of
volcanic activity  (in an  area not currently  active) is  extremely low,  but
none care  to estimate that  probability (Schneider  and Platt,  1974; Energy
Research and  Development  Administration,  1977; Claiborne and  Gera, 1974).

     Erosion—The average  rate  of  erosion  for the entire  United Sta<:as is a
few centimeters per  thousand  years,  while maximum rates are  of the order of
meters per  thousand  years  (Claiborne  and Gera, 1974).  However, as Schneider
and Platt  point out,  in"the  arid  and-semi arid areas-of  the United States,
river  channel  erosion can occasionally be  rapid  (e.g.,  of  the orcer of a
meter per  decade),  and cliffs (e.g.,  at the edges  of mesas) can also retreat
rapidly  (Schneider   and  Platt,  1974).   In  arid   areas,   the   flash  floods
associated  with  cloudbursts  do  not  occur in  well  defined channels and  can
transport  large quantities of material,  possibly causing changes in drainage
patterns (Schneider and Platt, 1974).

     Erosion could primarily  affect  the long-term  security of landfills,  and
the potential impact of flash floods and landslides on-lagoons should also be
considered  when  analyzing, the  siting  of any  land  disposal  facilities.   The
possibilities of glacial action and changes in sea  level  or  the development
of a  lake   have also been considered  for  the perpetual care of radioactive
wastes (Schneider and Platt,  1974).

     Development of an aquifer—Schneider and  Platt  (1974) include an inter-
esting  tentative  estimate  of  the  probability that  an aquifer  wil"  develop
where  none currently exists,  and  that water will  penetrate  a mine disposal
site (also  see  Deonigi,  1974).   These data are presented  in  Table B-7.  The
probability that the  water will penetrate the disposal  tunnel  and cavity is
based on-tunnel ing  and natural gas storage experiences (Schneider and Platt,.
1974).    The  development  of   an  aquifer  could affect all   forms   of  land
disposal.             ••  •                ...

     Hurricanes and tornadoesr-The  strong  winds associated  With  hurricanes
and tornadoes  would  primarily affect  surface structures  (e..g.t  engineered-
storage) and  could  transport  the  contents of lagoons.  Structure design  to
withstand  the  forces  of  high  velocity  winds  is  a  normal  engineering
procedure.

     The potential for damage from tornadoes has been evaluated wr:h respect
to nuclear power plants (see Rasmussen,  1975).   The average incidence of a
tornado  at a nuclear power plant  site has been calculated to  be 5x10  4  per
year,  but only 1 percent  of these tornadoes are expected to exceed the design
criteria (Rasmussen, 1975).
                                     183

-------
               TABLE B-7.   SAMPLE COMPONENTS OF RELEASE SEQUENCE
                           PROBABILITIES • FQ.R GEOLOGIC DISPOSAL*
Aquifer develops- in the
  region where one did
  •not exist previously

Wa.ter finds path into
  disposal site

Water flow cannot be
  control-led by man

Water is flowing

Cumulative,.release
  probabi.l ,ity.- -in- the
  time given;
                                          Probability of waste release


Fai lure event
•During
operational'- •
period

During 1,000
years
During
1,000,000
years
         10~8.   .,10>.to 10~4     10'r-tolO"1
10~4 to.10"2     -10~4 to 10"2 .   ' 10"4 to 10"2
1

1
1

1
    4 to lQ~io "'  10"10 to 10"6   •  10.".s  to  10."3
* Source:   Schneider  and  PIatt,  1974:1.17.


     Meteorite  impact—The  probability  of .the impact of  large  meteorites is
-•low   ^CTaiborne and Gera  indicate that the  probability of a meteor of mass in
excess  of ~2xl07 ' kg  striking'the'earth  is  about 10 13 per-km2-per year.  A
meteorite of  the above size would'make a crater.,of about 1 km in-diameter and
300  m  deep (Claiborne and Gera^  1-974).   Schneider and Platt-(1974) quota_data
suggesting that  the  .probability,  of  a meteorite  .being capable  of forming a
crater ..more,  than  400 m deep is. 10 21-per  km2 per year.   Such  craters could
threaten-any form  of land- disposaT.   Smaller meteorites are more common, and
-the  probabiTity::of. a  meteorite; :in the:.10-tonne .range; striking a  large surface
•'builUtng .has been, estimated to .-be Tess:,than 10 .I0. per year.   Even so,  this
'appears "to^be" an event  that can be  neglected'in  comparison with many other
 uncontrollable  risks.                                   .,

 Uncommon Man-caused Initiating Events

      Aircraft impacts—The  effects  of an  aircraft crash (including possible
 fuel  combustion)  have been  investigated for nuclear  power plants and waste
 processing facilities  and  for  the transport of nuclear materials  (Rasmussen,
 1975;   Energy Research and Development Administration,  1977).   The risk  from
 aircraft  during surface transport  is very  small  compared  to other types of
                                      184

-------
transport accidents,  but  due to the large quantities of energy that might be
involved, this  risk  is  of interest for  materials  (such as high-level radio-
active  materials)  that  are containerized securely  enough  to  resist soill age
in most  accidents.   The probability of a potentially damaging air crash at a
reactor  location  in  the United States, that  is_more than five miles from any
airport  has  been estimated  to be  10"6  to  10"8 per  year  (Rasmussen, 1975).

     Sabotage--Several  authors have  attempted  to  estimate  the  probability
that  facilities  containing   radioactive wastes  will  be   sabotaged  (Energy
Research  and Development  Administration, 1977;  Schneider and Platt,  1974;
Claiborne and Gera, 1974).  They conclude that the probability is low because
there  are more  attractive targets  for  saboteurs.    On the other  hand,  the
security  taken  at   any facility  where  sabotage   could  lead to  disastrous
consequences will reflect  the attractiveness of the  target,  which  will  tend
to even  out  the probabilities between different targets.  Nevertheless, most
techniques for  the ultimate  disposal  of hazardous wastes would not appear to
be  particularly  vulnerable   to   sabotage,   but  treatment  facilities  and
engineered  storage  could be.   Given  a  reasonable  level  of  security  at
hazardous waste management facilities,  it is difficult to  predict  the prob-
ability of a successful  sabotage attempt without knowing more about the local
political climate and alternative targets at the time.

REFERENCES

Angel 1, G.R., and A.S. Kalelkar.  llThe Cost of Relative Spill  Hazards Associ-
     ated With  the Modal  Transport of Hazardous Materials."   In:  Control of
     Hazardous Material  Spills.  Proceedings of the 1974 National_ Conference
     on Control of Hazardous  Material Spills, Anon,  ed. American Institute of
     ChemicalEngineers and  the U.S.  Environmental  Protection  Agency,  San
     Francisco, California, August 25-28, 1974.  pp. 90-98.

Anon.  ed.   Control  of Hazardous  Material Spills.   Proceedings  of the 1974
     NatipnaT  Conference on  Control of Hazardous Material  Spills.American
     Institute  of  Chemical  Engineers  and the U.S.  Environmental  Protection
     Agency,  San Francisco, California, August 25-28, 1974.  387 pp.

Arthur  D.  Little,  Inc.   Alternatives  to the Management  of Hazardous Wastes
     at National Disposal  Sites.Volume II.EPA/530/SW-46c-2, U.S. Environ-
     mental  Protection  Agency,  Washington,  D.C.,  1973.   [NTIS:   PB237264]

Arthur D. Little, Inc.  A Modal Economic and Safety Analysis of the Transpor-
     tation of Hazardous Substances in Bulk.    COM7411058,  U.S.  Department of
     Commerce, Maritime Administration, 1974.

Arthur  D.  Little,  Inc.   Safety  Considerations  in Siting  Housing Projects.
     Contract  No.  H-2178R Modification  1,  Department  of  Housing  and  Urban
     Development,  Office   of  Policy  Development  and  Research,  Washington,
     D.C., 1975.  124pp.   [NTIS:   PB253786]

Battelle Memorial Institute,  Pacific Northwest Laboratories.  Program for the
     Management  of Hazardous Wastes.   Volume I.   EPA-530/SW54c-l, U.S.  Envi-
     ronmentalProtection .Agency,  Washington,  D.C.,  1974.   403pp.  [NTIS:
     PB233630]

                                     185

-------
Booz, Allen,  and  Hamilton.   An Appraisal  of  the  Problem  of the Handling,
     Transportation,  and  Disposal  of Toxic and Other  Hazardous Materials.
     Contract No.  05-00033,  Environmental  Quality Council, Washington, D.C. ,
     1970.   180 pp.   [NTIS:   PB236599]

Buckley, J.L., and S.A.  Wiener.  Hazardous Material Spills:  A  Documentation
     and Analysis  of Historical Data.EPA-600/2-78-066,U.S.Environmental
     Protection Agency,  Cincinnati, Ohio, 1978.

Burkholder, H.C.,  M.O.  Cloninger, D.A. Baker, and G.  Jansen.  "Incentives for
     Partitioning    High-Level   Waste."    Nuclear Technology.   31(November):
     202-217, 1976.

California State Water Resources Control Board.   Waste Discharge Requirements
     for Nonsewerable Waste Disposal to Land:  Disposal Site Design and Oper-
     ation Information.    California  State  Water Resources  Control  Board,
     Sacramento, California, 1976.   60 pp.

Carter,  C.D., R.D.  Kimbrough,  J.A.  Liddle, R.E.  Cline, M.M.  Zack, Jr., W.F.
     Barthel, R.E.  Koehler,  and P.E.  Phillips.   "Tetrachlorodibenzodioxin:
     An Accidental  Poisoning Episode in Horse  Arenas."   Science.  188(4189):
     738-740, 1975.

Claiborne,  H.C.,  and F.  Gera.   Potential  Containment Failure Mechanisms and
     Their  Consequences  at a RadioactiveWasteRepositoryin Bedded Sait
     in New Mexico"!    ORNL-TM-4639,  Oak  Ridge  National   Laboratories,  Oak
     Ridge, Tennessee,  1974.

Coffman, J.L.,  and  C.A. von Hake.  United States Earthquakes.   U.S.  Depart-
     ment  of  Commerce,  National   Oceanic  and Atmospheric  Administration,
     Silver Spring,  Maryland, 1972.

Dawson,  G.W., and  M.W.  Stradley.  "A Methodology for Quantifying the Environ-
     mental  Risks From  Spills  of Hazardous Material."  A  paper presented at
     the AIChE  Conference,  Boston,  Massachusetts, September 8, 1975.   35 pp.

Dawson,  G.W., A.J.  Shuckrow,  and W.H. Swift.  Control of Spillage of Hazard-
     ous Polluting Substances.   15090FOZ 10/70,  Federal Water Quality Admin-
     istration,  Department of the Interior, Washington, D.C., 1970.

Dawson,  G.W.,  M.W.  Stradley,  and  A.J.  Shuckrow.   Determination  of  Harmful
     Quantities  and Rates of Penalty for Hazardous Substances.   Volumes II,
     III.  IV.EPA-440/9-75-005,   U.S.EnvironmentalProtection   Agency,
     Washington, D.C.,  1975.

Deonigi, D.E.  "Evaluation Methodology of Waste Management Concepts." Nuclear
     Technology. 24(3):331-338, 1974.

Elzy, E.,  and  F.T. Lindstrom.  "Model  of  the Movement  of  Hazardous Waste
     Chemicals  for  Sanitary  Landfill Sites."   In:   Proceedings of the Con-
     ference on Environmental Modeling and Simulation, W.R. Qtt, ed.  EPA-600/
     9-76-016, U.S.  Environmental  Protection Agency,  Washington, D.C., 1976.
     pp. 609-613.

                                     186

-------
Elzy, E.,  F.T.  Lindstrom,  L.  Boersma, R. Sweet, and P. Wicks.  "Model of the
     Movement of Hazardous Waste Chemicals In and From a Landfill Site."  In:
     Disposal of Environmentally Hazardous Wastes.  Oregon  State  Un'versity
     Environmental  Health  Sciences  Center  Task  Force  on  Environrientally
     Hazardous Wastes, Corvallis, Oregon, 1974.  pp. 111-201.

Energy  Research  and Development  Administration.   Alternatives for Long-Term
     Management  of  Defense  High-level  Radioactive  Waste, Savannai  •River
     Plant. Aiken. South Carolina.  Volume 1.  ERDA, 77-42/1, Energy Research
     and Development Administration, Washington, D.C., 1977.

Fenn, D.G.,  K.J.  Hanley,  and T.V. OeGeare.  Use of the Water  Balance Method
     for  Predicting  Leachate  Generation  from  Solid Waste Disposal Sites.
     EPA/530/SW-168, U.S.  Environmental  Protection  Agency,  Cincinnati, Ohio,
     1975.  40 pp.

Ghassemi, M.  Analysis of a Land Disposal Damage Incident  Involving  Hazard-
     ous Waste  Materials,  Dover Township, New Jersey.  'Contract  Nc.  68-01-
     2956,  U.S.   Environmental   Protection  Agency,  Washington, D.C.,  1976.
     127 pp.

INTERCOM?  Resource  Development and  Engineering,  Inc.   A Model for Calculat-
     ing Effects  of Liquid Waste Disposal in Deep  Saline Aquifers. Pa*-c_I--
     Development.Water-ResourcesInvestigations76-61,U.S.Geological
     Survey, Reston, Virginia, 1976.

Jones, G.P., et al.   Risk  Analysis  in  Hazardous  Materials Transportation.
     Volume I.   Contract  No.  DOT  OS 20114,  Department of  Transportation,
     1973.  297 pp.

Kalelkar, A.S.,  L.J.  Partridge,  and R.E. Brooks, Jr.   "Decision Analysis in
     Hazardous Material Transportation.". In:.  Control  of Hazardous Material
     Spills.  Proceedings  of  the  1974  National  Conference  on Control
     of Hazardous Material Spills.Anon. ed.American Institute of Chemical
     Engineers and  the U.S. Environmental  Protection  Agency.,. San Francisco,
     California, August 25-28, 1974.  pp. 336-344.

Konikow, L.F.  Modeling  Chloride  Movement  in  the Alluvial  AquiFer at the
     Rocky Mountain Arsenal, Colorado.Water-Supply Paper 2044, U.S. Geolog-
     ical Survey, Washington, D.C., 1977.  48 pp.

Lazar,  E.G.   "Damage  Incidents  From Improper Land  Disposal."  Journal  of
     Hazardous  Materials. 1:157-164, 1975/76.

Lazar,  E.C.   "Summary of  Damage Incidents From  Improper Land Disposal."  A
     paper presented at the National Conference on Management and Disposal of
     Residues from the Treatment of Industrial Wastewaters,  Washincton, D.C.,
     1975.  16 pp.

Lazar,  E.C.,  R.  Testani,  and A.B. Giles.  "The Potential for National Health
     and  Environmental  Damages from  Industrial  Residue Disposal."   A paper
     presented at the National  Conference on Disposal of  Residues  on Land,
     St. Louis, Missouri, September 15, 1976.  21 pp.

                                     187

-------
Miller, D.W., F.A. DeLuca, and T.L.  Tessier.  Ground  Water  Contamination in
    ~the Northeast States.  EPA  660/2-74-056,  U.S.  Environmental  Protection
     Agency, Washington, D.C., 1974.

Munn,  A.   "Health Hazards  to Workers  from Industrial  Chemicals."  Chemical
     Society (London) Reviews, 4(l):82-89, 1975.

Nace, R.L.  "Problems of Underground Storage of Wastes."  Journal of Research
     of U.S. Geological Survey. 1(6):719-723, 1973.

National  Academy  of  Sciences.   Assessing   Potential   Ocean   Pollutants.
     National Academy of Sciences, Washington,  D.C. , 1975.  458 pp.

National  Academy  of Sciences, Committee  on Hazardous  Materials, Analysis of
     Risk  in  the  Water  Transportation  of  Hazardous  Materials.  COT-CG-
     41680-A, U.S. Coast Guard, Washington, D.C., 1976.  118 pp.

Ocean Disposal Study Steering Committee.  Disposal in the Marine  Environment:
     An Oceanographic Assessment.   National  Academy  of Sciences, Washington,
     D.C., 1976.   85 pp.

Ott, W.R., ed.  Proceedings of  the EPA  Conference on Environmental Modeling
     and Simulation.EPA 600/9-76-016, U.S. Environmental Protection Agency,
     Washington,  O.C., 1976.  861pp.

Rasmussen,  N.C.,  director.   Reactor Safety Study:  An Assessment of Accident
     Risks in U.S. Commercial  Nuclear Power Plants.WASH-1400,  U.S.  Nuclear
     Regulatory Commission, 1975.207 pp.[NTIS:  PB248201]

Raleigh,  C.B.   "Earthquakes  and  Fluid Injection."   In:   Underground  Waste
     Management  and  Environmental  Implications, T.D. Cook, ed., Memoir 18,
     The American Association of Petroleum Geologists, Tulsa, Oklahoma, 1972.
     pp. 273-279.

Reeder,  L.R.,  J.H.  Cobbs,  J.W.  Field, Jr.,  W.O. Finley,  S.C.  Vokurka, and
     B.N. Rolfe.   Review  and Assessment of Deep-Well Injection of  Hazardous
     Waste.   Volume I.£PA600/2-77-019a,U.S.EnvironmentalProtection
     Agency, Cincinnati, Ohio, 1977.  215 pp.

Schneider,  K.J.,  and A.M. Platt,  eds.  High-level  Radioactive Waste Manage-
     ment Alternatives.   4  volumes.   BNWL-1900,  Battelle N.W.  Laboratories,
     Richland, Washington, 1974.

Schultz, D.W., ed.   Land  Disposal  of Hazardous Wastes:  Proceedings  of the
     Fourth Annual Research Symposium.EPA-600/9-78-016,  U.S.  Environmental
     Protection Agency, Cincinnati, Ohio, 1978.

Shuster,  K.A.   Leachate  Damage Assessment:  Case  Study  of  the  Fox Valley
     Solid  Waste  Disposal  Site in Aurora. Illinois.   EPA/530/SW-514.  U.S.
     Environmental Protection Agency, Cincinnati, Ohio, 1976a.  41 pp.
                                     188

-------
Shuster,  K.A.   Leachate Damage Assessment:  Case Study of the Peoples Avenue
     Solid  Waste Disposal Site in Rockford. Illinois.   EPA/530/SW/517.  U.S.
     Environmental Protection Agency, Cincinnati, Ohio, 1976b.  33 pp.

Shuster, K.A.  Leachate Damage Assessment:  Case  Study of the Sayville Solid
     Waste Disposal Site In Is lip (Long Island). New York.    EPA/530/SW-509,
     U.S.  Environmental  Protection Agency, Cincinnati,  Ohio, 1976c.  25pp.

Smith, D.D., and R.P.  Brown.  Ocean  Disposal   of Barge-Delivered Liquid  and
     Solid  Wastes  from  U.S.  Coastal  Cities.SW-19c,  U.S.  Environmental
     Protection Agency, Washington, D.C., 1971.  128 pp.

U.S. Department of  Labor,  Bureau of Labor Statistics.  Occupational Injuries
     and Illness in the United States by Industry. 1974.  Bulletin 1332, U.S.
     Department or" Labor, Wasnington, O.C., 1976.  147 pp.

U.S.  Environmental  Protection  Agency.   Report  to  Congress:   Disposal  of
     Hazardous  Wastes.   SW-115,   U.S.'  EnvironmentalProtection  Agency,
     Washington, D.C., 1974.  120 pp.

U.S.  Environmental  Protection  Agency.   Hazardous  Waste  Disposal  Damage
     Reports.    EPA/530/SW-151,    U.S.    Environmental   Protection   Agency,
     Cincinnati, Ohio, 1975a.   12 pp.

U.S.  Environmental  Protection  Agency.   Hazardous  Waste  Disposal  Damage
     Reports.  Document No. 2.  EPA/530/SW-151.2, U.S.  Environmental Protec-
     tion Agency,  Cincinnati,  Ohio, 1975b.  16 pp.

U.S.  Environmental  Protection  Agency.   Hazardous  Waste  Disposal  Damage
     Reports.  Document No. 3.  EPA/530/SW-151.3, U.S.  Environmental Protec-
     tion Agency,  Cincinnati,  Ohio, 1976.  17 pp.

U.S. Environmental Protection Agency.  Fish  Kills  Caused  by  Pollution  in
     1975:   Sixteenth  Report.   EPA-440/9-77-004, U.S.  Environmental Protec-
     tion Agency,  Washington,  D.C., 1977a.  88 pp.

U.S. Environmental  Protection Agency.   The  Report  to Congress:  Waste Dis-
     posal  Practices   and Their Effects~on Ground Water.EPA-570/9-77-001,
     U.S.  Environmental  Protection  Agency,  Washington, D.C., 1977b.  531pp.

Venugopal, B.,  and  T.D.  Luckey.   "Toxicology of Non-Radioactive Heavy Metals
     and Their Salts."  In:   Heavy Metal Toxicity. Safety and Hornology. T.D.
     Luckey, B. Venugopal,  and  D. Hutcheson,  eds.Academic Press,  New York,
     New York, 1975.  pp. 4-73.

Walker,  W.H.   "Where  Have  All   the Toxic  Chemicals  Gone?"   Ground Water.
     11(2):11-20,  1973.

Warner, D.L., and D.H.  Orcutt.  "Industrial Wastewater-Injection Wells in the
     United States—Status of Use and Regulation, 1973."  Proceedings: Second
     International  Symposium on Underground  Waste Management and Artificial
     Recharge.    Volume 2.,  J.  Braunstein, ed.   The  George Banta  Company,
     Inc., Menasha, Wisconsin, 1973.  pp. 687-697.

                                     189

-------
                                  APPENDIX C

                     VALUATION OF THE EFFECTS OF HAZARDOUS
                          WASTE MANAGEMENT TECHNIQUES


     This appendix  discusses the  nature  of control costs  and  indicates how
some of the common environmental impacts can be valued.

CONTROL COSTS

Generator's Costs

     Generator's costs Include all costs that the generator of a waste incurs
specifically  in  dealing  with its waste disposal  problem.   Gen°.-ator's costs
are the  only  costs  that  are internal to  a  firm, and therefore, according to
classical  microeconomic   theory,  they  are   the  only costs that  a  firm  is
assumed to consider in its decisionmaking.

     Generator's costs include  any costs  incurred by the  generating firm to
reduce the generation  of hazardous waste, as well as those involved in waste
separation,  treatment,  and  disposal.   They should  include legal  costs and
taxes  directly  applicable  to  the  waste handling  and disposal  operations
(e.g., property  taxes on  relevant facilities or  equipment),  but  not those
that cannot readily be attributed to waste handling or disposal  (e.g., income
taxes).  If the waste is subjected to any form of resource recovery, a credit
for the value of the materials recovered (based on the cost of an alternative
supply of that quality of material in that location) should be included, even
if  no  market  transaction is involved.  Where a waste is consigned to another
firm  for  treatment,  transport,  and/or  disposal,  the  charges   for  these
services are part of the generator's costs.

     Generator's costs are comparatively  easy to determine for a given waste
and  location.    Hazardous  waste  management  costs  have been determined for
various  industries  and   for  various  methods  of  treatment  and  disposal.
However,  many of  these  data are highly  location-specific  or  waste-stream-
specific  and  are  scattered throughout  the  literature.   A-  few  have  been
collated.   For  example,  Battelle  Memorial  Institute,  Pacific  Northwest
Laboratories  (1974,  Vol.II)  provides costs  for some widely applicable treat-
ment  modules;  and  Talley  and Albrecht (1974) provide  a comparison of costs
for some  disposal  techniques,  (Table C-l).   An EPA-sponsored study to update
and  collate   cost  data  for  a  wide range of  hazardous waste  treatment and
disposal  techniques has   just  been completed  (U.S.  Environmental  Protection
Agency, in progress).
                                     190

-------
                        TABLE C-l.   ESTIMATED COSTS OF SOME DISPOSAL TECHNIQUES*
           Technique
                                                                  Estimated costs
      Range
   Average
"Normalized"
LAND BURIAL TECHNIQUES

 : Conventional sanitary landfill
  Near-surface land burial
    containerized wastes

  Sludge encapsulation and
    burial
      Operating experience of
        one company
      Polymer encapsulation
      Cement encapsulation
      Asphalt

  Deep burial

ENGINEERED STORAGE

DEEP WELL INJECTION

OCEAN DISPOSAL

  Bulk
  Containerized
$1.98-$4.74/tonne
$3.30/tonne

$26.50/m3
$7.80-$31.80/m3
                       $26.40/m3
                     •  $66.10/m3
                       $35.20/tonne

$11.30-$33.90/m3

$106-$177/m3

$0.13-$0.53/m3
$0.66-$10.50/tonne    . $1.87/tonne
$5.50-$143.OO/tonne    $26.40/tonne
$0.99/m3

$26.50/m3
                $7.80-$31.80/m3
                $27.20/m3
                $66.00/m3
                $42.40/m3

                $11.30-$33.90/m3

                $106-$177/m3

                $0.13-$0.53/m3
                $1.87/m3
                $26.40/m3
* Source:.  Talley and Albrecht, 1974.   (Data were converted from U.S.  customary units.)

-------
      A major  problem  that may  be encountered  In attempting  to  determine
 generator's  costs  for planning purposes is the  variability  of  such costs  with
 local  conditions,  even  if  the level  of environmental  protection  is  not  con-
 currently varied.   Although disposal  costs  (for  a given technique)  are  not,
 in  general,  sensitive  to  the  precise composition of  the wastes  involved,
 treatment costs can vary considerably.   For example,  Arthur 0. Little,  Inc.
 (1973, Vol.1)  showed  a variation  in  total  operating  costs from 0.3 to  15.6
 cents per liter (1973  data)  for treatment of various  wastes containing heavy
 metals.   (These wastes  were  dilute heavy metals, concentrated heavy metals,
 and  organics with  heavy metals.   Data  excluded  sludge  disposal  costs  and  were
 for   waste   stream   volumes   typical   of   large-sized   industrial   plants.)
 Furthermore, major economies "for  scale are possible.  Arthur  0. Little,  Inc.
 (1973, Vols.I  and  II)  assumed  an  exponent  of -0.4  in  the  average  unit
 processing cost/throughput relationship  for all  treatment processes consid-
 ered.   (The"term "processing cost"  appears to  refer  to  labor  and capital
 related  costs only,  and  to exclude chemicals and utilities which  were taken
 as constant  unit, costs.)

      Disposal  costs will,  however, depend on geologic and climatic  factors,
 such as  the depth and  injection-pressure needed  for deep well injection,  the
 suitability  of'the location for evaporation ponds,  and the  relative ease  with
 which secure (chemical)  landfills  can  be constructed.   Talley and Albrecnt's
 'data (Table  C-l)  are  based  on a single source of  data  for each tecnnique,  and
'yet  typically they show variations of  2:1"to 4:1, with a far larger range for
 ocean disposal.  This  varies  "according  to geographical  area,  type of waste,
 distance   [from port]  to  the  disposal   area  and  annual  volume  of wastes
 handled"   (Smith  and  Brown, 1971).

 Administrative  Costs

      Administrative costs include the  costs of  planning,  promulgating regula-
 tions, monitoring compliance,  and prosecuting noncompliance with regulations,
 studying  incidents and  problems,  and  advising  industry.  These costs will  be
 incurred  by various   levels  of government  and by watchdog bodies  that  are
 concerned with environmental   quality.   Research  and development costs  should
 not   be  included  unless they  are integrally  associated  with  a  particular
 approach  to  hazardous waste control.

      It  is often  difficult  to derive  precise data on the administrative costs
 •involved, in a. .particular program area, because of multiple coverage by  many
 government departments.  For  example,  a legal department or a  laboratory  will
 probably'handle'the work.of  many program areas and will  not be restricted to
 hazardous 'wastes.    Fortunately,   the  analyst  is  primarily   interested  in
 changes  in  administrative  costs  as the approaches to  hazardous waste control
 are  varied,  and these  can  more readily be estimated by  assessing  the cost of
 employing  the  additional   staff  directly  needed to   implement  programs,
 together  with an  allowance  for support'from service departments, consultants,
 etc.                                         •••••

      Few   empirical  data have been  published  on  administrative   costs,  and
 those that  have are  generally broad  in coverage.  For  example, 1977 Federal
                                      192

-------
expenditures  for  standard setting  and  enforcement  in  relation to  al_^ pol-
lution  control  activities  were estimated  to  be  $422  million (Council  on
Environmental Quality, 1976).  However,  it is widely held that administrative
costs  can  vary  substantially  with  different  regulatory approaches.   The
National  Academy  of Sciences  (1975)  highlighted  this  variability,  and sug-
gested  that  the  magnitude  of  regulatory  administrative costs  should  be
considered in selecting an appropriate regulatory action.  Also, in assessing
the potential efficacy of alternative institutional approaches to the manage-
ment  of  hazardous  wastes,  Battelle  Memorial  Institute,  Pacific  Northwest
Laboratories  (1974,  Vol.1) gave Federal  cost a weighting of one-seventh in
deriving  an  effectiveness ranking.   Administrative costs can be expected to
increase  as  the  extent  of regulation is increased  (as  opposed to achieving
objectives by use of market forces) and as  the degree of environmental pro-
tection or quality is increased.

     Although crude, an  approach  that is sometimes adopted is to take admin-
istrative costs as a fixed proportion of some other major cost.  For example,
Moll  et  al.  (1977)  suggested  that administrative  and  enforcement  costs
incurred  by  governments  for  pollution control activities might amount  to 10
percent  of  the  annualized  costs  of   the  relevant  industrial   pollution
controls.

     Effluent fees  or fines  for  the violation  of standards  may  be used to
internalize  costs external to  a  firm and  can thereby  affect its behavior.
However,  because  fines and effluent charges are only  transfer payments bet-
ween different  categories  of control  costs, any contribution that they make
to  overall  economic  efficiency must  be by means  of changes  in pollution
control activities.   In work oriented toward the control of fly ash emissions
from power plants. Downing and Watson have analyzed the effect of an agency's
enforcement activities on a firm's projected behavior under the assumption of
cost minimization (Downing and Watson, 1974, 1977;  Watson and Downing, 1976).
They compared the effects of effluent fees with standards and fines,  and they
examined  the effects  of  using  different  compliance  tests  and  numbers  of
inspections,  as  well as  different  levels  of  fines and effluent  fees.  One
conclusion was that for the parameters involved in the .fly ash control situa-
tion, a firm's optimum strategy would involve frequent violation of standards
(Watson  and  Downing, 1976).   In  recent work, Downing  and Kimball  (no  date)
examined  the pollution  control behavior  adopted  by  firms  in a number of
industries and  found  that the firms appear  to  be  controlling pollution to a
greater  degree   than would  be  implied  by  a- cost  minimization  strategy.
Although Downing and Kimball examined a-number of possible explanations, they
were unable to satisfactorily explain the observed behavior.-

Social Control Costs

     The  category of  "social  control  "costs" is introduced to eaten hidden or
intentional  subsidies  provided  to  enterprises  that  have   to  dispose  of
hazardous  waste.   For  example, to  encourage  adequate disposal  of wastes,
governments might elect to provide landfill  services at no cost, or at a cost
below  that  which would  be justified  on an economic basis.   This  would con-
stitute  an  intentional   subsidy  that  should  be   taken into account  when
comparing the total  costs of alternative disposal  techniques.  A more subtle


                                     193

-------
subsidy—and  one  that  may  not  be  restricted  to  services   provided  by
governments—can occur where the  opportunity  cost or the replacement cost of
a facility  (such as  a landfill)  is higher than the historic  cost or account-
ing cost.   For  example,  even  in  the absence of inflation,  an accounting cost
could readily understate the capital component of landfill  costs.   Often,  the
land employed was  obtained  at  low cost;  but with the growth  of  the community
that it serves,  both the opportunity cost reflecting the value of the land in
alternative uses, and the cost of replacing the facility with another that is
functionally equivalent would be  far higher than the historic cost.   Govern-
ment charters may preclude landfill operation in such a way as to make a book
profit, and  individual  firms  often base  their charges on historic costs,  not
recognizing  the  need to  consider opportunity costs or to plan  for facility
replacement.

     Social control  costs—both  deliberate  and unintentional—can arise with
most hazardous waste management  techniques.   Highway users,  for example,  may
not pay the full cost of providing and maintaining the highway system.   Where
this type of cost exists, it is important that its presence is recognized and
that it  is  included  in  the evaluation  if it  is  a  significant  factor.   In
cost-benefit analysis, a "shadow price" is sometimes used to  reflect the true
cost of  a  transaction,  as opposed to the actual money cost,  if  any.   This is
equivalent to adding the social  control cost to the generator's  cost.

     Social  control  costs  are  external  to  the  waste  generator,   who  will
therefore  not  take  them  into  account in his  decisionmaking.   However,  such
costs might deliberately be incurred for some techniques to steer generators'
decisions  toward  ones that are  efficient  from  a societal  viewpoint  (i.e.,
when all costs and damages are considered).

VALUATION OF ENVIRONMENTAL IMPACTS

     This  section  discusses the  valuation of  environmental  impacts  that may
arise  from  hazardous wastes.   Some economic theory  that may be used in such
valuations  is presented  followed by discussion on  the  valuation  of commonly
encountered  categories  of  impact.   These  categories  have been  organized
largely  on the basis of  the  valuation techniques that might be  used,  and a
given  waste management  technique  may  result  in  impacts  in more  than  one
category.

Theoretical Considerations

Wi 11 i ngness-to-Pay—
     The valuation of environmental impacts is frequently more difficult than
the valuation of control  costs,  as there are no established markets for many
effects  (e.g.,  a  change  in water  quality).   Although  some  authorities con-
sider  that  monetary  values  should  not  be  assigned to  certain  effects
(Tihansky,  1975),  in principle  it is almost  invariably possible  to impute a
value  from  some  sort  of willingness-to-pay  survey or equivalent measure.
This involves asking affected consumers how much they would be willing to pay
to receive  a specified benefit,  even though it might be impossible or unfea-
sible  to collect such payments.   Because it  is  often difficult to develop a
willingness-to-pay  survey that  is not subject to bias, willingness-to-pay is
frequently  imputed  from some observed behavior.

                                     194

-------
     The  term  "willingness-to-pay"  is  frequently  used  to  encompass  two
separate  concepts  that  can  lead  to quite  different results  under certain
circumstances.   Thus   there  is  a  distinction   between   an  individual's
wi11ingness-to-oay to obtain  increased  utility,  and the compensation that an
individual would  demand to  accept  a reduction  in utility  (Freeman,  1978).
(The  term  ''wi11ingness-to-pay"  is  commonly  employed  to   encompass  either
measure,  except where  the distinction  is  important.)   If an  individual's
marginal  utility  of income  is  approximately constant (as where  the changes
are very  small  compared to total   income)  the two measures can  be  :aken as
equivalent (Freeman, 1978).   However,  in some circumstances, the distinction
can be important, as willingness-to-pay is limited by an individual's income,
whereas  compensation demands  are  not.   Consequently  some  individuals  may
indicate  that they would not accept certain  situations  at  any level of com-
pensation  (Prato,  1974).   Thus  the  compensation  that  an  individual  would
require  to accept  a reduction  in  his  life expectancy  or  to undertake, an
activity  that carries a given chance of death could  be  far greater than the
payment  that  he  would   be  prepared or  able  to  make  to  increase  his  life
expectancy by  a similar -amount,  or to  avoid an  activity that has  the  same
chance of death (Hirshleifer, Bergstrom and Rappaport, 1974).

     The  measurement  of willingness-to-pay  for  environmental  improvement is
difficult,  as   individuals  can  have strong  incentives for  concealing  their
true preferences (MSler  and Wyzga,  1976).   Thus  in responding to a question-
naire  or  in a  bidding  game, it  is in  the  interest of those responding to
overstate  their willingness-to-pay  if  their responses  carry no  financial
obligations.   If the amount they will actually pay is essentially independent
of their  responses,  they are.likely to promote  the perceived  improvement in
the quality of the environment'to the maximum extent possible-by. exaggerating
their  responses.    On   the  other  hand,  if  there  is  a  direct  individual
financial obligation associated with the responses, it is in the interest of
an individual  to  understate his  response in the  hope that other persons will
bear the brunt of improvement costs.  (This is an example of the "free-rider"
problem that  can arise  with  public goods.)   In  the  context of  valuation of
amenity  benefits,  Maler and  Wyzga (1976) suggest ways  in  which it may be
possible to
-------
to  pay  for  the  benefit.    If  pollution  reduces  the  attractiveness  of
recreational opportunities  in  the  park,  then the demand curve will be lower,
and  the  cost  of  that  pollution  (as  far  as  its  effect  on  the  park  is
concerned) will be the reduction in the consumers'  surplus.

     Where the distinction  between willingness-to-pay to obtain a procuct or
service and the  compensation  required to forego it is significant, there ara
two  different  measures  of   the   consumers'   surplus.   The  compensating
variation  measures  the  consumers'  surplus  in  the  former   case,   ana  the
equivalent variation measures  it  for  opportunities  foregone.    For  a normal
(as opposed to inferior)  good, the compensating variation is smaller than the
area under  the demand curve,  which  in  turn is smaller  than  the equivalent
variation.  However, where  income  effects  are negligible, the three measures
coincide.    In  practice,  although  goods   having  zero  income  effect  are
uncommon,   the  effect  is often  sufficiently  small  for  the   area  under tr.e
demand  curve   to   be  used  with  acceptable  accuracy   in  many  cost-oenefit
analyses (Mishan, 1971a).

     Unfortunately, analysis of  environmental  problems is complicated oy the
fact that  consumers can be highly emotional  about giving up their ngnts,
with the  result  that  the  two measures  are  not equivalent.   This s^cuation
was, for  example,  found  to  pertain to waterfowl hunting  (Browr and Hai-,,iack,
1972)  and  to  the public's  stated  views  about  changes  in air  quality :n tns
Four Corners area where Arizona, Colorado,  New Mexico and Utah T.eet (RandaT,
Ives,  and Eastman,  cited   in  Prato,   1974).   Because  of the  a-f:iculty  of
designing survey instruments that elucidate genuine responses  to willingness-
to-pay questions (and especially those that relate to compensation required),
there  may  be  considerable  uncertainty  about  the  true  magnitude  of  a
consumers' surplus.

Destruction and Damage to Manmade Property

     Destruction and damage to manmade property can be a comparatively easy
category  of  environmental  damage to  value.    If some  replacement of the
damaged  item   is  required,  then  replacement  cost will  provide  a   suitable
method  of  valuation where  the replacement is  functionally equivalent to the
damaged item.  Though  in many cases this can be assessed on the basis of the
cost  of a  similar  item (less  an allowance  for  depreciation  where appro-
priate), there will  be situations  where direct replacement is inappropriate.
For  example,   where wells  are  irreversibly contaminated,  the  damage  cost
should include the cost of providing an alternate water supply.  Provision of
an  alternate  water supply  from  a  different  source is  likely  to  be  far more
costly than drilling a direct replacement for the original well.

     On the other hand, there could be circumstances in which the appropriate
replacement  cost was  less  than that  of replacing the damaged  item with  a
direct  equivalent.   Due to changing  technology,  the  direct  reolacement  of
some  structures   (e.g.,  a  fire-damaged barn)  with  a physically  identical
structure  could  be more costly than  replacement  with  a  functionally equiv-
alent  structure   of different design.  Of  course,  replacement  of  damaged
property  with  something  that  is functionally equivalent  but  physically dif-
ferent  could   also  cause a social impact by  changing the  aesthetic  value
associated with the property.

                                     196

-------
     In other  cases,  because of  changing demand  or  technology,  replacement
would be  inappropriate,  as  there  might be some better means of utilizing the
resources or solving the problem.   For example, if an abandoned mine that was
being used  for hazardous waste disposal  were destroyed by  an  explosion,  it
would be  inappropriate  to  calculate  the cost  of replacing  the  mine,  since
some  completely  different  disposal  technique  would  probably be  adopted.
However,  in  this  case,  a  "value  in  use" could  be  attributed to  the  mine,
probably  based on  the  costs  of  an  alternative  waste disposal  technique.

Damage to Human Life and Health

Loss of Human Life--
     The valuation of human life  has recently been discussed by Hirsnleifer,
Bergstrom,  and Rappaport  (1974),  Zeckhauser  (1975),  and  Linnerooth  (1S76,
1975a,b).   Linnerooth (1976) identifies six approaches, as  follows:

      1.  The human-capital  approach:   Each life  is  valued according to
          the discounted future earnings of the beneficiary;

      2.  The insurance approach:   Each  life  is valued on  the  basis of
          individual life insurance  decisions;

      3.  The court-decided  compensation  approach:    Here  information
          from court awards  for fatal  accidents or diseases  is used to
          value each life;

      4.  The implicit-value approach:  • Each life is valued according to
          values  implicit  in past decisions  affecting human mortality;

      5.  The portfolio approach:    Changes   in   mortality  risk   are
          compared  with  the  entire  portfolio  of  risks  assumed  by
          society;

      6.  The utility,  or wi11ingness-to-pay approach:    This   approach
          values risk reduction by  the public s preferences or willing-
          ness to pay for this  reduction.

     The human-capital   approach is  probably the best  known  and widely  used,
although it  has been criticized  on  the grounds that it values livelihood and
not  lives  (Linnerooth,  1976).   Within  this approach,  there  is an  important
subdivision between the gross value  version,  which evaluates an individual's
lifetime  earnings,  and the  net value  version,  which only   considers  the
economic  loss  that  would be suffered  by others should that individual die,
(i.e.,  it excludes  his  consumption  or maintenance cost) (Linnerooth, 1975b).
Despite this  significant philosophical distinction (as an  individual's con-
sumption is  likely  to  be a large  portion of his earnings), values derived by
both versions of the approach have clustered around $200,000 (in 1953 to 1974
dollars) (Linnerooth,  1975b).

     The next three  approaches  (2, 3,  and 4) have not received as much atten-
tion as that of the human-capital approach and  can  lead  to a  wide  range  of
results.   For  example,  life  valuations  derived from  the  implicit-value


                                     197

-------
approach have  ranged from $9,000  to  $9 million (Linnerooth, 1976).   On  the
other hand, there  is evidence that court-decided compensation  is  related to
the human-capital  approach,  as  court  awards for loss of  life  are frequently
based on the expected future  earnings  of the deceased.

     A major difficulty  associated  with all of the  first  four  approaches is
that they can  lead to the conclusion  that for some sectors of  society, lives
have  no value.    These  sectors are  non-wageworkers  (e.g.,  homemakers  and
retired persons) and persons  with no dependents (for the insurance approach).
Although there are  ways  to  circumvent  this  problem,   it  does  point  up  the
inadequacy of these approaches.

     The last  two  approaches (5 and 6)  are of particular  interest for risk-
benefit analyses  as they  express  mortality risk in probabilistic  terms, as
opposed  to  dealing  with  ex ante  identifiable  persons.   Thus  they  are
considered appropriate to  decisions where life expectancy or the probaoility
of death is altered (Linnerooth,  1976, 1975b), as  is the case  with hazardous
waste management decisions.

     By far  the most widely known  example of the portfolio approach  is  the
work of Starr  (1969, 1972),  who drew the important distinction between risks
undertaken  voluntarily  and  those  that  are  involuntary.   This  approacn  has
been  criticized by several   authors  (Slovic,  Fischhoff,  and  Lichtenstein,
1976; Linnerooth,  1975b;  Mishan,  1971b), essentially on the grounds that  thp
sociopolitical  decisionmaking process  reflected in  these data does not reveal
the  true   public  preference for  risk-benefit trade-offs.  (This  topic  is
discussed in more detail  in Appendix F.)

     The willingness-to-pay approach is advocated by some authors as the only
conceptually valid way  of valuing  life—i.e., by  basing  the  valuations on
utility  (Mishan,  1971b;  Schelling,   1968).    The   approach  is  potentially
capable of recognizing nonlinearities  in the risk-benefit trade-off relation-
ship, but  it  is  fraught with  difficulties  in use."  There have been a  few
questionnaire-type  surveys  (see  Linnerooth,   1975b),  and  Thaler  and  Rosen
(1973)  have  attempted  to  use  the  "risk  premium"   associated  with  the
remuneration for hazardous jobs to obtain a value  of life.  Once again, both
survey  and  risk premium  results  suggest  a  value  of  life in   the  region of
$200,000 (based on small  increases in mortality).
* For  example,  an individual  might accept a 0.1 percent decrease in survival
  probability to receive $200.  If this applied to many individuals, it would
  lead to  an  ex post life valuation of $200,000.   However, the same individ-
  ual  might  require a  million  dollars to  accept a 50  percent  reduction in
  survival probability  (leading  to  an ex post life valuation of $2 million),
  and  he would  probably be  unwilling to  accept  any compensation—however
  great—in exchange for certain death.   Thus in  this example  the  ex ante
  risk-benefit   trade-off  relationship   is   nonlinear   (see  Hirshleifer,
  Bergstrom, and Rappaport,  1974).
                                     198

-------
Damage to Human Health—
     The valuation  of sickness and  accidents appears to have  received less
attention  than  loss  of life.   Most valuations  appear to  be  based on  the
human-capital  approach,  plus  an  allowance  for  medical   expenses.    This
approach  is,  of  course,   implicit  in  Workman's  Compensation  and  other
insurance schemes, and  such  data  could  be used as  a measure of work-related
damages to life and health.

     Values of lost workdays  derived from the human-capital  approach do  not
include  an  allowance  for  pain  and suffering.   This factor (together with
allowances  for  medical  expenses)  is often  present in court-awarded  injury
compensation.  In some empirical  studies,  an allowance for pain  and suffering
has been included as a proportion of the other costs of accidents and illness
(e.g., 25  percent of  monetary  costs [Garwood and  Newby, cited  by Tihansky,
1976]), but in most studies no explicit  allowance  has been made.

     The National  Academy  of Sciences  (1977) recently used  $50  per working
day as the lost income that would result from a disability.   They .argued that
the nonworking  population  was  included  indirectly  in their  calculations  by
means of resource  transfers.  Unfortunately,  this study did not estimate the
associated  reduction  in medical  costs  that would  result from  reduced dis-
abilities,   nor  did  it make  an  allowance for  pain and  suffering.   Several
other authors have  also used the $50 per  lost  workday figure (e.g., Brewer,
1976; Hub et  al.,  1973).   However, Hub et al.  (1973) restrict  the valuation
of accidents  to those  that  are  "external"  to  the  risky situation  to  avoid
double counting (i.e,  they assume that the labor cost for a worker exposed to
health risks  reflects at least some of the  costs of these  risks).   This  is
the same concept as the wage risk premium mentioned above.

     The distinction between internal and external risk-bearers  is important.
It implies  that  if the risk premium correctly  reflects  the additional  risks
undertaken,  it  is  unnecessary  to  evaluate .the  risk  to  life  and  limb  of
workers  normally  involved  with  hazardous waste  management  activities,  and
only  involuntary  risks to  bystanders need be considered.  .Thus  part of  the
wages paid  to an  operator  at a hazardous  waste disposal  facility should,  in
theory, reflect the higher  level  of job-related  risk  over  that of a similar
job  dealing with  nonhazardous  materials.   While  the  risk  premium may  be
reasonably accurate in an occupation like underground .coal mining where  there
is a history of accidents and illness (e.g.,  black lung disease) known to the
workers (see  Otway  and Cohen, 1975), it is less likely to reflect accurately
the risk in a hazardous waste management facility, because the  risks will  in
most cases be poorly defined.

     Involuntary risks to bystanders' lives and health can arise in many ways
from  hazardous  waste  management activities.   For example,  the  risk incurred
by individuals  along  waste  transport  routes are  involuntary,  as  are  those
that arise  from  undetected  contamination  of water supplies or foodstuffs due
to waste leaching or overflow.
                                     199

-------
Damage or Destruction to Animals, Vegetation and Land Ecosystems

Commercial Crops and Livestock--
     Commercial crops could  be  damaged or destroyed by overflow from lagoons
or treatment  facilities,  and could  be affected  by airborne  pollution from
nearby hazardous waste  facilities.   Land  application could affect subsequent
crops  on  the  same  site,  while  animals  could  eat  or  drink  contaminated
materials.

     It  is  probably reasonable  to  assume that agricultural  products  face a
market of pure  competition.   In this event, destruction  of commercial  crops
and  stock animals  can  readily  be valued at market  prices, although savings
that arise  from not  needing to  harvest and market the crop or products could
be subtracted  if significant.   Cleanup costs should be included where neces-
sary, and adequate  allowance should  be made for  long-lived effects, such as
reduced crop yield  or  the need  to plant less valuable crops in future years.

Valuation of Wildlife,  Etc.-
     Although damage to land-based ecosystems from hazardous waste spills and
severe air pollution is likely to be quite localized, the damage potential to
aquatic systems  is  probably one of the more  important environmental impacts
that can  arise from hazardous  wastes (see Appendix B).   Methods  of valuing
wildlife  and  associated ecosystems,  both  land and aquatic, largely amount to
the  valuation  of recreation opportunities  in  natural  surroundings.   Recrea-
tion  might  include  hunting,  fishing, camping,  hiking,  boating,  etc.   In
addition, there may  be  an "option value"  associated with the preservation of
natural environments, which will be discussed later.

     Ashton, Wykstra,  and Nobe  (1974)  discuss six  recognized approaches to
the  valuation of non-market supplied recreational  opportunities.   Although
some of these approaches are seriously flawed,  they are included here because
the  reader  may find them in use  and should therefore be  able  to recognize
them and understand their limitations.  They are as follows:

      1.   The expenditure method:  Measures  the value  of recreation in
          terms of the  total direct  private expenditures on recreation;

      2.   The gross-national-product method:  An attempt  to measure the
          contribution of recreation to GNP;

      3.   The consumers'-surplus method:   Determines  the  willingness of
          individualstopayforvarious  quantities   of recreation;

      4.   The cost method:   Uses  the cost of  supplying  recreational
          facilitiesas  a  measure of  the benefits  derived  therefrom;

      5.   The market-value method:  Is based on  fees charged at private
          resorts  as   a   proxy   for  the  value   of  public-supplied
          facilities;

      6.   The monopoly-revenue method:  Relies on  the  estimated revenue
          that  would  be  obtained  by a  nondiscriminating  monopolist
          owning a recreation site as a measure of the benefits.

                                     200

-------
     The  most  Important approaches  are  the  expenditures method  and  the
consumers'-surplus method.  The  expenditure  method is simple to app'y and is
widely used;  Ashton,  Wykstra,  and Nobe (1974) cite a variety of studies that
have  resulted in valuations of  $3.89  to  $28 per recreation day  (in  1952 to
1966  dollars).   One  approach to  the  consumers'-surplus  method  is  to  ask
consumers what  they  are  willing to pay  for  the  recreational  opportunities.
While  attractive  theoretically,  as  already  indicated,   this  approach  is
fraught with  practical  difficulties.   Alternatively,  in both the consumers'-
surplus and  monopoly-revenue  methods, travel distances, and hence differen-
tial  travel  costs can be used to  impute  a value  (or  rent)  for rec-eaiional
opportunities.

     The cost method  is conceptually unsound and is rarely used nowadays; but
its  successor,  the market-value method,  is difficult to  apply in sractice.
The appropriate version of the gross-national-product method involves  tracing
the contributions  to  regional  or national income that arise from tha  "screa-
tional activities  of  interest.   Thus this method  includes secondary  effects
and would  be inconsistent  with  other valuations used in  this  study.   Note,
however, that each method  discussed here is  not merely  a  different  way of
arriving at  the same  result,  but  it  is  in  fact a different measure  of the
value of recreation.   Fortunately,  the consumer-surplus and monopoly-revenue
methods  generally  give   results  that  are  similar  in  magnitude  to  those
obtained by  the  expenditure method (see Ashton, Wykstra and Nobe,  1974), so
that the significant  philosophical  differences  need not cause too many prob-
lems in practice.

     Other useful  discussions  on the valuation of recreation may be found in
Clawson  (1972),   Clawson and  Knetsch  (1966),  Krutilla  and Fisher  (1975),
Krutilla (1972),  Edwards et  al.  (1976),  and  Pearse  (1968).   Most work has
concentrated  on  the  valuation of  a recreation-day, as this  is most  readily
measured.    However,   in  many  situations  it  may be desirable  to be able to
value individual  species  (this  is  particularly useful for fish as it  permits
fish  kill  data  to be used).   For game  animals  (known as  "consumative" or
"harvested" species),  it is common to deduce an average animal  value from the
number of hunter-days (often estimated from license data),  the average expend-
iture or some other measure of the  value  of a  hunter-day, and the number of
animals harvested  (Ashton,  Wykstra, and Nobe,  1974).  Valuations  yielded by
this  procedure  have  been  used within  a  cost-benefit analysis  framework to
determine  the cost  of destroying  areas  of wildlife  habitat,  based  on the
number of  animals that  the habitat would support (Norman  et al.,  no  date).

     Unfortunately, the  average  animal-value approach has some serous draw-
backs.  First,  for the  purpose  of evaluating  any  specific  wildlife  kill,  a
marginal value should ideally be used.  Second,  it values animals as  though
they  were  harvested and does  not  take account  of the need  to  propagate the
species  for  future years.    Finally,  it cannot  be applied to nonharvested
species.

     There have been  some attempts to overcome these difficulties.   Brown and
Hammack (1972)  have developed a questionnaire-based  methodology (applied to
migratory waterfowl)  to  derive a marginal value for game animals that recog-
nizes the  variability of hunting success and the  effort  of constraints such


                                     201

-------
as the  bag  limit.   Inevitably,  a major difficulty associated with determina-
tion  of  marginal   rather than  average  values  is  the  more  complex  data
requirements; this is probably why most analysts use average values.

     The  need to  propagate  the  species  is  usually  taken into  account  by
valuing breeding areas  or young in terms of  the  yield to harvestable adults
and the value of  these adults.   For example,  Brown and Hammack (1972) derive
a  valuation  for nesting  areas  based on  survival probabilities  and hunting
yields, while the Oregon Department of Fish and Wildlife (1977) has developed
similar methods  for  valuing spawning  gravel and  brood  fish  in anadromous
fisheries.   This  approach does  not take account of  changed  competition for
food,  or   the effects  of  reduced  populations   on   hunting  activities  and
yields; but   to  attempt  to  do  so  would  indeed  be a  profound undertaking.

     Valuation  of   nonharvested  wildlife   presents   greater  difficulties.
Though  similar  methodologies to  those  used  for  arriving at the  value  of a
hunter-day can be used to determine the value of a recreation-day, the valua-
tion of specific  species  is  very difficult.  As  far  as the author is aware,
nonharvestable species valuation has only been attempted once (Norman et al.,
no date)  and in  this case, the values were derived by reviewing the harvest-
able values  and making subjective comparisons.   Thus in this case, the value
of a harvested  black bear was established  at $6,400;  the vali-e of a grizzly
bear (which  is not harvested) was subjectively set at $20,000 on the basis of
comparison with the  black bear.   (Telephone  interview, R.L. Norman. Colorado
Division of Wildlife, Grand Junction, Colorado,  November 21, 1977.)

Fish and Other Aquatic Life Kills in Surface Waters

Damage to Noncommercial Fishing, Etc.--
     The  considerations that apply  to the valuation of noncommercial fishing
and  to  water-related  recreation  have  already   been discussed  under  the
valuation of wildlife, etc.

Damage to Commercial Fishing-
     While  land-based crop  and  animal  losses can  reasonably be  valued  at
market prices, valuation  of  lost commercial  fishing is more difficult, as in
this  case,  resources  are   primarily  expended in harvesting  the  fish  (as
opposed to raising  the crops or animals) and therefore, it would be possible
to arrive at a  zero valuation  for reduced  fishing  yield,  if  the resources
could  be  employed  in  alternative  uses  that were  equally productive.   In
practice,   the resources  are  likely to  be immobile, except in  the very long
run,* and Brown et al. (1976) concluded that  the benefits of increased yields
  Gordon (1972) argues that the operation of a typical competitive fishery is
  such as to yield no net economic rent, and that some fishing grounds may be
  exploited  at a level  of negative marginal  productivity.   He  states  that
  fisherman  are  one  of  the least mobile of occupational groups and they will
  work  for  less  than  the "going  wage."   Bromley  (1969) confirms  the low
  mobility  of  resources and  argues  that some of the unique  features of the
  commercial   fishing  industry  justify  the  observed   labor  immobility  on
  grounds of social efficiency.  Much additional information on the economics
  of  fisheries may be  found  in  Peterson  and Fisher's  (1977)  survey of the
  economics of extractive resources.

                                     202

-------
could  be  valued at  market prices to  the fishermen plus  an  increase in the
consumers' surplus.  The  change  in the consumers'  surplus arose because the
demand  for  fish (salmonids  in  this case)  was assumed  to be  local,  and an
increased catch would cause a decrease in price.

     Similar  arguments  could  be  applied  to  decreases  in  yield  due  to
hazardous waste incidents,  but  unless the  incident  was  such  as to  have  a
substantial  effect on the local  catch,  the decrease in the consumers'  surplus
could  be  neglected.   If the physical  effects were  permanent,  the  loss to
society could probably  be restricted to several years  or at the most to one
generation,  as fishing resources would not be immobile forever.

The Impacts  of Ocean Dumping--
     Valuation of  the  environmental  impacts of ocean dumping is made diffi-
cult by the  diversity of possible effects (see Appendix B).   It may, however,
be  useful to identify  and  value  the activities that are  at  risk from ocean
dumping.  These include commercial and sport fishing,  ocean recreation oppor-
tunities  (skin  diving,  swimming,  etc.), and general  tourism.  The valuation
of  fishing   and recreation  has  already  been  discussed; tourism   is  most
commonly valued in terms of its  commercial impact by using tourists'  expendi-
tures  as  a  base.   Because tourists'  expenditures are  usually  assumed  to be
exogenous to the area,  secondary (multiplier) effects are commonly included.
In  general,  this  study  disregards  such  effects,  but  the  local  impact of
multiplier effects probably  should be  noted when lost tourism'is considered.

     Some estimates  of  economic  damages from  actual  hazardous  materials and
oil  spills  are  available.   For  example,  cleanup  costs for  oil spills have
ranged  from $0.13  to $4 per liter (Enk, 1974).  However, available estimates
have rarely, if ever, attempted to evaluate all costs.

Changes in Property Values

     The  concept  that  pollution will  affect  land values and   that  land or
property values can  thereby be  used as- some measure  of the public's percep-
tion of pollution  has  received  much attention from environmental economists.
Unfortunately,  analysis of empirical  data is far from simple and has met.with
only  limited  success (Fisher and Peterson, 1976).  Most  attention  has been
paid to the  impact of air pollution on property values, it being argu&d that
property value differentials provide an indication of individual-willingness-
to-pay  to reduce  pollution.   The theoretical  underpinning and  the practical
problems  associated  with  the use  of .land .rents  or-property values..-as  a
measure of  external  effects are  described by  Schmalensee  et  al.  (1975).
However, there  remains  some debate as to  exactly  what  can be  inferred about
pollution from property value-data (see Polinsky and Shave!1, 1975).

     The  results  of several  empirical.,studies, of  air  pollution on property
values  are  discussed by  Waddell  (1974).  .An  indication of  the magnitude of
the  effect  is  provided  by Waddell's conclusion  that an  increase of 0.1 mg
S03/100 cm2 per  day  (which would represent a doubling of the background rate
for sulfation in the U.S.) would decrease residential  property values by $100
to $600.
                                     203

-------
     Tihansky  (1975)  cites  some  studies that  attempted to  relate  property
values  to  water  pollution.   Unfortunately  none of  these  studies  provided
clear-cut results that could be helpful in evaluating hazardous waste manage-
ment problems.

     Schmalensee  et  al.  (1975) endeavored  to assess the  impact  of  sanitary
landfills  on  residential   property   values.    In  a  study  of  values  of
residential  property  adjacent  to  four  sanitary  landfills  in  Los  Angeles
County,  they  found a  general  lack of  significant  detrimental  environmental
effects.   In  one case truck noise associated  with the fill  had a  negative
effect  on  property  values,  but  both  proximity and  view  of  the  fill  had
positive effects, presumably because of the anticipated transformation of the
fill to a  recreational  site.   Note,  however, that these  results related to
well operated  sanitary  landfills,  where  air and  water pollution  were  not
problems.

     Except  perhaps  where  there  is  severe  air pollution,   it  appears  that
changes  in property  values  are  quite  local.   For sanitary  landfills,  the
effects  do not  appear  to extend beyond  one  or  two  miles (see Schmalensse et
al., 1975).   Thus  this  aspect  of the adverse  impacts of  hazardous  waste
management techniques  can  be minimized by locating any site  well away  from
residential >and   other  susceptible  properties.   As  landfills  ^.sed  for  the
disposal of  hazardous waste are  unlikely  to  be converted  to recreational
sites,   increases  in property values would not be anticipated.

     Though  property values  may reflect actual  physical pollution,  they may
also reflect psychological  influences.   The fear that property values may be
reduced  is frequently  encountered when  landfill sites  are being considered,
and local citizens will often oppose the establishment of a landfill  for this
reason  (U.S.  Environmental  Protection Agency, 1976).   Despite the safeguards
used in  a well operated chemical landfill or other waste management facility,
the situation may be exacerbated when it comes to hazardous waste operations,
reflecting a greater degree  of perceived threat.  The author is aware of two
instances in which industrial waste disposal sites were closed as a result of
public  pressure.   While  incidents  of this type are complex,  it does appear
that concern about property values was involved  in both cases.*
*  The Antioch,  California,  landfill  and  lagooning  facility operated  by a
  subsidiary  of Industrial  Tank Company was  closed as  a result  of  public
  pressure  after  a  housing  subdivision  was  built adjacent to  it.   The
  residents of  the  subdivision  complained of air pollution but  continued to
  press  for  closure  after  the  lagoons  had been filled  in and  cappe:.   The
  same company  operates two  other  treatment and  disposal sites  in  the San
  Francisco Bay  area, but  after strongly contested public hearings, the firm
  was unable  to  secure a land use permit for a proposed site near Brentwood,
  California.  (Personal interview,  C.G.  Schwarzer, California Department of
  Health, Berkeley,  California,  February 7, 1977.)   To avoid a repetition of
  the Antioch situation, the company has obtained control of a buffer zone of
  several thousand acres (used for agricultural  purposes) around its 280-acre
                                     204

-------
     Changes in property  values  are of particular importance with respect to
equity  since  changes that  are  caused by  a hazardous waste  activity  can in
part be regarded  as  transfer  payments.    (These  transfer  payments would be
between  those  whose  property  values  had  fallen  and  those whose  property
values  had  risen; and,  to the  extent that  the  price  changes did not reflect
changes, in utility between a seller and a buyer of a  property.)  In addition
to  these transfer  payments,  there  is  the  overall  external  effect  of  the
hazardous waste management  activity  on the welfare of  the  householders.   An
adverse  impact  will, in  aggregate,  reduce the utility of  the  properties to
the householders,  thereby decreasing welfare.

     To  summarize:   One  can  be  reasonably sure  that some  hazardous  waste
management techniques will—like many industrial activities—affect the value
of  adjacent  properties.   These  changes  in value are  indicative  of external
effects  associated  with  the hazardous waste management activities,  but they
only  provide a  true  measure  of  the external  effects under  ve-y  limited
circumstances (Schmalensee et al.,  1975;  also  see Fisher and Peterson,  1976).
In practice, changes in property values are largely important with respect to
equity.

Aesthetic Factors  and Option Values

     This category  of  effects  is included to cover individuals'  values that
do  not  normally enter the marketplace.   The effects  are highly indirect; an
individual does not necessarily  have to observe the environmental feature of
interest  to possess  these values.   They may  conveniently be  divided into
aesthetic value, existence  value,  and two types of option  value.   There are
no zero levels of  aesthetic and existence values, so for these factors, costs
must be expressed in terms of  differences.  Option values,  however,  do have
zero  levels,  and  hence  it is possible  to use  absolute  values  if desired.

     Aesthetics may enter  into  the  valuation  of  most of  the  environmental
effects already discussed—for example,  property values and the willingness-
to-pay  for  recreation  opportunities  partly reflect aesthetic considerations.
However,  there  may   be  situations  where-only -aesthetic values  are involved,
such as the  attractiveness  of  a river with clear  water as opposed to turbid
water,  and the discussion here, is intended to  cover those situations.
  Benica,  California,  site.    (Personal  interview,  F.  Balisteri,  Pacific
  Disposal Systems, Inc., Martinez, California, February 3> 1977.)
       Another  industrial  waste   disposal   site  near  Pasco,  Washington,
  operated  by Resource  Recovery  Corporation was  closed  after  the Franklin
  County Commissioners refused to renew its land use permit.   The company has
  attempted  to  establish operations elsewhere in eastern Washington  but has
  met with  opposition to  its  proposals (Stradley,  Dawson,  and Cone,  1975).
  The Pasco  site closure  occurred after a newspaper had suggested that its
  handling of the herbicide 2,4-0 could threaten local grape crops.   However,
  it  was  later demonstrated  that  no  such  physical  damage  had  occurred.
  (Personal  interview,  T.  Cook et  al.,  Washington Department  of Ecology,
  Olympia, Washington, March 8, 1977.)
                                     205

-------
     Since  changes  in these  nonmarket values  between  different hazardous
waste vmanagement schemes  are  difficult to measure, the  analyst may in some
cases simply wish to identify these costs rather than place a dollar value on
them.  However,  there  are  techniques that can permit values to be  assigned to
these costs, and these are discussed below.

Aesthetic Value—
     The  aesthetic  value  of  concern here is what  might  be termed "environ-
mental  aesthetics,"  as  opposed  to  the  aesthetics  of   art,   design,  city
planning,  etc.   Thus  it is  the  value  that  an  individual  places  en  the
physical  environment  to  reflect  that  individual's  taste  and  desire  for
beauty, and for  natural surroundings, etc.

     No  direct  costs  can be  associated  with most  changes that offend or
please a  person's  environmental  aesthetic values.  Virtually the  only «ay of
placing a dollar figure on this type of aesthetic value would be some measure
of willingness-to-pay, derived either directly from questionnaires or bidding
games,  or  indirectly  from  studies  of  human  behavior  such  as  the  travel
studies used to  value  recreation benefits.

     Some  authorities  propose  the  use  of  a Delphi  technique  to  estimate
aesthetic  damages  (Kennedy et al.,  1976;  CONSAD  Research Corporation, 1975;
Maler and  Wyzga, 1976).   Delphi  techniques are widely  used in  technological
forecasting and  to some extent in technology assessment.  The method involves
the  use  of a panel of experts who, through a controlled  interactive process,
attempt 'to  reach  consensus  in  answering  certain  questions   that usually
involve prediction.   For more details of the Delphi method, see Linstone and
Turoff (1975).   Though the Delphi method might have some promise  for identi-
fying different  aesthetic considerations,  the use of the opinions of a panel
of experts as  a  surrogate  for those of the public is questionable.

Existence Value-
     In addition to the aesthetic value of the environment to an  observer of
that  environment,  there  may  be  an  existence  value associated with  the
environment.   This  is the value that  some' persons  place on  knowing that
something  exists,  even  if they  never  expect to see  it  or benefit from it
(Fisher  and Peterson,  1976).  This value  is  largely  associated  with major
irreplaceable  features of the environment (e.g., the  Grand Canyon)  and with
rare  or  endangered   species.   Existence  value  can  include   the  possible
-scientific  benefits  from preserving  a  -species, an  ecosystem  or a natural
feature,  and also  includes  an aesthetic  aspect  that  values diversity, etc.

     A good example of existence value was provided by one of the  comments on
the  Environmental  Impact Statement  for  the  prototype  Federal  oil  shale
leasing program.   It appeared to come from an old man in Ohio, who evidently
did  not  want some of  the  canyons in the vicinity of the  oil shale formations
destroyed  by  depositing spent shale in them.   He said, "I have never seen a
canyon,  let  alone those in Colorado, Wyoming,  and  Utah.  .  .  .  please do not
destroy  the beauty of what God  put there ... if it  cost  more to  keep the
land and animals I would  rather spend the extra money" (Burn's,  1973).
                                     206

-------
     Another  illustration  of existence  value  was the drive  to  preserve the
only known  habitat  of  the  snail darter  on  the  Little Tennessee Rive'", which
halted a major dam project (Cook, Cook, and Grove, 1977).  In this case, some
individuals believed  that  the  preservation of this  species was  worth more
than  the  tens  of  millions  of  dollars  irretrievably  committed to  the dam
project.   Note, however, that in this case,  the costs of preserving the snail
darter would  not  fall  on those advocating preservation.   This is  akin  to
demanding compensation for an environmental  change, whereas the first example
illustrates genuine willingness-to-pay.

     Determination of existence values again involves attempting to ascertain
the  individual's willingness-to-pay  to preserve that  existence,  which  is
probably even more difficult to accomplish than determining aesthetic values.
Nevertheless,   confrontations  such  as   the  dispute  over  the   snail  darter
illustrate the very high existence  values that some people attribute to rare
environmental  features.  •

Option Values--
     Option value  refers  to the  benefits of  keeping an  option  available.
Like existence value,  it  is  only relevant to irreversible actions t!iat fore-
close some  future use  of  the feature of interest.  The best known example is
probably Krutilla's analysis of the  benefits of. maintaining Hell's Canyon-(on
the Oregon-Idaho  border)  in  its natural state (Krutilla  and Fishe", 1975).-
Key aspects  of the  analysis are that  a decision to dam  the Snake  River  in
Hell's Canyon would be  irreversible,  and that in the future the value of the
recreational benefits conferred by maintaining the river in its natural state
is  likely  to  increase, whereas technological  advance  (-in alternative'power
sources) is likely  to  reduce the real value of hydroelectric power generated
by the dam(s).   Thus  in this case,  the-implications  of technical  change are
said to  be "asymmetric" (because  the supply  of  wilderness  areas  cannot  be
increased)  (Krutilla,   1967),  and   maintaining  a  natural  environment  is
expected to make an important contribution to man's future welfare.

     There  has  been some  debate as to  exactly what  option value  "epresents
and how  it differs from the consumers1'  surplus.   It  is .generally held.-that
option value  can take  two  forms:   One  (often called  "option  demand") that
represents  a  risk  aversion  premium associated with  uncertain  future demand.
(and which  is. in addition  'to consumers'  -surplus'), whereas the other reflects
the benefit  of not  undertaking an  irreversible  project  while  society waits
for  improved   information  about the -benefits of alternative-  uses  of  our
resources (usually  the  environment) (Fisher and Peterson, 1976).  It is this
second form of  option  value  that has  potentially important implications for
hazardous waste  management.   As Arrow and  Fisher (1974)  point  out,  where a
pollutant is nondegradable  (and may  therefore have irreversible effects), any
decision about  its  discharge to the environment should take into account the
option value  associated with  delaying any  action until more data  about its
effects and about  alternatives are  obtained.    This option  value   must  be
balanced against any increased costs associated with the delay.

General —
     While  it is  easy to postulate  existence, values  and option demand,  a
pervasive difficulty with  such nonuser effects is to determine the number of


                                     207

-------
people  who  hold these values.  These numbers can be very  large--for example,
the  Alaska  pipeline issue must have been  known to a large proportion of the
U.S.  population,  and  hence  even  if only a  small  percentage  of those people
were concerned about  the potential damage  to  the environment,  the nonuser
benefits-of  environmental  preservation could have  been  substantial  (see
Bishop  and  Cicchetti,  1975).

      Another  aspect of aesthetic  values, existence values, and option demand
arising from  risk aversion that should be mentioned Is that values are likely
to  vary substantially  among individuals.    It  is  even  possible to postulate
changes 'that  would  provide a positive aesthetic  benefit to some individuals
and  a  negative one  to others.   Consider,  for example,  the  choice between
leaving land  as a wilderness and farming it.   Some people may prefer wilder-
ness because  they see farming as the introduction of a  nonnatural ecological
monoculture.   Others  may prefer to see  the  land  farmed, associating faming
with a  satisfying,  traditional  way  of  life.   While  individuals  may place
differing  valuations  on  market-traded  goods (giving  rise to the consumers'
surplus),  in  most instances the variation is probably larger with nonmarket-
traded   goods,  e.g.,   recreational  oppportunities.    If  the  results  of
willingness-to-pay  surveys  on improving the environment can be trustad, this
variation.can be  very large  (several orders of magnitude) for aestnefic and
existence values.   (See  Appendix D.)

      This  variation in  individuals'  values  raises  some difficult questions
•for  decisicomakers.   These questions  essentially revolve  around equity.  For
example,  to'  what extent  is it  reasonable   to provide  or deny  very large
.benefits to a'few individuals, when the  benefits to most others are small?  A
similar problem arises where the individual's  true willingness-to-pay (or to
take avoiding action) is limited by a low discretionary income.  If this low
willingness-to-pay  is  accepted  at its  face  value,  we  may  be  making  the
judgment  that society should value the welfare of that  person less than that
of  an  affluent person.   This type of  implication  should be considered when
alternative policy  approaches are evaluated.

REFERENCES

Arrow,  K.J.,  and A.C. Fisher.  "Environmental  Preservation, Uncertainty, and
      Irreversibility."   Quarterly Journal  of Economics.  88(2):312-319, 1974.

Arthur  D.- Little, Inc.   Alternatives  to the Management of  Hazardous Wastes
      at National" Disposal Sites.2  volumes.EPA/530/SW-46c, U.S. Environ-
      mefltal  Protection  Agency,  Washington,  O.C.,  1973.   [NTIS:   PB225164,
      PB237264]

Ashton, P.M., R.A.  Wykstra,  and  K.C. Nobe.  Optimum  Supplies  of Recreation
      Days  Under Conditions of Uncertainty:   A Case Study Application to Wild-
      life  Resources.Department  of Economics,  Colorado  State University, Ft.
      Collins, Colorado,  1974.  98 pp.

Battelle  Memorial Institute,  Pacific Northwest  Laboratories.  Program for the
      Management of  Hazardous Wastes.  2  volumes.   EPA-530/SW-54C,  U.S. Envi-
      ronmental   Protection   Agency,    Washington,   D.C.,   1974.    [NTIS:
      PB233630/1]

                                     208

-------
Bishop, J.,  and C. Cicchetti.   "Some Institutional and Conceptual Thoughts on
     the  Measurement  of  Indirect  and  Intangible  Benefits and  Costs."   In:
     Cost  Benefit  Analysis  and  Water  Pollution  Policy, H.M.  Peskin and
     E.P. Seskin,edTTTheUrbanInstitute,Washington,   D.C.,   1975.
     pp. 105-126.

Brewer, S.T. "Quantification  and  Comparison of External Cost  of Nuclear and
     Fossil  Electrical  Power  Systems."   In:   Energy  and  the  Envronment:
     Cost-Benefit  Analysis.  R.A.   Karam,   and  K.Z.   Morgan,eds.Pergamon
     Press,  New York,  New York,  1976.   pp.  506-530.

Bromley, D.W.   Economic  Efficiency in Common Property  Natural Resojrce Use:
     A Case  Study  of the Ocean  Fishery.   Working  Paper No.   28,  Bureau of
     Commercial  Fisheries,  Division  of  Economic  Research,   Oregon  State
     University, Corvallis, Oregon, 1969.  174 pp.

Brown,  G.M.,   Jr.,  and  J.  Hammack.    "A   Preliminary  Investigation  of  the
     Economics of Migratory Waterfowl."   In:   Natural Environments:  Studies
     in  Theoretical   and  Applied  Analysis. J~Krutilla,  ed.The  Johns
     Hopkins  University  Press,  Baltimore,  Maryland,  1972.    pp. 171-204.

Brown, W.G.,  D.N.  Larson,  R.S.  Johnson, and  R.J. Wahle.   Improved Economic
     Evaluation of Commercially and Sport-Caught  Salmon and Steel head of the
     Columbia  River.   Special  Report 463,  Agricultural  Experiment Station,
     Oregon State University,  Corvallis, Oregon, 1976.  33 pp.

Burn's, T.  "Letter No.  90."   In:   Final  Environmental  Statement  for  the
     Prototype  Oil  Shale Leasing Program.Volume V.U.S. Department of the
     Interior.    (TT  Government   Printing  Office,  Washington,  D.C.,  1973.

Clawson,  M.    "Methods  of  Measuring the  Demand  for and  Value of  Outdoor
     Recreation."  Paper  presented at a meeting  of  the Taylor-Hiboard Club,
     University  of  Wisconsin,  January  13, 1959.   Resources  for  the  Future
     Reprint #10.  Washington, D.C., 1972.   10pp.

Clawson,  M.,  and J.L. Knetsch.   The  Economics of Outdoor  Recre&tion.   The
     Johns Hopkins Press,  Baltimore, Maryland, 1966.348 pp.

CONSAD   Research  Corporation.    Cost/Benefit   Techniques  for   Hazardous
     Materials.  Toxic  Substances,  Solid  Waste  Control.ContractRbT
     68-01-3116,  U.S.  Environmental   Protection   Agency,  Washington,  D.C.,
     1975.  383 pp.

Cook.  S.G.,  C. Cook, and D.  Grove.   "What They  Didn't  Tell   You  About the
     Snail Darter and the Dam."   National  Parks  and Conservation  Magazine.
     51(5):10-13, 1977.                                         '

Council  on   Environmental  Quality.   Environmental  Quality -  1976:  Seventh
     Annual  Report.   U.S.  Government Printing Office, Washington, D.C., 1976.
     398 pp.
                                     209

-------
Downing, P.B., and J.N.  Kimball.   Enforcing  Environmental Regulations. Work-
     ing Paper  No.  CE-77-3-3, Department of  Economics,  Virginia Polytechnic
     Institute and  State University, Blacksburg,  Virginia,  no date.   50 pp.

Downing, P.B.,  and W.D.  Watson,  Jr.  "Economic  Considerations  in Enforcing
     Environmental    Controls."     Advances  in  Environmental  Science  and
     Technology.    Volume 7.   J.N.  Pitts,  Jr.  and  R.L.  Metcalf,  eds.   John
     Wiley and Sons, Inc., New York, New York, 1977.  pp. 423-514.

Downing, P.B., and  W.D.  Watson,  Jr.  "The  Economics  of  Enforcing Air Pollu-
     tion  Controls."   Journal  of  Environmental   Economics  and Management.
     1:219-236,  1974.

Edwards, J.A.,  K.C.  Gibbs,   L.J.  Guedry,   and  H.H.  Stoevener.   The  Demand
     for  Non-Unique  Outdoor  Recreational  Services:   Methodological Issues.
     Technical Bulletin  #133, Agricultural Experiment Station,  Oregon Stats
     University,  Corvallis, Oregon, 1976.   60 pp.

Enk, G.A.  "The Social  Impacts of Hazardous Materials Spills."  In:  Control
     of Hazardous  Materials Spills:  Proceedings of the 1974 National Confer-
     ence on  Control   of  Hazardous  Material Spills.  American Institute of
     Chemical Engineers  and  the  U.S.  Environmental  Protecticn  Agency,  San
     Francisco,  California, August 25-28,  1974.   pp. 143-148.

Fisher, A.C., and  F.M.  Peterson.   "The Environment in Economics:  A Survey."
     Journal  of Economic Literature. 14(1):1-33, 1976.

Freeman, A.M., III.   "Benefits of Pollution Control."  In:  Critical  Review
     of Estimating  Benefits of Air and Water Pollution Control. A. hershaft.'
     ed. EPA  600/5-78-014,  U.S.  Environmental Protection Agency, Washington,
     O.C., 1978.

Garwood, F.,  and  R.F.  Newby.   "Use of Chi-squared Distribution for Comparing
     Accident Frequencies."  In:   Road Research. Proceedings of the Symposium
     on the  Use  of  Statistical  Methods in the Analysis of Road Aceicents.
     Road Research Laboratory, Crowthorne,  England, 1969.

Gordon,  H.S.   "The  Economic  Theory  of a  Common  Property  Resource:   The
     Fishery."    In:    Economics  of  the  Environment:   Selected  Readings,
     D. Dorfman, and  N.S.  Dorfman, eds.   W.W.  Norton and  Company,  Inc.,  New
     York, New York, 1972.  pp. 88-99.

Hirshleifer,  J.,   T.  Bergstrom,   and  E.  Rappaport.   Applying  Cost-Benefit
     Concepts to Projects Which Alter Human Mortality.  UCLA-ENG-7^78, School
     of  Engineering  and  Applied  Science,  University  of  California,  Los
     Angeles, California, 1974.

Hub,  K.A.,  J.C.   Asbury,  W.A.  Buehring,   P.F. Gast,  R.A.  Schlenker,  J.T.
     Weil Is, K. P.  DuBois, and J.L. Gardner.   Social Costs for Alternate Means
     of  Electrical   Power   Generation  for  1980  and  1990.    4 volumes.
     ANL-8093,Argonne  NationalLaboratories,Illinois,1973.   1050  pp.
                                     210

-------
Kennedy, R.,  R.  Lowrey,  A.  Bernstein, and F. Rueter.  A  Benefit-Cost System
     for  Chemical  Pesticides.  EPA 540/9-76/001, U.S. Environmental Protec-
     tion Agency, Washington, O.C., 1976.  322pp.   [NTIS:  PB250988]

Krutilla,  J.V.   "Conservation  Reconsidered."   American  Economic  Review,
     57(4):777-786, 1967.

Krutilla,  J.V. ,   ed.    Natural  Environments:  Studies   in  Theoretical  and
     Applied  Analysis.   The  Johns  Hopkins  University  Press,  Baltimore,
     Maryland, 1972.362 pp.

Krutilla, J.V.,  and A.C.  Fisher.   The  Economics  of  Natural  Environments:
     Studies in the Valuation of Commodity and Amenity Resources"The Johns
     Hopkins University Press, Baltimore, Maryland,  1975.  214 pp.

Linnerooth, J. 'A Critique of Recent Modelling Efforts to Determine ~:he Value'
     of Human LifT!RM-75-67,International  Institute  for  Applied Systems
     Analysis, Laxenburg, Austria, I975a.  51 pp.

Linnerooth, J.   The  Evaluation of  Life-Saving;   A Survey.   RR-75-21, Inter-
     national  Institutefor  Applied  Systems  Analysis,  Laxenburg,  Austria,
     1975b.  31 pp.

Linnerooth, J.  "Methods for Evaluating Mortality Risk."  Futures.  8(4):293-
     304, 1976.

Linstone,  H.A.,  and  M.  Turoff,  eds.   The  Delphi  Method: . Techniques  and
     Applications.  Addison-Wesley  Publishing  Co.,  Reading,  Massachusetts,
Applli
19757
Maler, K.G., and R.E. Wyzga.  Economic Measurement of Environmental Damage; A
     Technical Handbook.  Organisation  for  Economic Cooperation and Develop-
     ment, Paris, France, 1976.  151 pp.       •                         .  •  -

Mishan,  E.J.   Cost-Benefit   Analysis:   An Introduction.   Praeger Pjblishers*
     New York, New York, 1971a.  374 pp.

Mishan, E.J.  "Evaluation of Life antf Limb:   A Theoretical Approach." Journal
     of Political Economy. 19(4):687-705, 1971b.                 •          ~~

Moll,  K.,  S.  Baum,  E.  Capener,  F.W.. Dresch,. and R"M.  Wright.   Hazardous
     Wastes:  A	Risk-Benefit  Framework  Applied  to  Cadmium and  Asbestos.
     EPA-600/5-77-002,U71TEnvironmental   ProtectionAgency,Washington,
     O.C., 1977.  268pp.  [NTIS:  PB257951]

National Academy of Sciences.  Considerations of Health Benefit-Cost Analysis
     for  Activities  Involving Ionizing Radiation Exposure and Alternatives.
     EPA 520/4-77-003,  U.S.  Environmental Protection Agency,  1977"199 pp.

National Academy of Sciences.  Decision  Making  for  Regulating. Chemicals in
     the Environment.  National  Academy  of  Sciences, Washington, O.C., 1975.
     242 pp.
             ' *

                                     211

-------
Norman,  R.L ,  L.A.  Roper, P.O. Olson, and R.L. Evans.  Using Wildlife Values
     in  Benefit/Cost  Analysis  and  Mitigation of  Wildlife Losses.  Colorado
     Division of Wildlife, Denver, Colorado, no date.  21  pp.

Oregon  Department  of  Fish  and Wildlife.   "Economic Data Used  by Fishery
     Biologists."   In:   Manual  for   Fish  Management.   Oregon Department  of
     Fish and Wildlife,  1977:pp. 168-192.

Otway, H.J.,  and  J.J.  Cohen.   Revealed   Preferences:  Comments  on  the  Starr
     Benefit-Risk   Relationships"!    RM-75-5,   International  Institute   for
     Applied Systems Analysis,  Laxenburg, Austria, 1975.   17 pp.

Pearse,  P.H.   "A New  Approach to the  Evaluation of Non-Priced Recreational
     Resources.-"-  Land Economics, 44(l):87-99, 1968.

Peterson, F..M., and A.C. Fisher.  "The Exploitation  of Extractive Resourcas:
     A Survey."  The Economic  Journal. 87(348):681-721, 1977.

Polinsky,  A.M.',  and  S.  Shavel.l.   "The Air Pollution   and  Property  Value
     Debate."   The Review of  Economics and Statistics.  57(1):100-104,   1975.

•Prato,  A.A.  "Internalizing  Environmental  Externalities  of Erargy  Resourca
     Developments."  In:   Economics  of Natural   Resource   Develccnent  ^n  tha
     West..  Human Resources and-Natural  Resources  Committees  or' ins .vestarn
     Agricultural Economics  Research Council,  1974.   pp. 51-67.

Randall, A.,  B.C.  Ives,  and C. Eastman.  Benefits of Abating Aestnetic  Envi-
     ronmental  Damage   from the  Four  Corners   Power  Plant.   Agricultural
     Experiment Station  Bulletin 618, New Mexico  State University, Fruitland,
     New Mexico, 1974.

Schmalensee,  R.,   R.  Ramanathan,  W.  Ramm,  and  D. Smallwood.    Measuring
     External  Effects   of   Solid  Waste  Management.   EPA-600/5-75-010, U.S.
     Environmental  Protection  Agency, Washington, D.C. , 1975.  448 pp.

Schelling,  T.C.   "The   Life You Save May  Be Your  Own."   In:   Problems   in
     Public Expenditure  Analysis.  S.B.  Chase,  Jr.,   ed.   The   BrooKings
     Institution, Washington,  D.C., 1968.  pp. 127-176.

Slovic,  P.,  B.  Fischhoff,  and S,  Lichtenstein.   "Cognitive" Processes  and
     Societal -Risk  Taking."    In:  Cognition and  Social Behavior. J.S.Carroll
     and J.W.  Payne,  eds.-  .Lawrence Erlbaum  Associates,   Potomac,  Maryland,
     1976.  pp. 165-184.

Smith, D.D., and  R.P. Brown.   Ocean  Disposal  of Barge-Delivered  Liquid  and
     Solid  Wastes   from  U.S.   Coastal   Cities.   SW-19c,   U.S.  Environmental
     Protection Agency,  1971.   128 pp.

Starr,  C.   "Benefit-Cost Studies  in  Sociotechnical   Systems."   In:   Perspec-
     tives  on Benefit-Risk Decision Making.  National Academy of Engineering,
     Washington,  D.C. ,  1972.   pp. 17-42.
                                      212

-------
Starr,  C.   "Social Benefit  Versus  Technological Risk:  What  is Our Society
     Willing to Pay for Safety?"  Science. 165(3899):1232-1238, 1969.

Stradley, M.W., G.W. Dawson, and B.W. Cone.  An  Evaluation  of the Status of
     Hazardous Waste Management in Region X.  EPA 910/9-76-024, U.S. Environ-
     mental Protection Agency, Seattle, Washington, 1975.  202 pp.

Talley, R.J., and O.W. Albrecht.  "Preliminary Economic Analysis of Hazardous
     Waste Control."  In-house research project, Program Element IDA312, ROAP
     OIAOC.   U.S.  Environmental Protection  Agency, Cincinnati,  Ohio,  1974.
     79 pp.

Thaler, R., and R.  Rosen.   Estimating  the Value  of Saving a Life:  Evidence
     from  the  Labor  Market.   Preliminary  report  for  the  Conference  on
     Research  in   Income  and Wealth,  National  Bureau of  Economic Research,
     Inc., New York, New York, 1973.

Tihansky,  O.P.  "A Survey of Empirical. Benefit  Studies."   In:   Cost-Benefit
     Analysis and Water Pollution Policy,  H.M. Peskin,  and E.P.  Seskin, eds.
     The Urban Institute,  Washington, D.C., 1975.  pp. 127-172.

Tihansky,  D.P.    "A  Conceptual  Framework for  Risk-Benefit Assessments  of
     Hazardous Materials  Transport."  In:  Proceedings  of the Fourth Inter-
     national  Symposium on Transport of.Hazardous  Cargoes by Sea and Inland
     Waterways.   CG-D-24-76,  Department -of Transportation, Washington, O.C.,
     1975.  pp.  103-112.

U.S. Environmental  Protection Agency:  'Decision-Makers Guide  in Solid Waste
     Management.   SW-500,  U.S. Environmental Protection Agency, 1976  158 pp.

U.S.  Environmental   Protection  Agency..   "Cost-Effectiveness   Analysis  of
     Treatment/Disposal Alternatives  for Hazardous  Wastes."    Contract  No.
     68-03-2754 (awarded  to  SCS Engineers),  U.S.   Environmental  Protection
     Agency, Cincinnati,  Ohio, in progress.-

Waddell, T.E.  The Economic Damages of Air Pollution.  EFA-600/5-74-012, U.S.
     Environmental  Protection Agency, Washington, O.C., 1974.  167 pp.

Watson, W.D., Jr.,  and P.B.  Downing.   "Enforcement of Environmental Standards
     and  the  Central  Limit  Theorem."  Journal-  of the  American  Statistical
     Association.  71(335):567-574,  1976.

Zeckhauser,  R.    "Procedures  for  Valuing  Lives."   Public  Policy.  24(4):
     419-464, 1975.
                                     213

-------
                                  APPENDIX D

                 PUBLIC ATTITUDES TOWARD ENVIRONMENTAL ISSUES*
INTRODUCTION
     The objective  of this  literature  survey was  to obtain  a  broad under-
standing of  the  likely attitudes  and behavior  of the  various  parties-at-
interest toward  situations  involved in hazardous waste management.   As only
one study (Lackey, Jacobs,  and Stewart,  1973--see Section 4) has specifically
addressed attitudes  toward  hazardous waste management  facilities,  a broader
scope  was  necessary.   Initially,   it  was  hoped  that  a  study  of  attitudes
toward  nuclear  power and  associated wastes,  and  toward hazardous  waste-
generating  industries such  as  the  chemical  industry would  prove  fruitful.
The  intention  was to  use  the chemical  industry as a proxy  for  a  hazardous
waste management  facility and to draw parallels between  nuclear  power, with
its  radioactive  waste  disposal  problem,  and  nonnuclear  hazardous  waste
disposal.  However, it appears that there is very little relevant material on
attitudes toward waste-generating industries other than nuclear power.

     Though  there have  been many  attitudinal   studies  relating to nuclear
power  (several  of  which  have  addressed the  question of  nuclear  waste
disposal),  it  was  concluded  that attitudes toward  nuclear  power  would be of
very  limited  value  in  predicting  attitudes  towards  hazardous  wastes,  and
their management.  This  is  because it appears that the public associates the
possible discharge of radioactivity from nuclear power plants and wastes with
the  effect  of nuclear  weapons  (Pahner,   1976;  Louis Harris  and  Associates,
Inc., 1976;  Rappeport and Labaw, 1975; Slovic  and Fischhoff, no date).  While
hazardous wastes  comprising  obsolete chemical  warfare and ordnance materials
could to some  extent pose  an analogous type  of  threat,  in general  reactions
to  nuclear  facilities appear  to be  far  too  extreme to  apply to nonnuclear
hazardous wastes.  Only  a  very small proportion of hazardous  waste would be
as  combustible as petroleum  (which the  public appears to  readily  accept in
the  form of gasoline),  while few  would  approach   the  toxicity  of  materials
stored on the premises of a pesticide formulator.

     Because the  approaches  discussed above were unfruitful,  the survey was
widened  to  encompass  attitudes  toward  environmental   issues   in  general.
Knowledge of such attitudes is useful, as hazardous waste management involves
making  decisions  about  the  environment—for  example,  attitudes  favoring
preservation   of   natural   environments   could  be  expected   to  correlate
  This appendix is based on material prepared by Ms.  N.S.  Avitable.


                                     214

-------
with  disapproval  of  ocean  dumping.   Widening  the  scope  of  the  survey
introduced  some  difficulties,  and  caution  must be  used  in  interpreting the
reported  results.   Many of  the  studies used extremely broad  or imprecisely
worded questions, and  consequently  it is hard to distinguish between ccncern
about pollution  and  concern  about the environment (which is a broader issue,
including resource use,  etc.)-   Unfortunately,  the term "pollution" is often
used to describe environmental  conditions having "unfavorable quality" (Swan,
1973), which suggests that respondents may have considered only the aesthetic
or  cosmetic side of  an issue and  not  concerned  themselves  with  health or
safety aspects.  Another difficulty is that quality of the environment may be
equated  with  quality  of life  (White,  1966).   Milbrath's  (1975)  conclusion
that perceptions of  environmental  quality incorporate views other than those
on  the  natural  environment reinforces this concern.   Lowenthal  (1966),  com-
menting  on  White's  (1966)  paper  mentions  some  other  pitfalls  that  nay
influence findings, as  follows.   Some people are not  always  able to present
their feelings articulately.   Others  consider that their environmental pref-
erences  should  not  be  subject  to  scrutiny.   Also,  people  have  variable
attention spans,  which  influences  what  they learn  about a situation;  thus
their opinions may be  based on incomplete or erroneous  knowledge.   Finally,
the values  inferred  from  a  study  are  dependent on how  the  respondents are
asked to express them (Dillman  and Christenson,  1972).

     Perhaps  inevitably  there  are numerous differences among  the  results of
the various surveys  reviewed.    In  addition to geographic and demographic
variations between the populations sampled,* the wording of the questions and
the timing  of  surveys  doubtless  influenced the  responses.  Thus the results
reported here  can probably be  used to provide  broad indications of views on
the environment,  but would  be  inappropriate if  fine detail were  required.

     This appendix attempts  to. group  material by subject .category to facili-
tate easy reference.   However,  because of varying objectives and the frequent
broad nature of individual  studies,  this often results in a given study being
mentioned several times, while the  allocation-of data  to a particular topic
is occasionally somewhat arbitrary.

PUBLIC CONCERN FOR THE ENVIRONMENT

Development of Environmental  Concern

     Concern about the  environment  is not a recent development.   Problems of
manmade  pollution  date  back to  Roman times  (Sewell and  Foster,  1971), and
conservationists such  as James  Fenimore Cooper and John  James  Audubon  were
active  in  the  early  nineteenth  century  (McEvoy,   1973;  Albrecht,  1975;
Morrison, Hornback,  and Warner, 1972).  In the early 1950's, certain dramatic
events such as  the  Donora, Pennsylvania, and the London,  England, air pollu-
tion deaths caused widespread  awareness  of the dangers of pollution.   By the
late   1950's,   some   air   and   water  pollution-abatement   programs   were
*  Except  where   indicated  otherwise,  all  the  surveys  discussed  in  this
  appendix  attempted   to  draw   representative   samples  from   the   local
  populations.


                                     215

-------
 Instigated,  but these did not receive widespread public support.   However,  by
 the late 1960's, the public had become sufficiently concerned about pollution
 t$-exert pressure  to have  something done,  and the turnaround in  the  Nixon
'Wrtfinlstrati on's environmental  policies  has  been credited to  public  opinion
 (Trop and Roos, 1971).

      Heberlein  (1972) implies  that  this  sudden growth  in  environmental  con-
 tarn might  be  attributed  to  the   moral  turbulence  of  the  1960's  and  a
 resulting shift from economics  to  ethics  as  a  basis  for  decisionmaking,
 together with the universal  appeal  of the  environment.   This may  nave caused
 p'edple  to begin to  feel  responsible for the consequences of  their actions.
 An alternative  explanation is that  this concern developed as a  response  to
 the rapid visible deterioration of the environment (Koenig, 1975).   This view
 is supported by McEvoy  (1973),  who  relates enhanced environmental  concern  zo
 increased contact  with   nature  (through  such  activities  as camping,  etc.),
 which provided a contrast between relatively  unspoiled  surroundings and man's
 generally deteriorating  environment.   In  the  early 1970's, interest  in the
 environment  was accelerated when the mass media found that it made good copy,
 particularly  as dramatized  by  "doomsday  sayers"  such  as  Barry  Commoner,
 Rachel   Carson,  and  Dennis Meadows  (Sewell  and  Foster,  1971; Downs,  1972;
 Swan, 1973;  Albrecht, 1975).   One result  was  that it became fashionable to  be
 environmentally concerned  (Swan, 1973;  Morrison, 1973; Means,  ._372;  Koenig,
 1975).

      The consensus  view of  studies  conducted  in the mid-1970's  is  that the
 environment  was perceived  as  being  worse at the  time  of  the studies than  it
 had  been  at any time in  the  past  (Milbrath, 1977, 1975;  Althoff,  Greig and
 Stuckey, 1973; Brunner,  Gravengaard  and Bennett, 1973).   Results of one study
 suggested that  recent decay  had  been most  evident to persons in metropolitan
 areas with populations  of more than one million  (Joseph  M.  Viladas Company,
 1973).   Although many respondents expected the environment to improve in the
 future,   there  was  pessimism about  the  degree of  improvement and  the  time
 needed to achieve it (e.g.,  Milbrath, 1977,  1975;  Brunner,  Gravengaard, and
 Bennett, 1973).

 Factors  Influencing Environmental Concern

      Concern for the  environment is not felt by everybody, and the degree  of
 concern   varies  among those who are concerned.   This  may  stem  from  man's
 relationship with the environment.   Historically, it has  been necessary for
 tnan to  master  the  environment merely to survive,  let  alone  prosper (McEvoy,
 1973; Albrecht, 1975, 1972).   The  technology that has  helped man do this and
 that has given  him  many conveniences has  at the  same  time created pollution
 (Heberlein,   1972;   Swan,  1973).   According  to  Brunner,  Gravengaard,  and
 Bennett   (1973),  it  is  man's  continued  desire for the  benefits of affluence
 that has made  pollution such a problem.  This may have made it hard for many
 to become more than mildly concerned about  environmental conditions.  Concern
 may  also be  lessened by man's ability to adapt to adverse conditions (Dubos,
 1956; Swan,  1973).   Means (1972) claims that there are  four factors that keep
 man  from changing traditional  attitudes  (i.e., pro-technology and pro-growth
 attitudes),   and that limit concern.   They are:   (1)  The  relationship  to
                                      216

-------
nature  is  assessed  in quantitative terms, not qualitative  ones,  (2) popula-
tion  growth  is seen  as good  for  the economy, (3)  the  present trend toward
urban existence separates  man from the land and associates him with economic
growth, and (4) social mobility further separates man from nature.

     The spectrum of  attitudes that the public exhibits towards the environ-
ment  reflects  differences  between  individuals (Maloney  and Slovonsky,  1971;
Costantini  and Hanf,  1972).   Concern  for  the  environment  is   related  to
attitudes  which  are  a  function  of such  factors  as  an  individual's percep-
tions, values, priorities and aspirations (Swan,  1973; White, 1966; Albrecht,
1975; Groves  and  Kahalas,  1975;  Klee, 1973).  White  (1966)  identifies four
attitude   formation   factors:   The  decision  situation,  the  individual's
experience with the  environment,  his  perception of  his  role,  and his compe-
tence  in   dealing  with  its  complexity.   Munton  and  Brady (1970,  cited  by
Albrecht,   1975)  consider that education  is an important  factor  in attitude
development, since it is necessary to be taught what is good and what is; bad.

     Attitudes are also  influenced by needs.   According to  Mas low,  man must
first satisfy  his physical  needs  (e.g., food, shelter) before he will become
concerned  about other needs  (e.g., safety, the environment) (Albrecht,  1975;
Swan, 1973).   This may explain the lack of environmental  concern exhibited by
people in  low socioeconomic groups (Swan,  1973).

The Role of the Mass Media

     The mass  media  can affect an individual's knowledge, as they constitute
a  source   of  environmental   information  (Milbrath,   1975).   It was  shown  by
McEvoy  (1973)  that  in  the  period  1953-69, when there was  a  growing concern
over environmental problems  by the public, there was not only an increase in
the number of  environmental  articles  (by a factor of 4.7), but a tripling of
the number of  magazines  carrying  such articles.   McEvoy's  review of a study
by  Russell  (1970)  indicated  that most  of  this  increase  was  in  articles
dealing with the urban environment.  Munton and Brady (1970) noted a dramatic
increase between 1964 and  1969 in the number of letters-to-the-editor of the
New York Times on environmental problems.   These  authors also demonstrated a
similarincrease  in  the coverage  devoted to  environmental  matters by  the
New York Times, and  they concluded that  public concern was  at least partly
explained  by the  attention  given  to environmental matters  by  the mass media
(Munton and Brady, 1970).

     O'Riordan (1971a)  found a high  degree of awareness  regarding  a sewage
disposal  issue in his  study.   He  attributed  this   to  the influence of  the
local newspaper,  as  more  than 65  percent of his  interviewees said  it  was
their most important  source  of information.   On the  other  hand,  data in one
of Milbrath's  (1975)  surveys indicated that the  public  perceived themselves
to be poorly informed about environmental  problems.

     The amount of  space given by the media  to  a topic depends on how news-
worthy it is felt to be.  In a basically rural state (Kansas) the environment
was not seen  as particularly newsworthy (Althoff, Greig, and Stuckey, 1973),
even though a  majority  of  media managers felt it was a real problem.  On the
other hand,  Maloney  and Slovonsky's  (1971)  survey of editors  in  areas with


                                     217

-------
high air pollution  found  the environment to be an important news item mainly
because pollution was  a  serious  problem that affected  everyone  and that had
become a  political  issue.   Although  pollution is  seen as  a  local proolem,
only the  larger newspapers  obtain  most of their information  from  local  and
state governmental  agencies.   The  smaller  ones rely on  the  wire services, a
fact which  influences  the  nature  of their coverage (Maloney  and Slovonsky,
1971).

     No matter  how  newsworthy the environment  is  felt  to  be,  Maloney  and
Slovonsky (1971)  indicate that the effect of  coverage depends  on audience
predisposition,  and  that interpersonal  relationships contribute  most  to  the
shaping of  public opinion.   This  view is  supported  by the work of Sharma,
Kivlin and  Fliegel  (1975).   Respondents in their sample  indicated  that they
did not use information  from the local  newspaper to influence their opinions
about local  pollution.   In some instances,  the news  media may add to existing
anxiety by  its  enthusiastic  coverage  of  an issue.   For example,  media bom-
bardment  heightened the   controversy  over radiation  in  the  early  1960's.

Who Is to Blame for Pollution?

     The public  does not  attribute pollution to any  one preeminent  cause,
although industry is often rated as a major source.   Respondents taheve that
the blame  can be  shared  by  the  public,  emphasis  on  growth,  and  a lack of
planning (Erskine, 1972a; Milbrath, 1975,  1977; National Analysts Inc., 1972;
Langowski  and  Sigler  [Belden  #2  study],*  1971; Murch,  1971;  Dunlap,  1975).

     The results  of  various  studies indicate that different groups emphasize
different sources.  Analysis  of  data  presented by Milbrath (1975), found the
public to give  the  highest mean-cause  ratings to industry and its own care-
lessness (first  and second)  out  of several  possible  causes.    In  contrast,
community leaders  rated   lack  of  long-term planning and  high  consumption as
causing  the  greatest  amount  of  pollution.   Dunlap  (1975)  found  that
Democratic  and  liberal students  blamed emphasis on  growth, while Republicans
and conservatives blamed  society's  habits  and desires.  This  may be indica-
tive of the nonbusiness  versus business emphasis of the groups, although it
could  be  two ways  of  saying the same  thing.   The  response  frequencies from
the  1970  Opinion  Research  Corporation survey  (cited  by  Erskine,  1972a)t
showed  the  chemical  and  oil  industries  receiving the  greatest  amount  of
blame.   Yet,  closer inspection of the data indicates that people put most of
the  blame   on what  is  familiar.   For  example, people  in urban  areas blame
industry and  automobile  emissions,  while  those in rural areas blame dust and
insecticides.  Finally, 45 percent of the  respondents  to  a  1972 Harris poll
*  Langowski  and Sigler  abstracted data from several  surveys,  most  of which
  were unpublished.

t Erskine reproduces data from various surveys, and does not provide specific
  citations for these surveys.
                                     218

-------
said pollution was  the  biggest problem created by  science  (Erskine,  I.972a).
However, Taviss  (1972)  found that opinion was divided over the inevitability
of technology causing pollution.

How Can Improvement Be Effected?

     Questions about how the environment can be improved are usually multiple
response questions,  and  there  is often no  consensus  solution.   In one study
(Murch, 1971),  almost half  the  respondents either  skipped the  ques;ion  or
said that  they could not  decide.  A  frequent response has been to  suggest
government control  at some  level—local,  state or  Federal  (Althoff,  Greig,
and Stuckey,  1973; National Analysts,  Inc., 1972;  Langowski  and Sigler, 1971;
Milbrath,   1975).   Other suggestions have  included  controlling  auto exhausts
and  industrial  wastes   (Erskine,  1972a)  by  individual   effort  (National
Analysts,   Inc.,  1972; Koenig,  1975;  Langowski  and Sigler  [Belden #1  study],
1971) and through advanced technology (LaPorte and Metlay,  1975).

     A  majority  of  the Democrats  and liberals  in Dunlap's  (1975)  student
sample  felt  that  a  significant change would be needed  in  our sociopolitical
structure for environmental improvement to occur.   The Republicans, moderates
and conservatives felt  that  a  solution could  be obtained within  the  present
system.  Dunlap and Gale (1974, cited by Albrecht,  1975) suggest that  this is
because the  Republicans  are  business  oriented, and  are  therefore not  likely
to want government  intervention or  control.   The  need for  sociopolitical
change also appears to be favored by many environmentalists  (Morrison,  1973).

     Environmental  planning  with  public  involvement  was  an  acceptable
solution to  a  majority  of  Milbrath1s  (1977) respondents.  However, -ndustri-
alists were  split equally  between public involvement and leaving planning to
the  specialists.   When  the  public  was  asked if  it still  wished to  have
planning as  opposed  to  certain freedoms (e.g., to do what they wish on their
own property), the percentage in favor of planning decreased slightly.   Those
in areas  not expected  to  experience  .economic, growth ana1  low  income  whites
were  most  likely  to  favor  freedom.    Environmentalists   and  developers,
however, were-uncompromising in their views.

     For another group sampled by Milbrath (1975),  pollution elimination (air
and  water),  obtaining  adequate  energy supplies,  educating the  puolic,  and
land use  planning were  at the top  of the priorities list  for  improving the
environment.   Individuals'  perceptions  of  the  adequacy of present efforts in
solving the  problem affect  their estimates  of the future  kind and level  of
effort needed to obtain  results.   Most of those in rural  areas  appeared to be
satisfied  with  present  efforts,  whereas  most of  those  in  industrial  areas
found them inadequate.   Community leaders tended to see existing efforts  to
be more effective than did the general public,  but it was suggested that this
might be because  they themselves were usually involved  in  these  efforts and
thus were  more aware of progress.  Of all the solutions suggested, more  of
the  public were  willing to  rely  on technology than were the  leaders,  and a
majority of  both  groups  saw  a need for some  social  change (Milbrath,  1975).

     There appear  to be no  well defined  attitudes  regarding  the amount  of
enforcement  needed  to   make  environmental  policies  effective   (Joseph  M.
Viladas Company,  1973).   The  results  indicated  that strict enforcement  is

                                     219

-------
more  likely  to  be  favored by  those  who are  the most  concarned  about  the
problem,  those  living in metropolitan areas, and  those seeing their environ-
ment  as  least good.   Also, the  more  a person was willing  to pay to  reduce
pollution,  the  greater the enforcement he desired.  However,  the researcners
concluded that  where  people  were faced with the  facts,  they first balanced
the  cost  of pollution abatement  against  its  benefits  and then took a  stand.

The  Relationship Between Concern  and Action

      Attitudes are not directly  related to action; they  may only establish  a
predisposition  to  behave  in  a particular manner  (Otway  and Fishbein,  1975;
White,  1966).   The  action that an  individual   takes   regarding  a problem
depends on  the concern he  has about the problem and its priority in  his  life,
and  on what he  feels he  can  do (Swan,  1973;  Costantini  and  Hanf,   1972).
Responses to risk situations  depend  on  how the  individuals  interpret  the
potential  perils  (real  or imagined) and the benefits, if any  (Maderthaner et
al.,  1976).   For example,  one person  may  fear  that the  nearby location of  a
chemical  landfill  could 'lead  to his  death  from  some  accident,  whereas  his
neighbor  may see the landfill as a source of employment.  Pahner (1976) in  a
paper  discussing  societal  responses  to  technological  change  (primarily
nuclear power)   concluded  that  individuals  learn  to construct  and ocrciive
their environment  in such a way  as to reduce their anxiety.   Ivaryboay  needs
to feel secure, and when anxiety  is intense enough, action results.

      Although  the  operation of  the National  Environmental  Policy Act  calls
for  citizen input  to environmental decisions (Smith, 1973), this has not  "ed
to  significant  public participation in many situations.   A  common pattern is
that  a  first decision  is made with little public input, even though  a special
interest  group  may  have  been responsible  for  the pressure that  led to  the
decision.   This  is then followed  by some kind of action because people  do  not
like  the decision  (O'Riordan,  1971a).  Milbrath  (1977)   feels this lack of
participation may  be a result of inadequate media emphasis  and the  fact that
people  do not have a  sustained interest in the issue.

      In 1972, Downs  demonstrated how the environment  fits  into the "issue-
attention cycle"  that the public appears to  follow on domestic problems.  He
suggested that  interest  intensity  would decline  as  the  public realized  how
.difficult,   and  especially how  costly,   it  would  be to  solve environmental
problems  (Downs,  1972;  also  see Sewell  and Foster, 1971).   However,  Downs
noted that once a problem  has been prominent, it  can easily recapture  atten-
tion.   This is  not  inconsistent  with  Mitchell's  (1978)  claim  that  the
environment is an  enduring concern.

      Although  most  of those  interviewed by  Murch  (1971)  said  they  would
consider  the pollution  issue  when voting for  candidates for public office,
action  on   environmental  issues is  usually initiated   by  special  interest
groups  who  attempt  to  influence decisionmakers  or  others  who can have an
impact  on  decisions  (O'Riordan,  1971a).   Draper  (1973)  implies that  action
groups  emerge because people feel remote  and their voices can only be  heard
in  this manner.    In meetings held  regarding  sanitary landfills,  Klee  (1973)
found that  the  percentage of the  public attending was  higher for those in
rural areas than  urban  areas.    Similar turnout levels were found  for  urban
                                     220

-------
ethnic and  working-class  areas.   Objections  were usually  raised by  ad  hoc
groups.

     Studies of  individuals  involved in environmental  activities  have found
that  such  people  were more  knowledgeable about the  environment and  less
optimistic about what  could  be done than  those who  were relatively inactive
(Levenson,  1974; Milbrath, 1975).   Those involved also had a higher level  of
education,  more complex belief structures, were more aware  of  and concerned
about  the  problem, and felt  it to be more  serious  and in  urgent  need  of a
solution than others  (Milbrath,  1975).   Milbrath's study found  that  most  of
the activities in which individuals said they participated were of a personal
rather than political  nature.   Generally,  people in his sample felt a higher
level  of  responsibility to engage in  such behavior than  they  actually  did,
but  they  were  less  inclined  to  feel  a need  to  participate  in politically
oriented activities than personal ones.

     Trigg et al.  (1976)  studied the anti-pollution behavior  of "internals"
(those who believe their rewards are dependent on themselves) and "externals"
(those who  believe  their rewards  are  dependent  on outside  forces).   They
found that optimistic internals had the highest mean number of anti-pollution
activities and  optimistic  externals the lowest.  The two  pessimistic groups
were in the middle, with an equal mean  number of activities.

     In some cases,  environmental  intervention has  led  to  positive results.
For example, Commonwealth  Edison's (Illinois)  stepped-up program to c'ean the
air (Maloney and Slovonsky, 1971), and  the prevention of a West Virginia  firm
from disposing  of  their chemical wastes in Louisiana (Mackay,  1976).  There
has also been a  shift in the  burden of  proof requirements from those environ-
mentally concerned to the  proponents of development (Goldman, 1973).  Thus  it
is  now the  developers who have  to  show  that  a project will not cause unac-
ceptable environmental degradation, whereas previously the converse was true.

     In other cases,  however,  public  concern has accomplished  little.   The
Reserve Mining  case has  received  much  attention,  but  while  protest action
started in 1969  (Carlson,  1975), the problem has only recently been resolved.
In  1978,  the Reserve  Mining  Company obtained  the necessary  permits  to  use
land  disposal   for  its  asbestos-bearing  taconite  tailings,  scheculed  to
commence in  1980.   (Telephone interview, A.  Samuel, Reserve Mining Company,
Silver Bay, Minnesota,  August 7, 1978.)   Carlson concluded that the concerns
of  the citizens  (over the quality of their drinking water) had difficulty  in
competing with  economic considerations.   She  implied that the Reserve Mining
Company was using  delaying tactics in  the hope that the public would tire  of
the  issue  and  to  delay any  costly remedial  actions  such as  land disposal
(Carlson,   1975).   In  a similar situation, although concerned  about sewage
seeping  into a local  lake,  most  of   O'Riordan's  (1971a)  sample had  not
bothered to become involved in any action that could change the situation,  as
they  felt  that  they  would not  be able to  influence  the political  system.

     Furthermore,  it  appears  that  people do not necessarily take avoiding  or
mitigating  action,  even   where  the  appropriate  action  (at  least as   a
precaution) is  obvious.   Data from a pilot study  done  in  Duluth, Minnesota,
on  people's attitudes  and  behavior regarding  asbestos fibers in their drink-
ing water  (originating from  the Reserve Mining Company's  taconite tailings)

                                     221

-------
were  not encouraging.   The vast  majority  still appeared  to be  using  this
water for everything (Franz, 1976).

ATTITUDES AND BEHAVIOR OF SPECIFIED PARTIES

The General  Public

Level of Concern—'
     Concern over the environment varies from none to a great deal, while the
public's perception of  pollution  as a problem ranges  from  little or none to
very serious.  When  asked  about the most important problem facing the United
States today, pollution generally is ranked behind such areas as the state of
the  economy and  crime; i.e.,  it  is  not  the  top  priority  (Koenig,  1975;
Brunner,  Gravengaard,  and  Bennett,  1973;  National  Analysts,   Inc.,  1972;
Milbrath,  1975;  Dillman  and  Christenson,  1972;  Erskine,   1972b;  Mltcnell,
1978; Trop  and Roos,  1971;  Joseph M.  Viladas Company,  1973).   Between  1965
and  1970,  public opinion polls  such  as those conducted by  Harris,  showed a
dramatic  increase  in concern  over the problems  of air and water pollution
(Erskine, 1972b).   A  review  of polls  from 1970 to  about  1977  by  Mitchell
(1978)  found that  although there  was  an  increase in  "very serious"  as  a
response to  questions  about the seriousness of  the  pollution problem,  there
was  a decrease  in the proportion of  those  responding "pollution"  to the
open-ended  question  of "What  are  the  important  problems  that  we are faced
with today?"

     Concern about  pollution  does  not necessarily correlate  with the levels
of  pollution  experienced.   Ludwig,  Morgan,  and  McCullen  (1970)  analyzed
survey data and found that concern for such things as air pollution had risen
in  a broad   sample of  urban areas, although the  levels  of  common pollutants
had  declined.  This  has been attributed, at least  in  part, to the attention
being given by the mass media to pollution (Albrecht, 1975).

Relationship Between Concern and Demographic Characteristics, Etc.--
     The  perceived  seriousness  and  concern  felt  about  pollution  and  the
environment  has   been  studied in  relationship   to  various  demographic  and
personality  characteristics.   There is  a consensus that high levels  of  both
perceived seriousness and concern are  exhibited by the young, those with  high
incomes,  those with  a high  level of  education  (usually  some  college and
above),  and  those  who are  classified  as professional  (white collar)  workers
(Winham,  1972;  Brunner, Gravengaard,  and  Bennett,  1973;  National Analysts,
Inc.,  1972; Trop  and  Roos,   1971;  Langowski  and  Sigler,  1971; Joseph  M.
Viladas Company,  1973; Tognacci et al.,  1972).

     There  also  appears to  be a rural/urban  dimension  (Althoff,  Greig, and
Stuckey,  1973;  Erskine,  1972b;  Milbrath,  1975).   The  greatest  concern  is
exhibited by those in large metropolitan cities (Sigler,  1973; Trop ana Roos,
1971; Langowski and Sigler, 1971; Joseph M.  Viladas Company, 1973).  A review
of the polls presented by Erskine (1972b) showed the following patterns:   The
most concern or awareness was in suburban areas,  followed by the core cities,
then  serai rural  areas,  with  the  least  being shown  in  rural   areas.   This
pattern  was  consistent,  no  matter how the  interview questions  were phrased.
However, at  least  one study (Koenig,  1975) did not find this dimension.   The


                                     222

-------
rural/urban dimension  may actually reflect some  obvious  differences between
rural  and  urban dwellers.   Rural  people are  most likely to  be  directly or
indirectly dependent  on natural resource industries  (including agriculture)
for  their  livelihoods, and  to  be  exposed to lower levels of  pollution than
urban  dwellers.    According  to  a  recent  analysis  by  Tremblay  and  Ounlap
(1978), this hypothesis is supported by study data which show that the urban/
rural  dimension  cannot be  accounted  for   by age,   income  or  educational
differences.

     Some evidence supports a regional factor.   Erskine's (1972b)  poll  review
found  that  peopTe in  the  South were  consistently low in their  concern and
awareness responses.    Trop and  Roos (1971)  feel that this is due  to the fact
that there  is little  industry  in  the  South,  and that  these  people  have a
tradition of States rights.

     Dillman  and  Christenson (1975) studied the  association  between concern
about pollution, the  extent to which various types of corrective actions were
desired (e.g., more government  spending),  and the actual  level of air pollu-
tion.  The results showed that the higher the pollution level  the  greater was
the concern,  as did  the analysis of Tremblay and  Dunlap (1978).  Dillman and
Christenson   (1975)   also  found  weak  correlations  between concern .and
activities desired, but  correlations  between  these variables and  demographic
and  community  identification   characteristics   were  inconclusive.   These
findings led  Dill man  and Christenson  to suggest  that people  'accommodate to
pollution.

     People tend  to  view pollution  as  being worse  elsewhere than at home
(Murch,  1971;  Milbrath,  1975;  Langowski  and Sigler,   1971;  Treirblay  and
Dunlap, 1978), although Winham (1972) found  that Hamilton, Ontario,  residents
rightly perceived  the pollution problem to be worse  in  their own  community
than in other parts  of Canada.   Murch1s (1971) respondents  not only did not
believe  that  pollution was  as  bad  a  problem in  their  city  (Durham,  North
Carolina) as elsewhere, they also perceived  it as  less severe in their neigh-
borhoods than in  the  rest of the city.   Murch attributed this to  the media's
concentration on  national  environmental  problems  and not local  ores.   This
tendency by  residents  to discount  local pollution  (unless  the problem is
really bad) was analyzed by  Milbrath (1975), who  suggested  that  many people
live as close as  they can to their ideal locale.   Consequently, they find it
difficult to perceive that where they are living is worse than anywhere else.
There  is some indication that  an individual's view of pollution  in his com-
munity is correlated  with whether or not he  is an  established-resident, which
tends  to  confirm  Milbrath's suggestion  above.   Sharma,  Kivlin,  and Fliegel
(1975) found  that  longtime  residents in their study did  not  view  the problem
as seriously as  others.  Murch1s (1971) more committed residents (e.g., those
owning homes)  also tended to most underestimate the problem.

     Perceptions also tend to be based on what individuals are used  to.   This
is  illustrated  by the  rating  given  to drinking water  by  citizens  in  two
neighboring Georgia  towns  using water  from the same  stream.   Citizens from
one town found the water satisfactory while  the other group of citizens found
it  unsatisfactory.   The unsatisfied residents sometimes  received water from
an  alternative  source with  a  less pronounced  taste,  whereas  the  satisfied


                                     223

-------
group only  drew  water from the source branded as obnoxious by the displeased
citizens (White, 1966).

     Political ideology  has  also  been found to  have  some influence on views
on  the  environment.   In  general,  the  more liberal  a  person is,  tne more
concerned he  seems  to be (Koenig, 1975; Tognacci  et al ,  1972; Dunlao, 1975;
Dunlap and  Gale,  1972).   Ounlap's (1975) study of students at the University
of- Oregon  found that  liberals  not only saw pollution as being more serious
than did the  conservatives,  but also expressed  greater  interest  in environ-
mental issues, gave  greater  approval  to associated  movements,  and were :nore
likely to have  been involved in some kind of environmentally related action.

Other Factors in Environmental Concern —
     There  are  conflicting findings on  the  relationship  between  tne ^ncaTi
shown and   an  individual's  understanding  of  the environment.   Swan  (1573)
found no  relation between concern and  knowledge, and Milbrath (1975) found
that though over  90 percent  of those sampled had some environmental concern,
a  majority  had  no  environmental knowledge, or knowledge  that was only  s^ner-
ficial.   Other  work  in  this  area  by  Trigg  et al.  (1976)  found  'ntarnals
'(i.e., those  who  believe they can have  some  effect  on outcome) to  
-------
concept  but  were  less  opposed  to  sewage  treatment  than  less  alienated
individuals.   He also  found that  anemic  people tend  to evaluate  the terra
"sewage treatment" negatively.

     In a  longitudinal  study,  Oonohue,  Olien,  and  Tichenor  (1974) surveyed
the residents in four Michigan communities (Duluth,  Ely, Silver Bay and Grand
Rapids) in  1970  and again 18 months later in 1972.   They found that accurate
knowledge on  three  local  environmental  issues had decreased, as  well  as had
the support  for  the belief that technology could solve the problems.  People
were  turning their  attention  to  the (adverse)  consequences of  control  of
environmental quality  and were  more  likely  to  believe that the issues had
been given  too much attention.   Residents felt  that  the  burden of resolving
the  environmental  problem  had shifted  from the  individual  to  th =  firms
involved and  to  public  agencies, and in three of the four communities there
was reduced willingness to pay additional taxes.   The investigators conceded
that  attitudes  are  affected by community  structure and  issue  remoteness.
Large communities tend to be more concerned and more willing to accept neces-
sary controls than  small  ones, which are  faced with  a  trade-off involving
development values.

Attitudes to Specific Environmental Issues--
     Some research  has  been  done on attitudes toward certain specific envi-
ronmental   issues.   In  one  such study,  Salcedo  et  al.  (1971)  investigated
attitudes  toward the  pesticide industry.   They found what  they  called  a
rural/urban   dimension,  but examination   of  the   results  indicates  that
"agriculture/non-agnculture"  might  be  a  better  description,  as  farmers
possessed more  positive attitudes  towards  the  pesticide  industry  tian city
and town dwellers.

     Langpwski and  Sigler's  (1971)  survey of studies found that  as distance
to  a  sanitary  landfill decreased,  the percentage  of  people  with negative
attitudes  increased.    Klee's  (1973)  review  of several  EPA  studies -about
sanitary  landfills  showed  that the  nature  of  opposition to  landfills  is
influenced  by community  social status.   Low social  status  communities -are-
worried about becoming  a "run-down  community," whereas  those  with  higher
status  are  primarily   concerned with  nuisance  and,  to  a  lesser  extent,
property values.   Twenty sanitary landfill siting proposals were rev-ewed and
protest was  found to  be  strongly  correlated with visibility.   In  turn, the
success/failure  of   the proposal  was found  to  be   related  to  protest  and
location  (in  or  out of the  community it  was meant  to serve).   Seven  of 15
protested and all nonprotested  proposals were successful.   However, only one
of  seven  outside proposals  was successful.   Communities  do  not want  to be
garbage dumps for others.   In  another  study reviewed by  Klee,  it  was  found
that there  was a low  general level of awareness about  solid waste disposal
activities,  although  awareness  increased when controversy was  present.   Low
awareness of the nature of environmentally adequate  waste disposal activities
may explain why  negative attitudes  to waste  appear  to be transferred to
related correction proposals, such as sanitary landfills.

Summary—
     O'Riordan (1971b)  discusses public  involvement in environmental issues,
and lists a number of factors that inhibit the clear expression and political
                                     225

-------
articulation  of  public  preferences   about  the  desired  quality  of  the
environment.  This  list,  which  is  reproduced below,  points up  many  of the
aspects already discussed.

     (a)  People  rarely  act  until   they  are  directly  affected  and
          threatened;
     (b)  People  rarely  help  each  other  unless  they are  bound by  a
          common cause or faced with  a common threat;
     (c)  People  become  tolerant of gradually worsening  situations  and
          are able to develop a number of defensive social psychological
          and behavioral mechanisms which help them to accept  or avoid
          the full intensity of  the  deteriorating environmental impact;
     (d)  People  are  environmental  gamblers  discounting  heavily  any
          future uncertainties  for the transitory pleasures of immediata
          gains;
     (e)  People  are  schizophrenic   with  regard  to  the  environment
          because they are  usually unaware of the delicate interactions
          linking the various environmental subsystems, and will pollute
          on the one hand and yet demand increased environmental quality
          on the other;
     (f)  People are confronted with  all sorts of personal and community
          problems  in life  and  always delay  the  difficul*-  decisions
          hoping  that when  they have  to be  faced,  solutions  will  be
          easier to find and decisions easier to make.   Thus people tend
          to leave the most complex  decisions to the politicians and the
          experts, yet  are  surprised and  not  infrequently  annoyed when
          they  do not  always  come  up  with the  right answers  in  the
          absence of a clear expression of public desires;
     (g)  People  play a  variety  of  roles in their  economic,  social  and
          political  lives,   and  frequently  a  number of  these  roles
          conflict simultaneously.   This tends to  distort  the  rational
          reasoning process and may  lead to an inconsistency of attitude
          toward environmental  phenomena.  For example, the president of
          a large paper mill polluting a river may also be a director of
          a  local  community  organization  pressing  for  an  off-river
          swimming pool  to  protect  his children  from  a possible health
          hazard.   We  see  here  clearly  the  conflict  of   values  and
          resulting inconsistent actions of an individual  who is playing
          simultaneous roles as a private entrepreneur and as a socially
          concerned citizen.  (O'Riordan, 1971b:101)

Leaders

     Leaders are  those  people  who are probably in a better position than the
general public  to have an  impact on environmental  decisionmaking and  on the
environment  itself.   They   include businessmen,  politicians,  and industrial-
ists.   Some researchers have  put environmentalists in this  group,  but they
will be treated separately in this appendix.

     Like the general  public,  leaders  tend not to see the environment as the
main  problem facing  the nation.   Miller's  (1972) survey  found them  more
concerned  with  such  issues  as  race   relations,  unemployment   and  poverty.


                                     226

-------
Sofranko  and  Bridgeland   (1975)  theorize  that leaders  (e.g.,  mayors,  who
downplay environmental problems)  do  so because they  feel  they need to main-
tain a good community image.

     There  is  not always  agreement  between leaders  and  the  general  sublic.
Although there was  some  basic consensus on water  quality issues and causes,
and need for  environmental  planning, Milbrath (1977) found the two grouos to
rank the  remaining water  problems  and related causes  differently,  with the
public being more  pessimistic about  regional  (Niagara  Frontier  in New York)
water  quality  than the leaders.   In another  survey  carried  out  in  the same
area,   Milbrath  (1975) found  leaders to be more aware  of "the environmental
problem" and  able to  see  the problem  more as a  long-range  threat  than the
public.   Leaders  in  urban areas  saw  the environment  as being  worse than
perceived by the  general  public,  while the reverse was found in rural areas.

     In  two studies,  Milbrath asked  leaders  how they thought  the  general
public would respond  to  certain  questions, and compared their estimates with
those  actually  received.   In one  of  the studies,  although not generally
accurate, elected  leaders  were found  to  be fairly good  in  predicting which
government  body should be  responsible  for water quality, that the government
should  do   some  environmental planning,  and that people prefer  freedom to
planning (Milbrath, 1977).   In the second, the leaders  were  correct in pre-
dicting that  the  public  saw  pollution as  a  component  of the environmental
problem, that they  put  the blame  on industry,  and that they prefer economic
growth and  jobs  to control.   However,- the leaders did  not think  the public
would  perceive  the environmental  problem  to  be as serious as they  did, and
they incorrectly predicted them to desire better  gas mileage over clean air
(Milbrath,   1975).   Milbrath  inferred that leaders' estimates of the public's
views might be based  on  media presentations.   There was also some indication
of  self-consciousness  on the  part of  the leaders, as  they thought  that the
public would  find  ineffective government  to  be  an  important cause  of the
environmental  problem, whereas the public had a low mean rating on this item
(Milbrath,   1975).   Jensen  and Stormes  (1971,  cited by..Klee,  1973) el so found
that officials  were  not  very  accurate  in their  perceptions of citizens'
rankings of reasons for objecting  to  sanitary landfills.

     Costantini and Hanf  (1972) surveyed  only those who could have an impact
on  decisionmaking  in  the  Lake Tahoe  area,  and  determined a  .score on a scale
that they  developed for environmental  concern.   This score  was  found to be
positively  related to  the  sense  of urgency felt for the area's proolems, the
amount of response  action  an individual would support, the importance placed
on  the issue  of  environmental degradation, and the  level  of  education.  The
environmental   score  correlated negatively with the  emphasis placed  on the
area's environmental  problems and with length of residence.  Those  in the
high concern  group rated  the urgency of  quality  problems (e.g.,  deforesta-
tion)   greater  than that  of  practical  problems (e.g., .traffic  congestion),
were pessimistic  about the  environmental  future,  and  tended to  be  profes-
sionals or  government  officials (particularly Federal).  They also tended to
have  no faith  in  local  government,   to  value rural  life,  to  have  great
aesthetic appreciation, and  to be critical of  technology.  Those  in the low
concern group saw  practical  problems as more urgent,  tended  to  perceive the
environment  in functional  or  utilitarian  terms, were  more  lively  to  be


                                     227

-------
 businessmen  or  local  government  officials, and  were  more  likely  to  have
 conservative  political views.   There  was no  correlation with  socioeconomic
 status,  possibly  because  the  sample  was  almost  exclusively middleclass.
 Respondents who defined the environment  in  natural  terms saw the  Lake  Tahoe
 area  as  being mpre polluted  than  those who  keyed onto  health hazard  aspects.
 The  apparent  lack of  concern  by  local  officials was cited as the  reason wny
 people had turned  to outside  help  for.their  environmental problems.

      Carter et  al.'s (1974) national survey  of city  (population  > 10,000) and
 county  (population  >  50,000)  governments  included  the following findings.
 There was no  consensus on a definition  of environment; the  two  levels  of
 government  rated  problems  differently  as  to  their severity,  but the  more
 severely  a  problem was seen,  the  more  likely it was that something  would  oe
 done  to  correct it.   A regional and urban/rural dimension was also detectaa,
 with  the  environment being rated  the most important  problem  in  the West, and
 in the suburbs  followed by the core cities.

 Environmentalists

      Environmentalists,  or  "protectionists"  as  they   are  sometimes  caned
 (Milbrath, 1975),  can  be characterized as generally having  higher leve's  of
 education,   income,   and   occupational    status  than  the  gene V   populace
 (Albrecht,  1975;   Morrison,   Hornback,   and  Warner,  1972;   Milorath.   1975;
 Morrison,  1973).   They tend   to live  in  newer, better kept homes in  better
 neighborhoods and  to like an  outdoor lifestyle (Milbrath, 1975).  It  has been
 suggested  that  their education allows these people  the conceptual  tools with
 which to  better   comprehend  environmental  problems,  and that  their  higher
.levels  of income  not  only make  them  economically independent  of  industries
 that  might  conflict  with  the  environment,  but  gives  them the   leisure  to
 cultivate  aesthetic values and  be environmentally involved (Albrecht,  1972,
 1975);  Morrison,  Hornback,   and  Warner,  1972;  Morrison,  1973).   Morrison,
 Hornback  and  Warner  (1972)   found  that  their  sample  of  environmentalists
 contained  a  large number of  young persons,  which  they  attributed  to general
 student  activism  and  to  the  students'  instructors  being part  of the  elite
 (leader)  group  described above.

      Responses  of protectionists  in Milbrath's  (1975)   survey indicated that
 they  believed environmental  quality to be low and  that they  saw the  environ-
 ment  as not just  one issue, but  as an interrelated  set  of problems.   They saw
 the  causes  of the problem as  basically  economic and social  in  origin  (e.g.,
 high  consumption,  lack of planning).  They were  more  likely to blame  technol-
 ogy  and to be  more  pessimistic  about  future improvement.  Environmentalists
 tend  to be anti-growth (Milbrath,  1977; Albrecht, 1972).  One Milbrath  (1977)
 study found that  though  environmentalists  believed  that community  rule was
 preferable  to individual  rule and strongly  favored  planning, they  thought  it
 should be  at  a  regional rather than  local  level.  A  majority believed that  we
 can  have  both jobs and a clean environment,  and  that  money (e.g., from  taxes)
 should  be  spent  on environmental improvement.   Milbrath (1977)   found  that
 environmentalists  were not  at all close in predicting the  attitude of the
 general  public  on matters of  planning  versus freedom,   on a jobs vs  environ-
 ment  compromise,  and on the growth issue.  In fact,  they appeared to  perceive
 the  public as holding  almost  exactly opposite views  to  their own!


                                     228

-------
     All  environmentalists  are  not alike  (Morrison,  Hornback,  and  Warner,
1972),  probably  because of individual value  preferences.   Schnaiberg (1971,
cited by  Albrecht,  1975) has  classified them  into four  groups.   The first
group (which  he calls  "cosmetologists")  is  made  up of those  who  deal  only
with  the  by-products  of consumption,  and are  rather  superficial  in their
outlook.  They  relate well to  activities  like  anti-littering  campaign  and
tend to be  well  liked by the pollution generators.   The  second group is'the
"meliorists,"  who  focus   largely   on  consumption-related  activities  and
resource  recovery.    Schnaiberg  calls the  third  group  "reformists."   This
group  perceives the  relationships  between consumption  and production  and
environmental decay  and is more knowledgeable  about the  physical/biological
processes  involved.    Reformists  attempt   to  influence  decisionmaking,  and
instead of  focusing on litter  and recycling, they  are  likely to oppose auto
pollution,  dam  construction,  nuclear power facilities, etc.   Finally, there
are  the "radicals,"  who aim  at a  total  restructuring   of  the. social  and
economic  system.   Radicals  do  not  separata  environmental  problems  from  the
corporate or  industrial  state, and  its  attendant  inequalities (Schnaiberg,
1971, cited in Albrecht, 1975).   Thougn this range in orientation of environ-
mentalists is very wide, implying a wide  range of tactics,  Gale (1972) argues
that  to  date most  action  has  been within the  confines  of  middle-class
politics,  such  as  lobbying,  petitions,  and court suits.   (These  are  the
tactics  of   the  reformists,  who  probably have the greatest potential  to
influence hazardous waste management decisions.   Consequently reformists  are
emphasized  when  the behavior of  environmentalists  is discussed  in  the main
text of this report.)

     The  dramatic  growth in membership  in environmental  organizations that
occurred in  the  late  1960's (McEvoy, 1973) has  slowed  somewhat.   One reason
is  a shift  in  concern of  many  such groups  from recreational  interests  to
general   environmental  quality  issues  (Faich  and  Gale,  1971,  cited   by
Albrecht,  1975) and to problems of resource  depletion (Morrison,  1973).   In
some  cases,  this  has  put recreationalists  at  odds  with  preservationists
(Albrecht, 1975).   Another reason is that it has become  apparent to many that
some costs  will  be necessary  to  reach their goals.  Although  some expendi-
tures will  only be  incurred over the short term,  they  may be substantial  in
dollar  magnitude;  other  costs  may  be incurred in  terms of  lost  personal
freedoms and  the added social  complexity inherent with  new laws.   Along with
this shift  in focus  has come  a shift  in  strategy—from  participation (e.g.,
Earth Day)  to  power   (to  coerce change)   (Morrison, Hornback,  and  Warner,
1972).

Growth-oriented Groups

     It  is  not  surprising  that  the  anti-growth  attitudes  of the  protec-
tionists have engendered the  formation of opposition groups sometimes called
"growthists"  (Albrecht,  1972).   These   opposition  groups  comprise  both
specific  interests  threatened  by   environmental  action  and  disadvantaged
groups (e.g., blacks,  working  class) who  hope that continued growth will help
them overcome their current status  (Morrison,   1973; Buttel, 1978}.    It  is
interesting  to  note that Milbrath  (1977)  found that it was  not irdustrial-
ists, but developers who were  more likely to favor growth.
                                     229

-------
     These growth-oriented  groups have mounted counterattacks  of  two kinds.
The  first  is that of  implying  that  the environmental movement is  part  of a
larger conspiracy  (i.e.,  communism).   The second uses scare  tactics  such as
implying that job loss is inevitable if certain pollution prevention policies
are  instituted  (Albrecht, 1972).  The  classic economy/environment trade-off
provides good examples of the development of growthist groups.  When environ-
mentalists  were  successful   in  blocking  power  plant  construction   in  the
economically  depressed Southwest,  a pro-growth  movement  developed.   These
people  were  willing  to  take chances  with  the  environment  to   obtain  the
economic  benefits that  would come  from the  construction and operation of
these plants (Albrecht, 1972).

Industry

     Not only has industry received a great amount of blame for pollution, it
has  also  been  accused by  some of  not doing very  much about it (Erskine,
1972a;  Althoff,  Greig  and  Stuckey,  1973).   In  response,  corporations  have
included accounts of what they are doing to protect the environment or return
it to  its  natural  state in their  annual  reports  and advertisements (see Fry
and Hock, 1976).

     Industry, according to a 1970 survey of "Fortune 500" c:inpany executives
(Diamond,  1970),  was  worried  that under public  pressure  (they do not  feel
that  environmentalists  represent the public), government  will  set standards
that  could  hurt  it financially.   However,  these executives  were  not inher-
ently  opposed  to Federal regulation of  activities  and  indicated  that  they
would  particularly   like   to  see  programs  instituted  that  include  tax
incentives.   They felt that  any action taken should  be  on an industry-wide
basis so that competition would be maintained.  Some executives, mostly those
in the South or  with transportation utilities,  would prefer  regulation by
local governments.  Although a substantial number of large firms had budgeted
for  anti-pollution  activities, their executives said  that pollution control
was not a top priority problem.

THE ECONOMICS OF POLLUTION ABATEMENT

     It  is  evident  that all  actions  to improve  the environment  will  cost
something.  The cost may be a conventional economic cost, or it may only be a
cost  in terms  of time and inconvenience.  Some surveys have asked who should
take  responsibility  for  reducing  pollution.   Langowski  and  Sigler's (1971)
review  found that people  felt financial responsibility  rested primarily on
industry.   This  finding  receives some  support  from Eastman, Randall,  and
Hoffer  (1974/75).   Their  survey  in the  Four Corners  Region (which has a
number  of  power  plants  and mining operations) found  that  Indians living on
reservations  felt that financial  responsibility  lay  with  the producers and
users.  Many of these respondents were not users of electricity.   A majority
of nonreservation  residents  and tourist/recreationists felt that the respon-
sibility  lay  with a combination  of  electricity  users,  producers,  local
residents, and recreationists.
                                     230

-------
     Many  survey  respondents favored  more  government  spending  on pollution
abatement.   Others  felt  a  combination  of  all   three  parties  Involved-
Industry,  government,  and consumers—should be  responsible  (Erskine,  1972b;
Mitchell, 1978; Dillman and Christenson, 1972,  1975).

Vlilllngness-to-Pay for Pollution Abatement

     Most  people  are willing to  pay some amount  for environmental  improve-
ment,  even  if they do not  feel  a responsibility to  do so  (Eastman,  Randall
and Hoffer,  1974/75;  Milbrath,  1975).   Since 1965 there has been an increase
in  this proportion,  and  by 1971  they represented  more  than 50  percent of
those  polled  (Erskine, 1972b).   During this same period  of  time (1965-1971)
there  was  a  drop  in the number of people willing to allow tax incentiv/es for
industry (Erskine, 1972a).

     Willingness-to-pay may  be  in the  form of higher electric bill-;  or  'a
higher  fee for  use of recreational facilities  (Eastman, Randall, and Hoffer,
1974/75).  In some cases,  an increase  in  taxes seems to  be  an acceptable
means  of  providing  necessary  pollution  control   funds (O'Riordan,  1971a;
Milbrath,  1977;  Murch,   1971).   However,  some  would  prefer higher  prices
(Joseph  M. Viladas Company,  1973;  Mitchell, 1978),  and  others  would  accept
more regulation (Milbrath, 1977).

     Of  course, not everybody wishes to  increase  his out-of-pocket expenses
to improve environmental  quality.   Some would prefer to take personal  action,
such as to carpool or source-separate wastes.  For example,  90 percent of the
housewives interviewed  by  National  Analysts, -Inc.  (1972)  said  they  would
perform  in-home  source  separation of  wastes rather  than pay  even a  minimal
fee to have this done by the municipality.  It  should, however, be noted that
intentions do not  always translate  into action.   Although this  sample of
housewives was  concerned  about the  solid waste problem, it  turned  out that
very few had actually done anything about it (National Analysts, Inc.,  1972).
This   sort  of  discrepancy  had  been  noted  elsewhere.   In  the  Erunner,
Gravengaard  and Bennett  (1973)   study  in  Toledo,  Ohio, 61 percent of the
respondents  said  that they  would be  prepared  to  use  unleaded gas, and 41
percent  said  that  they would ride the  bus  to  reduce pollution.   But only 24
and 23 percent, respectively,  showed any sign  of acting on  these intentions.

     The dollar sums  that individuals  are willing  to pay to abate pollution
or  to   protect  the environment  vary widely.   For example,  Milbrath  (1975)
found  that about half of-his respondents-were not prepared to pay any.hing to
protect  the  environment.   Other  responses ranged up  to $415  per month, with
about  20 percent  of  the  respondents prepared  to  give up more  than  $25'per
month.    In a  1973 nationwide  study,  the Joseph  M.  Viladas- Company  (1973)
found  that the  average  amount volunteered  to  reduce  auto  pollution  was  a
capital  expenditure  of  $62  per car  and an operating expenditure  of  $27 per
year.  Jo reduce air pollution from electric power plants, people volunteered
to  pay an average  increase  of 22 percent (or $3.84)  in their  monthy  elec-
tricity  bill.   They were  also prepared to pay $37.43  per year  to eliminate
water  pollution from food processing  and an extra 15 percent to recycle solid
waste.    O'Riordan1s  (1971a)  respondents  were  prepared  to  pay a  12  percent
increase in  local  taxes  to prevent the seepage  of  sewage into a local  lake.


                                     231

-------
     Those with  the highest  incomes  and the greatest  concern  are generally
willing to  pay the  most  to insure environmental quality  (Joseph  M.  Viladas
Company, 1973; Milbrath, 1975; Auld, 1974).   One study found the amount to be
strongly affected by the interaction of income and concern (Joseph M.  Viladas
Company, 1973).   The amount  has  also been  found to be  directly  related to
socioeconomic  status and  the degree of  understanding  of  the  environmental
problem,  its  ramifications,  and  potential  solutions.   More  whites  than
blacks, more  males  than  females,  and more of  those  active  in  environmental
groups  than  not   were  willing  to  pay  something  (Milbrath,  1975).   Other
factors found  to  affect willingness-to-pay  were awareness of the EPA (Joseph
M.  Viladas  Company,  1973),  level  of  education (particularly  in  industrial
settings)  (Auld,   1974) and  perception  of  the seriousness  of the  problem
(Murch, 1971).  Auld (1974),  found that the sum that an individual was will-
ing  to pay was  inversely related  to  age.   However,  the Joseph  M.  Viladas
Company (1973)  found that the young  and  the old were  prepared  to spend the
most,  which  they  attributed  to  the fact  that  these people  are  on  the low
commitments parts  of the  "life cycle of family spending."  In one study, the
sum was found  not to be related  to proximity  to pollution (Auld,  1974).  In
all the studies reviewed,  data suggest that people are most willing to pay an
amount of money  that will  not significantly reduce their standard of living.

     The rural/urban and  regional  dimensions  again  appeared in analysis of
data  regarding  the  amounts  people were willing to pay.   Evidence indicates
that  people  in the South  and  in  small  towns and rural areas are  willing to
pay slightly  less  than  those in other areas  (Erskine,  1972b).   However, one
of  Milbrath's  (1975) studies  found those in rural  areas  willing to pay more
than those living  in cities, suburbs, and towns.

Trade-offs with Business and Growth

     Costs  of protecting  the environment  could  also  be  in  the  form  of  a
decline in  economic growth  or a  loss  of jobs.  The proportion of  those in
favor  of  closing  a  polluting plant  is  low (particularly  in  the  industrial
East)  when  a job  loss  is  associated  with  such an action (Erskine,  1972a).
One study  found  that unless there was some kind of government assistance for
those  losing jobs  to find new employment, people had difficulty deciding that
a  plant  should be shut down  (Joseph  M.  Viladas Company,  1973).   They con-
cluded  that  lack  of contact with  such  a situation  would make it  hard for
people to  give  a valid answer to this type of question.  Thus it  is easy for
somebody to  be in favor  of  closing  a plant when  the  consequences  will not
affect them, but it  is  hard for an  individual to support such an action  if he
will  himself  suffer an  economic  loss  as  a  result.   A  recent review by
Mitchell  (1978)  illustrates  this difficulty.   Though  42 percent   of the
respondents to  a  survey favored jobs and 29  percent  said it was more impor-
tant to reduce pollution,  18 percent said both were important.   However, most
of  those interviewed by Milbrath (1977) felt that improvement of the environ-
ment  would  create  jobs,  thus no economic trade-off  would  be  necessary.  In
his  other  survey,  he  found 43  percent in  favor of  jobs  and  25  percent
preferring environmental  protection,  the  remainder being undecided or seeing
no  trade-off as necessary (Milbrath, 1975).
                                     232

-------
     In  a  survey of residents  in  a community where the  closure  of a water-
polluting meat packing  plant was likely, Sharma,  Kivlin,  ana Fliegel (1975)
found a low oositive correlation between intolerance of water aolluticn and a
desire  to  have the  plant closed.   Forty-two percent  of  tnose  surveyed were
willing to  close  the plant, and these  people were found to have unfivorable
(disapproving) attitudes  toward pollution,  to  ciscuss it  more,  and to nave
white collar  occuoations.   Noticeably few young people rfere  in favor of tne
closure.  This,  the authors  theorized, was due  to the  fact that  tl-ey were
mostly short-time residents looking for work.

     The public preferred economic growth to environmental protection in two
studies  (Milbrath,   1975;  Winham,   1972),  but  the results were  reversed  in
another (Milbrath,  1S77).   Growth  was  strongly favored cy  uroan  slacks,  and
protection was favored by rural oeople  (Milbrath, 1977).   Buttel (15~:) Jound
only a  modest correlation  between welfare-state  liberalism  and  support for
economic growth among his working-class'respondents.  From this he sosculated
(by  implication)  that .environmentalists  may be able  to  convinca this group
that economic growth is  not a prereauisita to achieve  their aspira;ions  of
wealth  redistribution,  thereby  persuading  them  to  shift tneir  support  'to
environmental  protection.    Interestingly,  many leaders  saw  no neea  for  an
environment/growth  trade-off  (Milbrath,   1975).   V/innam  (1972)  found  that
those in favor of growth were  less  likely  to  see pollution as a prcolem and
were  not concerned  about  aoatement  costs.  On  a slightly  different note,
Murch's (1971) respondents  felt that one way to  protect  the environment was
not to bring  in any polluting industry.

REFERENCES

Albrecht,  S.L.   "Environmental  Social  Movements  and  Counter-Movements:   An
     Overview and an Illustration."  Journal  of Voluntary  Action  pesgarch.
     l(0ctober):2-ll, 1972.

Albrecht, S.L.  "The Environment as a Social Problem."  In:  Soc-ial   Problems
     as Social Movements, A.L. Mauss, ed.  J.3.  Lippincott Company,  .New rork,
     New York, 1975.pp. 556-605.

Althoff, P.,  W.H. Greig,  and F.  Stuckey.   "Environmental  Pollution Control
     Attitudes of Media  Managers  in  Kansas."   JournaliSiii  Quarterly, 50(4):
     666-672, 1973.

Auld, D.A.L.   "Willingness to  Pay for Pollution  Abatement: • A Case Study."
     Alternatives. 3(1):34-36, 1974.

Brunner, J.A.,  P.  Gravengaard, and  G.  Bennett.   Pollution  in  the  Local
     Community:   Citizen   Attitudes  and   Willingness  to  .Make   Personal
     Sacrifices in  Abatement.   Occasional  Paper  No.  23.   Business Research
     Center,  College of  Business  Administration,  The University  of Toledo,
     Toledo, Ohio, 1973.  50 pp.

Buttel, F.H.   "Economic Growth and the Welfare State:   Implications for the
     Future  of Environmental ism."   Social Science  Quarterly, 58(4):692-699,
     1978.
                                     233

-------
Carlson,  K.T.   "The People's  Lake."   Environment.  17(2):16-20,25,26,  1975.

Carter, S.,  M.  Frost, C.  Rubin, and L.  Sumek.   Environmental  Management  and
     Local  Government.   EPA-600/5-73-016,   U.S.   Environmental   Protection
     Agency, 1974.   390 pp.   [NTIS:  PB232955]

Costantini,  E.,  and K.  Hanf.   "Environmental Concern and Lake Tahoe:  A Study
     of  Elite   Perceptions,  Backgrounds,  and Attitudes."   Environment  and
     Behavior.  4(2):209-242,  1972.

Diamond,  R.S.   "What Business Thinks.   The Fortune  500-Yankelovich Survey."
     Fortune. 81(2):118,119,171,172, 1970.

Dillman,  O.A.,  and J.A.   Christenson.   "The Public  Value  for  Pollution Con-
     trol."   In:  Social   Behavior, Natural  Resources, and the  Environment,
     W.R. Burch,   N.H.  Cheek,  Jr., and  L.  Taylor, eds.   Harper ana  Row,
     Publishers, New York, New York, 1972.   pp.  237-255.

Dillman,  D.A.,  and J.A.   Christenson.   "The  Public  Value  for  Air Pollution
     Control:   A  Needed  Change of Emphasis   in  Opinion  Studies."  Cornell
     Journal of Social  Relations. 10(l):73-95, 1975.

Donohue,  G.A.,  C.N. Olien,  and P.J.Tichenor.  "Communities,  Pollution, and
     Fight for Survival."  Journal   of  Environmental  Education. 6(1):29-37,
     1974.

Downs,  A.    "Up  and Down  with  Ecology—The  'Issue-Attention  Cycle."'   The
     Public  Interest. No.  28:38-50, 1972.

Draper,  D.   Public Participation in Environmental  Decision Making.  Exchange
     Bibliography #396, Council of Planning Librarians, Monticello, Illinois,
     1973.  28 pp.

Dubos,  R.   "Promises and  Hazards of Man's Adaptability."   In:   Environmental
     Quality in a Growing Economy,   H. Jarrett,  ed.   The Johns Hopkins Press,
     Baltimore,  Maryland, 1966.  pp. 23-39.

Dunlap, R.E.  "The  Impact of Political  Orientation on Environmental Attitudes
     and Actions."  Environment and Behavior,  7(4):428-454, 1975.

Dunlap, R.E., and R.P.  Gale.   "Party Membership and  Environmental Politics: A
     Legislative Roll Call Analysis." Social Science Quarterly, 55(December):
     4, 1974.

Dunlap,  R.E., and  R.P.  Gale.  "Politics and Ecology:  A Political  Profile of
     Student   Eco-Activists."    Youth and Society.   3(June):379-397,   1972.

Dunlap,  R.E.,   and  R.B.   Hefferman.   "Outdoor  Recreation  and  Environmental
     Concern:   An  Empirical  Examination."   Rural  Sociology.  40(1):18-30,
     1975.
                                     234

-------
Eastman,  C.,  A.  Randall,  and P.L.  Hoffer.   "How Much  to Abate Pollution?"
     Public Opinion Quarterly. 38(4):574-584, 1974/75.

Erskine, H.  "The Polls:  Pollution and Industry." .Public Opinion Quarterly.
     36(2):263-280, 1972a.

Erskine, H.  "The Polls:  Pollution and its Costs." Public Opinion Quarterly,
     36(1):120-135, 1972b.

Faich, R.G., and R.P. Gale.  "The Environmental Movement:  From Recreation to
     Politics."  Pacific Sociological Review. 14(July):270-287, 1971.

Franz, R.E., Jr.   Public  Behavior  and  Attitudes  in  Response to  Reported
     Hazardous  Drinking Water - A Feasibility Study.EPA-600/1-76-026, U.S.
     Environmental  Protection  Agency,  Cincinnati, Ohio, 1976. 55 pp.   [NTIS:
     PB257981]

Fry, F.L., and R.J. Hock.  "Who Claims Corporate  Responsibility?  The Biggest
     and the Worst."  Business and Society Review, No. 18:62-65, 1976.

Gale,  R.P.   "From  Sit-In  to  Hike-In:  " A Comparison  of  the  Civil  Rights and-
     Environmental   Movements."   In:   Social  Behavior,  Natural  Resources,
     and  the Environment. W.R.  Burch.'N.H.  Cheek, Jr.,  and L  Taylor, etis.
     Harper  and  Row,  Publishers, New York,  New York,  1972.   pp. 230-305.

Goldman,  C.R.   "Environmental Impact  and Water  Development."   In:  Environ-
     mental Quality and Water Development. C.R.  Goldman,  J.  McEvoy, III, and
     P.J.  Richerson,   eds.    W.H.  Freeman   and  Company,   San  Francisco^
     California,  1973.  pp. 1-11.

Groves,  D.L,  and  H.  Kahalas.   "Values:  Assimilation  and Accommodation."
     Cornell Journal of Social Relations. 10(1):111-137, 1975.

Heberlein, T.A.  "The Land Eth'ic Real'ized:  Some Social Psychological  Expla-
     nations   for   Changing   Environmental   Attitudes."    Journal of  Social
     Issues. 28(4):79-87, 1972.

Jensen, B.T., and J.M. Stormes.  A  Psychometric  Study  of Attitude  Factors
     and  Problems  of  Solid  Waste.  Contract No. CPE 69-107, U.S. Environ-
     mental Protection Agency, 1971.

Joseph M. Viladas Company.  The  American  People and Their Environment-1973.
     A Study of National  Opinion and Attitudes About Environmental  Problems
     and Their  Solution  for  the  United  States  Environmental  Protection
     Agency,  Volume I.OfA-EPA 68-01-0905,U.S.EnvironmentalProtection
     Agency, Washington, D.C., 1973.   [NTIS:   PB227020].

Klee,  A.J.  • "Attitude -Concepts and Applications  to  Solid Waste Management."
     U.S.  Environmental  Protection'Agency,  Cincinnati, Ohio,  1973.    39pp.
    .' (Also see Klee,. A.J..  .."Aspects of-Sampling Attitudes Towards Soltd-Warste
     Problems."  In:'  Statistical 'and  Mathematical  Aspects  of  Pollution
     Problems. J.W.   Pratt,e
-------
Koem'g, D.J.   "Additional  Research on  Environmental  Activism."   Environment
     and Behavior. 7(4):472-485, 1975.

Lackey,  L.L.,   T.O.   Jacobs,   and  S.R.  Stewart.   Public  Attitudes  Toward
     Hazardous   Waste  Disposal Facilities.   EPA-670/2-73-086,  U.S.  Environ-
     mental   Protection  Agency,  Cincinnati,  Ohio,  1973.   195 pp.  [NTIS:
     PB223638]

Langowski, A.,   and J.  Sigler.   Citizen  Attitudes  Toward  the  Environment:
     An  Appraisal  of  the  Research.IIEQ-40.004,Illinois  Institute for
     EnvironmentalQuality,Chicago,   Illinois,   1971.    69  pp.    rNTIS:
     PB214190]

LaPorte,  T.R.,  and D. Metlay.   "Technology Observed:   Attitudes of  a  dary
     Public."  Science. 188(4184):121-127,  1975.

Levenson,  H.   "Ecological  Knowledge  and Perception  of  Environmental Modifi-
     ability."  American Psychologist.  29(4):274-275, 1974.

Louis Harris and  Associates,  Inc.   A Second Survey of Public and  Leadership
     Attitudes   Toward  Nuclear  Power  Development in tne Urn tea Statac    A
     study conducted  for Ebasco Services  Incorporated,  NewTT"*.  Ne
-------
Means,  R. L.   "Public Opinion  and Planned Changes  in Social  Behavior:   Ttie  "
     Ecological Crisis."  In:  Social  Behavior,  Natural  Resources, and the
     Environment. W.R.  Burch,  N.H.  Cheek,  Jr., and  L.  Taylor,  eas.  Harper
     and Row, Puclishers, New York, New York, 1972.   pp.  203-213.

Milbrath,  L.W.    Environmental  Beliefs:  A  Tale  of  Two Counties.   Social
     Science Research Institute ana  Social  Science Measurement Center, State
     University of New York, Buffalo, New York, 1975.  136 pp.

Milbrath,  L.W.   An  Extra  Dimension  of  Representation  in  Water  Quality
     Planning:   A Survey Study of Erie and Niagara  Counties, New York. 1976.
     Environmental Studies Center, State University of New York, Buffalo, New
     York, 1977.  50 pp.

Miller, D.  "The Allocation of Priorities to Urban and Environmental Problems
     by Powerful Leaders and Organizations."  In:  Social  Behavior.  Natural
     Resources, and  the Environment, -W.R.  Burch,  N.H.  Cheek,  Jr.,  and L.
     Taylor, eds.   Harper  and Row,  Publishers, New  York,   New  York,  1972.
     pp. 306-331.

Mitchell, R.C.    "Environment:  An Enduring Concern."  Resources. No. 57:1-2,
     20-21, 1978.

Morrison,  O.E.   "The Environmental  Movement:   Conflict Dynamics."  Journal.
     of Voluntary Action Research, 2(Spring)': 74-85,  1973.
                                                                             • .

Morrison, O.E., K.E. Hornback,  and W.K.  Warner.  "The Environmental Movement:
     Some  Preliminary  Observations and  Predictions."   In:  Social  Behavior.
     Natural Resources,  and the Environment, W.R. Burch,  N.H. Cheek, Jr., and
     L. Taylor,  eds.  Harper and  Row, Publishers, New York,  New York, 1972.
     pp. 259-279.

Munton,  D.,  and  L.  Brady.   American  Public  Opinion  and  Envi-onmental
     Pollution.   The  Behavioral   Science.Laboratory  Research Report, Umver-
     sity of Ohio, 1970.

Murch,  A.W.   "Public Concern  for Environmental  Pollution."   Public  Opinion
     Quarterly. 35(1):100-106,  1971.

National  Analysts,  Inc.  Metropolitan  Housewives'   Attitudes  Toward  Solid
     Waste  Disposal.   EPA-R5-72-003,  U.S.Environmental  Protection Agency,
     Washington, D.C. , 1972.  90 pp.

O'Riordan, T.   "Public  Opinion and  Environmental Quality:   A Reappraisal."
     Environment and Behavior.  3(2):191-214, 1971a.

O'Riordan, T.   "Towards a Strategy of Public Involvement."   In:  Perceptions
     and Attitudes in Resources Management. W.R.D. Sewell and I. Burton, eds.
     Information Canada, Ottawa,  Canada, 1971b.  pp.  99-110.

Orr, R.H.  "The Additive and Interactive Effects of Powerlessness .and Anomie
     in Predicting Opposition to Pollution Control."  Rural Sociology, 39(4):
     471-486, 1974.

                                     237

-------
Otway, H.J.,  and  M.  Fishbein.   The  Determinants  of Attitude Formation:  An
     Application  to  Nuclear  Power.   RM-76-80,  International  Institute for
     Applied Systems Analysis, Laxenburg, Austria, 1976.  31 pp.

Pahner, P.O.  A Psychological Perspective of the Nuclear  Energy Controversy.
     RM-76-67,InternationalInstituteforAppliedSystemsAnalysis,
     Laxenburg, Austria, 1976.  27 pp.

Rappeport, M., and P. Labaw.  Public  Attitudes and Behavior Regarding Energy
     Conservation:  Detailed  Tabulations  by  U.S. Population and Population
     Segments.  Waves  28  and  29.FEA/D-76/252, Federal Energy Administra-
     tion, Washington, O.C., 1975.  42 pp.  [NTIS:  PB254592]

Russell,  C.   The  Evolution  of the Nature Magazine:  A Case Study of  Three
     Publications:   Outdoor  America.  National   Audubon,  American  Forests,
     1954-1969.   M.A.  Thesis, University  of  Michigan, Ann  Arbor, Michigan,
     1970.

Salcedo, R.N., H. Read, J.F. Evans, and A.C. Kong.  "Rural-Urban Perspectives
     of  the  Pesticide  Industry."   Rural Sociology.  36(4):554-562,  1971.

Schnaiberg,  A.   "Politics,  Participation  and  Pollution:   T^a 'Environmental
     Movement.1"  Mimeographed, Northwestern University, 1971.

Sewell, W.R.D.,  and H.D.  Foster.   "Environmental  Revival:   Promise and Per-
     formance."  Environment and Behavior, 3(2):123-134, 1971.

Sharma, N.C.,  J.E.  Kivlin, and F.C.  Fliegel.   "Environmental  Pollution:  Is
     There  Enough  Public  Concern  to  Lead   to   Action?"   Environment  and
     Behavior. 7(4):455-471, 1975.

Sigler, J.  Attitudes  of   Illinois  Citizens  Toward  Solid  Waste  and  the
     Environment.IIEQ-73-6,Illinois Institute for Environmental Quality,
     Chicago, Illinois, 1973.  140 pp.   [NTIS:  PB223457]

Slovic, P.,  and  B.  Fischhoff.   "How  Safe is  Safe  Enough?   Determinants of
     Perceived  and  Acceptable Risk."   Draft  chapter in:   The  Management of
     Nuclear  Wastes,  L.  Gould and C.A. Walker,  eds.   Yale  University Press
     (no date).39~pp.

Smith, W.J.J.  Environmental Policy and  Impact Analysis:  Origins. Evolution.
     and Implications  for Government, Business, and the Courts.  The Confer-
     ence Board,  Inc., New  York, New York, 1973.

Sofranko, A.J., and W.M. Bridgeland.  "Agreement and Disagreement on Environ-
     mental   Issues  Among  Community   Leaders."   Cornell  Journal of Social
     Relations. 10(1):151-162, 1975.

Swan,  J.A.   "Psychological  Response to  the Environment."  In:  Environmental
     Quality  and  Water  Development.  C.R. Goldman,  J.  McEvoy, III and P.J.
     Richerson,  eds.   W.H.  Freeman  and Company,  San  Francisco, California,
     1973.  pp. 95-110.


                                     238

-------
Taviss,  I.   "A Survey  of Popular  Attitudes  Toward Technology."  Technology
     and Culture. 13(4):606-621, 1972.

Tognacci, L.N.,  R.H.  Wei gel,  M.F.  Wideen, and D.T.A. Vernon.  "Environmental
     Quality:  How Universal is Public Concern."  Environment  and  Behavior,
     4(2):73-86, 1972.

Tremblay,  K.R.,  and  R.E.  Dunlap.    "Rural-Urban  Residence  and  Concern with
     Environmental Quality:  A  Replication and Extension."  Rural Sociology,
     43(3):474-491, 1978.

Trigg,  L.J.,  0.   Perlman,  R.P. Perry,  and  M.P.   Janisse.   "Anti-Pollution
     Behavior:  A  Function  of  Perceived Outcome and Locus of Control." Envi-
     ronment and Behavior, 8(2):307-313, 1976.

Trop,  C.,  and  L. L.   Roos,  Jr.   "Public Opinion  and the  Environment."  In:
     The  Politics  of Ecosuicide,   L. L.  Roos,  Jr.,  ed.   Holt,  Rinehart and
     Winston, Inc., New York, New York, 1971.   pp.  52-63.

White,  G.F.   "Formation and  Role  of  Public  Attitudes."  In:   Environmental
     Quality in a Growing Economy,  H.  Jarrett,  ed.   The Johns Hopkins  Press,
     Baltimore, Maryland, 1966.   pp. 105-127.

Winham,  G.   "Attitudes on  Pollution and Growth in  Hamilton,  or 'There's an'
     Awful  Lot of  Talk  These  Days About  Ecology.1"   Canadian  Journal  of
     Political. Science. 5(3):389-401, 1972.
                                     239

-------
                                  APPENDIX  E

                     FACTORS  INFLUENCING DISCOUNT  RATES  FOR
                             ENVIRONMENTAL PROJECTS
 INTRODUCTION

      This  appendix  discusses  tne  selection of  a discount  rate for  puonc
 projects,  presenting the arguments for  and  against the use of  a soc-ial  dis-
 count rate that  is lower  than  the rate  of discount used to assess  private
 sector projects.    The  discussion also  addresses  the question of whether  or
 not the discount  rate used on  social  projects  should be raised to  ra^'lact
 risk.   (In this  context,  i.e.,  capital markets,  the tern "risk" ^s  usac  to
 denote any uncertainty.)

      Where a project  can  have  irreversible effects  that  extend beyond  •:•'
 generation, the choice of  a discount rate  becomes  more complex,  and  the whai:^
 concept of discounting  has been  questioned.   This appenaix  discusses  chese
 issues and develops a  pragmatic  solution to  the problem.

 THE SOCIAL DISCOUNT RATE

      If for  the  moment  one disregards the question  of  risk and  considers
 risk-free  projects, there  are two principal opposing views  on  the use  of a
 social discount rate that  is  lower than the business  rate.   On one hand,  it
 is argued  that  private market  decisions  generally  favor the short term and  do
 not make sufficient provision for  the future,  leading to  a  rate of consump-
 tion that  is  too  high  (Krutilla  and Fisher,  1975).   Hence  a lower or "social"
 discount  rate   is   proposed  in  order  to  adjust  private  preferences  for
 consumption versus  investment  or conservation (as expressed in  the  private
 discount rate)  to a time preference that is  deemed appropriate for society  as
 a whole (Marglin,  1963).   The  requirement  for a lower discount rate  to stimu-
 late  social  projects  (as  opposed  to  private projects)  stems from the  long
 time  scales  over  which some  social  projects operate (e.g.,  water  resource
 developments)  and   the pattern  of   expenditures  and  benefits  (which  are
 equivalent to income).  This  pattern  usually involves heavy  expenditures  in
 the early  years  of  a project with benefits,  usually small  in  the early years,
 that continue for a long time.

      Another  argument  for  a  low  social  interest  rate  relates  to  positive
•externalities.   Therefore,  capital put to  public  use  often  has  a  higher
 social rate of return  than  the  same capital put  to private use.   Hence it  is
 claimed that a  lower discount  rate  is needed  to stimulate  social  projects
 (U.S.   Congress,  Senate,  1974;  Musgrave and Musgrave,  1976).   Clearly,  the


                                     240

-------
extent of any 'externalities  depends  on where the boundary  for evaluation of
the  project is  drawn.   However,  as  Arrow  and Kurz  (1970)  point out,  the
benefits of cleaner water,  for  example,  may be so widely  dispersed  tiat it
would be  impracticable  to devise  a  way of charging for these  benefits,  and
correspondingly it would probably be  difficult to value the benefits.

     On the other hand, it  is argued  that the correct discount  rate  is  the
opportunity cost  of  capital  based on the  returns when  the  project resources
are  put  to  alternative  uses.   Thus  it  is claimed  that to use a low social
rate  of  discount  on public projects  will divert  capital  from  the  private
sector to the public sector, leading to a distribution of  investment that is
not  optimally  efficient  (Musgrave and  Musgrave,   1976).   In  a  real  world
situation,   further complications  are  introduced by  the distorting effects of
corporate income  taxes.   Some authors  consider that the  appropriate social
discount rate is  the before-tax  return in the private  sector (Baumol, 1963).
However,  others  argue  that  the  relevant discount rate  for use  on  social
projects could  be  either  the  before-tax or  the  after-tax  private  sector
return,   depending  on  whether  the  funds are  diverted from investment  or
consumption (Musgrave and  Musgrave,  1976).   In addition, there are a number
of ways  of adjusting the investment criteria when social and private discount
rates diverge.    Discussion  of  these  topics  is beyond  the  scope of  this
appendix,  and the reader  is referred  to the literature (see  Herfindahl  and
Kneese,  1974; Mishan,  1971;  Oasgupta,  Sen,  and Marglin, 1972; Krutilla  and
Fisher,  1975).

ADJUSTMENT OF THE DISCOUNT RATE FOR RISK

     In  assessing  business projects,  it  is a  common  practice to  raise  the
discount rate  (or the minimum acceptable  rate  of return)   in  order to allow
for  uncertainty  associated  with  the  project.   Uncertainty  can arise  from
numerous factors such as the  magnitudes of the expenditures and revenues,  the
timing and  duration  of  the phases of the  project as well  as the  possibility
of "catastrophic"  events—a  major uninsured  accident,  for  example.   Corre-
spondingly,   it  is  assumed  that  an  investor  will  demand  a  higher  rate  of
return (expressed as an  expected value) from what is perceived to be a risky
project  than  from  a safe one.   Thus  investment criteria  will  refect  the
investor's  risk  aversion by including  a risk premium  in the  discount rate.

     Similar arguments  can be applied  to public projects, and some  authors
hold -that   public  projects  should be   assessed in  exactly  the   same  way  as
private  projects  to avoid the capital  diversion  effect already mentioned.
The  alternative  contention  is  that  public  investment criteria should  not
include   a   risk   premium.   Several  arguments  can  lead  to  this  position,
including claims  that the  private capital markets are  so imperfect that they
give no useful  information about the individual's risk preferences, and that
many of  the risks in the private sector (such as moral  risks)  do not exist in
the  public  sector (Arrow  and  Lind,  1970).   However,  there  are  three major
arguments for a  risk-free  approach.   The first is that governments 'nvest in
a great number of diverse  projects,  which enable them  to pool risks to a  far
greater extent than  the  individual investor (i.e.,  a government  acts as  its
own  insurer).  Second,  a government  distributes the risk associated with  any
one  project over such  a wide range of individuals that  the total  cost of


                                     241

-------
risk-bearing is  insignificant.  The  third  argument is that the state is more
than  a  mere  collection of  individuals  and has  an  existence  and  interests
apart  from  its   individual  members; government  policy  therefore  need  not
reflect the  risk  aversion  of individual  preferences (Arrow and Lind.  1970).
Again,  the reader  is  referred to the  literature for  more  details of  this
debate  (see  Herfindahl  and  Kneese,  1974;  Mishan,  1971;  Dasgupta,  Sen,  ana
Marglin, 1972; Krutilla and Fisher, 1975).

INTERGENERATIONAL EFFECTS

     The preceding discussion on the  choice of a discount rate used  arguments
that  are  appropriate to the viewpoint  of the  present  generation.   However,
where a longer  time span is involved, society should also consider  the view-
point of future generations.  This is particularly important for any projects
that could or do result in  (1) irreversible changes,  such as moaifications to
the  natural   environment,   or  (2)  the   use  or  disposition  of  nonrenewaole
resources.   Construction (e.g., of a  dam or landfill) is irreversible when it
causes environmental changes.   Thus  almost any effect that persists for more
than one generation could be regarded as irreversible.

     Two of  the arguments  already presented  for a  low  social  discount rate
are  particularly  appropriate when intergenerational effects  ar^> considered.
These are:  (1) The view that society  is  more than a collection of individuals
and  hence  does  not have to be  risk  averse, and,  more important (2) the view
that  market  decisions  stress present consumption to the detriment of conser-
vation.   (Page  [1977]  provides,  in the  context of resource  conservation, an
excellent  and  extensive  discussion  of  the  implications  of  the  various
approaches to the choice of a discount rate.)   Though the second argument can
be  applied directly (in assessing the  benefits  of  resource  recovery),  the
general   view  that  one  generation   should  not  unduly  mortgage  another's
activities in return  for immediate benefits is applicable  to  the management
of any "nontreatable" waste.

      Krutilla and Fisher (1975) have  analyzed  some intergenerational problems
and  conclude that  where there  is less than  perfect altruism,  the  overall
optimum  use  of  a  limited  resource   will  not  be achieved,  because  of  the
inability of different generations to bargain  with each other.   Consequently,
each  generation  will  optimize  the use  of resources from  its  own viewooint,
and  each  succeeding generation will  wish to  revise the plan  to provide it
with  the maximum  utility available from the remaining resources.  Of partic-
ular  interest is  the  case  where an  option  demand (see Appendix C)  increases
with  time,  and a  project  that  is   justified  (using  the potential  Pareto
criterion) at t=0, may cease to be justified when evaluated at t=i^  (t^O) as
a  result of  the increasing benefits  of  maintaining the status quo.   Further-
more,  the  magnitude  of the benefits  (viewed  from t=tit  or  later)  of  not
undertaking  the  project could  be  sufficient  to  permit the  compensation of
early beneficiaries of the  project  and still  satisfy the potential  Pareto
criterion.    Thus where  an  irreversible action  is  contemplated, some addi-
tional  test  of efficiency  is  required  if  society is to  aspire to  altruism.
This  test,  attributed  to   Scitovsky,  is  discussed  in  Krutilla and  Fisher
(1975).   (Anderson  [1978]  also provides  a useful discussion,  from  the view-
point  of  resource  conservation, of  intergenerational  effects  with  more
emphasis on distributional  considerations.)

                                     242

-------
OBJECTIONS TO DISCOUNTING

     In the  above discussion,  the  concept of discounting was  not  an issue,
although Krutilla and  Fisher  (1975) show (in the  context  of the consumption
of a fixed  resource  stock)  that to achieve  altruism,  the  discount rate must
be  uniform  between  generations.    However,  several  authors  (concerned  with
empirical  cost-benefit  analyses)  have questioned  the  concept of discounting
on the grounds that it unreasonably penalizes future generations, or even the
later welfare of  the  present  generation.   The problem  is  that with any con-
ventional  rate of discount, even  a low social rate, the future is sc heavily
discounted that  after a  few  decades, distant  events  can be disregarded in
virtually every  case.   This  is no  great problem  for  a  conventional  project
(such as  the construction of an  incinerator), where  technological  oosoles-
cence is expected to  limit  the project's useful  life by a couple of decades;
but  it  can  present  difficulties  where  human  life or the  environment  are
involved.

     This  issue   has  been  pointed  up  by  a National  Academy  of  Sciences
committee as follows:

          There have  been long-standing debates as  to  the appropriate-
     ness of applying  a  discount  rate to effects  on future generations,.
     since any positive  rate  of discount will directly  discriminate in
     favor of choices  that  involve  bad impacts on later generations but
     not on earlier ones.   Again by way of example, if  the discount rate
     were 5  percent,  100 cases  of toxic poisoning 75 years  from  now
     would be equivalent  to about 3 cases today;  or 1 case today would
     be valued the  same  as 1,730 cases  occurring in 200  years,  or  tne
     same as the  current world population (more than 3  billion cases) in
     450 years.  Clearly, intergenerational  effects of these magnitudes
     are  ethically  unacceptable;   yet  they  might  be  made  to  appear
     acceptable if the traditional  social  rate of discount concept were
     used to discount future costs  to  compare  with  present benefits.
     Some   other   method   of  ethically  weighting  intergenerational.
     incidence of effects must-be  devised,
     (National Academy of Sciences,  1975:177)

     The committee  also  noted that  there  is as  yet  no  generally  accepted
method  for weighting  the intergenerational  incidence 'of benefits and costs.

     Another  National  Academy  of  Sciences  (1977) committee  also  indicated
that it had  problems  with discounting with respect to  the-valuation of .lives
exposed to radiation and proposed  that:                                   ... .

          Weighting factors"should  be applied to  those  terms which  may
     be undervalued  by  market place economics.   Typically,  these  are
     likely to include the  terms  which have  a component which involves
     people not able  to  take  part  in the decision-making  process.   The
     values of the weighting  factors  have to be established by society
     in general,  whether through  the political process,  public survey,
     or other means.   (National Academy of Sciences, 1977:69,70)
                                     243

-------
     Sociologists have,  however,  developed some alternatives  to  the utility
approach inherent in  discounting.   Thus  if the present generation desires to
minimize the  regret  of future  generations as  to the  present  generation's
choices, then  the  appropriate social  rate of  discount is  zero~i.e.,  all
generations  are   valued equally  over a  finite  planning horizon  (Scnulze,
1974).   Rawls (1971) argues that society  should focus attention on maximizing
the welfare of the  poorest individual.   This approach  has  been  developed by
Solow (1974) and  Phelps and Riley (1978),  but their arguments concentrate on
the consumption of nonrenewable resources,  and although they are  interesting,
they cannot be applied  directly to hazardous waste management problems (also
see Page,  1977).   Epp et al.  (1977)  suggest  the  use of a zero discount rate
for cost-benefit  analysis  of  pesticide  use, although they do not justify the
proposal.

APPLICATION TO HAZARDOUS WASTE MANAGEMENT

General

     The use of  discounting in empirical  studies involving intergenerational
effects  is clearly  problematical.   However, in a mixed economy,  the complete
abandonment of the  discounting concept  for public  decisions  could c^ll  for
some difficult  and  perhaps  unsatisfactory second  best ana, j-'s,^  as  dis-
counting is  implicit  in virtually all  business  decisions.   Discounting is
clearly  not  a topic  upon  which wide  agreement will readily  be  "eached,  yer
some solution  must  be  adopted to  proceed  with any numerate analysis.   The
author  will  therefore  offer  some  pragmatic  suggestions and  arguments  that
could  be  appropriate to  the particular  characteristics of  V.ontreatable"
hazardous wastes.   Note that  many decisions relating to treatable wastes are
reversible (assuming  that  costs sunk  in  physical  facilities are  disregarded)
and only involve a conventional time scale as opposed to an intergenerational
one.  Hence  in these  cases, the only problem  is  the choice between a social
and a market discount rate.

Intergenerational Discounting

     Where  intergenerational  effects  are  possible (e.g., effects  that stem
from environmental  threats),  two  approaches  could be  employed:   (1) not to
discount the effects  of threats,  or  (2) not  to further discount the effects
of  any  threats that occur after one generation.  In the second approacn, the
intention  is  that  all  costs  and  impacts  that occur during  a normal (single
generation) project  life are  discounted  in the usual way, but that no effect
is  further discounted if it occurs past this time.  Both approaches nave the
added  practical   attraction  that  they  eliminate  or  reduce  the  problems
associated  with   deciding  upon the time  at  which  a   threat  is  assumed to
materialize.   It is  difficult to  predict  the  time at which a  threat  that
arises  from  hazardous  waste  disposal (as  opposed  to  treatment,  etc.) might
become   a   reality.    It  is   true,  for   example,  that  given  sufficient
*  If the  social  and business  discount  rates diverge,  tnen  this  will cause
  deviations  from  Pareto  optional  conditions (see Section 5) with the result
  that  any optimum  reached  will be  what economists  refer  to as  a "second
  best" condition.

                                      244

-------
precipitation  and 'geohydrologic  data,  the  emergence  and  movement  of  a
leachate  from  some form  of  land disposal could be  predicted.   In practice,
however,  adequate data  are  not  likely  to  be  available—and  the  greatest
threats may  come from  unanticipated  sources such as  an  unrecognized inter-
connection  between  two  aquifers.   Effects  may be  cumulative  and  may  not
become apparent  until  seme  (probably  ill-defined)  threshold level is massed.
Where  the threat  relates to  an  irregular  or  random process  (such  as  an
uncommon  natural  event),  the  timing cannot be predicted,  although of course
it  might  be  possible  to  derive  an  expected value.    Mence   use  of  the
methodology  is  simplified  if  threat  timing  does  not  affect  the  results.

Rationale for Not Discounting

     While neither  of the two  approaches suggested above  can  be ricorcusly
justified  (unless  one   accepts   the   minimum  regret  criterion  mentioned
earlier),   arguments   in  their  favor can  be put forward.   First,  little  is
known  about  man's future uses  of  the  environment,  and  especially  those  to
which  it  may be  put after two or three decades (i.e.,  one generation later).
Society may  need  to   employ  some  resources  that  are currently  unused  and
little valued.   For example:   When might we need to extract either freshwater
or minerals from a saline aquifer?  When may we need and be ready to  farm the
ocean?  When  may  we   need  to use  the  land where  a  landfill   is presently
located?  An allowance  for  the  unknowns can be made by placing some form of
option value  on  them, and it is  not  unreasonable  to  suggest that i':s  value
increases with time.   This  increase would stem from the  increasing relative
scarcity  of  the  fixed  supply  of  environmental resources in comparison  to
nan's  growing  real wealth.   Thus the further into  the  future that one looks,
the higher are likely to be the opportunity costs  associated with irrevers-
ible  decisions.   Put  another way,  the materialization of  a  threat  (such  as
contamination  of an  aquifer that  might  be the only  low-cost water supply
available  to  meet  future increased  demand)  may  prove  to be  increasingly
costly as one  moves  further  into the future  (and alternative water sources
are put to other uses, raising the cost of the  next feasible supply).

     The  increasing opportunity  cost  hypothesized  above could be regarded as
balancing the  discounting effect,  leading to an argument for not discounting
when  the  impacts  of   these  threats are  expressed  in today's  values.   The
drawback  to  this  argument  is  that there  is  no  particular  reason  why  the
discount  rate  should  become  zero,  i.e., that the  two  effects should exactly
balance.   The  appropriate rate  (i.e.,  that which would  be  determined with
hindsight after the events have  occurred) might turn out to be a low positive
rate, zero, or even a negative discount rate (i.e., a growth rate).  But this
information  is  not  available,   and 'a  zero   rate  has   the  attraction  of
simplicity.

Rationale for Discounting Over One Generation Only

     The  concept  of discounting  over  a "normal"  project life, but thereafter
holding the discount  factor  constant  (i.e.,  not continuing to discount)  has
an even stronger pragmatic attraction.   The lives of many industrial projects
are  limited  by technological obsolescence,  either directly  via  the process
technology used,  or indirectly through changes  in the marketplace.


                                     245

-------
Consequently, industrial project  lives  are usually taken as  one  and a half,
two, or at  the  most three decades.*  Technologies  pass  through a succession
of  phases  as  they move  from  basic  research  or concept  development  to
commercial use.  Technologies  that  are  likely to be used  in the next decade
or  so will  generally  be well  advanced  along this  progression and  hence
comparatively easy  to  identify.   By contrast, some of the  technologies that
will be important after three decades, for example,  may not yet be conceived,
or they may  be  in  the very  early stages  of development making technological
predictions over this time scale most uncertain (Taylor,  1978).

     Thus it can be argued that current  valuations of resources will  probably
reflect their  utility over  the  next decade or so  with  reasonable accuracy,
but that beyond this period,  uncertainty becomes so great that some resources
could  be  seriously undervalued.  (Corresoondingly,  some  resources currently
in  use may be of  little  value by that  time.)  Discounting  during tne first
generation (say 20  or  30  years, which is about the same duration as  a normal
project life) but not continuing to  discount environmental  effects thereafter
is  equivalent  to  postulating  an  impact or  opportunity  cost  that starts to
rise when the period of high uncertainty is reached at  the end of the conven-
tional project life.  A major attraction of the approach, however, is tiKat it
simulates  tne normal  industrial  decisionmaking  procedures  (which   in  -lost
cases  do  not consider times  beyond one generation), yet it  doe~ -ot overly
discount the very distant future.

Conclusion

     These two proposed alternative  approaches for handling intergenerational
effects will  lead  to  present values for these effects  or threats that differ
by  up to  one order  of magnitude.    (For  example,  by  ratios of  3.4:1,  and
10.8:1 for discount rates  of 5 and  10 percent respectively, over a period of
25 years.)   Since   estimates  of the magnitudes (and, if used,  of the proba-
bilities) of  environmental  threats  are  likely  only to  be  order-of-magnitude
estimates, the difference need not be of great concern.   The author finds the
arguments  for  the  second  approach  (i.e.,   discounting   during the  first
generation,  but  not  thereafter)  particularly  appealing,  but  the  first
approach (not discounting at all) has the advantage of  simplicity.

     Both approaches could be difficult to apply in some  circumstances.   If
the  period  of evaluation   is  infinite,  any  cost that   continues  for  an
indefinite period will  have  an infinite present value,  unless discounting is
employed.   This presents  no  problem where a  lump-sum valuation  of an effect
is possible  (e.g.,  a  one-shot valuation of some damage,  such as a cleanuo or
replacement  cost);  but where  some  continuing cost is involved  (such as  the
maintenance  cost  for  perpetual   care,  or  a  loss  of  existence  value),  the
evaluation or planning  horizon would have to be  limited  if discounting were
not used.
* Of course, it is also true that at the high rates of discount commonly used
  to evaluate industrial projects, the contribution to net present value made
  by any cash flow beyond this point is often very limited.   This reduces the
  importance of  accurately predicting  the  technological life  of a project.


                                     246

-------
REFERENCES

Anderson, R.C.   Evaluation  of Economic  Benefits  of  Resource  Consecution
     Through  Recycling.    EPA-600/5- 78-015,  U.S.   Environmental  Prooac^i on
     Agency, Cincinnati,  Ohio, 1978.   51 pp.
Arrow, K.J. ,  and M.  Kurz.   Public Investment, The Rate of Return, and j
     Fiscal   Policy.   The  Johns  Hopkins Press,  Baltimore,  Mary lane  1970.
     246 pp.

Arrow,  K.J.,   and  R.C.  Lind.   "Uncertainty  and  the  Evaluation  of Public
     Investment  Decisions."   American Economic Review,  60(3): 364-363.  1370.

Baumol ,  W.J.    "On the  Social  Rate of Discount."  American  Economic  3av>w.
     58(4): 788-802,  1968.

Dasgupta, P.,  A.  Sen, and  S.  Marglin.   Guidelines  for Project  Eva'uation.
     United Nations,  New York, New York,  1972.   399 pp.

Epp, D.J., F.R. Tellefsen,  G.A.  Shute, R.M.  Bear, and K.P. Wilkinson. Icenti-
     ficatipn  and  Specification  of  Inputs  for  Benefit-Cost  Modeling of
     Pesticide Use.   EPA-600/5-77-012, U.S.  Environmental  Protection /Agency,
     Washington, D.C., 1977.  154 pp.

Herfindahl,  O.'C. ,  and A.V. Kneese.   Economic  Theory of Natural  Resources.
     Charles  E.  Merrill  Publishing  Company,  Columbus,  Ohio,  1974.   415 pp.

Krutilla, J.V. ,  and  A.C.   Fisher.   The  Economics  of Natural  Environments:
     Studies  in the Valuation of Commodity and Amenity Resources.   Tre  Jonns
     Hopkins  University Press, Baltimore, Maryland, 1975.  214 pp.

Marglin, S.A.   "The  Social  Rate of Discount and  the  Optimal  Rate of Invest-
     ment."  Quarterly Journal  of Economics,  77(1): 95-111, 1963.

Mishan,   E.J.    Cost-Benefit Analysis:  An  Introduction.   Praeger Publishers,
     New York,  New York,  1971.  374 pp.

Musgrave, R.A.  , and  P.B. Musgrave.   Public  Finance  in Theory and Practice.
     McGraw-Hill Book Company, New York,  New York, 1976.  796 pp.

National Academy of  Sciences.  Decision Making  for Regulating  Chemicals in
     the Environment.  National Academy  of  Sciences,  Washington, O.C. , 1975.
     242 pp.

National Academy of Sciences.   Considerations of Health Benefit-Cost Analysis
     for  Activities   Invo1vinq"Tonizing Radiation Exposure and Alternatives.
     IPA  520/4-77-003,  U.S.   Environmental  Protection  Agency ,  Was n i ngton ,
     D.C., 1977.  199 pp.

Page, T.  Conservation  and  Economic  Efficiency:    An .Approach to Materials
     Policy.    The Johns Hopkins University  Press, Baltimore, Maryland, 1977.
     286 pp.


                                     247

-------
Phelps,  E.S.,  and  J.G.  Riley.   "Rawlsian  Growth:   Dynamic  Programming  of
     Capital and  Wealth for  Intergeneration 'Maximin1  Justice."   Review of
     Economic Studies. 45(139):103-120, 1978.

Rawls, J.  A Theory of Justice.   The Bel knap Press, Harvard University Press,
     Cambridge, Massachusetts, 1971.  600 pp.

Schulze, W.  "Social  Welfare  Functions for the Future."  American Economist.
     18(1):70-81,  1974.

Solow, R.M.  "Intergenerational  Equity and Exhaustible Resources."  Review of
     Economic  Studies,   Symposium on the Economics of Exhaustible Resourcas,
     1974.   pp. 29-45.

Taylor,  G.C.   Methodologies   for   Characterising   Technologies.   EA-320.  ~^e
     Electric Power Research Institute. Pi!o Alto, California, 1973.   1-3 ?o.

U.S.   Congress,  Senate.   Evaluation  of  Techniques for Cost-Bene^'t •Va'ys-Js
     of Water  Pollution Control Programs and Policies,  Document  No.  33-132.
     U.S. Government Printing Office, Washington,  D.C., 1974.
                                     248

-------
                                  APPENDIX F

                           RISK AND DECISIONMAKING
     Considerations  of  risk  appear in  several places  in this  study.   The
purpose of  this appendix  is  to  interrelate the various aspects  of  ris'<,  to
provide  limited further  information about  some  of  these  aspects,  and  to
provide the  interested  reader with an entry  to the wealth  of risk-related
literature  that may  be  relevant to  decisionmaking about  hazardous  wastes.

ASPECTS OF RISK

     The  term  "risk"  has  different implications to different  people.   Rowe,
in a broad discussion of many aspects of risk,  defines risk as

      .  .  .  the  potential   for  realization  of  unwanted  negative  con-
     sequences of an event or combination of events to individual  groups
     of people or to physical and biological systems.
     (Rowe,  1975:1)

     Risk only  becomes  relevant  when  it  affects decisionmaking,  and  all
decisionmaking  involves some  element of risk,  even if the potential  negative
consequences of a decision are as  simple as failing to maximize utility  or
satisfaction in  a  choice  between  two competing products.  However, r'sk does
not  play  a  significant role  in every decision,  and  hence only a  limited
proportion of  decisions would be  perceived by  a decisionmaker  as being risk
related.  It is possible  to classify decisionmaking on  the basis of whether
the viewpoint used is predominantly personal, economic,  or societal.

     Risk taking in personal decisions might involve social behavior, such as
the  acceptability  of drug  taking  to peer  groups;  or it  might relate  to  an
individual's well-being  or  safety, such as  a decision  to risk  the  conse-
quences of swimming in polluted water.   Note that risks-are taken voluntarily
as a result of personal  decisions.            '   •         .             .

     Many business decisions involve .predominantly economic risks.  Tie basic
uncertainty need not be economic;  it could,  for example, be a question of the
ability of  the firm's management  to adequately control  a  venture, or it may
be that a project involves an unproven technology.   However, to the Business
firm,  the ultimate  consequences  of an  unsatisfactory  performance  TI  these
areas  will   be measured  in  economic  terms.    The  upwards  adjustment  of  a
required  rate  of  return to  reflect the perceived riskiness of a project  is
one way in which decisionmakers allow for economic risks.
                                     249

-------
     Another important aspect of economic risk-taking is insurance decisions.
Insurance spreads over many  individuals  the risk of  adverse  economic conse-
quences  from  what is usually  a comparatively unlikely event.   The  basis  of
insurance is that many individuals (or organizations) prefer to pay a premium
(which is in  excess  of  the expected value of the loss due to the transaction
costs involved) rather than  risk the chance of serious economic consequences
should specified events  occur.   These events are usually those over whicn the
insured  has little or no control, for example,  automobile  accidents (little
control) or  natural  disasters  such  as  floods  (no  control).   The  fact that
individuals  will  pay a  premium  to  obtain  insurance  illustrates   a  widely
observed phenomenon—that many people are risk averse, and prefer to minimize
the  potential  for adverse consequences.   However,  individuals  oftan exhibit
risk  proneness  in   some  decisions;  for example,  research  has shown  that
individuals   frequently   do  not  carry  insurance  against  high-loss,  low-
probability  events (discussed later in this  appendix).

     Where risks  have the  potential  to affect large numbers of persons, they
become  societal  risks.    However,  perhaps  the most important  aspect  of
societal  risk  is  that   the  risks are  largely incurred  involuntarily.   The
distinction  between  voluntary  and  involuntary risks appears to  be  of prac-
tical as well  as  philosophical  significance, as it appears that, for a given
economic benefit,  the public  is  willing to accept voluntary  risks  tnat are
greater than involuntary risks (discussed later in this appendix).

     Another  key  feature  of societal  risks  is  that  individuals   can  make
decisions that affect many others.  In some  cases these decisions are made by
persons  who  (implicitly  or  explicitly)  represent the  public,  but  in others
they  are made  by individuals  in  the light  of their  own  specific interests.
In  the  latter  case,  society often  finds a way to constrain  or control the
decisions so  that the public  interest  is  not unduly  damaged;  for example,
certain methods of disposing of hazardous waste might be prohibited by law as
being too harmful to the environment.

     There are many   situations  in which decisions are not purely  personal,
economic or  societal; in  some  cases all  three facets can be  present.   For
example, if an individual  decides to transport some obsolete and potentially
dangerous explosives over  the public highways,  this  could  be viewed  as  a
personal  risk-taking decision.    However, it also has  societal implications
via  the risks  to bystanders,  and  economic  implications  to  the individual
transporting  the  explosives, to  his dependents, and to  society  (by way of
road use costs, etc.).

     Many  risks  have  a  technological   element,   even  though  the  threat-
initiating event may be natural.  It may be  that the technology itself is not
completely  reliable   or  understood  (e.g.,  weather  modification),   that the
man-technology  combination  is  not  secure  (e.g., is  a  nuclear  power station
completely fail-safe?),  or  that  technology  is used  to mitigate or minimize
the   effects   of   natural   risks  (e.g.,  design  of  buildings  to  resist
earthquakes).

     Many  risk  data for  specific  technologies,  e.g.,  transport  accident
statistics,   are  readily  available.   Some  of these  that  are  applicable  to
hazardous waste  management  are  presented  in Appendix  8 on  environmental

                                     250

-------
threats.  A problem sometimes encountered is the quantification of risk where
historic data are unavailable or inadequate.  One approach proposed to remedy
this  deficiency  uses envelope curves to describe  risk-consequence relation-
ships (Kastenberg, McKone, and Okrent,  1976; Okrent and Whipple, 1977; Starr,
1972).

     Recently there  has  been much research on the risks that complex techno-
logical systems pose  to  mankind,  the preeminent example being electric power
generation  from  nuclear  or other  fuels.   (For  example,  see  Okrent,  1975;
Starr, Greenfield, and Hausknecht,  1972;  Rasmussen, 1975;  U.S.  Atonrc Energy
Commission,  1974;  Barrager,  Judd, and North, 1976.)   An  associated develop-
ment has been the growing use of technology assessments.  While the objective
of  a  technology  assessment  is  to evaluate all  the effects of the  use  of a
technology,  including direct  costs  and  benefits, the  emphasis  is  on  long
range consequences and side  effects  (National Academy of Engineering, 1972).
Consequently,  technology  assessment  is   closely  associated   with   risk
evaluation.   Technology   assessments  have  been  performed  in  such  areas  as
precipitation augmentation (Weisbecker,  1974), transport of liquified natural
gas and oil,  and  nuclear proliferation and safety  (U.S.  Congress,  Office of
Technology Assessment, 1977).               '           •

DECISIONMAKING UNDER UNCERTAINTY,  AND RISK AVERSION

     It is widely held that most individuals are risk averse.   In  the context
of  economics, this  follows  from the assumption that  the  marginal  utility of
wealth decreases  as an individual's wealth increases,  i.e., it is  of the form
shown in Figure F-l.
                             UTILITY +
                                               WEALTH
     Figure F-l.  Utility of wealth for a risk-averse individual.

(Source:  Slovic et al., 1977:238.)

     Consider  a decisionmaker faced with  a situation where he' has  a  choice
between (A) a  certain  $1 million payoff and  (B)  a probability p=0.5 of a $2
million payoff  together with  a p=0.5 of a zero payoff.   If the. decisionmaker
is  indifferent between choices  (A)  and  (B),  he is "risk  neutral"  since the
                                     251-

-------
expected value  of choice  (B)  is $1  million,  i.e.,  the same as  (A).   If he
prefers (A) to (B), he is termed "risk averse"; however, should he prefer (3)
to (A), he  is  considered "risk prone."  Similar  arguments  apply to expendi-
tures and losses of utility.   A risk-averse decisionmaker might, for example,
prefer to  spend $1  million  to avoid  a potential waste problem rather than
face  a 90-percent  probability of  needing no  expenditures together with  a
10-percent probability of having to spend $10 million to contain the problem.

     The above  statements  do  not provide any information about the extent of
the risk aversion  (or risk preference) that a  decisionmaker  exhibits.   This
could be measured  by determining the expected  value  of the uncertain situa-
tion at which  the  decisionmaker was indifferent between this and the certain
outcome.   To extend  the  waste containment example above, a decisionmaker who
will spend $2 million to avert the specified risk is risk averse; a aecision-
maker who will  spend $3  million to avert  the  same risk is more risk averse.
Maurer (1977)  discusses  ways  of  characterizing the  extent of  risk aversion
and discusses some common empirical  risk tolerance functions that can be used
to quantify risk aversion.

     Though  individuals  may  exhibit risk-prone behavior in some situations,
human behavior is generally characterized by risk aversion  for monetary risks
(Maurer,  1977).  Consequently,  some  authors have suggested  that ..Here choices
can be couched  in  purely economic terms, a specified degree of risk aversion
can be built into the decisionmaking process (Okrent and Whipple, 1977).  The
concept  includes  the suggestion  that  we  should  be  less  risk  averse  in
decisions concerning  essential  services (for example,  in deciding  to  have  a
public electricity supply) than in  peripheral  services that are not particu-
larly beneficial to  society  or where an almost equivalent  function could be
performed by alternative low-risk means (for example, substitutes for the use
of PCB's in transformers).

     The above discussion of risk aversion dealt with comparatively straight-
forward choices  that are  primarily economic in  nature.   Utility theory has
traditionally  been  used to  provide  a  theoretical basis  for decisionmaking
under uncertainty.   It also  facilitates the analysis of problems in which it
is difficult to  place a  monetary value  on  some outcomes.   (For example, see
Von  Neumann and Morgenstern,  1953;  Ellsberg,  1961; Keeler  and Zeckhauser,
1969; Fishburn,  1964, 1965; Raiffa,  1968; Fischer, 1975.)

     Research  on  insurance  and gambling  decisions  has developed empirical
data  about  an  individual's  behavior  when faced with a variety of uncertain
s-ituations,  many of  which are somewhat complex  (Slovic,   1972a;  Slovic and
Lichtenstein,  1968;  Slovic et  al.,  1977).   If the  individual  is assumed to
make logical decisions on the basis of utility, some of these findings can be
surprising.  For example,  it appears  that people are less  inclined to insure
against  low-probability,  high-loss events  than   high-probability,  low-loss
risks.   This could  be  construed  as  implying risk-prone  behavior, but  an
alternative  explanation  could  be  that  there  is  a  threshold  level  of
probability  for  a  damaging event below which  individuals  do  not consider it
worth  carrying   insurance  (Slovic et  al., 1977).   Extending  this result and
drawing  on  other work,   Slovic  et  al.,   conclude   that  when  dealing  with
insurance  against risks  from  natural  hazards,  there  are  many  social  and


                                     252

-------
psychological  factors  that bear  on  an individual's decison  (Slovic  et al.,
1977;  Slovic,  Kunreuther,  and White,  1974;  Kunreuther  and  Slovic,  1978).

     Because utility theory alone is not entirely satisfactory for explaining
observed  behavior,  an  alternative   psychology-based  approach,  now  termed
"bounded  rationality,"  has  been  proposed  (Simon,  1956,  1959).   Bounded
rationality  postulates  that individuals  do not think  probabilistically and
that  they  try to  avoid  the  necessity  of directly  facing  uncertainty  in
decisionmaking (Slovic,  Kunreuther,  and White, 1974).   It is assumed that an
individual's cognitive  limitations force  him to construct a simplified model
of the world,  and  that the decisionmaking  goal  becomes  "satisficing"  (i.e.,
finding an  acceptable  solution)  rather than optimizing.  There  is empirical
evidence that  the  way  in which individuals process information is central  to
their  decisionmaking  procedures,  which  supports  models   of  the  bounded
rationality type (Slovic, Kunreuther, and White, 1974;  Slovic, 1972b;  "versky
and Kahneman, 1974; Fischhoff,  1976).

     It appears  that  there is  a tendency for  economic decisions made  by
groups to be more  risky than by individuals, and that the risk-taking levels
of some individuals  increase following group  discussion  (Clark,  1971,  Kogan
and Wallach, 1967;  also see Slovic,  1972b).  This phenomenon has been called
"risky shift."  However,  it is a complex  subject,  and  groups will make less
risky decisions under some circumstances (Cartwright,  1973);

RISK AND SOCIETY'

     One'of  th'e most difficult tasks with which a decisionmaker can be faced
is the need  to decide  on-a  level of risk for society to bear.  (A decision-
maker  will  have  a  personal  risk aversion [related  to  the  perceptions  of
others as  to how  well   he  has  performed his job] that may  not coincide with
the  level  of  risk aversion  that  is  optimal for society.   But this personal
factor is disregarded  in the following discussion.)  The dilemma of choosing
between  comparative  safety  together  with  low  cost-effectiveness  versus
greater  risks   but  higher effectiveness—provided  nothing goes  wrong--has
spawned article titles such as  "How Safe is Safe Enough?"* (Fischhoff et al.,
1976;  Slovic  and  Fischhoff, in  press), and "Balanced Risk:   An  Approach  to
Reconciling  Man's  Need with his Environment"  (Wiggins,- 1975).   Ideally, one
would  like  to  be  able  to  optimize the  level  of risk to which the  public  is
exposed,  but  it  is not an easy task  to  determine what the public  really
wants, the difficulty being greatly compounded by the "fuiz.iness" of the data
(see Shinohara, 1976).

     Otway  and Pahner, who  addressed, risks in-the context of technological
systems, see risk on a series of levels:

     1.   Physical, biological  risks  to man and the environment;
     2.   The perception of these risks by individuals;
* It appears that  Starr (1969) was the first author to specifically pose the

  now well-known question "How Safe is Safe Enough?"


                                     253

-------
      3.    The  potential   risk   to  the  psychological  well-being   of
           individuals  based upon  these  perceptions;
      4.    The  risks  to   social   structures   and   cultural   values   as
           influenced  by   the   collective  psychological   states   of
           individuals.
           (Otway and Pahner,  1975:123)

      Otway uses the term "risk  assessment"  to  describe the process of incor-
 porating social  values  into risk-related,  societal  decisions  (Otway,  1977;
 Otway and Pahner, 1976; Otway,  1975).   Risk assessment is  seen as consisting
 of  two   major  components:    Risk  estimationandFTsk  evaluation.    Risk
 estimation involves identifying  and  quantifying the risks associated with  a
 particular option; that is,  it  is the  largely  technical  side of risk assess-
 ment.   Some  material  on risk  estimation  has  been  presented  in  Appendix  B.
•Risk evaluation  involves  determining  the  acceptability  of  these risks  to
'society  (Otway and Pahner,  1976;  Slovic and Fischhoff,  in press).

      There are two principal  approaches to risk evaluation  (Fischhoff et al.,
 1976):

      1.    The revealed preference method;
      2.    The expressed  preference method.

      The  basis  of  the  revealed preference method  is  the  assumotion  that
 through  trial and error  society has arrived at  a satisfactory balance between
 the risks and  benefits  connected with  any  activity.   Past  economic  risk and
 benefit  data  may therefore  be  used to reveal  patterns of  acceptable  risk-
 benefit  trade-offs.   Acceptable risk for a new  technology is considered to  be
 that  level   of safety  associated  with  ongoing activities  having  similar
 benefit  to society (Fischhoff et al., 1976).

      The  "expressed preference"   method,   which has   also  been  called  the
 "controlled experiment"  method (Otway and  Pahner, 1976) and the "psychometric
 survey"  (Fischhoff et al.,  1976;  Okrent and Whipple,  1977)  depends on using a
 survey or questionnaire  to  find individual  responses to risk or specific risk
 attributes (Okrent and Whipple,  1977).

      Starr (1976,  1972,  1969) is  the  best  known  proponent of  the  revealed
 preference approach.  He essentially compared  risk  of  death with a valuation
 of  the  benefits received  by the individual from the  relevant  activity.   In
 most  cases,  benefits were  valued by  assuming  that they  were equal to  the
 average  expenditures  by an  individual  on that activity.  This  analysis  led
 Starr to the following conclusions:

      . . . (i) The  indications are that the public  is  willing to accept
      'voluntary1  risks  roughly  1000  times  greater than  'involuntary'
      risks,   (ii) The statistical  risk of death from disease appears to
      be  a psychological  yardstick for establishing the  level  of accepta-
      bility of  other  risks,   (iii) The acceptability of risk appears to
      be  crudely proportional  to the third  power of the  benefits (real  or
      imagined).   (iv)  The   social  acceptance of risk  is  directly
                                      254

-------
     influenced by  public  awareness  of the benefits  of  an activity, as
     determined  by advertising,  usefulness,   and  the number  of people
     participating. .  .  .   (Starr, 1969:1237)

     The principal  criticism  of this approach is that the actual risk levels
are not  known  to  the  public, and  hence  judgments  will be based on perceived
risks  rather  than actual  risks.*   The literature  presents  several  other
criticisms  of  the revealed preference approach, including the argument that
the  sociopolitical  decisionmaking process  does not  reveal the  true  public
preference  for benefit-risk  trade-offs,  that revealed  preferences do  not
differentiate between what  is best and what is traditionally acceptable, and
that they illustrate past values, not current ones (Okrent and Whippla, 1977;
Linnerooth,  1975;  Fischhoff  et  al.,  1976).   Also,  Lave  (1972)  argues  that
Starr's  distinction between voluntary and  involuntary exposure is explained
by  the  "public good" nature  of  the  involuntary exposure  situations,.   In a
public  situation,  a  large  number of  persons  are   exposed  to  the hazard,
thereby  reducing  the  acceptable per  person risk in comparison with that for
an individual activity.

     In  addition  to the conceptual difficulties associated with the revealed
preference approach, Starr's quantitative findings are also s'ubject to doubt.
Otway  and  Cohen  (1975).  have , reanalyzed  Starr's data,   and  show  that  the
results  are  excessively .sensitive to the assumptions  made and to the way in
which the data are handled.  They conclude that simple mathematical relation-
ships, of the type  suggested  by Starr are unlikely (Otway and Cohen,  1975).
In particular, Otway  and  Cohen did not replicate Starr's well known ratio of
1,000:1  .for the  required-  benefits  from  risks  undertaken..involuntarily as
opposed  to  voluntarily.  Some  recent unpublished work  has  suggested  that a
3:1  ratio  may be  more  appropriate.    (Personal  communication,  P.  Slovic,
Decision Research, Eugene; Oregon, September 5, 1978.)

     The expressed  preference approach has been the subject of investigation
by  workers  at  Decision  Research  (Fischhoff  et  al.,  1976;  Slovic  and
Fischhoff,  in  press;  Slovic, 1978)  and by Otway  and others  (Otway,  1975,
1977; Otway  and  Pahner, 1976).   However, this approach has not, as yet, been
widely applied.   The  principal  disadvantage  is that it measures attitudes,
not  behavior,  and  that  it  is  difficult  to  project attitudes  to behavior
(Otway and Pahner, 1976; Okrent and Whipple, 1977).     .   .

     In  one  particularly  interesting  study, Fischhoff et al. (1976) examined
the way  in  which, for a given level  -trf benefit, risk attributes affected the
* This  criticism  is supported by work  that  illustrates that public judgment
  on  the  relative  mortality  risk from  various  causes (diseases, accidents,
  etc.) is  far  from accurate (Lichtenstein et al.,  1978;  Slovic, Fischhoff,
  and  Lichtenstein, 1976).   The  biases in the  responses  were -attributed to
  disproportionate  'exposure - to  the  various  causes  (due  to  variation in
  coverage  by the  news media)  and also  to  differences  in -memorability or
  imaginability  of-various  events (Lichtenstein  et  al.., 1978;  Slovic and
  Fischhoff, in press.)..   .
                                     255

-------
acceptability  of  a  risk.   Although  this  study  was  restricted to  a  sma'l
sample of members  of the League of Women Voters and their spouses in Eugene,
Oregon,  it  confirmed  Starr's  general finding  that risks  undertaken volun-
tarily were  more  acceptable than  nonvoluntary  ones; it also  found  that the
accentability  of  risk increased where  the risk was controllable,  familiar,
known, and  had  immediate  consequences.   These results  are  illustrated  in
Figure F-2  and have  the effect of  favoring  low-technology activities  over
high-technology activities  (Fischhoff et al.,  1976).  In addition, the degree
to which an  activity's risk was potentially catastrophic,  dread,  and likely
to be fatal  also negatively influenced acceptability (Slovic, 1978).

     Although the expressed preference  approach reflects the public's views,
it  has  already  been  shown  that  these  views  are   liable  to  be colored  by
inaccurate ideas on the magnitude of risks, etc.  One way to corract for this
would be  to provide  appropriate,  accurate data to  the respondents.   Slovic
and  Fischhoff  (in press)  provide  a  useful  listing of the  factors  that can
cause  both   erroneous views  and  erroneous analysis   in  dealing with  risk
estimates.

     The two  methods  discussed  above provide,  in  theory,  means  far deter-
mining optimum levels of risks  acceptable to society.  A method that i selves
aspects of both has  been advocated by Rowe (1977),  who propose  that for any
given activity the risk levels  inferred from historical  data must be adjusted
according to  the  degree of the  inequity caused by  a specific action U-2.,
risks to those  individuals who do not share the benefits) and the ability to
control each  risk.    Rowe  suggested  that  greater  inequity and  lass  control
should dictate lower permissible risks.

     Another method of determining risk acceptability is to argue that a risk
that  is reduced to  well  below the noise level  should not be of significance.
Thus  if  a   risk  is  far  smaller  than  that  posed  by  natural,  unavoidable
hazards, it is claimed that it should be acceptably  low* (Otway, 1975, 1977).
A variety of  historic data have been collected on  various risks that can be
used  for  this approach, many  of  which  are included in  the materials  cited
above.  In  addition,  Baldewicz et al. (1974) provide a fairly comprehensive
listing, and  Tonnessen and Cohen (1977) provide data on  naturally occurring
hazardous  materials  in  deep  geologic  formations  as   perspective   for  the
relative  hazard  of  burial  of  nuclear  wastes.   The  disadvantage  of  this
"comparative"  approach is  that the public may not perceive the relevant risk
to be  insignificantly low, and further, even if they  did,  they may not find
this argument acceptable.
*  Note  that while  Rasmussen (1975) showed that the  predicted  nuclear power
  risks  were  much lower than  the  risks  from a variety of  other  manmade and
  natural  hazards,  he  did  not  make  any judgment  on the  acceptability  of
  nuclear power risks.
                                     256

-------
                               HICK
                         ACCEPTABLE
                               RISK
                               LOW
                                   LOW        BENEFIT      HIGH
           HIGH
      ACCEPTABLE
            RISK
            LOW
                                                  HIGH
ACCEPTABLE
      RISK
                                                   LOW
       B
                LOW
                         BENEFIT
                                     HIGH
                                                      LOW
                                                               BENEFIT     HIGH
            HIGH
      ACCEPTABLE
            RISK
            LOW
                                                  HIGH
ACCEPTABLE
      RISK
                                                   LOW
                LOW      BENEFIT
                                     HIGH
                                                      LOW      BENEFIT     HIGH
   Figure F-2.   Determinants  of acceptable risks as indicated by revealed
                  and expressed preferences.
Source:   Slovic, P.   Decision  Research,  Eugene,  Oregon.
                                         257

-------
     The widely observed  and continuing opposition to nuclear power, desoita
a variety of  assurances from the  so-called  experts,  suggests thera ara seme
risks that much of the public simply does not want to accept.*  AltJiougn seine
people consider that there can be  unacceptable risks, the National Acaceiiy of
Engineering (1972) study on benefit-risk decisionmaking argues that 31' ris
-------
Fischhoff, B.,  P.  Slovic,  S.  Lichtenstein,  S.  Read,  and B.  Combs,   how Safe
     is Safe Enough?  A Psychometric Study of Attitudes Towards  Techno logical
     Risks and Benefits.Report #76-1,  Decision  Research,Eugene, Oregon,
     1976.29*'pp.[Also  published  in   Policy  Sciences.  9,  1973.    pp.
     127-152.]

Fishburh',. -P:C.   Decision and Value Theory.  John  Wiley  and  Sons,  Ir.c. ,  Mew
     York, New York, 1964.469 pp...

Fishburn,  P.C.   Modern Utility Theory:  1940-1970.   Technical  Paper RAC-TP-
     160,  ResearchAnalysis  Corporation,McLean,  Virginia,   1965.   47 pp.
     [NTIS:  A0625047]

Kastenberg, W.E.,  T.E. McKone,  and D. Okrent.  On  Risk  Assessment  :n the
     Absence  of Complete  Data.   UCLA-ENG-7677,  SchoolofEngineering  and
     AppliedScience,University  of  .California,  Los  Angeles,  California,
     1976". 54 pp.

Keeler, E., and  R.  Zeckhauser.   Another'.'Type of  Risk Aversion..   RM-5996-PR,
     The,Rand Corporation, Santa Monica, California,  1969'.  22~pp.

Kogan, ti.,  and M.A.  Wallach.   "Risk Taking as  a  Function of the  Situation,
     the Person, and  the. Group."   In:  New"  ' Directions  .in- Psychology:   III.
     Hol-'t,  Rinehart  arid''Winston,  Inc.., x New..York,   New  York,  1967.    pp.
     111-278.        '    .        ' •          "...   '..          	'
                        ••••••.
Kunreuther,  H.,  and  P.   Slovic.   "Economics','  PsychoTogy1,' and  Protective
     Behavior."  The American Economic Review. Papers and Proceedings,  68(2):
     64-69, 1978.                                      .  .    •

Lave,  L.B.   "Risk, Safety,  and the Role  of Government."  In:   Perspectives
     on Benefit-Risk Decision Making,  National  Academy  of Engineering,  eds.
     National   Academy of  Engineering, Washington, D.C.,  1972.    pp. 96-108.

Lichtenstein,   S.,  P.  Slovic,  B.  Fischhoff,  M.   Layman,  and  B.   Combs.
     Perceived Frequency of Lethal  Events.   Report if76-2,  Decision  Research,
     Eugene, Oregon, 1978.82 pp.

Linnerooth, J.   The  Evaluation  of  Life-Saving:  A Survey.    RR-75-23,  Inter-
     national   Institute for  Applied -Systems7 Analysis,  Laxenburg,  Austria,
     1975.  31 pp.

Louis  Harris  and Associates,  Inc.   A Second Survey  of  Public and Leadership
     Attitudes Toward Nuclear Power Development  in  the Dm ted-States!   Ebasco
    •Services. Incorporated, New York,  New York,  1976.

Maderthaner,.  R.,  P.   Pahner,. G.  Guttmarin",'"and H.J.  'Otway.'   Perception'of
     Technological  Risks:   The  Effect-of...  Confrontat-i-cm:• .-RM-.76-53,.  Inter-
     national   Institute for  Applied  Systems  Analysis,  Laxenburg,  Austria,
     1976.  24 pp.
                                     259

-------
Maurer, K.M.  A Parametric Utility Comparison of Coal and Nuclear Electr^'ty
     Generation.  UCLA-ENG-7719,  Scnool  of Engineering  ana  Applied Science,
     University 'of California, Los Angeles, California, 1977.  31 pp.

National Academy  of Engineering,  ed.   Perspectives of Benefit-Risk Decision
     Making.  The  National  Academy  of  Engineering, Wasnington,  O.C. ,
     155 pp.
Okrent,  D. ,  ed.   Risk-Benefit  Methodology  and  Application:   Some
     Presented  at  the  Engineering   Foundation   Workshop.  Seotemoer  ZI-
     26, 1975. Asilomar, Cali formal   UCLA- ENG- 7598,   Scnool  of  Engineering
     and Applied  Science,  University  of California, Los Angeles,  California,
     1975.  655 pp.

Okrent, D.  , and C. Whipple.  An Approach to Societal Risk Accaotanca Cr^a^a
     and Risk Management.   UCLA-ENG-7746,  School  of  Engineering  ana Apor'ea
     Science,  University  of  California,  Los  Angeles,  California,  1977.
     35 pp.

Otway,  4.J.   Risk  Assessment and Societal Choices.   RM-75-2,  International
     Institute for Applied Systems Analysis,  Laxenburg, Austria,  19~5. 23 op.

Otway, H.J.   "The Status  of  Risk Assessment."  A paper presented at the 10th
     International TNO  Conference on  "Risk  Analysis:   Industry,  Goverrment=
     and Society,"   February 24-25, 1977.   Rotterdam,  Netherlands.   15+ pp.

Otway, H.J., and J.J.  Cohen.   Revealed  Preferences:  Comments  on  tha Starr
     Benefit-Risk  Relationship?!   RM-75-5,   International   Institute  for
     Applied Systems Analysis, Laxenburg, Austria, 1975.

Otway, H.J., and M. Fishbein.  The Determinants  of  Attitude  Formation:  An
     Application  to Nuclear  Power.   RM-76-80,  International  Institute for
     Applied Systems Analysis, Laxenburg, Austria, 1976.  31 pp.

Otway,  H.J.,  and  M.  Fishbein.   Public  Attitudes  and  Decision  Making.
     RM- 77-54,   International    Institute   for  Applied   Systems   Analysis,
     Laxenburg, Austria, 1977.  29 pp.

Otway,  H.J.,  and P.O.  Pahner.   "Risk Assessment."   Futures .  Apri'  1976,
     pp. 122-134.

Pahner,' P.O.   "The  Psychological Displacement of Anxiety:  An Application to
     Nuclear  Energy."   In:  Risk-Benefit  Methodology  and Application: Some
     Papers  Presented  at the Engineering Foundation Workshop,  Septemoer 22-
     26. 1975. Asilomar. California. D. Okrent. ed.   UCLA- ENG- 7598, School of
     Engineering  and  Applied Science,  University of California,  Los Angeles,
     California, 1975.  pp. 557-580.

Raiffa,  H.   Decision  Analysis:  Introductory  Lectures  on  Choices  Under
     Uncertainty.   Addi son-Wesley, Reading,   Massachusetts,  1968.   332 pp.
                                     260

-------
Rasmussen,  N.C.    Reactor  Safety  Study:   An Assessment  of Accident  Risks
     in  U.S.  Commercial  Nuclear  Power  Plants.    WASH-1400,  U.S.   Nuclear
     Regulatory   Commission,   Washington,   DTE.,   1975.     207 pp.   [NTIS:
     PB248201]

Rowe,  W.D.    An  "Anatomy"  of  Risk.    U.S.  Environmental  Protection  Agency,
     Washington,   O.C.,  1975.   221pp.  [An updated  and extended  version  of
     this  has  been published as Rowe (1977)]

Rowe, W.D.  An Anatomy of  Risk.   John  Wiley  and Sons,  New York, New York,
     1977.  502 pp.

Shinohara,  Y.   Fuzzy  Set  Concepts  for  Risk Assessment.   WP-76-2,  Inter-
     national   Institute for  Applied Systems  Analysis,  Laxenburg,  Austria,
     1976.  25 pp.

Simon,  H.A.   "Rational  Choice   and  the  Structure  of  the  Environment."
     Psychological Review.  63(2):129-138, 1956.

Simon,  H.A.   "Theories  of  Decision-Making  in  Economics  and  Behavioral
     Science." The American Economic Review, 49(3):253-283, 1959.

Slovic, P.  "Information Processing Situation Specificity, and the Generality
     of Risk-Taking Behavior."  Journal  of Personality and Social  Psychology,
     22(1):128-134,  1972a.

Slovic, P.  "Psychological  Study of Human Judgment:   Implications for Invest-
     ment  Decision Making."  Journal of Finance, 27(4):779-799, 1972b

Slovic, P.  "Judgment,  Choice and Societal Risk Taking."   In:  Judgment and
     Decision  in  Public Policy Formation, K.R.  Hammond,  ed.   Westview PressT
     Boulder,  Colorado, 1978.   pp. 98-111.

Slovic, P., and  B.   Fischhoff.   "How Safe  is Safe  Enough?   Determinants of
     Perceived and Acceptable  Risk."   In:   The Management of Nuclear Wastes,
     L. Gould and  C.A. Walker,  eds.   Yale  University  Press,   (in  press).

Slovic, P., and  S.  Lichtenstein.  "Relative  Importance  of Probabilities and
     Payoffs in Risk Taking."   Journal  of Experimental Psychology Monograph.
     78(3, pt. 2):1-18, 1968.

Slovic, P.,  B.   Fischhoff,  and  S.   Lichtenstein.   "Cognitive Processes and
     Societal  Risk Taking."    In:    Cognition  and   Social  Behavior,   J.S.
     Carroll  and J.W.  Payne, eds.   Lawrence  Erlbaum  Associates,  Potomac,
     Maryland, 1976.   pp.  165-184.

Slovic, P.,  H. Kunreuther, and G.F.  White.   "Decision Processes, Rationality,
     and  Adjustment  to  Natural  Hazards."    In:   Natural  Hazards:   Local,
     National  and Global.  G.F. White, ed.  Oxford University Press, New York,
     New York, 1974.pp.  187-205.
                                     261

-------
Slovic,  P.,   B.  Fischhoff,  S.  Lichtenstein,  B.   Corn'gan,   and  B.  Combs.
     "Preference  for  Insuring  Against  Probable   Small  Losses:   Insurance
     Implications."  The Journal of  Risk  and  Insurance.  44(2):237-258, 1977.

Starr, C.   "Social  Benefit  Versus  Technological Risk:  What  is Our Society
     Willing to Pay for Safety?"  Scienca. 165(3899):1232-1238, 1959.

Starr, C.   "Benefit-Cost Studies  in Sociotechnical Systems."   In:  Perspec-
     tives of Benefit Risk Decision Making, National Academy  of Engineering,
     eds.    National   Academy  of   Engineering,  Washington,   D.C. ,   1972.
     pp.  17-42.

Starr, C.   "General  Philosophy  of  Risk-Benefit Analysis."   In:  Energy and
     the Environment:   A Risk-Benefit Approach, H.  Ashley, R.L. Rudman =nc! C.
     Whipple,  eds.   Pergamon   Press,  New  York,  New  York,  1976.  pp. 1-30.

Starr, C.,  M.A.  Greenfield, and  D.F.   Hausknecht.   "A Comparison  of  Public
     Health  Risks:    Nuclear  vs.  Oil-Fired   Power Plants."   Nuclear News,
     October 1972.   pp.  37-45.
                t
Tonnessen, K.A., and J.J.  Cohen.   Survey  of Naturally  Occurring  Hazardous
     Materials  in Deep  Geologic  Formations:  A Perspective  311 the Relative
     Hazard of Deep Burial  of Nuclear Wastes.   UCRL-52199, Lawrenca Livermore
     Laboratory,UniversityofCalifornia,   Livermore,  California,   1977.
     15 pp.

Tversky,  A.,  and  0.  Kahneman.   "Judgment  Under  Uncertainty:   Heuristics and
     Biases."  Science. 185(4157):1124-1131, 1974.

U.S. Atomic Energy Commission.   Comparative Risk-Cost-Benefit Study of Alter-
     native  Sources  of  Electrical   Energy.     WASH-1224,   U.S.  Government
     Printing Office,  Washington, D.C., 1974.

U.S.  Congress,  Office  of  Technology  Assessment.   List  of  Publications,
     OTA-P-58,  U.S.   Government  Printing  Office,   Washington,  D.C.,  1977.
     20 pp.

Von Neumann,  J.,  and  0.  Morgenstern.  Theory of Games and Economic Behavior.
     Princeton University Press, Princeton, New Jersey, 1953.  661 pp.

Weisbecker,   L.W.    Snowpack.   Cloud-Seeding,  and  the  Colorado  Rive**:  A
     Technology  Assessment of Weather Modification.   University of Oklahoma
     Press, Norman, Oklahoma, 1974.   102 pp.

Wiggins,  J.H.   "Balanced Risk:  An  Approach  to Reconciling Man's  Need with
     His  Environment."   In:   Risk-Benefit Methodology and Application:  Some
     Papers Presented at the Engineering  Foundation Workshop,   Septemoer 22-
     26. 1975, Asilomar. California.  D. Okrent. ed.UCLA-ENG-7598, School of
     Engineering and Applied Science,  University of California, Los Angeles,
     California, 1975.   pp. 521-555.
                                     262

-------