-------
-------
ACKNOWLEDGEMENT
The authors acknowledge the substantial contribution of ICF,
Inc. and Swep Davis to the appendicies on cost and financing of
indemnification programs. Preliminary work done by David
Alexander, Mark Grady, Carolyn Link, and Larry Seidel of AMS,
particularly on existing indemnification programs and financing,
proved to De a valuable basis for the report. The authors
gratefully acknowledge the extensive assistance and direction
provided by the project officer, Ellen Selonick Berick.
-------
DISCLAIMER
This report was prepared under contract to an agency of the
Pnited States Government. Neither the United States Government
nor any of its employees, contractors, subcontractors, or their
employees makes any warranty, expressed or implied, or assumes
,a.ny legal liability or responsibility for any third party's use
or the results of such use of any information in this report or
represents that its use by such third party would not infringe on
privately owned rights. Mention of trade names or commercial
products does not constitute endorsement or recommenation for
use.
-------
TABLE OF CONTENTS
Page
EXECUTIVE SUMMARY i
I. INTRODUCTION 1
II. BACKGROUND ON INDEMNIFICATION . . . . 11
III. EVALUATION OF EPA PROGRAMS 29
IV. EVALUATION OF POTENTIAL LOSS CATEGORIES 118
V. CONCLUSIONS 164
APPENDIX A—
BACKGROUND ON INDEMNIFICATION A-l
APPENDIX B—
REGULATORY AND LEGISLATIVE ALTERNATIVES TO
INDEMNIFICATION •. B-l
APPENDIX C—
CONSULTATION WITH INDUSTRY C-l
APPENDIX D~
INDEMNIFICATION FINANCING SYSTEMS D-l
APPENDIX E—
COST AND IMPACTS OF INDEMNIFICATION E-l
-------
INDEMNIFICATION STUDY
Executive Summary
EXECUTIVE SUMMARY
Page
A. Introduction ..... ii
B. Origin and Meaning of the Study Mandate ii
C. Background on Government Indemnification ill
D. Criteria for Evaluating the Need for
Indemnification Within EPA Programs iv
E. Existing EPA Indemnification Programs v
1. FIFRA § 15 Pesticide Indemnities v
2. Clean Water Act § 202(a)(3) Encouraging
Innovative Technology vi
3. Clean Air Act § 113(b) Unreasonable
Judicial Enforcement vii
4. Clean Water Act § 311(1) Encouraging
Spills Cleanup vii
F. Other Possible Indemnification Categories vill
1. The Concept of Loss Categories viii
2. Disclosure of Confidential Business Information. . . vili
3. Delays in Permit Processing ix
4. Conflicting and Overlapping Regulatory
Requirements ix
5. Emergencies ix
6. Unreasonable Administrative Enforcement x
7. Changes in Agency Policy x
G. Conclusions x
-1-
-------
INDEMNIFICATION STUDY
Executive Summary
A. INTRODUCTION
This study seeks to fulfill the mandate of § 25(a) of the Toxic
Substances Control Act (TSCA) to determine "whether and under what
conditions, if any, indemnification should be accorded any person as a
result of any action taken by the Administrator" of EPA. The study
began with a detailed investigation of existing indemnification pro-
grams (both within and outside EPA) and moved on to a consideration of
the potential need for other indemnification programs under any
current or projected activity at EPA. Both phases involved extensive
discussions with EPA staff and informal consultation with knowledge-
able people in the industries regulated by EPA. We could not conclude
that there is a total lack of situations justifying indemnification,
nor could we conclude that there are conditions which always Justify
payment. In the context of the strong bias against compensation of
regulated parties for the cost of government regulations, there was
insufficient evidence of a need for new indemnification programs at
this time.
B. ORIGIN AND MEANING OF THE STUDY MANDATE
The § 2S(a) study mandate has roots in the 1972 amendments to the
Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA). Those
amendments require EPA to indemnify pesticide owners suffering losses
when EPA suspends and then cancels a pesticide registration without
giving the owners an opportunity to sell or use their pesticide
stocks. The Senate had originally opposed this FIFRA provision, and
later attempted to repeal it and to prevent its inclusion in TSCA.
Instead of either repealing the FIFRA provision or extending it to
TSCA, Congress enacted the mandate for an EPA study of indemnifi-
cation.
The term "indemnification" refers to a particular kind of compen-
sation or reimbursement. One of the most well known examples is an
insurance contract, in which the insurer agrees to indemnify the
insured against loss arising from a specific event. An equitable
right of indemnity may arise where a person has discharged a duty
owed by him, which should have been discharged by another.
Indemnification, as contemplated by the study mandate, has a very
limited meaning, distinct from many other kinds of payment and compen-
sation that the government makes to private parties. For example,
awards made by a court of law, payments under contracts for services
or materials provided to the government, entitlement payments and sub-
sidies are all different from indemnification under this study.
Indemnification cannot be justified for all losses due to regula-
tory action by the government. But some losses do merit compensation
under our laws, (e.g., the taking of land to build a military base).
At the other end of the spectrum, most of the routine regulation by
-ii-
-------
INDEMNIFICATION STUDY
Executive Summary
the government imposes costs that are accepted as normal costs of
doing business and are not considered to be compensable. Clearly, the
government cannot protect the public health and welfare without
causing some burdens; and it could never afford to pay for all of
them. The mandate of the study is to explore the gray arias beyond
those losses already covered by compensation to determine whether
there are additional circumstances which might merit indemnification.
C. BACKGROUND ON GOVERNMENT INDEMNIFICATION
American law has generally not allowed compensation for losses
resulting from regulation. Even those laws abrogating immunity have
narrowly defined the areas in which the government can be sued.
Compensation under the Fifth Amendment has been limited to circumstan-
ces when full title to property is actually taken, all the normal
benefits of owning the property are destroyed, or an important prop-
erty right is transferred to another private person. Exceptions to
the Federal Tort Claims Act's waiver of sovereign immunity have pre-
vented tort suits for the results of most policy decisions, by
excluding activities involving discretionary functions, misrepresenta-
tions, and many intentional torts.
Congress has only infrequently provided indemnification, and a
decision to grant indemnification generally has not been regarded as a
precedent for its use in other circumstances.
Instances of congressionally-granted indemnification can be cate-
gorized according to three indemnification goals:
• compensating some inequity;
• deterring an agency from undesirable action; and
• encouraging or permitting desired private activity or other-
wise facilitating a public goal.
Most equitable indemnification situations have been congressio-
nally handled on a case-by-case basis through use of the private-bill
mechanism. However, when there are a sufficient number of indem-
niflable incidents of a predictable and regular type, Congress may
establish some type of indemnification process, rather than deal with
each situation individually. On the other hand, most prospective
indemnification programs established by public laws have been intended
as disincentives to government action or'as incentives to private-
party action. Some of the instances for which Indemnification has
been granted include:
• Indemnification for the consequences of product bans
(pesticides under § 15 of FIFRA); :"
-iii-
-------
INDEMNIFICATION STUDY
Executive Summary
• Agricultural indemnities to farmers (beekeepers, contaminated
milk, diseased livestock, and the contaminated cranberry inci-
dent of 1959);
• Reimbursement for oil and hazardous substance spill removal
costs (§ 311(1) of the Clean Water Act; the Outer Continental
Shelf Lands Act of 1978; and the Comprehensive Environmental
Response, Compensation and Liability Act of 1980);
• Indemnification to encourage use of innovative pollution
control equipment (§ 202(a)(3) of the Clean Water Act
authorizing EPA funding of replacement or modification of
failed innovative wastewater treatment facilities);
• Arts and Artifacts Indemnity Act (losses incurred by museums
exhibiting internationally loaned art);
• Government contractors' indemnification (defense contractors
are promised indemnity for unusually hazardous activities,
NASA reauthorization of 1981 covering indemnities for certain
catastrophic losses);
• Swine flu vaccine program (acceptance of liability exposure
rather than reimbursement of losses);
• Indemnification for government misrepresentation (private
bills covering the Mitzokami spinach and Marlin toys cases);
and
• Attorneys' fees and court costs for certain unreasonable
agency actions (§ 113(b) of the Clean Air Act Amendments of
1977 and the Equal Access to Justice Act of 1980).
Congress is currently considering an indemnification mechanism for
certain firms who suffered losses when the chemical flame-retardant
Tris was prohibited in children's sleepwear. Also, the Trial
Commissioner of the Court of Claims should soon file an opinion in the
case of California growers and canners seeking indemnification for
losses suffered when cyclamates were banned in 1969.
D. CRITERIA FOR EVALUATING THE NEED FOR INDEMNIFICATION
WITHIN EPA PROGRAMS
No detailed formula was developed which could automatically deter-
mine whether indemnification might be necessary in any particular set
of circumstances. Instead, a review of previous incidents and
existing indemnification programs revealed relevant factors for ana-
lyzing EPA indemnification programs and assessing whether their use
should be expanded or whether entirely new indemnification programs
are needed or any specific situations require indemnities. These fac-
tors include:
-iv-
-------
INDEMNIFICATION STUDY
Executive Summary
• The nature of the reasons supporting indemnification
(compensation on equitable grounds; disincentive to government
action; to encourage or facilitate private sector or regulated
party activity);
• Whether an anticipatory indemnification program (as opposed to
ad_ hoc decisions) Is necessary;
• Availability of preferable alternatives to indemnification;
• Whether Congress has considered the issue of compensation or
other remedies in the situation at hand and expressed a posi-
tion which is determinative of the question;
• The extent to which indemnification supports or conflicts with
other program objectives;
• Cost predictability and the amount of indemnity payout and
administration cost compared with the benefit derived;
• Whether the particular indemnification can be effectively
financed and administered in a manner which does not unaccep-
tably interfere with the administration of other programs.
E. EXISTING EPA INDEMNIFICATION PROGRAMS
The study is grounded in an examination of the indemnification
programs already in existence at EPA, and consideration of whether
they should be expanded to other situations. This has provided a more
concrete basis for evaluation than a theoretical construction would.
1. FIFRA § 15 Pesticide Indemnities
The § 15 provision appears to have been the result of legislative
compromise and a unique, multifaceted combination of factors. It does
not appear to be typical of other Indemnification programs in the
government. Therefore, it is difficult to generalize from the
experience under § 15.
The essential elements of § 15 are: that the government suspend
and then cancel the registration of a pesticide without allowing
owners to sell or use their inventories. There have been very few
emergency suspensions since 1972, and only three have met the criteria
for indemnification. Only one of those was a commercial pesticide.
EPA has never developed regulations or guidance documents to
implement § 15. Instead, claims-processing procedures were developed
in 1979 during the Silvex case, through an agreement negotiated pri-
marily between EPA and Chevron Chemical Company. Under the agreements
signed by 18 registrants, indemnity claims were funneled through each
-v-
-------
INDEMNIFICATION STUDY
Executive Summary
flc funding reserve to assure the availability of money when It is
needed. In addition, payment of indemnification is entirely
discretionary.
In the context of demonstration projects, a substitute for govern-
ment indemnification exists In the form of private Insurance which
could be purchased out of demonstration-project funds In order to pro-
tect against the risk of failure or of damage to the source's
equipment. In addition, contingent Indemnification clauses have been
provided to EPA clean-up contractors under Superfund. Nevertheless,
if the use of innovative technology is to be strongly encouraged, and
if Congress is willing to write an indemnification guarantee, then
such a provision might provide a substantial savings compared to the
use of commercial insurance. c
3. Clean Air Act § 113(b) Unreasonable Judicial Enforcement
Section 113(b) of the Clean Air Act provides that a federal court
may award costs of litigation (including reasonable attorney and
expert witness fees) to a party against whom erroneous or harassing
enforcement action Is brought by EPA. The legislative history of this
provision Indicates that It was intended to encourage persons who
would not otherwise do so for economic reasons to raise a legitimate
defense to such actions, and to serve as a deterrent against unreason-
able enforcement litigation.
Congressional concern over the potential for harassing or other-
wise "unreasonable" litigation under the Clean Air Act does not appear
to have been borne out in fact. Only one case has ever resulted in a
§ 113(b) award, and there are no indications that EPA's judicial
enforcement efforts are perceived to be arbitrary or harassing. EPA
enforcement policies and concurrence procedures also seriously
restrict the potential for such action in the future. For these same
reasons, indemnities for judicial actions in other EPA enforcement
programs are also not necessary.
4. Clean Water Act § 311(1) Encouraging Spills Cleanup
Section 311(1) of the Clean Water Act encourages nonliable
dischargers to assist in oil spill cleanup efforts by promising to
indemnify such dischargers for reasonable cleanup costs. Thus, a
cleanup incentive is provided to the party most knowledgeable of the
content and amount of a spill, and the party most likely to be first
aware that it has occurred.
Hazardous substance spills are also covered under § 311(1), but
cleanup is largely managed under the terms of the recently enacted
Comprehensive Environmental Response, Compensation and Liability Act
(CERCLA). This new statute provides up to $1.6 billion over a five-
year period to respond to harmful releases of pollutants in all of the
-vii-
-------
INDEMNIFICATION STUDY
Executive Summary
registrant. While the administrative processing of claims was still
very difficult for EPA, this arrangement made processing feasible
despite the large number of Silvex indemnity claims received from
retailers and distributors. After final review of each claims
package, EPA informed the registrant of the claims amount which the
Agency considered "appropriate." However, the Agency also stated that
it had no money with which to pay the claims. Subsequently, some of
the registrants successfully obtained indemnity payments from the
Court of Claims Judgment Fund. The Department of Justice recently
settled seven Silvex cases for a total of $18.7 million.
A variety of arguments fueled the congressional debate over § 15.
In the ten years since the provision's enactment, some of these argu-
ments have proved to be insubstantial, while others have taken on new
weight. Although existing evidence does not warrant repeal of the
§ 15 indemnity program, there are significant operational problems,
especially the program's lack of a specific funding mechanism. There
are also administrative problems that result from the infrequent need
to handle a very large volume of claims.
Extension of an indemnity program to the toxics area for loss of
inventory like that in FIFRA § 15 does not appear to be necessary. In
the toxics area there is no evidence of the kind of losses envisioned
by the drafters of § 15, due to the differences between the statutes
and their subject matter. TSCA contains many methods for preventing,
limiting and sharing the costs of regulation. Experience in imple-
menting TSCA is of course limited, but so far there are no patterns of
inequities that would justify establishment of an indemnification
program. Decisions on whether isolated inequities justify an indem-
nification payment should be made based on the facts of the case, and
cannot be forecast.
2. Clean Water Act § 202(a)(3) Encouraging Innovative
Technology
The indemnification provision of § 202(a)(3) of the Clean Water
Act was designed, in part, as an Incentive to encourage the use of
innovative and alternative wastewater treatment technology by local
governments. The statute gives EPA discretion to fund 100% of the
costs of modifying or replacing a facility that fails to meet perfor-
mance specifications and has significantly increased operating or
maintenance expenses. Although the provision has only very limited
application to provide replacement or modification funds for any
failed treatment facilities, its very existence has encouraged gran-
tees to use innovative or alternative technology. Funding of replace-
ment or modification costs has not yet occurred, because eligible
facilities have only recently become operational.
The practical effectiveness of this provision as an incentive for
the use of innovative technology is diminished by the lack of a speci-
-vi-
-------
INDEMNIFICATION STUDY
Executive Summary
pollution media. While CERCLA has no provision identical to § 311(1),
a similar incentive is possible under § lll(a), which provides for the
payment of certain claims, including necessary response costs incurred
by any person as a result of carrying out the national contingency
plan. However, these expenses must be approved in advance by EPA,
according to current EPA policy. This is to ensure cost control, and
effective management of cleanup operations.
The experience under § 311(1) demonstrates that it has been suc-
cessful in accomplishing congressional purposes, although on only a
limited basis. It cannot be determined yet whether CERCLA ultimately
will provide an effective alternative to § 311(i); and any con-
sideration of expanding § 311(1) should await further experience in
implenting CERCLA. c
F. OTHER POSSIBLE INDEMNIFICATION CATEGORIES
1. The Concept of Loss Categories
In addition to considering possible extensions or transplants of
existing EPA indemnification programs to other EPA program areas,
otter possible circumstances under which indemnification might be
d were evaluated. In this regard, categories more likely to
indemnifiable situations were identified and specific indemni-
fiable event losses were sought in order to sharpen the analysis and
assess the seriousness of the problems. In this process, each
environmental statute and its major regulations were reviewed, discus-
sions were held with past and current EPA staff, and industry sources
were consulted for identification of potentially indemnifiable situa-
tions. Five loss categories were selected for more detailed evalua-
tion to determine the need for indemnification.
2. Disclosure of Confidential Business Information
Industry has been quite concerned with the problem of losses
resulting from government disclosure of confidential business informa-
tion (CBI J required to be submitted to government agencies. Industry
has therefore attempted to persuade Congress and the federal judiciary
to define CBI better and to narrow the scope of authorized CBI disclo-
sure or use. However, government indemnification has not been sought
for losses due either to authorized or to unauthorized government
disclosure of CBI.
An EPA Indemnification program does not seem warranted to remedy
such losses. Both negligent and intentional unauthorized disclosures
are already arguably subject to a suit for money damages under the
Federal Tort Claims Act. Furthermore, EPA has detailed security pro-
in the pesticides and toxics areas designed to safeguard CBI
uthorized disclosure. There are also substantial criminal
pwities for intentional unauthorized disclosure. No claims have
-viii-
-------
INDEMNIFICATION STUDY
Executive Summary
been filed against EPA seeking compensation for these types of losses,
and there have been no prosecutions or disciplinary actions for inten-
tional and unauthorized disclosure. With regard to authorized
disclosures, congressional consideration is being given to amendments
to the CBI portion of the Freedom of Information Act, as well as to
amendments to FIFRA which relate to CBI disclosure.
3. Delays in Permit Processing
An indemnification program also does not seem warranted to deal
with the alleged problem of losses due to delays in government permit
processing. Both industry and EPA representatives concerned with the
problem of permit delays have focused on remedial solutions other than
indemnification. Furthermore, several studies have indicated that the
causes of delays in construction of new power plants and other
industrial facilities are not limited to regulatory problems.
Instead, many delays result from other problems such as financing,
labor and equipment availability, and management difficulties.
A. Conflicting and Overlapping Regulatory Requirements
The problem of conflicting and overlapping regulatory requirements
does not appear to be sufficiently significant to warrant an indemni-
fication program. No conflicting regulatory interactions warranting
indemnification appeared in a review of EPA programs. At worst, the
instances found represent compliance costs made necessary by reason of
other regulatory standards. This is not normally regarded as a basis
for compensation.
5. Emergencies
Although emergency action seems theoretically to provide increased
chances for creating an indemnifiable loss, our search did not reveal
any such Indemnifiable circumstances. The agency has rarely used its
emergency authority to curtail private operations. One situation
involved an air pollution episode in Birmingham, Alabama in 1971 in
which a temporary restraining order, obtained in the middle of the
night under § 303 of the Clean Air Act was in effect for 32 hours.
The other incident involved the temporary halting of operations of an
FMC plant in West Virginia that was leaking carbon tetrachloride into
the Kanawa River. Where emergency authority is used, a judge deter-
mines that the imminent threat to public health requires the unusual
action. A new and comprehensive law (CERCLA) governs response to
emergencies, generated by release of hazardous substances. It appears
to limit the possibilities for unwarranted burdens on private parties.
Until more experience indicates a need for assistance beyond Superfund
and the contingent Indemnity provision for clean-up contractors, no
new indemnification in this area program should be recommended.
-ix-
-------
INDEMNIFICATION STUDY
Executive Summary
6. Unreasonable Administrative Enforcement
An expanded indemnity program to compensate persons subject to
unreasonable administrative enforcement is probably not warranted.
This is primarily because there is no evidence that clearly unreason-
able enforcement exists on more than a very infrequent basis, and
because EPA's current enforcement policies and concurrence procedures
make such occurrences less likely than ever before. In addition, it
would seem preferable to accomplish Agency restraint through tradi-
tional avenues of judicial review, rather than indirectly by the
threat of indemnification. The former route would encourage loss
anticipation and avoidance, while the latter route, if designed
effectively, could result in excessive restraint and thus interfere
with the Agency's enforcement mission.
7. Changes in Agency Policy
No indemnification program appears to be warranted to deal with
the problems associated with changes in Agency policy. Such changes
are incidental to the activities and responsibilities of almost every
federal agency. For the most part, the parties subject to regulatory
programs at EPA have long been aware of the potential losses they may
incur because of changes in the programs. The regulatory process
allows for adequate forewarning of changes and such changes are a nor-
mal commercial risk of doing business.
G. CONCLUSIONS
No determinative indemnity formula was devised. Our study of sta-
tutory and other legal indemnification precedents and theories and our
consideration of the numerous factors that might be applied,
demonstrated that indemnity factors vary in applicability and impor-
tance depending on individual circumstances, and that these cir-
cumstances will almost always be different in some important aspect.
Congress must make such exceptional decisions, as it has in the past,
based on legislative criteria that are best defined at the time the
issue arises and the unique facts of the situation are known.
No new indemnification programs are warranted at this time. This
conclusion is based on findings that (a) no justification currently
exists for expanding existing EPA indemnification programs into new
areas of EPA jurisdiction, (b) an analysis of likely loss areas
demonstrates that there is no justification for concluding that there
will actually be losses that should be indemnified under any of the
criteria used by Congress in the past, and (c) there is little current
interest on the part of Industry in the availability of such a remedy
for actions taken by EPA.
-------
INDEMNIFICATION STUDY
Executive Summary
Other important conclusions concerning indemnification based upon
equitable grounds include:
• Whether indemnification is warranted for an inequity raises
complex issues that are best considered in the context of the
specific case and not resolved in advance by legislating a
prospective indemnification program*
• Procedural protections may reduce the need for indemni-
fication.
• Indemnification for inequities has not resulted in domino-like
extensions of Indemnification to to other situations.
• Some alleged inequities are not unique to EPA, and solutions,
if warranted, may need to be government-wide in scope.
• The private bill process can continue to provide relief of
last resort for Inequities arising from EPA action*
Other important conclusions regarding indemnification to provide
incentives or disincentives are:
• Indemnification has worked to provide incentives to parties
outside the government
• Indemnification for the purpose of restraining certain unde-
sirable agency action is not clearly justified based on
experience under existing EPA programs. Moreover, the poten-
tial for adverse impact warrants very careful consideration
before legislating an indemnification program for this
purpose.
-xi-
-------
INDEMNIFICATION STUDY
Section I
I. INTRODUCTION
Study Approach.
1. Overview. . 2
2. Phase I 2
3. Phase II 3
4. Phase III 4
5. Report 5
B. Major Issues. 7
I. "Any Person" 7
2. "Indemnification" 7
3. The Question of an Indemnification Formula 7
4. "Federal Laws Administered by the Administrator". . . 8
5. Other Environmental Losses. ... 8
C. Other Information . * 10
-1-
-------
INDEMNIFICATION STUDY
Section I
A. STUDY APPROACH
1. Overvlev
In 1976, Congress passed § 25(a) of Che Toxic Substances Control
Act (TSCA) requiring, in part, that the EPA Administrator conduct a
study of the need for indemnification under EPA-admlnistered statutes.
This study would be "for the purpose of determining whether and under
what conditions, if any, indemnification should be accorded any person
as a result of any action taken by the Administrator under [all
Federal laws administered by EPA]." This report responds to the
§ 25(a) mandate by clarifying the government indemnification concept
and by evaluating its suitability as a remedy for losses that have
been or might be sustained under specific EPA programs.
During this study, Interviews were conducted with persons repre-
senting the agency, industry, Congress, and envlromental groups.
Indemnification precedents (both in and outside EPA programs) were
studied in detail, and consideration has been given to the potential
need for indemnification for current and projected EPA activities.
The study was performed in three separate (but partially overlap-
ping) phases. Phase I involved an analysis of the indemnification
concept and a study of principal indemnification precedents. Phase II
involved a detailed analysis of existing EPA indemnification programs.
Finally, Phase III Involved an analysis of all EPA programs to deter-
mine whether an indemnification remedy might be warranted where it
does not already exist. For the most part, these three phases were
conducted independently, and have been treated separately in the
report.
2. Phase I
During Phase I, a thorough study of government indemnification was
conducted. This study included a review of non-EPA public and private
bills in which indemnification was granted, and cases referred (by a
special procedure) to the Court of Claims for recommendation. To
help define the vague language of the study mandate it also included
research into the legislative debates behind § 25(a) of TSCA and § 15
of the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) (a
major indemnification provision under EPA authority which was con-
sidered as a possible model for § 25(a) indemnification). Finally, it
included a thorough review of previous and current government policies
pertaining to the compensation of persons adversely affected by
government activity. This entailed an analysis of the Constitution's
Fifth Amendment, legislative exceptions to the principle of sovereign
immunity (especially the Federal Torts Claims Act), and judicially
created compensation precedents. It also entailed a review of all
potentially relevant legislative issues currently before Congress,
including such general topics as regulatory reform, regulatory relief
-------
INDEMNIFICATION STUDY
Section I
and federal budget control, as well as specific topics, such as
currently proposed bills to provide relief to asbestos-exposed workers
and to provide black lung benefits.
From this review, factors were identified which Congress appears
to have applied when determining the appropriateness of indemnifica-
tion. These are the compelling characteristics that influence a deci-
sion to deviate from the normal policy of not compensating for the
cost of complying with regulations. In broad categories they include:
• Fairness, equity, due process (e.g., detrimental reliance,
government error);
• The availability of alternative loss avoidance or relief
mechanisms;
• The potential impact (Interference or support) on agency
missions;
• Consistency with other congressional policies (e.g., the
policy against compensating regulated parties, the policy of
subsidizing certain constituencies); and
• Feasibility (e.g., costs, financing, administration).
As a part of this review, Instances of congresslonally supported
indemnification were categorized according to their major goals.
These categories include:
• Indemnification to remedy some inequity (e.g., losses due to a
government mistake that are not otherwise compensable under
federal law).
• Indemnification to deter an agency from undesirable action.
• Indemnification to promote some desirable activity by a party
other than the federal government (e.g., destruction of
diseased livestock to prevent spread of the disease).
These efforts provided an analytical framework for evaluating EPA
programs In Phases II and III. During the latter phases, indemnifica-
tion factors and goals were continually reevalutated and refined as
their use in actual application became more clearly understood.
3. Phase II
During Phase II, all EPA-implemented statutes were reviewed to
identify and categorize existing indemnification programs. Four EPA
indemnity programs were identified: (1) § 15 of FIFRA (compensation
for losses due to emergency suspensions and cancellations of pesti-
-3-
-------
INDEMNIFICATION STUDY
Section I
cide registrations); (2) § 202(a)(3) of the Clean Water Act (CWA)
(government grants to cover the costs of replacing innovative waste
treatment facilities that do not work); (3) § 311(i) of CWA
(reimbursement of cleanup costs for certain oil and chemical spills);
and (4) § 113(b) of Clean Air Act (CAA) (reimbursement of attorneys'
fees for unreasonable enforcement).
In each case, the legislative goals and the major issues of
legislative debate were researched and examined. Then, interviews and
research explored (a) how the program has worked, (b) whether
congressional goals were achieved, (c) whether the congressional
debate was accurate in its consideration of advantages and
disadvantages, (d) whether new advantages and disadvantages have been
suggested by actual experience, and (e) whether modification (or a
reevaluation of the need for) the program is warranted. This research
included discussions with staff from EPA, the Department of Justice,
and other governmental agencies involved In implementing indemnifica-
tion programs; discussions with industry officials affected by the
programs; and a review of appropriate regulations, guidelines,
memoranda, and indemnification decisions.
4. Phase III
Based on the research conducted in Phases I and II, a review of
all EPA programs was conducted to determine whether there have been or
might be instances in which Indemnification is appropriate. This was
approached from three angles. First, research was conducted to deter-
mine whether existing indemnification programs should be expanded into
other statutory areas. Essential justifications for each of the four
existing EPA indemnification provisions were defined and then compared
to other EPA programs to determine whether similar circumstances might
exist.
The second approach was to look for patterns of loss due to pre-
sent or threatened EPA action under any EPA statute. The logical
places to find mention of losses were in Industry complaints about
environmental regulation, administrative enforcement cases, rulemaking
'comments and petitions, and actions for judicial review, and through
discussions with EPA, industry and environmental organization
representatives. This research resulted In a list of six loss cate-
gories meriting close examination:
(1) disclosure of confidential business information,
(2) delays in EPA action (e.g., permit or variance review),
(3) conflicting requirements,
(4) emergency actions,
-4-
-------
INDEMNIFICATION STUDY
Section I
(5) unreasonable administrative enforcement, and
(6) changes in policy.
Major EPA programs were screened to identify possible contexts
where losses would be most apt to occur. Then, each category was ana-
lyzed for the programs involved by applying the factors identified
during Phases I and II, and conclusions were reached regarding the
probable need for indemnification.
A third task involved consultation with industry representatives
and other persons. One critical indication*of a need for indemnifica-
tion in specific instances is that affected persons express a need for
compensation. This, in fact, has been the traditional manner in which
any kind of special subsidization or compensatory relief has been
granted in the past.
No comprehensive survey was conducted. However, major trade
associations were consulted regarding their knowledge of incidents for
which indemnification might be considered appropriate. These persons
were also asked their opinion on the general desirability of new indem-
nification programs. Trade association representatives often referred
us to specific industry contacts. In addition, industry representa-
tives were called to discuss specific issues Involving their industry
as they arose. The list of organizations contacted during this study
is provided as Appendix C to this report.
5. Report
The results of all three phases of study are reported here. A
background section introduces the reader to the concept of government
indemnification, explaining its historical basis and tracing its
development. Factors affecting any indemnification decision are
briefly described, and an analytical method for evaluating EPA
programs is presented.
The evaluation of EPA programs follows and is presented in two
separate sections. First we examine existing EPA indemnification
programs and the need for extending these programs to closely anala-
gous circumstances. Second, other loss categories are evaluated for
all EPA programs.
At the end of the report, conclusions are presented, along with a
summary of findings supporting the conclusions and an explanation of
the reasoning involved.
Appendix A contains detailed background information which supple-
ments Section II. Appendix B discusses various remedial or loss
avoidance mechanisms which may be alternatives to indemnification. As
noted, Appendix C presents a list of the organizations contacted
-5-
-------
INDEMNIFICATION STUDY
Section I
during this study. Appendix D discusses means of financing indemnifi-
cation programs. Appendix E presents a methodology for estimating the
cost, including two examples that illustrate the problems.
-6-
-------
INDEMNIFICATION STUDY
Section I
B. MAJOR ISSUES
In addressing the requirements of § 25(a), numerous issues of
interpretation were raised. The major issues are described below.
I
1. "Any Person"
According to § 25(a), the goal of the study would be to determine
whether indemnification should be accorded "any person as a result of
any action taken by the Administrator." An issue for interpretation
was whether "any person" should include all persons potentially
affected by EPA actions, including not only the directly injured
party, but also persons dependent on that party to some degree (e.g.,
employees, purchasers), or persons incurring injury indirectly because
of the action's effect on the directly injured party (e.g., consumers,
local community).
This question was resolved by limiting the study to directly
injured parties. For example, a ban on the sale and use of pesticides
would directly affect both manufacturers and growers, but only
indirectly affect the farm product purchasers. The latter type of
injury has not been covered In this study. This restriction in scope
was necessary to allow for achievement of reasonable project goals
within an acceptable budget and timeframe. Moreover, the need to
reimburse indirect loses may be evaluated, if necessary, in the same
analytical framework applicable to directly injured parties.
2. "Indemnification"
Another major issue involved the definition of "indemnification."
Should the study be restricted to circumstances involving reimburse-
ment of a person for claims paid by that person (the traditional
definition), or should it Include, in addition, other costs incurred
as a result of the agency action? A review of the legislative history
of § 25(a) indicated congressional interest in the need to reimburse
any costs associated with EPA action, and the study has been struc-
tured to consider all circumstances involving any reimbursable costs.
3. The Question of an Indemnification Formula
One way to fulfill the mandate of § 25(a) would be to define
criteria, assign weights to them and fashion them into a formula that
could be applied to any loss situation to determine whether or not
indemnification was warranted. An attempt was made during this study
to design such an indemnification formula. In the end, however, no
predictive formula could be stated. From this attempt we concluded
that certain criteria should always be considered, but the issue of
whether indemnification Is warranted is ultimately a question of
policy and circumstances.
-7-
-------
INDEMNIFICATION STUDY
Section I
Much of the report focuses on whether a prospective indem-
nification program is warranted. The criteria governing decisions to
establish indemnification programs are more objective and easier to
generalize than those applying to individual payments. However, this
does not reflect a conclusion that there are no Individual circumstan-
ces in which indemnification is appropriate. Instead, this report
attempts to provide some guidance on how such decisions should be
made.
4. "Federal Laws Administered by the Administrator"
Although the study requirement pertains to all federal laws imple-
mented by EPA, only major environmental statutes were considered, and
among these, only those provisions pertaining directly to environmen-
tal protection were evaluated. Thus, among the laws and provisions
excluded from consideration in this study are those concerning civil
rights, contracts, civil service, employment, and environmental con-
sultation to other agencies.
5. Other Environmental Losses
In the course of research for this project, several examples of
environmentally related losses were brought to our attention. For a
variety of reasons, they do not fall into the scope of the study man-
date which Is to consider losses caused by actions of the
Administrator under any of the laws administered by EPA. As the
reader of this study is likely to be familiar with at least some these
examples, we set forth the reasons that they do not appear in this
study:
• Health effects of old hazardous waste sites. Injuries from
hazardous waste sites are directly caused by those who build
and maintain them and only indirectly by the government. If
the Injury was caused before EPA had been given regulatory
authority, there would be no basis for a claim against the
Agency. This would not fall into the category of formal deci-
sions not to take action which have been considered Agency
actions for the purposes of this study.
• Conflicts with state common law. A state court ordered the
owner of a hazardous waste disposal site to close it and
dispose of the waste elsewhere. Even though the disposal site
had been in compliance with EPA regulations, it was found to
be a nuisance under state law. The loss suffered by the owner
was caused by state action and not by any action of EPA.
• Urea-formaldehyde foam. The Consumer Product Safety Commis-
sion recently banned the use of UF foam, and this has caused
direct losses to firms who manufacture and install it. These
are the kinds of losses which this report considers, but the
-8-
-------
INDEMNIFICATION STUDY
Section I
ban came after we had concluded our research and the situation
was too uncertain at earlier stages to analyze. The losses
caused to homeowners by reduction in the value of their homes
is speculative and indirect and also does not fall within the
scope of the study.
Asbestos insulation. There are several pieces of legislation
proposing compensation for workers who were exposed to
asbestos on the job and have suffered health affects as a
result. .The reason for the bills Is that neither the
insurance market nor workers' compensation now appear likely
to provide a satisfactory response to this problem. In
addition, the government was the employer for some of the
exposed workers in its shipyards. Obviously, this is not
related to any EPA action. However, the problem of school
districts having to identify and remove insulation in their
schools pursuant to EPA regulation is mentioned in this study.
-9-
-------
INDEMNIFICATION STUDY
Section I
C. OTHER INFORMATION
Information for this study was collected from November 1978
through May 1982. During this time, significant legislative and
agency activity relevant to this study occurred — including the pro-
posal and elimination of government indemnification programs, the pre-
sentation of major indemnification claims under § 15 of FIFRA, and the
initiation of new federal regulatory and budget policies that will
affect the future consideration of indemnification programs by
Congress. This report is largely up-to-date on such issues. However,
the political backdrop to regulatory relief is still changing, and the
indemnification issue primarily involves policy judgments affected by
evolving considerations.
-10-
-------
INDEMNIFICATION STUDY
Section II
II. BACKGROUND ON INDEMNIFICATION
Page
A. The Study Mandate Under TSCA § 25(a) 12
1. The Statutory Provision 12
2. Meaning and Origin of the Study Mandate 12
B. History of Indemnification by the Government 15
1. Traditional Limitations 15
2. Examples of Government Indemnification 16
Footnotes 22
-11-
-------
INDEMNIFICATION STUDY
Section II
A. THE STUDY MANDATE UNDER TSCA § 25(a)
1. The Statutory Provision
Section 25(a) of TSCA provides:
(a) INDEMNIFICATION STUDY.—The
Administrator shall conduct a study of all
Federal laws administered by the Administrator
for the purpose of determining whether and
under what conditions, if any, indemnification
should be accorded any person as a result of
any action taken by the Administrator under
any such law. The study shall-
(1) include an estimate of the probable
cost of any Indemnification programs which
may be recommended;
(2) include an examination of all viable
means of financing the cost of any recom-
mended indemnification; and
(3) be completed and submitted to
Congress within two years from the effec-
tive date of enactment of this Act.
The General Accounting Office shall review the
adequacy of the study submitted to Congress
pursuant to paragraph (3) and shall report the
results of its review to the Congress within
six months of the date such study is submitted
to Congress.
2. Meaning and Origin of the Study Mandate
The genesis of the § 25(a) study mandate occurred In 1972 when the
92nd Congress enacted § 15 of FIFRA requiring EPA to indemnify certain
pesticide owners suffering losses due to EPA's emergency suspension
and cancellation of a pesticide registration. This indemnification
provision was supported by the House Agriculture Committee and Initi-
ally opposed by the Senate, but the Senate ultimately accepted the
provision in order to get the pesticide bill passed.2
Subsequently, in the 93rd Congress the Senate attempted to repeal
the pesticide indemnification provision and prevent its inclusion in
toxic substances control legislation.-* The attempt was a part of a
Senate amendment to an unenacted toxic substances control bill, which
also would have required EPA to commission an independent study of
indemnification to be conducted by a university or independent
research center.^ Although TSCA did not pass during the 93rd Congress,
it was enacted in 1976 during the second session of the 94th Congress
and included the study mandate noted above.^
-12-
-------
INDEMNIFICATION STUDY
Section II
While the term "indemnification" was not expressly defined in
§ 25(a) aa enacted, a predecessor bill which passed the Senate defined
"indemnity" as:
. • . any payment made to a person as reim-
bursement for loss or damage other than a
payment made in accordance with the judgment
of any Court in an action brought at common
law or under the [Federal Tort Claims Act].6
Thus, the "indemnification" with which this study is concerned
Involves compensation or reimbursement for loss or damage incurred by
any person as a result of an action by the EPA Administrator in admin-
istering Federal laws. Excluded from this definition is any payment
that the party might receive from the government through a judgment in
a court of law. Also excluded are a variety of federal government
payments which do not Involve reimbursement for loss, such as:
• Payments to government employees or contractors for public
services.
• Entitlement payments and subsidies to particular groups,
Including the poor, the elderly, veterans, and farmers.
Sometimes these benefits take the form of loans, loan
guarantees, or tax credits.
In addition, some federal programs are designed to cover the
risk of various losses and provide reimbursement payments out of an
insurance reserve funded by premiums paid by the parties protected
from potential loss. The federal government is involved in insurance
or reinsurance covering crops, floods, crime, urban riots, nuclear
disasters, bank or savings and loan failures, war risks, mortgages,
health and medical, unemployment and exports. To the extent that
payouts are funded by premiums, instead of by payments from the
federal treasury, these insurance programs are distinct from govern-
ment indemnification.
Similarly, restoration funds such as those available under
Superfund which are supported by an "injury tax" or a surcharge on a
product or activity related to the purpose of the fund (e.g., cleaning
up hazardous waste sites) are distinct from government
indemnification." Furthermore, disaster relief payments or other
financial support such as loans or loan guarantees in emergency cir-
cumstances are not within the scope of government indemnification con-
templated by this study. Nor is victim compensation or emergency
relief in environmental Impairment type situations.'
Indemnity payments by the federal government thus consitltute a
very small portion of the overall total of government payments. Even
among federal mechanisms designed to protect against risk, government
-13-
-------
INDEMNIFICATION STUDY
Section II
indemnification is only one small part, because premium-funded federal
insurance mechanisms and restoration funds cover a wide variety of
potential loss categories.
-14-
-------
INDEMNIFICATION STUDY
Section II
B. HISTORY OF INDEMNIFICATION BY THE GOVERNMENT
The following brief background information is documented and
discussed in detail in Appendix A and in the portions of this report
covering existing EPA indemnification programs.
1. Traditional Limitations
The concept of sovereign immunity — which bars a suit against
the government except with its permission — provides a key to
understanding the question of indemnification by the government.
Sovereign immunity protects the government from being legally forced
to provide compensation to injured parties.1*-* Nevertheless, there are
areas where, either by constitutional provision or by statute, the
government has waived its immunity and can be sued for money damages.
For example, the Fifth Amendment requires compensation when private
property is taken for public use, and contract claims and some tort
suits against the government have been expressly authorized by
statute.
Such judicial actions for damages, however, are narrowly pre-
scribed, and regulated parties usually are not paid by the government
to comply with regulatory requirements. Nor are they normally compen-
sated for the burdens of regulation or the adverse effects of govern-
ment actions. Although the Federal Tort Claims Act allows suits
against the government for operational negligence, such as careless
driving, liability suits against the government are not authorized for
negligence in performing discretionary functions such as promulgating
regulations.11 The government is also protected from damage suits
based on misrepresentations made by government employees.1^ Further-
more, courts have consistently refused to interpret statutes as
implying private rights of action against the government for
damages." Moreover, the Anti-Deficiency Act1* precludes executive
branch employees from making a promise of indemnification without
an express congressional authorization and appropriation.
These traditional limitations on indemnification for the adverse
effects of governmental actions stem, to a large extent, from a fear
of creating precedents which would require government indemnity pay-
ments for a wide range of regulatory activities, and from the belief
that situations that might require indemnity should be avoided so as
to make such payments unnecessary. Some of the arguments against
paying compensation to parties injured by regulatory activities are:
(1) Government could not function if it had to compensate par-
ties for all the regulatory burdens which it imposes, *
(2) The essential character of government regulation involves
the uncompensated adjustment of individual rights for the
public good;1"
-15-
-------
INDEMNIFICATION STUDY
Section II
(3) Business risks and regulatory liabilities have tradition-
ally been placed upon private parties;
(4) Indemnification can be disruptive of governmental program
objectives;
(5) A decision to indemnify in one circumstance can set a
costly precedent requiring compensation payments in many
additional circumstances; and
(6) Alternative remedial devices which avoid loss or lessen
harm are usually preferable to indemnification.*'
Since the enactment of § 15 of FIFRA in 1972, no new prospective
programs have been enacted for the purpose of compensating a private
party for commercial loss due to anticipated agency action. Instead,
Congress has tended to put the risk of loss with the party who engages
In activity requiring regulation. To protect against unforeseeable
harm, Congress has established procedures requiring ample notice to,
and participation of, affected parties in the development of regula-
tions; and to protect against unintended harm, Congress has provided
for waivers and exceptions, as well as for timely administrative peti-
tions and judicial review. Most regulatory statutes implemented by
EPA contain an ample variety of these loss avoidance and special
relief mechanisms. Moreover, inquires of representatives of
Industries affected by EPA statutes disclosed no demands for Indemni-
fication from unforeseeable or erroneous regulation (other than events
subject to existing indemnification authority).
The option of indemnifying regulated parties for losses due to
EPA action has received very little attention from the current
Congress, since industrial losses can be prevented by relaxing regula-
tory requirements, and an indemnification program could pose an even
greater drain on the federal budget. The concern over budget is par-
ticularly evident in the recent proposals to Congress for fiscal year
1983, which would cut back existing indemnification programs at the
Department of Agriculture for diseased livestock and contaminated
milk.18 Last year, Congress agreed to eliminate the Agriculture
indemnification program for beekeepers.1'
2. Examples of Government Indemnification
a. Indemnification Categories
Indemnification has been used in a variety of circumstances
and for one or more of the following reasons:
• remedying inequities;
-16-
-------
INDEMNIFICATION STUDY
Section II
• creating disincentives to inappropriate or unreasonable
government action; and
• creating incentives for action by parties outside the
federal government to facilitate achievement of a
public goal.
Some congressional Indemnification determinations have
occurred after the fact of loss. Others have Involved prospective
programs designed as incentives or disincentives. Most of the equity
determinations have been made on a case-by-case basis, unless a clear
and regular pattern of loss has been Identified which would warrant an
indemnification program.20
b. Indemnification on Equitable Grounds
Indemnification for equitable reasons includes the payment
of compensation for the taking of private property for public use.
Under the Fifth Amendment of the Constitution, a compensable taking
occurs only when full title to property is actually expropriated, the
normal indicia of property ownership are all destroyed, or an impor-
tant property element is expropriated for the use of another.21 The
seizure of contraband or substances injurious to public health does
not constitute a taking for which compensation is required.
The Federal Tort Claims Act, originally passed in 19A6, par-
tially waives the government's Immunity from suit and permits negli-
gence actions to be brought against it, except for discretionary acts,
misrepresentations, and certain other specific exceptions.23
Similarly, the Tucker Act authorizes contract and other non-tort money
claims to be brought against the government in the Court of Claims.24
Some parties injured by government actions have successfully
sought equitable relief from Congress by means of private bills
covering specific cases.25 Sometimes Congress refers such matters to
the Court of Claims commissioners for an advisory determination, and
sometimes it confers special jurisdiction on the court for a particu-
lar case.26 Examples of such congressional relief for the adverse
consequences of government error include the Mlzokami spinach case^'
and the Marlin toys case.28 in each of those situations, the misrep-
resentation exception to the Federal Tort Claims Act prevented a
negligence action from being maintained. In the spinach case, the
Mlzokami brothers were awarded more than $300,000 in damages suffered
when the FDA incorrectly found certain spinach to be contaminated by
the pesticide heptachlor. In the Marlin toys case, more than $40,000
was awarded as compensation for loss that occurred when a toy was
incorrectly listed by the CPSC as being banned.
The CPSC ban on children's sleepwear containing the chemical
flame retardant Tris is a recent instance in which Congress has
-17-
-------
INDEMNIFICATION STUDY
Section II
indicated that indemnification can be appropriate to correct an
inequity.^ Although the first Trie bill was vetoed by President
Carter, a second bill passed the Senate last year and is awaiting
action by the House. In the Tris case, the CPSC had established flam-
mability standards for children's sleepwear, and many manufacturers
met the requirement by using fabrics treated with Tris. When Tris was
later determined to be carcinogenic, a ban was imposed on further
sales and repurchase by the clothing manufacturers was required.
In most cases where no Court of Claims or Federal Tort
Claims action is available, it is necessary to petition Congress for a
private bill. However, in the cranberry case of 1959, the USDA tapped
an existing fund without having to go to Congress.^0 More than $8
million was paid to cranberry growers adversely affected by government
publicity concerning cranberries grown In the states of Washington and
Oregon which were contaminated with a dangerous herbicide. The indem-
nity payments compensated the growers of untainted cranberries in
other states who were unable to sell their crops due to the scare over
pesticide contamination.
Finally, the EPA pesticide indemnity program under § IS of
FIFRA is based partially upon equitable concepts in that compensation
is to be paid to pesticide owners who suddenly are forced to dispose
of stocks of pesticides which the government had previously
registered.31
c. Disincentives to Inappropriate or Unreasonable Government
Action
A prospective indemnification program can be intended as a
disincentive to undesirable government action. Indemnification can
force the government agency to internalize the expense of losses suf-
fered as a result of agency action, and thereby stimulate a more
thorough consideration of whether or not such losses can be avoided.
Often an accompanying rationale for indemnification under these cir-
cumstances is a desire to compensate the injured party for equitable
reasons.
Indemnification under § 15 of FIFRA was intended by some as
a disincentive to hasty or ill-considered pesticide registration. It
has also been suggested that § 15 operates as a disincentive to EPA's
cancellation of pesticides without the use of the "Special Rule" under
§ 15 which allows existing stocks of pesticides to be used or sold
even after an emergency cancellation of the registration.
The Equal Access to Justice Act, passed in 1980, is a recent
example of a statute designed to prevent or punish unreasonable gov-
ernment action.32 it accomplishes this by awarding court costs and
attorneys' fees to successful small business litigants (in court and
informal agency adjudications), unless the agency can prove that its
action was "substantially justified."
-18-
-------
INDEMNIFICATION STUDY
Section II
A similar disincentive provision exists under § 113 of the
Clean Air Act which authoriEes a court to award attorneys' fees and
costs if EPA brings an unreasonable Clean Air enforcement action.33
d. Indemnification as an Incentive for Action by Parties Out-
side the Federal Government to Facilitate Achievement of a
Public Goal
Prospective indemnity programs may be established as incen-
tives to prompt action by parties other than Federal agencies. An
indemnity program can also facilitate achievement of public goals by
removing disincentives to outside party action.
There are numerous agricultural indemnity programs that
operate as incentives.-^ Pesticide problems are the basis for most of
the agricultural indemnity programs. Two of them, in addition to
FIFRA § 15, address the problem of a farmer caught between the con-
flicting goals of using pesticides to increase productivity and pro-
tecting the public food supply from contamination by dangerous
residues. When food is barred from the market because of unacceptable
pesticide residues, notwithstanding the farmer's lawful use of such
pesticides, part of the loss has been shifted to the government.
The beekeepers program compensated for damage done to bees
by pesticides developed as substitutes for DDT.35 xhe program was
officially terminated after Congress apparently decided last summer
that twenty years was sufficient time to adjust to the cancellation of
DDT.
The dairy indemnity program compensates dairy farmers pre-
vented from marketing milk with unacceptable pesticide residues.36 At
one point dairy farmers received $350,000 a year as compensation for
such losses. Apparently, current pesticides are not as persistent and
do not prevent milk from being marketed as often. Only $40,000-
$50,000 a year has been paid in recent years. However, a major con-
tamination incident in Hawaii this year may overwhelm the program with
millions of dollars in claims.
The cranberry case mentioned above is a fourth instance of
pesticide-related indemnities.
The animal disease program is the oldest and most costly
agricultural indemnity program.3' it accompanies and supports a
cluster of USDA programs aimed at the eradication of particular
livestock diseases. The money paid for destruction of diseased and
exposed animals serves as an incentive for farmers to cooperate with
the program. Between 1970 and 1981, more than $180 million of indem-
nities were paid under the diseased livestock program.
Another health-related, but non-agricultural, indemnity
program is associated with the swine flu vaccine.38 AS an incentive
-19-
-------
INDEMNIFICATION STUDY
Section II
to get vaccine manufacturers to produce swine flu vaccine, the govern-
ment accepted some of the liability risk of the immunization program.
In a complicated arrangement, the government agreed to accept the vac-
cine manufacturers' potential strict product liability for personal
injuries due to the vaccine. Under this scheme, an injured victim is
permitted to sue only the federal government under state tort law.
Later, if negligence is proved, then the government can in turn
recover from the parties responsible for the negligence. In effect,
the federal government has accepted the product liability exposure of
the vaccine manufacturers for liability in excess of that for negli-
gence. Thus far, more than $40 million of awards or settlements have
been made by the government under the program, out of more than 4,000
claims seeking $2.95 billion.
The Arts and Artifacts Indemnity Act enables the federal
government to indemnify American museums for losses incurred in exhi-
biting international loaned art or artifacts. * Under the Act the
program my not obligate more than $400 million at any one time and
there is a limit of $50 million per exhibit. Although no indemnity
payments have yet been made, the fact that museums do not have to
purchase private insurance for the amount covered by the federal
Indemnification program has saved an estimated $800 million in
insurance premiums as of the summer of 1981.
Indemnification clauses In government defense contracts^
are similar in many ways to the Arts and Artifacts Program. Although
the Anti-Deficiency Act usually prevents an Executive Branch official
from promising to pay a contingent and uncertain liability, Congress
and the President have made exceptions for contracts that "facilitate
the national defense" and are "unusually hazardous or nuclear in
nature." Although such contractual indemnification clauses have fre-
quently been used, no claims have actually been paid by the government
to date.
Indemnification for government contractors is a matter of
current congressional concern. Objections to classifying certain
activities as "unusually hazardous," resulted in 1981 legislation per-
mitting an Indemnification clause in NASA contracts for the benefit of
space shuttle users. It has also led to an interagency task force,
co-sponored by the Office of Federal Procurement Policy and NASA,
which recently completed a draft report recommending expansion of
indemnification authority to include catastrophic losses beyond those
covered by reasonably available insurance but not "unusually hazard-
ous." There is also a proposed bill to indemnify all suppliers of pro-
ducts to the government.
Indemnification has also been used as an incentive to
encourage prompt private-party clean up of oil spills or hazardous
waste spills.^ Section 311(1) of the Clean Water Act encourages an
Innocent discharger to assist in clean up by promising to reimburse
-20-
-------
INDEMNIFICATION STUDY
Section II
him for the reasonable costs involved. The Outer Continental Shelf
Lands Act Amendments of 1978, the Deepwater Port Act of 1974 and the
Trans-Alaska Pipeline Authorization Act all contain provisions per-
taining to oil spills that are similar in basic design to § 311. The
Comprehensive Environmental Response, Compensation and Liability Act
(CERCLA) contains provisions relating to hazardous substance spills.
While CERCLA contains no provision identical to § 311(1), a similar
result is accomplished under § lll(a) which provides that the
President may pay claims for necessary response costs incurred by any
person as a result of carrying out the national contingency plan, so
long as such costs are approved under the plan and are certified by
the responsible federal official.
Another incentive program is that under § 202(a)(3) of the
Clean Water Act which authorizes EPA to provide 100 percent funding of
modification or replacement costs of failed municipal wastewater
treatment facilities using innovative or alternative technology.^
This provision acts as an insurance policy designed to encourage use
of new and risky technology.
-21-
-------
INDEMNIFICATION STUDY
Section II
FOOTNOTES TO SECTION II
the Federal Environmental Pesticides Control Act of 1972,
P.L. 92-516, 86 Stat. 973, contained, in Section 15, a provision
amending FIFRA, the Federal Insecticide, Fungicide, and Rodenticide
Act. P.L. 80-104, 61 Stat. 163.
§ 15 indemnities provision was part of H.R. 10729. See
H.R. Rep. No. 92-511, 92d Cong., 1st Sess. (1971) (accompanying
H.R. 10729). Two Senate Committees amended H.R. 10729 to strike out
the pesticide Indemnity provision, and the Senate as a whole approved
this change. S. Rep. No. 92-838, 92d Cong., 2d Sess. (1972). Specif-
ically, the Senate Committee on Agriculture and Forestry adopted the
amendment, and the Senate Commerce Committee on June 23, 1972 reported
the legislation out without any changes to the Agriculture and
Forestry Committee's decision against indemnities. The Senate on
September 26, 1972, passed H.R. 10729 in this form. However, the
pesticide indemnification provision was restored in Conference and
became law in 1972. Committee of Conference, S. Rep. No. 92-1540, 92d
Cong., 2d Sess. (1972).
Senate toxic substances control bill originally made no men-
tion of indemnification. However, this bill (S. 426) was amended by
the Senate to provide that no indemnity payments should be made under
TSCA or under § 15 of FIFRA. 119 Cong. Rec. 24494-95 (1973). See
also Toxic Substances Control Act of 1973; Hearings on S. 426 and
Amendments 1, 8, and 9 Before the Subcomm. on Environment of the
Senate Comm. on Commerce, 93d Cong., 1st Sess. (1973) and Toxic
Substance Control Act of 1973; S. Rep. No. 93-254, 93d Cong., 1st
Sess. (1973).
^S. 426 as reported out of the Senate Commerce Committee contained
the described provisions of Senator Phillip Hart's Amendment No. 9 on
indemnification. The bill passed the Senate on July 18, 1973 with
this provision intact.
House toxic substances control bill in the 93rd Congress made
no mention of indemnification, and no bill emerged from Conference in
the 93rd Congress. See Toxic Substances Control Act of 1973; Report
of the House Comm. on Interstate & Foreign Commerce, H.R. Rep. No.
93-360, 93d Cong., 1st Sess. (1973). Neither H.R. 5087 nor H.R. 5350
contained any indemnification provisions.
The idea of a study of indemnification reemerged in the first ses-
sion of the 94th Congress in 1975. The Senate bill (S. 776) again
called for an independent study of indemnification but, unlike the
amended version of S. 426 in the 93rd Congress, did not attempt to bar
indemnification under TSCA or under § 15 of FIFRA. A House bill (H.R.
7229) contained a provision requiring a study by the General Account-
ing Office (GAO) of Indemnification under all laws administered by
-22-
-------
INDEMNIFICATION STUDY
Section II
EPA. See Toxic Substances Control Act: Hearings on H.R. 7229,
H.R. 75A8, and H.R. 7664 Before the Subcomm. on Consumer Protection &
Finance of the Corem. on Interstate & Foreign Commerce, 94th Cong., 1st
Sees. (Serial No. 94-41) (1975). H.R. 7229, H.R. 7548, and H.R. 10318
contained indemnification provisions, but H.R. 7664 did net. There
was no debate on the floor of either body in 1975 on the issue of
indemnification, and again nothing emerged from Conference.
In 1976, during the second session of the 94th Congress, the House
bill (H.R. 14032) contained the § 25(a) provision that was finally
adopted into law. Toxic Substances Control Act; Report by the House
Comm. on Interstate & Foreign Commerce [H.R. 14032], 94th Cong., 2d
Sess. (1976). The Senate bill (S. 3149) contained a similar section,
with the difference that the GAO, Instead of the EPA, was to conduct
the study. S. 3149 was cleaned-up version of S. 776. Toxic
Substances Control Act: Report of the Senate Comm. on Commerce [H.R.
14032], 94th Cong., 2d Sess. (1976). In Conference, the Senate bill
was largely adopted, although the House indemnification study provi-
sion was used instead of the language from the Senate bill. Toxic
Substances Control Act; H.R. Rep. No. 94-1679, 94th Cong., 2d Sess.
100-101 (1976). Again, there was no debate in either the Senate or
the House focusing on § 25(a). The bill passed both bodies on
September 28, 1976 and was signed into law by President Ford on
October 11, 1976. P.L. 94-469, 15 U.S.C. §§ 2601 et seq.
^Senator Hart's Amendment No. 9 to S. 426, supra note 4, included
this definition of indemnity. A review of case decisions and secon-
dary sources defining "indemnity" and "indemnification" reveals that
the terms "indemnity," "compensation," "contribution," and "insurance"
are closely related, and frequently one term is used to define
another. Nevertheless, each term has a distinct legal meaning that
sets it apart from the other terms.
"Compensation" is the broadest of the four terms. ISA C.J.S.
"Compensation" at 102 (1967); Black's Law Dictionary 256 (5th ed.
1979). It refers to a transaction between two or more persons in
which one person restores another person to his former position,
usually through the payment of money. In the case of liability for
physical injuries, compensation involves the payment of money to make
the injured person whole, i.e., through reimbursement for medical
expenses, lost earnings, etc. Compensation may also relate to payment
for services rendered, or for property taken or destroyed.
The right of "indemnity" may arise either in equity (to avoid
unjust enrichment) or in contract, where one party obligates himself
to insure a loss or to discharge a duty or obligation. In the context
of an equitable remedy, "indemnity" is a right which inures to a per-
son who has discharged a duty owed by him, bujt which should have been
discharged by another person. Restatement of Restitution § 76; 41
Am. Jur. 2d "Indemnity" § 1, at 687 (1968); American Mutual Liability
-23-
-------
INDEMNIFICATION STUDY
Section II
Insurance Co. v. Reed Cleansers, 265 Minn. 503, 122 N.W.2d 178, 182
(1963). To illustrate, an employer orders an employee to repossess
the car of a third person. In fact the employer had no right to
repossess the car, and thus the repossession amounted to a conversion
of the car. The employee is liable to the third party for his tor-
tious act in taking the car. However, since the employee acted within
the scope of his employment, he would be entitled to indemnification
from his employer who is primarily liable to the third party for the
conversion of the car.
In the context of "indemnity" as a contractual remedy, the insur-
ance situation is a common example. 42 C.J.S. "Indemnity" § 1 at 564,
and § 3 at 566 (1944); Ballantine's Law Dictionary at 608; Black's
Law Dictionary at 692. By contract, the Insurer agrees to indemnify
the insured against loss or liability. While the insured is liable to
the third party for the loss which the third party sustained, the
insurance carrier by contract has obligated itself either to pay the
third party for the loss, or to reimburse the Insured for the amount
which he paid to compensate the third party. In this instance, the
contract Imposes the responsibility upon the insurance carrier and
creates the right of Indemnity for the insured.
"Contribution" is the right of one who pays more than his propor-
tionate share of an obligation to be compensated by the others who
were also liable. Restatement of Restitution § 81 and Comment b;
Black's Law Dictionary 297 (5th ed. 1979); Shannon v. Massachusetts
Bonding & Insurance Co., 62 F. Supp. 532, 537 (W.D. La. 1945).
Examples of such parties are joint tortfeasors or joint obligors on a
draft. Contribution differs from indemnity in that the liability Is
shared, rather than the whole burden being shifted from one party to
another.
"Insurance" is simply a contract by which the insurer undertakes
to indemnify the insured against loss arising from a specific
event. 44 C.J.S. "Insurance" § Ib at 473 (1945); Physicians' Defense
Co. v. Cooper. 199 F. 576, 578 (9th Clr. 1912); Black's Law DictionaFy
at 721. Ballantine's Law Dictionary at 642; Jordan v. Group Health
Ass'n, 107 F.2d 239, 245 (1939). The contract of insurance has three
basic elements. First, there must be consideration or value given to
the Insurer in exchange for its obligation to pay. In most instances
consideration will take the form of the premiums which the Insured
pays on his policy. Second, there must be a risk which is unknown or
contingent, and against which the insured is protected. The risk may
be in the form of harm to the insured himself or of liability to a
third party. Finally, there must be indemnity, i.e., a specified sum
to be paid by the insurer in the event loss Is suffered from the
occurrence of the event or contingency specified in the contract.
Thus, by contract responsibility for payment of the loss is shifted
from the insured to the insurance carrier.
-24-
-------
INDEMNIFICATION STUDY
Section II
^Federal Crop Insurance Corporation, 7 U.S.C. §§ 1501 et seq.;
National Flood Insurance Program, 42 U.S.C. §§ 4001 et seq. ; Federal
Crime Insurance, 12 U . S . C . §§ 1749 bbb-lOa et seq.; Urban Property
Protection and Reinsurance Act, 12 U.S.C. §§ 1749 bbb et seq.; Nuclear
Energy Liability Insurance, 42 U.S.C. §§ 2210 et seq.; Federal Deposit
Insurance, 12 U.S.C. §§ 1811 et seq. (FDIC), 12 U.S.C. §§ 1725 et seq.
(FSLIC), and 12 U.S.C. §§ 1781 et seq. (share insurance for credit
unions); war risks insurance, 46 U.S.C. §§ 1281 et seq. (marine war
risk insurance) and 49 U.S.C. §§ 1531 et seq. (air war risk insur-
ance); mortgage and housing insurance, 12 U.S.C. § 1703 (insurance of
institutions which finance housing); 12 U.S.C. §§ 1706c et seq. (mort-
gage insurance); 12 U.S.C. §§ 1736 et seq. (war housing insurance); 12
U.S.C. §§ 1747 et seq. (insurance of investments in rental housing);
42 U.S.C. §§ 1471 et seq. (federal housing); 46 U.S.C. §§ 1271 et seq.
(ship mortgage insurance); health and medical Insurance, 5 U.S.C.
§§ 8901 et seq.; 42 U.S.C. §§ 1395 et seq.; unemployment insurance, 45
U.S.C. §§ 351 et seq.; Export-Import Bank tangible property insurance,
12 U.S.C. § 635. See also Overseas Private Investment Corporations 's
insurance of corporate Investments abroad, 22 U.S.C. § 2191.
See Appendix B on Alternatives to Indemnification, page B-6.
9 Recently, Congress has been willing to assist industry in
the face of potentially enormous claims due to personal or property
injury by establishing liability limits and assisting with payment of
the claims. The Comprehensive Environmental Response, Compensation
and Liability Act of 1980, P.L. 96-510, 94 Stat. 2767, establishes
liability limits for damages and cleanup costs associated with inac-
tive chemical disposal sites and chemical spills and it establishes a
fund (financed primarily by industry) to facilitate, among other
matters, cleanup and remedial activity where fault or responsibility
cannot be determined. Recently proposed legislation in both the House
and the Senate involving occupational injuries related to asbestos
would utilize the same method: a limit on liability and an industry
financed fund to cover claims. See H.R. 5224, sponsored by Rep.
Millicent Fenwick (R-N.J.) and H.R. 5735, sponsored by Rep. George
Miller (R-Cal.).
In both the case of chemical discharges and asbestos injuries,
however, there is an apparent congressional attitude that those per-
sons responsible for damages and injuries should absorb the costs
involved. Government assistance is justified because it primarily
benefits the public or the class of injured persons, rather than
industry. The prevailing attitude appears to be that the government
may consider stepping in where traditional market incentives and reme-
dial systems (e.g. , insurance, liability law, judicial administration
and resolution of disputes) fail to address effectively a significant
and widespread problem. An additional motivating factor involving
asbestos injury claims is that the government has also been an
employer of asbestos exposed workers, particularly in public shipyards
during World War II.
-25-
-------
INDEMNIFICATION STUDY
Section II
A legislative assistant to Representative Millicent Fenwick who is
familar with the current debate over proposed asbestos legislation
thought that the primary purpose of such legislation is to assist the
worker exposed to asbestos. If this results in underwriting or
limiting industry costs to some extent, she thought it would be
acceptable to Congress if that were the only way to provide the needed
assistance. Both industry and the workers face huge legal costs and
uncertainty in the court system. The worker who is injured is often
compensated after he dies, loses a significant amount of the award in
legal fees, and may often be barred from recovery entirely by state
law which is not designed to handle claims resulting from injury over
a 20- to 30-year period. The evidence suggests that the majority of
exposure cases have yet to appear and the problem for Industry, the
courts and the worker will become significantly larger in the years to
come. G
See also § 301 (e) of CERCLA which mandated a study concerning com-
mon law and statutory compensation remedies for harm to man and the
environment caused by release of hazardous substances, and Six Case
Studies of Compensation for Toxic Substances Pollution; Alabama,
California, Michigan, Missouri, New Jersey, and Texas, Congressional
Research Service Report for the Senate Committee on Environment and
Public Works, 96th Cong., 2d Sess. (June 1980) (Serial No. 96-13).
^Appendix A, page A-2.
1 * Appendix A, page A-4.
^Appendix A, page A-6.
Eastport Steamship Corp. v. United States, 178 Ct. Cl.
599, 608-11 (1967) (damages denied in claim against Maritime Commis-
sion for an alleged abuse of its discretion to grant a license or
approval. The Court of Claims referred to the "persistent refusal by
the federal legislature to permit damage claims against the United
States resulting from wrongful regulatory activity").
1431 U.S.C. 665(a), 41 U.S.C. ll(a).
15Andrus v. Allard. 444 U.S. 51, 100 S. Ct. 318, 326 (1979) (5th
amendment taking case).
"id..
l?Some additional arguments against compensation are:
(7) Parties affected by government regulations should be
encouraged to foresee and prevent or mitigate the poten
tially negative effects of those regulations;
-26-
-------
INDEMNIFICATION STUDY
Section II
(8) An Executive Branch promise to indemnify In the future
at an unknown cost would interfere with the legislative
prerogative to authorize and appropriate money;
(9) Money damages are often not an appropriate or effective
remedy for harms due to regulatory activities;
(10) Private parties should not be paid for compliance with
health and safety regulations.
*"Appendix A, page A-8.
*^Appendix A, page A-8.
2^A prime example of the use of an indemnification program to deal
with recurring loss situations is the Federal Tort Claims Act. Once a
pattern or a sufficient number of loss incidents becomes clear,
Congress will usually consider establishing a regular mechanism to
handle the claims.
^Appendix A, page A-7.
22Appendlx A, page A-7.
23Appendix A, page A-2.
^Appendix A, page A-15.
25Appendix A, page A-12.
2^Appendix A, page A-15.
27Appendix A, page A-20.
28Appendix A, page A-21.
2^Appendix A, page A-22.
30Appendix A, page A-20.
31Report, infra page 30.
32Report, infra page 67.
33Report, infra page 63.
34Appendix A, page A-18.
35Appendix A, page A-19.
-27-
-------
36
'Appendix A, page A-19.
37Appendix A, page A-18.
•10
JOAppendix A, page A-23.
39Appendix A, page A-25.
*°Appendix A, page A-26.
^Report, infra page 75.
A2Report, infra page 50.
INDEMNIFICATION STUDY
Section II
-28-
-------
INDEMNIFICATION STUDY
Section III
III. EVALUATION OF EPA PROGRAMS
A. Criteria for Evaluating the Need for Indemnification
Within EPA Programs 30
I. General Nature of the Criteria 30
2. Specific Factors 30
B. Indemnification Under § 15 of FIFRA 33
1. The Existing Program. 33
2. Extension of FIFRA § 15 to New Situations 46
C. Indemnification Under § 202(a)(3) of the Clean Hater Act. . 53
1. The Existing Program 53
2. Possible Modifications of the Existing Program 58
3. Possible Applications of the § 202(a)(3) Model
in Other EPA Programs 60
D. Indemnification Under § 113(b) of the Clean Air Act .... 66
1. Legislative History and Background 66
2. Application to Date 71
3. Summary Assessment of § 113(b) 74
4. The Applicability of the § 113(b) Model to
Other EPA Statutes 76
E. Indemnification Pursuant to § 311(1) of the
Clean Water Act 78
1. Legislative History And Background 78
2. Application To Date 84
3. Summary Assessment of 311(1) 91
4. The Applicability of § 311(1) to Other EPA Programs . . 93
Footnotes 95
-29-
-------
INDEMNIFICATION STUDY
Section III
A. CRITERIA FOR EVALUATING THE NEED FOR
INDEMNIFICATION WITHIN EPA PROGRAMS
1. General Nature Of The Criteria
Consideration was given in this study to developing a general rule
or formula for deciding whether indemnification would be warranted in
a particular circumstance. After efforts at logical analysis, deduc-
tion of patterns from previous indemnification incidents, and use of
tentative formulas tested by application to actual cases and to
hypothetical situations, it was concluded that a formula approach
would be extremely complex and probably ineffective. Thus, while
general criteria and relevant factors for analysis were identified, no
clear-cut and automatic formula was developed. Examples of past
indemnities for the consequences of government regulations were too
few and varied to provide conclusive patterns of important criteria.
Precise indemnity determinants also cannot be derived theoretically
because real cases appear to depend on unpredictable combinations of
factors. All of the actual indemnity situations, and particularly
those involving equitable considerations seem to be dependent upon the
facts of an individual situation.
What follows, therefore, is not a mathematically precise for-
mula that can be used automatically to decide when indemnification
should be afforded a party who has suffered a loss. It is instead a
list and brief description of criteria that we found to be most useful
in evaluating the situations that we studied. Our conclusions about
the way these factors have worked in specific instances may be found
with the other conclusions in Section IV, infra. In evaluating these
specific situations the factors either supporting or weighing against
indemnification were assessed. Then, the factors supporting indem-
nification were balanced against those weighing against indemnifica-
tion. Generally, we found that many of the factors involved an
enormous range of possibilities and are often interrelated with other
factors.
2. Specific Factors
a. The Justification for Indemnification
The balancing of arguments for and against indemnification may
be expected to vary depending on the overall objective intended. The
strength and nature of the justification for indemnification is a key
element in deciding whether to grant it. Although many indemnity
programs appear to have multiple goals, there are three possible cate-
gories of reasons for indemnification:
• Equity: Equitable considerations may favor after-the-fact
indemnification when the government is "responsible" for
the loss or has created an expectation that the government
r30-
-------
INDEMNIFICATION STUDY
Section III
should pay for the loss. Important considerations
include: whether the loss was within an expected scope of
business risk, whether the risk was insurable, whether the
event causing loss was caused in part by the injured
party, and if the loss resulted from reliance on the
government, whether such reliance was reasonable. Another
equity aspect is whether the party had procedural due
process. This usually concerns opportunities to comment
on or challenge the agency decision before the loss
occurs.
• Disincentive; Indemnification may be intended to discour-
age certain government action, tinder this category,
equity considerations involving the nature or existence of
a loss would be less0important, since the major purpose
would be to prevent agency action causing losses rather
than subsequently remedying any inequities. Key con-
siderations are whether such a disincentive is really
needed and whether a practically effective disincentive
can be structured which would not involve undesirable side
effects.
• Subsidy or incentives; Indemnities may be used as subsidy
payments for certain parties or financial incentives to
encourage or permit a government objective or policy.
Under this category, too, equity considerations would be
less important, since the major purpose would be to
encourage action by parties outside the federal govern-
ment. Again, an important issue is whether an incentive
is really necessary and workable.
b. Need for an Anticipatory Indemnification Program
A useful distinction in the indemnity analysis is whether the
problem can be dealt with on a case-by-case basis or whether it
requires an indemnification program. The factors relevant to this
threshold question are relatively easy to apply in that three possible
considerations may support the establishment of a program. First,
there may be a need for a strong incentive to influence behavior.
Second, a program can usually provide much faster processing of claims
and payment than an ad hoc approach. Third, an institutionalized
response may be required in order efficiently to handle frequent,
similar claims.
c. Availability of Alternatives to Indemnification
This appears to be a very important factor in deciding whether
or not to provide indemnification to deal with an alleged problem.
Consideration must be given to whether compensation is already
available from an existing mechanism such as a lawsuit under the
-31-
-------
INDEMNIFICATION STUDY
Section III
Federal Tort Claims Act. Also, there may be relief mechanisms such as
exemptions and waivers which could be used to avoid loss and these may
be viewed as a preferable alternative to indemnification. (Alterna-
tives to indemnification are discussed in detail in Appendix B, infra.)
d. Impact on Other Policy Objectives
Another important factor is whether indemnification would sup-
port or conflict with other program objectives or legislation.
Indemnification cannot be viewed in isolation. Its usefulness in a
particular situation must be evaluated in light of potential negative
side effects or related benefits.
e. Legislative Predetermination
This factor concerns whether Congress has already considered
the issue of compensation or other remedies in the situation at hand
and expressed a position that should be viewed as determinative.
Alternatively, there may be relevant but inconclusive indications of
congressional attitude that should be considered.
f. Cost of Indemnities
The dollar costs of indemnification must be compared to the
overall benefits of providing indemnities. An indemnity payout may be
unacceptably expensive to administer. Indemnification cost estimates
may also be too unpredictable or speculative. However, there are a
variety of ways in which the cost to the government can be contained,
deferred, or shared, though they may result In less than full
indemnity. Cost estimates may be more concrete if based upon analysis
of the nature and frequency of losses that have already occurred,
rather than upon abstract predictions. Consideration should also be
given to whether an Indemnity payment would establish a precedent for
indemnification that could not readily be distinquished.
g. Ability to Finance and Administer
The feasibility and problems of administering and financing
indemnification In particular circumstances should be assessed.
Similar to the issue of dollar costs, administrative and financing
difficulties must be compared to the potential benefits of an indem-
nity mechanism.
Indemnification payments need to be financed and administered
in a manner that does not interfere unacceptably with the administra-
tion of other programs. The choice among the many ways that a program
can be financed must be made carefully so that it furthers the goals
of the indemnity programs and the underlying mission of the agency.
Failure to choose a financing mechanism at the time the indemnity
program is established may seriously undermine it, particularly if it
is Intended as an incentive.
-32-
-------
INDEMNIFICATION STUDY
Section III
B. INDEMNIFICATION UNDER § 15 OF FIFRA
1. The Existing Program1
a. Introduction
Section 15 of FIFRA provides for the indemnification of owners
of a pesticide who suffer economic loss from the suspension and subse-
quent cancellation of the pesticide's registration. Section 15 was
first enacted as part of the Federal Environmental Pesticide Control
Act of 1972 (FEPCA),2 which amended earlier versions of FIFRA.3
Indemnification has been retained without change in the 1975, 1978,
and 1980 versions of the Act.^ The legislative history of the § 25(a)
indemnification-study mandate shows that Congress was interested in
the type of indemnification represented by § 15 of FIFRA.
Under § 15, indemnity for the cost of the pesticide must be
paid by EPA to any pesticide owner in those rare circumstances where
the owner can show that:
(a) The pesticide registration was suspended on the ground
that it posed an imminent hazard;
(b) It was subsequently cancelled;
(c) The suspension or cancellation of a registration caused a
loss to the owner of the pesticide; and
(d) If the claimant is the pesticide producer, that it
notified EPA or discontinued production once It learned of
facts giving rise to the need for the cancellation.
Section 15 also Includes a "Special Rule" under which EPA may avoid
causing economic losses by providing a reasonable time for use or
disposal of a suspended and cancelled pesticide.^
No prior government program had ever provided indemnification
to manufacturers of potentially hazardous products (automobiles, toys,
or pharmaceutlcals, for example).^ However, government indemnities to
farmers whose products have been destroyed or determined to be danger-
ous to health are common. These have included payments to cranberry
growers,^ beekeepers,^ farmers with pesticide-contaminated milk,' and
farmers with diseased livestock.10 Section 15 substantially expanded
the traditional agricultural indemnification subsidies by authorizing
indemnification of manufacturers and sellers of pesticides, In addi-
tion to farmers and other and users of pesticides.
-33-
-------
INDEMNIFICATION STUDY
Section III
b. Statutory Language
Section 15 of FIFRA, 7 U.S.C. § 136, provides:
SEC. 15. INDEMNITIES.
(a) REQUIREMENT.—If—
(1) the Administrator notifies a regis-
trant that he has suspended the registration of
a pesticide because such action is necessary to
prevent an imminent hazard;^
(2) the registration of the pesticide is
canceled as a result of a final determination
that the use of such pesticide will create an
imminent hazard; and
(3) any person who owned any quantity of
such pesticide Immediately before the notice to
the registrant under paragraph (1) suffered
losses by reason of suspension or cancellation
of the registration,
the Administrator shall make an indemnity payment
to such person, unless the Administrator finds that
such person (1) had knowledge of facts which, in
themselves, would have shown that such pesticide
did not meet the requirements of section 3(c)(5)
for registration, and (ii) continued thereafter to
produce such pesticide without giving timely notice
of such facts to the Administrator.**
(b) AMOUNT OF PAYMENT.—
(1) IN GENERAL.—The amount of the Indemnity
payment under subsection (a) to any person
shall be determined on the basis of the cost of
the pesticide owned by such person Immediately
before the notice to the registrant referred to
in subsection (a)(l); except that in no event
shall an indemnity payment to any person exceed
the fair market value of the pesticide owned by
such person immediately before the notice
referred to in subsection (a)(l).
(2) SPECIAL RULE.—Notwithstanding any other
provision of this Act, the Administrator may
provide a reasonable time for use or other
disposal of such pesticide. In determining the
quantity of any pesticide for which indemnity
shall be paid under this subsection, proper
adjustment shall be made for any pesticide used
or otherwise disposed of by such owner.
-34-
-------
INDEMNIFICATION STUDY
Section III
c. Legislative History of § 15
Indemnification was not included in the predecessor pesticide
laws of 1910, 1947, or 1964.13 -j^e indemnity provision was developed
during 19 closed business meetings held by the House Agriculture
Committee to work on H.R. 4152 in February 1971.^ xhe indemnifica-
tion section was later deleted from the Senate version of FEPCA passed
in September 1972, but a slightly revised version of the House
Committee Print section was included in the conference bill that
passed both houses in October 1972.
Section 15 was a focal point of the debate over FEPCA among
members of the agricultural community, pesticide producers and distri-
butors, and representatives of environmental and consumer organiza-
tions.*^ As enacted, the indemnification provision largely reflected
the position of the House Agriculture Committee. However, the Confer-
ence Committee did add a provision denying indemnities to manufactur-
ers, acting in bad faith, who withheld information showing that their
pesticides should not be registered.1°
d. Implementation of § 15
(1) Overview
Because EPA has never developed regulations or guidance
documents to implement § 15, ad hoc claim procedures were formulated
for the Silvex case. To a great extent they followed the agreement
negotiated primarily between EPA and Chevron Chemical Company.^
Before the Silvex case, the only circumstance meeting the
requirements for indemnification under § 15 was a minor one involving
vinyl chloride, a propellant in aerosol pesticide cans.*° EPA paid
three indemnity claims, totaling approximately $53,000, from EPA funds
originally allocated to other purposes.
Two other emergency suspensions (Aldrin/Dieldrin and
Heptachlor/Chlordane) occurred during the period from 1974-1976.^ In
both cases, existing stocks of pesticides were permitted to be used or
sold, thereby avoiding any indemnifiable economic loss by the owners
of the pesticides.
One claim has been paid to a fanner after the suspension
and cancellation of DBCP. It appears that the total value of DBCP
claims will not be large.*0 in addition, depending upon the outcome
of the 2,4,5-T/Silvex pesticide cancellation proceedings, indemnifica-
tion claims may also be made for 2.4,5-T and for Silvex products that
were not cancelled by agreements.
During the legislative hearings, an estimate of the value
of nationwide inventory of one chemical was made at $30 million.^2
-35-
-------
INDEMNIFICATION STUDY
Section III
This is within the range of the expected total indemnities for Silvex.
The amount is approximately equal to one half of the total annual
budget for EPA's Office of Pesticide Programs.
(2) Funding
No financing mechanism was provided for § 15. The statute
states that, when certain requirements are met, "the Administrator
shall make an indemnity payment," thus implying that funds are to come
from EPA's budget.23 However, no corresponding budget authorization
or appropriation has ever been provided by Congress.
Four potential methods of funding EPA indemnification are:
• Reprogramming of Agency funds to meet indemnities as
they occur;
• Submitting a supplemental budget request, following
use of Agency funds previously appropriated for other
purposes;
• Contingency budgeting for estimated indemnities with
funds held in reserve until needed; or
• Requesting case-by-case appropriations by Congress in
EPA's annual budget, with payment deferred until
funds are actually appropriated.
However, none of these options is free from significant practical or
policy problems and, by default, a fifth option of payments from the
Court of Claims Judgment Fund has occurred.
Because it had no funds available, and no process for
obtaining them, EPA refused the Silvex claims, while acknowledging
their validity. The claimants therefore sued the government in the
Court of Claims. While the settlement of several of these claims has
provided for payment from the Court of Claims Judgment Fund, some
question remains whether EPA will be required to reimburse the Fund.
(3) Silvex
On February 28, 1979, the registrations of many products
containing 2,4,5-T and Silvex were suspended by emergency order (and
notice of Intent to cancel) because of imminent hazards deemed so
serious that suspension could not await a hearing.^ This emergency
suspension was effective immediately. The Silvex suspension covered
products registered for forestry, rights-of-way, pasture, home and
garden, commercial/ornamental and turf, and aquatic weed control uses.
(Not all of the uses of Silvex were suspended. It may still be used
for rice, range, sugar cane, orchards, and non-crop sites.) Silvex
-36-
-------
INDEMNIFICATION STUDY
Section III
was considered an imminent threat because new information on human
reproductive and oncongenic risks was made public just before the
spring growing season, and EPA concluded that it had to act quickly if
it was to prevent the pesticide from being sold or used that season.
Led by Chevron Chemical Company, one of the largest manu-
facturers of Silvex products, 19 registrants subsequently entered into
agreements with EPA which, in effect, cancelled the registrations of
the horae-and-garden use Silvex pesticides. The first of these, agree-
ments was signed in May 1979. The agreements stipulated that it was
infeasible to relabel, recover, recycle, reprocess, or otherwise use
the Silvex pesticides for any of the non-suspended uses. This enabled
claimants to show economic loss required by § 15(a)(3). It precluded
the government from arguing that since all uses were not suspended, a
market remained, and the loss would only be the cost of relabeling for
resale.
Other 2,4,5-T and Silvex registrants, however, opposed
cancellation and requested a hearing. Those proceedings are currently
recessed to allow for settlement talks.
The agreements, that were reached stipulated that the stat-
utory criteria for indemnification under § 15 had been met, estab-
lished procedures for processing claims and formulas for calculating
the value of the pesticides, and confirmed EPA's statutory obligation
to provide for disposal of the pesticides. Under the agreements, all
claims of the registrants, distributors and retailers owning cancelled
products were funneled through each registrant and presented to EPA in
a package. Distributors and retailers were asked to fill out claim
affidavits and to give a power of attorney to the registrants to make
their claims to EPA.
EPA relied heavily on the registrants in processing
claims, particularly Chevron, which had approximately 11,000 distribu-
tors and retailers of lawn and garden pesticides. Chevron mailed a
notice to every dealer carrying Chevron agriculture products, telling
them to return their stocks of Silvex home and garden products to the
distributor. The distributors gave cash payments or credits to the
dealers. The distributors then delivered all they had collected to
Chevron warehouses across the country. Upon delivery, distributors
were given credits on their accounts with Chevron.
Apparently, the registrants had sought the agreements with
EPA to speed processing of their indemnification claims, and in the
belief that indemnification would ultimately be paid, even though no
funds for payment were appropriated. They also apparently wanted to
be able to empty their warehouses of the pesticides and to get EPA to
dispose of them quickly. They agreed to collect claims and assist in
processing them, apparently out of concern with protecting their busi-
ness relations with distributors and retailers.
-37-
-------
INDEMNIFICATION STUDY
Section III
Pursuant to the agreements and to § 19 of FIFRA, EPA
became responsible for disposing of a very large percentage of the
horae-and-garden Silvex on the market. EPA has already disposed of the
39 million pounds of solid pesticide in a hazardous waste landfill in
Alabama. Approximately 900,000 gallons of liquid pesticide should be
destroyed soon.25
EPA finished processing the first Indemnification claims
by early 1980 and has now reviewed the claims presented by 15 regis-
trants. Even with the help of the major registrants, it has been a
lengthy and difficult job. Using the cost formula in the agreements,
EPA has found approximately $18.2 million in indemnification claims
for Silvex products to be "appropriate." The .largest portion of this
amount is the $12.8 million of appropriate claims in the package sub-
mitted by Chevron. o
After final review of each claims package, EPA sent the
registrant a letter Indicating the amount that the Agency considered
appropriate. However, the letters also stated that the Agency lacked
funds to pay the claim and did not expect to receive such funds.
Several registrants subsequently filed suit in the Court
of Claims, seeking payment from the Court of Claims Judgment Fund.
The Department of Justice, representing the government, has recently
settled seven of the larger claims, for a total of $18.7 million (the
remaining claims total less than $200,000). Except for the Ag-Way
claim, which was settled for approximately $500,000 when EPA had
approved only about $375,000, the Justice Department's settlements
have been for amounts roughly equal to those found appropriate by EPA.
Section 15 does not define the "cost" for which owners are
to be indemnified. The Justice Department questioned the pesticide
cost-calculation formula contained in the negotiated agreements.
These formula specified that the cost to a retailer or distributor was
100% of the invoice price plus 15% for handling, and the cost to the
manufacturer was 70% of the cost to the dealer. The position of the
Justice Department was that actual cost data was required for a calcu-
lation of Indemnification amounts and that the formula (particularly
the 15% allowance for handling) was not consistent with the language
of § 15. However, an FBI Investigation of actual cost data revealed
that in nearly every circumstance the formula in the agreements was
too low and the registrants' claim totals, which sometimes exceeded
EPA's approved figure, were also too low. Therefore, in each of the
seven settled claims, the Justice Department agreed to the amount
claimed.
e. Evaluation of § 15
There has not been sufficient activity under FIFRA § 15 to
evaluate its effectiveness conclusively. However, a start can be made
-38-
-------
INDEMNIFICATION STUDY
Section III
by comparing the actual operation of the program to the assumptions,
goals, and arguments made before enactment. In the ten years since the
debate, some of the arguments have proven insubstantial, while others
have taken on new weight.
(1) Arguments in Support of Indemnification
(a) Pesticide manufacturers, dealers, and owners of
pesticides rely in good faith on the government's
pesticide registration
Arguments were made during the legislative debate on
§ 15 that indemnification would be necessary to protect manufacturers,
dealers, and farmers who rely on the validity of a pesticide's regis-
tration, if EPA changed its mind and canceled a registration. ^ This
argument rests on the premise that an Innocent party has relied, in
good faith and to its detriment on the registration.
The pesticide manufacturers apparently hoped to be
able to rely on the registration and tried to insert into the 1972
FIFRA amendments a provision that would use the registration to shield
them from third-party liability. They suceeded only in obtaining the
very conditional language of § 3(f)(2) that says "provided that as
long as no cancellation proceedings are in effect, registration of a
pesticide shall be prima facie evidence that the pesticide, its
labeling and packaging comply with registration provisions of the
Act."
In reality, there is little factual basis for relying
on the registration of the average pesticide as conclusive proof of
its safety. Most current pesticides were either "grandfathered" in
from previous pesticide laws or involve state, rather than federal,
registration.2' Many registrations are granted by the state on the
basis of limited test data, with nominal review by EPA. Some experi-
mental use permits are not registrations.
Moreover, most pesticide manufacturers do not in
fact appear to rely upon the continuation of registration. The
registration section (§ 3(f)) disclaims reliance. The statutory
requirements applied to most existing pesticides have been relatively
lax.28 Moreover, in making its registration determination, EPA relies
almost exclusively upon data submitted by the registration applicant.
Thus, the pesticide manufacturer is typically relying upon its own
data, rather than that of the government, when making financial deci-
sions concerning pesticide production and inventory.
Nor is there a basis for relying on the indefinite
continuation of the registration. The structure of the Act makes
clear the intention of Congress that a pesticide may be cancelled
whenever new information shows that it may be an unreasonable risk.2*
-39-
-------
INDEMNIFICATION STUDY
Section III
Farmers may more reasonably be expected to rely on
EPA's registration of a pesticide. However, farmers have a problem
after a deregistration that Is bigger than the loss of Inventory —
finding a comparably priced substitute that is equally efficacious.
Sometimes this is not immediately possible, and the fanner has to pay
more for a substitute that results in a lower yield, at least for a
couple of years. Although these costs are not addressed by the
current indemnification program, they are taken into account when the
Agency makes cancellation decisions. FIFRA § 6(b) requires the Admin-
istrator to consider restricting uses as an alternative to cancella-
tion and to take into account the "... impact of such final action
on production and prices of agricultural commodities, retail food
prices, and otherwise on the agricultural economy. . . ."
Indemnifying for°all farmers' losses in the event of
an emergency suspension and cancellation may be very expensive.
Furthermore, although the loss is not entirely speculative, it would
be very difficult in practice to measure and verify it. One could not
predict transition costs, production drops and the time needed for
pesticide manufacturers to develop and market a lower priced substi-
tute. Thus, it might be possible to make payments only some years
after the original pesticide cancellation when the costs were known.
Even then, it would always be difficult to distinguish the production
yield decrease due to other factors, such as weather, from the
decrease due solely to the different pesticide.
Under these circumstances it might be more helpful to
the farmers to prevent or limit the loss by providing for a more
orderly transition period and stimulating the development of alter-
native pesticides. By giving the industry some advance warning of the
cancellation, allowing time to Increase production of existing substi-
tutes and accelerate the development of others, the losses are
minimized. This is the procedure for ordinary, rather than emergency,
cancellations — the way most pesticides are de-registered.
There was no problem of agricultural transition after
the Silvex cancellations because only home and garden uses were can-
celled. Because some other uses were not suspended, and the emergency
suspension of the remainder was challenged by registrants, cancella-
tion has not gone into effect. The anomaly of the first major indem-
nification case excluding farmers makes it difficult to base any
conclusions concerning the operation of § 15 on it*
(b) Banning of a Pesticide to Benefit the Public
Constitutes a Taking of Property for Which the
Government Should Pay Compensation
Some asserted that the sudden banning of a pesticide,
when accompanied by the requirement that the pesticide be turned over
to EPA for disposal, constitutes a taking of property for which the
-40-
-------
INDEMNIFICATION STUDY
Section III
government should provide compensation.^2 A converse formulation of
this argument Is that the dereglstratlon of an unsafe pesticide con-
fers a benefit on the public that should be paid for by the public.
Important to this argument is the fact that § 15 *s compensation of all
owners of pesticides makes this remedy very similar in form to compen-
sation for a taking of property.
The problem with the taking argument is that the
seizure of contraband or items that are hazardous to public health
does not legally qualify for compensation under the Fifth Amend-
ment.-^ Nor is the government legally estopped by its registration of
the pesticide from asserting that the pesticide is unsafe.-" By
registering the pesticide, the government has clearly not stated that
it believes the pesticide to be safe forever or that no evidence of
potential danger will be found in the future. Rather, the pesticide
registration indicates only that, at the time of the registration, the
government did not have sufficient evidence to refuse registration.
(c) Indemnification is needed to promote new pesticide
research and production
Arguments were strenuously made during the legisla-
tive debate on § 15 that, without financial protection from the possi-
bility of pesticide bans, chemical companies would cease to undertake
research and development of new pesticides.36 A less extreme argument
was also made that substantial financial losses due to pesticide can-
cellations would impede research and development. In fact, however,
no evidence has been found either that the availablity of Indemnifica-
tion stimulates new pesticide research and development or that pesti-
cide companies would reduce their research and development efforts if
indemnification were unavailable. Rather, it appears that the manu-
facturers view the possibility of dereglstration of any particular
pesticide as remote. The possibility that deregistration will involve
a pesticide during the first five or ten years of its useful life,
before It "breaks even" is even lower. Thus, it does not seem to
affect research decisions.
The business of pesticides is so volatile that the
manufacturers must constantly innovate in order to stay in business.
Its high profit margins indicate the risks of the business, but also
provide an incentive to do research necessary to produce new products.
Furthermore, the indemnification under § 15 is only for the inventory
on hand when a deregistration occurs. It does not necessarily compen-
sate the owner for the cost of the putting the chemical on the market.
One analysis of the effect of paying for lost inven-
tory on a pesticide's net present value shows that such payments have
only a small impact compared to the effect of the lost sales.
Paying Indemnification for lost inventory in the event of cancellation
increases the net present value of a pesticide by less than one
-41-
-------
INDEMNIFICATION STUDY
Section III
percent.This amount will probably not be sufficient to influence
decision-making for most companies. ^ Thus, loss of inventory is a
crude measure which not only fails to reflect a registrant's actual
losses from a cancellation, but also may fail to provide the incen-
tives that Congress desired.
Other indemnification formula could be devised to
have a greater impact on research. One possibility would be to pay
manufacturers for their total investment cost, less net revenue, with
respect to a cancelled pesticide. This formula is based on the fact
that a relatively new chemical, not long on the market, would not yet
have earned back its cost of research and development. Promising to
pay for investment cost less net revenues would at least guarantee the
manufacturer that its research costs would not be lost. This presum-
ably would encourage manufacturers to continue to spend money to deve^
lop new pesticides.
There would be problems with this approach, however.
If the chemical has been on the market for some time, total sales will
be greater than Investment. A judgment would have to be made whether
the owner should be compensated for the inventory under those circum-
stances. If the exclusive concern of indemnification is with covering
research and development costs, then indemnities should not be paid
for such inventory losses. Furthermore, the cost of paying for devel-
opment costs is likely to be much higher than the cost of inventory.
Also, it is difficult to Justify paying for research and development
only for some cancellations and not others. Inventory indemnification
can be justified on the basis of the emergency suspension.
A much more direct impact is exerted by the expendi-
tures of the federal government to support pesticide research and
development. The Department of Agriculture and experiment stations
spent $330 million in 1978, approximately matching the amount of pri-
vate funds spent. This federally funded research even results in
industry-owned patents.^
(d) Indemnification encourages EPA to be more careful
with its pesticide registrations
A minor point raised during the § 15 debate was that
ificatlon for deregistration would promi
ment to be more careful with the original registration.
requiring indemnification for deregistration would prompt the govern-
The desired feedback mechanism probably cannot work
because the registration process is so far removed from possible
indemnification. Of course the agency would not issue a registration
if It were contemplating revoking it. Deregistration and indemnifica-
tion occur many years later. More careful application of statutory
pesticide registration requirements at the outset would not have pre-
vented the problems of new information years later.
-42-
-------
INDEMNIFICATION STUDY
Section III
However, there is some evidence that the current
lengthy review process already discourages sone potential product
manufacturers from pursuing the registration process, even though,
since 1972 EPA has actually denied registration for only five active
ingredients.^
(e) Indemnification provides an incentive to comply with
deregistered pesticide disposal regulations
One Congressman argued in the § 15 debate that
indemnification would increase compliance with EPA's regulations
governing disposal of cancelled pesticides, because owners would wish
to comply with these regulations in order to collect their indem-
nities.44
In order for indemnification to serve as an effective
incentive to assure proper disposal, pesticide owners must know that
indemnification is available. While indemnity payments have been used
as an effective incentive when the federal government has sought to
destroy diseased livestock, indemnification has not actually been used
by EPA in a similar way under § 15. Indeed, when EPA sends out stop-
sale orders to registrants and dealers of suspended and potentially
cancelled pesticides, no mention is made of the possible availability
of indemnification. Nevertheless, pesticide registrants may have
viewed § 15 as an incentive to agree to cancellation of their regis-
trations, and the agreements signed with EPA did establish procedures
for retrival and transfer of Silvex to the government in addition to
procedures for filing indemnification claims.
Pesticide manufacturers are generally large com-
panies, well known to the government, and it is relatively easy to
monitor their compliance with the deregistration. Processors,
dealers, and users are much more numerous, small and scattered. Since
the users are even more difficult to Identify, it is difficult to
notify them of cancellation. Indemnification theoretically provides a
more efficient means of passing notice down the chain of market trans-
actions. A requirement that manufacturers repurchase stocks of sus-
pended and cancelled pesticides would accomplish some of the same
purposes achieved by indemnification. The third parties who had
relied on the continued use of the pesticide would be compensated.
They would have an incentive, comparable to the present one, to dis-
pose of a dangerous substance properly. However, what would be lost
in such a scheme would be the motivation of the manufacturer to do all
that it can, rather than merely what it must do to achieve minimal
compliance with a repurchase requirement. The substantial dependence
of the Silvex disposal and claims processing operation on the coopera-
tion and good will of the registrant illustrates the importance of
maintaining their motivation for participation.
-43-
-------
INDEMNIFICATION STUDY
Section III
Indemnification makes it easier for EPA to decide in
favor of an emergency suspension because adverse
econon'i.c impacts will be partially covered by Indem-
nification
The idea behind this contention Is that EPA must con-
sider potential adverse economic impacts when it Is making a suspen-
sion decision, and indemnification might lessen those adverse economic
impacts.
The problem with this argument is that it assumes a
pool of available funds from which EPA can make indemnity payments.
Such a pool of funds does not exist. The Court of Claims Judgment
Fund might serve as such a pool, but its availability to respond to
repeated § 15 claims is certainly open to question. Moreover, con-
sideration of indemnification in making suspension decisions appears
to be contrary to the thinking of Congress because the Agency is
directed to study and take into account only the economic impacts on
the agricultural sector, not on pesticide manufacturers.
(2) Arguments Against Indemnification
(a) Indemnification can discourage suspension decisions
by EPA
This argument is based upon the idea that EPA may
hesitate to ban deadly and hazardous pesticides if it knows that such
a decision may cost the taxpayers millions of dollars, and especially
if indemnity funds must come from EPA's budget.
Although the specter of a "chilling effect" on EPA
has been raised as a serious problem, the prospect of indemnification
does not appear to have been an Important factor in previous suspen-
sion and cancellation decisions. Rather, decisions have thus far been
made on their scientific and policy merits. Clearly, however, the
possibility that EPA might have to reimburse the Court of Claims Judg-
ment Fund for millions of dollars in Silvex indemnification claims
could affect future suspension decisions. Moreover, even if the
prospect of having to pay Indemnification does not deter EPA from
suspending and cancelling a pesticide, the Agency might tend to reduce
any margin of safety by permitting existing stocks of pesticides to be
used or by allowing limited use of the pesticide so as to avoid the
need for indemnification.
(b) Indemnification reduces industry incentives for
safety
During the legislative debate on § IS, the point was
made that indemnification would curb incentives to the chemical
industry to test their pesticides.^ Because the government pays for
-44-
-------
INDEMNIFICATION STUDY
Section III
losses that the manufacturer suffers because of a § 15 pesticide can-
cellation, but pays nothing for stocks produced after the manufacturer
has knowledge disclosing the pesticide's hazard, § IS arguably dis-
courages any testing beyond what is minimally necessary for registra-
tion.
There is, as yet, no clear evidence concerning
whether indemnification actually reduces industry safety activity.
Nevertheless, the possibility of indemnification theoretically provi-
des an incentive to pesticide registrants to wait for government
action, rather than to move voluntarily to withdraw a pesticide from
the marketplace and thereby lose the opportunity for indemnities.
On the other hand, there is ample evidence that at
least some pesticide manufacturers conduct safety research more exten-
sive than is mandated by EPA. Often a manufacturer and its individual
professional employees will desire to conduct such research in order
to protect their business or scientific reputations.
(c) Indemnification for pesticides is unduly preferen-
tial or will prompt a flood of such indemnification
schemes
The contention here is that there is no reason to
treat pesticide manufacturers differently from drug manufacturers or
other regulated parties that do not receive indemnification if their
products are banned.^ An alternative formulation of this argument
Is that the enactment of pesticide indemnification sets a precedent
for allowing indemnification of other dangerous products that are
banned from the marketplace.
As previously noted, there Is a long history of gov-
ernment indemnities to farmers when their products have been destroyed
or determined to be dangerous to health. Pesticide indemnification
seems merely to continue this history, but only a tiny proportion of
indemnities actually paid under § IS have gone to farmers. (Probably
no more than one or two farmers have been indemnified ever under
§ IS.) Indemnification obviously benefits the pesticide manufacturers
and dealers, and gives them an advantage over drug manufacturers, for
example, who do not receive indemnification if their products lose FDA
approval.
Since the enactment of § 15, there has been no flood
of new Indemnification programs designed to cover the losses of manu-
facturers or dealers whose products prove to be dangerous and must be
removed from the marketplace.^ Thus, there Is reason to believe
that § 15 is not widely regarded as a precedent for indemnification of
such parties.
-45-
-------
INDEMNIFICATION STUDY
Section III
(d) For reasons of economic efficiency, the risk of a
pesticide ban should be borne by pesticide manufac-
turers and sellers, rather than by the government
This argument makes the point that the possibility of
cancellation based on new scientific evidence of hazard is an ordinary
business risk which should properly be internalized by the manufac-
turer or seller of pesticides. Indemnification instead allows produ-
cers to gain the benefits and profits of the pesticide, while some of
their risk is underwritten by the government.
Indeed, the substantial profit margins in the pesti-
cide industry appear to reflect this and other risks. The possibility
of cancellation Is only one of the risks of the pesticide industry
along with the risk of product liability, the risk that insects will
develop Immunities to a particular pesticide, and the risk that a com-
petitor will develop a superior product. All are routinely absorbed
by the industry. If the risk is reduced by the government, the pesti-
cide manufacturers would enjoy a windfall profit in the form of risk
premiums without the risk, not an efficient allocation of resources.
f. Conclusion
The indemnification provision has not been used often enough
to draw any specific conclusions about its effect on the behavior of
the agency or of the pesticide industry. In its present form, § 15
operates more like individual payments made after a loss and after an
arduous process of pursuing the claim than like a more automatic pro-
gram that can be counted on before a loss occurs. If it is desired
that either the agency or the industry should take indemnification
into account in their planning, some stable and reliable financing
mechanism will have to be instituted.
The other conclusion that can be made is that the presently
constituted indemnity section does not reach farmers, who have
received almost no payments so far. Furthermore, inventory loss is
not a problem for them because, even if farmers were sent stop use
orders, they hold so little inventory, that it would not be worth their
while to file a complicated and lengthy procedure to obtain reimburse-
ment. Their real problem is with the disruption caused by losing the
pesticide. If present protections are judged to be Insufficient to
help them make a transition, some adjustment of the indemnification
provision would be necessary.
2. Extension of FIFRA § 15 to New Situations
a. Application to TSCA
The Toxic Substances Control Act is the first place to con-
sider an extension of a FIFRA-like indemnification program. The study
-46-
-------
INDEMNIFICATION STUDY
Section III
mandate grew out of a disagreement over whether there should be Indem-
nification under TSCA. There are also similarities in the laws that
make the question of extension to TSCA natural. TSCA is the only
other statute administered by EPA that significantly regulates partic-
ular products rather than processes or wastes. EPA has the authority
under both laws to ban a product, including the existing stocks. An
analysis of the cost of a FIFRA-like indemnification program under
TSCA indicates that its cost would be in the same range as that of
FIFRA § 15.^° However, the transplant appears to be unnecessary
because there are significant differences both in the statutes and in
the substantive area that each regulates.
(1) Statutory Differences Between TSCA and FIFRA
TSCA is primarily an information development, collection
and coordination apparatus, rather than a comprehensive regulation of
hazardous chemicals.^ it differs from FIFRA in many respects. One
Important difference is that EPA does not register new chemicals under
TSCA. There is a requirement that manufacturers give EPA a 90 day
notice before beginning commercial production of a new chemical.
However, this is just a notification, and no test information need
accompany the notice. If the Agency finds reason to suspect a
chemical, it may order the manufacturer to develop additional informa-
tion before beginning production. Such orders, under § 5(e), have
only been issued on nine of the 1,450 chemicals reviewed since the
program began. (Of course, not even this limited review was available
for the 50,000 chemicals introduced before TSCA's pre-market review
was implemented.) There is no factual or legal ground for a company to
rely on the Agency's continued approval of any chemical that passes
through the preraanufacturing notification review process. The Act
specifically provides for several kinds of follow-up regulation of new
and existing chemicals if new information, new uses, or increased pro-
duction volume signal reasons for concern.
Another important difference between FIFRA and TSCA is the
procedure for emergency bans. Under FIFRA, emergency suspensions are
Agency orders. They are reviewable by court on an expedited basis,
but only on grounds of whether there is an imminent hazard and whether
the suspension order was arbitrary or capricious. In contrast,
perhaps in an effort to prevent some of the mistakes that can result
from emergency action, Congress has given the decision to an indepen-
dent decision-maker. Under TSCA, If EPA wants to take action against
an existing chemical that is causing an Imminent hazard, it must peti-
tion a federal district court for seizure or relief. The hearing pro-
vides an opportunity for regulated parties to challenge the action and
its factual and legal basis. The court may order notification of
purchasers, recall, repurchase, or replacement.50
A third important difference in TSCA is the inclusion of
authority to require recalls or repurchases. If it were necessary to
-47-
-------
INDEMNIFICATION STUDY
Section III
remove a chemical from the market, some kind of recall or repurchase
scheme probably would be very helpful. The chemical market is a very
•complex one, and chemicals are most often used as intermediates in the
production of other chemicals. This sometimes happens within one
plant, even within one continuous and enclosed process. Some chemi-
cals are produced as by-products of others. The large volume chemi-
cals typically have many uses, and thus many types of customers with
different purposes. It would be difficult for a mandatory enforcement
program to ferret out a chemical that follows a convoluted path
through the marketplace. A repurchase scheme aids in identifying and
notifying the proper people by tracing them through market trans-
actions. The success of Chevron in obtaining an estimated 90t of the
Silvex products held by its distributors and retailers is due, in
part, to the organization of the voluntary repurchasing effort, as
well as to the prompt payment of cash or credit for the goods.
The comprehensive authority of TSCA, allowing regulation
of chemicals from manufacture through processing and use or disposal,
would allow a broad enough repurchase order to avoid the problems that
occurred in the Trie situation.^ The authority of TSCA can be used
to spread the cost of any ban among all of those involved in manufac-
ture and distribution, rather than concentrating on those who may not
be able to pay.
But repurchase requirements alone might not be enough to
produce this speedy and efficient result. Conversely, it probably
would not be necessary to indemnify for the full cost of the inventory
in order to produce the necessary incentive. (The animal disease
program, for example, gets good results with amounts below the
replacement value of livestock because the rigorous enforcement also
acts as an incentive.)
(2) Alternatives to Indemnification Under TSCA
Section 2 of TSCA declares that the development of ade-
quate data on the health and environmental effects should be the
responsibility of the chemical manufacturers and processors. Never-
theless, Congress provided many cost limiting and cost sharing mea-
sures; illustrating the alternatives to indemnification discussed in
Appendix B. For example, § 8 exempts small businesses from many of
the reporting and recordkeeping requirements. Section 9 contains ela-
borate consultation procedures to prevent conflicting or overlapping
regulations of chemicals. Nearly every section provides expanded
opportunities for participation and comment before agency action
(including the emergency action under § 7). Section 6 mandates a
balance of the risks of a chemical against the costs of various regu-
latory alternatives, and requires the Administrator to choose the
least costly alternative. The cost of testing existing chemicals can
be shared among all the importers, manufacturers and processors of
the chemical.
-48-
-------
INDEMNIFICATION STUDY
Section III
(3) Impeding Statutory Incentives for Self-Regulation
The existence of TSCA, even in advance of full implementa-
tion, has encouraged ami supported various voluntary and private
efforts to reduce the risk of chemicals to health and the environment.
There is evidence that many chemical manufacturers do test new
chemicals, even though the premanufacturing notice does not require
any tests. A successful effort is being made to negotiate with the
manufacturers of chemicals listed by the Interagency Test Committee to
arrange voluntary testing of existing chemicals, without promulgation
of a formal test rule. Some companies are reporting the voluntary
remedial measures taken in response to problems reported to the Agency
under § 8(e). There are advantages for both the chemical Industry and
the Agency in making such arrangements. The chemical companies have
more control over their response to the chemical risk, and in some
cases are able to restrict the amount of confidential Information
given to the Agency. The industry does not have to spend as much time
and money commenting on regulations. Advantages to the Agency to
include decreased cost, speedier results, and increased cooperation of
the regulated industry.
This delicate web of agreements, arrangements, and incen-
tives could be disrupted by the counter-incentive of an indemnifica-
tion program. The possibility of more costly compliance with
regulations and of possible adverse publicity is presently an incen-
tive for the companies to detect and begin to correct problems with
their products. Although indemnification for inventory would not
fully compensate a manufacturer for the loss associated with a ban, it
would be an Incentive to wait for the Agency to take action rather
than to respond immediately to the perceived problem. The impact of
an indemnification program is hard to gauge, but its direction may run
counter to the incentives in the statute and the present method of its
implementation.
(A) Frequency of Bans Under TSCA
One of the primary reasons for setting up an indemnifica-
tion program is to handle a large volume of indemnification payments.
There are several differences between the chemicals regulated under
TSCA and those regulated under FIFRA which may lead to fewer emergency
bans under TSCA. Pesticides are designed to be dangerous to certain
forms of life; they are called "economic poisons." The average chemi-
cal under the broad jurisdication of TSCA is not harmful either to
human health or the environment. Whereas non-pesticide chemicals have
a multitude of uses and vast differences in the exposure to people,
most pesticides are used on the food chain. Anything in the food
chain has a measurable, if attenuated, exposure link to people. In
addition, the exposure to the environment is much greater with pesti-
cides since they are used in fields, forests, along roads and in
homes.
-49-
-------
INDEMNIFICATION STUDY
Section III
In contrast, most industrial chemicals are converted
within a factory into a stable end-use product. Thus, the exposure is
generally only to workers within the factory, and not to the customers
or the general environment. The considerations of toxicity and expo-
sure are the important ones in determining whether and how quickly to
regulate. The lower average toxicity and exposure of industrial chem-
icals indicates less frequent need for emergency bans.
Even when a chemical does need to be regulated under TSCA,
it is more likely to be a regulation of some uses, rather than a total
"ban on all uses. Industrial chemicals often have a wide variety of
uses. Some of them may cause much more exposure or may be more harm-
ful than others. Section 6 requires the Administrator to regulate by
using the least burdensome means that will adequately protect against
the risk. Thus, it is likely that the Administrator would regulate
some uses and not others, leaving a continued market for the chemical.
This was true, for example, of the only ban to date, that of CFC
aerosals. Many other uses of CFC's, in refrigeration, air condition-
ing, sterilization, and plastic formation have not been regulated at
all. Aerosals were considered to cause the most Immediate threat to
the ozone layer, and to be the least necessary to the economy.
(5) Equitable Basis for Indemnification Under TSCA
Many operational errors possible under TSCA would be the
basis for a cause of action under the FTCA. A mistake like the one in
the Mlzokaml Brothers case, that could not be the basis for a FTCA
suit, seems unlikely to occur under TSCA. The chances for error are
reduced by the fact that the Agency does not itself test chemicals, as
the FDA does food. Furthermore, EPA cannot seize a chemical without a
hearing. If there had been a hearing before the seizure in the
Mizokami case, it seems likely that the hearing officer would have at
least stayed the order until another test could been conducted. Most
of the Mizokami's loss was spinach that would not keep for six weeks
while the parties argued over it. Chemicals on the other hand, prob-
ably could survive a dispute without loss of intrinsic value (although
there might be associated losses such as lost business opportunities).
It is also unlikely that there will be a problem like the
FDA announcement that eliminated the market for cranberries in 1959
because it did not distinguish between contaminated cranberries and
untainted ones. The strength of the public reaction to that announce-
ment apparently took everyone by surprise. Both Congress and the
agencies are now more careful in assigning and using the power of pub-
licity. EPA may only add the name of a chemical which it suspects may
present an unreasonable risk to TSCA's "risk list," § 5(b)(4), after a
procedure more elaborate than that required for most rulemakings.
There is one instance in which Congress has considered
indemnification for consequences of a potential action under TSCA. In
-50-
-------
INDEMNIFICATION STUDY
Section III
1980, Congress authorized a grant program to cover the costs of
detecting and correcting problems of asbestos insulation in school
buildings. The aid was for school districts that had to correct an
environmental problem they had not created. The legislation was
passed before EPA had proposed rules requiring inspection of schools,
but the funding has never been provided. In the meantime, many states
have established their own grant programs, or some other form of
assistance for the school districts.
(6) Conclusion on TSCA
There appears to be no reason to transfer an indemnity
program like that in FIFRA § 15 to TSCA. There is no evidence of the
kind of losses envisioned by the drafters of § 15 due in part to the
differences between the statutes and their subject matter. TSCA con-
tains many other methods of preventing, limiting, and sharing the
costs of regulation, and it does not appear that there will be fre-
quent or complete bans. Experience in implementing TSCA is limited,
but so far there are no patterns of inequities that would justify
establishment of an indemnification program.
b. Product Recalls
(1) Description
The other logical extension of § 15 is to product recalls.
However, there are very few such recalls in the EPA-administered
statutes. As mentioned, TSCA contains recall authority, but it has
not yet been used. The only significant recall provision that has
been used by EPA is that in § 207(c) of the Clean Air Act.
Automobiles may be recalled for adjustment if they do not
meet the applicable emission limitations throughout their useful life,
when properly maintained. The recall may occur up to five years or
50,000 miles after their initial sale. Of the 223 recalls since 1973,
163 have been voluntarily undertaken, without any action by EPA, when
the manufacturer discovered a defect. EPA has issued formal recall
orders only 21 times, two of them contested. The remainder were
"influenced recalls," in which EPA discovered a problem and began
working on a recall order, but the manufacturer voluntarily recalled
before it was issued.
The cost per car of recall and repairs has been $15-20 on
average. However, recalls may apply to millions of cars, and the
total cost to the manufacturer can be quite large. It is estimated
that about 50% of the cars in the categories that are recalled are
actually brought in for repair. One reason for indemnification in
this situation might be to increase the participation rate. The
situation is a classic one in which the cost"(or annoyance to the
owner) is not balanced by a benefit to the individual. The benefits
-51-
-------
INDEMNIFICATION STUDY
Section III
are derived only from Che participation of a significant number of
others. It is possible that, if the manufacturers were indemnified
for their costs, they would change their approach to notification or
otherwise make a greater effort to increase participation.
(2) Evaluation
There does not appear to be a high probability of surprise
or agency error in the recall of automobiles. The automobile manufac-
turers are notified whenever EPA test their cars, and are invited to
attend as observers. Thus, they are aware of the data on which EPA
would base a recall from the beginning. Frequently they conduct inde-
pendent tests to confirm it. The statistics suggest that EPA Is
usually correct, because the automobile makers often take voluntary
action at considerable expense, rather than waiting for EPA to order
it. When an order is contested, It can be stayed pending the outcome
of the suit. Therefore, the manufacturer does not have to incur the
expense of complying with what might be a mistaken order, unlike the
case of emergency action to curtail production during the Birmingham
air pollution alert.5^
It is not clear, of course, that indemnifying the manufac-
turer will, in fact, increase participation by car owners. There is
no out-of-pocket cost to the owner in a recall, just the inconvenience
of being without a car. The indemnification cannot change the fact
that there is no individual benefit to the owner. Unless the repair
of the emissions system makes the car run better or more efficiently,
the incentive to the individual Is only that he is contributing to the
public health. An alternative way to increase participation would be
institution of mandatory inspection and maintenance programs. The
threat of failing inspection may be a better incentive for all owners.
(3) Conclusion on Recalls
Indemnification does not appear to be an appropriate
adjunct to the automobile recall program. Recalls are generally not
emergencies. There is ample notice, opportunity to inspect data and
confirm it, and to contest the order'before compliance* The procedure
also reduces the possibility of error on the part of the Agency.
Indemnification would probably not be an effective incentive to
increase the participation of automobile owners.
-52-
-------
INDEMNIFICATION STUDY
Section IIL
C. INDEMNIFICATION UNDER § 202(a)(3) OF THE CLEAN WATER ACT
1. The Existing Program
a. Introduction
Indemnification under § 202(a)(3) of the Clean Water Act of
1977," as amended in 1981,^ was designed, in part, as an incentive
to encourage the development of innovative and alternative wastewater
treatment technologies." The provision was originally drafted in
Conference in 1977 to supplement the incentives created by § 202(a)(2),
which provides federal grant money for up to 85% of the cost of treat-
ment plants, or unit process and techniques thereof, that employ inno-
vative or alternative technology, rather than the 75% (or up to 55%
beginning in FY 1985) normally provided.56 Congressional debate on
§ 202(a)(3) characterized it as an "Insurance policy" covering the
risk of failure of such technology.^
Under § 202(a)(3), EPA is given discretion to fund 100% of the
costs of modification or replacement of an innovative or alternative
facility that fails to meet performance specifications and has signi-
ficantly increased capital, operating or maintenance expenditures. An
additional statutory prerequisite is that the plant failure not be
attributable to negligence. EPA's May 1982 interim final rule under
the section also requires that the failure occur within a two-year
period following Initial operation of the project. °
Section 202(a)(3) has only been utilized in one or two instan-
ces to provide replacement or modification funds for treatment works
failure.59 However, it has been used to encourage grantees to use
innovative or alternative technology.
b. Statutory Language and Regulations
Section 202(a)(3) of the Clean Water Act of 1977, 33 U.S.C.
§ 1282(a)(3), provides:
"(3) In addition to any grant made pursuant to
paragraph (2) of this subsection, the Administrator
is authorized to make a grant to fund all of the
costs of the modification or replacement of any
facilities constructed with a grant made pursuant
to paragraph (2) if the Administrator finds that
such facilities have not met design performance
specifications unless such failure is attributable
to negligence on the part of any person and if such
failure has significantly Increased capital or
operating and maintenance expenditures.
-53-
-------
INDEMNIFICATION STUDY
Section III
Regulatory implementation of § 202(a)(3) is codified at 40
C.F.R. § 35.2032(c), which provides:
(c) Modification or replacement of innovative
and alternative projects. The Regional Administra-
tor may award grant asistance to fund 100 percent
of the allowable costs of the modification or
replacement of any project funded with increased
grant funding in accordance with paragraph (a) of
this section if he determines that:
(1) The innovative or alternative elements of
the project have caused the project or significant
elements of the complete waste treatment system of
which the project is part to fail to meet project
performance standards;
(2) The failure has significantly increased
operation and maintenance expenditures for the pro-
ject or the complete waste treatment system of
which the project is part; or requires significant
additional capital expenditures for corrective
action;
(3) The failure has occurred prior to two
years after Initiation of operation of the project;
and
(4) The failure is not attributable to negli-
gence on the part of any person.
c. Legislative History of § 202(a)(3)
Section 202(a)(3) did not appear in either the original House
or Senate versions of the 1977 Water Act, but was added later in
Conference.60 The subsequent House debate on the Conference Report
characterized the provision as an "insurance policy" designed to cover
the risks created by forcing technology and to provide an incentive
for the use of innovative or alternative technology in wastewater
treatment works.^ It was enacted in order to facilitate achievement
of the benefits of the new technology, despite higher risks associated
with it.
Congress in 1977 wanted to strengthen its encouragement of
Innovative and alternative wastewater treatment systems by adding
financial incentives.^2 It was the sense of the Congress that there
had been insufficient progress in response to the encouraging language
of the 1972 Act. Apparently, the Conference Committee felt that it
was also necessary to reinforce and supplement the incentive provided
by the 10% additional federal grant money provided under § 202(a)(2).63
-------
INDEMNIFICATION STUDY
Section III
In enacting § 202(a)(2) and § 202(a)(3), Congress clearly intended
that local governments be permitted and encouraged to use wastewater
treatment technology Involving higher risks of potential failure.
d. Implementation of § 202(a)(3)
(1) Regulations and Guidance Documents
As quoted above, an interim final rule implementing
§ 202(a)(3) was published May 12, 198264 as part of the new regula-
tions made necessary by the December 29, 1981 enactment of the
Municipal Wastewater Treatment Construction Grant Amendments of
1981." xhe previous regulations were published on September 27,
1978.66 The new (1982) interim rule provisions dealing with modifica-
tion and replacement costs are essentially the same as the 1978
regulations, although the language has been changed for reasons of
clarity and to conform with related changes in the grants program.
A May 1982 draft Guidance Document entitled "Construction
Grants - 1982" contains procedures for awarding 100% modification or
replacement grants. This document establishes the criteria for the
exercise of EPA's discretion in funding replacement or modification
costs. The overall document generally emphasizes an "ongoing effort
to simplify and delegate the municipal construction grants program."
Section 15.2 at 110 states that:
The key to implementation of I&A
[Innovative & Alternative] projects Is the
acceptance of an acceptable level of risk by
you that the I or A project may not work as
predictably as a more conventional treatment
process.
For the purpose of 100 percent M/R
[Modification/Replacement] funding, failure is
defined as the Inability of the entire system
or significant components to meet design per-
formance specifications, where such failure is
due to higher risk elements of design as
determined in the original design documents.
M/R 100 percent funding is not available where
the failure of an I or A system or component
is covered by a warranty or caused by
negligence.
The Guidance Document further explains that evaluation of
requests for 100Z modification or replacement grants will involve a
determination that:
-------
INDEMNIFICATION STUDY
Section III
(a) design performance specifications have not
been met; (b) the failure results in a signi-
ficant increase in O&M [Operation & Mainten-
ance] costs and/or requires additional
significant capital expenditures to correct
the problem; and (c) the failure is not attri-
butable to negligence on the part of any
person. Negligence should be evaluated as the
last item since the determination is based
upon both technical analysis and legal
findings.
Initial screening will reject from further
consideration projects which do not qualify
for obvious reasons and, therefore, do not
require a detailed evaluation. Such reasons
may include expiration of the two-year period,
hydraulic or organic overloading of the system
or lack of an adequate O&M program.
(2) Potential Utilization
Section 202(a)(3) has only been in effect long enough to
result in one or two limited instances of replacement or modification
costs for any failed innovative or alternative systems.
The total facility cost for projects funded under
§ 202(a)(2) through September 30, 1981 is approximately $1 to $1.5
billion. This includes approximately 250 grant awards for innovative
technology and 1000 for alternative technology.^ Because EPA
believes that § 202(a)(3) funds should be used to cover the risk of
failures in unproven technology, it plans to focus on innovative uses
of waste water technologies.
Thus, EPA's exposure for total replacement of innovative
technology funded through September 30, 1981 may be between $200-$375
million, if replacement costs equal the corresponding initial expendi-
tures. Of course, such exposure would require failure of all of the
innovative facilities, as well as complete replacement at an amount
equal to the original costs. It would also require that all of the
failures occur within two years of the date of final inspection.
Therefore, actual funding needs are likely to be far less than the
maximum exposure of $375 million for these projects. If they are as
high as 10% of this amount, total indemnification costs would be on
the order of one percent of the total FY 1980 Construction Grants
Aft
appropriation.00
-56-
-------
INDEMNIFICATION STUDY
Section III
With the 1981 legislation, Congress made the innovative/
alternative program a continuing part of the Construction Grants
program.69 EPA's exposure for replacement or modification will there-
fore also continue*
e. Benefits and Problems of § 202(a)(3)
Although § 202(a)(3) was designed to assure the correction of
a malfunctioning Innovative or alternative wastewater treatment system,
the provision also was enacted to stimulate the use of innovative or
alternative technology. However, its practical effectiveness as an
Incentive is diminished by the lack of a specific funding reserve to
assure the availability of money when it is needed.
Because of the lack of specific funding, the prospect of
indemnification has been viewed with some skepticism by grantees. For
most grantees who have decided to utilize Innovative or alternative
technology, the primary motivating factor seems to have been the
availability of the 10Z higher grant assistance under § 202(a)(2).70
This incentive, however, provides a stimulus for somewhat different
conduct than is promoted by § 202(a)(3). Section 202(a)(2) encourages
grantees to adopt the minimum level of innovative and alternative
technology that will qualify for the extra grant assistance. Section
202(a)(3), in contrast, Is supposed to encourage true risk-taking by
indemnifying against the risk of technological failure of innovative
and alternative treatment works. The former incentive has outweighed
and probably negated the second.
There also is the problem that a potential indemnltee under
§ 202(a)(3) must compete for funding priority with other potential
grantees. In addition, payment of indemnification is entirely discre-
tionary under the Act, which is thus vastly different from a legally
binding commercial insurance policy with specific reserves and
detailed criteria for payment.
f. Problems with § 202(a)(3) That Might be Remedied
As with § 15 of FIFRA, a possible deficiency with § 202(a)(3)
is the absence of guaranteed funding. While a general funding mecha-
nism exists, and replacement and modification grants are authorized to
be paid out of overall grant funds for construction of treatment works,
no money has actually been set aside to cover the contingency of a
facility failure. In other words, there is no assurance that the
grant program itself or any necessary funding under it will actually
exist when failure occurs. Further, a grant application for replace-
ment or modification funding will have to compete on the priority list
against grant applications for original construction. Thus, rather
than being an insurance policy that guarantees funding, § 202(a)(3)
merely establishes eligibility if funding is available.
-57-
-------
INDEMNIFICATION STUDY
Section III
One way to correct this would be to set aside specific money
In a government trust fund reserve, similar to a commercial insurance
reserve, with use limited to replacement or modification funding. An
alternative would be to use a portion of the grant money set aside for
innovative technology to purchase a commercial insurance policy
covering the risk of system failure. Either of these approaches would
provide a greater incentive for use of innovative technology since
potential grantees would be more assured of protection from the risk
of failure due to the use of a new and unproven technology*
A second issue is that EPA has placed a two-year limitation on
the time period during which an idemnifiable failure must occur. EPA
concluded that a two-year period would provide ample operational
experience with the treatment works, so that a failure due to the use
of innovative technology would very likely occur, if at all, during
this period. Nevertheless, certain types of failures might not become
evident until after the two-year period — for example, unrealized
life-cycle cost savings — and some type of allowance might be made
for this contingency. Comments on EPA's regulation before it was
promulgated indicated substantial support for longer time periods —
as long as ten years — but this is not the only possible modification
to alleviate this alleged problem.
g. Summary Evaluation of the Existing Program
Section 202(a)(3) is an example of the use of indemnification
to support a government policy objective. This provision represents a
type of indemnification somewhat similiar to private contractual
indemnification.
Even though § 202(a)(3) has only had very limited application
to pay indemnification for a failed innovative facility, it has been
offered as an incentive to stimulate the use of new technology.
However, Its value as an incentive has been limited by the lack of
guaranteed funding, and by the more important and partially counter-
vailing incentive of innovative technology construction grants under
§ 202(a)(2).
2. Possible Modifications of the Existing Program
a. Expansion of Eligible Costs
The Municipal Wastewater Treatment Construction Grant Amend-
ments of 1981^1 increased the number of technology innovations poten-
tially eligible for funding under § 203(a)(2), and for "insurance"
under § 202(a)(3), by making innovative or alternative unit processes
and techniques eligible for the first time. Under the amendments, it
Is no longer necessary for innovative technology to comprise a sub-
stantial portion of the treatment works to be eligible for the indem-
nification of § 202(a)(3).
-58-
-------
INDEMNIFICATION STUDY
Section III
Further expansion of § 202 along these lines, however, now
seems unworkable. Virtually all innovation is now covered within
the program, and any further additions, therefore, would necessarily
involve technology that is not innovative or alternative. While an
argument could be made that all failures of wastewater treatment tech-
nology cause losses worthy of EPA reimbursement, the principal ratio-
nale for § 202(a)(3) remains the creation of an incentive to take the
special risk of innovative or alternative technology. That incentive
would be undermined if similar indemnification were available for
treatment works, processes or techniques that are not innovative or
alternative.
b. Expansion of Eligible Grant Recipients
In addition to its construction grants program, EPA has also
conducted a much smaller demonstration grants program under §§ 104 and
105 of the Clean Water Act.'2 The grantees under this program were
both municipal treatment works and parties seeking to demonstrate par-
ticular pollution-control technology which thereafter could be mar-
keted to industrial sources. The function of the demonstration grants
from their perspective was to reduce the costs of demonstrating a par-
ticular technology where these costs might otherwise interfere with
the technology's ever reaching the market. From EPA's perspective,
the function of demonstration grants was analogous to the function of
§ 202 grants: to promote innovation in water pollution control.
Approximately $20-30 million was budgeted during 1967-70 for indus-
trial demonstration projects and a similar amount for municipal demon-
stration projects.'-* However, the relatively low level of municipal
participation in the demonstrations ultimately led to the enactment of
§ 202(a)(2), and § 104 funding for municipal demonstrations was cur-
tailed thereafter.
The industrial demonstration grant program, as a practical
matter, is limited to technological innovations that cannot possibly
be utilized on a publicly-owned facility because, if such utilization
Is possible, § 202(a)(2) provides more favorable terms to the control
device manufacturer. Since some publicly-owned facilities in fact
treat a single finns's wastes almost inclusively, the opportunity may
exist for industrial pollution control technologies to be demonstrated
as innovative technology under § 202(a)(2). Thus, the scope of the
demonstration grants program Is limited by definition. Moreover, Its
practical scope is further limited by the virtual termination of
funding In recent years.^
The possibility that the offer of Indemnification might pro-
duce a practical incentive effect on technology innovation under the
demonstration grants program is thus quite small. It is reduced fur-
ther by the fact that it is not the risk of failures that seems to
inhibit pollution control demonstrations, but rather the cost of the
demonstrations themselves. Thus, in most instances, the barrier to
-59-
-------
INDEMNIFICATION STUDY
Section III
final marketing of the technology is effectively removed if the grant
is made, and no further incentive for the manufacturer is needed. The
industrial source similarly appears to have no special disincentive to
participate because of the risk of failure since, In most circum-
stances, it is merely providing an effluent stream to be cleaned. At
most, If the risk of failure proves to be a special disincentive in a
peculiar circumstance, some individual remedial measure — such as a
special insurance policy — might be called for. In fact, however,
there have apparently never been any instances where indemnification
was sought, or Insurance against failure thought to be needed, under
the clean water demonstration grants program. ->
3. Possible Applications of the § 202(a)(3) Model In Other EPA
Programs
a. Introduction
The Clean Water Act's innovative and alternative technology
program in § 202 integrates a construction grants program with an
indemnification provision applicable to publicly-owned wastewater
treatment facilities. It defines desired behavior, not otherwise
required to meet EPA18 regulatory requirements (innovative and alter-
native technology). That behavior involves a risk that may deter some
parties who would otherwise consider undertaking the behavior (risk of
plant failure). If harm actually occurs, the indemnification program
offers reimbursement for some of the damages incurred (the costs of
retrofitting or replacing the wastewater treatment plant — but not
the damages from pollution caused by the failed plant). Thus, the
incentive is created by the promise of reimbursement (sometimes made a
specific "selling point" in negotiations over construction grants).
b. Candidate Indemnification Measures
(1) Clean Air Act
A fundamental difference exists between the Clean Water
Act's § 202 incentive and the various innovative and alternative
incentives under the Clean Air Act: there is no substantial govern-
ment funding of the cost of compliance with the Clean Air Act.
Rather, in virtually all instances, the Clean Air Act contemplates
that the economic burden of the air program should be borne by regu-
lated parties. This policy does not preclude reliance on Indemnifica-
tion incentives to foster desired behavior, but such Incentives are
logically more defensible where the judgment has been made to share at
least some compliance costs.
In contrast to § 202 of the Clean Water Act, a primary
incentive to use innovative and alternative technology to comply with
Clean Air Act requirements is the possibility of obtaining a limited
waiver of applicable emission limitations under § lll(J) (sources sub-
-60-
-------
INDEMNIFICATION STUDY
Section III
ject to New Source Performance Standards) and § 113(d) (existing
sources subject to state implementation plan requirements). Such
waivers eliminate some of the risk a source incurs by utilizing inno-
vative rather than conventional technology, since the principal harms
that result from Che failure of such technology are pollution (the
effects of which are largely external to the source and arguably sanc-
tioned by the waiver) and regulatory violation costs (which are
waived). Thus, the need for an indemnification incentive Is not
established.
Another mechanism, however, is somewhat more analogous to
§ 202. Under the Clean Air Act, EPA uses demonstration grants to
stimulate innovative pollution control technology.76 Typically, these
demonstration projects are cost-shared, with the government picking up
as much as half the cost, and the rest being financed by private
parties, such as a vendor of a new control device whose demonstration
costs are prohibitive without EPA's assistance.77 However, unlike the
innovative and alternative technology program under the Clean Water
Act, and like the Clean Water Act Industrial demonstration grants
program, many of the ideas for demonstration funding come from such
private parties. This differs from the "classic" incentive indem-
nification model, since EPA does not need to recruit private party's
participation in the demonstration. On the other hand, EPA does
instigate some demonstration projects on its own. Nevertheless,
because funds for demonstrations depend on the existence of a tech-
nology to be demonstrated, the program depends more on Initiatives
outside EPA than does the § 202 program.
The potential benefits to vendors of a successful demon-
stration are clear. Thus, reduction of the cost of the demonstration,
by means of an EPA grant, may be a sufficient Incentive for such
parties, especially if they can proceed without fear of liability. On
the other hand, host sources may be less inclined to participate. To
be sure, some of these sources may wish to help solve a current or
potential pollution control problem, such as finding a more cost-
effective method of control, a method by which a future emission limi-
tation can be achieved or an appropriate control method for future
facilities. This is quite different, however, from the public reci-
pients of § 202 funds, for which compliance with EPA standards without
EPA grant money is a practical impossibility.
The risks of participation are similar for the demonstra-
tion grants program as for § 202(a)(3). A demonstration of equipment
that fails will often require that the equipment be replaced or con-
verted to its original or some other configuration. Another potential
risk of failure is that the demonstration technology may interfere
with the production processes of the emission source. Both of these
types of harms Involve costs that are normally not a part of the ori-
ginal grant. On the other hand, they may be similar to other business
risks, commonly absorbed by industry but quite impossible for public
facility owners to accommodate.
-61-
-------
INDEMNIFICATION STUDY
Section III
In theory, these types of risks can be remedied through
the use of Indemnification and, In at least one demonstration grant
circumstance, the need for Indemnification was actually perceived. In
an EPA demonstration contract with Combustion Engineering Involving a
demonstration project at Utah Power & Light Co., indemnification was
regarded as essential by the host source before it would agree to par-
ticipate in a project involving the alteration of boiler combustion
with a new type of burner and related equipment.'" In particular, the
host source required indemnification against any damage to the boiler
or loss of production due to the project. EPA did not agree to Indem-
nify the host source directly, however, since it chose instead to
include approximately $500,000 of contract funds for a privately-
placed insurance policy to cover potential losses.79
(2) RCRA
Under RCRA, an analogy to the objectives of Innovative,
alternative or demonstration programs can be articulated, and the pos-
sible utilization of an indemnification incentive is thus suggested.
Hazardous waste technology does Involve risks, and those risks argu-
ably contribute to the barrier to hazardous waste technology implemen-
tation. To the extent that they do, indemnification could plausibly
be used to stimulate wider implementation. Such indemnification could
not presently be attached to a grant from EPA, since no such grants
are presently made. However, the lack of an existing administrative
mechanism for effecting indemnification would not necessarily preclude
the adoption of a new mechanism to achieve that purpose.
c. Rationale for Indemnification
Indemnification for failed innovative technology does not
directly stimulate the use or testing of such technology. Instead, it
serves to remove certain disincentives, such as the risk of potential
failure of the innovative technology, the risk that the source's
equipment will in some way be damaged, and the risk of environmental
impairment. Thus, indemnification can function only to complement
whatever primary incentives support the use of Innovative technology.
These incentives may be the prospect of commercial rewards to a manu-
facturer of a successfully tested new pollution control technology,
the anticipated benefits to a source in meeting regulatory require-
ments, or the potential for a higher level of government grant funding
than would be available for conventional technology. In other words,
In order for indemnification to operate in this context there must be
an underlying commercial, regulatory, or government funding incentive
which supports the use of Innovative technology in the first place.
The Clean Air Act demonstration project program offers a
potential context for the use of indemnification, closely analogous to
§ 202(a)(3). The program involves activity which EPA desires to
encourage in support of its Clean Air Act objectives (demonstration of
-62-
-------
INDEMNIFICATION STUDY
Section III
new technology). That activity may involve a risk that makes it
uneconomical to the party who would otherwise engage in it (equipment
failure, production interruption, environmental impairment liability).
If that risk were to result in harm, significant costs might be
imposed (for retrofit or replacement, for lost production or for the
payment of damages). Thus, the promise of indemnification provides an
incentive for participation (for example, by contractural arrange-
ment) .
The analogy to RCRA is less precise. EPA has not yet adopted
a policy of directly fostering Improved hazardous waste technology
implementation, although the policy may be implied as operating stan-
dards for treatment, storage and disposal are developed. Hazardous
waste technology does Involve risk, but the technology EPA may wish to
promote would probably involve fever environmental risks than conven-
tional technology, while the more important barrier to its implemen-
tation is its higher cost and the comparatively low demand for it.
Nevertheless, if the technology proves faulty, the losses that result
may be substantial and, for this reason, a promise of indemnification
might provide some incentive for implementation of the technology.
d. Evaluation of Indemnification
The desirability of Incentive indemnification must be eval-
uated in each distinct context where it is proposed. In each context,
a threshold question Is whether this type of indemnification is viable
without a corresponding government funding program similar to
§ 202(a)(2). That question can be answered affirmatively, only if
government policy to encourage use of innovative technology is sup-
ported by a willingness to fund modification or replacement costs if
necessary.
In contexts where there is such support, the next question is
the choice between indemnification and commercial insurance covering
the same risks. The choice of indemnification will defer the actual
expenditure until after the harm has occurred and, depending upon the
financing structure used, may also place the expenditure In some
budget other than that of the demonstration project. Such costs,
however, should properly be internalized to the demonstration project
program if a decision is made to provide government funding of such
costs. Where coverage is needed, its cost should be absorbed by EPA
in the same way, and only to the same extent, as other innovative
technology costs. Thus, the principal rationales for using indemnifi-
cation rather than commercial insurance would be that such insurance
Is commercially unavailable, that the premium costs are comparatively
much higher than indemnification expenditures would be, or that there
are significant benefits to deferring the expenditure.
Since much of the innovative technology that is the subject of
demonstration grants Is designed to operate at the "end of the pipe,"
-63-
-------
INDEMNIFICATION STUDY
Section III
any continuing harms to be incurred are often remediable merely by
removing the device being demonstrated. After such a removal, the
host source is often no worse off, other than any liability incurred,
than if it had needed to invest in conventional technology in the
first place. Since its Investment in the demonstration technology is
small or zero, the host source's losses from failure of that technol-
ogy will often be readily absorbed by the pollution control equipment
manufacturer, who is hoping that the demonstration project will prove
successful.
Only in the case where actual production processes must be
altered to accommodate the demonstration project technology and
altered back when the technology fails, will the host source suffer a
substantial loss for the failure. In such circumstances, the purchase
of private insurance may be more appropriate than an offer of indemni-
fication by EPA. While the type of coverage required may be unusual
and costly, this reflects thu relative rarety of the need for
coverage, and hence also for indemnification.
The future of the use of demonstration projects to spur the
use of Innovative and alternative technology under the Clean Air Act
is very uncertain at this point, however, recent budget cuts have
reduced the funding for demonstration projects to the point that it
may no longer be a viable stimulus. The predictions of budget support
for demonstration projects may indicate that support for an indemnifi-
cation program would be lacking.
In the absence of either construction grant funding or demon-
stration project grants, it is still theoretically possible that
indemnification would be desirable, If there were a sufficient regula-
tory or commercial Incentive for the use of innovative technology, and
the risk of failure or equipment damage were deemed to be a deter-
mining impediment. While the evidence does not Indicate that a
situation of this type has yet arisen, use of Indemnification might be
considered more closely if the circumstances seem to warrant it. At
this point, the need for indemnification has not been shown in the air
pollution control context*
The technological-feasibility problems presented by RCRA
differ from those of other innovative technology programs principally
in that the existing implementation barriers to hazardous waste treat-
ment, storage or disposal technology development appear to be economic
rather than technical. Thus, the ample list of unused or underuti-
lized hazardous-waste technologies derives, not from their need to be
demonstrated or improved, but from their relatively high cost, com-
pared to conventional disposal practices. The main barrier to
improved technology is the lack of demand for that technology at its
current price. That price, in turn, reflects not only the relative
complexity of the improved technology, compared to conventional
measures, but also the difficulties of siting the technology in the
face of public opposition.
-64-
-------
INDEMNIFICATION STUDY
Sect lo'n "ill
For this reason, it may be unreasonable to believe that a pro-
nise of indemnification would substantially alter the net Impact of
existing Incentives and barriers to hazardous waste technology imple-
mentation. The magnitude of the other factors influencing hazardous
waste technology decision making appears to be so much greater than
the magnitude of the risk of technology failure, that Indemnification
can be expected to alter decision processes In only rare circum-
stances.
e. Conclusion
Use of Indemnification as an incentive to encourage use of
innovative technology In either the air pollution or hazardous waste
areas has limited potential applicability. Based upon current
evidence, the extension of a § 202(a)(3) type of indemnification to
other EPA Program areas does not seem warranted.
-65-
-------
INDEMNIFICATION STUDY
Section III
D. INDEMNIFICATION UNDER § 113(b) OF THE CLEAN AIR ACT
1. Legislative History and Background
a. Introduction to § H3(b)
Section 113 of the Clean Air Act80 provides the primary
enforcement authority pertaining to stationary source requirements of
the Clean Air Act. Section 113(a) provides for administrative notices
of violation and orders; § 113(b) provides for civil relief; § 113(c)
provides for criminal relief; and § 113(d) provides for delayed
compliance orders*
Section 113(b), requires EPA in certain cases, and authorizes
it in others, to bring a civil action for an Injunction or for a civil
penalty, or both. It also provides for partial indemnification In the
case of unreasonable enforcement:
[l]n the case of any action brought by the
Administrator under this subsection, the Court
may award costs of litigation (including
reasonable attorney and expert witness fees)
to the party or parties against whom such
action was brought In any case where the court
finds that such action was unreasonable.
This provision was enacted as a part of the 1977 Clean Air Act
Amendments, which Instituted major changes to the basic federal
enforcement scheme established in 1970.°l The 1970 Amendments
stressed administrative enforcement at the federal level for the first
time under federal pollution control legislation. By 1977 EPA had
proceeded with hundreds of administrative enforcement actions to
require compliance with applicable emission limitations. However,
some of these actions, for practical reasons, lengthened compliance
timetables past deadlines established in the 1970 Amendments. The
1977 Amendments substantially constrained EPA's enforcement authority
by limiting the criteria under which such extensions might be granted
(§ 113(d)) and by requiring that EPA go to court to obtain long-term
injunctlve relief, at the same time, Congress provided a judicial
civil penalty mechanism to facilitate economic sanctions for unwar-
ranted compliance delays. These amendments seemed to forecast more
substantial reliance on judicial enforcement by the agency; therefore,
it is not surprising that Congress would have included a provision
under $ 113 designed to constrain unreasonable enforcement litiga-
tion.82
b. Legislative History to the § 113(b) Indemnity Provision
Section 113(b)'s indemnification provision originated in the
House Interstate and Foreign Commerce Committee as an amendment to
-66-
-------
INDEMNIFICATION STUDY
Section III
H.R. 10498 in the 94th Congress (1976 session).83 The House Committee
Report on that bill explained that the provision was Intended to pro-
tect parties against "wholly unwarranted enforcement actions" and to
restrain "overzealous enforcement;" the report also commented that,
"Only If the bringing of the action Is arbitrary, capricious, frivo-
lous, harassing, or wholly without basis in fact or law should the
court consider such an action 'unreasonable'."8^ The Senate followed
the House version, as did the Conference Committee.
No bill was ultimately enacted In 1976. When Congress began
again in 1977, the House bill did not include the indemnity provision
under § 113(b). Instead it arose in the Senate. As introduced and
ultimately enacted, It was identical to the 1976 House version. The
Report of the Senate Committee on Environment and Public Works
explained, simply, "The intent of the provision is to prevent harass-
ment of parties by the Government."85 The Conference Report provided
no further explanation.""
c. Legislative Issues Involving Indemnification
There is very little.recorded debate in the legislative
history on the issue of indemnification; and no mention is made of
specific circumstances warranting the inclusion of such a provision in
1977 Amendments. However, the issue has been debated in other
contexts, as discussed below.
(1) Background
The American private litigation system is one of the few
in the world that do not award court costs and attorney fees to the
prevailing party as a matter of course. The "American Rule," removes
a potential barrier to litigation. A party must expect to pay his own
way, win or lose, but he need not forego a legitimate cause of action
out of fear that he may lose and incur the litigation costs of his
opposition. The rule thus reflects a strong American orientation
toward the importance of encouraging litigation.87 In this view the
power to litigate is the power to ensure not only that the law is
applied, but that it is continually refined to suit new or changed
circumstances.
Discretionary exceptions to the American Rule are recog-
nized in at least three broad categories; (1) to penalize frivolous or
intentionally harassing litigation, delay tactics, and other actions
amounting to abuse of process; (2) to effect an equitable distribution
of litigation costs when the litigant has performed a benefit for a
larger class of persons (e.g., a stockholder suit that benefits all
stockholders); and (3) to encourage litigation to facilitate accom-
plishment of statutory goals (e.g., encouragement of the public to act
as private attorneys general where statutes expressly allow for such
an award). In all three categories, the exception continues to pro-
-67-
-------
INDEMNIFICATION STUDY
Section III
mote the underlying philosophy of the American Rule by discouraging
unwarranted litigation (in the case of the first) and by encouraging
desirable litigation that might otherwise be unaffordable (in the case
of the last two). These exceptions are BO well established as to be
considered part of the Rule.
According to the principle of sovereign immunity, the
government is subject only to such legal claims as it may expressly
allow by legislation. Although there are a fairly large number of
statutes"' in which Congress has authorized recovery of attorneys'
fees among private litigants, few statutes have allowed recovery
against the Federal Government and even fewer in enforcement cases.
Most EPA implemented statutes, for example, allow recovery of attor-
neys' fees by citizens who sue to require the Administrator to perform
a nondiscretionary duty,90 but only the Clean Air Act and the Toxic
Substances Control Act allow recovery of attorneys' fees in other cir-
cumstances, and only the Clean Air Act allows recovery in an enforce-
ment action.
In 1981, however, Congress enacted general legislation
waiving immunity for attorneys' and expert witness' fees. This legis-
lation, the Equal Access to Justice Act, applies to all litigation by
and against the Federal Government involving small business (it also
applies to certain formal administrative adjudications).^1 Its pri-
mary purpose is to erase an economic barrier to legitimate litigation
although the legislative history also indicates that small industry
may be more susceptible to unfair litigation tactics pursued by the
government. ^ This statute applies to all EPA programs, including
enforcement under the Clean Air Act.
(2) Section 113(b) in Contrast to Other Approaches
Key aspects of the § 113(b) indemnification option are
that the award is discretionary, It may be made only to the party or
parties against whom the enforcement action is brought (i.e., not to
the Administrator), it may be made only where the court finds that
such action was "unreasonable," and the award for attorney and expert
witness fees must be "reasonable."
(a) The American Rule
Under the American Rule an award is usually made only
if a party ultimately prevails, and it is not necessarily conditioned
on a determination that the action was unreasonable (for example, an
award might be made if a large class of persons has benefited by the
suit). Conceivably, a successful defense to EPA enforcement might
benefit other potential defendants, for example, by exposing the
weakness or error in an underlying regulation. However, an award for
this reason is not allowed under § 113(b).
-68-
-------
INDEMNIFICATION STUDY
Section III
The question of restricting the award only to pre-
vailing parties arose as a legislative issue on several occasions
under the 1977 Amendments — relating to § 307 as well as § 113(b).
Ultimately, this restriction was rejected on the basis that a har-
assing suit might nevertheless be won by the government, and that a
suit under § 307 may provide a benefit to others even if It is lost by
the plaintiff. In both cases the legislative objective would be
thwarted by denying attorneys' fees, perhaps on the basis of a simple
legal technicality.93
Under the American Rule a much clearer concept of
"unreasonable" agency action than exists under § 113(b) has evolved.'^
Under the "bad faith" exception, action meriting such an award is
almost always limited to circumstances when the judge perceives an
intent to harass or delay — in other words, action amounting to an
abuse of process. The policy behind the American Rule approach is to
establish a disincentive to misconduct, but also to preserve existing
incentives to litigate (the fundamental purpose of the rule); thus,
litigation pursued in good faith should not be restrained even if it
may later prove to be unfounded, but litigation pursued in bad faith
should be restrained.^
(b) The Surface Mining Control and Reclamation Act
Section S2S(e) of the Surface Mining Control and
Reclamation Act^ is the only other federal environmental protection
statute that allows an award of attorneys' and expert witness* fees in
the context of enforcement. In contrast to § 113(b), it allows a dis-
cretionary award by the Secretary for costs resulting from administra-
tive proceedings, including formal as well as informal adjudications,
and by a court for judicial review of those proceedings. Judicial
enforcement proceedings, however, are not expressly covered.
According to § 525(e), an award may be made to either
party ". . .as the court . . . deems proper." Both the legislative
history and implementing regulations promulgated by the Department of
Interior's Office of Surface Mining (OSM) make it clear that an award
should be made only when the proceeding has been initiated "in bad
faith and for the purpose of harassing or embarassing the permit-
tee."^7 The clear purpose of the provision was to establish a disin-
centive to such action. Although several petitions have been filed
under this section, no award has yet been made.98
A major difference between § 113(b) and § 525(e), as
interpreted and applied by OSM is that an award may be made under
§ 525(e) for costs during administrative proceedings. Another varia-
tion pertains to what circumstances justify an award. The language of
§ 113(b) requires a finding that the action was "unreasonable;"
§ 525(e), on the other hand, only requires that the award be deemed
"proper." The legislative history to § 525(e), however, strongly sup-
-69-
-------
INDEMNIFICATION STUDY
Section III
ports an Interpretation consistent with the American Rule — that an
award he made when the action is frivolous or for harrassraent; and it
has heen restricted by rule only to circumstances clearly involving
bad faith."
(c) The Equal Access to Justice Act
The Equal Access to Justice Act 100 provides for
attorneys' fees awards to small businesses and, like the Surface
Mining Control and Reclamation Act, authorizes such awards in admi-
nistrative as well as judicial actions, although the administrative
actions are restricted to formal adjudications. When administrative
action Is involved, the award is mandatory, subject to demonstration
that the agency position was substantially justified or that special
circumstances make the award unjust. The result Is the same when
judicial action Is involved; in addition, sovereign Immunity Is
waived, and the common law adopted for circumstances not already
covered under the Act. In both cases a party must prevail to receive
an award. Guidance is given on how to determine reasonable fees. An
agency may pay from appropriated funds, and funds are specifically
authorized for that purpose.
Since § H3(b) of the Clean Air Act does not cover
administrative actions, the Equal Access to Justice Act expands the
circumstances where attorneys' fees can be recovered under Clean Air
Act proceedings (although formal adjudications occur in only a few
narrow circumstances under the Act). With respect to judicial
enforcement, § 113(b) may be in conflict with the new statute, since
it allows for discretion In making the awards and since the § 113(b)
standard may be somewhat more favorable to the agency (although this
would be a matter of interpretation).
Substantial legislative debate was given to the issue
of what circumstances would justify an award under the new Act. A
purely discretionary standard, similar to the Surface Mining Control
and Reclamation Act and the American Rule, was rejected in part
because it was thought that an agency would be reluctant to award fees
against itself in an agency proceeding. It was also thought that an
agency required more specific legislative direction.1^1
Also rejected was a standard proposed by the Depart-
ment of Justice, that would limit recovery to circumstances in which
the government action was "arbitrary, frivolous, unreasonable, or
groundless, or the United States continued to litigate after it
clearly became so." It was thought that this standard would be too
difficult for the regulated party to prove. Instead, the burden was
shifted to the government to prove that its action was "substantially
justified" (or that other special circumstances would make an award
unjust).102
-70-
-------
INDEMNIFICATION STUDY
Section III
According to a Department of Justice spokesperson,
only one petition has been filed to date (the effective date of the
Act was October 1, 1981). The Department is currently processing that
claim.103
2. Application to Date
a. Cases
There is only one case to date In which an award under
§ 113(b) has been made or requested.10^ This case involved Apache
Powder Co., a manufacturer of explosives in Arizona, who had been
cited by both the State agency and EPA for violating Arizona nitric
acid plant emission requirements.10^ Compliance problems were docu-
mented as early as 1972 by the state, and In response to a request for
assistance by the State, EPA began documenting violations soon after.
Ultimately, both the State and EPA agreed to a compliance plan which
involved replacement of the noncoraplying facility and retirement by
January, 1976. When, as late as 1978, the facility was still
operating in violation of emission requirements, EPA initiated a suit
to recover civil penalties, to require compliance with the Arizona
regulation and to enforce a § 113 abatement order previously issued by
the Agency.
Apache moved for summary judgment on the ground that EPA had
failed to approve variances submitted by the State to EPA, which, if
approved, would have excused the violation. Although information is
not clear on the point, it seems that during consideration of this
motion, Department of Justice lawyers determined that an independent
basis for challenging the case existed — that EPA had not satisfied a
jurisdictlonal prerequisite to the action by approving the underlying
regulation as a part of the State Implementation Plan. Although the
Regional EPA Office argued that the regulation had been approved, the
Department, with EPA concurrence, filed a Confession of Error and
agreed to pay costs and counsel fees.
A review of the reported facts in the case and subsequent
legal proceedings indicates a strong likelihood that throughout all of
the proceedings the Agency acted in good faith. The potential proce-
dural deficiency that resulted in withdrawal from litigation had been
identified by regional lawyers long before initiation of litigation,
and had been resolved on the basis of legal arguments that probably
would not be considered frivolous.10^
b. Indemnification Factors
(1) Administration
There is no indication in the Apache Powder case that
administration of the indemnity provision was burdensome or presented
-71-
-------
INDEMNIFICATION STUDY
Section III
specific problems. Costs of litigation are normally fairly easily
documented. While one might anticipate dispute over the issue of
whether specific costs are "reasonable," a Department of Justice
attorney Indicated that consideration of such claims is routine, not
time-consuming, and rarely controversial. When they are controver-
sial, he indicated that courts generally follow the Department's
recommendation, and the controversy is usually resolved without a
substantial additional effort by the Department. The preferred route
Is to negotiate a proposed award out-of-court. ^'
A General Accounting Office staff person, who actually
processes the award, indicated that payment is also a very simple
administrative procedure. Notice of the Award is forwarded by the
Department, and a check is drawn on the Judgment Fund of the U.S.
Treasury. This fund, which supplies wide-ranging claims, is main-
tained at a high enough level so that attorneys' fee awards represent
an almost insignificant depletion.^8
(2) Financing
The financing mechanism for § 113(b) is through the Judg-
ment Fund, authorized by 31 U.S.C. § 72A(a) for the purpose of paying
judgments against the United States, administered by the GAO, and
financed from general revenues. According to a GAO staff person,
Congress routinely replenishes the fund with specific appropriations
and does not usually scrutinize individual payouts.^9 Since this fund
serves as a reservoir for judgments and fees in a wide range of cases,
it is doubtful that the costs in a single claim for attorneys' fees
would be significant in comparison to all claims against the fund.
(3) Costs
Approximately $10,000 was awarded Apache Powder in fees
and costs.*10 These were considered reasonable and in line with costs
Incurred in suits of similar scope and duration according to a
Department of Justice attorney.*" Since this is the only award to
date, the total costs of Implementing the § 113(b) indemnity provision
are probably not significant. There is no indication that EPA resour-
ces have been devoted to the provision — no regulations address the
provision, and persons Interviewed were not aware of any formal or
informal agency efforts to consider the provision; Indeed, no agency
policy memoranda which address indemnification under § 113(b) could be
identified.
(4) Alternative Relief Mechanisms
The experience to date suggests § 113(b) provides a relief
mechanism that is not required. This is Indicated primarily by the
infrequency of awards (one). However, it may also be reflected in
recent agency indications that air pollution enforcement for station-
ary sources will shift more toward a State assistance role.112
-------
INDEMNIFICATION STUDY
Section III
(5) Equity, Fairness, Due Process
Although § 113(b) provides partial relief to a person who
has suffered wrongdoing on the part of EPA, this is not its major
purpose. Rather, Congress intended primarily that § 113(b) deter the
Agency from future wrongdoing; and the award of attorney fees works
both as an Incentive to the economically disadvantaged party to raise
legitimate defenses to unreasonable litigation, and as a disincentive
to unreasonable litigation in the first place. By compensating for
the smaller inequity (unreasonabale litigation costs) § 113(b) pre-
vents the larger inequity (submission to unreasonable enforcement
without defense). Viewed In this manner, an award in Apache Powder
would not have been appropriate if Apache would have defended anyway.
This would have been the case, according to an attorney who repre-
sented Apache during the proceedings.*•"
It should also be noted that the company considered the
amount of the award to be satisfactory.^* It was not determined,
however, whether additional uncompensated expenses were incurred, or
whether the company feels that attorney costs (and other costs) prior
to litigation deserve compensation. EPA's enforcement action began
almost five years prior to the 1978 court action. During that time,
extensive negotiations occurred between the company and EPA. Evidence
strongly indicates, however, that negotiations with the state agency
would have continued to occur in any case, and that the company may
not have reduced its overall resource commitment In these negotiations
even if EPA had not been involved.
(6) Statutory Indemnification Objectives
A primary purpose of the indemnity provision was to reduce
the likelihood for "unreasonable" enforcement litigation by EPA. Fee
shifting might accomplish this by allowing for an effective defense,
where no defense (or an ineffective defense) would otherwise be
pursued: knowledge that a defendant would pursue its case might deter
the Agency from using litigation as an unwarranted threat; pursuit of
a' legitimate defense would also protect against a miscarriage of
justice; finally, a fee payment obligation might serve as an economic
or moral deterrent to unreasonable litigation.
Apache Powder alone provides little basis for testing the
objectives of § 113(b).Even If one assumes that as a result of
Apache Powder more careful attention would be given to state implemen-
tation plan status in future cases, It is not clear what role the
indemnity clause has played. It does not appear that the agency
action was taken in bad faith. Furthermore, discussions with agency
officials indicate greater embarassraent over having to abandon the
case than over the payment.115 In addition, it Is likely that Apache
Powder would have pursued its defense even if no such award were
allowable.116
-73-
-------
INDEMNIFICATION STUDY
Section III
The infrequency of § 113(b)'s usage (only one case to
date) may be viewed as an Indication that the provision is not neces-
sary. Although it could also be viewed as an indication that the
provision Is actually working effectively as a deterrent, this inter-
pretation is of doubtful merit. Most agency persons interviewed were
either unaware that the indemnity clause existed, or indicated that It
had never been actively considered when planning and pursuing litiga-
tion. Internal clearance procedures are so rigorous .that there is
very little risk that actions will be groundless.H7
Further, a review of recent congressional hearings and
other reports involving the Clean Air Act and numerous interviews with
Industry representatives disclosed little evidence that EPA judicial
enforcement is viewed as harassing, erroneous or without legal justi-
fication. H8 The vast majority of all cases filed under § 113(b), in
fact, go to court in the form of negotiated consent decrees.H*
Finally, If the indemnity clause is needed to encourage companies to
defend against the agency when the imbalance of litigation resources
would otherwise be dissuasive, then one would expect to see more
reliance on the provision than has occurred.
(7) Impact on Other Statutory Missions
An area of concern in any prospective Indemnification
program, especially those designed to have an impact on specific
agency actions, is whether the program will result in other unintended
adverse impacts. A logical area of concern under § 113(b) would be
the potential for interference with EPA enforcement objectives. There
has been no indication, however, that the indemnification provision of
§ 113(b) has had either a positive or a negative impact on enforcement
under the Clean Air Act.
3. Summary Assessment of § 113(b)
a. It Is Questionable Whether The Indemnity Provision In § 113(b)
Is Needed To Accomplish Congressional Purposes.
Concern over the potential for harassing, or otherwise
"unreasonable" enforcement litigation under the Clean Air Act seems
unfounded. Only one case has Involved such an award, and research
disclosed few instances of complaint regarding the good faith of, or
errors involving EPA enforcement litigation. In fact, the great
majority of enforcement actions are resolved by negotiated consent
decree. In addition, EPA budget requests, enforcement clearance pro-
cedures, and other air program guidance within EPA suggest that liti-
gation occurs only after rigorous screening procedures designed to
ensure that it is justified. Finally, it is doubtful that the § 113(b)
fee-shifting provision would actually serve as a disincentive to
unreasonable litigation, since EPA is not accountable for the fee
under the present system. A ruling against EPA on the case would pro-
bably be a more persuasive deterrent.
-74-
-------
INDEMNIFICATION STUDY
Section III
Experience to date also does not support the need for an
•Indemnity provision to overcome an inequitable distribution of litiga-
tion resources between EPA and enforcement case defendants* Even if
such an inequity has existed, the limited usage of § 113(b)'s indem-
nity clause indicates that the inequity has been an unimportant factor
in Judicial enforcement proceedings. Although it Is possible that
such limited usage has occurred because the indemnity provision provi-
des an insufficient enticement to defend, the more likely reason is
that enforcement cases have been justifiable. Few cases, in fact,
have occurred which were not settled by consent decree. Agency
spokespersons also expressed their sentiment that inequities do not
actually exist — industry that is subject to enforcement litigation
normally has more resources for defense at its disposal than the
Agency has for prosecution. This view could not be confirmed with
actual data, however.
b. Potential Problem Areas
(1) Definition of "Unreasonable" Enforcement
The legislative history tends to support two interpreta-
tions of the term "unreasonable." In one interpretation, action may
be "unreasonable" only when it is pursued in bad faith (an interpreta-
tion consistent with common law application of the American Rule on
fee shifting). In the other, it may be "unreasonable" for less
restrictive grounds — when the action was taken in good faith, but a
court determines the decision to be arbitrary or without adequate
legal basis.
(2) Who pays?
Under § 113(b) the agency is not accountable for payment.
Rather, the award is paid by GAO from the Judgment Fund. It is possi-
ble, that payment by agency would create a more significant disincen-
tive to engage in "unreasonable" enforcement. There is no present
indication, however, that a more significant disincentive (or, in
fact, any disincentive at all) is needed.
(3) Equal Access to Justice Act Overlap
This overlap exists only to the extent small businesses
are involved. If so, a "prevailing" small business may attempt to
obtain more favorable treatment under the Equal Access to Justice Act
than is provided under § 113(b). The award in such a case Is manda-
tory, subject to a demonstration by the Agency that its enforcement
action was "substantially justified," or that other special factors
warrant no award. Although the burden of proof has clearly shifted to
the agency, it may be possible for the agency to argue that the sub-
stantive standards are essentially the same as under § 113(b).
-75-
-------
INDEMNIFICATION STUDY
Section III
4. The Applicability of the § 113(b) Model to Other EPA Statutes
Major characteristics of § 113(b) indemnification are: (1) the
need to restrain unreasonable judicial enforcement pursued by EPA, and
(2) the need to encourage persons to litigate who would otherwise
refrain from defending an enforcement action In court for economic
reasons. While the need for such a program under the Clean Air Act is
not clearly justified, it could nevertheless be justified In enforce-
ment litigation under other statutes administered by EPA. For most of
the same reasons, however, a similar indemnification provision is pro-
bably not warranted under other EPA statutes.
a. There Is Little Evidence That Judicial Enforcement Undertaken
By EPA Is Arbitrary, Harassing, Without Legal Authority, Or
Otherwise Unreasonable.
Discussions with industry and agency spokespersons, and a
review of current literature disclosed very few judicial enforcement
cases that could be termed "unreasonable" in the sense that the agency
brought the action erroneously or without legal justification.*2^
Most Industry complaints relating to unreasonable EPA enforcement con-
cern underlying regulations or policies, or they relate to enforcement
target selection or negotiating practices of the Agency.
b. There Is No Reason to Expect That Ureasonable Judicial
Enforcement Will Occur in the Near Future.
Although enforcement authorities and mandates vary among the
different EPA statutes, the internal control over enforcement actions
recommended for litigation is the same. This control is rigorous,
requiring a careful demonstration of facts and legal theories that
must be cleared through numerous layers of regional, headquarters, and
ultimately Department of Justice concurring authority.121 Interviews
with present and past enforcement officials indicated that this
screening process works well to prevent unwarranted enforcement.'-22
EPA*s enforcement policies also emphasize that only the major
violators will be aggressively pursued, and the states will be relied
on more to fulfill the overall compliance goals established by
Congress. Even the major violators will be encouraged, as in the
past, to settle out-of-court.*2-*
c. There is No Indication That Such a Provision Would Actually
Work as a Disincentive to Unreasonable Enforcement.
Persons involved in the implementation of enforcement programs
who were interviewed in the course of research agreed that payment of
attorney fees for unreasonable enforcement would be irrelevant as a
disincentive to unreasonable litigation. Prevention of such action
would occur primarily because of the internal clearance proce-
dures.124
-76-
-------
INDEMNIFICATION STUDY
Section III
If the agency were forced to divert substantial funds from
previously authorized activities in order to cover litigation Indem-
nities, it is possible that this would serve more as a disincentive to
litigate whenever there Is a risk of loss, rather than as an effective
disincentive to engage In unreasonable enforcement.^^
d. An Alternative Exists Which May Obviate the Need for Extend-
ing § 113(b) to Other Statutes, If Such a Need Exists.
The Equal Access to Justice Act promises an award to pre-
vailing parties in all EPA enforcement litigation if EPA cannot
justify Its action.12° This Act applies only to small businesses*
However, it is the small business that is most likely to have economic
reasons not to defend aggressively against EPA enforcement. Larger
companies, which are more often subject to enforcement, typically have
adequate resources to defend against EPA action, and the availability
of an indemnity for litigation costs will not usually be the persua-
sive factor in a decision to defend.
-77-
-------
INDEMNIFICATION STUDY
Section III
E. INDEMNIFICATION PURSUANT TO § 311(1) OF
THE CLEAN WATER ACT
1 . Legislative History And Background
a. Introduction to § 311(1)
Section 311 of the Clean Water Act127 ia the primary regula-
tory mechanism under that statute for restricting one-time or episodic
discharges of oil and hazardous substances1^8 that may be accidental
or intentional, and are not susceptible to regulation by permit. It
contains both preventive and remedial requirements, as well as sepa-
rate sanctions and a revolving fund to finance cleanup. This fund is
administered by the United States Coast Guard (USCG) and financed by
special appropriations, penalties, and moneys recovered from respon-
sible parties.
Section 311 's prevention requirements are implemented by EPA
for non-transport dischargers and by the USCG for transport dis-
chargers. However, its spill response requirements are divided geo-
graphically: the USCG maintains jurisdiction in the coastal zone,
ports and harbors, and the Great Lakes, while EPA maintains jurisdic-
tion in other U.S. waters. Other federal agencies are Involved, as
needed, in the formulation and implementation of a national contin-
gency
A major portion of § 311 is structured to ensure a rapid,
effective response to a spill. Its provisions require immediate noti-
fication when a spill occurs and impose strict liability (with certain
limited defenses) on the discharger for the costs of cleanup. If the
discharger's actions are inadequate, the Federal Government may ini-
tiate cleanup on its own and later sue the discharger for reimburse-
ment. A limit on liability exists unless the discharge Involved gross
negligence or willful misconduct. If the discharger is able to assert
a defense under §§ 311(f) and (g), however, he is not required to
clean up.^0 TO the extent that he does clean up, § 311(1) entitles
him to reimbursement for the reasonable costs involved. That section
provides:
(i)(l) In any case where an owner or opera-
tor of a vessel or an onshore facility or an
offshore facility from which oil or a hazar-
dous substance is discharged in violation of
subsection (b)(3) of this section acts to
remove such oil or substance in accordance
with regulations promulgated pursuant to this
section, such owner or operator shall be
entitled to recover the reasonable costs
incurred in such removal upon' establishing, in
a suit which may be brought against the United
-78-
-------
INDEMNIFICATION STUDY
Section III
States Government in the United States Court
of Claims, that such discharge was caused
solely by (A) an act of God, (B) an act of
war, (C) negligence on the part of the United
States Government, or (D) an act or omission
of a third party without regard to whether
such act or omission was or was not negligent,
or of any combination of the foregoing causes.
(2) The provisions of this subsection shall
not apply in any case where liability is
established pursuant to the Outer Continental
Shelf Lands Act, or the Deepwafer Port Act of
1974.
(3) Any amount paid in accordance with a
judgment of the United States Court of Claims
pursuant to this section shall be paid from
the funds established pursuant to subsection
(k) of this section, ^l
The primary benefit and legislative intent of § 311(1) is that
it encourages rapid cleanup by the party most knowledgeable of the
content and amount of a spill, and the party most likely to be first
aware that it has occurred. It accomplishes this result primarily by
providing an economic incentive where there may be no other adequate
Incentive to take corrective action because of the cost involved and
the confidence that there would be no liability.132 in addition, it
removes a potential barrier to rapid private cleanup by postponing the
need to resolve potential issues of liability.*^
Section 311(1) thus utilizes government indemnification to
encourage a regulated party to take immediate steps to mitigate the
harmful impacts of a sudden pollution occurrence. Such a program is
potentially relevant to all areas of pollution control supervised by
EPA where unforeseeable and unavoidable discharges might occur, and
'where the availability of an indemnity might be persuasive in the
encouragement of immediate mitigation efforts to avoid imminent
environmental damage. In such a case, it is possible that regulated
parties might provide a more effective, cost efficient public service
than the Agency assigned primary responsibility to do so.
b. Legislative Background
Although the oil spill provisions of § 311 date back to the
Oil Pollution Act of 1924, most of the oil spill programs in effect
today originated in or are modeled after the Water Quality Improvement
Act of 1970.134 Coverage of hazardous substances was added by the
Federal Water Pollution Control Act Amendments of 1972.l35 However,
implementation of hazardous substance removal authority was delayed
-79-
-------
INDEMNIFICATION STUDY
Section III
until after amendments to the Clean Water Act were enacted in 1978,
which clarified EPA's authority and responsibilities relating to the
designation of hazardous substances and the determination of harmful
quantities*13^ Currently, however, EPA authority over hazardous sub-
stance spills is exercised primarily under the Comprehensive Environ-
mental Response, Compensation and Liability Act of 1980 (CERCLA),137
and § 311 is used primarily for the prevention and regulation of oil
spills.138
The 1970 Water Quality Improvement Act provisions relating to
oil spills derived from numerous House and Senate bills offered in the
wake of the Torrey Canyon, Santa Barbara, and other major oil spills
of the late 1960's. Extensive congressional hearings at the time
reviewed the adequacy of existing statutory authority to prevent, con-
tain and remove such splllso Ultimately, comprehensive legislation
was developed In both the Senate and the House. However, a § 311(1)
equivalent arose only in the Senate.*3* This provision was modified
slightly In conference and retained as § 11(1) of the Act.14^
Available legislative history does not indicate who proposed
the original Senate version of § 311(1) or the reasons offered at the
tlme.14^ A further review of House and Senate Hearings on the pro-
posed legislation discloses almost no discussion of the issue and it
appears not to have been controversial.14^ However, a primary thrust
of oil spill cleanup policy has always been to encourage prompt pri-
vate action. This policy was articulated in the first "National
Multi-Agency Oil and Hazardous Materials Pollution Contingency Plan"
(September 1968), and has been repeated in many statements of legisla-
tive and regulatory policy since.*43
The Senate report accompanying S. 7 noted simply:
S. 7 provides an opportunity for the owner
or operator of a vessel or onshore or offshore
facility to immediately remove any oil dis-
charged. It is the intent of the committee to
encourage removal of oil by the owner or
operator of the discharging source.
In many instances, the owner or operator of
a vessel or onshore or offshore facility will
know of a discharge prior to any agency of the
U.S. Government and be in the best position to
take early action to prevent or minimize
damage. As testified to by the oil and
shipping industry it will be in the best
Interest of the owner or operator to take
immediate measures to reduce damage from an
oil spill.144
-80-
-------
INDEMNIFICATION STUDY
Section III
c. Other Statutory Approaches
There are three other statutes pertaining to oil spills and
one other statute pertaining to hazardous substance spills which are
similar in basic design to § 311 of the Clean Water Act. These
include, for oil, the Trans-Alaska Pipeline Authorization Act of 1973
(TAPAA),145 the Deepwater Port Act of 1974 (DWPA),14* the Outer
Continental Shelf Lands Act Amendments of 1978 (OCSLA),147 and for
hazardous substances, the Comprehensive Environmental Response, Com-
pensation, and Liability Act of 1980 (CERCLA).148 Each of these
statutes imposes strict liablity for certain costs associated with
spills and establishes a separate fund, in part, for the purpose of
providing an effective response to spills not adequately provided
under § 311.
(1) The Existing Oil Spill Fund Statutes149
Like § 311, each of the other oil spill fund statutes
(TAPAA, DUPA and OCSLA) authorizes government cleanup and provides the
government with a right to obtain reimbursement from the responsible
party unless a defense is available. Again like § 311, each statute
also establishes a limit on liability, so that to the extent the limit
is exceeded, the fund pays for certain costs that might otherwise have
been incurred by the responsible party.
There are numerous differences, however. Most notably,
the § 311(k) fund size Is much smaller; it is financed primarily by
appropriations, while the other funds are financed primarily by fees
assessed against the oil facility and vessel owners; the scope of
liability is much smaller under § 311, excluding property damages,
third party cleanup costs, and other types of claims covered under the
other statutes. In addition, discharger liability limits vary, as do
defenses that may be asserted against the government in cleanup cost
recovery proceedings.
Although no provisions identical to § 311(1) exist, both
OCSLA and DWPA allow for the fund to cover discharger cleanup costs in
two ways. First, both statutes allow for a claim against the fund for
such costs If a defense to liability exists.15^ Second, both statu-
tes limit the discharger's liability in such a way that cleanup costs
exceeding that limitation might be charged against the fund. 51
TAPAA, on the other hand, makes the right-of-way holder absolutely
liable for cleanup costs,^2 an
-------
INDEMNIFICATION STUDY
Section III
to resolve these differences (along with other objectives) by pro-
posing oil Superfund legislation. For several years such legislation
was proposed and debated (most recently as H.R. 85) but never passed
Congress. 155
A review of the most recent legislative debates disclosed
no express concern over the need for a change in approach to § 311(1)
or to similar provisions under the other funding statutes. However,
significant concern was expressed regarding the need for more strin-
gent sanctions to encourage dischargers to take prompt action.15^
Although no provision identical to § 311(1) emerged, a
somewhat similar approach was adopted which was closer in design to
the OCSLA and DWPA models. Section 103 of H.R. 85 would have allowed
a claim against the fund by the discharger for "removal" costs if a
defense to liability existed or if such costs exceeded the statutory
limit on liability. (A similar claim could also be made by a non-
responsible third party.) Under § 104 limits on liability would not
apply if the incident were due to willful misconduct or gross negli-
gence or a violation of applicable federal regulations, or if the
owner or operatior failed or refused to assist in cleanup. Defenses
would have included acts of war, natural phenomena and third party
responsibility (a "third party" would Include governmental
entities).157
(3) The Comprehensive Environmental Response, Compensation
and Liability Act (CERCLA)
CERCLA addresses a much larger variety of pollution inci-
dents than any of the oil spill legislation. Ironically, it does not
include oil spills, even though early initiatives supporting Superfund
legislation were largely directed at coordinating and improving the
various oil spill statutes.15® In 1979, however, Superfund momentum
shifted toward hazardous substance pollution problems. Primary con-
gressional concern was focused on the thousands of abandoned or Inac-
tive hazardous waste disposal sites which required expensive remedial
action. However, concern was also expressed over chemical spills.
In congressional Superfund hearings EPA Deputy Adminis-
trator, Barbara Blum testified that as many as 1,800 hazardous sub-
stance spills occur each year that are not subject to regulation under
§311 because they do not reach navigable water.15' She went on to
testify that other EPA authority to deal effectively with such inci-
dents was also limited in some major respect — usually having to do
with funding or procedural constraints.1^
r
CERCLA addressed these problems directly by providing an
extensive funding mechanism ($1.6 billion over a five year period,
compared to the $35 million revolving fund under § 311(k)) and stiff
sanctions for failure to report a spill or to take required cleanup
-------
INDEMNIFICATION STUDY
Section III
steps. At the same time, many of the other issues that have plagued
implementation of the oil spill statutes were resolved.
While CERCLA contains no provision identical to § 311(i),
a similar result could be accomplished under § lll(a), which provides
that the President may pay any claim for necessary response costs
incurred by any person (other than the government) as a result of
carrying out the national contingency plan, so long as such costs are
approved under the plan and certified by the responsible federal
official. However, the approval requirement in this section has been
interpreted by EPA to require pre-authorization.^1 This requirement,
along with a current EPA policy to authorize only a very minor amount
of cleanup costs by persons other than governmental organizations or
approved contractors marks a significant difference between CERCLA and
§ 311(1).1^2 in effect, these policies seriously restrict voluntary
cleanup action.
Three primary factors have been cited in support of this
Agency position: (I) cleanup techniques for chemical spills (as
opposed to oil spills) require highly specialized knowledge, not only
to prevent wasted effort, but also to prevent mistakes that may com-
pound the environmental threat or complicate further cleanup, or both;
(2) containment and cleanup action require special safety precautions
because of the dangerous nature of many chemicals (more so than oil),
and the government should not establish economic incentives to engage
in dangerous conduct, especially when the risks may be minimized by
encouraging the use of specific contractor organizations known to have
the requisite skills; and (3) it is desirable to maximize EPA cost
control in order to protect the fund — unanticipated cleanup costs,
not subject to federal supervision, would be an unacceptable drain on
limited resources. ^
Undoubtedly, cost control is of primary significance.
There is a tremendous federal incentive (reflected in the legislative
history to CERCLA, as well as agency regulations)^ to require pri-
vate cleanup whenever possible, since the total anticipated cleanup
costs, according to one estimate, may exceed fund resources by several
hundred percent.165 Offering reimbursement for voluntary cleanup
could result In numerous claims, ultimately draining the fund and
interfering with EPA prioritization of fund expenditures — and per-
haps resulting in an unacceptable commitment of resources to the
administration and adjudication of claims (without preauthorization of
expenses, a greater number of non-legitimate claims would be
expected).
Persons directly involved in spill cleanup, moreover, feel
that inherent differences in the characteristics of chemicals and oil
increase the opportunity for error and dangerous exposure and that the
potential benefit in encouraging inexperienced private voluntary
cleanup in the limited circumstances that might be managed effectively
without preauthorization would not justify the risks involved.
-83-
-------
INDEMNIFICATION STUDY
Section III
Nonetheless, others have Indicated that the potential for
increased environmental harm cannot be dismissed, and the costs of
cleanup for certain events may be more extensive. It may be expected
that spills will occur for which the discharger Is clearly not liable
and for which he will be unwilling to take immediate cleanup steps
without federal reimbursement. For some of these spills, federal
cleanup will be more difficult and cost more because of the delay, and
greater environmental damage will occur.
It Is perhaps too soon to tell whether the approach under
CERCLA is more appropriate than the approach under § 311(1). Although
Superfund has been in effect for almost one and one-half years, imple-
mentation has proceeded slowly and conservatively.*-""
2. Application To Date
a. Cases167
Since 1973 there have been 45 cases brought under § 311(1) of
the Clean Water Act, all involving oil spills, and totalling approxi-
mately $6.3 million in claims. Fourteen of the cases have resulted in
settlement and four have resulted in an award by the U.S. Court of
Claims. Fourteen cases were dismissed by the court and 13 cases are
currently pending.
Approximately 80 percent of the cases filed to date have
Involved an act or omission of a third party. Approximately half of
these cases have involved vandalism. Over 60Z of the completed third
party cases have been settled In favor of the plaintiff.
Only one claim involving negligence on the part of the U.S.
Government has been filed. In that case, the USCG wrongfully accused
the plaintiff of an oil discharge. The plaintiff filed a claim for
$936, the smallest claim submitted to date, and the case was settled.
Half of the remaining cases (approximately 10Z of the total) were
claimed to involve acts of God. Half of these were settled In favor
of the plaintiff. Finally, eight of the cases involved insurance com-
panies that had paid for a loss incurred as a result of an oil spill
and then sought reimbursement from the Court of Claims.
b. Indemnification Factors
(1) Costs168
Total anticipated payments for claims filed to date are
approximately $1.2 million, or $117,000 per year for the 10 year
period. If 1972 Is excluded (it was the first year of the program and
there was only one case) and 1979-1981 is also excluded (additional
cases may still be filed for these years), the annual average is
approximately $188,600.
-84-
-------
INDEMNIFICATION STUDY
Section III
While the average claim has been about $119,000, the
average claim for which payment has been made was approximately
$69,000. The average payment to date has been $39,000.
The future number and types of cases brought to the U.S.
Court of Claims under § 311(1), and resulting costs may reasonably be
expected to be similar to those submitted to date. There is always
the possibility that a very large case will be filed which greatly
modifies the payments for a given year. However, there have been a
sufficient number of cases to date to allow for an estimate which,
over the long run, should be fairly accurate.
In addition to payments of claims, additional costs are
incurred in administering the program. Most of these involve Depart-
ment of Justice lawyers responsible for defending against the claims
and Court of Claims adjudication costs. In addition, there are minor
costs involved in administering the revolving fund under § 311(k).
These costs are not readily available for inclusion in this report.
Although the costs to date and predictable future costs
are not insignificant, no basis has been identified for concluding
that the costs are unacceptable In view of the benefits derived. In
fact, the actual reimbursed cleanup costs incurred by the discharger
may be less than what they would be if the state or Federal Government
were to conduct the cleanup. This is not simply because the private
polluter may be able to manage cleanup more efficiently, but because
the delayed reaction time inherent in government initiation of cleanup
may result in greater costs because of the Increased difficulty
involved in containment and removal. However, this is a matter of
speculation, and no data was available which addressed this issue.
(2) Financing
The financing mechanism for § 311(1) Is through § 311(k)
which establishes a revolving fund operated by the USCG and financed
through a combination of appropriations, penalties against § 311
violators, and costs recovered by the government. The fund, which Is
currently maintained at $35 million, is available for all cleanup
action, including reimbursement under § 311(1), federal cleanup, and
certain state and local activities as well.
Although it was hoped initially that the combination of
penalties and federal cost recoveries would be sufficient to replenish
the fund, this has not happened. Based In part on this experience,
funds established to finance cleanup In subsequent legislation (the
Outer Continental Shelf Land's Act, Deep Water Port Act, Trans-Alaska
Pipeline Authorization Act, and CERCLA) are financed in large part by
special fees imposed on potential dischargers. Proposed oil spill
Superfund legislation (H.R. 85) would also follow these later models.
-85-
-------
INDEMNIFICATION STUDY
Section III
The total payout associated with § 311(1) to date, approx-
imately $700,000, appears to represent a small burden on the § 311(k)
fund and would not in itself appear to justify revision of the
financing system for such payments.
(3) Administration
The administration of claims under § 311(1) Is straight-
forward. Suit is filed in the Court of Claims. An adversary pro-
ceeding ensues, with Department of Justice lawyers representing the
§ 311(k) fund. Claims may be settled; judgments are appealable; and
payments are made by the USCG from the § 311(k) fund upon presentation
of a request In accordance with 33 C.F.R. § 153.411.
Persons from the USCG have suggested, however, that the
Court of Claims procedure may be too lengthy and cumbersome for the
small number of cases involved. Docket delays occur, and resolution
of many cases might occur sooner If less formal or non-adversarial
procedures were utilized to consider whether prerequisites to recovery
are met. It is possible that these delays interfere with a primary
statutory objective — that the promise of recovery serve as an incen-
tive for the non-liable discharger to take rapid cleanup steps.
However, this could not be documented.
In other circumstances involving statutory incentives to
proceed rapidly with oil spill cleanup (e.g., the Outer Continental
Shelf Lands Act, the Deepwater Port Act, and the Trans-Alaska Pipeline
Authorization Act) funds available for private cleanup efforts are
financed primarily by a fee levied against the owner of the oil, and
claims procedures are administered by the agency or a third party.
Although this alternative might allow for a more efficient processing
of claims under § 311(1), the USCG may be reluctant to assume such
duties: the budget implications would probably be considered unfavor-
able; furthermore, the USCG has traditionally viewed claims adjustment
as an area of responsibility that is inconsistent with and might
detract from its primary operational duties, and the USCG has con-
tended in the past that it does not have adequately trained persons to
assume such duties.*"'
(4) Fairness, Equity, Due Process
Fairness is not ordinarily an issue for consideration in
the Incentive Indemnification situation, because the purpose of the
mechanism Is not to remedy some inequity that may have occurred, but
to encourage a nonliable discharger to assist in eliminating an
environmental hazard. However, experience indicates that an inequity
periodically occurs. Occasionally, the USCG makes a tentative deter-
mination that a party is liable, that party proceeds with cleanup, and
later it is determined that some other party was actually responsible.
Often, in such a case, the discharger Is not certain whether or not he
-86-
-------
INDEMNIFICATION STUDY
Section III
is responsible, and the need for immediate response does not allow
time for determining liability prior to initiating cleanup. It is
also possible that the Federal Government will clean up a spill, then
assess and collect costs from a party erroneously determined to be
responsible (the USCG reports that this happened on one o-casion)
In both cases the party may turn to § 311(1) to be indemnified
for the government error. However, If it turns out that the party was
not the discharger, he cannot recover, since § 311(1) limits recovery
only to a discharger.m This restriction does not exist under OCSLA
and was not Included In H.R. 85.
(5) Statutory Indemnification Objectives
Section 311(1) has worked to encourage timely, cost-
effective private cleanup of certain oil spill incidents and thus has
fulfilled its primary purpose on those occasions. However, the small
number of claims since 1972 (45) compared to the large number of
reportable oil spills since 1972 (one estimate exceeds 125,OOO)172
indicates that § 311(1) is not a major factor in accomplishing
cleanup. Most cleanup is initiated by the federal, state or local
government, or by a responsible party who admits or is found to be
liable for cleanup by the government or who is Ineligible for indemni-
fication.173
This suggests that other mechanisms utilized under § 311
designed to accomplish cleanup or establish cleanup incentives have
been working well (e.g., national contingency plan, surveillance,
notification, state indemnification, penalty provisions, etc.). Most
oil spills are now detected relatively quickly and cleanup usually
occurs, although whether cleanup is always effective is a matter of
debate.174
With respect to chemical spills, there is no record of any
attempt to obtain reimbursement under § 311(1). Yet, such spills have
occurred in significant numbers during the years since enactment of
§-311.175 Several factors appear to account for this. First,
§ 311(1) could not actually be used for chemical spills until
September 28, 1979, when EPA promulgated regulations on reportable
quantities for hazardous substances — a necessary prerequisite to
determining the existence of a violation under § 311(b)(3), which, in
turn, is a precondition to a cause of action for a claim under
§ 311(i)(l). Second, EPA and the USCG have traditionally discouraged
unsupervlsed private response efforts Involving chemical spills
because of the complexity and dangers involved.
-87-
-------
INDEMNIFICATION STUDY
Section III
Other possible explanations for the limited use of
§ 311(1) involving chemical spills include:
• Until recently very little containment and removal
action was considered possible for chemical spills.
• When small spills are Involved (most chemical spills
are small In comparison to oil spills) the cost of
cleanup rarely merits the cost of pursuing a claim
under § 311(1).
• Since enactment of CERCLA (December 11, 1980) the
USCG and EPA have relied on CERCLA for response to
chemical spills, rather than~§ 311, and dischargers
may have concluded that they no longer have status
under § 311(1).1>6
(6) Alternative Mechanisms
Primary alternatives to § 311(1) include (1) more elabo-
rate surveillance and response capability on the part of government,
(2) more effective spill disincentives and spill prevention regula-
tions; (3) more effective private cleanup incentives; and (4) elimina-
tion of the Incentive altogether along with substitution of a
provision requiring absolute liablity for cleanup.
(a) Expanded Government Surveillance and Response
Improvements in government surveillance of and
response to oil and chemical spills have been areas of continual pro-
gress. As a practical matter, however, budget and other administra-
tive constraints virtually eliminate effective governmental oil spill
response management without discharger assistance. It Is well
accepted that all of the reportable oil spills (over 8,000 a year
according to one estimate)^" and their location (throughout most of
the states and territorial waters) could not be effectively addressed
by available governmental response resources. This, in addition to
the fact that the discharger is most often in the best position to
take effective immediate mitigating action, has resulted in the
current legislative strategies to promote private action — strict
liability for government cleanup except when certain defenses apply,
and when those defenses apply, compensation under § 311(1).
While It is conceivable that the government might
undertake cleanup In the few cases that merit a claim under § 311(1),
there are at least two drawbacks. There is a risk that in some
Instances the failure of the discharger to take immediate stop-gap
measures will create a more complicated and costly cleanup problem;
also, the government would be forced to make an early determination of
liability in close cases to allow for an appropriate choice of action.
-88-
-------
INDEMNIFICATION STUDY
Section III
This would result in less flexibility and a greater administrative
burden than currently exists at the time that a spill occurs.
(b) Enhanced Spill Prevention Strategies
Section 311 already gives substantial emphasis to
spill prevention.178 Regulations on oil spill prevention currently
exist. * In addition, fines and civil penalties are allowed under
both § 311 and CERCLA.'80 It Is conceivable that more highly sophis-
ticated prevention techniques and a severe sanctions policy would
diminish spill Incidents. However, there Is general agreement that a
certain number of spills are Inevitable, regardless of the exercise of
preventive care.181 As long as this is the case, spill prevention
techniques and measures to encourage their application will never be
adequate substitutes for spill response techniques.
(c) Enhanced Cleanup Incentives
Alternative Incentives are probably inappropriate, at
least for non-liable dischargers, since it has been demonstrated Chat
Indemnification under § 311(i) actually works to encourage desired
cleanup that might not otherwise occur as effectively or as effi-
ciently. °^ Some doubt must remain on this issue, however, in view of
the different approach recently adopted under CERCLA for chemical
spills. Under CERCLA, EPA will allow for reimbursement of a non-
liable party only if the expenses have been preauthorized as consis-
tent with the National Contingency Plan.18-*
A possible drawback to the CERCLA approach is that
certain desirable containment and removal actions could be delayed or
would not occur because of the time required to obtain preauthoriza-
tlon and because of possible monetary limits that would be Imposed on
the private response. Possible advantages, on the other hand, include
more effective safety and cost control under government supervision,
and more effective allocation of fund resources toward higher priority
incidents. Whether the CERCLA approach will prove to be the better,
one.must await further experience in the implementation of CERCLA, and
will require in any case that consideration be given to differences in
the statutory goals of CERCLA (use of the fund for long-term and non-
emergency remedial activity and hazardous waste sites) and to Inherent
differences Involving dispersal and cleanup characteristics of the
pollutants (chemicals under CERCLA, oil under § 311(1)).
Nonetheless, certain changes instituted in the oil
spill fund statutes following § 311, as well as in CERCLA, might prove
beneficial under § 311 as mechanisms to encourage or facilitate pri-
vate cleanup assistance. These include:
• Expansion of § 311(1) to include non-dischargers
(with restrictions to prevent clearly inappropri-
-89-
-------
INDEMNIFICATION STUDY
Section III
ate or Ineffective cleanup, or counter productive
profiteering).
• Agency administration of claims rather than
administration of claims In the Court of Claims.
• Authority to order cleanup and stiff penalties
for failure or refusal.
• Expanded funding (financed, perhaps, by potential
dischargers).
(d) Elimination of § 311(1)
Elimination of the § 311(i) incentive altogether
would not seem wise without concurrently imposing absolute liability
for cleanup on the owner or operator. This liability would be neces-
sary to ensure that the government Is not solely responsible for
cleanup when the owner is clearly not at fault. If the government
were solely responsible in such a situation, there would be a greater
risk of environmental and other property damage, not only because of
the natural delay inherent In a notification/response system, but also
because the government would be forced to make a more careful determi-
nation of liability prior to any cleanup action. This determination
would be necessary, since the discharger would have no ability to
recoup his expenses if he cleaned up, then was able to show later on
that he was not liable. Since a determination of liability would pro-
bably be opposed by the discharger, it could prove to be difficult and
time consuming, and the government might be encouraged to clean up the
spill and Incur the costs involved rather than become bogged down in
dispute over the issue of liability.
The no-fault, absolute liability for clean-up alter-
native, on the other hand, presents a more sensitive political issue.
Presumably, under such an alternative, the discharger could proceed
against a responsible third party for damages. But this would not be
an effective solution if unforeseeable and unavoidable acts of God or
war were involved. In the past, Congress has rarely been willing to
impose absolute liability without the benefit of any defense, and has
chosen instead to adhere to the progressive strict llablity doctrines
adopted by many courts and state legislatures which recognize certain
limited defenses out of fairness to those who could not have foreseen
or avoided the incident and are wholly without fault.^
(7) Impact on Other Statutory Missions
In general, implementation of § 311(1) appears to have
promoted cleanup goals of the Clean Water Act, at least for the claims
which have been rendered. A brief survey of the facts In each case
decided to date indicates that there have probably been few, if any,
-90-
-------
INDEMNIFICATION STUDY
Section III
unwarranted recoveries. Persons who fail to take reasonable pre-
cautions against a spill usually do not recover in the Court of
Claims.185
It is possible, on the other hand, that vigorous defense
by Department of Justice lawyers in the Court of Claims, -combined with
the long delay in resolution (suits are still pending where claims
were filed as early as 1975) has served as a disincentive to voluntary
cleanup, even when it seems that the spiller would qualify for compen-
sation under § 311(1).
It has also been suggested that third parties who are
actually responsible for spills may benefit by § 311(i), because the
Innocent spiller will not attempt to recover from such a person, and
the government, who is subrogated to the spiller1s rights in such a
situation, may forego further action because of the time and resource
commitment that is often required in identifying the responsible party
and proving the case*
Finally, It has been suggested that encouragement of
voluntary cleanup may actually work against the goals of CERCLA.
Encouragement of such action might introduce costs that will burden
the hazardous substance response fund and interfere with the alloca-
tion of available resources to higher priority actions. These
objections, however, were limited to hazardous substance spills.
(8) Other Congressional Policies
No significant conflicts with other congressional policies
were identified.
3. Summary Assessment of § 311(1)
a. Is § 311(1) Appropriate and Needed?
Information directly supporting the appropriateness and need
for § 311(1) was obtained in this study. Representatives of industry
and the Federal agencies have indicated that § 311(1) actually serves
as encouragement to conduct voluntary cleanup after an oil spill and
to proceed with greater willingness to clean up when probable liabil-
ity has been established by the government. In addition, spokesper-
sons for the USCC indicated that the availability of § 311(1) makes it
easier to persuade potentially responsible persons to conduct cleanup,
without being certain that they are actually responsible, since they
will be eligible for an indemnity. Assignment of responsibility
without a complete Investigation is not often possible. Such an
option is particularly Important because the USCG may have limited
resources, or the USCG cleanup response time may be unreasonably long
when compared to the likelihood for environmental damage and the more
rapid response that may be undertaken by private parties who are
closer to the Incident. (A joint effort is often appropriate.)
-91-
-------
INDEMNIFICATION STUDY
Section III
Private cleanup may also be more cost effective than when
undertaken by the government because of private economic efficiency
incentives, as well as because of the possibility that continued delay
will allow for a spill to spread and make the cleanup task more
difficult.
Although some potentially negative impacts of § 311(1) have
been suggested, no such impacts were actually documented.
Whether § 311(i) is needed for hazardous substance spills is a
question that may not be effectively addressed for two reasons: (1)
there have been no claims under § 311(1) for such spills to date, and
(2) hazardous substance spills are covered under extensive new
authority In CERCLA, which may obviate further need for § 311(1).
b. Potential Problem Areas
(1) Administration in the Court of Claims
The oftentimes long delay In resolving a claim, coupled
with the adversarial nature of the proceeding, may discourage volun-
tary cleanup initiatives because of the actual difficulty In obtaining
reimbursement of costs. Alternatives have been suggested In H.R.
85.187
(2) Availability of Funds Under § 311(k)
The lack of funds under § 311(k) to finance oil spill
cleanup on a large scale (e.g., similar to the chemical spill cleanup
financed under CERCLA) may prevent complete cleanup and other
appropriate remedial activity.
(3) Conflicting and Overlapping Statutes
All four oil spill statutes contain different liability
standards, different financing requirements, different administrative
.requirements, and other Inconsistencies. Although perhaps justified
Individually when enacted, all together they produce a complex web
that Invites difficult and inconsistent implementation. These prob-
lems are heightened by CERCLA, which Imposes a fifth set of liability
and reimbursement rules for chemical spills.
(A) Voluntary Cleanup of Chemical Spills by Innocent Parties
CERCLA policies prevent large-scale financing of cleanup
efforts by dischargers, including those who are uncertain of their
liability. Section 311(1) has been used effectively In such circum-
stances to enable the USCC to persuade a discharger to clean up
without having to determine liability, then pay later, if it turns out
the discharger can assert a defense to liability. CERCLA policies may
-92-
-------
INDEMNIFICATION STUDY
Section III
not permit such an approach; and the agencies may be less willing to
order needed cleanup if there is reasonable doubt regarding the
discharger's liability.
On a more fundamental level, the limited, procurement
approach to cleanup that EPA has taken under CERCLA suggests that
there is little, if any, role in chemical spill cleanup for voluntary,
unsupervised efforts. This approach is strongly defended within EPA
and may in actuality be the better approach for chemical spills;
however, insufficient experience exists to make such a determination
at the present time.
A. The Applicability of § 311(1) to Other EPA Programs
a. Introduction
The following elements are Important to § 311(1): (1) an
unanticipated spill, discharge or release of a pollutant that is a
threat to the environment; (2) the possibility that effective Imme-
diate action may be taken by the discharger to contain or remedy the
problem after the spill or release has occurred; (3) greater cost
savings or environmental benefit if the discharger undertakes the con-
tainment or remedial action quickly; and (4) no other Incentive for
the discharger to take such action voluntarily.
If there exist other situations that Include the four
criteria, the question presented is whether It is reasonable to
attempt to apply a mechanism such as § 311(1). This question depends
on whether the situation is already being addressed by a different
mechanism, whether that mechanism appears to deal adequately with
situations of that type (or could be made to do so), and whether a
§ 311 model could more profitably be utilized.
. b. Evaluation of Other EPA Programs
Sudden, unanticipated pollution occurrences exist in all of
the pollution media subject to EPA jurisdiction. Many of these are
hazardous and require some type of immediate action. In testimony on
Superfund, for example, Deputy Administrator Barbara Blum pointed out
that as many as 1,800 spills a year are not regulated under § 311
because they go into groundwater, the air, or onto land. Moreover, as
many as 1,000 hazardous substance spills that reach navigable waters
were not subject to § 311 jurisdiction at the time because they
Involved substances not yet designated under § 311. °°
This gap in authority was one primary reason for the enactment
of CERCLA. Its provisions cover virtually every type of hazardous
substance release for which application of a § 311(1) approach might
be appropriate. These Include hazardous substances designated under
§ 31l(b)(2)(A) of the Clean Water Act, hazardous wastes listed under
-93-
-------
INDEMNIFICATION STUDY
Section III
§ 3001 of the Resource Conservation and Recovery Act, hazardous pollu-
tants listed under § 112 of the Clean Air Act, imminently hazardous
chemical substances or mixtures which have been acted on pursuant to
§ 7 of the Toxic Substances Control Act, and any other substance regu-
lated pursuant to § 102 of CERCLA. The most significant coverage,
however, Is provided under § 104. According to § 104(a)(l), whenever
such a substance, or any other pollutant or contaminant which may pre-
sent an Imminent and substantial danger to the public heath or welfare
Is released or there Is a substantial threat of release into the
environment, the President is authorized to act in a manner consistent
with the national contingency plan to remove the substance and provide
remedial action, unless he determines that removal and remedial action
will be done properly by a responsible party.
Under § 107, the discharger Is liable for removal and remedial
costs unless he can prove that the release and damages were caused
solely by an act of God, an act of war, or an act or omission of a
third party. In such a case, the statute is silent on the dischar-
ger's prerogative to remove or conduct remedial action voluntarily.
However, § lll(a) provides that the President may pay any claim for
necessary response costs Incurred by any person as a result of
carrying out the national contingency plan so long as such costs are
approved under the plan and certified by the responsible federal
official. Under this provision, EPA has Indicated that It will
require prior approval of private response efforts,"•"* and will
strictly limit the extent to which CERCLA funds may be spent In that
regard."0
The reasons for EPA's position Include concerns over safety
and the chance of error (Inherent differences In the nature of oil and
chemical releases are noted), but perhaps the most important concerns
relate to the potential costs Involved if private parties may conduct
cleanup actions that have not been preauthorlzed, and the potential
Interference with EPA fund allocation priorities. In effect, EPA
feels that the risks of Increased environmental damage associated
with the rejection of a § 311(1) approach under CERCLA are outweighed
by the need to ensure safety, effective cleanup management, and
control over fund allocations.
At this stage, CERCLA implementation has not proceeded far
enough to evaluate the successfulness of CERCLA policies as an alter-
native to § 311(1).191 Until such time, it would be Inappropriate to
recommend the extension of § 311(1) to other EPA programs, without an
indication of specific circumstances in which cleanup efforts are
falling because there is no voluntary action by nonliable parties. No
such Information surfaced during this study. Therefore, no recommen-
dation for extending § 311(1) to other EPA programs has been made.
However, in view of the success of § 311(1) in the area of oil spills,
it is recommended that CERCLA policies on this Issue be reconsidered
after sufficient implementation experience has been gained.
-94-
-------
INDEMNIFICATION STUDY
Section III
FOOTNOTES TO SECTION III
following description and analysis draws on an extensive
series of interviews with Involved parties at EPA, the Department of
Justice, industry and congressional staff.
2P.L. 92-516, 86 Stat. 973.
2a444 U.S. 51, 100 S. Ct. 318, 326 (1979) (citations omitted).
•'The Federal Insecticide, Fungicide, and Rodenticlde Act of 1947,
P.L. 80-104, 61 Stat. 163, repealed and replaced the Insecticide Act
of 1910, P.L. 61-152, 36 Stat. 331. Amendments to FIFRA were enacted
in 1964 as P.L. 88-305, 78 Scat. 190.
4P.L. 94-140, P.L. 95-396, and P.L. 96-539.
^Allowing farmers to use their stocks is sometimes spoken of as
"disposal by use." Spreading the pesticide out all over the country,
rather than transporting and then disposing of it all in one place,
can sometimes reduce risks to human health and the environment.
"See Conner, "Federal Indemnification for Losses Resulting from
the Suspension of Hazardous Products — the Lessons of FIFRA," 32
A.B.A. Admin. L. Rev. 441 (1980).
'Appendix A, page A-20.
"Appendix A, page A-19.
'Appendix A, page A-19.
l°Appendlx A, page A-18.
HImminent hazard Is not a statutory basis for cancelling a pesti-
cide registration; it Is the basis for suspension. The language of
.§ 15 appears to be a drafting error.
* This provision was Included by the Conference Committee to
assure that Indemnification would be denied to persons who had not
produced a pesticide in good faith. However, the language goes beyond
Inquiry Into the claimant's knowledge of the safety risks posed by the
pesticide. It seems to require an investigation Into whether EPA
would conclude that the risks are outweighed by the benfits of use,
because that is the standard for registration. As a practical matter,
It would be difficult to deny Indemnification to any producer under
this standard because the evidentiary requirements are too difficult.
Thus, It Is probably ineffective for its Intended purpose of weeding
out those claiming indemnification in bad faith.
-95-
-------
INDEMNIFICATION STUDY
Section III
Supra note 3 .
*• The only public records of these meetings are three revised cora
ralttee prints of H.R. 4152. Committee Print No. 2, dated June 22,
1971, contained a very brief indemnity provision, while Committee
Print No. 3, dated July 13, 1971, contained a more detailed indemnifi
cation provision that eventually passed the House as H.R. 10729 in
November 1971.
e.g-, testimony of American Farm Bureau Federation, Federal
Pesticide Control Act of 1971; Hearings before the House Comm. on
Agriculture, 92d Cong., 1st Sess. 462 (1971); Dow Chemical Company,
id. at 302; Union Carbide Corporation, id. at 332, 349-50; Sierra
Club, Federal Pesticide Control Act 1972: Hearings Before the
Subcomm. on Agriculture Research and General Legislation of the
Senate Comm. on Agriculture and Forestry, 92d Cong., 2d Sess. 123, 127
(Part II) (Mar. 7-8, 1972); Environmental Defense Fund, id. at 156,
157; and the Public Interest Research Group, id. at 354-55. It was
during the second day (February 23, 1971) of these hearings that the
indemnification Issue was raised in an exchange between Representative
Coodling (R.-Pa.) and J.P. Campbell, Undersecretary of Agriculture.
House Hearings at 67-68.
16Coramittee of Conference, S. Rep. No. 92-1540, 92d Cong., 2d
Sess. (1972).
1'The agreement was signed on May 7, 1979 after negotiations with
EPA's Office ot General Counsel and Chevron Chemical, the largest
manufacturer of Silvex products.
18See 39 Fed. Reg. 14753 (Apr. 26, 1974) (emergency suspension),
EPA PR Notice 74-5 (Apr. 30, 1974), and 40 Fed. Reg. 3494 (Jan. 22,
1975) (cancellation).
19For Aldrin/Dieldrin see 39 Fed. Reg. 37265-72 (Oct. 18, 1974)
(suspension), Environmental Defense Fund v. EPA, 510 F.2d 1292 (1975)
(suspension upheld but use of special rule ordered to be reconsid-
ered), and 40 Fed. Reg. 24232 (June 5, 1975) (status of existing
stocks). For Heptachlor/Chlordane, see 39 Fed. Reg. 41298 (Nov. 26,
1974) (intent to cancel), 41 Fed. Reg. 7552 (Feb. 19, 1976) (intent to
suspend), 548 F.2d 998 (1976) (suspension upheld but special rule
ordered to be reconsidered), 43 Fed. Reg. 12372 (Mar. 4, 1978)
(cancellation allowing phase-out).
20Interview on November 5, 1981 with Ralph Colleli, formerly EPA
Office of Pesticide Programs.
^Administrative hearing suspended for settlement negotiations.
22House Hearings, supra note 15, at 349-50 (Mr. Wellman of Union
Carbide).
-96-
-------
INDEMNIFICATION STUDY
Section III
^Section 15 does not specify the source of funds for paying
indemnification claims. In response to a question from Senator
Curtis, David D. Dominick, Assistant Administrator for Categorical
Programs of EPA, stated that EPA would "provide the Indemnification"
under § 15. There was no discussion of the matter. Senate Hearings,
supra note 15, at 93. Anita Johnson of the Public Interest Research
Group argued in her testimony before the Senate Subcommittee that,
"the pending bill Is abhorrent because it requires indemnities to be
paid from EPA funds." Id. at 354-55.
2444 Fed. Reg. 15874 (Mar. 15, 1979)
^Telephone conversation on March 11, 1982 with Ray Krueger, EPA
Office of Pesticides, Hazard Evaluation. As of March 11, 1982, four
million containers of liquid Silvex were still at the collection
points established by the registrants. The disposal process will
Involve shipment of the containers of Silvex to the same waste facil-
ity in Alabama that was used for the solid pesticide. EPA now has a
contract with the Alabama facility for disposal of the liquid. The
containers will be opened and the liquid removed and incinerated, but
it is estimated that this will not occur prior to September 1982.
26117 Cong. Rec. 40047 (1971) (e.g., remarks of Rep. Teague).
27Interview on October 21, 1981 with Herb Harrison, Chief of
Insecticide/Rodenticide Registration Branch.
28Since 1972 EPA has only denied registration for five active
ingredients (used in 216 applications). Four of the five had already
been cancelled when application for the new registration was made:
DDT, sodium cyanide, sodium fluoroacetate, and heptachlor/chlordane.
The fifth was a petition to use sodium arsenate in bottle tops.
^Telephone conversation on May 19, 1982 with Edward Gray, EPA
Acting Associate General Counsel for Pesticides. (The Acting
Associate General Counsel for Pesticides has said that this is so
cleax to all Involved that he has never heard an argument to the
contrary).
pesticide Industry Is so competitive that substitutes are
constantly under study, even in the absence of regulation. The
National Research Council study Regulating Pesticides estimates the
economic life of a pesticide to be only about 10 years.
example, the total impact of the 2,4, 5T/Sllvex suspensions
was estimated to be $89 million (nearly all attributed to the 2,4,5T
uses). The Economics Analysis Branch, Office of Pesticide Programs,
EPA Pesticide Cancellations/Suspensions: A Survey of Economic Impact,
March 1980.
-97-
-------
INDEMNIFICATION STUDY
Section III
32Supra note 26 at 40053 on October 21, 1971 (Rep. Abernathy).
33Id. at 40052 (Rep. Poage).
•^Appendix A, page A-7. See, e.g. , Mugler v. Kansas, 123 U.S. 623
(1887) (elimination of a nuisance).
e.g. , United States v. Pennsylvania Industrial Chemical
Corp.. 329 F. Supp. 1118 (W.D. Pa. 1971), rev'd & remanded. 461 F.2d
468 (3d Cir. 1972), modified, 411 U.S. 655 (1973) (discharge of
industrial waste without permit); and United States v. Articles of
Drug. . . Hormonin, 498 F. Supp. 424 (D.N.J. 1980) ("new drug"
required registration despite previous FDA letter Indicating other-
wise). See also, Thompson, "Equitable Estoppel of the Government," 79
Colum. L. Rev. 551 (1979); Apnot, "Estoppel Against Government," 27
A.L.R. Fed. 702 (1976).
36Supra note 26 at 40049-50 (Rep. Goodling and Rep. Kyi).
37House Hearings, supra note 15 at 349-50 (Mr. Wellman of Union
Carbide Corp.).
•^Appendix E, page E-ll et^ seq.
39Id. at 11.
^"Product Regulation and Chemical Innovation," (The Conservation
Foundation, March 1980), Interview with Dr. Arnold Aspelln, Chief
Economic Analysis Branch, Office of Pesticide Programs.
42117 Cong. Rec. 40052 (1971) (Rep. Andrews and Rep. Poage).
3Four of the five (used in 216 applications) had already been
cancelled when application for the new registration was made: DDT,
sodium cyanide, sodium fluoroacetate, and heptachlor/chlordane. The
fifth was a petition to use sodium arsenate in bottle tops.
44Supra note 42 at 40052-53 (Rep. Bergland).
4 5ld± at 40046 (Rep. Evans).
46Id. at 40050 (Rep. Obey).
The Trts case is the only product ban Indemnf ication situation
in which Congress has supported compensation, and that action was
vetoed by the President. The Senate has now passed a new bill. See
Appendix A, page A-22.
48Appendlx E, page E-5.
-98-
-------
INDEMNIFICATION STUDY
Section III
^Section 5 of the Toxic Substance Control Act requires manufac-
turers of chemicals to notify the Agency 90 days before beginning com-
mercial production of a new chemical. If the Agency has concerns
about a new chemical, It can order the manufacturer to do specific
tests though this happens rarely.
Section A established an Interagency Testing Committee that recom-
mends to the Administrator which existing chemicals should be given
priority consideration for testing. The Administrator has 12 months
to respond to the recommendations by inflating a rulemaking or
publishing in the Federal Register a reason why It will not be
initiated.
Section 8 contains authority for several kinds of recordkeeplng
and reporting requirements. The most important is the Inventory
established by 8(b). Until this section was Implemented, there had
been no comprehensive knowledge of what chemicals were being produced
in the United States. It now provides the baseline for determining
what chemicals are "new." In addition, there is a mandatory duty to
report to the Administrator "any information which reasonably supports
the conclusion that such substance or mixture presents a substantial
risk of Injury to health or the environment" under § 8(e).
Sections 6 and 7 provide authority to regulate existing chemicals
when there is a reasonable basis to conclude that they provide an
unreasonable risk of Injury to human health or the environment. The
Administrator is required to chose the least burdensome requirement
that will adequately protect against the risk. The economic con-
sequences of the regulation are to be taken Into account, as are the
possibilities of regulating under another statute. In fact, § 9
requires the Administrator to use other EPA authorities to protect
against the risk unless it is in public interest to use TSCA. If the
risk may be reduced to a sufficient extent by action taken under a
Federal law administered by another agency, the Administrator is
required to give the other agency an opportunity to act, before taking
action under TSCA.
50TSCA § 6(d)(2) provides for immediately effective proposed
rules, but If the rule proposes a total ban, it must be proceeded by a
court order under § 7. Thus, the only way to immediately ban a chemi-
cal under TSCA is to go to court.
^Appendix A, at page A-22.
page 134.
5333 U.S.C. §§ 1251 et seq.
97-117 (Dec. 29, 1981).
-99-
-------
INDEMNIFICATION STUDY
Section III
technology Is defined, In part, In terras of its ability to
produce:
• reduced life-cycle costs of treatment facilities;
• reduced primary energy requirements for treatment; or
• improved environmental quality
as compared to conventional technology* U.S. Environmental Protection
Agency, Innovative and Alternative Assessment Manual, (EPA
430/9-78-009] (1980).
56The Clean Water Act § 202(a)(l), as amended by P.L. 97-117, con-
tains the new 55% limitation.
^information regarding the intent of the Conference Committee in
including this section is found in the House debates on the Conference
Report (the Senate debates do not mention § 202(a)(3)). Rep. Roberts
explained the Committee's action:
In recognition of the risks involved In
encouraging this departure from buslness-as-
usual [as evidenced by the provisions encour-
aging development and deployment of innovative
and alternative technology], the conferees
adopted another amendment authorizing grants
to cover the total cost of modifying or even
replacing systems that do not perform as
Intended and result in capital or operating
and maintenance costs significantly higher
than anticipated. This, in effect, constitu-
tes an insurance policy.
Innovative technology, as referred to in
the conference report, means new and promising
technology which has not been fully proven
under the circumstances of this contemplated
use. ... It would fall between the extremes
of unproven technology, and the old conven-
tional methods which have been duplicated and
repeated endlessly in this country. We think
the key lies in the recognition and acceptance
of a certain element of risk, which Is why we
provide for the Federal Government to assume
100 percent of the costs of modifying or
replacing systems which fall to perform as
intended* These risks are deemed acceptable
in light of the potential benefits in terms of
environmental enhancement, lower capital or
-100-
-------
INDEMNIFICATION STUDY
Section III
maintenance costs, and reclamation, recycling
and use of water which Is becoming
increasingly costly to provide.
3 A Legislative History of the Clean Water Act of 1977, A Continuation
of the Federal Water Pollution Control Act 299, (emphasis added 307-08
(1978) (hereinafter cited as "CWA Legislative History"). Later in the
debate, Rep. Cleveland emphasized the insurance nature of the
provision:
However, because we encourage Innovation
and because we recognize there is risk, new
subsection 202(a)(3) provides for payment of
all of the costs of modification or replace-
ment of facilities which do not meet design
performance specification. This provision, an
Insurance policy, recognizes there will be
risks where technology is pushed ahead.
3 CWA Legislative History 299, 385 (1978).
58*0 C.F.R. § 35.2032(c), 47 Fed. Reg. 20462 (May 12, 1982).
^Telephone conversation on December 7, 1982 with Richard Thomas,
EPA Office of Water Program Operations, Municipal Technology Branch.
60Section 202(a)(3) was not present in either S. 1952 or H.R. 3199
as originally passed by each House of Congress. The indemnification
provision enacted into law was added by the Conference Committee:
Additionally, the Administrator is authorized
to make a grant to fund all costs of modifica-
tion or replacement of facilities constructed
with such a grant if they fail to meet design
performance specifications, unless this
failure is attributable to negligence, and has
significantly increased capital or operating
and maintenance expenditures.
H.R. Rep. No. 95-830 at 55, 95th Cong., 1st Sess., reprinted in, 3 CWA
Legislative History 185, 239 (1978).
61-Supra note 57.
^Concern for the encouragement of Innovation was very evident in
the Senate Report on S. 3199:
More than any other issue concerning the
construction grant program the committee
hearings focused on the need to encourage
-101-
-------
INDEMNIFICATION STUDY
Section III
alternative and innovative systems. The prob-
lems of small communities coping with expen-
sive capital-intensive waste treatment systems
and the wastefulness of discharging valuable
nutrient resources to the Nation's waters were
stressed throughout the country* The need for
new Industrial processes which produce no
waste was emphasized*
S. Rep. No. 95-370 at 5, 95th Cong., 1st Sess., reprinted in, 4 CWA
Legislative History.
note 57.
6ASupra note 58.
65Supra note 54.
6643 Fed. Reg. 44049, 40 C.F.R. § 35.908(c). The proposed new
regulations were published November 6, 1981. 45 Fed. Reg. 55220, pro-
posed 40 C.F.R. § 35.2032. The 1978 regulations provided:
(c) Modification or replacement of innovative
and alternative projects. The Regional Administra-
tor may award grant assistance to fund 100 percent
of the eligible costs of the modification or
replacement of any treatment works constructed with
85-percent grant assistance If:
(1) He determines that:
(1) The facilities have not met design perfor-
mance specifications (unless such failure is due to
any person's negligence);
(ii) Correction of the failure requires signi-
ficantly increased capital or operating and main-
tenance expenditures; and
(lii) Such failure has occurred within the
2-year period following final inspection; and
(2) The replacement or modification project is
on the fundable portion of the State's priority
list.
^Figures provided by Richard Thomas, supra note 59.
n FY 1980, the Construction Grants appropriation was $3.4
billion.
-102-
-------
INDEMNIFICATION STUDY
Section III
69Supra note 54.
7(^In addition, there may be cost savings relative to conventional
technologies in the operation or maintenance of the plant.
7233 U.S.C §§ 1254 and 1255.
7 telephone conversation on March 23, 1982 with William Cawley,
Deputy Director, U.S. Environmental Protection Agency, Industrial
Environmental Research Laboratory.
U.S.C. § 1254 provides the authorization of the grants.
77Telephone conversation on January 25, 1982 with Mike Maxwell,
Chief, Emission and Effluent Technology Branch, U.S Environmental
Protection Agency, Industrial Environmental Research Laboratory.
^Telephone conversation on February 1, 1982 with Malcolm P.
Huneycutt, Contracting Officer, U.S. Environmental Protection Agency,
Contract and Management Division (Research Triangle Park).
79The mechanism providing for this insurance Is found in Clause 42
of the standard EPA contract which provides:
(a) The Contractor shall procure and maintain such
Insurance as is required by law or regulation,
including that required by Subpart 1-10.5 of the
Federal Procurement Regulations as of the date of
execution of this contract, and such insurance as
the Contracting Officer prescribes by written
direction.
(b) At a minimum, the Contractor shall procure and
maintain the following types and amounts of insur-
ance:
(1) Workmen's compensation and occupational
disease insurance in amounts sufficient to
satisfy State law;
(2) Employer's liability insurance, where
available;
(3) Public liability insurance, on the com-
prehensive form of policy, in the amount of
-103-
-------
INDEMNIFICATION STUDY
Section III
$200,000 per claimant and $500,000 per
incident;
(4) When aircraft are used In the performance
of the contract, aircraft public and passenger
liability Insurance, in such form, in such
amounts, and for such periods of time as the
Contracting Officer may require or approve;
(5) When vessels are used in the performance
of the contract, vessel collision liability
and protection and Indemnity liability Insur-
ance, in such form, in such amounts, and for
such periods of time as the Contracting
Officer may require or approve.
(c) With respect to any insurance policy all or
part of the premiums of which the Contractor
proposes to treat as a direct cost under this
contract, and with respect to any proposed
qualified program of self-Insurance, the written
approval of the Contracting Officer shall be
obtained prior to any claim for payment therefor.
The Contractor shall be reimbursed for the protion
allocable to this contract.
(d) The term of any other insurance policy held by
the Contractor shall be submitted to the Contract-
Ing Officer for review and/or approval upon request
of the Contracting Officer.
(Emphasis added.)
8°A2 U.S.C. § 7413.
81Federal clean air legislation dates back to 1955. See Act of
July 14, 1955, 69 Stat. 322. Numerous amendments have occurred since.
The most significant, however, were the 1970 and 1977 amendments.
Pub. L. 91-604, 84 Stat. 1676 (December 31, 1970); and Pub- L. 95-95,
91 Stat. 712 (Nov. 16, 1977). See, e.g., "The Clean Air Act Amendment
of 1977: A Selective Legislative Analysis" 13 Land & Water L. Rev.
747 (1978); "Enforcement and Litigation Under the Clean Air Act
Amendments of 1977," 12 Natural Resources Lawyer 435 (1979); "Clean
Air Act Amendments of 1977," 19 Natural Resources J. 475 (1979).
e.g., "Enforcement and Litigation under the Clean Air Act
Amendments of 1977," 12 Natural Resources Lawyer 435, 482-87 (1979).
83As first offered by Representative Satterfield (D. Va.), it pro-
vided that a court could make an award whenever it found the action
-104-
-------
INDEMNIFICATION STUDY
Section III
"unwarranted" or determined that such an award was "appropriate."
Although Satterf ield' s amendment was rejected, the committee passed a
similar amendment (and one identical to the provision ultimately
enacted) offered by Representative Eckhardt (D. Tex.)i and amended by
Representative Dlngell (D. Mich.), which restricted the award to
actions found to be "unreasonable." A separate amendment offered by
Representative Frey (R. Fla.) to provide an award only to a
"prevailing" party was also rejected in Committee. Minutes of the
House Committee on Interstate and Foreign Commerce (Full Committee
Mark-up Session, Clean Air Act Amendments of 1976), 94th Cong., 2d
Sess. (March 19, 1976).
84H.R. Rep. No. 94-1175 at 277, 94th Cong.,_2d Sess., reprinted In
7 CAA Legislative History 6547, 6826 (1978).
o
85S. Rep. No. 95-127 at 99-100, 95th Cong., 1st Sess., reprinted
^n 3 CAA Legislative History 1371, 1473-74 (1978).
the amendment to § 113(b) was under consideration, a
separate, parallel provision involving suits against the Administrator
was also under consideration. The 1970 Amendments had provided In
§ 304(d) for an award of attorneys' fees In citizen suits, whenever
the court determined that such an award was "appropriate." The pur-
pose of this provision was both to encourage legitimate public Inter-
est suits (against industry or the agency) and to discourage harassing
suits by allowing for costs to be imposed against the plaintiff in
such circumstances. S. Rep. No. 91-1196 at 38, 91st Cong., 2d Sess.,
reprinted in 1 CAA Legislative History 397, 438 (1974).
Prior to the 1977 Amendments interest was expressed by industry
and environmental organizations in extending this provision to suits
brought for the purpose of reviewing agency rulemaking, and other
similar actions, under § 307 of the Act. As with § 304, it was
argued, such suits would serve to promote the public Interest by
serving as a check against unreasonable rulemaking. This ultimately
resulted in the enactment of § 307(f) which allows for a similar
award, again whenever a court determines that it is "appropriate;" and
the stated purposes are essentially the same as in § 304: "to prevent
harassment of innocent parties," and to "reimburse citizens who seek
to enforce the law." S. Rep. No. 94-717 at 82-3, 94th Cong., 2d Sess.,
reprinted in 6 CAA Legislative History 4701, 4783-84 (1978).
87See Alyeska Pipeline Co., v. Wilderness Society, 421 U.S. 240
(1975); "Awards of Attorneys' Fees by the Federal Courts (Part I)," 6
ALI-ABA Course Materials J. 95 (1982).
e.g. E. Larson, Federal Court Awards of Attorney's Fees
257, n. 5 and Appendix C (1981).
-105-
-------
INDEMNIFICATION STUDY
Section III
9°See, e.g., Clean Water Act § 505(d); Toxic Substances Control
Act § 20; Safe Drinking Water Act § 1449(d); Noise Control Act
§ 12(d); Resource Conservation and Recovery Act § 7002(e). The
Federal Insecticide, Fungicide, and Rodenticlde Act; and the Compre-
hensive Environmental Response, Compensation, and Liability Act of
1980 do not contain a similar provision.
91The "Equal Access to Justice Act" appears at Title II of the
"Small Business Export Expansion Act of 1980." Pub. L. 96-481; 94
Stat. 2321. Its provisions are codified at 5 U.S.C. § 504 and 28
U.S.C. § 2412. The effective date was October 1, 1981.
92H.R. Rep. No. 96-1418, 96th Cong., 2d Sess., reprinted in 1980
U.S. Code Cong. & Ad. News 4984-89.
93See H.R. Rep. No. 94-1175 at 277; Minutes of the House Committee
on Interstate and Foreign Commerce, March 18, 1976; S. Rep. No.
94-717 at 82-3; Transcripts of the Senate Committee on Environment and
Public Works, March 16, 1977. See Sierra Club v. Gorsuch. 16 E.R.C.
2113, 2120 (D.C. Clr. 1982) (Sierra Club awarded fees but did not pre-
vail under § 307(f) of the Clean Air Act).
94
See discussion of S 113(b), supra at 86.
discussion of cases on point in 6 AL1-ABA Course Materials
J_._ 95 (1982). This policy is also provided in Canon 7-4 of the Code
of Professional Responsibilities for Attorneys:
The advocate may urge any permissible con-
struction of law favorable to his client . . .
his conduct is within the bounds of law, and
therefore permissible, if the position taken
is supported by any good faith argument for an
extension, modification, or reversal of the
law. However, a lawyer is not justified in
asserting a frivolous claim.
9630 U.S.C. § 425(e).
97 See 43 C.F.R. § 4.1294; Colloquy between Seiberllng and Udall,
_ Cong. Rec. H3826, April 29, 1977; H. R. Rep. No. 95-218 at 90, 95th
Cong., 1st Sess.
98Interview on October 13, 1981 with Mark Squlllace, Office of the
Solicitor, Department of Interior.
9943 C.F.R. §§ 4.1294(c) and (d). See also 43 Fed. Reg. 34385-86
(Aug. 3, 1978); H.R. Rep. No. 95-218 at 311, 95th Cong. 1st Sess.
100Supra note 91.
-106-
-------
INDEMNIFICATION STUDY
Section III
101H.R. Rep. No. 96-1418, 96th Cong., 2d Sess., reprinted in 1980
!U..S. Code Cong. & Ad. News 4984, 4994.
rejected standard had been offered by the Department of
Justice as a compromise to the standard ultimately enacted* Depart-
ment opposition had been particularly strong. In a statement before
the House Committee on Small Business (May 1, 1980), Alice Daniel,
Assistant Attorney General, Civil Division, argued that such legisla-
tion could cost taxpayers well over 100 million dollars a year, that
it would be overreaching in its disruption of government enforcement,
that it would be more efficient to eliminate bad regulations and
improve the regulatory process than to establish disincentives to
enforcement; and that establishing a presumption against the reason-
ableness of a regulatory action departed from a long valid tradition
without justifiable cause.
In its report on H.R. 6429 (S. 265), the House Committee on Small
Business indicated that placing the burden on a prevailing party to
challenge the reasonableness of the underlying agency action would be
deterrent to small business, and would thwart Congress1 interest in
ensuring that such rights be vindicated. The Report goes on to note:
The standard and burden of proof in S. 265
represents an acceptable middle ground between
an automatic award of fees and the restric-
tive standard proposed by the Department of
Justice. It presses the agency to address the
problem of abusive and harassing regulatory
practices. It is Intended to caution agencies
to carefully evaluate their case and not to
pursue those which are weak or tenuous. At
the same time, the language of the section
protects the government when its case, though
not prevailing, has a reasonable basis in law
and fact. Furthermore, it provides a safety
valve where unusual circumstances dictate that
the government is advancing in good faith a
credible, though novel, rule of law.
H.R. Rep. No. 96-1418 at 13-14, 96th Cong. 2d Sess., reprinted in
1980 U.S. Code Cong. & Ad. News 4984, 4993.
103Interview on January 27, 1982 with Don Stevers, Department of
Justice.
involving the Apache Powder case were derived primarily in
the following interviews: interview with Barbara Brandon, former
Attorney, Department of Justice, January, 1982; interview with Craig
-107-
-------
INDEMNIFICATION STUDY
Section III
Litman, Attorney, EPA Region IX, January, 1982; Interview with Amy
Coy, Counsel for Apache Powder from Evans, Kitchel and Jenckes,
Phoenix, Arizona, Fehruary 3, 1982.
lo&Letter to Marvin B. Durnlng, Esquire, Assistant Administrator
for Enforcement, EPA, from James W. Moorman, Assistant Attorney
General, Lands and Natural Resources Division, Department of Justice
(May 26, 1978); "Region IX Position on the Confession of Error by the
Department of Justice In U.S. v. Apache Powder Co., Civ. No. 78-058,
In the United States District Court for the District of Arizona,"
unpublished EPA document.
__,>ra note 103.
""interview with Sharon Green, General Accounting Office, January
27, 1982. The Judgment Fund utilized for § 113(b) is the same fund
from which Silvex indemnification claims under § 15 of the Federal
Insecticide, Fungicide, and Rodenticlde Act have been paid. See
discussion supra at page 33.
Coy supra note 105.
^-Barbara Brandon supra note 105. A recent estimate of the
average award for judicial proceedings generally, excluding tax cases,
was approximately $16,000 for attorney fees and $2,000 for other
costs. Congressional Budget Office, "Cost Estimate for S.265, the
Equal Access to Justice Act," July 18, 1979, reprinted in Sen. Rep.
No. 96-253, 96th Cong., 1st Sess., at 10-12.
See Environmental Protection Agency, "Justification of
Appropriation Estimates for Committee on Appropriation, Fiscal Year
1983" A-106-117.
Coy supra note 105.
ll5Intervlew with Edward Reich, Director, Division of Stationary
Source Enforcement, Environmental Protection Agency on September 22,
1981.
116Amy Coy supra note 105.
internal concurrence requirements for all judicial actions
are rigorous. See "Draft Memorandum from EPA Administrator Anne M.
Gorsuch on General Operating Procedures for Civil Enforcement
Program," reprinted in BNA Environmental Reporter, Current Develop-
ments (May 21, 1982) at 78-85. In addition, Department of Justice
-108-
-------
INDEMNIFICATION STUDY
Section III
review is even more demanding. Interviews with Jeffrey Miller of
Bergson, Borkland, Margolis and Adler (Former Acting Assistant
Administrator for Enforcement, Environmental Protection Agency)
(December, 1981) and Angus Mac Beth of Bergson, Borkland, Margolis and
Adler (Former Assistant Attorney General, Lands and Natural Resources
Division, Department of Justice (December, 1981). See also "Memo-
randum of Understanding on Civil Enforcement Between the Justice
Department and the Environmental Protection Agency" (June 13, 1977),
reprinted in BNA Environment Reporter, Federal Laws 41:2401;
Memorandum to Regional Administrators from Kathleen Bennett (December
29, 1981) (emphasizes that enforcement should be focused on signifi-
cant violators that have been pre-selected with state assistance and
that states should be encouraged to take the lead).
118Infra Section IV, notes 73-76. Most complaints related to the
underlying regulations and EPA interpretations of congressional
policy. However, Attorney George Freeman of Hunton and Williams, a
Richmond, Virginia law firm that actively represents industrial and
electric power facilities In proceedings involving EPA regulations,
raised several complaints relating to EPA negotiating policies and the
use of civil penalty authority. Senate Oversight Hearings, June 4,
1981, part 2, No. 97-H12, 97th Cong., 1st Sess. at 479-482.
supra note 115; Miller supra note 117.
Section IV, notes 73-76. Other sources consulted include
BNA Environment Reporter, Cases (vols. 14-16 through May 21, 1982),
Current Developments (May 1, 1979 - May 21, 1982).
121Supra note 117.
122Id.
123Supra notes 112 and 117.
Reich supra note 115. Miller supra note 117. Interview with
Richard Wilson, Environmental Protection Agency, September, 1981.
provision in the Senate version of the "Equal Access to
Justice Act" would have required payment from previously appropriated
funds and would not have allowed an appropriation specifically for the
purpose of an attorney fees award. See S. 265 §§ 3(d) and 4(d)(4),
96th Cong. 1st Sess. However, this requirement was not Included in
final legislation.
discussion supra at 67.
12733 U.S.C. § 1321.
up of hazardous substance spills, but not oil, is now
covered primarily by the Comprehensive Environmental Response, Compen-
sation and Liability Act of 1980. kl U.S.C. §§ 9601 et seq.
-109-
-------
INDEMNIFICATION STUDY
Section III
129Executive Orders 11735 (August 3, 1973) and 12316 (August 14,
1981) delegate presidential authorities and responsibilities for
carrying out the provisions of § 311 and CERCLA. Executive Order
12316, § 1, provides that the National Response Team under the
National Contingency Plan will Include representatives of the
Department of State, Department of Defense, Department of Justice,
Department of the Interior, Department of Agriculture, Department of
Labor, Department of Health and Human Services, Department of Energy,
Environmental Protection Agency, Federal Emergency Management Agency
and United States Coast Guard. Other specific agencies with important
roles under § 311 include the Public Health Service and the Federal
Maritime Commission.
§ 311 defenses are very strictly construed and apply only
if the discharger has had no part at all in causing the discharge.
See U.S. v. Bear Marine Services. 16 E.R.C. 1953, 1956, n. 3 (E.D. La.
1980); Sabine Towing & Transportation Co., Inc. v. United States, 16
E.R.C. 2081 (U.S. Ct. Cl. 1981); United States v. LeBeouf Brothers
Touring Co.. 621 F.2d 787, rehearing den.. 629 F.2d 1350 (5th Cir.
1980), cert, den.. 452 U.S. 906 (1981).
13133 U'S.C. S 1321(1) (Supp. 1981).
Report of the Senate Committee on Public Works accompanying
S.7, No7~9~l-351 (August 7, 1969) 91st Cong., 1st Sess., at 17-19.
Note, that the Court of Claims recently rejected an argument that the
remedial purpose of § 311(i) required a permissive reading of an
exception listed in § 3ll(i)(l). Sabine Towing & Transportation Co.,
Inc. v. United States, supra at 2084.
*33See discussion infra at 90. It has also been argued that cost
savings in cleanup is an Important factor. Early response may avoid
the cost of more extensive cleanup. But also, it is a common percep-
tion in the private sector that a profit oriented enterprise will
attain more effective cost control. In addition, federal procurement
requirements may result In additional costs when the USCG or EPA
undertakes cleanup. On the other hand, it is possible that § 311(1)
would encourage more extensive cleanup costs by a private party than
the government would be willing to incur. See discussion infra at 91.
13;>P.L. 91-224, 84 Stat. 91. See Sabine Towing & Transportation
Co., Inc. v. United States, supra at 2082.
135P.L. 92-500, 86 Stat. 862.
. 95-576. Regulations listing hazardous substances were
promulgated on March 13, 1978 at 43 Fed. Reg. 10479 (40 C.F.R. § 116);
regulations establishing reportable quantities were promulgated August
29, 1979 at 44 Fed. Reg. 50776 (40 C.F.R. § 117).
-110-
-------
INDEMNIFICATION STUDY
Section III
13742 U-S'C. §§ 9601 et seq.; P.L. 96-510; 94 Stat. 2767.
*™CERCLA does noc necessarily preempt the regulation of hazardous
substances under § 311. See CERCLA § 304(c). Except for the transfer
of funds provided in CERCLA § 304(b), EPA and the USCG have agreed to
utilize the Hazardous Substance Response Fund under CERCLA rather than
the § 311(k) fund for hazardous substance response because of the
additional funds available under CERCLA. See EPA-USCG "Reimburseable
Agreement for Funds Expended by 311(k) Pollution Fund on Behalf of
Section 104 of CERCLA," signed by Michael B. Cook, EPA (April 24,
1981), and by W.E. Caldwell, USCG (April 20, 1981). Nevertheless, it
is arguable that claims under § 311(1) may still be made for the 297
designated substances under § 311. See infra, note 170. .
139S. 7, § 12(j). See 1970 U.S. Code Cong. & Admin. News at 2720.
at 2726.
model appears in an oil Industry sponsored non-statutory
solution to public concerns involving the increasing frequency of oil
spills. Tanker owners devised an international voluntary plan
entitled "Tanker Owners Voluntary Agreement Concerning Liability for
Oil Pollution (TOVALOP)." Under this plan, participating tanker owners
would agree to reimburse the government for cleanup costs up to a
stated maximum in the case of negligent spills and would expect reim-
bursement for cleanup expenses for non-negligent spills.
Another model exists In the 1969 Convention of Civil Liability for
Oil Pollution Damage, Article V, 8, reprinted in BNA: International
Environment Reporter, Reference File, Vol. 1, 21:1501. Although an
individual ship owner Is responsible for constituting the fund under
this convention, Article V, 8 treats voluntary cleanup expenses
Incurred by the owner with equal rank to other claims against the
fund. (As of July 1980 the United States was not a party to the
Convention.) See also, the 1971 Convention on the Establishment of an
International Fund for Compensation for Oil Pollution Damage,
reprinted in BNA: International Environment Reporter, Reference File,
Vo. 1, 21:1701, which established an oil spill damage compensation
fund financed by contributions of member nations. Article 4 provides
indemnification for voluntary cleanup in certain circumstances. (As
of September 1977 the United States was a party to this Convention.)
for the provision Is mentioned only once — by the
American Waterways Operators, Inc.
143S_ee_, for example, Hearing on H.R. 85, No. 96-26, 96th Cong. 1st
Sess. at 194 (testimony of Swep Davis) September 26, 1979; Committee
Report on H.R. No. 96-172, 96th Cong. 1st Sess., reprinted in 1980
U.S. Code Cong. & Admin. News at 6184-6186 (legislative history to
CERCLA); Regulations on Determination of Reportable Quantities for
-111-
-------
INDEMNIFICATION STUDY
Section III
Hazardous Substances, Preamble, 44 Fed. Reg. 50776 (August 29, 1979);
Memorandum from Christopher J. Capper and William A. Sullivan, Jr. to
Regional Administrators (February 23, 1982) at 5, 9; EP.\ pamphlet,
"EPA's Emergency Response Program" (HW-3) (April, 1982); it is also an
implicit facet of EPA's recently published "Guidelines for Using the
Imminent Hazard, Enforcement and Emergency Response Authorities of
Superfund and Other Statutes," 47 Fed. Reg. 20664 (May 13, 1982).
l^Report of the Senate Committee on Public Works accompanying
S. 7 , supra at 17.
U.S.C. §§ 1651 et seq.
IA633 U.S.C. §§ 1501 et seq.
U743 U.S.C. §§ 1801 et sag.
U.S.C. §§ 9601 et seq. There are numerous other legal
authorities that may apply to the prevention and cleanup of oil and
hazardous substance spills. See, e.g. , 40 C.F.R. Part 1510, Annex
VII.
a brief comparison of the four oil spill fund statutes, see*
Musgrove, Lee, Reisch, Blodgett and Armltage, "Compensation and
Liability Legislation for Victims of Toxic Substance Pollution: A
Comparison of Proposals," (September 16, 1979), reprinted in hearing
on H.R. 85, No. 96-26, 96th Cong. 1st Sess. (September 26, 1979) at
442-44.
I500uter Continental Shelf Lands Act Amendments of 1978 (OCSLA)
§ 303(b)(l)(A); Deepwater Port Act of 1974 (DWPA) § 18(f)(5).
151QCSLA S 303(b)(l)(B); DWPA § 18(f)(2).
1 52Trans-Alaska Pipeline Authorization Act of 1973 (TAPAA)
§ 204(b). See Alyeska Pipeline Service Co. v. U.S., 16 E.R.C. 1812
(U.S. Ct. Cl. 1981).
153See TAPAA § 204(c).
l54Proposed oil "Superfund" legislation In 1979 was justified In
part as a means of unifying a "patchwork of overlapping, inconsistent
Federal and State laws." Letter to Hon. Thomas P. O'Neill, Jr. from
Brock Adams (Secretary of Transportation), March 20, 1979, forwarding
H.R. 29 to Congress, reprinted in House Hearings No. 96-16, March 13,
1979, at 105.
155H.R. 85 was passed by the House In 1980, but not the Senate.
It was Introduced again In 1981 but was not acted on by the House or
the Senate. It has not been relntroduced In 1982. Interview with
-112-
-------
INDEMNIFICATION STUDY
Section III
Suznnne Bolton, Subcommittee Staff, May 25, 1982. A more recent bill,
JI..R. 5906, currently under active consideration, would amend the Outer
Contenttal Shelf Lands Act to impose nn overall limit of $75 million
on the liability of owners and operators for damages and cleanup
costs. As explained, this would facilitate the insurability of oil
drilling facilities. Id.
issues arising in a legislative debate on the oil Super-
fund legislation include:
• the ability to minimize defenses to liability;
• the need to impose stricter penalties to help prevent
spills, as well as to encourage prompt mitigation and
cleanup efforts;
• the need for a larger fund to finance cleanup when respon-
sible parties cannot be Identified;
• the need for expanded provisions relating to consequential
damages and damages incurred by third parties;
• the need for state reimbursement and legislative
preemption;
• the use of fees to finance the fund (amount, to whom
charged, how managed, etc.); and
• limits on liability and defenses.
House Hearings No. 96-16 (March 13, 14, July 31, September 13, 1979);
No. 96-26 (September 26, 1979); No. 96-114 (June 19, August 17,
October 10, 11, 1979). Most of these issues were ultimately addressed
under CERCLA, which grew out of the oil Superfund legislative efforts
but ironically was enacted without oil spill coverage.
• 157House Hearing No. 96-16, supra at 66-77. Under § 102 the
Department of the Treasury would be responsible for administering the
fund, which would be financed by a fee charged to refiners and
export/import terminals. Other aspects of the law would be imple-
mented by the Department of Transporation, except that private
insurance or claims adjustment organizations would be relied on to
process claims. A formal administrative review procedure would apply
in the case of claims disputes, while federal cost recovery and
penalty collection proceedings would be handled through the Department
of Justice.
158See Congressional Quarterly, Almanac — 1980 at 584-93.
-113-
-------
INDEMNIFICATION STUDY
Section III
^'Hearings Before the Subcommittee on Transportation and Commerce
of the House Committee on Interstate and Foreign Commerce on 11. R.
4571, H.R, 4566, and H.R. 5290 (June 19, 1979), No. 96-1U (1980) at
216-246.
l60Section 7003 was thought to be Ineffective because of the dif-
ficulty In Identifying the responsible party, the financial Inability
of responsible parties to cover the costs of remedial action, the dif-
ficulty in sustaining the legal burden of demonstrating that the
hazard is "imminent and substantial," and the time lapse that occurs
during trial* Aside from the problems of limited coverage, § 311 and
the emergency powers provision of the Clean Water Act, § 504, were
criticized because they contained inadequate funds to allow for effec-
tive action by EPA — $35 million under § 311, and $10 million under
§ 504 compared to an estimated need of as much as $6.1 billion to
address Inactive and abandoned hazardous waste dump sites alone.
Similar procedural and funding constraints were said to exist in other
hazardous substance control statutes, such as §§ 112 and 303 of the
Clean Air Act. Id.
161See 40 C.F.R. § 300.25(d), 47 Fed. Reg. 31207 (July 16, 1982).
types of funding mechanisms are available for preauthorized
expenses. A cooperative agreement may be established with states anfl
political subdivisions; and contracts may be given to private parties.
If an emergency exists up to $2,500 may be made available to persons
not already on EPA's preferred contractor list; and up to $50,000 be
made available to preferred contractors. Interim Emergency Procure-
ment for Hazardous Substance Response Program (undated EPA document).
factors were suggested in discussions with numerous state
and federal agency staff involved In the implementation of CERCLA.
16^Supra notes 159, 161, and 163. See also "Guidelines for Using
the Imminent Hazard, Enforcement and Emergency response Authorities of
Superfund and Other Statutes," 47 Fed. Reg. 20664 (May 13, 1982);
House Hearing on "Implementation of the Comprehensive Environmental
Response, Compensation, and Liability Act of 1981" No. 97-H31, 97th
Cong. 1st Sess. (July 8, 1981), statement of Congressman Florlo, at
5-6.
l^Intervlew uith Swep Davis, Environmental Testing Corporation
(Former Associate Assistant Administrator for Water and Waste
Management, EPA), May 3, 1982.
S«£ House Hearings, No. 97-H31, supra note 164.
^^Information derived from a review of case files In the U.S.
Court of Claims. ICF Draft Memorandum, "Cost Associated with § 311(1)
of the Clean Water Act," March 8, 1982 for EPA's Office of Toxic
Substances, Regulatory Impacts Branch.
-114-
-------
INDEMNIFICATION STUDY
Section III
_ e . g . , House Hearings on Coast Guard Authorization, F.Y.
1982 (H7R7 2559) and Oversight, March 16, 1981, H. Rep. No. 97-2, 97th
Cong. 1st Sess., 251, 298, 302. It should be noted that such respon-
sibilities exist under DCS LA and DWPA. However, Implementation
experience is too limited to allow for evaluation at this stage.
I70lnterviews with: Commander George Brown, USCG, 12th District
and Commander Skip Omstead, USCG, llth District, January 27, 1982.
l71See_, e.g.. Quarles Petroleum Co. v. United States, 551 F.2d
1201, 1204 (U.S. Ct. Cl. 1977) (sets out elements of § 311(1) claim).
172Interview with Commander Adams, USCG, March 22, 1982.
175In 1979 EPA estimated that from 700-1,200 chemical spills
occured each year which .would be subject to § 311 regulation.
Statement of Barbara Blum, Deputy Administrator, House Hearings on
H.R. 4571, H.R. 4566, and H.R. 5290, No. 96-113, 96th Cong. 1st Sess
at 225.
continued availability of a remedy under § 311(1) to a
discharger of contaminants covered under CERCLA depends on whether
§ 311(1) in a particular case is determined to be In conflict with
CERCLA. See CERCLA § 304(c). Such an argument might be made in the
case of expenses not authorized as consistent with the national con-
tingency plan. See CERCLA § lll(a)(2). Note, however, that special
relief in such a situation might be available under CERCLA § lll(b).
^ Adams supra note 172.
178See
17940 C.F.R. part 112; 33 C.F.R. part 154. Hazardous substance
spill prevention regulations do not exist and are not included on the
most recent regulatory agenda. See 47 Fed. Reg. 15702 (April 12,
1982).
l80Section 311(b)(5) provides for up to $10,000 In fines and one
year of imprisonment for failure to report a discharge, and
§ 311(b)(6) requires an assessment by the USCG of up to $5,000 as a
civil penalty for each offense and, alternatively, allows for EPA to
pursue in court penalties not exceeding $50,000 per discharge (or not
exceeding $250,000 If willful negligence or misconduct was involved).
In addtlon, § 311(j) allows for an administrative penalty of up to
-115-
-------
INDEMNIFICATION STUDY
Section III
$5,000 for each violation of prevention, containment and removal
regulations •
Section 103(a) of CERCLA provides for up to $10,000 in fines and
one year of imprisonment for notification violations, and $20,000 In
fines and one year of imprisonment for record keeping violations*
Failure to comply with an abatement order under § 106 may result in up
to a $5,000 penalty for each day of violation; and under § 109 up to
$10,000 per day may be assessed against a person who has failed to
comply with financial responsibility requirements under § 108. If a
person fails to conduct required removal or remedial actions, pursuant
to § 107(c)(3) he may be liable for punitive damages that amount to
three times the amount of costs incurred by the Hazardous Substance
Response Fund*
was a conclusion of the First Report of the President's
Panel on Oil Spills. Executive Office of the President, Office of
Science and Technology, "The Oil Spill Problem," (undated publication)
at 4.
other than § 311(1) already exist, including fore-
knowledge that defenses to liability are very narrowly construed,
possible liability for unreasonable costs of cleanup if undertaken by
the government, treble damages (under CERCLA § 107(c)), potential
state and common law liability and concerns over public image.
discussion supra at 83.
however, that the Trans-Alaska Pipeline Authorization Act
of 1973, 43 U.S.C. §§ 1651 et seq., does not provide an act of God or
third-party-negligence defense. The legislative history notes that
stricter liability standards are warranted because of the enormous
size of tankers expected to transport oil from Alaska and the
increased risk, therefore, of high damages to property and natural
resources. 1973 U.S. Code Cong. & Admin. News, at 2530.
185See_, e.g., City of Pawtuckett v. United States, 211 Ct. Cl.
324 (1976); Proctor Wholesale Co. v. United States. 215 Ct. Cl. 1049
(1978); Chicago, Milwaukee. St. Paul & Pacific Railroad Co. v. United
States^ 575 F.2d 839 (U.S. Ct. Cl. 1978).
l86See text and notes, supra at 80.
text and notes, supra at 78.
188Supra note 136.
l89See 40 C.F.R. § 300.25(d), 47 Fed. Reg. 31207 (July 16, 1982).
"Guidelines for Using the Imminent Hazard, Enforcement and
Emergency Response Authorities of Superfund and Other Statutes," 47
Fed. Reg. 20664 (May 13, 1982).
-116-
-------
INDEMNIFICATION STUDY
Section III
l^llt should be noted that persons involved in the administration
of both CERCLA and § 311, in EPA and the USCG, have stated that they
prefer the administrative ease of § 311. Objections range from the
complexity of the claims and appeals procedures, to the difficulty of
Implementing procurement requirements. It is generally recognized
that a program of the size of CERCLA ($1.6 billion over a five-year
period) merits more careful control than one the size of $ 311 ($100
million over a ten-year period). However, it was suggested that a
spill response program requires special flexibility; and Chat this
need is overshadowed by combining it with the long-term remedial
program required for abandoned waste sites. It is possible that such
concerns will be less Important as the respective EPA and USCG staffs
gain experience in implementing the new law.
-117-
-------
INDEMNIFICATION STUDY
Section IV
IV. EVALUATION' OF POTENTIAL LOSS CATEGORIES
Page
A. Introduction 119
B. Indemnification for Disclosure of Confidential
Business Information 122
1. Description of the Problem 122
2. Evaluation « 123
3. Conclusion v 124
o
C. Indemnification for Delays in EPA Permit Processing . . . 126
1. Description of the Problem 126
2. Evaluation 127
3. Conclusion 128
D. Conflicts and Overlaps 130
1. Description 130
2. Candidate Indemnifiable Situations 130
3. Rationale for Indemnfication 131
4. Evaluation of Indemnification 132
E. Emergency Powers 134
1. Birmingham Air Pollution Alert 134
2. Evaluation 135
3. Emergency Authority Under Other Statutes 136
4. Other Types of Emergency Resolution Provisions. . . . 137
5. Conclusion 138
F. Unreasonable Enforcement • 139
1. Introduction 139
2* The Potential for Unreasonable Administrative
Enforcement 139
3. Blacklisting 145
4. Alternative Remedies and Prevention Mechanisms. . • • 145
5. Feasibility and Impacts 146
6. Conclusion 147
G. Changes in Agency Policy 148
1. Description of the Problem 148
2. Evaluation 148
3. Conclusion • 150
Footnotes 151
-118-
-------
INDEMNIFICATION STUDY
Section IV
A. INTRODUCTION
This section of the report concerns the search for potentially
indemniftable EPA actions in areas other than merely extensions to
existing EPA indemnity programs. In order to determine whether there
are circumstances which might justify such indemnification, two
research techniques were used. First, categories of events were iden-
tified that appeared most likely to contain indemnifiable situations.
Second, specific examples of allegedly indemnifiable situations were
sought to sharpen the analysis and assess the seriousness of the
problems. In each specific case for analysis, indemnification factors
identified in previous research were applied to determine the poten-
tial arguments justifying and opposing the need for indemnification.
For the first technique, the easiest general category of events
to define were those in which some losses are already indemnified
under existing EPA programs. (This category was covered in the pre-
vious chapter as "extensions" of existing programs*) Another category
involved situations defined to be similar, but not identical, to
existing programs. Still another category concerned situations iden-
tified by the convergence of several of the factors that had been
determined to be important in earlier phases of the study. For
example, "emergency action" was chosen because it presents an
increased risk of agency error due to the speed with which the agency
decision must be made, and because procedural due process may of
necessity be suspended. In addition, the agency may desire to stimu-
late speedy action on the part of outside parties during emergencies,
and this may warrant use of indemnification as an incentive.
In undertaking the search for potentially Indemnifiable situa-
tions, every statute and major regulation implemented by EPA were
reviewed. Discussions were held with past and current EPA staff to
learn about the practical operation of the regulations, and particu-
larly whether there have been or might be indemnifiable losses. The
categories identified in the course of the study guided the search,
but were also refined and redefined as additional incidents were iden-
tified or reconsidered.
Industry sources were also consulted for identification of
potentially indemnifiable losses. This effort included a review of
surveys and reports on regulatory problems in general, including many
that had not specifically addressed indemnification as a solution to
any regulatory problem.* In conjunction with this review, major
industry trade associations were contacted. Association represen-
tatives often referred us to particular members who also provided com-
ments and information. As a result, a fairly broad perspective of
current Industry attitudes toward EPA regulation and the need for
indemnification was obtained.2
Based on the review of EPA statutes and regulations and con-
sultation with Agency and industry spokespersons, the following cate-
-119-
-------
INDEMNIFICATION STUDY
Section IV
goriee were selected for more detailed evaluation to determine the
need for Indemnification for potential losses due to:
• Disclosure of confidential business information;
• Delays in Agency action;
• Conflicting and overlapping requirements;
• Error or overbreadth in Agency emergency action;
• Unreasonable administrative enforcement action; and
• Changes in Agency policy.
Each loss area was selected partly because it was a topic of major
complaint or concern on the part of industry or the agency. The loss
categories provide a fairly diverse representation of loss types for
purposes of analysis, but an evaluation of every industry complaint or
potential loss would, of course, have been impossible within the scope
of this study.
Of the three types of indemnification identified and evaluated
in the course of this study (equitable, disincentive, incentive),
almost all of the categories chosen for analysis fall within the
"equitable indemnification'' category. This Is consistent with the
actual historical distribution of indemnification claims presented to
Congress over the years.
In considering the following analysis, one must keep in mind
that government generally does not pay for the cost of compliance with
regulatory requirements. Nor does government normally pay for the
consequences of the burdens of regulations unless those burdens are so
great as to constitute a "taking" of property through the almost
complete destruction of its value. As the U.S. Supreme Court recently
restated in Andrus v. Allard, a Fifth Amendment taking case:
Suffice it to say that government regulation
— by definition — involves.the adjustment of
rights for the public good. Often this
adjustment curtails some potential for the use
or economic exploitation of private property.
To require compensation in all such cir-
cumstances would effectively compel the
government to regulate by purchase. "Govern-
ment hardly could go on if to some extent
values incident to property could not be dimi-
nished without paving for every such change in
the general law."*a
-120-
-------
INDEMNIFICATION STUDY
Section IV
There are very limited exceptions to these general rules. For
example, the federal government provides grant money to assist munici-
palities in complying with the Clean Water Act's wastewater treatment
requirements. However, this involves the federal government assisting
local governments, rather than aiding private regulated parties. A
narrow example of government paying for the indirect burdens of regu-
lations is the recently-eliminated beekeepers indemnification program
in which payments were made to beekeepers for the consequences of the
elimination of DDT from the marketplace.
One must also keep in mind that the federal government already
has provided a broad-based indemnity scheme covering many areas of
potential loss to private parties. Specifically, the Federal Tort
Claims allows suits for negligence with certain exceptions, and the
Tucker Act allows for recovery of contract claims and implements reco-
very of claims under the Constitution's Fifth Amendment Taking Clause.
Thus, our search for potentially indemnifiable circumstances
focused on the narrow, gray area of exceptional circumstances which
theoretically fall outside the existing indemnity schemes provided by
the FTCA and the Tucker Act, and yet do not fall into an area of loss
which is considered not to be compensable. In other words, for indem-
nity to be recommended under this study, either the limitations to the
existing indemnity schemes must be removed (e.g., by repealing the
discretionary function and misrepresentation exceptions to the FTCA),
or unique circumstances must be found which Justify an exception to
those limits.
An example of the possible wholesale removal of an existing
limitation would be to permit compensation for governmental "error" in
the area of policy formation (rather than for just operational
negligence). This would involve repeal of modification of the discre-
tionary function exception to the FTCA's waiver of sovereign Immunity.
This, of course, would pose very serious practical problems of deter-
mining whether a discretionary policy judgment among several alter-
natives can be deemed "erroneous." Unlike the question of operational
negligence which involves measuring government action against a more
definite standard (e.g., the terms of the regulation or statute
itself), the issue of policy error is complicated by the lack of any
clear-cut way of deciding that the policy judgment was "erroneous" at
the time it was made. Furthermore, the threat of having to pay com-
pensation might seriously impair the policymaker's judgment and have a
chilling effect upon the consideration of various policy alternatives.
In sum, there is no support for a recommendation to repeal the
discretionary function exception and authorize a vague, broad standard
of government liability for "erroneous" policy decisions.
-121-
-------
INDEMNIFICATION STUDY
Section IV
B. INDEMNIFICATION FOR DISCLOSURE OF
CONFIDENTIAL BUSINESS INFORMATION
1. Description of the Problem
Concern over possible industry losses resulting from disclosure
of confidential business information (CB1) submitted to regulatory
agencies has resulted in efforts to get Congress and the federal judi-
ciary to define CBI more precisely and to narrow the scope of
authorized CBI disclosure or use.
Although unauthorized disclosures may cause some worry, industry
seems to be primarily concerned with authorized disclosures under the
Freedom of Information Act (FOIA) or other statutes. This is clear
from efforts to restrict or block such authorized disclosures. Almost
no industry interest has been shown in providing indemnification to
private parties suffering damage from either authorized or unauthor-
ized government disclosure of CBI.
Much industry legislative effort is now aimed at strengthening
the CBI exemption to the Freedom of Information Act.3 Amendments to
FIFRA have also been advanced that would alter the disclosure provi-
sions of §§ 3 and 10^ by providing for treble and punitive damages
against a private party who properly receives CBI for a limited pur-
pose but then improperly discloses the information.^
Extensive, but generally unsuccessful, industry efforts have
also been undertaken in the courts to disallow or restrict intentional
Agency use and disclosure of pesticide data under §§ 3 and 10 of
FIFRA.6 Some companies have also challenged EPA's authority to permit
contractors to gain access to CBI while assisting EPA with inspection
efforts, but only one such judicial challenge has been successful thus
far.^ No complaints were found concerning EPA contractors wrongfully
disclosing CBI.
The CBI issue at EPA has focused on pesticides and toxics
because of Agency acquisition of sensitive product information in
those programs, and because the CBI issue received extensive congres-
sional scrutiny while TSCA was being debated•** Elaborate FOIA proce-
dures for authorized EPA release of CBI are now provided under 40
C.F.R. Part 2, and the Agency has developed both TSCA and FIFRA
security manuals designed to safeguard CBI obtained under those
statutes.^ In addition to EPA's provisions for employee discipline,
there are criminal sanctions for intentional unauthorized disclosures.
Title 18 U.S.C. S 1905 Is a general trade secrets disclosure prohibi-
tion, and TSCA, FIFRA, RCRA, and CERCLA also contain criminal sanc-
tions against wrongful disclosure.^
Despite detailed EPA security procedures and apparently very few
disclosure incidents, there are indications that some industry repre-
-122-
-------
INDEMNIFICATION STUDY
Section IV
sentatives are still not comfortable with Agency assurances of ade-
quate protection from unauthorized CBI disclosure. One industry
representative contends that EPA's disciplinary procedures are not
sufficiently severe to deter unauthorized disclosure by employees.**
He also contends that potential criminal sanctions for unauthorized
intentional disclosure do not serve as an effective deterrent.^
Knowledgeable EPA and Justice Department personnel knew of no
claims filed against EPA seeking government compensation due to
unauthorized (either intentional or inadvertent) release of CBI.
Nor do there appear to have been prosecutions or disciplinary actions
for intentional and unauthorized disclosure of such information.
The true number of improper [disclosures may be difficult to
ascertain however. Some disclosures may not be detected, and an
affected company may be reluctant to come forward with a complaint for
fear of bringing the information to the attention of competitors or
weakening its trade secret status.^ Furthermore, company management
may fear that its stockholders would be upset that management had
allowed the disclosure to occur.
2. Evaluation
Indemnification has not generally been put forward as a remedy
for the disclosure of CBI, though one knowledgeable industry represen-
tative has suggested use of such a remedy. Most industry parties,
however, seem to believe that payment of money damages does not deal
adequately with the problem. Those concerned with the consequences of
disclosure of CBI believe that the solution to such a problem is to
prevent disclosure rather than to provide compensation. Industry thus
places more emphasis on preventing submission of the information in
the first place, or limiting the ability of the Agency to make legally
authorized disclosures, than on ensuring effective EPA security and
disciplinary provisions to protect against unauthorized or negligent
disclosures.
Evaluation of an indemnification mechanism for CBI disclosure
requires that distinctions be made about the various circumstances in
which CBI can be disclosed. First, there is an important distinction
between unauthorized and authorized disclosures* Second, unauthorized
disclosures can be either negligent or Intentional.
The rationale for Indemnification can focus on the equitable
need to provide compensation to the injured party, or on affecting the
behavior of agency employees so as to prevent unauthorized disclo-
sures, or on both. The former rational is most important with respect
to authorized disclosures. It may be possible to permit necessary
disclosure in a way that minimizes or eliminates the loss to the party
whose CBI Is released.18 Thus, the key question with respect to
authorized disclosures may be whether the burden imposed on the
-123-
-------
INDEMNIFICATION STUDY
Section IV
Injured party for the benefit of the public is one for which compensa-
tion should be paid. °
Because indemnification under the § 25(a) study mandate is not
intended to cover payments that might otherwise be recovered in a tort
action under the Federal Tort Claims Act, it is necessary to consider
whether particular types of disclosure losses are already actionable
under the FTCA. If disclosure would be judicially actionable, then
consideration of indemnification is unnecessary. For the purpose of
an FTCA analysis, there are three different categories of disclosure.
In a case of intentional and unauthorized disclosure, the exception in
28 U.S.C. § 2860(h) arguably bars an FTCA action, although this pre-
cise issue has not yet been decided in the courts.^0 Exception (h)
might similarly bar an action for negligent disclosure. Even If it
does not, the injured party might have difficulty establishing the
duty of care and proving that it had been breached.^ Finally, cer-
tain intentional disclosures might be authorized or mandated by
statute. For example, TSCA provides authority for release of CBI if
it is necessary to protect the public health.22 Because such disclo-
sures are expressly authorized or required, they would not be action-
able under the FTCA. Thus, only this last category of disclosure is
clearly within the scope of indemnification under this study.
To the extent private parties are injured by wrongful government
disclosure of CBI, the solution may not be to establish a special EPA
indemnification program, but instead to amend the Federal Tort Claims
Act specifically to permit suits for unauthorized disclosures.^ in
other words, if there is a problem, it is government-wide in scope and
not unique to EPA, and should be dealt with through amendments to the
FTCA or FOIA.24 Any attempt to compensate for release of CBI would
have to solve the administrative problem of how to measure the party's
loss. The value of Information, and expecially the loss of secrecy is
speculative and difficult to ascertain. A compensation program vould
also have to overcome the reluctance of a company to advertise the re-
lease of its information by making a claim.
3. Conclusion
Two of the criteria discussed in Section III-A are particularly
pertinent to this evaluation. CBI releases are clearly addressed more
effectively by alternatives preventing them than by compensation
afterwards. Also, an indemnification program would have very diffi-
cult administrative problems in defining the value of the release and
in protecting the company from further damage in the course of pur-
suing Its claim. Where existing alternatives are not fully satisfac-
tory, government-wide solutions through amendments to the FOIA or the
FTCA may be more appropriate to consider than an isolated EPA remedy.
No EPA indemnification program seems warranted for losses
resulting from unauthorized disclosure of CBI. Evidence of such
-124-
-------
INDEMNIFICATION STUDY
Section IV
unauthorized disclosures appears to be quite limited, and there are
both criminal sanctions against intentional disclosure and elaborate
security provisions under FIFRA and TSCA designed to safeguard CBI.
Relevant judicial decisions involving EPA have established that
authorized disclosures by the Agency do not amount to unconstitutional
takings or, in some instances, that the party does not have a property
interest in the information.2* Congressionally-authorized disclosures
designed to protect the public health or welfare are comparable to
other regulatory burdens not requiring compensation. In many cases,
it may be possible to accomplish the needed disclosure, while at the
same time protecting the affected party from loss or damage. Further-
more, much of the conflict over CBI disclosure involves defining the
term. Indemnification would not be appropriate where Congress has
already defined certain information as not constituting protected CBI
or where Congress had given the Agency the discretion to make such a
determination.
-------
INDEMNIFICATION STUDY
Section IV
C. INDEMNIFICATION FOR DELAYS
1. Description of The Problem
(a) Delays in Promulgating Regulations
Although all types of delays in government action were con-
sidered in the course of this study, failure to meet congressionally-
mandated deadlines for the promulgation of regulations were determined
at an early stage to be Inappropriate bases for indemnfication. Cer-
tainly indemnification on this basis would involve severe practical
problems. Measuring losses and determining causation would be diffi-
cult. Ascertaining the particular regulation that would have been
promulgated, had delay not occurred, would amount to pure speculation.
Costs may be unacceptable, since great numbers of people are arguably
adversely affected by the delay and therefore potentially entitled to
indemnification. Such an indemnity scheme might also be highly
disruptive of agency objectives. Moreover, a mandatory injunction
forcing the agency to discharge a statutory and non-discretionary duty
within a particular timetable is available and a preferable alter-
native relief mechnism that has been used in the past to deal with ,
this problem. For example, in February 1982, EPA was ordered by the
U.S. District Court for the District of Columbia to fulfill its statu-
tory duty under the Comprehensive Environmental Response, Compensa-
tion, and Liablity Act of 1980 to promulgate guidelines for emergency
response actions and to revise and republish the National Contingency
Plan.26 The court relied on the statute's mandatory requirement that
EPA act within 180 days of the statute's enactment.
(b) Delays in EPA Permit Processing
Industry representatives have complained about delays in
government permit processing, saying that these delays have increased
costs of financing or construction and caused losses in opportunities
and profits.^ Industry sees the permit-delay problem as particularly
accute in the environmental area because of the complexity of the
environmental laws and the Involvement of many different regulatory
parties. This complexity, coupled with slow government action, has
allegedly stretched out the total time period necessary for siting
energy facilities and other industrial operations.
Several studies have indicated that the causes of delays in
construction of new power plants and other facilities are not limited
to regulatory problems. In one study, equipment-related and labor-
related delays appeared to be much more important factors than regula-
tory delays such as changes in requirements or delays in obtaining
permits.2° Studies have also Indicated that EPA permit processing
appears to be reasonably swift, considering the complexity of the sta-
tutory requirements. For example, the average permit processing time
for PSD permits has been reported to be less than six months from the
-126-
-------
INDEMNIFICATION STUDY
Section IV
time of submission of a complete application to the time of issuance.
(It takes about three months from the initial submission to the time
?Q
the application is considered to be complete.) Despite statistical
evidence that EPA has a reasonably good permit-processing performance
record, the Agency has undertaken substantial efforts to assess the
validity of industry complaints and to evaluate possible improvements
to make the permit-processing system more efficient.^
Theoretically, strict time deadlines could be established,
and indemnification could be required whenever EPA misses these dead-
lines. However, such an approach would subject EPA to indemnification
regardless of its fault and for delays outside the Agency's control.
On the other hand, indemnification for losses due to permit-processing
delays could be based upon EPA fault. Such indemnification would be
premised on the permit applicant's right to timely review of its per-
mit application and EPA's unreasonable failure to process the applica-
tion within a certain period of time.
2. Evaluation
Both industry and EPA parties concerned with the problem of per-
mit delays have focused -on remedial solutions other than Indemnifica-
tion. For example, the Business Round table has concentrated on
identifying potential structural changes to the Clean Air Act which
would lessen the complexity and severity of permit requirements.^
Within the Agency, efforts have been focused upon tracking the permit
process and setting average and maximum performance times, reviewing
possible changes in substantive and procedural requirements, reviewing
particular regulations concerning the consolidated permit process, and
attempting to manage the existing permit process more effectively.32
The Importance of the alternatives to indemnification is height-
ened because of the substantial administrative difficulties with
indemnification. Unlike the factual determinations necessary to FIFRA
§ 15, the determination that a loss was caused by permit delays and
the measurement of damages would be extremely difficult. While
increased construction costs resulting from delay might be measurable,
the determination of lost opportunities and profits would be much more
problematic. Of course, indemnification could be limited to construc-
tion cost or other readily measurable increases.
Another administrative problem would be to determine whether a
delay had actually been unreasonable. Although the establishment of a
strict time period for permit processing might provide a simple device
by which to determine whether an indemnifiable delay had occurred,
there would be a substantial problem in ascertaining the starting
point for the time period to begin to run, because processing of the
permit application may have to await submission of a more complete
application. Frequently, applications are judged so incomplete that
they must be returned to the applicant for additional information. A
-127-
-------
INDEMNIFICATION STUDY
Section IV
similar problem arose in the debates over the proposed Energy
Mobilization Board when it was supposed to. establish a deadline after
which a permit would become effective by default.^ This approach was
concluded to be impractical because of the difficulty in setting the
starting date of the permit processing time period.
Another serious problem is that the incentives and disincentives
resulting from indemnification may have an adverse impact upon the
environmental objectives of the agency. An indemnification program
might substantially shift the focus of agency personnel from the
quality of permit review to the speed of that review. The allowable
permit processing time might be Inadequate for complex and controver-
sial projects requiring careful evaluation and Involving difficult
judgments, and thus might cfoster the poorest decisions in the hardest
cases. Other consequences might be to prompt EPA employees to waste
time developing a "paper trail" with regard to the completeness of the
application or in order to establish evidence that its delay was not
unreasonable. EPA might also refrain from getting states to act as
EPA'a agent in certain situations because of the fear of being subject
to indemnification for state actions.
A reverse type of Indemnification is theoretically possible
where fees are charged to regulated parties for the costs of permit
processing. Under such an approach, the fee could be waived if EPA
failed to process the permit within a specified time period. Alter-
natively, a higher fee might be charged for "speedy" permit proces-
sing. Rather than using indemnification as a penalty disincentive to
the agency, a fee approach would Involve a positive incentive analo-
gous to the bonus that Amtrak pays rail lines for "on time" perform-
ance. Of course, such a scheme could not be used unless EPA were to
impose permit-processing fees.^
3. Conclusion
Assuming that industry is suffering losses due to delays in EPA
permit processing, an indemnification program would not be an appro-
priate remedial device to deal with the problem. There are five pri-
mary factors that weigh against the use of an indemnification program
to remedy problems with permit delays.
First, the evidence does not show, in the aggregate, that an
unreasonable amount of permit-processing delay is actually occurring.
To the extent that there are delays in the development of new power
plants or other industrial facilities, regulatory problems (including
permit processing delays) are usually not significant factors in those
delays. While there are some Infrequent situations in which delay
results solely from regulatory problems, the difficulties often result
from the congressionally-determined burden of the requirements and the
complexity of the system, rather than from unreasonable agency permit
processing procedures or from convoluted regulations created by the
agency.
-128-
-------
INDEMNIFICATION STUDY
Section IV
Second, regulated parties usually do not have (statutory or
regulatory) rights to have a permit processed within a particular
time. If a required time period for processing is exceeded, then the
affected party can bring an action for a mandatory injunction to
remedy the problem.
Third, to the extent that permit losses actually are resulting
from permit delays, the generally preferred corrective solution is to
Improve and speed up the permit process instead of paying indemnifica-
tion. Numerous agency efforts are now underway to accomplish the
objective of Improved permit processing, even without whatever stim-
ulus is derived from the fear of indemnification.
Fourth, there would be substantial administrative problems
involved in the use of indemnification as a remedy for permit delays.
In particular, determination of whether the delay was unreasonable and
the cause of the loss would be extremely difficult* Furthermore, the
determination of the appropriate quantum of damages would be very dif-
ficult unless standards limiting claimants to measurable losses in
specified categories were used.
Fifth, indemnification would establish counter-productive incen-
tives and disincentives for the agency and the permit applicant.
Indemnification would frequently also conflict with agency objectives.
-129-
-------
INDEMNIFICATION STUDY
Section IV
D. CONFLICTS AND OVERLAPS
1. Description
Conflicting and overlapping regulatory requirements have been a
principal target of regulatory reformers and, to a lesser extent, may
serve as a basis for the payment of indemnification to those caught in
the conflict, or forced to comply with two sets of regulatory require-
ments. In fact, however, regulatory conflict and overlap do not
appear to be an important problem at EPA, relative to other regulatory
issues. This is partly because all of the major environmental statu-
tes include provisions directing that EPA coordinate its activities,
both between separate divisions and with other agencies of the federal
and state governments.
Regulatory conflict occurs when compliance with one regulatory
requirements results in the inability to comply with another require-
ment. Regulatory overlap occurs when separate regulations with simi-
lar objectives are aimed at a single activity or process.
2. Candidate Indemnifiable Situations
(a) Interacting Requirements
Interacting requirements are those where compliance with one
requirement brings a party within the scope of regulations in another
area. If the second regulation conflicts with the first, such that it
is impossible to comply with both, then the interacting requirements
have caused a loss which is Impossible to avoid, and thus a candidate
for indemnification. The Trie situation is alleged by some to fit
this model of Interacting requirements.37
A recent GAO study identified 20 instances of interacting
requirements involving federal agencies, including four that involve
EPA.3® An example involving EPA is presented by the fact that as many
as 26 of the wastes listed as hazardous under EPA's hazardous waste
regulations are generated bv the proper operation of water and air
pollution control devices. These'devices capture toxic pollutants
in the form of sludge which must be disposed of in accordance with
EPA's RCRA regulations. This example is typical of interacting
requirements in that compliance with air or water regulations creates
the need to comply with RCRA standards. It differs from the Tris
example, however, because RCRA does not prohibit the generation of
toxic sludge, but merely subjects it to' additional regulatory
standards.
(b) Overlapping Jurisdiction
Overlapping jurisdiction occurs when two or more agencies or
offices have requirements regarding the same product, substance or
-130-
-------
INDEMNIFICATION STUDY
Section IV
process. The GAO identified 14 examples of overlapping jurisdiction
problems, including six involving EPA. ^
An illustrative example is presented by the fact that sac-
charin is listed as a hazardous waste under RCRA, because of its car-
cinogenic properties, while the FDA continues to approve it as a food
additive. When two or more agencies or offices have jurisdiction over
a single product or activity, the potential for inconsistent treatment
such as this is increased. Overlapping jurisdiction therefore, may be
thought of as forcing regulated parties to incur the added costs of
complying with two sets of regulations, when it would arguably be more
efficient for there to be a single "integrated" set of requirements.
(c) Duplicate Enforcement
Duplicate enforcement occurs when two or more agencies have
responsibility for investigating complaints, issuing permits, or con-
ducting inspections for similar or related purposes. The GAO found 18
examples of duplicate enforcement in its report, including 10
involving EPA.*1
For example, both EPA and the states make inspections under
both the Clean Air Act and the Clean Water Act. These inspections
occur at different times and require that Information be presented in
different formats. A similar problem may be presented if a single
enterprise must apply for and obtain two or more permits for separate
agencies or offices. To the extent that these permits serve similar
purposes, or enforcement actions are aimed at similar conduct, they
force regulated parties to incur expenses not otherwise imposed.
3. Rationale for Indemnflcatlon
The rationale for indemnification is slightly different for
interacting requirements than for the other types of conflicts and
overlaps. Where interacting requirements are the asserted basis for
indemnification, the rationale is that compliance with one requirement
forces the regulated party to comply with another requirement. Thus
the regulated party has incurred a double expense, made necessary only
because of the manner in which the first regulation operates. If the
second regulation amounts to a total prohibition, then the affected
party is caught in an impossible situation: compelled by one regula-
tion to do what is prohibited by a second regulation.
With respect to overlapping jurisdications and duplicate
enforcement, the rationale for indemnification is that regulatory
activity designed to deal with the particular regulated party affected
could be designed to achieve the same effect at lower cost to that
party if only a single agency or office were responsible for the
party's activity or product. Since it is the structure of EPA, rather
than the nature of the regulated activity, that imposes the double
compliance costs, these costs should be indemnified.
-131-
-------
INDEMNIFICATION STUDY
Section IV
4. Evaluation of Indemnification
The GAO has concluded that the problems of regulatory conflict
and overlap are not as serious as regulatory reform advocates
assert.^ Moreover, where they exist, they do not necessarily produce
a loss, as that term is ordinarily defined. While compliance costs
may be greater where regulatory conflicts and overlaps exist, when
compared to the situation where a single office is responsible for
controlling all aspects of a party's activities, these increased costs
are not losses, in the traditional sense, because there is no reason
to suppose that regulated parties are entitled to regulatory programs
designed to minimize compliance costs. Indeed, EPA is certainly
entitled to structure its regulatory'efforts so as to minimize its own
administrative costs or
-------
INDEMNIFICATION STUDY
Section IV
interference with EPA objectives, are more readily administered by the
Agency, and do not involve the potential for enormous expenditures.
For these reasons, and indemnifty program can not be recommended in
preference to the available alternatives.
-133-
-------
INDEMNIFICATION STUDY
Section IV
E. EMERGENCY POWERS
Any action taken quickly in an emergency is more likely to have
some untoward consequences than actions taken with enough time to
investigate, deliberate and consult. If, in addition, agency action
requires an immediate response from regulated parties, the possibility
of loss is increased. The parties cannot avoid or reduce the loss by
objecting before implemention of the agency order. A well known
example of emergency agency action, requiring immediate response,
Illustrates the theoretical problems.
1. Birmingham Air Pollution Alert43
Birmingham has had a long history of air pollution problems
because It is a mining and metallurgical center located between moun-
tains which often prevent the dispersion of air pollution created by
industry. At the time of the air pollution emergency in November,
1971, the state had just enacted, but had not yet implemented, an air
pollution control law. The state was also drafting Its State
Implementation Plan, including Its plans for handling air pollution
emergencies. Thus, there was neither a federal nor a state regulatory
structure, and no authority delegated to local officals to respond to
the episode.
The local health department tried to obtain voluntary reduction of
the particulate emissions causing the emergency. Some of the indus-
trial sources responded, but those contributing most to the problem
did not.^* As it became evident that the efforts of the county health
department were failing, and the Weather Service predicted a con-
tinuing inversion, EPA officals decided to use the emergency authority
in § 303 of the Clean Air Act to force emission reductions. The pro-
vision had been in the Act since 1967, but had never been used.
Nevertheless, EPA and Justice attorneys and local health officials put
together the necessary papers in six or seven hours.
For reasons that are not entirely clear ten years later, none of
the 23 industrial sources named in the Temporary Restraining Order
(TRO) were informed that it was being sought. The judge granted the
TRO at 1:45 a.m. on November 18 after an ex parte hearing. Soon after,
the sources were telephoned to inform them of the order. Most were
asked to postpone or reduce operations to the extent possible without
damage to equipment, rather than to shut down completely. For
example, the U.S. Steel Falrfield Works was required to Increase
coking time to the maximum extent possible, and to reduce emissions
from open hearth furnaces by ceasing feed, but was allowed to maintain
heat in the furnaces.^ The TRO was in effect for 32 hours. By the
time of the hearing set for the preliminary injunction, the weather
had changed, bringing rain.
The attorneys.for the affected facilities expressed indignation at
the lack of notice and the midnight hearing. Some also complained
-134-
-------
INDEMNIFICATION STUDY
Section IV
that their clients should not have been included in the order because
they had installed pollution control equipment or were too far away to
be contributing to the emergency. One attorney who represented eight
or nine of the sources said that, if the order had lasted longer and
greater losses had been incurred, they certainly would have sought
indemnification for their losses. "
2. Evaluation
If it is true, as alleged, that some of the named facilities were
not contributing to the air pollution emergency and thus were errone-
ously included in the order, an argument can be made for indemnifica-
tion. Errors made in such a situation are more serious than at other
times because they lead to irreversible harm. The affected parties
had no opportunity to pursuade the Agency or the Judge that they
should be removed from the list. There was no notice, and no oppor-
tunity to prepare for the new regulation. (On the other hand, it was
clear to all that the situation was serious. Apparently, visibility
was so poor in places that driving was nearly impossible.) The
order, though denominated a "temporary restraining order" had irre-
versible impact. There was no way to appeal or to have the TRO
reviewed before it took effect.
For those properly addressed, indemnification would not be
justified. Their costs were only those of complying with a proper
Agency action taken to protect the public from an imminent health
hazard. This was precisely the situation in which Congress had
intended § 303 to be used.*7
In this case, though there appears to have been some procedural
unfairness and possible error, the loss was limited by the short
period, and the way in which the order was written. It should be
remembered that drastic action was taken only after requests for
voluntary action had failed, and the situation threatened the health
of a large population. Similar losses are possible in the future, but
unlikely. The primary response to future emergencies will be made by
the 'state under their SIP's. This will cure many of the problems of
the emergency response in Birmingham.^ The industry has had a good
deal of notice of what will be expected as each level of pollution is
reached. The triggers for successive stages of control are designated
ahead of time* The regulations are developed with opportunity for
comment and with the substantive assistance of the affected parties.
Therefore, error will be less likely because of the consultation, and
cooperation more likely since the sources helped write the plans.
Application of the criteria for decisions discussed in Section
III-A, to a situation like Birmingham provides guidance for future
Incidents, though it Is not clearly for or against payment of
indemnification.
-135-
-------
INDEMNIFICATION STUDY
Section IV
Providing indemnification for an individual incident would most
likely have the objective of correcting an inequity. Thus, the first
determination would be whether the agency action was indeed inequit-
able. This requires a collection of all the facts. In the Birmingham
situation a decision-maker balanced the refusal of some of the sources
to cooperate and the clear danger to the public health against the
government's precipitous and perhaps overbroad action. No formula or
criteria can substitute for an analysis of the particular facts.
Alternatives to Indemnification are fewer during emergencies than
at other times. For example, exceptions or waivers to agency orders
might undermine the purpose of the action, and take too long to pro-
pose and approve. There are however, many ways to limit losses that
might be incurred during an emergency, and the environmental statutes
contain many examples. TSCA requires hearings before or within five
days of emergency actions thus limiting the time an unchallenged
emergency order could be In effect*
The language of the emergency powers sections make it clear that
Congress has accepted greater compliance costs in an emergency than
for ordinary regulations. Such expressions of congressional opinion
should be taken into account in deciding what losses warrant
Indemnification.
The possibility that the payments of an indemnity might have a
chilling effect on future agency action is more serious in relation to
emergencies because the threat to public health is by definition
imminent and substantial.
3. Emergency Authority Under Other Statutes
Virtually all pollution control statutes Implemented by EPA have
provisions similar to § 303 of the CAA. These provisions respond to
the common need to act quickly to alleviate immediate threats to
health or the environment that could not have been prevented by ordi-
nary regulatory mechanisms. These extraordinary measures are most
needed before the rest of the regulatory structure is in place, as
demonstrated in the Birmingham incident, but there will continue to be
accidents or occurrences not controlled by regulation.
Like the Air Act, the Clean Water Act now regulates most emissions
of pollutants. Emergencies, however, are probably most likely with
respect to unregulated emissions, although they can occur as the
result of breakdowns of otherwise functioning processes, or other unu-
sual situations. Under the Clean Water Act as well, internal and
external review is required before taking action, even in emergency
situations. Both Agency officials and the Department of Justice must
approve requests to seek court orders-. The judge acts as independent
decision maker, balancing the threat to the public against the cost.
The power to shut down an operation has only been used once under the
-136-
-------
INDEMNIFICATION STUDY
Section IV
CWA, to temporarily halt operation of an FMC plant in West Virginia
that was leaking large amounts of carbon tetrachloride into the Kanawa
River. (It took three days to clear the request to apply for a TRO
even though the drinking water of large populations downstream was
seriously threatened.)
The Comprehensive Environmental Response, Compensation and
Liability Act (Superfund) provides a coordinated response to emergen-
cies caused by release of hazardous substances into the air, water, or
land. It authorizes direct action by the government in cleaning up or
containing a spill, as well as providing authority to issue a
presidential order or seek injunctive relief. Private parties
assisting in the cleanup may be reimbursed for their efforts under
certain conditions. Many of the features of the new law should help
to reduce unreasonable losses that might be Imposed on private parties
during environmental emergencies. Action based on the National
Contingency Plan, like action under a regulation, will not surprise
those whom it affects. It should also be less likely to be erroneous
or unfair since the NCP was developed over a decade with wide partici-
pation. The ability of the agency to take direct action should also
reduce burdens on private parties. Even in cases in which the govern-
ment later sues responsible parties for reimbursement, the more
leisurely determination of responsibility should prevent errors and
overbroad orders.
The ability of the agency to take direct action creates a new
source of liability: the losses that may be incurred by firms hired
to clean up hazardous waste sites or spills. Handling hazardous waste
is a dangerous and unfamiliar process, and potential contractors have
refused to undertake cleanups without some indemnification from the
government for their potentially large liability to third parties.
Because commercial insurance is not available in the quantities neces-
sary to cover potential losses, EPA has indicated that It will try to
indemnify contractors for losses above the required one million
dollars of commercial liability insurance. ®D The indemnification
clauses in the contracts are conditioned on the availability of funds.
4. Other Types of Emergency Resolution Provisions
Although Congress has provided a catch-all emergency mechanism in
each environmental statute to address Imminent and substantial harms,
there are also usually other, more specific emergency response provi-
sions in each statute. Often the agency can prevent or respond to an
emergency without Invoking emergency authority, for example through
its enforcement powers. In many cases a pollution emergency will
result from a release that violates a specific statutory prohibition
(e.g., § 301 of the CWA) or the terms of a permit* Therefore adequate
abatement authority will exist under the enforcement provision of the
statute. In such cases the underlying standards are usually clear,
and industry will have participated in the development of the regula-
-137-
-------
INDEMNIFICATION STUDY
Section IV
tions or in permit negotiations* Enforcement actions are subject to
internal review and provide opportunity to contest the action in
court. Thus, there should be less of a chance of error or unreason-
able action by the agency*
5. Conclusion
There does not seem to be a justification for further Indemnifica-
tion programs related to emergencies at this time. EPA has rarely
used its emergency authority in the past, and there is no reason to
believe that use of the conventional authorities will increase* A new
and comprehensive law governs response to emergencies generated by
release of hazardous substances* It appears to limit the posslblities
for unwarranted burdens on private parties. Until more experience
indicates a need for assistance beyond Superfund and the provisions
for clean up contractors, no new Indemnification program should be
recommended.
-138-
-------
INDEMNIFICATION STUDY
Section IV
F. UNREASONABLE ENFORCEMENT
1. Introduction
Under § 113(b) of the Clean Mr Act, and other similar statutes, a
person subject to unreasonable enforcement may recover attorneys fees
and other reasonable costs of litigation. He may also recover similar
costs associated with formal agency adjudicative actions under the
Equal Access to Justice Act (although the standards of recovery are
different). The purposes in such provisions are to encourage a person
to present a legitimate defense who would otherwise not do so because
of the expense involved, and to dissuade the agency from taking
unreasonable action. '
These provisions, however, stop short of remedying and discour-
aging unreasonable administrative action that will never proceed to
trial — perhaps precisely because the agency has determined that its
initial action was unreasonable. It is acknowledged at EPA that there
is potential for error and harrassment in administrative enforcement,
although they are thought to be infrequent and very rarely inten-
tional. Given the clear expression of congressional policy to remedy
and prevent unreasonable judicial enforcement by the use of an
attorneys' fees indemnity provision, a logical inquiry is whether
there should be a similar policy to remedy and prevent unreasonable
administrative enforcement (especially since experience indicates that
unreasonable judicial enforcement will rarely occur).
The following discussion explores the potential for unreasonable
administrative enforcement in EPA programs, explores the potential for
alternative remedies to Indemnification, and considers the feasibility
and impacts of such an indemnification program.
2. The Potential for Unreasonable Administrative Enforcement
The potential for "unreasonable" enforcement is inherent in any
administrative program that relies on enforcement to ensure implemen-
tation of standards. Almost all of the major statutory programs under
EPA jurisdiction contain substantial enforcement authority, and
Congress has traditionally authorized significant staff levels and
funds for enforcement purposes. In addition, many of the statutes
contain enforcement imperatives, and Congress has frequently reviewed
EPA*s enforcement record with an eye toward promoting more action.
The Term "Unreasonable"
What constitutes "unreasonable" enforcement is necessarily a
matter of interpretation. Many would probably agree if enforcement
amounts to intentional harrassment or if the agency has committed an
error in determining the existence of a violation or the applicability
of a regulation, or proceeded to enforce without a good faith legal
justification, that these actions would be "unreasonable."
-139-
-------
INDEMNIFICATION STUDY
Section IV
Undoubtedly, however, there are other circumstances which persons
would argue amount to unreasonable actions, but over which there would
be legitimate disagreement. Typical past complaints against EPA
enforcement policies include: (1) enforcement of technically or eco-
nomically infeasible requirements (where the Agency intends to Imple-
ment statutory mandates); (2) Inconsistent enforcement practices, for
example, in the assessment of penalties (where the Agency utilizes its
inherent discretion to enforce as it sees fit under the circum-
stances); (3) discriminatory selection of enforcement cases —for
example, the Agency may publicize an enforcement target list, as under
RCRA and CERCLA, that includes certain violators who may pose a less
significant environmental risk than others who are excluded from the
list (in such a case the Agency's enforcement interest may be based on
other factors, such as location, adequacy of evidence, visibility for
purposes of deterrence, etc.); and (4) the use of harsh bargaining
techniques (for example, the Agency may threaten to pursue all
available or particularly burdensome sanctions as leverage to obtain
acceptance of certain unfavorable terms in a consent decree).
For purposes of the current evaluation, the latter circumstances
are excluded from further consideration, and a narrow view of
"unreasonable" enforcement has been chosen. Such a view is consistent
with congressional consideration of the issue under § H3(b) of the
Clean Air Act and the Equal Access to Justice Act.50 Moreover, the
latter circumstances are almost always within the broad province of
enforcement discretion typically provided by Congress and accorded by
the courts.^
The Term "Enforcement"
Typically, an enforcement program embraces all aspects of regula-
tion design and implementation, from assistance in regulation drafting
(or review) to ensure enforceability, to the initiation and completion
of appropriate administrative and judicial proceedings. The list of
enforcement responsibilities in any one of EPA's enforcement programs
is quite long, Including such activities as preparing guidelines and
manuals; conducting workshops; reviewing state regulations and
enforcement actions; reviewing permit applications and compliance
reports; conducting inspections and other* compliance Investigations;
responding to private complaints and inquiries; preparing and issuing
informal compliance evaluations, formal notices, and abatement orders;
conducting formal and informal meetings and hearings for purposes of
due process, negotiation; assessing and collecting penalties; pre-
paring and negotiating consent decrees; assisting in litigation; and
numerous other activities. *
Of principal concern here are administrative enforcement pro-
ceedings that impose or could lead to the imposition of specific
sanctions. Inspections and other preenforcement investigations are
excluded,53 aa well as informal negotiations, and the numerous formal
-140-
-------
INDEMNIFICATION STUDY
Section IV
activities that involve rulemaking, permitting, and similar activities
(e.g., delegations of enforcement authority, approval and issuance of
delayed compliance orders and nonferrous smelter orders under the
Clean Air Act, response to State submissions and the issuance of NPDES
permits under the Clean Water Act, etc.)* Emergency actions have also
been excluded because they are discussed separately in this chapter.
In general, four types of administrative enforcement activities
have been given consideration: (1) notices of violation, (2) compli-
ance orders, (3) administrative penalties, and (4) the use of
blacklisting. The following table indicates which statutes contain
these four administrative enforcement techniques.
Administrative Enforcement Mechanism
Statute
CAA
CWA
RCRA
SDWA
FIFRA
TSCA
CERCLA
NOV
Authority^*
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Compliance
Orders
Yes
Yes
Yes
Yes
Yes
_—
Yes
Penalties
Yes
Yes
—
Yes
Yes
Blacklisting
Yes
Yes
a. Existing Constraints to Unreasonable Administrative Enforce-
ment
There are numerous factors that protect persons from unreason-
able administrative enforcement action by EPA. These Include statu-
tory enforcement restrictions; Agency concurrence and review
procedures; Agency policies favoring state enforcement leadership,
informal disputes resolution, and concentration on only the most
"significant" violators; as well as overall resource constraints. It
should be noted that these factors, either alone or In combination,
offer no assurance that unreasonable enforcement will not occur; and
they vary in significance from program to program. Nonetheless, they
offer a general perspective on the potential for unreasonable enforce-
ment at EPA and help to explain why there are few complaints relating
to EPA enforcement that are being raised by industry.
(1) Statutes
EPA's statutes include two types of constraints on
enforcement that would limit the occurrence of unreasonable action:
(1) provisions that limit EPA's enforcement authority; and (2) proce-
dural requirements that enable persons subject to enforcement to
influence the Agency's action.
-141-
-------
INDEMNIFICATON STUDY
Section IV
There are numerous statutory constraints on administra-
tive enforcement authority. Under § 113(a) of the Clean Air Act, for
example, EPA has interpreted the statute to require that administra-
tive orders provide for compliance within thirty days. If this is not
feasible, the Agency must pursue judicial action under § 113(b).
Thus, the flexibility to take formal administrative enforcement action
under the Clean Air Act is often restricted. Another type of restric-
tion exists under §§ 1414 and 1423 of the Safe Drinking Water Act:
EPA findings of noncompliance must be turned over to any state which
has been given primary enforcement responsibility by EPA; then, unless
EPA finds that the State has abused its enforcement discretion, EPA's
role is limited to advice and assistance.55 These, and similar con-
straints which exist under other statutes limit the potential for
unreasonable action by limiting the occurrence of administrative
enforcement altogether.
On the other hand, procedural prerequesites to enforce-
ment often permit or induce an independent check on the validity of an
action and allow for errors to be identified and resolved prior to the
occurrence of any significant adverse impact. For example, many of
the statutes impose notice of violation and conference or hearing
requirements prior to the issuance of an order or assessment of a
penalty;56 and special statutory hearings and appeals procedures exist
for certain enforcement actions." Where statutes do not provide for
these procedures, agency policies often do.58
(2) Policies
Other areas which might raise the potential for unreason-
able enforcement have been deemphasized or eliminated by current
agency policies. For example, under the pesticides law EPA has
adopted a policy emphasizing compliance assistance to industry as well
as state agencies and deemphaslzing independent enforcement.59 Shifts
of emphasis may also be noted under other programs, particularly under
the Clean Air Act, the Clean Water Act, and the Safe Drinking Water
Act.60
This is considered appropriate not only as a matter of
fiscal constraint and efficient resource allocation, but also as an
important element of the current administration's policy of "coopera-
tive" or "new" federalism. As an example of how this policy would be
Implemented in one program, a recent air pollution enforcement policy
memorandum instructs the Regional Offices to encourage states to take
the lead in enforcement against "significant violators," to offer
assistance, and to assume the lead only where a State cannot or will
not take the lead, despite whatever assistance EPA can provide.^
It is also apparent that the agency intends to concen-
trate its enforcement against predetermined "significant" violators.
Each program office has developed criteria for identifying such
-142-
-------
INDEMNIFICATON STUDY
Section IV
violators, and in some cases, lists have already been or are being
prepared. In general, this policy is an element of the Administra-
tor's new accountability system, intended to remove enforcement quotas
as a measure of performance and to instate criteria based on environ-
mental results.62 Guidance under the stationary source air program,
for example, provides that "significant violators" are existing faci-
lities located in or affecting non attainment areas with controlled
emissions exceeding a 250 tons per year potential, unless the magni-
tude and duration of the violation is minimal and non-recurring; new
and hazardous facilities are normally considered significant (again,
unless the violation is minimal and non recurring)." Identification
of facilities meeting these criteria and development of specific
enforcement strategies are to involve state participation.64
A coordinated approach toward identification of enforce-
ment targets In advance limits the potential for error associated with
hasty or negligent action. In addition, Administrator Gorsuch has
stated that a "non-confrontational" approach to enforcement is pre-
ferred — one that presumes good faith by the regulated industry and
emphasizes voluntary compliance.65 Such an approach may be Indicated
by the reduction in cases forwarded to the Department of Justice for
litigation since Gorsuch took offlee,66 as well as in recent EPA
efforts to involve industry in self-regulation programs.67
(3) Concurrence Procedures
To ensure that these national policies are observed, as
well as to ensure legal adequacy, internal enforcement concurrence
procedures are stressed. These procedures are especially rigorous
when litigation is involved.68 However, to preserve the option of
litigation, the litigation concurrence procedures also serve as an
incentive toward careful case consideration at the outset.69 For
administrative action they currently require that the Regional Counsel
approve the action as legally adequate and consistent with national
policy.^O Normally, the action must also be signed by an official
high enough within the managing hierarchy (e.g., Division Director) so
that several levels of review will have occurred before the action is
forwarded to counsel for concurrence.
(4) Overall Resource Constraints
It is possible that the potential for unreasonable admin-
istrative enforcement may also be influenced by enforcement resource
limitations. Fewer resources, for example, may result in more careful
case selection. Traditionally, EPA enforcement resources have per-
mitted only a limited number of enforcement cases, and these have been
chosen on a selective basis. Recently, moreover, as a part of the
overall administration initiative to conserve federal resources and
achieve a balanced budget, EPA's enforcement resources have been cur-
tailed even further.
-143-
-------
INDEMNIFICATION STUDY
Section IV
During F.Y. 1982, EPA's overall operating budget for
enforcement was reduced by 12% (reflecting efficiencies attained by
management and reorganization according to testimony by Gorsuch before
the Senate Public Works Committee in 1981).71 EPA's budget request
for 1983, moreover, Indicates that In certain program areas, more
reductions may occur.^ However, the Administration has indicated
that these proposed reductions for the most part will not result in an
actual decrease in enforcement activity from 1982 levels. Further, a
portion of the individual program decreases have merely been shifted
to a new Office of Enforcement Counsel, (within the recently estab-
lished Office of Legal and Enforcement Counsel) as a part of an
overall reorganization of enforcement operations. Thus, whether
currently planned reductions in enforcement resources will result in
more selective enforcement would be a matter of speculation.
b. Absence of Complaints
During the course of this study, substantial research was con-
ducted to determine the existence of actual complaints involving
unreasonable EPA enforcement. Enforcement officials currently with
the Agency (or who recently departed) were interviewed. Major
industry associations were consulted, and relevant reports and con-
gressional documents were reviewed.7* Although this research was
neither comprehensive nor designed scientifically to ensure statisti-
cal validity, very few Instances were identified in which complaints
pertained to unreasonable enforcement, as defined in the current
analysis. In fact, only in recent Senate Oversight Hearings involving
the Clean Air Act were complaints alleging specific losses due to EPA
enforcement actions raised.7* Even in these situations, however,
objections related to EPA's use of its legal discretion, not to
errors, negligence or action amounting to an abuse of process.
It is not possible to conclude from this research that no EPA
enforcement action to date warrants indemnification. To the extent
that such actions may exist, however, they have not been openly pub-
licized. Several respondents during the course of this study sug-
gested that the absence of Industry complaints may reflect concern
over bad publicity and fear of compounding existing agency relation-
ship problems by raising the issue.
On the other hand, testimony does appear throughout the past
several years criticizing EPA's failure to conduct adequate enforce-
ment. This is particularly noticeable in the 1981 enforcement over-
sight hearings, as well as in recent oversight hearings on the
implementation of the Clean Water Act, the Resource Conservation and
Recovery Act, and Superfund;75 and in legislative hearings on the
budget and the Clean Air Act reauthorization.7^
-144-
-------
INDEMNIFICATON STUDY
Section IV
3. Blacklisting
An area of concern frequently raised in interviews involved
the agency's blacklisting authority. Under the Clean Air and Clean
Water Acts,' no federal agency may engage in a contract with a person
who is convicted of a criminal violation (unless the President provi-
des a special exemption) until EPA certifies that the condition giving
rise to the conviction no longer exists. Expanding upon this author-
ity, Executive Order 11738, § 4,78 requires that all federal procure-
ments include provisions requiring compliance with the Clean Air and
Water Acts. Accordingly, EPA has established blacklisting criteria
under 40 C.F.R. § 15.20(a)(l) that include not only criminal convic-
tions, but other instances of non-compliance.
Blacklisting, of course, is intended to interfere with a
company's ability to obtain new procurements and thus may have a cal-
culable economic impact that would vary depending, among numerous
factors, on the amount of federal procurements anticipated by a given
company, the timing of these procurements, the Investment already
incurred, and the time period necessary to obtain delisting. The
stigma attached may also be more serious than that associated with
other administrative enforcement actions, since, in contrast to most
administrative and judicial enforcement cases, the list Is published
in the Federal Register, and the Administrator's listing determina-
tion would therefore be more widely publicized.
In general, however, this procedure has not often been
utilized. In part this is because by the time a criminal conviction
occurs, the underlying violation has been corrected; and most civil
actions are resolved by settlement or consent decree. In addition,
the agency rarely considers listing except in the context of exascer-
bated cases or repeated offenses, and the opportunity for error is
substantially limited in those cases. Moreover, listing does not
occur automatically because a facility is in violation. Rather, it
must be the result of a specific recommendation (by certain EPA
officials, governors or the public), followed by an independent EPA
case examination, which includes an opportunity for the owner of the
facility to present evidence, and it is ultimately effected by a
"listing review panel."80
4. Alternative Remedies and Prevention Mechanisms
If unreasonable enforcement actually occurs, and relief within the
agency is not available, a lawsuit to prevent further such action
would be the most logical alternative. However, if losses have been
incurred because of the initial agency action, they probably.cannot be
recovered. Further, unless the party qualifies as a small business
under the Equal Access to Justice Act, it may not be able to recover
court costs for his successful lawsuit. Compensation for other
losses, such as lost business, is not available in any case. In addi-
-145-
-------
INDEMNIFICATON STUDY
Section IV
tion, If the agency is not obligated to pay compensation, then one of
the primary purposes in current legislation may not be fulfilled — to
deter unreasonable action by forcing the agency to pay — although the
lawsuit, itself, may be an adequate deterrent In this regard.
Nonetheless, It is possible that injury may be avoided alto-
gether through judicial review. However, courts have often been
unwilling to allow suits at an early stage of the enforcement process,
on the theory that review should be limited to final agency action, as
defined under the Administrative Procedure Act or the specific statu-
tes involved.81 However, not all courts agree that review must await
some later stage of enforcement, if a demonstration of immediate, harm-
ful impact can be made: and the trend appears to be more toward flexi-
bility on this issue.82 Judicial review will usually be permitted if
it can be demonstrated that administrative remedies are inadequate,
futile, will result in irreparable injury, or would be void.8''
5. Feasibility and Impacts
If there is a right to such compensation for unreasonable adminis-
trative enforcement, or even if the indemnity is subject to court
discretion, an indemnification program could invite lawsuits at any
stage of administrative procedures prior to their completion, and thus
supplant administrative enforcement discretion entirely — or, at a
minimum, require a judicial proceeding on the issue of indemnification
prior to completion of the agency proceeding. This illogical result,
along with the high potential for chilling administrative enforcement
activity would argue for making an award only at the completion of the
agency proceeding. A more appropriate course would be to ensure the
availability of enforcement review whenever agency action, regardless
of the stage, poses a significant threat of immediate Injury, and a
facial demonstration is made that the agency action is erroneous or
amounts to an abuse of process.
A second important consideration relates to the amount of compen-
sation. Fees associated with enforcement defense are readily calcul-
able if they involve retaining outside experts or if the defendant has
hired such experts for the purpose of defending the case. However, if
inside staff are used, it may be difficult to show that an unantici-
pated expense has occurred as the result of unreasonable enforcement.
Lost profits and other opportunity costs may prove to be too specula-
tive. Again, the most logical approach would be to ensure that a
cause of action exists in clearly warranted cases so that judicial
review may be obtained prior to incurring business losses.
Finally, one should balance the need for compensation and the
interest in deterring unreasonable enforcement against the strong con-
gressional policy toward enforcement under most EPA statutes.
Especially in the absence of any evidence that unreasonable enforce-
ment occurs on more than isolated occasions, and in the presence of
-1A6-
-------
INDEMNIFICATON STUDY
Section IV
evidence that all enforcement will be subject to close scrutiny, it
may not be reasonable to establish unnecessary legislative constraints
to EPA enforcement.
6. Conclusion
An expanded indemnity program to compensate persons subject to
unreasonable administrative enforcement is probably not warranted.
This is primarily because there is no evidence that unreasonable
enforcement exists on more than a very infrequent basis, and because
EPA's current enforcement policies make such occurrences unlikely.
However, it cannot be stated with certainty that unreasonable adminis-
trative enforcement will not occur.
The most effective approach toward resolving potential problems of
unreasonable enforcement would be to ensure the availability of judi-
cial review (to the extent it is not already available) in cases where
the Agency has clearly engaged in abusive or erroneous activity. This
would preserve existing compliance incentives on the part of the
affected industry (and Incentives to take precautions against agency
errors), and it would preserve existing enforcement Incentives on the
part of the Agency at a time when there is little evidence of agency
abuse and substantial evidence of agency restraint. Any new indemni-
fication program should be established only when a pattern of error
and abuse is clearly discernible, and it is apparent that alternative
loss avoidance mechanisms are inadequate.
-147-
-------
INDEMNIFICATION STUDY
Section IV
G. CHANGES IN AGENCY POLICY
1. Description of the Problem
Stability and predictability of government action is Important to
regulated parties. Abrupt or unexpected changes in government policy
can disrupt planning and cause losses beyond those attributable to the
new policy itself. Furthermore, since an industry is rarely homo-
geneous, a change in policy can affect the relative competitive posi-
tions in ways that may not be intended or even foreseen by the Agency.
While the usual effect of a change of policy often times is much the
same as the effect of an original policy (i.e., the cost of complying
with a justified governmental regulation), the issue Is a topic of
considerable concern to Industry and some academics, and has been
raised in so many of our discussions that it must be addressed.
Nevertheless, we did not find, and none of our contacts brought to our
attention, any actual case of a change in policy causing a loss that
might justify indemnification.
There are two theoretical fact patterns that bear discussion. The
first is a situation in which a company must comply with a new or
changed regulation at the same time that it is challenging the regula-
tion in administrative or judicial review. Although a prevailing
party may be able to collect court costs and attorney's fees, no sta-
tute would entitle it to reimbursement for the costs of Interim
compliance with a regulation which is later overturned. The second
situation need not Involve an erroneous or arbitrary change in policy.
It might happen that even a fully justified change could impose an
unreasonable cost on a segment of the industry. For example, suppose
that the Agency changed from a relatively more stringent standard to a
relatively less stringent standard for emissions. Those who had
acquired equipment or otherwise invested to comply with the first
standard might well find that the investment was lost. Thus, those
who had been most diligent in complying would have the greatest loss.
This might also happen as standards become more stringent, if equip-
ment used to meet the old standard would not work for the new one.
2. Evaluation
Although immunity of the government and its officers has been
steadily strinking over the last forty years, there are still some
core activities protected from tort liability. Broad decisions to
adopt particular standards or regulations or other decisions calling
for balance of policy considerations are discretionary judgments
within the exemption from the Federal Tort Claims Act.8* The reasons
for protecting such policy decisions go to the very structure of our
government: maintaining the independence of the three branches. It
is not the proper role of the judiciary to remake the decisions of the
executive. Political questions are not considered amenable to testing
through tort law because "objective standards are notably lacking when
-148-
-------
INDEMNIFICATON STUDY
Section IV
the question Is not negligence but social wisdom, not due care but
political practicability, not reasonableness but economic exped-
iency. "1" Changes in policy are by definition part of this category of
protected government activity. Thus, any recommendation for Indem-
nification In this area either has to overcome a heavy presumption
against it, or find a way around the separation of powers problem.
The factors for analysis that were identified in Section III are
useful in evaluating the potential Inderaniflability of losses caused
by changes in policy.
Alternatives to indemnification can be used to reduce or spread
any cost of changes in regulation. Advance notice and phased-in
compliance can allow those involved to make an orderly transition.
Participation of a representative sample of all firms in an industry
during rulemaking could help avoid unwitting impacts on competitive
relationships that might result if the Agency did not understand all
the variations in the industry. Exceptions and waivers can permit
flexibility in application of the new regulation to individual
situations. Most of the statutes administered by EPA have some relief
by means of variance or exception for small businesses or those to
whom the regulation was unfairly applied. Of course, some costs will
remain no matter how long the transition period, or flexible the
implementation. Whether these are beyond the cost of compliance can
only be determined from an examination of the particular facts of a
specific situation.
Another powerful alternative to indemnification for changes in
policy is review of the decision by administrative or judicial
tribunals. The remedy sought is not money damages but modification or
retraction of the offending policy action. This may explain, in part,
why we found so few actual losses connected to changing policies. If
a projected loss really is severe or unreasonably allocated, it can
often be prevented by obtaining a stay of the action while the review
is ongoing.
An abrupt change might constitute a procedural inequity, if made
in the absence of compelling public health reasons, but the government
rarely acts abruptly. Even the emergency decision to cancel some uses
of pesticides containing 2,4,5-T and Sllvex came at the end of eight
years of regulatory action and study, Including an earlier can-
cellation of home and aquatic uses of 2,4,5-T (later withdrawn).
Though the manufacturers of the pesticides were surely aware of the
continuing Agency concern and new developments in the scientific
understanding, and thus the potential for regulation, it is possible
that the sellers and users of the pesticides were not prepared. This
points up a possible substantive inequity: that cost of the govern-
ment action might fall most heavily on those least responsible, and
least able to bear It. This would depend upon an analysis of the
case. Since a ban is the most drastic action the Agency can take, the
more common and less onerous actions would have less severe impacts.
-149-
-------
INDEMNIFICATON STUDY
Section IV
Changes In policy are often based on explicit Congressional
direction. Either the law itself is amended, or the law specifies
that the Agency must revise standards in light of new knowledge. In
the face of such clear directions, it would be difficult to justify
indemnification if Congress has not already authorized it. Occasion-
ally the question of compensation is addressed directly* After the
de-regulation of the trucking industry, the tax code was amended to
allow deductions for the book, value of trucking certificates rendered
useless by the de-regulation. This only partially compensated those
who had held the certificates for a long time and those whose taxes
were low. This treatment indicated that Congress regarded the loss as
a cost of doing business rather than an extraordinary loss.
The ability of the Ageiccy to balance policy considerations objec-
tively could be hampered significantly by the possibility of having to
pay for the costs. This is one of the reasons for exempting policy
decisions from the Federal Tort Claims Act. If, however, Congress
should decide that the cost of some change in policy ought to be born
by the taxpayers rather than by some segment of the regulated
industry, this is a different matter. It would be a political deci-
sion by those most accountable, and would be separate from the policy
decision of how best to implement a statute.
3. Conclusion
Change in policy is a category almost as broad as all Agency
action, and it is difficult to make generalizations about it.
However, it is clear that for most changes there is no extraordinary
loss suffered. The alternatives to indemnification work well here,
either to avoid the loss through participation in the creation of the
decision to change by exemption, or by judicial review and modifica-
tion of the policy.
-150-
-------
INDEMNIFICATION STUDY
Section IV
FOOTNOTES TO SECTION IV
surveys and reports reviewed were the following: ERT, "The
Effects of the Clean Air Act on Industrial Planning and Development"
(for the Business Roundtable); Vice President Bush's Regulatory Relief
Task Force survey and responses; The Small Business Hotline.
^This was not a thorough or scientifically designed survey of all
industries that deal with EPA. Instead, it was a series of informal
conversations in which we attempted to ascertain thoughts about indem-
nification and to learn about losses that the agency might have
caused. The first conversations were quite general. As we reached
people who were knowledgable about a particular event, or Interested
in a particular area they became more specific. Often, persons with
whom we spoke were not willing to commit their firms or organizations
to particular positions; and many requested that they not be quoted by
name. At this writing several persons have Indicated interest in a
further opportunity to provide input after reviewing a written report.
^Three Freedom of Information Act (FOIA) bills are before the
Congress. The Reagan Administration bill is H.R. 4805 and S. 1751.
Senator Orrin Hatch (R.-Utah) has introduced S. 1730, which is similar
to the Administration bill. S. 1730 has been approved by the Senate
Subcommittee on the Constitution and, as of May 19, 1982, was still in
markup before the Senate Judiciary Committee. According to a tele-
phone conversation on March 22, 1982, with Mr. Randy Radar, subcom-
mittee staff person, the focus of S. 1730 is on defining what is and
what is not confidential business information under the FOIA. Because
it is felt that the current law does not clearly define confidential
business information, the subcommittee has attempted to develop more
reliable standards for this determination. The definition of CBI
would affect what can and cannot be disclosed under the FOIA.
4H.R. 5203, The Federal Insecticide, Fungicide, and Rodentlcide Act
Amendments of 1982, was reported out of the House Agriculture Commit-
tee on May 17, 1982.
5Id. under Section 5(7) a new subsection (h) entitled "Civil
ActiorT^would be added to § 10 of FIFRA. Additionally, section 7 of
H.R. 5203 would amend § 14(b)(3) to provide criminal penalties to pri-
vate parties and federal employees and contractors for unauthorized
knowing or willful disclosure.
6E.g., Chevron Chemical Co. v. Costle. 641 F.2d 104 (3d Cir.
1981), cert, denied, 101 S. Ct. 3110 (1981) (concerning §§ 3(c)(l)(D)
and 10 of FIFRA); Mobay v. Costle, 517 F. Supp. 252 (W.D. Pa. 1981),
which expressly followed the holding of the Chevron case; Union
Carbide Agricultural Products Co. v. Gorsuch, (S.D.N.Y.., No. 76 Civ.
2913) (preliminary injunction against EPA issued by the district court
vacated on appeal by the Second Circuit); and Petrolite Corp. v. EPA,
-151-
-------
INDEMNIFICATON STUDY
Section IV
519 F. Supp. 966 (D.D.C. 1981) (EPA's motion for summary judgment
granted).
Due to the erroneous FOIA release of an EPA review of data on
Monsanto's herbicide glyphosate in May 1982, the Agency in August 1982
agreed to special procedures designed to protect Monsanto's interests.
Specifically, any future requests for registration of herbicides con-
taining active ingredients will be submitted to EPA's Scientific
Advisory Panel for a judgment whether the application is based upon
Monsanto data improperly disclosed by the Agency. The new procedures
to which EPA agreed were set forth in a court judgment issued by the
U.S. District Court for the Eastern District of Missouri, Monsanto v.
Gorsuch, No. 79-0366-0(1). See "Current Developments" at 758, BNA
Environment Reporter (Oct. 1, 1982).
^^^^^^••^^•^•••^^••••^•^H^^MKH^HM^^H k
7Telephone conversation on March 22, 1982 with James Nelson, EPA
Office of General Counsel.
°See TSCA Legislative history on CBI issue.
Q
E.g., U.S. Environmental Protection Agency, TSCA Confidentla
Business Information Security Manual (Office of Toxic Substances,
October 1981).
10TSCA § 14(d) contains a misdemeanor provision which applies to
any current or former United States employee, officer or contractor.
FIFRA § 10(f) contains a similar provision. RCRA § 3007(b)(2) and
CERCLA § 104(e)(2)(B) contain criminal sanctions applying to any per-
son (such as a state official) not subject to 18 U.S.C. § 1905.
HTelephone conversation on March 8, 1982 with Mr. James T.
O'Reilly, lecturer in law at the University of Cincinnati and senior
counsel for Proctor and Gamble Co. O'Reilly was particularly con-
cerned about the progressive nature of employee discipline sanctions
and the fact that the removal sanction is not usually applied until
the third offense. It should be noted, however, that "in unusual cir-
cumstances greater or lesser penalties may be applied unless otherwise
provided by law." TSCA Security Manual, supra note 9 at 31 (Appendix
I).
12Id. O'Reilly argued that non-use of 18 U.S.C. § 1905 Is indica-
tive of its lack of deterrent effect. Also, he noted that § 9-2.025
of the U.S. Attorney's Manual dated May 1979 requires that any prose-
cutions under § 1905 must be cleared by Department of Justice offi-
cials in Washington, D.C. O'Reilly contended that this restricts the
use of § 1905. An attorney at the Department of Justice, however,
emphasized that there is no policy not to use § 1905 and that the U.S.
Attorney's Manual provision is informationally oriented and probably
designed to permit national coordination on the use of § 1905. Tele-
phone conversation on March 10, 1982 with Mr. Frederick Hess, Acting
-152-
-------
INDEMNIFICATON STUDY
Section IV
Director of Legal Support Services (dealing with FOIA Matters), U.S.
Department of Justice. Another Justice Department attorney felt that
it would be speculative to conclude that use of § 1905 would be
restricted because of the U.S. Attorney's Manual requirement. Tele-
phone conversation on March 10, 1982 with Mr. Alan Carvtr, Public
Integrity Section (handling 18 U.S.C. § 1905), U.S. Department of
Justice.
^Telephone conversations on February 25, 1982 with Mr. Charles
Breece, EPA Office of General Counsel; on March 2 and 22, 1982 with
Mr. James Nelson, EPA Office of General Counsel; on March 22, 1982
with Mr. John Euller, Assistant Director, Civil Division, U.S.
Department of Justice; and on March 22, 1982 with Mr. James Klapps,
Torts Section, Civil Division, U.S. Department of Justice.
c
Id. Nelson, Hess, and Carver; and telephone conversations on
March TT7 1982 with Mr. Larry Swales, Security Officer, EPA Office of
Toxic Substances; and on March 23, 1982 with Mr. James Conn, Divi-
sional Office for the Mid-Atlantic Region, EPA Inspector General's
Office. The only known Intentional unauthorized disclosure of CBI may
have occurred in the area of EPA contractors voluntarily submitting
information needed for the competitive bidding process. Apparently,
EPA employees may have disclosed information to a contractor's compe-
titor. In response to "blistering" letters received from irate con-
tractors, EPA has undertaken careful investigations to determine
whether criminal prosecution might be warranted. So far, the evidence
has been insufficient to warrant criminal prosecution, although cer-
tain cases have in the past been referred to the Department of Justice
for consideration. Conn, id.
150'Reilly, supra note 11. Mr. O'Reilly claims that instances of
disclosure and loss have occurred and that such incidents have been
described to him "off the record" by affected parties.
17<
example, H.R. 5203, supra note A, permits limited disclosure
of health and safety test data to environmental groups, but then pro-
vides for substantial civil liability if that information is then
passed on improperly to a competitor of the party whose CBI is
disclosed.
example of statutory authorization where this problem might
arise is found under TSCA § 14(a)(3) which permits disclosure "if the
Administrator determines it necessary to protect health or the
environment against an unreasonable risk of injury to health or the
enviornment." This provision, however, has not yet been used.
-153-
-------
INDEMNIFICATON STUDY
Section IV
20Jayson, Handling Federal Tort Claims § 260.06 (Matthew Bender,
1964 and current supplements) states at 13-96:
The breadth of Section 2680(h)'s exclu-
sion of "any claim arising out of * * * inter-
ference with contract rights" is unclear
largely because the common-law concept of the
tort of Interference with contract rights is
not a clear-cut one. Interference with
contract rights may be but one aspect of tor-
tious conduct In the field of economic
relations. The broader concept includes a
variety of torts and improper conduct
comprising the body of law loosely described
as interference with business relations. The
Restatement of Torts, in discussing this
general area, deals with such subjects as
engaging in business for the purpose of
causing loss of business to another, fraudu-
lent marketing, trademark infringement,
disclosure of trade secrets, refusal to deal
with another, inducing breach of contract or
refusal to deal, and labor disputes, to men-
tion but a few.
Whether any or all of these other,
closely related torts are also embraced within
the Tort Claims Act phrase "interference with
contract rights" is still unsettled. The
legislative history of Section 2680(h) provi-
des no answer. Prosser states that there is a
definitely marked out tort "under the name of
inducing breach of contract, or interference
with contract." And\the Restatement sets out
a general principle of liability under the
heading "inducing breach of contract or refu-
sal to deal." It should be noted, however,
that the phrase used In the Tort Claims Act is
not precisely the same as that used by those
two leading works on torts, (emphasis added).
The FTCA's exclusion of claims arising out of "interference with
contract rights" is susceptible of covering not only interference with
existing contractual relations but also interference with precontrac-
tual relations or prospective advantage. There has been a lack of
unanimity among the cases on this question. Dupree v. United States,
264 F.2d 140 (3d Cir. 1959), a case involving the alleged negligent
withholding of a security clearance by the Coast Guard, held that the
exception covers interference with prospective advantage, as well as
existing contractual relations. But the Ninth Circuit in Builders
-154-
-------
INDEMNIFICATON STUDY
Section IV
Corp. of America v. United States, U8 F. Supp. 482 (N.D. Cal. 1957)
rev'd, 259 F.2d 766, 769 (9th Cir. 1958), indicated doubt about this
conclusion. See also Black v. United States, 389 F. Supp. 529 (D.D.C.
1975), appeal sub, nom, Black v. Sheraton Corp. of America, 564 F.2d
531, 540-41 (D.C. Cir. 1977), In which the Court of Appeals expressly
rejected the government's contention that the interference with
contract exception was applicable in a case Involving invasion of pri-
vacy and the dissemination of information to the effect that the
claimant was connected with organized crime. Another relevant case is
Quinones v. United States, 492 F.2d 1269 (3d Cir. 1974), in which a
former government employee was allowed to bring an action under the
FTCA based on the alleged failure of the government to use due care in
maintaining his personnel records, resulting in damage to his reputa-
tion. In both Black and Quinones the courts refused to stretch excep-
tion (h) beyond its specific coverage and allowed the claimants to
maintain their causes of action.
2^0'Reilly, supra note 11, noted in particular that tort law in
the District of Columbia is not highly developed. He also thought
that proof of the quantum of damages would be particularly difficult.
22
Supra note 19.
primary reason for expressly subjecting the government to suit
for either negligent or intentional unauthorized disclosures would be
to affect agency behavior by prompting it to be more careful in its
handling of CBI. With regard to criminal sanctions, the burden of
proof is much higher and so the impact on the agency and its employees
may be less with criminal sanctions than with civil tort liability.
The product liability area is frequently cited as one in which tort
liability may significantly affect behavior and prompt more careful
actions. Telephone conversation on May 19, 1982 with Dr. Nickolas
Ashford, Center for Policy Alternatives, Massachusetts Institute of
Technology.
example, one such general problem may be the lack of a
requirement in FOIA that notice be given to an information submitter
prior to its release by an Agency. O'Reilly, supra note 11, discussed
this problem with regard to the FDA in his testimony on July 16, 1981
at Freedom of Information Act oversight hearings before a subcommittee
of the House Government Operations Committee. See hearings at 596,
599, 609-10.
note 6.
26Environmental Defense Fund, Inc. v. Goreuch, 17 E.R.C. 1099
(D.D.C. 1982). EPA was ordered to notice the proposed National
Contingency Plan revisions within 30 days from the date of the court's
order on February 12, 1982. EPA met this requirement on March 12,
1982 (47 Fed. Reg. 10472). The public comment period closed April 28,
-155-
-------
INDEMNIFICATON STUDY
Section IV
1982 and EPA now has until July 16th to publish the revised National
Contingency Plan. See also NRDC, Inc. v. Train, 510 F.2d 692 (D.C.
Cir. 1975) (action to compel publication of effluent limitation guide-
lines under the Federal Water Pollution Control Act).
27gee, e.g., the EPA Office of Policy & Resource Management cata-
loging"^? of September 1, 1981 of industry responses to V'ice President
Bush's letter to Industry groups soliciting comments to the Cabinet-
Level Task Force on Regulatory Relief, especially at A-7, p. 2
(Prevention of Significant Deterioration — Set I Pollutants: recom-
mendation of mandatory deadlines In dealing with applications and
reviews); at B-5, pp. 1-2 (NPDES and Consolidated Permits Programs:
claims that "regulations placed great and unjustified burdens on
industry with specific concerns about delays in processing applica-
tions and issuing permits; recommendation to revise regulations to
establish reasonable time limits within which the entire permitting
process must operate; recommendation to have government agencies work
together to streamline permitting functions); at B-9 (Construction
Grants: recommendation tthat states should be delegated full review
responsibility because EPA review of every stage of project causes
extensive paperwork and delay; and at E, pp. 4-5 (Pesticide Programs
Under FIFRA: recommendation to establish procedures for reviewing
data more quickly in the pesticide registration process, with specific
deadlines for review and conditional granting of registration if EPA
does not meet the deadline). See also the Business Roundtable report,
infra note 31.
28M. Hamilton, "The Permit Explosion: Siting New Energy Facil-
ities In the Western United States," (Dep't of Political Science,
Colorado State University, Ft. Collins, Colorado; Presented to the
State Energy Permitting Workshop of the Western Governor's Policy
Office and the New Mexico Dep't of Energy & Minerals, Albequerque,
N.M.; Dec. 16-17, 1980).
See also Urban Systems Research & Engineering, Inc., "Causes of
Delay for Major Industrial Projects," (Draft Final Report prepared for
the Council on Environmental Quality, Jan. 11, 1981), which reviewed
the existing literature and attempted to provide a more comprehensive
look at the effects of regulations on project schedules than had been
done by previous studies. The Urban Systems analysis involved 41 case
studies of non-nuclear energy projects and non-energy projects. In
over half of the cases, project delays were attributable to a variety
of factors. The remaining cases "were equally divided between cases
where regulatory factors alone were responsible for the delays and
cases where economic factors alone were responsible. The availability
of labor and equipment was not a significant cause of delay. The
study concluded that "while the regulatory delays can to some extent
be managed (and minimized), they are often out weighed by economic
difficulties which are external to the project management."
-156-
-------
INDEMNIFICATION STUDY
Section IV
memorandum entitled "Briefing on Permitting, 9/1/81" from
Stuart Sessions, Chief Energy Facilities Branch, Office of Policy
Analysis, OPRM to Nolan Clark, Associate Administrator for Policy &
Resource Management. More current information provided by Mr.
Sessions as of April 30, 1982 indicated from a sample of 28% of the
recently Issued PSD permits that the average time from the application
submittal to permit issuance was 10.5 months and from the time of sub-
mission of the last information to permit Issuance was 4.9 months. It
was noted that the submission of the last information is often later
than when the application is declared by EPA to be complete. Thus,
"the most complex permit processing takes less than a year." Infra
note 30.
•^Telephone conversation on March 12, 1982 and interview on April
27, 1982 with Mr. Stuart Sessions, e Chief , EPA Energy Facilities
Branch, Office of Policy Analysis, OPRM. In 1980, EPA established a
specific branch in the Office of Policy Analysis for the purpose of
evaluating possible improvements to the permit-processing system.
See also text accompanying note 32 infra.
^Environmental Research & Technology, Inc., I & II Effects of the
Clean Air Act on Industrial Planning & Development; Analysis of Case
Studies (The Business Roundtable Air Quality Project) (July 1981).
See also EPA memorandum entitled "Review of BRT Report on 'Effects of
the Clean Air Act on Industrial Planning & Development," from Stuart
Sessions, Chief Energy Facilities Branch, Office of Policy Analysis,
OPRM to Kathleen Bennett, Assistant Administrator for Air, Noise &
Radiation (Oct. 27, 1981).
32Supra note 30.
33Supra note 30 and note 29, at 5-6.
has established a work group to conduct an Agency-wide eval-
uation of whether or not various fees might be appropriately levied.
A report is anticipated by mid-summer 1982. Supra note 30.
^Although § 165(c) of the Clean Air Act contains a mandatory
requirement that PSD permits "shall be granted or denied not later
than one year after the date of filing of [a] completed application,"
no mandamus action has ever been brought to force Agency action on a
PSD permit application. Nevertheless, such an action could be
brought. Interview on June 2, 1982 with Mr. Peter Wyckoff, EPA Office
of General Counsel (Handling PSD Permits).
36Appendix B, note 54.
37Appendix A, page A-20.
™U.S. General Accounting Office, Gaines & Shortcomings in
Resolving Regulatory Conflicts & Overlaps, Report to the Congress by
-157-
-------
INDEMNIFICATION STUDY
Section IV
the Comptroller General of the United Stated (PAD-81-76; June 23,
1981).
39Id. at 58.
40Id. at 60-68.
41Id. at 68-79.
42Id. at 44.
^Description generally based on Rendleman, "Legal Anatomy of an
Air Pollution Emergency," 2 Environmental Affairs (Boston College) 90
(1972); and Hardy, Pate, ejt al^, "First Use of the Federal Clean Air
Act's Emergency Authority, "A Local Analysis,1" 64 Amer. J. Public
Health 72 (1974).
44Hardy, Pate, id. at 74.
45United States v. U.S. Steel. CA 71-1041 (N.D. Ala., Nov. 19,
1971), Transcript of Hearing before Judge Pointer, quoted in
Rendleman, id. at 100.
^Telephone conversation with Fournier J. Gale, III, on January
28, 1982.
A7H.R. Rep. No. 90-728 at 19; S. Rep. No. 91-1196, 91st Cong., 2d
Sess. (1970).
an air alert was declared the following June, all regula-
tions and phased curtailment plans were in place. Inspections showed
that sources were complying with the plans, and that the particulate
count decreased markedly when the abatement plans went into effect.
Hardy, Pate, id. at 76.
48aTelephone conversation on May 25, 1982 with Jeff Miller, former
EPA Acting Assistant Administrator for Enforcement.
48bTelepnone conversation on May 26, 1982 with Pamela John in
Contracts Management Division of EPA.
discussion supra at 66-71.
discussion of enforcement discretion in City of Seabrook v.
Costle, 16 E.R.C. 1676 (5th Cir. 1981).
detailed description of current and planned enforcement
activities may be found in prepared materials supporting the Adminis-
-158-
-------
INDEMNIFICATON STUDY
Section IV
trations budget proposal for F.Y. 1982 and 1983. The F.Y. 1982
materials have been reprinted in an abbreviated form in SNA:
Environment Reporter, Federal Laws 51:1601-16-45. The F.Y. 1983
materials are available at EPA in an unclassified publication
entitled, "Justification of Appropriation Estimates for Committee on
Appropriation, Fiscal Year 1983."
boundaries of protection relating to inspections and other
investigations are well established under the Fourth Amendment. See,
e ; g . , Public Service Company of Indiana v. EPA, 15 E.R.C. 1939 (S.D.
Ind. 1981); EPA's inspection policy as stated In a memorandum to the
Regional Administrators from the Assistant Administrator for Enforce-
ment on "Conduct of Inspections After the Barlow's Decision" (April
11, 1979), reprinted in BNA: Environmental Reporter, Federal Laws
41:2451. Restrictions also exist in the individual pollution control
statutes and under the Federal Torts Claims Act. While Agency inspec-
tions may sometimes be inconvenient and occasionally cause injury, to
the extent that these impacts are unjustifiable, they may normally be
avoided by court action. To the extent that the inspection results
may be erroneous, EPA must enforce on the basis of the erroneous
results before any actual Injury would occur —thus, it is the enfor-
cement action,, rather than the inspection, that merits primary concern
in the current study.
authority may be express or implied. In general, the agency
provides notice of a violation prior to issuing an abatement order or
proceeding to litigation even when the statute does not require such
notice.
55See also §§ 26 and 27 of the Federal Insecticide, Fungicide and
Rodenticide Act.
56See, e.g., § 113 of the Clean Air Act and § 309 of the Clean
Water Act.
57See, e.g., § 120 of the Clean Air Act.
note 52; see also blacklisting procedures stated in 40
SuPra note •>*; see also blaclc;
.F.R. §§ 1520 et seq., Appendix A.
C
5"in the recent administration's budget proposal, EPA indicated
: .at the pesticides enforcement program would shift in emphasis to
compliance assistance to both industry and the states. EPA document
entitled "Environmental Protection Agency, Justification of Appropria-
tion Estimates for Committee on Appropriation, Fiscal Year 1983" at
P-48-49. This is consistent with §§ 26 and 27, which provide the
States with primary enforcement responsibilities.
60Id. at A-110-111; WQ-114; DW-49; under the toxics and Superfund
legislation there is no statutory analog to the State implementation
and delegation programs that exist under other statutes.
-159-
-------
INDEMNIFICATON STUDY
Section IV
"^•Memorandum to Regional Administrators from Kathleen Bennett
(December 29, 1981). Delegation to the states is the Administrator's
top priority management goal for F.Y. 1982. "Adminstrator's Manage-
ment Goals for F.Y. 1982," reprinted in Senate Oversight Hearing,
October 15, 1981, No. 97-H29, 97th Cong., 1st Sess., at 85.
62See Testimony of Administrator Gorsuch, Senate Oversight
Hearing, October 15, 1981, No. 97-H29, 97th Cong., 1st Sess. at 34.
63
64
Bennett memorandum, supra note 61.
Id. See also draft memorandum from EPA Administrator Anne M.
Gorsuch on General Operating Procedures for Civil Enforcement Program,
reprinted in BNA: Environment Reporter, Current Developments (May 21,
1982) at 78, 82. 0
See £NA: Environment Reporter, Current Developments (May 14,
1982) at 38.
^Senate Oversight Hearing supra note 62 at 33.
"'A substantial effort is underway to design and implement self-
auditing programs within the various regulatory media. Interview with
Karen Blumenfeld, Regulation Reform Branch, Office of Policy and
Resource Management, EPA. See also proposed self-certification rules
for PCB generators, proposed amendment to 40 C.F.R. part 761,
§ 761.185, 47 Fed. Reg. 24976-88 (June 8, 1982).
68See discussion supra, Section III note 117.
"'Gorsuch Draft Memorandum, § VI, supra note 64.
70EPA concurrence procedures have undergone several changes during
the past year. See Memoranda from Enforcement Counsel to Regional
Administrators (December 29, 1981 and February 26, 1982); Gorsuch
Draft Memorandum, supra note 64. As currently stated, these proce-
dures provide headquarters offices with the principal role in defining
national enforcement policy, but give the Regional Offices the primary
decision-making authority regarding .the Initiation of administrative
action; no headquarters concurrence is normally required for such
actions unless a new issue of national policy is raised. This rule is
not absolute, however. Variations exist among the different programs,
depending on prior guidance (and the need to maintain active headquar-
ters supervision). For example, procedures involving enforcement of
RCRA and CERCLA provide for headquarters participation in most aspects
of administrative enforcement activity involving inactive sites —
from the listing of priority sites to the issuance of administrative
orders. See Gorsuch Draft Memorandum § IV supra note 64, at 82;
Memorandum to Regional Administrators from Chlstopher J. Capper and
William A. Sullivan, Jr. (February 23, 1982) at 8.
-160-
-------
INDEMNIFICATON STUDY
Section IV
^Senate Oversight Hearing, supra note 62, at 11-32.
72See EPA Budget, supra note 59, at A-113, 114, 117; 1-27, 28, 31;
SF-24, 25, 29-30, 55. The most substantial reductions would occur in
stationary air (A-106-107); water quality (WQ-109-110); hazardous
waste (HW-59-60); pesticides (P-44-45) and toxic substances (TS-48-49).
Although the drinking water enforcement program would not experience
reductions of the same degree, it is already funded at a comparatively
low level — less than 2% of the overall enforcement budget. Only in
the areas of mobile air pollution, improper waste disposal and hazard-
ous substance response are increases anticipated.
Current resolutions from both the House and Senate Budget Commit-
tees would maintain the F.Y. 1983 enforcement budget authority at
approximately the same level as F.Y. 1982. See S. Con. Res. 92; H.
Con. Res. 345. However, Congress is only part of the way through the
budget process and has yet to reach agreement on an overall F.Y. 1983
budget or to consider what spending ceilings should be imposed. It
should be noted, moreover, that the Executive Branch has formal as
well as informal options for decreasing actual expenditures, regard-
less of what is finally approved by Congress. Under the Congressional
Budget Act, 31 U.S.C. §§ 1301 et seq., the Administration may rescind
a portion of the budget (subject to congressional approval), or defer
a portion (subject to congressional disapproval). Informally, the
executive office traditionally exercises its discretion in committing
funds to suit actual circumstances. As a practical matter, Congress
must rely on specific legislative directives and oversight proceedings
(as well as informal pressure) to ensure that its objectives are
achieved.
'•'Research included the following sources:
1. Complaints received in response to Vice President Bush
letter of March 25, 1981, regarding the need for regula-
tory relief.
2. Review of complaints raised by the Business Roundtable in
its reports entitled The Business Roundtable, Air Quality
Project, "Effects of the Clean Air Act on Industrial
Planning and Development," Vols. 1 and 2 (July 1981).
3. Consultation with industry associations identified in
Appendix C to this report.
4. Congressional Oversight Hearings: House Report,
"Implementation of the Federal Water Pollution Control
Act," No. 96-71, 96th Cong. 2d Sess. (December 1980);
Senate Hearing, "Environmental Protection Agency Over-
sight," No. 97-H29, 97th Cong. 1st Sess. (October 15,
1981); Senate Hearings, "Implementation of the Comprehen-
-161-
-------
LNDEMNIFICATON STUDY
Section IV
sive Environmental Response, Compensation, and Liability
Act of 1981," No. 97-H-31, 97th Cong. 1st Sess. (July 8
and 20, 1981); Senate Hearings, "Oversight of Hazardous
Waste Management and the Resource Conservation and
Recovery Act," (unnumbered publication) 96th Cong. 1st
Sess. (July 19 and August 1, 1979); House Hearings,
"Oversight — Clean Air Act Amendments of 1977," No.
96-110, 96th Cong. 1st Sess. (July 30, November 27 and
28, 1979); Senate Hearings "Clean Mr Act Oversight," No.
97-H12, 97th Cong. 1st Sess. (April 8, 9, May 20, June 2,
3, 4, 5, 9, 11, 22, 23, 24, 25, July 8 and 9, 1981);
Senate Hearings, "Toxic Substances Control Act Over-
sight," No. 95-H69, 95th Cong., 2d Sess. (July 20, 21,
1978); House Hearings, "Authorization and Oversight of
the Toxic Substances Control Act," No. 96-28, 96th Cong.
1st Sess. (March 8, 13, 20, 1979); Index to Publications
of the United States Congress, 1978-1982, Congressional
Information Service, Washington, D.C.
5. Report of the National Commission on Air Quality, "To
Breathe Clean Air," March 1981.
'^George Freeman, attorney for Hunton and Williams of Richmond,
Virginia, and counsel to numerous facilities subject to EPA air pollu-
tion regulations, discussed three enforcement cases that he described
as evidence of unreasonable EPA enforcement policies. All three
involved EPA's use of prosecutorial discretion in a manner that
resulted in significant burdens on the facilities affected. In a
later conversation, Freeman indicated that while he though EPA's
stance to be unnecessary, it was not wholly without legal support and
he felt the primary problem exists in the 1977 Clean Air Amendments.
Senate Oversight Hearings, June 4, 1981, part 2, No. 97-H12, 97th
Cong., 1st Sess. at 479-82. Interview with George Freeman, July 9,
1982.
^Environmental Protection Agency, Senate Oversight Hearing
(October 15, 1981) No. 97-H29, 97th Cong., 1st Sess. 1-11; Senate
Hearings on Implementation of CERCLA (July 8 and 20, 1981), No.
97-H31, 97th Cong., 1st Sess.; House Oversight Report on Clean Water
Act, No. 96-71, 96th Cong., 2d Sess. (December 1980) 53-54; Senate
Oversight on Resource Conservation and Recovery Act (unnumbered) (July
19 and August 1, 1979) 96th Cong., 1st Sess. 317-622.
also report of the National Commission on Air Quality, "To
Breathe Clean Air" (March 1981) at 3.8-14; BNA: Environment Reporter,
Current Developments (July 24, 1981) at 414; House Hearings, Health
Standards for Air Pollutants, No. 97-97, 97th Cong., 1st Sess.
(October 14 and 15, 1981); House Hearings, Clean Air Act Part 1, No.
97-102, 97th Cong. 1st Sess. (October 22, 28 November 5, 10, 1981);
House Hearing, Clean Air Act Part 2, No. 97-103, 97th Cong., 1st Sess.
(November 19, 20, December 14, 16, 1981).
-162-
-------
INDEMNIFICATON STUDY
Section IV
77See CAA § 306; CWA § 508; U.S. v. Sharon Steel Corp., 14 E.R.C.
1957, 1959 (W.D. Pa. 1980).
7838 Fed. Reg. 25161 (September 12, 1973).
79See 40 C.F.R. § 15.20(a)(3).
80See 40 C.F.R. § 15.20 et seq.; Appendix A; discussed in U.S. v.
Sharon Steel Corp., supra note 77, at 1958-59.
81See, e.g., Union Electric Co. v. EPA, 593 F.2d 299 (8th Cir.
1979); Fry Roofing Co. v. U.S. EPA, 554 F.2d 885 (8th Cir. 1977); West
Penn. Power v. Train, 522 F.2d 302 (3d Cir. 1975).
82See, e.g., Conoco Inc. v. Gardebring, 503 F. Supp. 49 (N.D. 111.
1980), and cases cited in text. See also Americ in Petroleum Institute
v. Costle, 17 E.R.C. 1334 (E.D. La. 1982).
, e.g., Humana of South Carolina, Inc. v. Califano, 590 F.2d
1070, 1081 (D.C. Cir. 1978); Porter County Chapter of the Izaak Walton
League of America, Inc. v. Costle. 571 F.2d 359, 363 (7th Cir. 1981);
Rhodes v. United States, 574 F.2d 1179, 1181 (5th Cir. 1978);
Winterberger v. Teamsters, Local Union 162, 558 F.2d 923, 925 (9th
Cir. 1977). The question of when judicial review of administrative
action may be sought also raises issues relating to the exhaustion of
administrative remedies and ripeness for review. See, e.g. , Hooker
Chemical Company v. EPA, 15 E.R.C. 1721, 1724 (3rd Cir. 1981); Conoco,
Inc. v. Gardebring, supra; U.S. V. Sharon Steel Corp., supra ; U.S.I.
Properties v. EPA, 16 E.R.C. 1408 (D. Puerto Rico 1981).
Blessing v. United States, 447 F. Supp. 1160 (1978), esp.
fns. 14, 18 and 26.
85Id. at 1170.
-163-
-------
INDEMNIFICATION STUDY
Section V
V. CONCLUSIONS
Page
A. Introduction 165
B. Statement of Conclusions 165
1. No New Indemnification Programs are Recommended
at This Time 165
2. Other Conclusions Concerning Indemnification
on Equitable Grounds 170
3. Other Conclusions Concerning Indemnification
Designed As An Incentive Or Disincentive • • • 175
-164-
-------
INDEMNIFICATION STUDY
Section V
CONCLUSIONS
A. Introduction
This section focuses on the principal question posed by
Congress — under what circumstances, if any, should indemnification
be accorded any person as a result of action taken by the Admin-
istrator? Conclusions pertinent to this question are stated, along
with a summary of the findings and reasoning supporting them.
Documentation appears in prior text. We have not made conclusions
with respect to cost or financing mechanisms because no new programs
were proposed for which cost or financing recommendations were needed.
However, cost and financing have been considered both in general, and
in the evaluation of existing and potential programs. Findings and
observations on these points are included in the body of the report.
The conclusions fall into two categories. The first is a series
of findings supporting the overall determination that no new indem-
nification programs are needed at this time. The second category of
conclusions are general ones that are more tentative and are perhaps
more correctly characterized as observations based on the overall
research and analysis involving the issue of government indemnif-
ication. They are included to provide assistance in the future con-
sideration of indemnification legislation.
B. Statement of Conclusions
1. No New Indemnification Programs are Recommended at This Time
This conclusion is based on findings that (a) no justifica-
tion currently exists for expanding existing EPA indemnification pro-
grams into new areas of EPA jurisdication, (b) an analysis of likely
loss areas revealed no Justification for recommending a new indem-
nification program, and (c) there is minimal current interest on the
part of the regulated industry in such a remedy.
When evaluating the need for a new indemnification program,
emphasis was given to the identification of current or past cir-
cumstances that would be arguable candidates for indemnification under
any of the criteria identified in prior analysis, as well as to an
assessment of the probability that such circumstances may arise in the
future. In general, few candidates were identified. While future
incidents could not be ruled out, the probability of such incidents
combined with past experience suggests that these incidents will be
infrequent, if they occur at all, and that they probably will not be
similar if they occur. In such a case little would be gained by
establishing a new program rather than by addressing each Incident
individually.
-165-
-------
INDEMNIFICATION STUDY
Section V
a. There is no justification for expanding existing
programs into new areas of EPA jurisdiction.
Section 15 of FIFRA
Evaluation of the need to extend § 15 indemnification to other
EPA programs indicated that TSCA would be the primary candidate for
possible extension. TSCA is the only other statute administered by
EPA that significantly regulates particular products, rather than pro-
cesses for wastes. Application of § 15 to TSCA would be unnecessary
because TSCA incorporates many safeguards for inventory losses. The
Administrator is given a large variety of regulatory options to reduce
risk under TSCA, but must choose the least burdensome one for each
situation. Unlike FIFRA, bans under TSCA may not be made by admi-
nistrative order. Instead, they oust either be by rulemaking or court
order, thus allowing substantial opportunities for hearing. The two
TSCA bans implemented so far have been phased in and did not cover all
the uses of the chemicals, thus avoiding the problem of unusable or
unsaleable inventory. Limited experience under TSCA prevents a final
conclusion, but at this time no FIFRA-like indemnification program
seems warranted.
Section 202(a)(3)
Based upon the current situation, § 202(a)(3) indemnification
also should not be extended to other EPA areas. Both of the primary
candidate areas —the Clean Air Act and RCRA — are fundamentally dif-
ferent from the Clean Water Act in that they do not involve initial
government funding of the cost of compliance. The potential for
failure of innovative air pollution or solid waste technology does not
appear to be a major disincentive to the use of innovative technology.
Furthermore, it is questionable whether an incentive-type indem-
nification provision can function effectively without either the type
of funding provided by § 202(a)(2) grants or a commitment to
demonstration project funding. Because construction grants funding is
available only for municipal wastewater treatment facilities, and
because private-sector water and air-pollution control demonstration
projects have been significantly reduced by budget cutbacks, such an
extended indemnification program does not seem warranted.
Nevertheless, if in the future significant encouragement of the deve-
lopment of innovative technology is deemed to be appropriate, then
indemnification comparable to that under § 202(a)(3) might be
considered.
Section 113(b)
Section 113(b) of the Clean Air Act allows for an indemnity in
the form of attorneys' fees and other court costs when a court deter-
mines that EPA has taken unreasonable enforcement action. This is
intended both to encourage persons (who would otherwise not do so) to
-166-
-------
INDEMNIFICATION STUDY
Section V
make a legitimate defense, as well as to deter unwarranted EPA
litigation.
This provision does not exist in other EPA statutes and is pro-
bably not needed. First, there Is no indication from current experi-
ence under the Clean Air Act that such a provision offers a meaningful
deterrent to unreasonable Agency action, or that it is needed to pro-
vide an incentive to defend against such action. Second, with rare
exceptions there is no evidence that judicial enforcement undertaken
by.EPA is "unreasonable" (although such a term may be interpreted in
various ways). Finally, under current EPA enforcement policies and
concurrence procedures, there is little probability that unreasonable
Judicial enforcement will occur. It should be noted that the Equal
Access to Justice Act already ensures for small businesses much of the
relief that might be provided if § 113(b) were extended to other EPA
programs.
Section 311(1)
Section 311(1) of the Clean Water Act allows a nonliable oil
spill discharger to recover voluntarily-Incurred cleanup costs. The
purpose is to encourage fast, effective action by the person who is
first likely to know of the spill and who may be able to prevent a
more costly and damaging problem by taking immediate containment and
cleanup steps. Section 311(1) has worked, but such a provision is
probably not suitable for spills and similar releases under other EPA
statutes at the present time. A somewhat different approach has been
adopted by the Comprehensive Environmental Response, Compensation and
Liability Act of 1980 (CERCLA) that applies to all releases (other
than oil spills). Although experience with CERCLA is limited, a need
for further change Is not now indicated.
b. An analysis of likely areas revealed no justification
for recommending a new Indemnification program.
The categories of losses specifically evaluated for
pot'ential indemnification were losses due to:
• Agency release of confidential business information;
• delays;
• conflicting requirements;
• emergency actions;
• unreasonable enforcement actions;
• change In Agency policy.
-167-
-------
INDEMNIFICATION STUDY
Section V
An evaluation Indicates little justification for a prospective indem-
nification program under any of these categories.
Confidential Business Information
No EPA indemnification program seems warranted for losses
resulting from unauthorized disclosure of confidential business infor-
mation (C81). Both negligent and intentional unathorized disclosures
are arguably subject to a suit for money damages under the Federal
Tort Claims Act. The extent of such disclosures appears to be quite
limited, and there are both criminal sanctions against intentional
disclosure and elaborate security provisions under FIFRA and TSCA
which are designed to safeguard CBI. Additionally, congressionally-
authorized disclosures by the Agency are comparable to other regula-
tory burdens that do not require compensation.
An indemnification program also does not seem warranted to deal
with the alleged problem of losses due to delays in government permit
processing. Both industry and EPA representatives concerned with the
problem of permit delays have focused on remedial solutions other than
indemnification. Furthermore, several studies have indicated that
regulatory problems are not often significant causes of delays in con-
struction of new power plants and other industrial facilities.
Instead, many delays result from such problems as financing, labor and
equipment availability, and management difficulties.
Conflicting Regulatory Requirements
The theoretical problem of conflicting and overlapping regula-
tory requirements does not appear to be a problem warranting an indem-
nification program. While the type of interacting requirements that
allegedly posed difficulty in the Tris case may well be deemed to
justify indemnification, no similar Interactions appeared in a review
of EPA programs. At worst, the examples found represent compliance
costs made necessary by reason of other regulatory standards. This is
not normally regarded as a basis for indemnification, but rather an
ordinary business consequence of any regulatory program.
Emergencies
Although emergency action seems theoretically to provide
increased chances for creating an indemnifiable loss, the Agency has
very rarely used its emergency authority to curtail private
operations. The enactment of CERCLA has provided EPA with authority
for a coordinated response to any emergency caused by release of
hazardous materials, according to the National Contingency Plan.
Together with Superfund, which permits the government Itself to take
action in some cases, the new law replaces an ad_ hoc response by court
-168-
-------
INDEMNIFICATION STUDY
Section V
Injunction with a planned, publicized and coordinated approach. This
reduces the possibility of inequity, to the point that there is no
current justification for recommending an indemnity program for losses
caused by emergency action.
Unreasonable Enforcement
An expanded indemnity program to compensate persons subject to
unreasonable administrative enforcement is probably not warranted.
This is primarily because there is no evidence that clearly unreason-
able enforcement exists on more than a very infrequent basis, and
because EPA'a current enforcement policies and concurrence procedures
make such occurrences less likelycthan ever before. In addition, it
would seem preferable to accomplish Agency restraint through tradi-
tional avenues of judicial review, rather than indirectly by the
threat of indemnification. The former route would encourage loss
anticipation and avoidance, while the latter route, if designed
effectively, could result in excessive restraint and thus interfere
with the Agency's enforcement mission.
Changes in Policy
Changes in policy may sometimes result in losses, but the
changes are usually foreseeable and the losses are usually within
expected risk parameters. In addition, policy changes are usually an
expression of congressional policy or the result of a rational and
defensible exercise of discretion delegated to the Agency and are
rarely if ever due to Agency negligence or misconduct.
c. Industry consultation disclosed little Interest in
indemnification as a means of regulatory relief.
Consultation with representatives of the regulated
Industry indicated:
o Given the controlling statutory provisions, most
persons felt that EPA regulatory action is usually
supportable. They thought that industry is given a
fair opportunity to participate in rulemaking and to
avoid unnecessary losses — especially in comparison
to other federal health and safety regulatory
agencies.
o Many felt that indemnification is not an Important
issue and not politically appropriate in the current
legislative environment.
o Some expressed greater interest in other regulatory
relief to solve particular problems and did not feel
indemnification would satisfactorily resolve the
problems experienced.
-169-
-------
INDEMNIFICATION STUDY
Section V
• No persons interviewed described past losses of the
nature chat would obviously result in an indemnity
under any of the theories and factors explored (with
the exception of losses related to programs already
in place).
• Interest was expressed in keeping existing indem-
nification programs in place.
2. Other Conclusions Concerning Indemnification on Equitable
Grounds
a. Whether Indemnification is warranted for an Inequity
raises complex issues that are best considered In the
context of the specific case and not resolved In advance
by legislating a prospective indemnification program.
(1) Prospective Inequity indemnification programs are
not usually desirable unless based on a specific
event.
Certain indemnification objectives require that a
program be prospective in nature. These objectives include
encouraging desirable private action, such as voluntary oil spill
cleanup, or discouraging undesirable agency action, such as unreaso-
nable enforcement litigation. However, if the objective is to relieve
or correct some inequity, an indemnification program would not nor-
mally be necessary until it is clear that an inequity has occurred or
will occur. When based on a specific event and the reasonable expec-
tation that there will be repetitions of that event, a prospective
program may be justified to avoid the costs and inefficiencies asso-
ciated with piecemeal legislation.
Otherwise, it is difficult to justify such a program
in view of the potential disadvantages. These include: (1) that the
program will encourage undesirable behavior on the part of potential
recipients (e.g., less carefulness to avoid Indemnifiable losses); (2)
that it will result in undesirable conduct on the part of the agency
(e.g., riskier conduct, knowing that any adverse impacts may be
compensated, or inappropriate restraint, in order to avoid the costs
or inconveniences involved); (3) that it will be inaccurate In its
forecast of the costs involved, the criteria that should be applied,
or the appropriate financing mechanisms; and (4) that other signifi-
cant adverse Impacts may have been ignored (e.g., interference with
other agency missions).
Once legislation is in place, an agency must act
within its scope, regardless of whether it proves to be Inappropriate
in some respect. Congress, on the other hand, may offer greater
assurance of its appropriateness by waiting until events requiring
such legislation are clear.
-170-
-------
INDEMNIFICATION STUDY
Section V
(2) Specific factors supporting or oppossing indem-
nification may be expected to vary from case to
case, and the issues raised are not readily
resolved in advance by the design of a general rule*
Whether an inequity has occurred is almost always a
function of circumstance and is highly dependent on the character of
the injury, its cause, who was at fault (if anyone), whether the
injury could have been foreseen or avoided, whether the loss could
have been mitigated, and other elements typical of tort law analysis
involving issues of fault and remedy. These issues are rarely clear.
In fact, tort cases have traditionally provided attorneys with excep-
tional opportunities to practice their skills of argument and per-
suasion precisely because each casecis unique in some respect. Under
tort law, it would not be expected that the answer would be simple or
fit precisely within a given rule, and there is little reason to
expect for the same issues to be resolved more easily in the context
of government indemnification, if any assessment of the inequity is to
occur at all.
Aside from the inability to determine the precise
elements of an inequity in advance of a specific event, there are
other significant barriers to the design of a general rule. Such a
rule requires a commitment to a specific future policy, reflecting not
only agreement that certain events amount to an inequity requiring
redress, but also that compensation (according to some rule of
calculation) is the most appropriate means of redress. Such a commit-
ment would be particularly suspect given the traditional rejection of
a compensation remedy by Congress except in those narrow circumstances
recognized under the Federal Tort Claims Act and the Fifth Amendment.
In fact, our research Indicates in every case to
date but one, indemnification legislation has been based on specific
injuries; and in almost every case the legislation has been excep-
tional in character and not indicative of a recurrent policy or series
of similar events. Moreover, Congress usually prefers some other form
of relief, such as an exemption or variance — or no relief at all,
thus creating an incentive among those susceptible to injury to avoid
it. In such a climate it appears impractical to forecast the need for
indemnification by resolving highly individual policy issues in the
absence of specific circumstances in which these Issues will arise.
The dilemma becomes all the more apparent if one
attempts the design of a theoretical model to use as a guide for ine-
quity indemnification legislation. General criteria pertaining to the
nature of the loss, responsibility for the loss, the amount of
compensation, a financing scheme, and administration of the program
may be readily identified* However, the relative importance of each
factor may vary significantly depending upon the perceived importance
of indemnification as a policy objective in comparison to other objec-
-171-
-------
INDEMNIFICATION STUDY
Section V
lives from which indemnification may detract and depending on the
availability of alternative loss avoidance and relief measures which
may obviate the need for indemnification.
For example, a difficult Indemnification administra-
tive burden may currently be considered unjustifiable If It will
require the reallocatlon of agency resources away from a higher
priority pollution control objective to administer the indemnification
program. In such a case, a variance may provide sufficient, although
only partial relief. However, priorities may later shift so that the
pollution control program involved is no longer rated more Important
when the loss occurs to a specific group, e.g., small fanners in an
economically depressed area — at least with respect to the realloca-
tion of program resources represented by administrative costs. This
problem is compounded by the number of factors other than administra-
tive costs which may have an important role in determining the advisa-
bility of an indemnification program, as well as the almost endless
alternative loss avoidance and relief mechanisms that might be made a
part of the rule; and it Is rendered virtually unresolvable by the
Infinite variations In policy preference which may arise due to unpre-
dictable events.
b. Procedural protections may reduce the need for
indemnification.
Equity-based indemnification payments have been made in
situations in which full procedural due process had not been provided.
While this is not the only factor supporting indemnification in those
situations, nor does due process or proper procedure always work
effectively to prevent a loss, suspension or absence of procedural
protections Is an Important factor which can weigh in favor of
granting indemnity.
The participation of affected parties and the public In
general In regulatory proceedings provides notice of the agency's
intentions, and thus a longer transition period within which to adjust
to the regulatory requirements. It provides opportunities to
challenge the assumptions or proposed solutions. There is less likely
to be an outright error or an unfair devolution of the whole cost of a
regulation on one group after public discussion of an issue.
Most environmental statutes facilitate involvement of
the public at the earliest possible stage of policy conceptualization.
This has been done in response to the need to avoid party actions or
unreasonable regulations. There is much organized public involvement
prior to any rulemaking, and the agency is frequently facilitating
this involvement even when it is not specifically required to do so.
However, when an emergency arises, then the procedural protections
such as a hearing may be deferred until after the needed action is
taken. Usually, though, either Congress or the agency recognizes the
-172-
-------
INDEMNIFICATION STUDY
Section V
potential danger and will undertake advanced planning to anticipate
the way in which such a matter should be handled when it arises. The
growing sophistication and sensitivity of the administrative process
to this type of problem will probably prevent, in most cases, blunders
like the DHEW announcement of the contaminated cranberries.
Nevertheless, as in the past, indemnity payments are most likely to be
warranted and needed where the process of orderly regulation has bro-
ken down.
c. Indemnification for inequities has not resulted in
domino-like extensions of indemnification to other
situations.
An argument often raised against the use of indem-
nification as a remedy is that it will establish a legislative prece-
dent that will be difficult to avoid in the future. Numerous requests
of a similar nature will arise as a result, with predictable adverse
impacts on budget and government administration at a minimum, and
possibly interference with other pressing government objectives in
addition.
Although it is not possible to state categorically that
such concerns are illfounded, it seems unlikely that events justifying
indemnification would result in such a domino effect. Two factors
would normally prevent it. First, inequities are almost always unique
in character. What makes an event inequitable may usually be
explained only by describing the entire event and its unique elements
of person, time, place, situation, intent and misfortune. The appro-
priate remedy in such a case, whether it is compensation or some
alternative measure of relief, may also be attributed to considera-
tions that apply at that time to that event — the availability of
financing, the availability of an appropriate administrative
mechanism, the political support or opposition, etc. These unique
aspects of event and remedy should pose for future circumstances only
the inconvenience of comparison, not the dilemma of inconsistency.
A second factor involves the capability of Congress to
define indemnification objectives and conditions in such a way that
the precedential effect is very limited. It is within the power of
Congress to pass or reject indemnification legislation for almost any
reason, including strictly political reasons (within constitutional
limitations); however, a clear description of the factual as well as
political basis would forewarn future indemnification constituencies
that arguments founded on legislative precedents would be of limited
persuasiveness.
A review of proposed and enacted indemnification
legislation points out that events for which indemnification has been
proposed in the past are unique, that the congressionally formulated
remedy has been different in most cases (not always involving indem-
nification), and that no domino effect has occurred.
-173-
-------
INDEMNIFICATION STUDY
Section V
d. Some alleged inequities are not unique to EPA, and
solutions, if warranted, may need to be governraent~wide
in scope.
Certain alleged problems which might merit consideration
of indemnification involve situations that are not unique to EPA and
may more appropriately require consideration of broad statutory solu-
tions outside the scope of this study. Such broad statutory solutions
might Involve either an Indemnification mechanism or they may involve
broad alternatives to indemnification.
For example, certain losses might be dealt with by
amendments to the Federal Tort Claims Act which would reduce some of
the Act's exceptions to the waiver of immunity. Specifically, the
discretionary function and misrepresentation exceptions might be
modified to permit tort suits against the government where certain
discretionary functions or misrepresentations are involved.
Furthermore, some actions which might not now be covered under the
Federal Tort Claims Act could specifically be authorized. For
example, it is not clear whether an action against the government for
breach of a trade secret could be maintained under the Federal Tort
Claims Act. If considered necessary, an amendment to the Act could
clarify the point. But limiting consideration and the effect of such
changes to EPA probably would not be wise. Piecemeal modification of
the Federal Tort Claims Act does not seem advisable. If there are
possible problems that are not unique to EPA, then such problems
should be handled in a government-wide manner.
e. The private bill process can continue to provide relief
of last resort for inequities arising from EPA action.
Private bills have historically provided a process to
redress inequities that are not susceptible to a program or plan
because they are so unpredlctiable in nature or infrequent in
occurrence. The existence of such a political process, though it is
neither well known nor simple, has sometimes provided a relief valve
for cases otherwise barred from the courts. In the past when Congress
has become aware of a pattern of similar meritorious grievances pre-
sented in private bills, it has enacted general legislation to deal
with the problem. Thus, the private-bill process can serve as an
interim remedial mechanism until such time as a general relief program
proves warranted.
The special jurisdiction procedure, by which Congress
removes a legal impediment to an otherwise valid claim in the Court of
Claims, is particularly appealing for regulatory indemnification.
Congress can waive sovereign immunity or create a special exception to
some barrier to relief, tailored to the particular circumstance and
having no precedential value. Then it can leave to the court the
determination of liability in what is usually a complex factual situa-
-174-
-------
INDEMNIFICATION STUDY
Section V
tion. Further, by being given jurisdiction, the court can pay any
award that it makes out of its own Judgment Fund. This technique is
•faster, and results in sore certainty of payment than the
•Congressional Reference procedure in which the Court of Claims only
advises the Congress, and then sends its recommendation back for final
congressional action.
Nevertheless, the private bill process can be lengthy,
and there are no guarantees that either Congress or the Court of
Claims will act favorably, because the pitfalls of the political pro-
cess are involved and because all private bills request special
exceptions. It should also be noted that private legislation cannot
provide the incentives to private action or the disincentives to
government action that are the rationale for some kinds of Indemnifi-
cation. Since the remedy is provided £d_ hoc, perhaps long after the
loss and cannot be counted on in advance, It will not affect the
behavior of the parties before the fact of loss.
3. Other Conclusions Concerning Indemnification Designed As An
Incentive Or Disincentive
a. Indemnification has worked to provide Incentives to par-
ties outside the government.
Several indemnification programs demonstrate the bene-
fits of using indemnification to encourage action by parties outside
the federal government. These include the diseased animal livestock
program at the Department of Agriculture (to encourage destruction of
diseased livestock to prevent spread of the disease); the arts and
artifacts indemnification program (to facilitate museum display of
foreign art for which losses would otherwise be uninsurable); the
Superfund contractors indemnification clause (to facilitate the remo-
val and destruction of hazardous waste in circumstances where liabi-
lity insurance is inadequate); the Clean Water Act innovative
technology replacement grants (to encourage municipalities to try
innovative waste treatment technology); and the Clean Water Act oil
spill cleanup indemnification provision (to encourage dischargers who
are not liable for cleanup to do so voluntarily in the interest of
ensuring a rapid, effective initial response).
Two major advantages in these indemnity provisions were
noted: in some instances they ensure accomplishment of public policy
objective that could not be undertaken by the federal government
(e.g., immediate, stop-gap cleanup measures); in others, they offer a
more economical option than federal action because of efficiencies
obtained in private enterprise. However, these programs are not
without potential disadvantages. Notable areas of concern include:
• Agency supervision. To the extent that the program
establishes an obligation to indemnify, design measures must
-175-
-------
INDEMNIFICATION STUDY
Section V
be included to ensure that only those actions consistent
with Agency (or legislative) priorities are actually
indemnified. Under the oil spill indemnity statute, for
example, unsupervised cleanup activity could result in the
allocation of funds to incidents or specific cleanup coste
for a lower priority category. A mechanism employed under
Superfund to prevent this occurrence is to require that all
expenses be preauthorized.
• Alternative mechanisms. It Is possible that the same con-
duct can be encouraged by other mechanisms, or that natural
Incentives already exist which make indemnification unnec-
essary. For example, if specific losses are predictable, it
is possible that a procurement proceeding, with concommitant
advantages gained by planning and supervision, would prove
more economical or more effective. Under Superfund, tough
sanction provisions and narrow defenses to liability are
thought by some to be more significant cleanup motivation
factors than Indemnification.
• Congressional and executive oversight. Since incentive
indemnification is necessarily prospective in nature, there t
is a greater risk that the mechanism, as designed, will be
off the mark in some respect. In the diseased livestock
program, for example, it was determined that federal
inspectors, and not the indemnification program, controlled
the spread of disease; and indemnification was viewed pri-
marily as subsidy. This factor suggests the need for more
careful monitoring and review.
b. Indemnification for the purpose of restraining certain
undesirable agency action is not clearly justified based
on experience under existing EPA programs. Moreover,
the potential for adverse impact warrants very careful
consideration before legislating an indemnification
program for this purpo-se.
The argument that indemnification may serve as an effec-
tive disincentive to undesirable Agency conduct was raised under both
§ 15 of FIFRA and § 113(b) of the Clean Air Act. In the case of
FIFRA, it was argued that the right to indemnification might prevent
precipitous suspension actions by inducing the Agency to compare the
costs of compensation to the environmental risks involved in per-
mitting existing pesticide inventories to be used up. In the case of
the CAA, it was argued that unreasonable judicial enforcement action
might be prevented by awarding attorney fees to the opposing party
whenever a court determined the action to be unreasonable.
Under both statutes, experience indicates that the
indemnification provision has had minor if any impact as a disincen-
-176-
-------
INDEMNIFICATION STUDY
Section V
tive, and chat other factors have been far more effective in accom-
plishing desired restraint. Under the CAA, for example, enforcement
screening procedures for all judicial actions are so thorough
(involving numerous layers of EPA and Department of Justice review),
that there have been very few complaints relating to error or
harrassment. Under FIFRA there is a greater likelihood that the
question of indemnification may have been persuasive within the Agency
as a negative factor when considering an emergency ban. However, in
the relatively few cases in which suspension proceedings have ensued,
the Agency has indicated that either such bans were imposed or the
scientific evidence supporting a ban was insufficient. Further, it is
widely felt that the burden and complexity of procedural requirements
under § 6 have served as a primary disincentive to the Agency's
reliance on suspensions, thus preventing the indemnification issue
from arising at all except on a very infrequent basis.
As an effective disincentive, however, the indemnifica-
tion provisions may be deficient for a more significant reason.
Currently, neither provision has actually posed the threat of a signi-
ficant adverse impact on the Agency. Administration has proven to be
feasible without significant costs, and the Agency budget has not been
responsible for the actual payout of awards.
In either case, it would be theoretically possible to
make the disincentives more relistic — for example, by taking award
payments out of the operating budget of the office involved (without
access to a supplemental appropriation, as one version of the Equal
Access to Justice Act would have required), or by forcing the offi-
cials involved to undertake such burdensome administrative require-
ments that their primary mission would necessarily be interrupted.
Such an approach would be exceedingly difficult to justify in most
circumstances, however, unless the underlying regulatory mission were
insignificant in comparison to the policy favoring indemnification.
Two risks are raised: that the disincentive will not work, in which
case the indemnification remedy would exact its toll on the Agency
mission (as intended); and that the disincentive will work too well,
in -which case the Agency may be deterred from taking any action at
all.
If restraints on Agency action are deemed necessary, it
would appear more logical to address this need directly by imposing
procedural hurdles, expedited review, or new statutory criteria for
the action rather than indirectly by establishing economic
disincentives, which may have unpredictable and costly results.
-177-
-------
INDEMNIFICATION STUDY
Appendix A
APPENDIX A
BACKGROUND ON INDEMNIFICATION
Page
I. General Compensation Provisions A-2
A. Introduction A-2
B. The Federal Tort Claims Act A-2
1. Background A-2
2. FTCA Program Administration A-3
3. Important Exceptions to the FTCA A-4
C. Fifth Amendment Taking A-7
II. Congressional References and Other Private Bills A-12
A. Private Bills A-12
B. The Court of Claims A-1A
C. Congressional Reference Cases A-15
D. Special Jurlsdication A-17
III. Non-EPA Instances of Indemnification A-18
A. Agricultural Indemnities — Subsidies and Incentives. A-18
1. Animal Disease A-18
2. Dairy and Beekeepers A-19
3. The Cranberry Scare of 1959 A-20
B. Indemnity Awarded for Government Error A-20
1. Mlzokaml Spinach A-20
2. Marlin Toy Co A-21
C. Indemnification Being Considered for Policy Change. . A-22
1. Tris Ban A-22
2. Cyclamate Ban A-23
D. Swine Flu Vaccine Program —
Specific Incentive Situation A-23
E. Indemnity in the Form of Ongoing Insurance Program. . A-25
1. Arts and Artifacts. . « A-25
2. Defense Contracts A-26
3. Non-Defense Contractors A-26
Footnotes A-28
A-l
-------
INDEMNIFICATION STUDY
Appendix A
I. GENERAL COMPENSATION PROVISIONS
A. Introduction
The concept of sovereign immunity, which bars a suit against the
government except with its permission, provides a key to understanding
the question of indemnification by the government. Although the doc-
trine protects the government from being legally forced to provide
compensation to injured parties, there are circumstances where, by
either constitutional provision or statute, the government has allowed
itself to be sued for money damages. Specifically, the Fifth Amend-
ment requires compensation when private property is taken for public
use; and contract claims and some tort suits against the government
have been authorized by statute.
Judicial actions for damages, however, are narrowly prescribed,
and compensation is not generally provided in circumstances not
involving a taking of property, a contract claim, an action under by
the Federal Tort Claims Act, or some other specific waiver of immu-
nity. Of course, there have been exceptions to the general tendency
against indemnification, but these have involved unique circumstances,
and the judgment to grant indemnification has been made on a discre-
tionary basis.
The traditional limitations on indemnification for the adverse
effects of governmental actions partly involve a belief that situa-
tions that might require indemnification should be avoided so as to
make such payments unnecessary. If a loss cannot be avoided, then
payment of an indemnity is generally viewed as Inappropriate. A large
part of the attitude stems from a fear of creating precedents which
would require the government to make indemnity payments for a wide
range of regulatory activity. This position has been articulated and
exemplified by numerous authorities, Including fifth amendment taking
decisions, hearings on congressional reference bills, court of claims
opinions in congressional reference cases, and Federal Tort Claims Act
decisions.
B. The Federal Tort Claims Act
1. Background
Over a century ago, the United States began selectively
waiving its immunity from suit by enacting various laws allowing con-
tract claims, patent infringement claims, claims arising out of the
activities of government-operated railroads and utilities, and admi-
ralty and maritime tort claims against the federal government. How-
ever, until the passage of the Federal Tort Claims Act (FTCA) in 1946,
if a person suffered personal Injury, property damage, or other finan-
A-2
-------
INDEMNIFICATION STUDY
Appendix A
'Ci.a'1 loss due to the negligence of a federal employee acting on behalf
of the federal government, no court action was available for the
asse-ssment of damages.^
Thus before 1946, a person who sustained injury from the tort-
ious actions of the federal government was required to seek compensa-
tion from Congress. Unless the claim fell within one of the limited
categories for which Congress had statutorily waived immunity (e.g.,
admiralty and maritime tort claims), it was necessary for the claimant
to convince Congress to pass a private act covering the specific claim
and appropriating funds for the payment of damages.
As early as 1832, John Quincy Adams complained that Congress
was spending half of its time considering private business, and
expressed concern that the system was inefficient and inconsistent.
Partially in response to this concern, Congress began, in the
mid-1800s, to waive Immunity for certain specified claims and to allow
courts to take over some of the burden. By the 1940's, Increased
efforts were made to find a way to relieve Congress of its burden of
private legislation and to find a more equitable way to adjust the
growing number of tort claims arising out of expanding federal
activity. The solution adopted was passage of the FTCA in 1946, which
provided that the federal government, with certain significant excep-
tions, could be sued for the torts of federal employees acting within
the scope of their employment.^
2. FTCA Program Administration
Any individual, corporation, or other legal entity, with the
capability to sue as such, is eligible to file a claim or pursue a
lawsuit under the FTCA. There are several prerequisites for a
claim:
• A personal Injury, property loss or death must have been
suffered;
• The injury must have been caused by the negligent or
wrongful act or omission of a federal employee acting
within the scope of his employment;
• The conduct causing injury must be a tort under law of the
state where the conduct occurred; and
• The conduct causing injury oust not be within any of the
FTCA exceptions.
Before filing a lawsuit, claimants must present their claims
to the appropriate agency for administrative adjustment. Agency heads
have the authority to compromise or settle such claims. If a claim
is denied, or if the agency head takes no action on the claim for six
A-3
-------
1980
Total
EXHIBIT 1
CLAIMS AGAINST TUB GOVERNMENT UNDER
THK FEDERAL TORT CLAIMS ACT
Asbestos
Aviation
Swine Flu
Radiation Admiralty General Torts Mlacellaneous
I
(U
Total Number
of Claims
Total Amount
at Issue
Judgments Against
the Government (I.e..
total amount awarded)
1981
2370
$7,931.523.925
$95.681,489
35
204
122
$50.725,000 $1,007.206.900 $348.968.775
$175,000 $25,298,176 $3,867,438
11
23S
1481
282
$65.000,000 $66.657,396 $3.409,160,495 $2,983.805,359
0 $9,375.215 $49,899,962 $7,0b5,t>98
Total Number
of claims
Total Amount
at Issue
Judgments Against
the Government (I.e.,
total amount awarded)
1819
$13,484.447,265
$74.999.783
171
122
212
$34,500,000 $522,133,307 $277.274.645 $34.750.000
0 $13.508.536 $10,407,423 0
1124
177
$67,179.290 $7.462,878,102 $5,085,731,921
$3,263.809 $42,425.471 $4,894,544
-------
INDEMNIFICATION STUDY
Appendix A
months, a lawsuit may be filed. Suits under the FTCA are tried in the
United States district courts by judges sitting without juries.
3. Important Exceptions to the FTCA
From the perspective of this indemnification study, the impor-
tant general exceptions to the waiver of Immunity are those involving
discretionary functions and misrepresentation.' The exception covering
interference with contract rights also is relevant.^ These exceptions
effectively shield most governmental regulatory activity from tort
suits, although there are a few cases where this shield has been
pierced to a very limited extent. In particular, some courts have
allowed personal injury suits against the government for negligently
conducting a government safety Inspection, such as a mine safety
Inspection.
a. Discretionary Function Exception
The discretionary function exception of the FTCA excludes,
from the waiver of immunity claims based upon "the exercise or per-
formance or the failure to exercise or perform a discretionary func-
tion or duty. . . . "^ The exception applies both to negligent conduct
and wrongful acts involving an abuse of discretion. '
While the courts have not established an absolute defini-
tion of discretionary functions, their opinions have focused on the
nature and quality of the discretion involved in the activities that
are the subject of the complaint. The key determinant seems to be
whether a policy judgment or the formulation of policy related to the
public interest is involved. One court has noted several considera-
tions or determinations that are usually involved in policy-oriented
discretionary decisions which are immune from suit:
• The evaluation of the feasibility or practicability
of government programs;
• The balancing of cost and benefit factors in govern-
ment programs;
• The establishment of priorities due to limitations of
available resources;
• The balancing of competing policy considerations in
determining the public interest; and
• The promulgation of new policies through regulation."
In recent years, courts have focused on a distinction
between policy Judgments and operational activities in order to inter-
pret and apply the discretionary function exception.'7 Where the
A-4
-------
INDEMNIFICATION STUDY
Appendix A
courts have found culpability in the execution of operational tasks
(such as maintenance of a road, a rescue operation at sea, auditing a
federal credit union, use of an aucomobile, and handling hazardous
materials such as explosives), they have generally found the govern-
ment liable under the FTCA. On the other hand, where th->y have dis-
covered culpability in the nature of the undertaking itself or in
decisions at the policy or planning level, the courts have usually
held that the exception applies. For example, agency decisions to
issue regulations, seize and detain property, prosecute, preserve
wildlife, accept bids and contract renewals, and run supersonic
flights have all been considered policy or planning matters, covered
by the exception and immune from liability.
Generally, when the government decides to establish and
operate a particular program, that policy decision Is an exercise of
discretion at the planning or policy formulation level, and exempt
from the waiver of immunity under the FTCA. However, once the discre-
tion is exercised and the decision made, government employees have no
discretion to operate the program negligently.^
The setting of standards and the application of those
standards through the establishment of regulations clearly Involve
discretionary policy judgments. On the other hand, application of the
regulation to particular parties and specific compliance, inspection,
and enforcement activities might, under certain circumstances, be
regarded as operational rather than discretionary. In particular,
limited functions of lower level officials, which are required by
rules and regulations and which clearly specify when official action
is required, fall within the operational category and are not covered
by the exception. Of course, a determination of whether the govern-
ment official has discretionary authority requires a careful case-by-
case evaluation of the particular statute or regulation involved. The
courts also seem to consider whether or not the statute, regulation,
or nature of the activity allows the claimant to rely upon the govern-
ment activity's being handled properly.9
Many claims under the FTCA have been for damages alleged
to have been caused by the negligent issuance or wrongful failure to
issue a permit, license, contract, certificate, or other government
clearance or authorization. In the great majority of cases, courts
have held that the decision whether or not to issue such government
authorization involves the consideration of a multitude of factors,
some of which may be political and non-Justiciable, thereby rendering
the activity discretionary within the meaning of the FTCA exception.*0
In governmental inspection programs, the shield of the
discretionary function immunity may be breaking down. During the last
decade, a number of personal injury claimants have been successful in
cases based upon negligent government inspections where the courts
have found the inspection to be an operational function involving pro-
fessional expertise but little discretion.^
A-5
-------
INDEMNIFICATION STUDY
Appendix A
Consistent with the policy/operational distinction, EPA
decisions to establish an Agency program, regulate a particular activ-
ity, issue policy guidelines or locate a facility would be discretion-
ary planning or policy decisions protected by the statutory exception.
Generally, licensing, permit and certification activities are simi-
larly protected. However, a negligent action or omission by EPA per-
sonnel in implementing the programs could be the basis of a suit under
the FTCA.
b. Hisprepresentation Exception
As with the discretionary function exception, the misrep-
resentation exception to the FTCA waiver of immunity relates to the
subject matter jurisdiction of the court. The misrepresentation
exception prohibits "any claim arising out of ... misrepresentation
. . .," and the exception applies both to negligent and intentional
misrepresentation.12 in construing this provision, the courts have
ruled that "misrepresentation" will be considered according to the
traditional and commonly-held legal definition of the tort as inter-
preted under federal, rather than state, laws.13 As decided by the
federal courts, misrepresentation includes the failure to provide
information as well as providing information that is wrong.
The misrepresentation exception can cover actions
involving personal injury, wrongful death, property damage, or finan-
cial or commercial loss. Most cases considering the applicability of
the misrepresentation exception also involve the discretionary func-
tion exception. Generally, the courts been more likely to deny claims
for commercial losses than claims involving bodily injury on the basis
of the misrepresentation exception.15
In cases where faulty flight information from air traffic
controllers has caused an accident, the controller's advice has been
regarded as an operational task negligently performed, and the misrep-
resentation exception has been held inapplicable.1° On the other
hand, one area in which the misrepresentation exception has consis-
tently been held to apply is in claims by disappointed bidders for the
construction or repair of government property.1'
A central issue in many misrepresentation cases is the
actual source of the claimant's harm. The courts have held repeatedly
that there is a distinction between negligent misrepresentation and
negligent conduct, with recovery as a result of the former being
barred by the misrepresentation exception to the FTCA.1** in the regu-
latory context, there have been a number of misrepresentation cases
involving inspections, certifications, and permits where the misrepre-
sentation exception has shielded the government from suit.19
A-6
-------
INDEMNIFICATION STUDY
Appendix A
governmental body. Only the second and third
situations are thought of as takings today.22
The Supreme Court recently explained the reason for a restrictive
approach to compensation for regulatory constraints:
Suffice it to say that government regulation—
by definition—involves the adjustment of
rights for the public good. Often this adjust-
ment curtails some potential for the use or
economic exploitation of private property. To
require compensation in all such circumstances
would effectively compel the government to
regulate by purchase. "Government hardly
could go on if to some extent values incident
to property could not be diminished without
paying for every such change in the general
law."23
The Takings Clause of the Fifth Amendment traces its origins to
the law of England, but it also reflects both the colonial experience
and dominant legal thinking of that time.2* Flowing first from the
Magna Carta, basic rights regarding property were brought to America
by the English colonists. In the eighteenth century, eminent domain
also became recognized and accepted in the American colonies as an
inherent and necessary power of government. At the same time, private
parties were afforded compensation for the loss of their property to
public use.
During the nation's first century, the taking issue was rarely
raised. When the issue did arise, the concept of a compensable taking
was limited by the courts to an actual seizure of land by the govern-
ment. The Supreme Court's first major treatment of eminent domain and
the police power came in 1887.25 jhe claimant there was a brewer
whose business was rendered worthless when Kansas enacted a prohibi-
tion statute. The Supreme Court concluded that the statute was a
valid exercise of police power, which involved no direct Invasion of
claimant's property. While the statute impaired the use of the pro-
perty, there was no taking found because the government had the power
to regulate property in order to abate a nuisance.26 With increasing
population, rapid economic growth, and urbanization in the late nine-
teenth and early twentieth centuries, the courts continued to uphold
regulations preventing nuisances to the community as a valid exercise
of police power and not a taking.
In 1922, the Supreme Court reversed its earlier line of cases and
stated that the constitutional question with respect to taking turned
primarily upon the degree of economic harm Imposed by the government
regulation.27 in the few cases since then, the Supreme Court has
refused to adopt any binding rule for determining where regulation
A-8
-------
INDEMNIFICATION STUDY
Appendix A
ends and a compensable taking begins.28 This ambiguity has permitted
the courts to judge each case on its own merits. However, it has left
no satisfactory rationale from which to define the distinction between
police power and taking.
The current approach of case-by-case determinations of whether a
taking has occurred began in 1922, when the Supreme Court dealt with a
statute that forbade the mining of coal in such a way that would
affect the structural soundness of any house.29 The effect of the
statute was to deny to the owner of mineral rights to a particular
property the right to mine for coal. Since the value of coal can only
be obtained if it is mined, the mineral rights became worthless* The
court held that the statute was an unconstitutional exercise of the
state's police power, with the diminution in value of the property
being the key determinant that the statute amounted to a taking. It
was the recognition that when the diminution "reaches a certain magni-
tude, In most if not in all cases there must be an exercise of eminent
domain and compensation to sustain the act."30 The decision thus set
out the mode of analysis that the constitutional question of a taking
depended upon the degree of injury imposed by the regulation. To a
certain extent the court also relied upon balancing the public gain
against the private loss, although the term "balancing of interests"
was not stated.3*
Thus, an unconstitutional taking can be implied from a substantial
and burdensome interference by the government with a private property
owner's use of the property. To constitute such an implied or de_
facto taking, government interference with private property rights
must be so substantial as to deprive the owner of all or virtually all
of his or her interest in the property. As a result, there is no
taking in the constitutional sense, unless the interference is so
substantial as to render the property worthless or useless.
In a 1958 Supreme Court case, owners of gold mines, closed in 1942
by direction of the War Production Board (WPB), unsuccessfully sued
the federal government to recover damages for an alleged taking of
property.32 xhe Court stated:
[T]he mere fact that the regulation deprives
the property owner of the most profitable use
of his property is not necessarily enough to
establish the owner's right to compensation.33
Since the mine closures did not amount to a disposal of the owners'
property or transfer of their workers, the Court concluded that the
UPB order did not constitute a taking.
In a 1979 decision, plaintiffs were charged with violations of
federal statutes prohibiting commercial transactions in protected
birds.3^ The Supreme Court explained that a reduction in the value of
A-9
-------
INDEMNIFICATION STUDY
Appendix A
property was not necessarily equated with a taking, even though the
most profitable use of the property was lost. In this case the speci-
fic regulation did not compel a surrender or physical invasion or
restraint upon the bird artifacts. Nor was all economic benefit
denied to the owners. The Court concluded that the cost of regulation
is a burden borne to secure "the advantage of living and doing busi-
ness in a civilized community."-" Similarly, in a 1980 taking case,
the Court upheld a rezoning of land overlooking San Francisco Bay,
because the ordinance in question only limited the number of residen-
tial units that could be built on the property and did not extinguish
a fundamental right of property.36
It is unlikely that EPA programs will result in a Judicial finding
of a constitutional taking. As a practical matter, the taking issue
is most likely to be raised in the regulatory context where EPA may be
exercising its powers to control air pollution, toxic chemicals, or
the use of confidential business information submitted to the Agency.
In regulatory cases, the courts seem to employ a sliding scale,
balancing the public health or interest against the individual's pro-
perty right. The regulatee must demonstrate a near total destruction
of the usefulness and value of this property right.
In one EPA case, a number of environmental groups, state govern-
ments and utility companies sought review of regulations promulgated
under the Clean Air Act to prevent significant deterioration of air
quality in areas that have air cleaner than the National Ambient Air
Quality Standards.^7 Responding to plaintiffs' constitutional chal-
lenge, the D.C. Circuit ruled that regulation of air pollution is
within the federal government's Commerce Clause powers. Concluding
that these regulations could not be an unconstitutional taking, the
court explained that, while the use of private land was limited, such
a limitation was not so extreme as to represent an appropriation of
the land.
The First Circuit similarly upheld an EPA transportation control
plan that mandated a 40% reduction in available off-street parking
spaces.38 in refusing petitioners' constitutional arguments, the
court focused on the fact that this was a regulation of uses, that no
title was taken, and that other uses remained available.
Recently, the 1978 FIFRA amendments relating to agency use of test
data submitted by one applicant to support a subsequent pesticide
registration application by another applicant has been the subject of
constitutional challenge. Thus far, the courts have concluded either
that plaintiffs in these cases do not have a property right in the
data or that the statute does not effect a taking.
A number of taking cases in the environmental field have been
deemed to be premature. For example, a pre-enforcement constitutional
challenge to the Surface Mining Control and Reclamation Act was held
A-10
-------
INDEMNIFICATION STUDY
Appendix A
not to be ripe because "mere enactment" of the statute did not con-
sititute a taking.4°
A-ll
-------
INDEMNIFICATION STUDY
Appendix A
II. CONGRESSIONAL REFERENCES AND OTHER PRIVATE BILLS
Congress has historically used private legislation to provide
relief to those who have suffered a loss at the hands of the govern-
ment and have no other means of recovery. Precisely because they are
available when exceptions to the Federal Tort Claims Act and other
limitations apply, private bills have constituted a safety valve for
exceptional cases.
There are three types of private legislation presently used by
Congress. All can be used as relief from the strictures of sovereign
immunity in exceptional circumstances. The largest category consists
of private bills handled entirely within the legislative process and
funded by special appropriations. Congress enlists the assistance of
the Court of Claims for the two other categories of private bills,
however. Congressional reference cases are those private bills
referred to the commissioners of the court of claims for hearings,
findings, and a recommendation to Congress. The cases are later
transferred back to the legislative body for final decision, and for
special appropriations if a claim is awarded. Bills of special juris-
diction provide relief indirectly In the form of lifting a Jurisdic-
tional barrier. A special jurisdication is conferred on the Court of
Claims for that case only, but it is final. Judgment, if any, is paid
out of the court's Judgment Fund.
A. Private Bills
Private bills are produced by a legislative process and passed by
a legislative body, but they do not have the general applicability
that is usually characteristic of legislation and fundamental in
guaranteeing equal protection. They are specific to a particular per-
son or situation. It can be difficult to draw a line between public
and private legislation because some public legislation (some tax
bills, for example) Is so narrowly drawn as to apply only to a very
few people, though the language seems to be general. Most private
bills have nothing to do with the costs of regulation, concerning
instead things such as moving expenses for government employees,
pensions, and permission to immigrate.
The Constitutional basis for private legislation is found in
Article I, § 8. Congress is empowered to "pay the debts ... of the
United States." This has been held by the Supreme Court for nearly 100
years to include moral or honorary debts as well as legal ones:
The term 'debts' includes those debts or
claims which rest upon a merely equitable or
honorary obligation, and which would not be
recoverable in a court of law if existing
A-12
-------
INDEMNIFICATION STUDY
Appendix A
facts, or written evidence that the Administration has withdrawn its
opposition. Also, as a practical matter, since there is not wide
interest in any bill, it would be difficult to muster the two-thirds
majority in both houses to override a veto.^9 There have, in fact,
been few vetos, most of them in cases in which the statute of limita-
tions had been waived for a tax case.^" These are explained by a very
strong administrative interest in the finality of tax determinations.
The number of private bills has declined for a number of reasons.
Many situations that once required private legislation may now be
remedied through the regular judical system. The Federal Tort Claims
Act is a prime example of a permanent waiver of sovereign immunity for
cases that had previously been handled by private bills. Other simi-
lar statutes provide causes of action to prisoners for injuries
suffered in prison, and to government employees for waivers of over-
payment of wages.51
B. The Court of Claims
The Court of Claims was established in 1855 to help Congress with
the large volume of private bills.** The original court was an advi-
sory board, hearing claims and reporting its findings and opinions,
together with recommended legislation, to Congress. It quickly became
evident that this would not provide the help needed. President
Lincoln thought the problem Important enough to address In his first
State of the Union message:
It is as much the duty of Government to render
prompt justice against Itself, in favor of
citizens, as it is to administer the same be-
tween private individuals. . . It was
Intended by the organization of the Court of
Claims mainly to remove this branch of busi-
ness from the Halls of Congress; but while the
court has proved to be an effective and
valuable means of investigation, it in great
degree fails to effect the object of Its
creation, for want of power to make its
judgments final.^3
Two years later, the Congress had acted on President Lincoln's sugges-
tion. The decisions of the Court of Claims were made final judgments
with appeal to the Supreme Court.54 However, the independence of the
court continued to be hindered because it could not pay judgments
until after appropriations had been "estimated" by the Treasury. This
provision was repealed in 1866.55 After several changes In the way
judgments of the court were funded, the present system was estab-
lished, giving the court maximum independence from Congress.5^ A
Judgment creditor need only submit a certified copy of the Judgment to
the General Accounting Office to be paid. Congress replenishes the
A-14
-------
INDEMNIFICATION STUDY
Appendix A
judgment fund periodically in lump sums. This minimizes the possibil-
ity that a particular claimant will be denied payment by Congress
after award by the Court of Claims.
The jurisdication of the court has been expanded by a number of
statutes. The best known Is the Tucker Act, which describes the
general Jurisdiction of the court:
The Court of Claims shall have jurisdiction to
render judgment upon any claim against the
United States founded either upon the Consti-
tution, or any Act of Congress, or any regula-
tion of an executive department, or upon any
express or implied contract with the United
States, or for liquidated or unliquidated
damages in cases not sounding in
tort. . . .57
The Court of Claims now has jurisdication over most money claims
against the government, with the prominant exceptions of tort and
pension claims.^ In all cases the government is the defendent and
there is no jury trial. In. the year ending September 30, 1981, the
Court of Claims heard 772 petitions involving 2,304 plaintiffs.59
C. Congressional Reference Cases
The desirability of having a judical body render assistance to
Congress in deciding private relief bills has long been evident, but
it was not until 1966 that a procedure was perfected. In 1962, the
Supreme Court found the Court of Claims to be an Article III court
(part of the Judicial, rather than the Legislative Branch of govern-
ment ).'^ Until then, both judges and commissioners"'- on the court had
participated in the resolution of reference matters for Congress.
Although the Court did not explicitly decide the issue, both the
majority and concurring opinions suggested that congressional refer-
ence jurisdiction was incompatible with the judicial power of an
Article III court because it involved advisory opinions rather than
active controversies.^2 xhe Court of Claims thereupon largely ceased
to accept new reference cases, though it continued work on those
pending.^ Four years later, Congress revised the congressional
reference statutes to provide that only the trial commissioners, not
the Article III judges, would take reference cases. The standard of
recovery in the statute did not change:
The trial commissioner . . . shall proceed in
accordance with the applicable rules to deter-
mine the facts, including facts relating to
delay or laches, facts bearing upon the ques-
tion whether the bar of any statute of limita-
tions should be removed, or facts claimed to
A-15
-------
INDEMNIFICATION STUDY
Appendix A
excuse the claimant for not having resorted to
any established legal remedy. He shall append
to his findings of fact conclusions sufficient
to inform Congress whether the demand is a
legal or equitable claim or a gratuity, and
the amount, if any, legally or equitably due
to the claimant.*>*
The procedure followed for the congressional reference cases is
similar to that for other private bills,6^ except that the committee
passes a resolution to refer a bill to the court, rather than to pay
compensation. These are one house resolutions. It is not until the
case is transmitted back to Congress from the commissioners that the
bill embodying the award is passed by both houses. Congress has
followed the recommendation of the commissioners in nearly all of
about 100 reference cases since World War II.66
The reference procedure is very little used now. It has been a
year since the last case was referred to the court, and only five
reference cases are under consideration in the Senate.6' This may be
due, in part, to the unfamiliarity of the legal profession with the
reference procedure, and to the fact that congressional committees aye
more likely to decide the cases themselves.
Although there is still some debate about what is meant by an
"equitable claim or a gratuity," the case that seems to embody the
majority view is Burkhardt v. United States, where the Court said that
[T]he term "equitable claim" as used in 28
U.S.C. 2509, is not used in a strict technical
sense meaning a claim Involving consideration
of principles of right and justice as adminis-
tered by courts of equity, but the broader
moral sense based on equitable considera-
tions.68
Many congressional reference cases have since read Burkhardt's defini-
tion of equity Into the statute. In spite of the expansiveness of the
standard, it by no means indicates that a favorable recommendation is
made in all congressional reference cases.6'
Many recent congressional reference decisions have added a more
concrete standard to help separate the equitable claims from the
merely gratuitous — that the plaintiff base its claim on some unjus-
tified act or omission of the government.^ The requirement of fault
is still a matter of controversy.
A-16
-------
INDEMNIFICATION STUDY
Appendix A
D. Special Jurisdication
The last category of private bills are those that confer special
Jurisdication on the Court of Claims for a particular case. These
usually remove some affirmative defense of the government such as res
judicata or an expired statute of limitations. The judgment of the
court is final in these cases, and they are not transmitted back to
Congress for approval or appropriations.
The Supreme Court has held that directing the Court of Claims to
decide a case on the merits after removal of a legal impediment does
not unconstitutionally interfere with the separation of powers.7^ The
basis of congressional authority is the power to determine and pay
debts. In Sioux Nation, the Supreme Court held that precedents
"clearly establish that Congress may recognize its obligation to pay a
moral debt not only by direct appropriation, but also by waiving an
otherwise valid defense to a legal claim against the United
States."72
An example of a past special Jurisdication bill granting indemni-
fication is the Hizokami Brothers spinach case. The private law
directed the court to decide the case notwithstanding the fact that it
clearly fell within the misrepresentation exception of the Federal
Tort Claims Act.7^ The court awarded damages for the mistaken deter-
mination that the spinach was contaminated with a pesticide.
In addition to waiving a defense, special jurisdiction bills some-
times direct the Court of Claims to take particular factors into con-
sideration in its decision. Such instructions are consistent with the
rule that the sovereign may condition its consent to suit.7^ The bill
providing compensation for losses due to the ban of Trie-treated
children's sleepwear Is a current example of a special jurisdiction
bill containing criteria for decision.'5
A-17
-------
US DA DISEASED LIVESTD^^INDEMNIFICATION PROGRAM
ESTIMATED OBLIGA^TONS FOR INDEMNITIES
(Dollars In Thousands)
1970
1971
1972
1973
1974
1975
1976
TQ
1977
1978
1979
1980
1981
- Actual
- Actual
- Actual
- Actual
- Actual
- Actual
- Actual
- Actual
- Actual
- Actual
- Actual
- Actual
- Actual
Brucellosis
$ 1.245
1,184
3,175
4,163
4,771
8,354
12,821
2,271
10,784
14,768
14,666
18,963
26,548
Cholera
$3,748
2,064
788
3,129
46
479
912
2
—
Scrapie
$ 55
33
17
21
12
28
38
2
10
124
68
347
358
Tuberculosis
$ 153
456
678
977
627
1,072
1.446
167
1,579
1,761
329
2,102
1,293
Newcastle
$ "~~"~
6,410
24,393
507
41
55
1
854
823
740
416
Total
$ 5,201
3,738
11,068
32,683
5,963
9,495
14,839
3,353
13,229
16,653
15,886
22,152
28,615
TOTALS
$123,713
$11,168 $1,113
$12,640
$34,240
$182,875
Source: Division of Veterinary Services, Animal and Plant Health Inspection Service, USDA.
-------
INDEMNIFICATION STUDY
Appendix A
III. NON-EPA INSTANCES OF INDEMNIFICATION
A. Agricultural Indemnities — Subsidies and Incentives
More indemnification programs are connected with agriculture than
any other single enterprise. To a certain extent, this fact reflects
a policy favoring agricultural subsidies, in keeping with the national
interest in a healthy agriculture sector. It may also be a reflection
of the political power of farm interests. There are also more speci-
fic and substantive reasons for the indemnity programs, however. One
is that the losses to which the agricultural indemnities respond are
frequent and predictable, both in the sense that fanners are certain
to suffer losses of these types and that there has been sufficient
experience to estimate the size and frequency of the losses. When
there are a large number of losses, it is much more efficient to set
up a program that can respond automatically rather than treating each
one separately.
Pesticide problems are the basis for most of the agricultural
indemnity programs. All except the animal disease program address the
problem of a farmer caught between the conflicting goals of using
pesticides to increase productivity and protecting the public food
supply from contamination by dangerous residues. When food is barred
from the market because of unacceptable pesticide residues, notwith-
standing the farmer's lawful use of them, part of the loss has con-
sistently been shifted to the government.
1. Animal Disease
The animal disease program is the oldest statutory indemnity
program,76 and probably the most costly. It accompanies and supports
a cluster of programs aimed at the erradication of particular live-
stock diseases. The money paid for destruction of diseased and
exposed animals acts as an incentive for farmers to cooperate.77 Over
the years the USDA has adjusted the payments so that they are high
enough to induce support, but not so high as to make diseased animals
a "good business."7° There this is still an incentive for farmers to
keep the herds healthy.
The program operates under regulations written by the
USDA,79 but Is jointly administered with the states. Animals are
Inspected by either state or federal offlcals and, if diseased, are
usually appraised at the same time. Indemnification payments are then
made quickly. The amount varies, but is usually less than the
replacement value of the animal. Regulations set a maximum dollar
amount for each disease, with a wide variety of factors to be consid-
ered. For example, cattle exposed to tuberculosis are eligible for up
to $450 per animal, but combined federal and state payments, plus
A-18
-------
INDEMNIFICATION STUDY
Appendix A
salvage, If any, may not exceed the appraised value. " The amounts of
the awards may be challenged, though they rarely are.81
The program is funded by regular appropriations. Each program
is separately considered in internal budget development, but none is a
line item in the budget sent to Congress. Therefore, up to ten per-
cent of appropriated funds may be shifted from one program to another
without congressional approval, if estimates prove incorrect. If the
Secretary of Agriculture declares a national disease emergency, funds
for erradication and indemnification may be borrowed from the
Commodity Credit Corporation, and later paid back by appropriations.
Administrative costs are estimated at 10-12 percent of the amount paid
out.82 As the table on the previous page shows, the total amount paid
out is substantial.
c
2. Dairy and Beekeepers
The dairy&3 and beekeeper*^ programs are not designed to
induce behavior changes but to compensate for pesticide-related
losses. The beekeepers program compensated for the damage done to
bees by pesticides developed as substitutes for DDT.**5 Congress
apparently decided last summer that the beekeepers had had enough time
to adjust to the change and find ways to protect their bees. The
program was thus officially terminated, after OMB had cut its budget
for several years. The dairy program remains, but is much reduced
from its peak, when $350,000 a year was distributed to dairy farmers
who were prevented from marketing milk that had unacceptable pesticide
residues through no fault on their part. Apparently, current pestici-
des are not as persistent, and do not prevent milk from being marketed
as often, because only $40,000-50,000 a year is now being paid."
One recent contamination incident threatens to upset the
orderly operation of the dairy program. Unacceptable amounts of hep-
tachlor appeared in milk from cows in Hawaii. The source appears to
be pineapple wastes. Pineapple companies have been given a temporary
exemption to use heptachlor to control mealy bugs. It was thought
that mulching and aging the skins and tops reduced the heptachlor con-
tent sufficiently. It is not clear whether the process was faulty or
whether the feed had simply not been aged long enough. The immediate
problem is that there is not nearly enough money in the indemnity
budget to cover the potential claims of millions of dollars. The
Hawaii congressional delegation is investigating possbilities for aid
to the dairy farmers.^
Both programs were administered by state and county agri-
culture stabilization and conservation committees under regulations
promulgated by the USDA.88 The delegation of authority only extended
to implementation, however, and the local committees were not per-
mitted to make modifications or waivers. Payments were made quickly,
from funds Included in the USDA's regular appropriations.
A-19
-------
INDEMNIFICATION STUDY
Appendix A
3. The Cranberry Scare of 1959
Early in November of 1959, then Secretary of Health, Educa-
tion, and Welfare Arthur Fleming, held a news conference to urge the
public not to buy cranberries grown in Washington or Oregon. It was
feared that these cranberries were contaminated with the herbicide
aminotriazole which had been found to cause cancer in rats. The
Secretary could not assure the public that cranberries from other
states were free of contamination. As a result, virtually the entire
cranberry crop remained unsold at the end of the year. This threat-
ened ruin for the cranberry growers, as almost all of their sales are
for Thanksgiving and Christmas.
Apparently, the farmers were able to mobilize a great deal of
political pressure and, by the beginning of February, 1960, the USDA
had designated the cranberry producing states as areas in which the
Farmer's Home Loan Administration could make emergency loans. During
the next six months, $333,000 was loaned to 30 growers.89 At the end
of March, the White House announced an indemnity program to compensate
the growers of untainted cranberries who had been unable to sell their
crops due to the scare over the contaminated ones. Under the program,
the growers of contaminated cranberries were ineligible to receive »
indemnification. The USDA used an established fund for the
payments.^0
This was the first time that the FDA had used its power of
national publicity to warn the public of an imminent danger. The
reaction of the consumers was so immediate and overwhelming, that just
the threat of similar publicity has since motivated many other com-
panies to institute voluntary recalls.^l
B. Indemnity Awarded for Government Error
1. Mizokami Spinach92
During the summer of 1962, the Mizokami Brothers shipped ten
railroad cars of spinach east from Colorado. The FDA conducted
routine paper chromotography tests to check for pesticide residues.
Eight of the carloads showed no residues in excess of established
tolerances, but tests on two carloads indicated unacceptable residues
of heptachlor. Shortly after Mizokami was notified, the two car loads
were seized. The growers could not understand what had produced the
contamination because they had not used heptachlor. Mizokami hired
two chemists (one was the inventor of heptachlor) to do additional
tests and to testify for them. Following a hearing, the FDA sent
samples to Washington, for a more sophisticated gas chromotography
test. The new tests showed no trace of heptachlor, and in a letter of
September 24, 1962 the FDA admitted that the original paper chromo-
tography test must have been in error.
A-20
-------
INDEMNIFICATION STUDY
Appendix A
In the interim, the Mizokami Brothers had plowed under a good
deal of their 1962 crop, presumably because spinach is a fragile com-
modity that could not be stored until the debate had concluded. The
rest of the 1962 crop was sold at reduced prices.
In 1964, a private bill conferred special jurisdiction upon
the Court of Claims to hear and render judgment on the claims of the
Mizokami Brothers.93 Although the misrepresentation exception would
have precluded recovery under the Federal Tort Claims Act, the court
held that the private law had waived sovereign immunity and conceded
liability, but had left leaving causation and damages for the court to
decide.
The Mizokami Brothers were awarded $301,974.33 in 1969 from
the Court of Claims Judgment Fund. Most of the award ($227,399) was
for the spinach that was plowed under in 1962, and the majority of the
rest for the reduced price sales in 1962 ($49,904). The court also
awarded attorney's fees and court costs.
2. Marlin Toy Co.
The only other case in which the Commissioners of the Court of
Claims recommended indemnification for an agency error was to the
Marlin Toy Company. Two of Marlin's toys were included on a holiday
list of banned products for 1973, one described as "without plastic
pellets."9* This was an error as only the version with pellets had
been banned, the other was approved. Marlin was unable to rectify the
mistake in time for Christmas sales. Because this was clearly a
misrepresentation, Marlin had no cause of action under the Federal
Tort Claims Act, and instead went to Congress for relief.
CPSC supported the private bills referring the case to the
Court of Claims under the congressional reference procedure.9^ On
December 19, 1978, the Review Panel advised Congress that Marlin had
an equitable claim of $40.000 against the government, payment of which
would not be a gratuity.9" A bill awarding this amount to Marlin was
passed by the Senate during the 96th Congress, but not acted upon by
the House. No bill was reintroduced during the current Congress.
The Incident also led Congress to add § 5(h) to the Consumer
Product Safety Act.97 It creates special exemptions that allow
claims based on misrepresentation and deceit otherwise barred by the
FTCA. However, the claim may not be made with respect to any agency
action, as defined by the Administrative Procedure Act. Although ori-
ginally applicable only to claims arising before January 1, 1978, the
provisions have since been indeflnately extended. It is not clear
that the Marlin Toy Company could have been successful in making a
claim under §5(h). Although the lists of banned hazardous substances
are the result of adjudication or rulemaking, they are considered by
the Commission to be official actions as they are published in the
Federal Register after a vote by the Commission.9**
A-21
-------
INDEMNIFICATION STUDY
Appendix A
C. Indemnification Being Considered for Policy Change
1. Tris Ban
In 1971 and 1974 the Department of Commerce and its successor
in administering the Flammable Fabrics Act, the Consumer Product
Safety Commission (CPSC), promulgated flame-retardant performance
standards for children's sleepwear.^ Early in 1976, the Environ-
mental Defense Fund petitioned the CPSC to require labeling for Tris-
treated garments, based on the results from salmonella mutagenicity
tests. On April 7, 1977, after analyzing two rodent feeding studies
done by the National Cancer Institute, the CPSC declared children's
apparel containing Tris to be banned hazardous substances.
The impact of the Commission's action did not fnll evenly on
the industry. On the upper end of the manufacturing chain, the chemi-
cal production of Tris had apparently ceased by the time of the ban so
that the chemical companies had little inventory.*00 On the lower
end, the repurchase provision of the Federal Hazardous Substances Act
worked well.^i The goods held by customers were repurchased by
retail stores, their merchandise was repurchased by distributors, and
the distributors returned it to the manufacturers. There the flow ^
stopped. The American Apparel Manufacturers Association brought suit
to compel the CPSC to extend the original ban from just garments to
include all fabric treated with Tris. Under court order, the Commis-
sion did so. But when it tried to enforce the repurchase provision by
having the fabric mills buy from the manufacturers, the mills chal-
lenged on the grounds that the procedure used to extend the ban to
them had lacked required notice and comment opportunities. Eventually
the suits were settled, with the mills agreeing not to challenge the
ban on fabric any further, if the CPSC would not enforce the repur-
chase provisions against them. ^2 Suits brought to enforce private
rights of repurchase also failed. This left the manufacturers holding
the entire inventory. According to testimony in the Senate, the manu-
facturers are a group of about 100 companies, each having annual sales
of under $10 million. The total loss in inventory due to the ban is
now estimated at between $50.1 and $56 million.^
A public bill to reimburse businesses adversely affected by
the Tris ban passed both Houses, but was vetoed by President Carter in
1978. A new legislative proposal would confer special jurisdiction on
the Court of Claims to hear and render judgment on claims for losses
resulting from the Tris ban.104 it does not concede federal liabil-
ity. The bill directs the court to consider many specific factors in
determining the validity of the claims and the extent of losses.
Among them are:
A) Whether reasonable alternatives to Tris existed at the
time of the flammabllity standard;
A-22
-------
INDEMNIFICATION STUDY
Appendix A
B) Whether the claimant could or should have tested Tris for
chronic hazards;
C) The degree of good faith of the claimant in complying with
the flammability standard, the ban, and export provisions;
and
D) The extent to which the claimant relied in good faith on
assurances from suppliers.
Chemical manufacturers of Tris may not make claims under this
statute. No compensation is allowed for lost profits, proceeds from
distress sales, attorneys' fees, or interest. Although special jur-
sldiction bills often give the Court of Claims some direction, the
Tris ban bill is unique in the specificity of its instructions. Most
of the factors are aimed at sorting out the equities of the compli-
cated factual situation.
2. Cyclamate Ban
Those holding stocks of cyclamate-sweetened goods when it was
banned in 1969 have been trying ever since to obtain Indemnification
for their losses. The House passed a bill in 1971 that would have
given the Court of Claims jurisdiction to hear claims, and specially
appropriated an estimated $120 million for payment.10' No bill passed
the Senate. 1-06 Several years later however, the Senate passed and
conveyed to the Court of Claims a congressional reference on the issue
of relief for the California Canners and Growers Association.1^
After years of motions, extended discovery, and briefs, the opinion of
the trial commissioner is expected soon.
The arguments revolve around the fact that cyclamates had been
on the FDA's GRAS list (Generally Recognized As Safe). The Industry
claims that it reasonably relied on that assessment of the substance,
and that it had insufficient warning of an adverse action. The
government argues that the listing implied no conclusive determina-
tion, and that the industry was well aware of the growing concerns
about the substance.108 ^8 U8ualt the equities of the situation are
not clear, and depend upon an examination of all of the facts.
Even if the trial commissioner recommends an award, Congress
will have to pass a special appropriation to fund it. Thirteen years
have passed since the loss.
D. Swine Flu Vaccine Program—Specific Incentive Situation
The possibility that liability problems would drive vaclne manu-
facturers out of the business had been a matter of concern at HEW for
some time. During the spring of 1976, as the agency planned a mass
innoculation against swine flu, it became a more immediate and more
A-23
-------
INDEMNIFICATION STUDY
Appendix A
critical issue. Vaccinating tens of millions of people with a new
vaccine has an unquantified, but certain risk of serious side effects
for some of them. In addition to this, the insurance companies
worried about groundless and ultimately unsuccessful suits that they
would have to face.
HEW's response to drug company concerns was proposal to assume the
duty to warn through the procurement contracts. The insurance com-
panies ultimately refused to provide liability coverage on this basis.
Congress discussed indemnity legislation, but failing to agree on a
solution, asked manufacturers to resume their negotiations. Into that
deadlock dropped news of Legionnaires' Disease, which was suspected for
four days of being caused by swine flu. New legislation was intro-
ducted giving victims a right to sue the federal government on the
model of the Federal Tort Claims Act, instead of providing an indem-
nity. In less than two weeks the bill was passed by both Houses and
signed by President Ford.^9
In a fairly complicated arrangement, the government agreed to
accept the vaccine manufacturers' potential strict product liability
for personal injuries due to the vaccine. Under this scheme, an
Injured victim is permitted to sue only the federal government under
the law of his state. If negligence is proved, then the government
can in turn recover from the parties responsible for the negligence.
The bill also eliminated the possbllity that the manufacturers could
earn a profit on the swine flu vaccine sold to the government.
Vaccines are a special case, probably not duplicated in situations
under EPA's jurisdiction. They are important to public health, highly
regulated, and produced by only a few manufacturers because the risks
are so great. In spite of the most thorough adherence to government
specifications and the greatest degree of care, vacines can cause a
certain number of severe adverse reactions in the vacinated popula-
tion. Swine flu is the only vaccine for which the federal government
has assumed liability, however.110
There is a national interest in having enough vaccine to prevent
an epidemic (even though a swine flu epidemic did not develop).
Indemnification was used as an incentive to induce the production of
swine flu vaccine, rather than as a subsidy to the entire industry.
There also is an element of joint responsibility in this situation,
since vaccine making is heavily regulated, and public health is a tra-
ditional concern of the federal government.
The Department of Justice has administered the entire swine flu
compensation program, Including claims processing, settlement and
litigation. The administrative costs are paid out of the regular DOJ
budget, and the awards are paid in the same manner as Federal Tort
Claims Act awards.111
A-24
-------
INDEMNIFICATION STUDY
Appendix A
As of December 18, 1981, over 4,000 claims totaling $2.95 billion
dollars had been filed. Of these, 2,773 (for $1.91 billion) have been
denied, and 263 settled administratively (for $6.24 million, rather
than the $115 million claimed). Some 1,500 suits have been filed; 845
of them still pending, 341 dismissed, 234 settled (for $22 million
rather than the $241 million claimed). DOJ stipulated Co liability in
35 cases resulting In awards of $13 million. Sixteen judgments have
been for plaintiffs (most are being appealed) and 74 judgments have
been for the defendant.^^
E. Indemnity in the Form of an Ongoing Insurance Program
1. Arts and Artifacts
The Arts and Artifacts Indemnity Act113 enables the federal
government to indemnify American museums for losses Incurred In exhi-
biting internationally loaned art or artifacts. It was passed after
the previous Congress enacted special legislation for two specific
exhibitions. Interested parties apparently convinced Congress that
the problem needed a more institutionalized solution. The cost of
insurance had grown to 2/3 to 3/4 of the total cost of putting on an
exhibit, and many institutions could not afford such premiums.*^
The program is administered under regulations promulgated by
the National Endowment for the Arts. A rigorous application process
insures that most of the 40 final applicants a year are accepted,
because those who could not succeed are weeded out before they com*
plete the process. Applications concern such things as packing,
shipping, climate control, and security arrangements rather than
aethetic merit are or political significance. The Endowment's Inter-
est in risk reduction prompts requests to add night guards and impose
other conditions. Small institutions which could not handle the
stringent requirements alone often go to the International Exhibition
Foundation which gives them advise as well as grants. The exhibitor's
valuation of the objects is reviewed by an indemnity advisory panel
and then by the Federal Council on Arts and the Humanities. A con-
tract between the Council and the museum pledges the full faith and
credit of the United States. Funding is provided through special
appropriations. When a loss occurs it will be investigated by the
government and an assessment certified to Congress. So far, no losses
have exceeded the amount of the deductible.
The federal indemnification program does not insure the entire
amount of the exhibit. The institutions are required to obtain insur-
ance from private sources to cover a deductible (which ranges from
$15,000-$50,000 depending on the value of the exhibit), and the risks
in excess of the federal limit of $50 million per exhibit. The pro-
gram may not obligate more than $400 million at any one time.
Although no payments have yet been made, the fact that the museums do
not have to purchase insurance for the amount covered by federal
A-25
-------
INDEMNIFICATION STUDY
Appendix A
indemnification had saved them an estimated $800 million in insurance
premiums by the summer of 198I.11*
2. Defense Contracts
Although a contingent and uncertain liability like indemnifi-
cation is usually prohibited by the Anti-Deficiency Act, Congress and
the President have made exceptions for contracts that "facilitate the
national defense" and are "unusually hazardous or nuclear in
nature."I1*1 Evidently the authority to use indemnification clauses in
such contracts has been used frequently, but no claims have actually
been paid out by the government. In this case also, the promise of
the federal government has been sufficient to accomplish the goal of
obtaining services and materials, important to the national defense,
that would otherwise not be obtainable because of a possibly unaccep-
table risk to the manufacturer.
Some agencies have apparently agreed to indemnify even without
express authority if the contractors are needed to perform a vital
service, such as evacuating plane loads of people from Viet Nam. The
Comptroller General sanctioned one such use of indemnification in a .
contract between the General Services Administration and a public
utility.117
3. Non-Defense Contractors
Some agencies, arguably within the scope of the "unusually
hazardous" exception carved out by the Executive Order, have not used
it because they did not want to label their activities. This has been
a particular problem for NASA. Its contractors demand some protection
against the possible catastrophic loss connected with an event such as
the Space Shuttle crashing In a populated area. In its authorization
of 1981, NASA received a special authority to indemnify users of the
space shuttle against claims for property damage, death or injury to
the extent the claims are not compensated by liability insurance, and
to the extent that they do not result from negligence or willful mis-
conduct by the user.
An interagency task force, co-sponsored by the Office of
Federal Procurement Policy and NASA, has recently completed its draft
report and recommended expansion of indemnification authority to cover
possibilities of catastrophic loss, which it defines as loss beyond
that can be covered by reasonably available insurance, but not neces-
sarily resulting from "unusually hazardous" activity.11** The National
Association of Manufacturers is sponsoring a related bill to indemnify
all suppliers of products to the government.11^
EPA has had several requests from contractors for indemnifica-
tion provisions in their contracts with the Agency. A potential "host
site" asked that the contractors who want to test new pollution con-
A-26
-------
INDEMNIFICATION STUDY
Appendix A
trol equipment In its plant be Indemnified by the government against
•possible damage to the plant or Interrupted operations. This request
was denied. In at least one case, EPA paid the insurance premiums on
a special policy for a contractor conducting field operations on haz-
ardous material cleanup. The premium cost $200,000 for $1,000,000
worth of coverage.
In one case EPA has agreed to directly indemnify contractors
hired to clean up hazardous waste sites under Superfund. Private
insurance for the contractors would have added 102 to the cost of the
contracts. (Insurance coverage of $10,000,000 would have cost one
contractor $400,000 for one year, and another $320,000 for a
$25,000,000 coverage for a year.)120 EPA designed an indemnification
clause to avoid Anti-Deficiency Act problems.121 The clause covers
all damages, even those arising from contractor negligence but, it is
conditioned on the availability of funds of the time a claim is made,
with the specific proviso that EPA will not promise to seek funds, and
that Congress is under no obligation to provide them. Apparently this
was sufficient to satisfy the concerns of the contractors.122 The
Indemnity clauses are excess liability only, all contractors are
required to purchase one million dollars of commercial liability
insurance which must be exhausted first.
A-27
-------
INDEMNIFICATION STUDY
Appendix A
FOOTNOTES TO APPENDIX A
177 Am. Jur. 2d "United States" § 117 (1975); 91 C.J.S. "United
States" § 184 (1955).
2See generally 35 Am. Jur. 2d "Federal Tort Claims Act" §§ 1 et
seq. (1967); Annot., "Federal Tort Claims Act," 1 A.L.R.2d 222, § 1
(1948); Jayson, Handling Federal Tort Claims (Mathew Bender 1964);
H.R. Rep. 1287, 79th Cong., 1st Sess. (1945); H.R. Rep. 2245, 77th
Cong., 2d Sesa. (1942); Hearings on Bills to Provide for the
Adjustment of Certain Tort Claims Against the United States, H.R.
5373 and H.R. 6463. House Judiciary Committee. 77th Cong., 2d Sess.
(1942); and Tort Claims Against the United States, Subcommittee of the
Senate Judiciary Committee, 76th Cong., 1st Sess. (1940).
Specifically, 28 U.S.C. § 1346(b) confers jurisdiction on the
federal district courts and under 28 U.S.C. §§ 2671 et seq. the United
States has consented to suit, subject to the following thirteen excep-
tions under 28 U.S.C. § 2680:
(a) Any claim based upon an act or omis-
sion of an employee of the Government, exer-
cising due care, in the execution of a statute
or regulation, whether or not such statute or
regulation be valid, or based upon the exer-
cise or performance or the failure to exercise
or perform a discretionary function or duty on
the part of a federal agency or an employee of
the Government, whether or not the discretion
involved be abused.
(b) Any claim arising out of the loss,
miscarriage, or negligent transmission of let-
ters or postal matter.
(c) Any claim arising in respect of the
assessment or collection of any tax or customs
duty, or the detention of any goods or mer-
chandise by any officer of customs or excise
or any other law enforcement officer.
(d) Any claim for which a remedy is pro-
vided by sections 741-752, 781-790 of Title
46, relating to claims or suits in admiralty
against the United States.
(e) Any claim arising out of an act or
omission of any employee of the Government in
administering the provisions of sections 1-31
of Title 50, Appendix.
A-28
-------
INDEMNIFICATION STUDY
Appendix A
(f) Any claim for damages caused by the
imposition or establishment of a quarantine by
the United States.
(g) [Repealed]
(h) Any claim arising out of assault,
battery, false imprisonment, false arrest,
malicious prosecution, abuse of process,
libel, slander, misrepresentation, deceit, or
interference with contract rights. Provided,
That, with regard to acts or omissions of
investigative or law enforcement officers of
the United States Government, the provisions
of this chapter and section 1346(b) of this
title shall apply to any claim arising on or
after the date of the enactment of this
proviso, out of assault, battery, false
imprisonment, false arrest, abuse of process,
or malicious prosecution. For the purpose of
this subsection, "investigative or law
enforcement officer" means any officer of the
United States who is empowered by law to exe-
cute searches, to seize evidence, or to make
arrests for violations of Federal law.
(i) Any claim for damages caused by the
fiscal operations of the Treasury or by the
regulation of the monetary system.
(j) Any claim arising out of the combatant
activities of the military or naval forces, or
the Coast Guard, during time of war.
(k) Any claim arising in a foreign
country.
(1) Any claim arising from the activities
of the Tennessee Valley Authority.
(m) Any claim arising from the activities
of the Panama Canal Company.
(n) Any claim arising from the activities
of a Federal land bank, a Federal intermediate
credit bank, or a bank for cooperatives.
^In adjusting claims administratively, agency heads must follow
any regulations prescribed by the Attorney General, and obtain his
prior written approval for any award greater than $25,000* If admi-
A-29
-------
INDEMNIFICATION STUDY
Appendix A
nistrative adjustment proves unsuccessful and a lawsuit if filed, the
Attorney General, or a designee of the Attorney General, may settle
the case out of court.
4Supra note 2. 28 U.S.C. § 2680(h).
518 U.S.C. § 2680(a). See Reynolds, "The Discretionary Function
Exception of the Federal Tort Claims Act," 57 Geo. L.J. 81 (1968); and
James, "The Federal Tort Claims Act and the Discretionary Function
Exception: The Sluggish Retreat of an Ancient Immunity," 10 U. Fla.
L. Rev. 184 (1957).
6Blessing v. United States. 447 F. Supp. 1160, 1178 (E.D. Pa.
1978).
7See generally. 35 Am. Jur. 2d "Federal Tort Claims Act" §§ 15 et^
seq. (1967); Annot., "FTCA-Permits, etc.," 35 A.L.R. Fed. 481 (1977);
Annot., "Law Enforcement as Discretionary," 36 A.L.R. Fed. 240
(1978); Annot., "Maintenance of Public Property," 37 A.L.R. Fed. 537
(1978); Annot., "Federal Tort Claims Act," 99 A.L.R.2d 1016 (1965);
Annot., "FTCA-Exceptions," 6 L. Ed. 2d 1428 (1962).
8For example, in Eastern Airlines v. Union Trust Co., 221 F.2d 62
(D.C. Cir. 1955), it was held that the government's determination of
the location of an air traffic control tower is a discretionary
planning decision which falls within the discretionary function excep-
tion to the FTCA. However, an operator's negligence in clearing two
planes for landing on the same runway at the same time is not pro-
tected by the discretionary function exception.
'For example, a party may be entitled to rely upon the
government's non-negligent handling of a rescue operation at sea or
the proper enforcement of mine safety standards under a safety inspec-
tion program. See, e.g., Eastern Airlines, id. and Raymer v. United
States, 482 F. Supp. 432 (W.D. Ky. 1979).
10Barton v. United States. 609 F.2d 977 (10th Cir. 1979)
(temporary discontinuance of livestock grazing permits, Bureau of Land
Management); Thompson v. United States. 592 F.2d 1104 (9th Cir. 1979)
(permit to sponsor motorcycle race on public property, Bureau of Land
Management); Myers & Myers, Inc. v. United States Postal Service, 527
F.2d 1252 (2d Cir. 1975) (award of transportation contract); Coast-
wise Packet Co. v. United States, 398 F.2d 77 (1st Cir. 1968), cert.
denied, 393 U.S. 937 (1968) (certificate of inspection, Coast Guard);
Boruski v. Division of Corp. Finance, 321 F. Supp. 1273 (S.D.N.Y.
1971) (registration approval, Securities and Exchange Commission);
Hooper v. United States, 331 F. Supp. 1056 (D. Conn. 1971) (permit to
construct power-generating station, Corps of Engineers); Marr v.
United States, 307 F. Supp. 930 (E.D. Okla. 1969) (pilot's license
and certificate of convenience and necessity, Civil Aeronautics
Board).
A-30
-------
INDEMNIFICATION STUDY
Appendix A
uln Griffin v. United States. 500 F.2d 1059 (3d Clr. 1974), a
decision involving negligent inspection and approval of oral polio
vaccine, the court found that the inspector's action was of a
non-policy, professional nature and that the discretionary function
exception was not applicable.
In a mine safety inspection case the discretionary function excep-
tion was held not applicable to protect the government against liabi-
lity resulting from injury when an inspector allowed an extension of
time for the correction of an obviously hazardous condition. Raymer
v. United States, supra note 9. Noting the clear statutory mandate
for vigorous enforcement, the court held that the inspector's function
was operational in that he had no discretion to grant a time extension
and he was negligent to do so.
1228 U.S.C. § 2680 (h). United States v. Neustadt. 366 U.S. 696
(1961). See generally Tort Claims Against the United States, Subcom-
mittee of the Senate Judiciary Committee, 76th Cong., 1st Sess. 39
(1940); 35 Am. Jur. 2d "Federal Tort Claims Act" § 46 (1967); Annot.,
"FTCA-Deceit," 30 A.L.R. Fed. 421 (1976); Annot., "FTCA-Exceptions," 6
L. Ed. 2d 1465 (1962).
"id..
14Cenna v. United States. 402 F.2d 168 (3d Cir. 1968).
has been a particular reluctance to apply the
misrepresentation exception to personal injury claims of medical
malpractice or negligent flight Information. United States v.
Neustadt, 366 U.S. 696 (1961); Redmond v. United States, 518 F.2d 811
(7th Cir. 1975); Marlval. Inc. v. Planes. Inc.. 306 F. Supp. 855 (N.D.
Ga. 1966).
16Ingham v. Eastern Airlines. Inc.. 373 F.2d 227 (2d Cir. 1967),
cert, denied, United States v. Ingham, 389 U.S. 931 (1967); United
Air Lines. Inc. v. Wiener, 335 F.2d 379 (9th Cir. 1964), cert.
dismissed.. 379 U.S. 951 (1964);
17Ramirez v. United States. 567 F.2d 854 (9th Cir. 1977).
example, in medical malpractice cases an erroneous diagnosis
communicated to the patient is simply a misrepresentation and falls
within the exception to the waiver of immunity. If the physician were
simply a diagnostic consultant and not responsible for the care of the
patient, the government would not be liable for the misrepresentation.
But if, as is more often the case, the diagnosing physician is also
the treating physician, who is under an affirmative duty to give
proper treatment, liability could be based on the negligent conduct of
the physician in treating the patient. Scanwell Laboratories. Inc.
v. Thomas, 521 F.2d 941 (D.C. Cir. 1975); Covington v. United States.
303 F. Supp. 1145 (N.D. Miss. 1969).
A-31
-------
INDEMNIFICATION STUDY
Appendix A
a case where the exception was found applicable, a grain
importer unsuccessfully claimed that the government had negligently
processed a request for amendments to its import permits and had
failed to advise of a written policy prohibiting the types of imports
which he planned to ship to the United States. Because of these
alleged events, the claimant had its corn shipped to this country only
to have it destroyed upon arrival. Cargill, Inc. v. United States,
426 F. Supp. 127 (D. Minn. 1976).
example of a taking is the situation in Kaiser Aetna v.
United States, 44 U.S. 164 (1979), in which the Supreme Court deter-
mined under the facts of the case that- the government could not, with-
out paying compensation, force the petitioner to admit the public to a
private lagoon which had been connected to a public bay. A second
taking example is Pete v. United States, 531 F.2d 1018 (Ct. Cl. 1976),
in which the government condemned all the land surrounding a lake and
prohibited use of gasoline powered engines on the lake. The court
concluded that the government's action constituted inverse condem-
nation compensable under the Fifth Amendment because it effectively
destroyed the usefulness of three large houseboats permanently placed
upon the lake. This was not a pure regulatory taking, though, because
the regulatory restriction accompanied the physical taking of real *
estate. A third example is Hernandez v. City of Lafayette, 643 F.2d
1188 (5th Cir. 1981), in which the Court of Appeals remanded for trial
a case where a city allegedly took property by refusing to rezone land
because it was going to build a highway through the property and
wanted to keep the depressed value of the land that the current zoning
represented.
20aSweet v. Rechel. 159 U.S. 380 (1895). See 58 Am. Jur. 2d
"Nuisance" § 206 (1971); 35 Am. Jur. 2d "Food"HT65-66 (1967).
21South Terminal Corp. v. EPA. 504 F.2d 646, 678 (1st Cir. 1974)
(transportation control plan which mandated a 40Z reduction in
available off-street parking spaces held not a taking).
22Id. at 679.
23Andrus v. Allard. 444 U.S. 51, 100 S. Ct. 318, 326 (1979)
(substantial restriction on Indian artifacts containing feathers of
protected birds not considered a taking because owner continued to be
allowed to possess and transport property, to donate or devise the
protected birds and could exhibit the artifacts for an admission
price). The court in And r us also -stated:
[T]he denial of one traditional property right
does not always amount to a taking. At least
where an owner possesses a full "bundle" of
property rights, the destruction of one
"strand" of the bundle is not a taking,
A-32
-------
INDEMNIFICATION STUDY
Appendix A
because the aggregate must be viewed in its
entirety. . . When we review regulation, a
reduction in the value of property is not
necessarily equated with a taking. . . At any
rate, loss of future profits — unaccompanied by
any physical property restriction — provides a
slender reed upon which to rest a takings
claim. Prediction of profitability is essen-
tially a matter of reasoned speculation that
courts are not especially competent to per-
form. Further, perhaps because of Its very
uncertainty, the interest in anticipated gains
has traditionally been viewed as less compel-
ling than other property-related interests.
Id. See Appendix A page A-9, A-10.
Bosselman, D. Callies & J. Banta, The Taking Issue; An
Analysis of the Constitutional Limits on Land Use Control (Gov't
Printing Office, Washington, D.C. 1973).
25Mugler v. Kansas. 123 U.S. 623 (1887).
26The Court in Mugler stressed the fact that there was no physical
invasion of property, ruling there was no taking since it was within
the government's power to regulate property to eliminate a nuisance.
The Court stated that:
the destruction of property which is itself a
public nuisance, or the prohibition of its use
in a particular way, whereby its value becomes
depreciated is very different from taking pro-
perty for public use. ... In the one case, a
nuisance only is abated; in the other, unof-
fending property is taken away from an inno-
cent owner.
Id. at 668-69.
27Fennsylvanla Coal Co. v. Mahon, 260 U.S. 393, 413 (1922). The
Court stated:
One fact for consideration in determining such
limits is the extent of the diminution. When
It reaches a certain magnitude, in most if not
in all cases there must be an exercise of emi-
nent domain and compensation to sustain the
act. So the question depends upon the partic-
ular facts. The greatest weight is given to
the judgment of the legislature, but it always
A-33
-------
INDEMNIFICATION STUDY
Appendix A
Is open to interested parties to contend that
the legislature has gone beyond its constitu-
tional power.
28Kaiser Aetna v. United States. 444 U.S. 164 (1979); Andrus v.
Allard, supra note 23; Penn Central Transportation Co. v. City of New
York. 438 U.S. 104 (1978); Goldblatt v. Town of Hempstead, 369 U.S.
590 (1962). For a recent overview of the taking issue, see Note,
"Reexamining the Supreme Court's View of the Taking Clause," 58 Tex.
L. Rev. 1447 (1980). See also Kanner, "The Consequence of Taking
Property by Regulation." 24 Prac. Law. 65 (1978). For an evaluation
of various theoretical models for analyzing the taking question, see
Berger, "A Policy Analysis of the Taking Problem," 49 N.Y.U.L. Rev.
165 (1974). c
2^Supra note 27.
30260 U.S. at 413.
31Id. at 413-14.
32United States v. Central Eureka Mining Co.. 357 U.S. 155 (1958).
33Id._ at 168.
>ra note 23.
W^BIm
3Supra note 23.
36Agins v. City of Tiburon. 100 S. Ct. 2138 (1980). See also
Goldblatt v. Town of Hempstead, 369 U.S. 590 (1960) (prohibition of
excavation of sand and gravel below water table not a taking); Penn
Central Transportation Co. v. City of New York, 438 U.S. 104 (1978)
(denial of right to exploit air space by building office building
above Grand Central Station not a taking).
37Sierra Club v. EPA, 540 F.2d 1114 (D.C. Cir. 1976). See also
Chesapeak Bay Village. Inc. v. Costle. 502 F. Supp. 213 (D. Md. 1980)
(no taking where plaintiff's plans to develop land into subdivisions
jeopardized because of insufficient sewer facilities due to federal
government giving insufficient grant for sewage treatment facility);
and Creppel v. U.S. Army Corps of Engineers, 500 F. Supp. 1108 (E.D.
La. 1980) (taking claim rejected in action to compel Corps of
Engineers to authorize flood control project in accordance with origi-
nal plans).
38Supra note 22.
39Chevron Chemical Co. v. Costle. 641 F.2d 104 (3d Cir. 1981),
cert, denied. 101 S. Ct. 3110 (1981) (concerning §§ 3(c)(l)(D) and 10
A-34
-------
INDEMNIFICATION STUDY
Appendix A
of FIFRA). See also Mobay v. Costle, 517 F. Supp. 252 (W.D. Pa., No.
CA 79-591D, June 12, 1981), which expressly followed the holding of
the Chevron case; and Union Carbide Agricultural Products Co. v.
Gorsch (S.D.N'.Y., No, 76 Civ. 2913), in which a preliminary injunction
issued by the district court was vacated on appeal by the Second
Circuit.
40Hodel v. Virginia Surface Mining & Reclamation Ass'n,
452 U.S. 264, 101 S. Ct. 2352 (1981) (steep slopes provision did not
prohibit surface mining, but only regulates conditions of performing
mining operations). See also Hodel v. Indiana, 452 U.S. 314, 101 S.
Ct. 2376 (1981) (same situation as Hodel above except challenge cen-
tered upon prime farmland requirement); Nance v. EPA, 645 F.2d 701
(9th Cir. 1981) (taking issue not ripe because not yet established
that strip mining on Northern Cheyenne Tribe's reservation is subject
to PSD nor that the land is suitable for strip mining). See also San
Diego Gas & Electric Co. v. City of San Diego. 450 U.S. 621, 101 S.
Ct. 1287 (1981) (Supreme Court found 5-4 in challenge to city's open
space plan and rezonlng that California state court decision was not
final in that California court decided monetary damages were not
appropriate but not if there was indeed a taking).
A1United States v. Realty Co.. 163 U.S. 427, 440 (1896). The
Supreme Court has recently reaffirmed the authority of Congress to
define and pay the debts of the United States, by appropriation or by
waiving an otherwise valid defense to a claim against the government.
Sioux Nation v. United States. 448 U.S. 371, 100 S. Ct. 2716 (1980).
The right to petition the government for redress of grievances Is also
cited as authority.
42Those of the House are codified In the Rules of the Subcommittee
on Administrative Law and Governmental Relations, House Committee on
the Judiciary.
^Telephone conversation on March 26, 1982 with Bill Shattuck,
Counsel to House Judiciary Committee, Subcommittee on Administrative
Law and Governmental Relations. The Comptroller General may also
recommend bills, though the authority is rarely used, 31 U.S.C. § 236.
44Note, "Private Bills in Congress," 79 Harv. L. Rev. 1684, 1689
(1966).
one member objects strongly to a bill, he can usually per-
suade one of his colleagues to object along with him. Thus, in prac
tice a single objection can kill a private bill, according to
Shattuck, supra note 43.
46In the 95th Congress 1,552 private bills were introduced, 283
passed. In the 96th Congress 1,334 were introduced, of which 181
passed. The 97th Congress is not yet over, but so far 962 private
A-35
-------
INDEMNIFICATION STUDY
Appendix A
bills have been introduced and 75 passed. The percent passing has
dropped from 18%, to 13.5% to less than 8X.
^Id. , Private Bills, supra note 44. For example, the CPSC
testified in favor of relief for Marlin Toy Co.
example, although 67% of the private bills introduced in the
97th Congress concerned immigration, 85Z of those passed were immigra-
tion bills. See also Private Bills, supra note 44 at 1693.
^'Apparently only six vetoes of private bills have been
overridden, none of them in the last-fifty years. Id. at 1693.
50ld. at 1701. °
51Sup_ra note 43.
32Act of Feb. 24, 1855, ch. 122, 10 Stat. 612.
53Cong. Globe, 37th Cong., 2d Sess., App. 2 (1862).
54Act of March 3, 1863, ch. 92 § 5, 12 Stat. 765, 766. Appeal /s
now available only on writ of certiorari.
"Repealed after the Supreme Court refused to review a judgment
on the grounds that Congress's remaining power over the court's Judg-
ments was inconsistent with its exercise of judical power. Glidden v.
Zdanok. 370 U.S. 530, 554 (1962).
5631 U.S.C. § 724a.
U.S.C. § 1491. The district courts have been given concur-
rent jurisdication on these cases where the claim is less than
$10,000. 29 U.S.C. § 1346(a)(2).
^However, the court of claims may hear appeals of FTCA cases from
federal district courts with the consent of appellees. 28 U.S.C.
§ 1504.
59"Annual Report to Congress, January 25, 1982" by Office of the
Clerk, Court of Claims.
60Glidden, supra note 55.
^In cases falling under general jurisdiction, trial commissioners
hold hearings and make recommendations to the court. 28 U.S.C. § 792
(1925).
62Glidden, supra note 55.
A-36
-------
INDEMNIFICATION STUDY
Appendix A
"Congressional Reference Cases in the United States
Court of Claims: A Historical and Current Perspective, 25 Am. U. L.
Rev. 595 (1976).
6428 U.S.C. § 2509(c); 28 U.S.C. § 1492. Congressional reference
cases are tried under rules which closely follow the regular rules of
the court. 28 U.S.C. § 2509(b). For an explanation of procedure, see
Glosser, supra note 63, at 605-17.
"For more detail on procedure, see Glosser, id.
66M.T. Bennett, "Private Claims and Congressional References," 9
A.F. Jag L. Rev. No. 6 (Nov. -Dec. 1967)
"'Telephone conversations on March 26, 1982 with Linda Nucessian,
Counsel to Subcommittee on Agency Administration, Senate Judiciary
Committee and Frank T. Peartree, Clerk, U.S. Court of Claims.
68113 Ct. Cl. 658, 667 (1949). Though the language is clear,
Burkhardt has caused confusion because the court found that the plain-
tiffs had more than a merely equitable or honorable obligation. The
more expansive definition was not necessary to reach the holding.
69168 Ct. Claims 318, Cong. Ref. No. 11-58 (1964). In Drake
America Corp. v. United States for example, the court found that a
claim for additional compensation for a contract to develop a high-
speed snow plow for the Air Force had neither a legal nor an equitable
basis within the meaning of the statute. The record did not support
the plaintiff's claim that it reasonably relied to its detriment on a
promise that additional funds would be approved, and the court found
that the plaintiff had been sufficiently compensated for the work
performed.
, Innocent Victims of the Occupation of Wounded Knee, South
Dakota v. United States. Cong. Ref. No. 4076 (June 10, 1981); Burt v»
United States. 199 Ct. Cl. 897; Glosser, supra note 63 at 620-22.
71Cherokee Nation v. United States. 270 U.S. 476 (1926), discussed
in Sioux Nation v. United States, supra note 41.
72Sioux Nation, id. at 397.
73Mizokaml v. United States. 118 Ct. Cl. 736, 414 F.2d 1375
(1969).
74Pope v. United States. 323 U.S. 1 (1944) discussed In Sioux
Nation, supra note 41, at 399.
7 ^s. 823 discussed supra at page A-22.
A-37 '
-------
INDEMNIFICATION STUDY
Appendix A
7621 U.S.C. § lU(a), 134(d)
''"Indemnification Under Animal Disease Control Programs with
Special Emphasis on Foot-and-Mouth Disease," Nasser A. Aulaqi and
W.B. Sundquist, Department of Agricultural and Applied Economics,
University of Minnesota, Economic Report ER 77-2, February 1977, at 4.
^Telephone conversation with Dr. M.J. Tillery, USDA Animal, Plant
Health and Inspection Service.
799 C.F.R. §§ 50-56.
81See Julius Goldman' c Egg City. 566 F.2d 1096 (1977).
^Estimate by Division of Veterinary Services USDA, contained in
Background Draft by ICF, Inc. "Costs Associated with USDA Animal
Disease Indemnification Programs" at 6.
837 U.S.C. 450J.
8A7 U.S.C. 135b.
"Although less harmful to humans, and less persistant in the
environment, they are more damaging to bees.
^Figures from a telephone conversation with Gerald Schiermeyer,
Agricultural Stabilization and Conservation Service, USDA.
87Telephone conversation on July 12, 1982 with Marcy Farden,
legislative aid to Rep. Akaka.
887 C.F.R. § 760.
89Gellhorn, "Adverse Publicity by Administrative Agencies," 86
Harv. L. Rev. 1380, 1408 (1973) (quoting from internal FDA
memorandum) .
year an amount equal to 30 percent of the gross receipts
from duties collected under the customs law the proceeding year is
appropriated to the fund authorized by 7 U.S.C. § 612c. It is used to
encourage agricultural exports, and domestic consumption, and to
"reestablish farmer's purchasing power." The food stamp and school
lunch programs are run under the authority of this section.
Approximately $8.5 million was paid to 12 claimants representing 1,215
growers, based on a schedule of prices per barrel of cranberries.
91Gellhorn, supra note 89 at 1408-10. The enabling legislation of
the Consumer Product Safety Commission requires notice to the parties
similar announcements on product safety are made. 15 U.S.C. § 2055.
A-38
-------
INDEMNIFICATION STUDY
Appendix A
of the following description Is based on Mizokami v. U.S.,
414 F.2d 1375, 188 Ct. Cl. 736 (1969).
93Priv. L. 88-346, 78 Stat. 1195 (1964).
9438 Fed. Reg. 29400, 29404 (1973).
95S. 3666, 120 Cong. Rec. 34000 (1974); H.R. 17652, 120 Cong. Rec,
41832-33 (1974).
96Cong. Ref. Nos. 2-74 & 1-75, (Dec. 19, 1978).
9715 U.S.C. 2054(h). P.L. 94-284.
(h) Civil action against the United States
Subsections (a) and (h) of section 2680 of
title 28 do not prohibit the bringing of a
civil action on a claim against the United
States which —
(1) is based upon —
(A) misrepresentation or deceit before
January 1, 1978, on the part of the Com-
mission or any employee thereof, or
(B) any exercise or performance, or
failure to exercise or perform, a discre-
tionary function on the part of the Com-
mission or any employee thereof before
January 1, 1978, which exercise, perform-
ance, or failure was grossly negligent;
and
(2) is not made with respect to any agency
action (as defined in section 551(13) of title
5).
In the case of a civil action on a claim based
upon the exercise or performance of, or fail-
ure to exercise or perform, a discretionary
function, no judgment may be entered against
the United States unless the court in which
such action was brought determines (based upon
consideration of all the relevant circum-
stances, Including the statutory responsibil-
ity of the Commission and the public interest
in encouraging rather than inhibiting the
exercise of discretion) that such exercise,
A-39
-------
INDEMNIFICATION STUDY
Appendix A
performance, or failure to exercise or perform
was unreasonable.
(Emphasis added.)
^Conversation with Mana Jennings, Office of General Counsel,
CPSC.
^FF 3-71, FF 5-74: The working paper, "Tris Ban" prepared for
the Regulatory Impacts Branch, OTS by The Conservation Foundation
notes only 40-502 of the garments sized 12 months to 6X used the flame
retardant Tris to meet the standard. After JTris was banned, clothing
manufacturers had to use a more expensive chemical flame retardant for
a short time, but by 1978 they tfere meeting flammability standards by
using inherently flame- retardant fabrics at costs comparable to Tris.
1 ^Telephone conversation with Mana Jennings, Office of General
Counsel, CPSC, March 1982.
1 01- The requirement to repurchase banned hazardous substances is no
longer automatic. 15 U.S.C. § 1261. The 1981 Authorization of CPSC
added an informal adjudicatory hearing on which the Agency order to
repurchase must be based.
102Supra note 100.
103Supra note 99, and S. Rep. 97-130, accompanying S. 823.
. 823 was passed by the Senate in June of 1981; the House
Judiciary Subcommittee on Administrative Law held hearings on June 16,
1982.
.R. 4264, H.R. 4180, Cyclamatea Compensation; Hearings Before
Subcommittee No. 2 of the House Committee of Judiciary, 92d Cong., 1st
Sess. 44 (1971).
the legislation never passed, it raised issues that
were deemed important enough to be considered by high level officials
in a series of meetings and memos. A memorandum for the Honorable
John C. Whitaker, The White House, on "Indemnification of Business
Losses Arising from Federal Regulatory Activities, (March 15, 1971)"
signed by Wilmot R. Hastings, General Counsel recounted some of this
activity, and the positions of various participants, including John
Dean, Richard Kleindiest and Jim Lynn.
107S. Res. 225 referred to the Court of Claims S. 1894 and S. Rpt.
95-576 on Nov. 28, 1977.
10^Briefs are filed at Court of Claims. According to defendant's
final brief, the proposed rule was published two months before the
A-40
-------
INDEMNIFICATION STUDY
Appendix A
Association had to make a decision on how to pack that year's crop.
Apparently the Association discussed the situation with many people,
including its bankers. One reason that it may have decided to go
ahead and take the risk was that for the period just before the ban,
15% of the sales of the Association involved cyclamate-packed
products, but they contributed 50% of the profits.
1°942 U.S.C. § 247(b), P.L, 94-380.
one account of the political, professional, and economic
stakes in the dispute over a swine flue vaccination program, see The
Swine Flu Affair; Decision-Making on a Slippery Disease, Richard E.
Neustadt and Harvey V. Fineberg, M.D. According to Neustadt the impe-
tus for swift passage of the swine flu act (through both Houses and
signed by Ford in two weeks) was a feared link to Legionnaire's
Disease.
^Telephone conversation with Jeffrey Axelrod, DOJ.
112!d_L
11320 U.S.C. 971. 89 Stat. 844 (1976).
1141975 U.S. Cong, and Admin. News 1640.
1 ^Telephone conversation with Alice Martin, Indemnification
Administrator for National Endowment for the Arts, February 1982.
11650 U.S.C. § 1431 and Executive Order 10789.
11759 Comp. Gen. 705 (Sept. 3, 1980).
118"Draft Report of the OFPP Interagency Task Force on Indemnifi-
cation," (January 28, 1982).
119H.R. 1504; cf. Stencel v. United States. 431 U.S. 66 (1977).
from Paul Martin, Acting Director Procurement and
Contracts Division to William Coleman, Indemnification Task Group,
OFPP, September 8, 1981.
1 ^Internal EPA Memorandum from Don Nantkes, OGC to Paul Martin,
OTC, (June 28, 1981).
122Telephone conversation on May 26, 1982 with Pamela John, EPA,
Procurement Contracts Management Division.
A-41
-------
INDEMNIFICATION STUDY
Appendix B
APPENDIX B
REGULATORY AND LEGISLATIVE ALTERNATIVES TO INDEMNIFICATION
Page
I. Introduction B-2
II. The Operation Of Alternatives B-3
A. Alternative Means of Eliminating Harms B-3
1. Financial Assistance r- B-3
2. Exceptions and Waivers B-4
3. Facilitated Participation and Review B-5
B. Alternative Means of Spreading or Shifting Harms. . . B-6
1. Restoration Funds B-6
2. Limitation of Liability B-7
C. Alternative Means of Reducing the Severity
of Harms B-8
1. Postponement of Compliance Deadlines B-8
2. Forebearance with Publicity B-9
D. Alternative Means of Improving Ability to Absorb
Harms B-9
1. Education and Information Transfer B-9
2. Resource and Technology Pooling B-10
3. Partial Financial Assistance B-ll
III. Legislative Treatment Of Loss Situations B-12
A. Emergency Powers B-12
B. Change of Regulatory Policy B-13
C. Tort Situations Potentially Not Covered Under the
Federal Tort Claims Act: Delays and Ineffective
EPA Action; Unreasonable Enforcement; and Disclosure
of Confidential Business Information B-13
D. Conflicting Requirements B-14
E. The Incentive Situation B-14
IV. A Framework For Choosing Between Indemnification and
Alternative Relief Mechanism B-16
Table I—Will an Alternative Produce a Remedial Result
Comparable to Indemnification? B-17
Footnotes ...... ..... B-18
•a
B-l :
-------
INDEMNIFICATION STUDY
Appendix B
I. INTRODUCTION
One important corollary to the general tendency against indemnifi-
cation by the government is that there are numerous alternative mecha-
nisms available to address many of the same losses, concerns, or
problems, to which indemnification could be made to apply. While
indemnification has rarely been provided in federal statutes and
administrative programs, some of the available alternatives are quite
common.
In this appendix, several major alternatives to indemnification
are identified and defined. Part II of the appendix describes the
operation of ten important alternatives that address the types of
losses with which indemnification may be concerned. Fart III examines
how alternative mechanisms are presently used to remedy or prevent
loss situations or to provide incentives under existing environmental
programs. Part IV suggests a framework for choosing between indemni-
fication and an alternative relief mechanism.
B-2
-------
INDEMNIFICATION STUDY
Appendix B
II. THE OPERATION OF ALTERNATIVES
EPA-sponsored indemnification theoretically may operate in two
ways. Either payment of indemnification arises after the fact of loss
resulting from a situation determined to be indemnifiable, or the
offer of indemnification arises prior to any loss and operates as a
pledge of compensation for harms resulting from defined indemnifiable
situations. The former type of indemnification may be an ad hoc
response to an unforeseen or unpredicted harm that is thought, in
fairness, to require indemnification. It requires no advance authori-
zation from Congress, but can be authorized entirely as a one-time
response to particular circumstances.1
c
In contrast, any indemnification program thft is established by
statute or regulation Is, by definition, of the second type. A pro-
gram may approximate the first type of indemnification by providing
only the most general description of indemnifiable situations or
harms.^ However, its more important effect may be to provide a pledge
upon which specified parties may rely in encountering the indemnlfi-
able situation. The pledge will affect those parties' conduct either
explicitly and intentionally^ or indirectly and diffusely.4
Alternatives to indemnification may also operate, In either of the
two ways described above, to achieve some or all of the same results
achievable through indemnification. The available alternative re-
sponses to a potential or actual loss situation may be loosely grouped
into four categories, depending on whether they are aimed at:
• Eliminating the harms resulting from otherwise indemnifiable
situations;
• Spreading or shifting such harms;
• Reducing the severity of such harms; or
• Improving the ability of affected parties to absorb harms
generally.
A. Alternative Means of Eliminating Harms
1. Financial Assistance
Many environmental regulatory programs Include some element of
financial assistance for persons affected by them.^ Such assistance
may take the form of direct cash payments — grants and subsidies, for
example — or they may be less direct forms of assistance, such as tax
credits. Both direct and indirect financial assistance is given out
to those who satisfy certain qualifying criteria.
B-3
-------
INDEMNIFICATION STUDY
Appendix B
Unlike some social welfare "entitlement" programs where aid IB
given to all who qualify, most financial assistance programs in the
environmental field involve some further limitation beyond the minimum
qualifying criteria. This may take the form of a rating system in
which various applicants are rank-ordered,^ or it may be no more than
first-come-first-served rationing. In any form, such financial
assistance is often subject to the exercise of a certain amount of
administrative discretion in selecting recipients. This discretion
may be structured by regulation in order to facilitate identification
of likely recipients in advance of their applying for assistance.
Federal financial assistance is nearly always accompanied by a
variety of conditions to which the recipient must agree. Often these
conditions limit, in minute detail, the disposition of the funds
involved. Some of the conditions may have very little relationship to
the main objective of the financial assistance itself, but may be
included to foster other governmental policy goals.8
Clearly, indemnification is one form of financial assistance.
However, in several important ways, indemnification differs from other
forms of assistance. Indemmification is generally measured by the
magnitude of the harm, while other forms of financial assistance may
be measured by whatever reasonable formula has been adopted.
Moreover, financial assistance awards typically place the government
in the role of benefactor who may restrict or condition assistance as
seems reasonable. Indemnification, however, differs from other finan-
cial assistance in that it places the government in the position of
remedying harms that result from certain indemnifiable situations, and
thus, when such situations arise, the government is unlikely to impose
such restrictions.
2. Exceptions and Waivers
An exception is a statutory or regulatory provision that
operates to restrict the general applicability of the language of the
provision.^ The effect of an exception is to excuse compliance with
the general provision by those to whom the exception applies.10 A
waiver is an individual determination that a particular party should
be excused from compliance.
Both exceptions and waivers can be alternatives to indem-
nification because they can have the effect of eliminating the harms
associated with application of a statutory or regulatory provision.
In general, statutory or regulatory exceptions are provided when it is
possible to define in advance the precise class of parties or situa-
tions to which a standard should not apply. In such a case, an
exception operates as an alternative to the second type of indemnifi-
cation noted above, since it serves as a statement that the harms of
regulation will not be imposed in certain defined "exceptional" cir-
cumstances.
-------
INDEMNIFICATION STUDY
Appendix B
When such a definition is not possible, and Che determination
of whether the standard should apply depends upon a balance of com-
peting considerations, the waiver mechanism is appropriate* A waiver
mechanism established by statute or regulation may specify, in general
terms or in minute detail, the considerations to be balanced in
granting or denying the waiver.^ When waiver considerations are
stated very generally, the waiver mechanism most closely approximates
an alternative to the first type of indemnification noted above, since
it can be an ad hoc response to an unforeseen harm caused by imposi-
tion of a regulatory standard.
The formal grant of a regulatory exception constitutes admi-
nistrative rulemaking; the formal grant of a waiver constitutes
adjudication. Both exceptions and waivers may also exist Informally,
however, In the exercise of administrative prosecutorial discretion.
The practical effect of an administrative decision not to pursue a
particular class of violators^ or an individual violator is the same
as if a formal exception or waiver had been granted.
Exceptions and waivers may be total, or they may involve the
concept of tiering. Tiering is an administrative mechanism which
allows the stringency of a regulation to vary in proportion to some
other factors — the degree of hazard, the impact on the regulated
community, etc.^ Thus, a provision may except one class of parties
from one regulatory requirement, but subject it instead to a less
stringent requirement.^ Similarly, a waiver may excuse compliance
with a particular standard provided that a lower standard is
met.l' Clearly, both exception and waiver mechanisms can be designed
with as many or as few tiers as seems appropriate.
3. Facilitated Participation and Review
Under the Administrative Procedure Act, EPA rulemaking activi-
ties are generally subject to the requirement that affected parties be
given, at a minimum, notice of proposed regulations and an opportunity
to comment on them. Similarly, affected parties are entitled to
participate in agency adjudicatory decisions.^ In a few
circumstances, however, EPA may take limited actions without affording
such participation, either because of the immediacy of the problem
being addressed20 or because the action constitutes neither rulemaking
nor adjudication. When this occurs, the risk of producing undesired
harms is, of course, increased since there is no mechanism for
bringing the possibility of such harms to the attention of EPA before
the action is taken.
One method of dealing with this increased risk is to provide
for an expeditious hearing on the EPA action after it is taken, so
that it can be rescinded or modified to avoid the harm, as appropri-
ate, at an early date.21 Even where a hearing has been held or notice
and comment procedure has been followed, the same objective of expedi-
B-5
-------
INDEMNIFICATION STUDY
Appendix B
tlous review, modification, rescission or confirmation may be
desirable in minimizing or eliminating untoward consequences of EPA
actions. In such circumstances, facilitated judical review may be
appropriate.
Every final administrative action — except for those com-
mitted to agency discretion by law — is subject to judicial
review.22 Thus, judicial review is virtually always available as a
mechanism for undoing or modifying certain actions in appropriate
cases. In a few situations, Congress has chosen to facilitate judi-
cial review of administrative action. This is often accomplished by
designating a particular district or circuit court (such as those of
the District of Columbia) as having exclusive jurisdiction over
challenges to the agency's action, or by providing the direct judicial
review is to occur at the Court of Appeals level." Such expedited
review is most often regarded as benefiting the affected administra-
tive agency, either by avoiding a proliferation of judicial eva-
luations of its actions or by permitting the agency to have a quicker
evaluation of its actions, less subject to reversal by a higher court.
However, in a few circumstances, expedited judicial review may operate
as a benefit to the party affected by EPA action, by accelerating the
implementation of modifications to that action. *
Both facilitated participation and expedited review serve as
alternatives to indemnification when they operate to modify or ter-
minate quickly a course of EPA regulatory action that is Imposing
harms on aggrieved parties. Like exceptions and waivers, these alter-
natives operate to eliminate such harms by removing the necessity of
complying with the standard at issue. The value of facilitated par-
ticipation and expedited review, therefore, is derived primarily when
an EPA action is ultimately modified, rescinded or overturned.
B. Alternative Means of Spreading or Shifting Harms
1. Restoration Funds
A restoration fund is a sum of money set aside to pay compen-
sation under specified circumstances. Moneys for the fund are nor-
mally collected by means of an "injury tax" or surcharge on a product
or activity related to the purpose of the fund. Thus, for example,
Superfund is a restoration fund financed (in part) from a tax imposed
on crude oil received at a United States refinery, imported petroleum
products, crude oil exports and uses,2* and 42 specific chemical
products.2^
A restoration fund differs from indemnification, although both
involve payments measured by certain defined harms, in that the amount
of the fund is collected in advance in anticipation of the necessity
of payouts. The fund is kept separate from other revenues and may con-
tinue to grow year after year until payouts are necessary.
B-6
-------
INDEMNIFICATION STUDY
Appendix B
Eligibility for compensation from the fund may be limited by the total
anount available^ at the time of the discovery of the loss, or it may
be measured only by the indemnification formula established. The
levels of the injury tax may theoretically be made to vary in propor-
tion to the fund's payout experience. In contrast, indemnification is
normally financed from general revenues and its amount is measured
only by the indemnification formula established.
Payments from a restoration fund are typically conditioned on
a triggering event.^8 Such events must be statistically foreseeable,
even if they are not actually predictable, and their probability may
vary with an activity upon which the tax or surcharge can be levied.
The existence of a restoration fund in the first Instance
clearly obviates the need to provide indemnification, and thus serves
as an alternative to indemnification. Restoration funds do not shift
the costs of harms to EPA, however, as indemnification would, but
rather distribute those costs among those contributing to the fund.
Requiring payments into a restoration fund In anticipation of a trig-
gering event such as might be the subject of indemnification thus for-
ces industry members to share the risks involved in their enterprises.
Restoration funds are therefore essentially mechanisms for spreading
the risks of certain activities among all those engaged in the acti-
vity, rather than letting the untoward consequences of those activi-
ties fall on an unlucky few, or on the government agency providing
indemnification.
2. Limitation of Liability
A limitation of liability is a statutory declaration that a
party whose conduct or product causes harms will not be held.account-
able for all of those harms.29 A limitation of liability is an alter-
native to indemnification when the harms being addressed would
initially fall on a diffuse group of parties who, in the absence of
such a limitation, would claim compensation from the responsible party
who, in turn, would be in need of indemnification from the government.
Like Indemnification, a limitation of liability shifts the relevant
risk away from the party engaged in the activity or producing the pro-
duct creating the risk. Thus, like indemnification, a limitation of
liability removes an economic impediment to a particular activity or
product in order to stimulate those who would otherwise be willing to
engage in the activity or develop the product.^ However, unlike
indemnification, the limitation of liability shifts that risk to those
who Initially suffer the harm,31 rather than to a government body pro-
viding indemnification.
A limitation of liability may be a feasible alternative to
indemnification when it operates to shift harms to a group who are
readily able to absorb them, and to spread those harms among members
of that group. Since this shift occurs without significant transac-
B-7
-------
INDEMNIFICATION STUDY
Appendix B
tion costs, it can prove to be a relatively trouble-free alternative.
Indeed, a limitation of liability will nearly always be an alternative
of lower cost than indemnification In situations to which it can be
made to apply•
C. Alternative Means of Reducing the Severity of Harms
1. Postponement of Compliance Deadlines
A postponement of compliance deadlines occurs when compliance
is excused for a period of time after a statutory or administrative
regulatory action is taken.3^ Such postponements are quite common,
and are usually conceived of as necessary in order to give notice to
the affected parties that the regulatory action has occurred, and an
opportunity for such parties to adjust their activities in response in
an orderly fashion within a reasonable time.
Many regulatory actions are taken with a general postponement
of their effective date accompanying them. Thus, for example, a regu-
latory standard may be promulgated with an effective date a specified
length of time after its Federal Register publication.33 Alter-
natively, the regulation itself may state that the affected activity
must comport to the regulatory standard by a specific future date.3*
In theory, if not often in practice, postponements may be made
applicable to some classes of regulated parties, and not to others.
More common are policies, adopted either by regulation or otherwise,
not to require compliance with a regulatory standard by all those
against whom enforcement is regarded as not immediately appropriate,
for example, because of their difficulty or expense in complying.36 A
formal grant of a postponement in such circumstances constitutes an
administrative adjudication. However, a less formal postponement may
be regarded as a mere exercise of prosecutorial discretion.
As a practical matter, postponements are also often bargained
for in the course of prosecution and defense of enforcement
actions.3^ Thus, even when a decision is made to prosecute such an
action, the key objective of enforcement — compliance with the stan-
dard — may be achieved (and more quickly) by means of a settlement
including a legally enforceable compliance deadline, rather than by
protracted litigation.
Postponements may be regarded as alternatives to indem-
nification to the extent that they will reduce the severity of harms
associated with a regulatory activity. In some circumstances, post-
ponements merely delay those harms and arguably do not reduce their
severity. However, where compliance is facilitated or made more
orderly by a postponement, harms may be made less severe without the
need for indemnification.
B-8
-------
INDEMNIFICATION STUDY
Appendix B
2. Forebearance with Publicity
There are a number of ways available to EPA to achieve some or
all of its regulatory goals without actually engaging in rulemaking or
adjudication.38 One such way is to act merely to publicize the fact
or action that might otherwise be the subject of regulatory
concern.39 Rather than to prohibit the action, or ban the product in
question, EPA may simply inform merchants, retailers and the public of
the possible environmental risks at issue.
Such an action would not constitute rulemaking, would not com-
pel any particular response from the affected parties and would not be
subject to judicial review. The affected party is left free to
behave, in response to the publicity, in whatever manner it deems
appropriate. By publicizing that behavior, however, EPA may stimulate
market or other forces to induce proper behavior. In an appropriate
case, the primary objective will be achieved in either event, but if
EPA has done no more than publicize the party's activity, it will have
done nothing that, in fairness, can be the basis for indemnification.
Indeed, the total harm to the affected party will likely be less,
since it will be left free to devise its own solution to the environ-
mental problem at hand.
Foregoing regulatory action in favor of an alternative such as
publicity operates as an alternative to indemnification in the sense
that it reduces the adverse consequences of regulatory activity. This
reduction may be achieved at some cost in the health or environmental
benefit to the public, but it allows regulated parties to design their
own response to the problem, in proportion to market incentives or
threats of liability.
D. Alternative Means of Improving Ability to Absorb Harms
1. Education and Information Transfer
One important function of regulatory agencies, closely related
to forebearance with publicity, is to educate those affected by their
actions and to provide information to the regulated community and the
public concerning the need for agency action, the risk being
addressed, the rationale for the regulatory strategy followed and the
means of complying with, or otherwise accommodating, the agency
action.^0 Such education and information transfer serve as an alter-
native to Indemnification by improving the ability of affected parties
to absorb harms they have incurred.
Effective educational and information programs can reduce
resistance to regulatory activity; produce a cooperative relationship
between the agency and the regulated community; improve the quality of
participation in a regulatory program by the regulated community and
the public, and lower the overall costs Imposed by regulatory action.
B-9
-------
INDEMNIFICATION STUDY
Appendix B
Thus, they can be an important alternative to indemnification. When
the harms arising out of regulatory action can be reduced signifi-
cantly through educational efforts by the agency (for example, on the
lowest cost method of meeting a new regulatory standard), the need for
indemnification is correspondingly reduced. If educational efforts
can reduce the adverse impacts of EPA action to a level reasonably
absorbed by affected parties, no indemnification will be required at
all.
2. Resource and Technology Pooling
Resource and technology pooling consists of joint efforts
among one or more federal agencies, other levels of government or pri-
vate firms to share technological Innovations, research and develop-
ment, economic or other resources in order to reduce the costs
associated with regulatory action and to foster innovation.^1 Resource
pooling occurs when funds or capital resources are contributed to a
single venture or activity addressed to a common regulatory impact.
Technology pooling occurs when specific equipment or technology is
shared with others who are engaged in parallel activity in response to
regulatory action.
Resource and technology pooling can occur without any gov-
ernment involvement at all — at least in those circumstances where
antitrust laws do not interfere. However, there is much that the fed-
eral government can do to participate in such pooling or to encourage
participation by others. For example, EPA might pursue a policy of
liberalized licensure of patents held by the government for those par-
ticipating in resource or technology pools. Similarly the Agency
could tie other benefits to such participation.
Resource and technology pooling is obviously a means of
improving the ability of affected parties to absorb the harms asso-
ciated with regulatory activity, and thus can be regarded as an alter-
native to indemnification in this sense. When pooling results in
lower individual costs being incurred in response to administrative
action — for example, when technology is shared to take advantage of
economies of scale — the desired objective is achieved without indem-
nification and resistance to the action may be reduced. This, for
example, appears to be the purpose of the pooling of data and study
results under TSCA.^2
However, creation of a pool does not, by itself, always
guarantee amelioration. For example, where administrative action is
designed to force technology, a resource pool designed to undertake
research and development may achieve the necessary breakthrough, but
it may also fail.
B-10
-------
INDEMNIFICATION STUDY
Appendix B
3. Partial Financial Assistance
Financial assistance may provide something less than full cash
payment te- its recipient. Thus, once it ia determined that a par-
ticular sum is required, that sum may be loaned, rather than paid, or
an Interest subsidy or repayment insurance^ may be written to facili-
tate private funding, or a smaller amount may be paid.44
Although partial financial assistance is generally a mechanism
for Improving the ability of assisted parties to absorb the impact of
regulatory activity, rather than to eliminate those harms as would
full cash assistance, they are structurally otherwise identical.
B-ll
-------
INDEMNIFICATION STUDY
Appendix B
III. LEGISLATIVE TREATMENT OF LOSS SITUATIONS
When the various statutes administered by EPA are examined in
relation to the possible loss situations that can arise under them,
the most striking conclusion is the consistency with which the various
situations have been treated by Congress across the range of environ-
mental legislation. A strong pattern emerges, reflecting a congres-
sional routine — whether it amounts to a congressional preference is
unclear — for avoiding or responding to these situations. With a few
exceptions, congressional consistency appears to be far greater than
one might have expected from a political forum.
Clearly Congress has often acted more to avoid loss situations
than to create a remedy for them. Repeatedly, Congress has demon-
strated its willingness to compromise primary statutory objectives
rather than face the likelihood of creating loss situations requiring
indemnification. This suggests a congressional willingness to assign
even important legislative objectives relatively less weight than loss
situation avoidance. To the extent that this desire can be translated
into a congressional policy, the implication to be drawn is that
indemnification ought not to be rejected in a particular case merely
because a legislatively mandated program objective would be compro-
mised by payment. Rather, congressional policy seems to be that the
harms suggested in the indemnification situations are usually of
greater concern than the objectives of the statutory scheme.
On the other hand, it is also clear that indemnification is not a
favored response to loss situations. Rather, other measures are
nearly always provided most of them designed to avoid the situations
entirely. Thus, a second congressional policy to be derived from this
analysis is that indemnification should be regarded as a strategy of
last resort, to be pursued only when no acceptable alternative appears
and compromise of the legislative objective is impossible.
A. Emergency Powers
This situation involves EPA emergency actions which prove to be
ill-conceived but which are excluded from the Federal Tort Claims
Act'*5 because they arise out of the exercise of EPA discretion.
Although, in emergency situations, EPA can make errors of judgment in
promulgating regulations, formulating policy, or taking remedial
action, Congress has never formally addressed such errors in its
legislation.
Where EPA error is not the basis for indemnification, the exercise
of emergency powers has given rise to the mandate for indemnification
in only one instance. This mandate appears in § 15 of the Federal
Insecticide, Fungicide and Rodenticide Act (FIFRA).46 No other
B-12
-------
INDEMNIFICATION STUDY
Appendix B
suggestion appears in any other environmental statute, however, that
the threat of emergency action should give rise to the need to provide
indemnification.
B. Change of Regulatory Policy
A loss situation may occur whenever new information or under-
standing produces a change in EPA's regulatory program. Such a change
may be most abrupt when EPA reverses a prior position generally
approving or even positively sanctioning a particular product or
activity. However, the loss situation may also arise after EPA
acquiescence or complete nonaction with respect to a product or
activity.
o
The usual response to the policy change situation is to grant a
temporary variance from, or postpone the deadline for compliance with,
the newly necessitated regulatory requirement.^? In some instances,
the effectiveness of any new requirement is explicitly delayed for a
period of time. In others, its application to those who seem to nave
the greatest stake in the prior state of affairs is delayed.^8 in
still others, elaborate safeguards are included before changes can be
made — thus eliminating the possibility of suddenness.*'
These treatments of the policy change situation reflect congres-
sional willingness, noted above, to compromise the primary objectives
of EPA's programs, rather than to disrupt affected activities with
policy or regulatory changes. This willingness may perhaps be
explained by the realization that environmental regulation continues
to be an inexact process, involving some measure of guess work as well
as hard data and analysis. Where technical uncertainty is acknowl-
edged, it is obviously most difficult to insist on regulatory action
as if such uncertainty were eliminated. Rather, it may be justifiable
to seek to minimize disruption when such uncertainty causes sudden
changes in EPA's regulatory program.
On the other hand, detrimental reliance by an affected party on a
regulatory policy or provision, which, when rescinded causes that
party to and suffer a harm is often cited as an Important regulatory
relief concern. None of the environmental statutes explicitly deals
with the detrimental reliance situation.
C. Tort Situations Potentially Not Covered Under the Federal
Tort Claims Act; Delays and Ineffective EPA Action;
Unreasonable Enforcement; and Disclosure of Confidential
Business Information
These situations occur whenever EPA commits a tortious act that is
not compensable under the Federal Tort Claims Act.50 Although there
are many conceivable tortious acts that EPA could conceivably commit,
the three cited are most often mentioned.
B-13
-------
INDEMNIFICATION STUDY
Appendix B
The problem of unreasonable enforcement is considered, but only
once, In the various enviornmental statutes. Section 113(b) of the
Clean Air Act,^ which deals with enforcement actions by the Adminis-
trator of EPA, authorizes the award of litigation costs to a party
against whom an unreasonable action was brought. No .obvious explana-
tion appears for the existence of this particular provision, limited
to air enforcement actions, when no similar provisions exist under
other environmental statutes. In general, however, the Clean Air Act
is the most complex of the environmental statutes and the one which
underwent perhaps the most extensive revision after Its initial
enactment. Regulatory relief provisions are thus more common in the
Clean Air Act than in the other environmental statutes, and this
political and historical reason may account for the existence of this
provision.
Other tortious EPA conduct is ignored throughout the various
statutes*
D. Conflicting Requirements
This situation occurs when a federal environmental regulation or
statute mandates or effectively compels conduct that is prohibited by
another state or federal provision. It also occurs when a federal
environmental provision prohibits an activity that is mandated or
effectively compelled by another federal or state program.
The conflicting requirements situation rarely occurs in practice.
There is an ample body of rules governing conflicts of law designed to
assure that such situations will not arise.52 Nevertheless, various
state and federal enactments do have differing (and arguably
conflicting) objectives, and these conflicts do need to be resolved.
The various envionmental statutes contain many provisions for
avoiding the conflicting requirements situation but not for remedying
it. These avoidance mechanisms consist of either (1) rules for
assigning primary responsibility to one regulatory regime when a
conflict appears;^^ and (2) requirements that regulatory actions be
coordinated.5*
E. The Incentive Situation
This situation occurs when EPA wishes to provide an incentive for
an activity or conduct involving risk, which the party to whom the
incentive is provided would not undertake because of that risk. The
situation seems most likely to arise where a new technology is desired
to be encouraged, but any desired activity involving risk may give
rise to this situation.
The "classic" example of this situation la represented by
§ 202(a)(3) of the Clean Water Act.55 That section provides for
B-U
-------
INDEMNIFICATION STUDY
Appendix B
indemnification for the cost of constructing innovative and alter-
native wastewater treatment facilities that later fail to perform as
expected. Clearly this section is designed to encourage investment in
such facilities by reducing the fear of failing to meet performance
specifications.
The environmental statutes are replete with provisions designed to
encourage particular types of conduct. Few, however, provide that
incentive by means of risk absorption, despite the high degree of
technical uncertainty in environmental regulation and the various
technology forcing policies reflected in the statutes.
Section 1441 of the Safe Drinking Water Act,56 offers an example
of an incentive situation. In an earlier paragraph of that section,
Congress provided for a certification of need to be issued to any per-
son who uses chlorine or other chemicals f'or treating public water
supplies and who determines that the amount of such chemicals reason-
ably available to him will be inadequate to treat that water
effectively. When the certification Is Issued, it is accompanied by
an order to an appropriate manufacturer, producer or processor,
directing sale of the needed chemicals to the certified treatment
facility.
Paragraph (d) then provides:
There shall be available as a defense to any
action brought for breach of contract in a
Federal or State court arising out of delay or
failure to provide, sell or offer for sale or
exchange a chemical or substance subject to an
order issued pursuant to subsection (c)(l),
that such delay or failure was caused solely
by compliance with such order.
Thus, this section authorizes EPA to eliminate one of the risks of
drinking water treatment — loss of chemical supplies — by a certifi-
cation procedure designed to guarantee the supply. In this situation,
the desire to guarantee a continuous flow of safe drinking water makes
indemnification an unsuitable device for risk absorption, and supply
guarantees a preferred strategy.
B-15
-------
INDEMNIFICATION STUDY
Appendix B
IV. A FRAMEWORK FOR CHOOSING BETWEEN INDEMNIFICATION
AND ALTERNATIVE RELIEF MECHANISM
The choice between indemnification and an alternative relief
mechanism depends largely on the circumstances of the particular loss
situation involved or the nature of the incentive desired. For any
particular situation defined, the choice depends on the answer to two
questions
• Will an Alternative Produce a Remedial Result Comparable to
Indemnification?
• How Vill Use of the Alternative Affect Program Objectives?
Table I illustrates how the answers to these two questions assists in
the selection between indemnification and an alternative.
B-16
-------
WILL AN ALTERNATIVE PRODUCE A REMEDIAL RESULT COMPARABLE TO INDEMNIFICATION?
How will use of
an alternative
affect program
objectives?
Beneficial
No or Neutral
Effect
Adverse Effect
No Worse Than
Indemnification
Adverse Effect
Not Unacceptably
Worse Than
Indemnification
Adverse Effect
Unacceptably
Worse Than
Indemnification
Yes At Least at
Effective as
Indemnification
A
A
A
A
B
Not as
Enough to
Obviate Need for
Indemnification
B
B
B
B
E
Effective as Indemn
Enough to
Narrow Range of
Indemnification
Situation
C
C
C
D
E
if Icatlon
Indemnification
Still Needed
E
E
E
E
E
I
»"—•
^J
A • reject indemnification; choose alternative (threshold analysis)
B • consider alternative as factor in deciding whether to Indemnify (balancing analysis)
C • choose alternative; redefine indemnification situation more narrowly; repeat Step 1 (Iterative analysis)
D « consider whether to choose alternative and redefine indemnification situation more narrowly; repeat Step 1
(balancing and iterative analysis)
E - reject alternative; other factors determine whether to 4"demnlfy
-------
INDEMNIFICATION STUDY
Appendix B
FOOTNOTES TO APPENDIX B
1If passed, the Trls Indemnification Bill, S. 823 (97th Cong.,
1st Seas.) would be an example of post-hoc authorization of indemnifi-
cation. See "Payments of Losses incurred as a Result of the Ban on
the Use of the Chemical Tris," Report of the Committee on the Judici-
ary, U.S. Senate, 97th Cong., 1st Seas. (1981).
2Section 113(b) of the Clean Air Act, 42 U.S.C. S 7413, is argu-
ably the least specific of the existing indemnification provisions in
environmental statutes. While it defines indemnif iable harms with
some precision ("costs of litigation"), its definition of the indem-
nifiable situation ("unreasonable" enforcement action) is fairly open
to interpretation.
3Section 202(a)(3) of the Clean Water Act. 33 U.S.C. S 1282(a)(3),
offers such an explicit and intentional incentive for constructing
innovative and alternative waatewater treatment facilities.
^It has been suggested, for example, that § IS of the Federal
Insecticide, Fungicide and Rodenticide Act (PIFRA), 7 U.S.C. § 126m,
provides an indirect incentive for pesticide manufacturers not to |
engage in certain research and development activities, but rather to
rely on the pesticide registration process, because indemnification
will be available if EPA determines that marketing and use of the
pesticide must end before stocks are depleted.
Construction Grants Program under the Clean Water Act, 33
U.S.C. §§ 1281-1297, for example, is generally regarded as a means of
overcoming the burden imposed on local governments by the Act's
requirement of secondary treatment of municipal effluent. 33 U.S.C.
§ 1311(b)(l)(B).
"Construction grant funds, for example are allocated according to
a formula set forth in 33 U.S.C. S 1285. As described by one
observer:
This program provides grants to states based
on (a) severity of problems, (b) existing
population affected, and (c) need for preser-
vation of high quality waters. . . .During FY.
1976 $4.9 billion were obligated for this pro-
gram. The data elements used in this program
are state population estimates for July 1975,
published by the Bureau of the Census, and
various components of need determined by a
survey to estimate cost of construction of
publicly owned wastewater facilities. In
determining these needs, the projected popula-
tion for 1990, prepared by the Bureau of
B-18
-------
INDEMNIFICATION STUDY
Appendix B
Economic Analysis (BEA) and based on Census
Series 1-E, is compared with the population
for July 1975. This program includes both a
House formula and a Senate formula. The House
formula is additive with three components:
(a) a needs factor that includes costs for
secondary treatment plants, weighted double
. . . ; (b) a needs factor that includes
rehabilitation of sewers and control of sewer
overflow . . . ; (c) and state population for
July 1975. Each factor is a ratio of the
state estimate to the sum for all states. The
Senate formula chooses the larger of the state
population ration and the ratio of a third set
of needs estimates for the state. . . .These
initial entitlement estimates are prorated so
that they sum to 100 percent. The next step
is to check that the Senate percentage for
each state is at least .5 percent. This pro-
cess is repeated until the sum of the state
percentages based on the Senate formula equals
100 percent. After both the House and Senate
formulas have been computed, the compromise
state entitlement percentage is obtained by
weighting the two estimates equally.
Gonzalez, "Characteristics of Formulas and Data Used in the Allocation
of Federal Funds, 34 American Statistician 200. 207-08 (1980).
'See Gilhooley, "Standards and Procedures for the Discretionary
Distribution of Federal Assistance," in Administrative Conference of
the United States, 3 Recommendations and Reports 422 (1974).
8 Some of these are enumerated In Cappalli, "Mandates Attached to
Federal Grants: Sweet and Sour Federalism," 13 Urban Law. 143 (1981).
9Jensen v. Garrison. 241 F. Supp. 523 (D. Ore. 1965); Sands,
Sutherland's Statutory Construction § 47.11.
statutory provisions except activities from regulation when
it would be nonsensical to compel compliance. For example, the Noise
Control Act prohibits removal of noise abatement devices from
products, 42 U.S.C. § 4909(a)(2), but except s from the prohibition the
removal of such devices for repair purposes. Id. Other exceptions
appear when the cost of compliance appears not to be justified in
light of the expense. TSCA's exception from requirements to submit
reports and data that EPA already has is an example. 15 U.S.C.
§ 2607(a)(2).
11See. e.g., 33 U.S.C. 5 1342(e), authorizing a waiver of the
requirement that states notify EPA of every permit application
B-19;
-------
INDEMNIFICATION STUDY
Appendix B
received under the National Pollution Discharge Elimination System
(NPDES) and of every action related to the consideration of such
permits.
*2Thus, for example, the Clean Air Act exempts certain non-profit
health or education facilities from "prevention of significant
deterioration" (PSD) review. 42 U.S.C. § 7479(1).
e*8- > the Clean Air Act's provision for waiver of the con-
tinuous emissions reduction technology requirement for primary non-
ferrous smelters where the requirement is found to be likely to cause
a cessation of operations. 42 U.S.C. § 7419(d)(2).
l^An example of such a decision occurred when EPA announced that
it would not enforce the 1980 annual report requirement for hazardous
waste generators and owners and operators of hazardous waste treatment
and disposal facilities. 46 Fed. Reg. 8395 (January 26, 1981). The
requirement has since been suspended.
key criticism of EPA's regulatory program under the Resource
Conservation and Recovery Act (RCRA), 42 U.S.C. §§ 6901 et seq.. for
example, has been its failure to incorporate tiered exemptions based
on the degree of hazard of each particular waste. Compare Cal. Health
& Safety Code § 25115 (defining "extremely hazardous waste"). See
Chemical Manufacturers Association, A System for Management of
Hazardous Wastes by Degree of Hazard under Subtitle C of RCRA
(updated) .
1 ^ RCRA, for example, excludes from solid waste regulation those
discharges that are already regulated under the arguably less
stringent Clean Water Act. 42 U.S.C. § 6903(27).
17For example, § 301 (c) of the Clean Water Act, 33 U.S.C.
§ 1311(c) authorizes EPA to excuse point sources from meeting effluent
limitations based on "best available technology economically achiev-
able" (BAT) if they achieve "maximum use of technology within the eco-
nomic capability of the owner or operator."
185 U.S.C. § 553.
195 U.S.C. § 554.
2°The Toxic Substances Control Act (TSCA), for example, permits
certain types of rules to be effective immediately upon their
proposal. 15 U.S.C. § 2605.
2*In the TSCA provision noted supra note 20, for example, such an
expeditious hearing is provided. 15 U.S.C. § 2065(d).
225 U.S.C. §§ 551, 702.
B-20
-------
INDEMNIFICATION STUDY
Appendix B
for example, provides for judicial review of most regula-
tions issued under the Act in the District of Columbia Circuit Court
of Appeals. 15 U.S.C § 26i8(a).
Judicial review may, as a practical matter, be facilitated in
other ways as well. For example, an agency can facilitate review of a
praticular activity by designating it as final agency action and thus
removing any question of its ripeness for review. Another example
occurs when an agency's governing statute sets a review standard for
agency action that is more precise or better defined than the
"arbitrary and capricious" and "substantial evidence" standards of the
APA.
24Internal Revenue Code. 26 U.S.C. § 4611.
25Internal Revenue Code, 26 U.S.C. § A661.
the Comprehensive Environmental Response, Compensation and
Liability Act (CERCLA), the Hazardous response Trust Fund is to be v,
funded from the tax revenues noted supra notes 24, 25, as well as by
direct appropriations of $44 million per year. 42 U.S.C. § 9631(b)(2).
27See, e.g.. 42 U.S.C. § 9631(c)(2).
28Cf. 42 U.S.C. § 9631(c)(l)i
29See. e.g.. 42 U.S.C. § 9611(a).
limitation of liability provision appears in § 141(d) of the
Safe Drinking Water Act, 42 U.S.C § 300j(d), but this provision was
allowed to expire on September 30, 1979.
3*In Duke Power Co. v. Carolina Environmental Study Group. Inc.,
438 U.S. 59 (1978), the U.S. Supreme Court characterized liability
limitation as "a classic example of an economic regulation—a legisla-
tive effort to structure and accommodate 'the burdens and benefits of
'economic life.'" 438 U.S. at 83, quoting Usery v. Turner Elkhorn
Mining Co.. 428 U.S. 1, 15 (1976).
32Forcing such parties to absorb such harms, however, is unlikely
to amount to a sufficient ground for declaring the limitation of
liability unconstitutional. See id., Duke Power Co.
33Sands, supra note 9, at § 20.24.
34
See,
the Noise Control Act's provisions postponing compli-
ance with product noice emission standards until six months after pro-
posed standards are published in order to facilitate orderly design
modifications. 42 U.S.C. § 4905(b).
B-21
-------
INDEMNIFICATION STUDY
Appendix B
35The Clean Water Act's effluent limitation provisions, 33 U.S.C.
§ 1311, for example, specify a timetable for achieving best prac-
ticable control technology (BPT), generally by July 1, 1977, and best
conventional pollutant control technology (BCT) generally by July 1,
1984. 33 U.S.C. §§ 1311(b)(l)(A), 1311(b)(2)(E).
36see, e.g. , 42 U.S.C. § 7419, which authorizes primary nonferrous
smelter orders postponing application of State Implementation Plan
(SIP) requirements to such smelters if they are unable to comply by
the applicable date because no means of emission limitation is avail-
able.
37Compare § 301(c) of the Clean Water Act. 33 U.S.C. § 1311(c),
with § 119 of the Clean Air Act. 42 U.S.C. S 7419.
38See, e.g., United States v. Homes take Mining Co., 26 Fed. Rules
Serv. 2d 194 (D.S.D. 1978); 33 U.S.C. § 1319(a)(5)(B).
existence of formats of agency action other than ruleraaklng
and adjudication was recognized by the U.S. Supreme Court in I.T.T.
Corp. v. Electrical Workers Local 134. 419 U.S. 600 (1975). In that
case a labor union appealed an order of the National Labor Relations f
Board based on a finding that the union had committed an unfair labor
practice. The appeal cited an alleged violation of § 5 of the APA, 5
U.S.C. § 554, and argued that under the section It was improper for
the same Individual to act first as a hearing officer at the hearing
concerning the charges held pursuant to § 10(k) of the National Labor
Relations Act, 29 U.S.C. § 10(k) , and then later to prosecute the
unfair labor practice charge.
The Court held that the § 10(k) determination was only a prelimi-
nary administrative determination to resolve a dispute, and therefore
was not within the scope of § 5:
If one were to start with the proposition that
all administrative action falls into one of
two categories, rulemaking or adjudication,
the Section 10(k) determination certainly is
closer to the latter than to the former. But
such light as we have on the intention of
Congress when it enacted the Act does not
Indicate that this Is a sound starting point.
Knowledgeable authorities in this field
observed shortly after the passage of the Act
that "certain types of agency action are
neither rulemaking nor adjudication." Ginane,
"Rule Making" "Adjudication" and "Exemptions
Under the Administrative Procedure Act, 95
U. Pa. L.Rev. 621 633, (1947); Netterville,
The Administrative Procedure Act: A Study in
B-22
-------
INDEMNIFICATION STUDY
Appendix B
Interpretation, 20 Geo. Wash. L. Rev. 1, 33
(1951); Cf. Attorney General's Manual on the •
Administrative Procedure Act 40 (1947). ..." i
I
The Ginane article cites, as an example of the point rade, the j
investigation of aircraft accidents by the Civil Aeronautics Board*
Such an investigation results in findings as to the accident's cause
and also in recommendations for operational changes, if necessary. It
does not result in the issuance of an order (ad juciation) , or of an
agency statement proscribing further conduct (rulemaking).
National Ornament & Electric Light Christmas Ase'n v.
CPSC, 526 F.2d 1368 (1975). Professor Ernest Gellhorn in "Adverse
Publicity by Administrative Agencies," 86 Harv. L. Rev. 1380 (1973)
(based upon report prepared for the Administrative Conference of the
United States), examines the lack of standards for issuing such publi-
city and proposes reforms intended to better control the use of
adverse publicity.
various environmental statutes are replete with provisions
of this type. See, e.g., the Noise Control Act. 42 U.S.C. § 4904(b),
requiring publication of reports on major sources of noise and on
techniques for noise control, in order to facilitate compliance with
noise emission standards.
., JRB Associates, Inc., Checklist Papers of Alter-
natives and Supplements to Direct Regulation for Environmental
Protection. 26-1 (EPA Contract No. 68.01.4991)(1980).
4315 U.S.C. § 2604(b)(2)(B).
See, e.g., 33 U.S.C. § 1293, providing loan guarantees for
POTWs where credit is otherwise unavailable.
A5See, e.g.. 33 U.S.C. § 1282(a)(l), providing for EPA to assume
a portion of the cost of construction of POTWs.
U.S.C. § 2680(a) [FTCA discretionary function exception] .
477 U.S.C. § 136m.
A8See, e.g.. Clean Air Act. 42 U.S.C. §§ 7413(d)(8) (delaying
termination of postponement orders to avoid undue hardship); 7419 (e)
(delaying termination of primary nonferrous smelter orders until no
undue hardship would result).
49See. e.g.. Clean Water Act. 42 U.S.C. §§ I316(d) (protecting
complying point sources from new performance standards for ten years);
1326(c) (providing similar protection from thermal discharge limita-
tions) .
B-23
-------
INDEMNIFICATION STUDY
Appendix B
50S_ee_, e.g., FIFRA, 7 U.S.C. § 136d. Other provided responses
include:
(a) TSCA
15 U.S.C. § 2605(d): expeditious hearing on immediately
effective proposed rules.
(b) Clean Air Act
42 U.S.C. § 7525(b): expedited Judicial review of
suspension or revocation of motor
vehicle emission certifications.
42 U.S.C. § 7525(g): nonconformance penalty option in
lieu of suspension or revocation of
heavy duty vehicle emission
certification.
S^The exceptions to the FTCA waiver of sovereign immunity are con-
tained in 28 U.S.C. S 2680.
5242 U.S.C. § 7413.
generally Sands, Sutherland's Statutory Construction.
example:
(a) TSCA
15 U.S.C. § 2602(b): excludes from TSCA those substances
regulated under FIFRA, the Atomic
Energy Act, FDA and other statutes.
15 U.S.C. S 2608(a): assigns primacy to all statutes not
administered by EPA.
15 U.S.C. § 2617: preemption of state law.
(b) Safe Drinking Water Act
33 U.S.C. § 300q-3: authorizes state regulation of safe
drinking water.
(c) Clean Water Act
33 U.S.C. S 1370:
authorizes state clean water
regulations.
B-24
-------
(d) Noise Control Act
42 U.S.C..S 4905(c):
(e) RCRA
42 U.S^.5 6905(b):
42 U.S.C.. § 6963:
(f) r.Tean Air Act
42 U.S.C. § 7402:
5633 U.S.C. § I282(a)(e).
TimcMMT FT CATION STUD1
Appendix B
requires consideration of other
federal regulatory standards before
promulgation of products standards.
provides for integration with other
environmental statutes.
requires all federal agencies
cooperate with EPA.
to
encourages cooperative state acti-
vities and uniform laws.
B-26
-------
INDEMNIFICATION STUDY
Appendix C
APPENDIX C
CONSULTATION WITH INDUSTRY
Page
I. Introduction C-2
II. Industry Contacts C-3
III. Example Letter C-A
C-l
-------
INDEMNIFICATION STUDY
Appendix C
I. INTRODUCTION
During the final phase of the study, representatives of industry
were contacted to obtain more information relating to specific circum-
stances warranting indemnification and the general reaction of
industry to the indemnification option.
In order to reach persons who might be familiar with EPA regula-
tory problems on an industry-wide basis, trade associations were the
primary target for consultation. Over 10 trade associations repre-
senting industries likely to have environmental concerns were
contacted; in most cases, telephone discussions with several people
within each association were conducted.
Many persons contacted were unfamiliar with the concept of
indemnification. Frequently, persons requested a written description
of the study and of information that might be provided. In response
to this type of request, nine letters were sent to trade associations
(see attachment). Most persons who requested that letter be sent to
them indicated their intention to distribute copies of the letter
among trade association members or to circulate the letter among the
professional staff. Therefore, it is likely that a much larger, but
indeterminate, number of persons are aware of the study and our
request for industry information.
Frequently, trade association representatives referred us to spe-
cific individuals within corporations who might be helpful to us or
have special experience or insight. Follow-up interviews of these
corporate representatives were conducted.
C-2
-------
INDEMNIFICATION STUDY
Appendix C
II. INDUSTRY CONTACTS
Alcoa
The Aluminum Association
Amax
American Consulting Engineers Council
American Insurance Association
American Iron and Steel Institute j ?j
American Mining Congress ];
American Paper Institute '
American Petroleum Institute
Atlantic Richfield
Battery Council International
Business Roundtable
California League of Food Processors
Chemical Manufacturers Association
Chemical Specialties Manufacturers Association
Chevron
Commercial Union Insurance
Consolidated Coal t
Dow Chemical Company
E.I. dePont de Nemours & Co.
Edison Electric Institute
Electrical Power Research Institute
Exxon Monsanto
Food and Drug Law Institute
Formaldihyde Institute
Motor Vehicles Manufacturers Association
National Solid Waste Management Association
National Agricultural Chemicals Association
National Association of Manufacturers
National Federation of Independent Business
National Food Processors Association
National Forest Products Association
National Insulation Certification Institute
National Small Business Association
NCASI—National Council of the Paper Industry for Air and Stream
Improvement
Pharmaceuticals Manufacturers Association
Procter and Gamble
J.R. Sinplot
Shell Oil
Smith Kline Corporation
U.S. Chamber of Commerce
U.S. Steel
Water Pollution Control Federation
C-3
-------
INDEMNIFICATION STUDY
Appendix C
III. EXAMPLE LETTER
Dear
I appreciate the interest in our study on indemnification that
you expressed earlier today* In response to your request for more
information, The Research Group is currently participating in a study
of the feasibility and desirability of indemnifying affected indus-
tries for significant compensable losses caused by EPA actions.
Indemnification may also be appropriate as an incentive for industries
to protect or enhance the environment or as a means of ensuring that
EPA will consider the economic impact of regulatory acts. This study
will ultimately be presented by EPA to Congress as mandated by § 25(a)
of the Toxic Substances Control Act.
We are reviewing all EPA programs including air, water, pesti-
cides, waste disposal, and toxic substances to identify those EPA pro-
grams to which Congress might apply an indemnification scheme. The
study includes an analysis of all environmental statutes, together
with their legislative history, implementation, and the potential eco-
nomic impact of government indemnification. We are conducting Inter-
views with Agency personnel, persons Involved in the legislative
process, and public Interest representatives to obtain relevant infor-
mation. In order to present a well-balanced study, we are also
seeking information and opinions from industry.
To date, indemnification has not been favored by Congress as a
general regulatory relief or remedial mechanism. Its use has been
confined to narrow circumstances under the Federal Insecticide,
Fungicide and Rodenticide Act, as amended; agricultural indemnities to
farmers for losses due to diseased livestock, contaminated milk, etc.;
the Consumer Product Safety Act; and several occasions warranting a
special Congressional indemnification bill* It is most often proposed
as a viable regulatory mechanism in situations where an Agency has
committed (or may commit) a mistake which causes substantial losses to
a regulated party. It has also been suggested that indemnification
may be appropriate where the Agency merely exercises bad judgment or
later reverses a policy decision creating losses within a given seg-
ment of industry which relied on the original policy decision. Apart
from a "fault" base rational, it has also been argued that indemnifi-
cation may provide a mechanism for ensuring rapid, cooperative
response in circumstances warranting emergency actions and thus assist
In accomplishing an Important goal of a statute. It might also serve
as a disincentive to Agency use of certain extraordinary relief
mechanisms, except in cases where alternatives are insufficient.
C-4
-------
INDEMNIFICATION STUDY
Appendix C
Mr. XXXXXXXXXX
Date
Page Two
In order to present a thorough and fully informed report to
Congress on this issue, it is very important that representatives of
industry, such as the Chemical Manufacturers Association, consider the
potential impact of an indemnification program for regulated parties
and provide us your opinion and any other relevant information that
would assist us in our consideration of the iBsue. If your organiza-
tion or its members support the concept of indemnifying regulated par-
ties in certain situations, we would like to document or discuss
specific factual events or circumstances in which compensation would
have been desirable, either as a matter of fairness or to obtain the
appropriate regulatory or legislative results.
We will be pleased to discuss the TSCA § 25(a) study or answer
questions you may have regarding situations possibly warranting indem-
nification. Please contact me at (804) 977-5690. We ask that you
respond to us, at least informally, by February 17, 1982. If you
would like to submit a formal or written response, please let us know
so that a workable deadline may be established that will ensure full
consideration of your views in the draft report to be submitted to
EPA.
Thank you very much for your Interest and cooperation.
forward to hearing from you.
We look
•
Sincerely,
C-5
-------
INDEMNIFICATION STUDY
Appendix D
APPENDIX D
INDEMNIFICATION FINANCING SYSTEMS
Pa
I. Introduction ...................... D-2
II. Financing System Options ................ D-3
A. Cash Payments From Regular Appropriations ...... D-4
B. Cash Payments From Special Appropriations After
The Indemnifiable Event ............... D-5
C. Cash Payments From Dedicated Tax Revenues ...... D-5
D. Cash Payments From a Revolving Fund ......... D-6
E. Direct Assumption of Liability by the Government . • D-7
F. Special Tax Credit Resulting in Reduced Tax Revenues
Revenues ...................... D-7
III. Selection of an Appropriate Financing System ...... D-9
A. Purpose of the Indemnification Payment ....... D-9
B. Need for Immediate Payment ....... . ..... D-10
C. Frequency and Predictability of Events ....... D-10
IV. Conclusion ........................ D-12
Exhibit 1 — Characteristics of Existing Indemnification
Financial Systems ....... . .............. D-13
Footnotes ........................... D-14
D-l
-------
INDEMNIFICATION STUDY
Appendix D
I. INTRODUCTION
Government programs can be financed in a number of ways* This
appendix provides a brief overview of financing alternatives most per-
tinent to indemnification programs. It also provides guidelines for
appropriately matching a financing system to a particular indemnifica-
tion program. A wealth of literature exists on the use of tax
credits, interest subsidies, and similar devices for use as incen-
tives. As much of the literature discusses the relative advantage of
one devise over another in specific circumstances, it would be helpful
in designing the financing for a particular indemnification program.
This appendix is not intended as a review of all of this information,
but as an introduction to the possibilities and their implications for
indemnification.1
D-2
-------
INDEMNIFICATION STUDY
Appendix D
II. FINANCING SYSTEM OPTIONS
Financing systems have two basic components:
• Benefit distribution component — mechanism by which economic
benefits are transferred to firms which suffer eligible
losses.
• Funding component — mechanism by which the economic resources
to be transferred to eligible firms are collected.
There are four basic options for distribution of benefits;
• Cash payments — direct payments to eligible parties.
• Assumption of liability — transfer of responsibility for a
potential liability to the government.
• Tax assistance — selected tax relief for eligible firms.
• Credit assitance — reduced cost of borrowing or access to
otherwise unattainable loans.
There are also four basic options for funding programs:
• Appropriations by Congress from general tax revenues -- either
regular or special appropriations.
• Appropriations by Congress from dedicated or "ear-marked"
revenue sources.
• Revolving fund account — budget initially may be provided by
general tax revenues or ear-marked tax receipts but fund is
perpetuated by repayments (e.g., of loans) or recoveries of
expended funds.
• Tax expenditures — reduced tax revenues resulting from a
special deduction, credit, deferral, or other special tax
concession.
Each of the benefit distribution and funding options has varia-
tions, resulting in many possible alternative structures for a finan-
cing system. For purposes of providing an overview of the range of
alternatives, we discuss the following systems:
• Cash payments from regular appropriations.
pi
D-3
-------
INDEMNIFICATION STUDY
Appendix D
• Cash payments from special appropriations after the indemnifi-
able event.
• Cash payments from dedicated tax revenues.
• Cash payments from a revolving fund.
• Direct assumption of liability.
• Special tax credit resulting in reduced tax revenues.
A. Cash Payments From Regular Appropriations
Under this alternative, indemnified parties would receive govern-
ment checks in the amount for which they are eligible. These payments
would be funded by appropriations from general tax revenues. The
appropriations could have many different combinations of features with
respect to: limits on the amount of funds available, the agency's
authority (or lack thereof) to carry over the unobligated funds from
one year into the next, and the timing of the agency's requirements to
request new budget authority.
An example of a regularly funded indemnification program is USDA's
payment of compensation for contamination of milk, milk products or
dairy cattle by pesticides that were registered under FIFRA at the
time of their use.2 The compensation awards are paid by U.S.
Government checks, drawn on a USDA general fund expenditure account.
In making its regular budget request each fiscal year, USDA includes
an amount to cover expected payouts plus a margin for unexpected
claims. At the end of each fiscal year, unobligated funds for this
account are returned to the Treasury.
This alternative can be relatively efficient in handling frequent
and predictable compensation claims. Although this alternative is
flexiblle in the amounts of funding it can handle, the expected
requirements should be large enough and frequent enough to justify the
permanent staff. Failure to provide adequate funding for expected
requirements will tend to disrupt normal agency operations (i.e., by
forcing normal operating budgets to be diverted to cover indemnifica-
tion claims) or to undermine the incentive effects of the program
since actual payment of claims cannot be assured.
Another example of this alternative is the Court of Claims
Judgment Fund which is entirely financed by appropriations from
general revenue. It is replenished from time to time in bulk and
funding decisions seem unrelated to Individual claims drawn from it.
Many different kinds of claims are paid from this fund after decision
by the Court of Claims. The Mizokami Brothers were paid from it for
their losses associated with an erroneous FDA determination that their
spinach was contaminated.
D-4
-------
B.
INDEMNIFICATION STUDY
Appendix 0
Cash Payments From Special Appropriations After The Indemnl-
fiable Event
This alternative differs froa the previous alternative in that the
funding is requested after an indemnifiable event occurs. By not
seeking funding until after the event, there is less certainty that
payment will be made but more certainty about the amount needed.
Therefore, this alternative may be less effective than the previous
alternative in motivating private parties or government agencies.
However, the certainty may be increased by statutory guarantees or
administrative pre-arrangements.
An example of special appropriations is the Art and Artifacts
Indemnification Program,^ administered by the National Endowment for
the Arts (NEA). Under this program, the NEA is authorized to enter
into agreements with museums and other institutions to Indemnify these
parties for loss or damage to works of art, rare manuscripts, and
other specified artifacts while on exhibition in the United States or
overseas. Any party that desires to receive coverage for a particular
exhibition must apply to the NEA. Upon approval of the application,
the NEA and the party wishing to be covered enter into a formal indem-
nity agreement, specifying the amount of coverage to be provided, sub-
ject to certain statutory limitations. No premiums are paid to the
NEA for this coverage. In the event of an eligible loss, the NEA
would certify its investigation of the loss to Congress under a statu-
tory arrangement. Congress would then appropriate money for compensa-
tion. Although appropriations are not made until after the loss, the
certainty of receiving them is high because the statute allows NEA to
pledge the full faith and credit of the United States.
For programs where claims will occur on a non-routine and infre-
quent basis and where the amounts of claims are unpredictable,
"after-the-fact" special appropriations usually will be more appropri-
ate. Backed by statutory authority to ensure eventual payment, this
approach provides crediblity to the program while avoiding unnecessary
budgeting of funds.
C'. Cash Payments From Dedicated Tax "Revenues
This alternative differs from the previous two in that revenues
come from dedicated sources such as special taxes or fees assessed on
a designated class of individuals (i.e., the users of a certain
product) or companies or a designated product (i.e., certain
chemicals). The revenues could be generated either continuously to
meet routine and predicted requirements or periodically to meet non-
routine and infrequent needs or to replenish the fund from previous
payouts.
The exact relationship between the use of the funds and the
individuals, firms or products on which the taxes are assessed can
D-5
-------
INDEMNIFICATION STUDY
Appendix D
vary and can be designed in such a way as to create a variety of
incentives or disincentives on the part of the taxed parties or the
users of affected products.
In the case of Superfund, most of the fund comes from taxes on
certain chemicals that are either directly or indirectly (as chemical
building blocks) most frequently associated with hazardous chemicals
that come into contact with the environment. The primary objective of
this approach to funding was to create an incentive for companies that
generate these chemicals to exercise greater care in managing them so
as to avoid spills or other releases into the environment.
A secondary objective was to free the fund to some extent from the
requirement to have new funds appropriated from Congress each year.
However, to prevent the loss of congressional control a portion of the
fund still comes from general revenues and the entire fund still is
subject to appropriations acts.
To establish this financing system, several issues have to be
decided. The first is whether the assessments of firms, individuals
or products will be a regular, ongoing activity, almost like an
insurance program, or will occur only after eligible losses. Other
Issues are the definition of the types of firms that would be assessed
and the basis for collecting the assessment (e.g.. a per-unit tax on
certain types of goods paid by the manufacturer). These issues have
to be decided as the program is established. Otherwise, funding may
never happen because agreement as to the correct magnitude of the
claims and as to the basis for contributing to the fund will not
occur.
D. Cash Payments From a Revolving Fund
A revolving fund is a pool of money out of which indemnity pay-
ments are made, which is regularly replenished through the repayment
or recovery of expended funds. It is not unusual for such funds to be
initially funded and even continually supplemented from general or
special appropriations or from special assessments, fees or penalties
from parties affected by or in some way related to the indemnifiable
events.
The key aspect of revolving funds is that they take on some form
of self reliance. This reduces the need for continued appropriations
and can provide an incentive for good fiscal management by the agen-
cies responsible for the program. This same aspect tends to make the
program somewhat independent of Congress, thereby circumventing one of
the more effective means of congressional oversight. For this reason,
expenditures from such funds still may be made subject to annual
appropriations acts even though funds are not coming from tax
revenues.
D-6
-------
INDEMNIFICATION STUDY
Appendix D
Section 311 of the CWA and Superfund both have revolving fund
characteristics. Funds expended for cleanup activities are later
recovered if possible (i.e., if a responsible party can be found who
is financially viable).
This type of system requires a pre-existing administrative appara-
tus to determine the rules for who pays, how much, and when.
E. Direct Assumption of Liability by the Government
An assumption of the underlying risk by the government can provide
a substantial economic benefit though no payment is made. The govern-
ment might or might not establish a means of paying for the loss
before it occurs.
The swine flu progran Is a good example. Vaccine manufacturers
were so concerned about potential suits that they demanded protection.
The Congress finally agreed to make the sole cause of action one
against the government. In addition to assuming the liability, the
Department of Justice has absorbed the considerable cost of defending
against the four thousand suits filed.
F. Special Tax Credit Resulting in Reduced Tax Revenues
A number of alternatives exist involving the government foregoing
tax revenues for specific purposes. Foregoing tax revenues or making
indemnification payments have the same net effect on the amount of
government funds available for other purposes.
Losses suffered by reason of agency actions are usually eligible
for deductions as ordinary business expenses. Thus, some tax
assistance is already available. At times Congress has been explicit
in declaring these losses eligible for a deduction, but not for
indemnification. The tax bill of the summer of 1981 contained such a
provision. Truck certificates which had formerly had considerable
value were rendered worthless when the trucking industry was
de-regulated. Congress used the tax bill to make clear Its view that
the trucking firms could take a deduction for the value of the
certificates.6
Tax credits could provide a much greater benefit for losses con-
sidered more than ordinary expenses. Eligible firms would be allowed
to reduce their tax liability by some specified amount. An important
aspect of this alternative is the formula or set of guidelines that
would specify the amount of the special tax credit. The amount of the
credit would be related to the eligible losses suffered. For firms
with tax liabilities that are less than the amount of the allowable
special tax credit, the program could be designed to allow application
of the credit to other years or could allow the credit to be sold to
another firm.
D-7
-------
INDEMNIFICATION STUDY
Appendix D
Tax credit devices are administratively simple to operate,
reducing the complexity of trensfering an economic benefit to the
indemnified parties. However, they should only be applied to very
large programs affecting many people or firms. The required changes
in tax forms and reference materials can only be justified by a
widespread use of the new provision.
D-8
-------
INDEMNIFICATION STUDY
Appendix D
III. SELECTION OF AN APPROPRIATE FINANCING SYSTEM
The five financing system alternatives provide a sense of the
spectrum of possible systems. In the discussion of each, dimensions
along which it seems most appropriate to differentiate programs
emerged, e.g., programs relying on regular appropriations may be most
appropriate when Inderanlfiable events are predictable. Exhibit 1 sum-
marizes important characteristics of the financial systems of existing j|l
indemnification programs. '•{'
In this section, guidelines are presented and discussed that can be
used to help define the kind of financing alternative most appropriate
for a particular indemnification program. The four aspects of an
indemnification program are as follows:
• Purpose of the indemnification program;
• Cost of the event relative to an individual party's financial
capability;
• Need for indemnification payments to be immediately available;
• Frequency and predictability of events.
A. Purpose of the Indemnification Payment
Indemnification programs usually exist for some combination of the
following objectives:
• Remedying Inequities;
• Creating disincentives to inappropriate or unreasonable
action;
- • Creating incentives for private parties or otherwise facil-
itating public goals.
For remedying inequities there is no Inherent bias on the selec-
tion of funding methods. The key is that relief reach the correct
parties and in an effective manner. Cash payments are the most direct j* |
distribution method, but tax credits or other forms of indirect relief '
can be Just as valuable and effective under the right circumstances.
The second two objectives are somewhat independent of distribution
methods, but can be affected substantially by the choice of funding
methods. For example, a program whose "objective is to encourage
greater industry care in handling hazardous materials" might choose
special taxes on chemicals as a source of revenues for indemnifying
D-9
-------
INDEMNIFICATION STUDY
Appendix D
parties against the voluntary assistance in cleanup activities.
Likewise, a program aimed at discouraging ill-founded enforcement
actions might draw some or all of its funds directly from agency
operating budgets.
Creating incentives for private parties also may depend upon the
degree of certainty the parties feel. If the indemnification program
is established and funded in advance, private parties are more likely
to believe in the program and thus to be motivated by it. However,
some parties require more certainty than others. The contractor hired
to clean up abandoned waste sites were satisfied with a contingent
promise to consider indemnifying them when a loss occurs. Swine flu
vaccine manufacturers transfered their entire potential liability
(except that resulting from negligence) before they began to manufac-
ture the vaccine.
B. Weed for Immediate Payment
The need for immediate receipt of funds to respond to the indem-
nifiable event has a bearing on the choice of financing system. Some
indemnifiable events involve large costs to individuals or firms. If
immediate payment is not forthcoming, serious economic harm may occui.
Also, for some incentive programs, immediate payment is important to
the success of the programs.
The USDA animal disease indemnification program5 provides an
example of this point. If the Secretary of Agriculture declares a
national disease emergency for some animal disease situation, funds
for eradication and indemnification are made available from the
Commodity Credit Corporation. These funds must subsequently be paid
back through congressional appropriations. The success of these
disease emergency situations depends on quick government action and on
industry cooperation. That cooperation is in turn dependent on the
availability of indemnification funds to replace animals destroyed as
part of the eradication program. Thus, speedy distribution of indem-
nification payments is essential, and hence, the need for a funding
arrangement such as borrowing the Commodity Credit Corporation funds.
However, the administrative costs necessary to provide quick response
can only be justified by the expectation of a series of losses.
C. Frequency and Predictability of Events
Another major factor to consider in choosing the financing system
is the likely frequency and predictability of the indemnifiable
events. Some events recur with a regularity that allows for esti-
mating the expected annual costs and requires an administrative
apparatus to handle claims. Other events might occur so infrequently
and with widely varying costs as to make the total required payments
unpredictable. In cases where the event is frequent, up-front
financing through insurance-like mechanisms or regular appropriations
D-10
-------
INDEMNIFICATION STUDY
Appendix D
can be appropriate. Up-front financing becomes less attractive, if
events occur infrequently and with widely varying costs.
'ill
r"
D-ll
-------
INDEMNIFICATION STUDY
Appendix D
IV. CONCLUSION
There are many financing mechanisms appropriate to indemnification
programs. Choosing between them should be done in the context of a
particular program to ensure that the financing components are compat-
ible with the aims of that program.
D-12
-------
EXHI!
CHARACTERISTICS OF EXISTING INDEMNIFICATION FINANCIAL SYSTEMS
USDA
EPA
Air
EPA
Clea
Program
Animal Disease
S Il3(b) Clean
Act
$ 202(a)(3)
n Water Act
Benefit
Distribution
Method
Cash payments
Cash payments
Cash payments
Funding
Method
Regular Approprl-
tlona
Appropriations to
Insufficient Claims
and Judgments Fund
Regular Appropria-
tions**
Purpose Costs of Need for
of Events Relative to Immediate
Program* Financial Capability Payment
3 Can be very high Yes
1,2 Low No
3 Uncertain No
Frequency and
Predictability
of Events
High
Low
Low
Unique
Aspects
Ability to borrow
funds from Comoodlty
Credit Corp. for
emergencies.
7 EPA 5 15 FIFIA
Cash payments None established
1, 2, 3 Low-Moderate
No
Low
Relied on Industry
consolidation of
claims In Sllvex
case.
Arts and Artifacts
Cash payments
Special Approprla-
atlons
Can be high
No
Low
EPA $ 311
Clean Water Act
Cash payments
Special Approprla-
atlona plus
recoveries of funds
from responsible
parties
Can be high
No
High
*1 • Remedying Inequities*
2 - Creating disincentives to inappropriate or unreasonable government action.
3 - Creating incentives for private parties or otherwise facilitating a public goal.
•* Payment of the indemnity la not a certainty. If a grantee qualifies for the indemnity, they must still compete for the funds with other potential uses.
-------
INDEMNIFICATION STUDY
Appendix D
FOOTNOTES
*A more detailed discussion of this area can be found in "Identi-
fication and Preliminary Evaluation of Basic Financing System Options
for an EPA Indemnification Program," EPA, Prepared by American Manage-
ment Systems, Inc., December 1979.
2Equal Opportunity Act of 196A, § 331(a) (rider to EOA).
3Arts and Artifacts Indemnity Act, Public Law 94-158, 20 U.S.C.
§S 971 et aeq.
4This assumes that the firms have played some part in bringing
about the situation requiring correction.
521 U.S.C.
, 134(d).
626 U.S.C. 165 (note), 95 Stat. $ 266 (Aug. 13, 1981).
D-14
-------
INDEMNIFICATION STUDY
Appendix E
APPENDIX E
COSTS AND IMPACTS OF INDEMNIFICATION
Page
I. General Approach E-2
II. Cost of a TSCA Indemnification Program E-5
A. Introduction ._ E-5
B. Costs E-5
c
I. Number of Banned Chemicals E-5
2. Value of Inventory of Banned Chemical E-6
C. Indemnification Impacts E-9
1. Government E-9
2. Manufacturers E-9
3. Distributors E-10
A. Users E-10
III. Some Indemnification Impacts of FIFRA Program E-ll
A. Manufacturers E-ll
B. Distributors E-14
C. Users E-15
Footnotes E-16
1
i'
ir
E-l
-------
INDEMNIFICATION STUDY
Appendix E
I. GENERAL APPROACH
An approach to estimating the cost and impact of an indemnifica-
tion program is first discussed in general, and then illustrated by
application to a hypothetical TSCA indemnification program, and a
hypothetical ban of a pesticide under FIFRA.
The costs to the government for an Indemnification program are
largely a function of (1) the number of indemnifiable events and (2)
the cost of each event. Estimates of the number of indemnifiable
events require:
• definition of what constitutes an Indemnifiable event,
• specificity of estimates by type of event, and
• estimates of expected values and likely upper bounds on the
number of events of each type.
Estimates can be formed on the basis of a review of past experience o
the program, review of similar programs, and opinions of experts.
The cost of each indemnifiable event must also be estimated.
These estimates require:
• definition of the coverage of the program,
• data on costs of similar events or other relevant data, and
• data by type of indemnifiable event.
Total costs of a program are estimated as the product of the
number of events and the cost of each event. A range can be estab-
lished around this estimate using ranges determined while estimating
the number and cost of events. The range of estimates provides a
measure of the uncertainty associated with the estimates.
Only after the costs are reliably estimated can any analysis of
the impacts of the cost be made. The impact is the effect that the
cost has on the economic viability and behavior of the parties
directly and indirectly affected.
Although the methodology is straightforward, applying it is not.
The § 15 model limits indemnifiable costs to inventory. This is a
comparatively concrete and verifiable measure. It is not the only
measure of loss that could be used. However, using any other measure
would add to the problems of estimating cost. For example if indemni
fication is desired to compensate for all of a company's loss when a
E-2
-------
INDEMNIFICATION STUDY
Appendix E
.product is banned, some way would have to be found to value the
research and development costs, the overhead sunk into the product,
the future profits now lost, the cost of substitutes for customers,
etc. This would be much more difficult to reliably estimate since the
lack of definiteness leaves room for argument, and since the amounts
would vary so much depending upon when in a product's life cycle the
regulation was imposed, what kind of product it was, and whether the
firm was a manufacturer, distributor, or user.
Estimating the frequency of unplanned events, especially new
programs dissimilar to existing programs, may produce widely varying
costs depending on assumptions. This is especially true for programs
involving events that occur infrequently. Another complicating factor
is that the existence of the program itself may alter behavior of the
government, individuals, or firmsC For example, a government agency
might be more likely to act in a situation which would result In
indemnification payments because the affected parties would be less
severely Impacted by the action with the availability of compensation.
The government agency could also be envisioned to be less likely to
act because it might have to make large indemnification payments*
These impacts on behavior are difficult to pin down since they must be
based on good cost estimates, but will also affect the cost of the
program* The descriptions of impacts also rely on the hypothesis that
a government agency faced with a potentially costly decision will
react in the same way that a "reasonable economic man" would. This
hypothesis may well be Incorrect. The only thing that can be said
with some certainty is that the total cost of an indemnification
program will not approach the magnitude necessary to have an impact on
the overall federal budget or the economy of the country. The cost of
a program might well be large enough to affect the budget of an indi-
vidual agency or program.
Estimating costs of each event is also difficult, especially where
there is great variability between different events. One can deal
with "average" events, but if the number of events is small, high
uncertainty will be associated with the estimate of program costs.
Another difficulty can occur in obtaining data for the analysis* Some
of the key data items may not be publicly available and those with the
data may not cooperate by supplying it to the analysis. Assumptions
and estimates can be made from the data that are available, but the
results will be less certain than otherwise.
The difficulty and uncertainty associated with estimates of indem-
nification program costs can be a deterrent to establishment of an
indemnification program. With high uncertainty in what the program
will cost, the actual costs could far exceed what is estimated to be
the expected cost. Over a number of years, the estimate of expected
annual costs could be expected to more closely approximate the running
average of the actual costs than would the expected average cost
approximate the actual cost for any particular year. It could be dif-
E-3
-------
INDEMNIFICATION STUDY
Appendix E
flcult to "sell" a program with greatly fluctuating costs even if the
expected average annual costs approximated the actual average annual
costs.
There are cost continment techniques that could be used to over
come the problem of uncertainty, though they would limit the amount of
indemnification received and thus po-ssibly reduce the effectiveness of
the program. A ceiling can be placed on the size of individual
claims, or on the amount of expenditures that the program could make
in any one year4 The indemnification program might only be authorized
for a year or two to give Congress an opportunity to reconsider based
on actual experience.
In part II of this appendix, the costs of a hypothetical TSCA
indemnification program are analyzed, and in part III an attempt is
made to analyze the impacts of a hypothetical pesticide ban under
FIFRA.
E-4
-------
INDEMNIFICATION STUDY
Appendix E
II. COST OF A TSCA INDEMNIFICATION PROGRAM
A. Introduction
The Toxic Substances Control Act (TSCA) does not provide indemni-
fication for any costs incurred by individuals and firms as a result
of TSCA actions taken by EPA. In this paper, we analyze the costs
that would be associated with a hypothetical TSCA indemnification
program — one patterned after the Federal, Insecticide, Fungicide,
and Rodenticide Act (FIFRA) § 15 indemnification program.
We hypothesize that the TSCA program would -apply only to chemicals
which have made it through the TSCA § 5 premanufacture notification
(PMN) review process and for which the manufacture, processing, dis-
tribution, use, or disposal is subsequently prevented by EPA action
under § 6 or 7. Indemnification would be provided to compensate for
the cost of the chemical held by any individuals or firms at the time
of the EPA action.
B. Costs
The cost to EPA of a TSCA indemnification program of this type is
a function of the number of chemicals likely to be banned and the cost
of the Inventory in those chemicals when the bans are Imposed. Each
of these factors will be discussed in turn.
1. Number of Banned Chemicals
The number of chemicals likely to be banned is a function of
the thoroughness of the PMN review and the frequency that EPA uses
product bans with no inventory sell.off as a regulatory device. The
thoroughness of the review process is of course relative. Because of
this, it is useful to compare the PMN review process with the pesti-
cide preregistration review process that occurs under FIFRA. With a
sense of how these review processes compare, we can review the FIFRA
registration cancellation experience and analyze what general implica-
tions it might have for the number of chemicals that make it through
the PMN process and would subsequently be regulated. Of course, there
are basic differences between pesticides and other chemicals which
limit the usefulness of comparisons between the review processess.
In a previous comparison1 of the PMN and FIFRA review
processes, we concluded that the two processes are very different.
The most significant difference is that TSCA requires none of the
testing that Is the backbone of the FIFRA process. The testing
required for approval of a new pesticide is quite lengthy and costly.
The PMN process, on the other hand, is intended as an Initial
screening and reporting mechanism and is not Intended to provide a
examination of all risks.
E-5
-------
INDEMNIFICATION STUDY
Appendix E
Even with the thorough review process before pesticides are
registered, there still have been over ten pesticide registrations
cancelled or notices of intent to cancel issued. This indicates that
a preregistration review process does not ensure that unreasonably
unsafe pesticides will be prevented from being registered.
This FIFRA experience implies that some number of chemicals
that make it through the PMN review process will be subsequently regu-
lated by EPA. The FIFRA experience, however, cannot reasonably be
used to estimate the number of chemicals involved. This is because
pesticides by their nature are more toxic than the average chemical
and therefore cannot be considered representative of chemicals in
general. The criteria for subsequent EPA regulation are also differ-
ent for pesticides than they are for other chemicals.
EPA has to date initiated very few regulatory actions under
§ 6 of TSCA. None of these § 6 actions have been directed at chemi-
cals that have made it through the PMN review process. Because most
of the chemicals now on the market were introduced before TSCA was
implemented, we would expect to see over the next decade very few EPA
actions taken against chemicals that have been reviewed prior to com-
mercial production. Additionally, product bans represent an extreme I
regulatory action and would be expected to occur infrequently. For
these reasons, over the long term an assumption of one product ban
every two years involving a chemical that made It through the PMN pro-
cess could be considered a conservative upper bound on the actual
number of such bans that will occur over time.
2. Value of Inventory of Banned Chemical
The value of the inventory of a banned chemical depends on the
chemical's production volume, price and use patterns. In a previous
analysis^ of a sample of 1,459 chemicals from the TSCA Chemical
Inventory which were candidates for a § 8 reporting rule, we deter-
mined the distribution of these chemcials across a series of produc-
tion categories. This information is shown in Exhibit 1. If we
assume that all chemicals within a production category are produced at
the production volume represented by the midpoint of the category, the
average production volume can be calculated to be approximately 88
million pounds.
E-6
-------
INDEMNIFICATION STUDY
Appendix E
EXHIBIT 1
PRODUCTION VOLUME DISTRIBUTION OF SAMPLE OF 1,459 CHEMICALS
FROM THE TSCA CHEMICAL INVENTORY
Production Range (Ibs)
0
1,000
10,000
100,000
1 million
10 million
50 million
100 million
500 million -
greater than
1
10
1,000
10,000
100,000
million
million
50 million
100 million
500 million
1 billion
1 billion
Percentage of Chemicals
in Category I a/
11.5
7.8
11.2
14.4
15.8
14.8
7.0
13.0
2.4
2.1
a/ These percentages relate only to the chemicals in the sample
of 1,459 for which production volume was reported.
Source; "Analysis of TSCA $ 8(a) Small Manufacturer Exemption,"
ICF Incorporated, May 1981.
There are three reasons why this average may be higher than
the average production volume of chemicals that have been banned after
having made it though the PMN process. Firstly, because economic
impact would be factored into consideration of a product ban, we would
expect higher volume chemicals to be less likely to be banned than
lower volume chemicals. Secondly, the sample of 1,459 chemicals may
have been biased toward larger production volumes because these chemi-
cals were selected on the basis of the degree of hazard they poten-
tially presented. This means that larger production volume chemicals
would be over-represented because, everything else being equal, they
represent more hazard than do lower production volume chemicals. The
third reason that the 88 million pounds might be biased upwards as an
estimate of the average production volume of PMN "cleared" chemicals
that are subsequently banned, relates to the difference In age of the
1,459 chemicals versus PMN "cleared" chemicals. The production volume
of a chemical typically follows a life-cycle as shown in Exhibit 2.
Newer chemicals will tend to be of lower production volume than more
mature chemicals. Therefore, one would expect the production volumes
for the 1,459 chemicals to be higher than a sample of relatively new
chemicals. However, over time the universe of chemicals which have
"cleared" the PMN process will Increase and their age profile will
become closer to the age profile of the 1,459 chemical sample. For
purposes of this analysis, we will use the 88 million pound average,
recognizing that it probably is an inflated estimate.
B-7
-------
INDEMNIFICATION STUDY
Appendix E
EXHIBIT 2
TYPICAL CHEMICAL LIFE-CYCLE
Volume
Growth
Mature
Aging
Growth
Embryonic
Time
Source; "Impact of TSCA Proposed Premanufacturing Notification
Requirements," Arthur D. Little, Inc., December 1978.
The volume of manufacturers' sales associated with the 88 *
million pounds of chemicals can be estimated by multiplying by the
average price of a pound of chemicals. Most,commercial chemicals sell
for under $1 per pound.•* For purposes of this analysis we will assume
a $.75 per pound price for the average chemical. Therefore sales for
the average chemical equals $66 million.
In 1979, for a sample of 75 chemical companies, the average
ratio of sales to inventories was 6.9. Applying this ratio to the
$66 million in sales for the average chemical yields $9.6 million in
manufacturers' inventories associated with the average chemical. Add-
itional inventories in the chemical are held by distributors and
users.
Most chemicals are intermediate materials sold to other manu-
facturers who process them into end products.5 Also, most large
chemical users buy direct from the manufacturers; less than 122 of all
chemicals are sold through distributors.^ We assume that the inven-
tory holding practices of distributors and users is the same as the
manufacturers. We therefore assume that the distributors and users
combined hold an additional $9.6 million in inventories of the average
chemical.
We assume that the values of inventory we have estimated are
based on firms valuing their inventories on retail prices. With that
assumption we can use the same factors EPA used to estimate inventory
cost in calculating the level of indemnification to be payed in the
Silvex registration cancellation under § 15 of FIFRA. In the Silvex
situation, the cost of the manufacturer's inventory was set at 701 of
-------
INDEMNIFICATION STUDY
Appendix E
the Invoice price, and the cost of the users' and distributors' inven-
tory was set at 115% of the invoice price. The cost of manufacturers'
inventories of the average chemical are therefore $6.7 million and the
cost of the distributors' and users' inventories are $11 million. The
total cost of inventories is therefore $17.7 million.
We have assumed that every two years there will be one produc-
tion ban involving the payment of indemnification funds for the cost
of inventory. Therefore $17.7 million in indemnification will be
paid out every two years or an average of about $9 million per year.
C. Indemnification Impacts
1. Government
c
It is not clear how the requirement to pay indemnification for
inventory losses associated with product bans would influence the
government's decision regarding the ban. Product bans are an extreme
regulatory action that are unlikely to be used except where their need
is clearly warranted. This would imply that an indemnification provi-
sion would not have a significant influence on whether or not to
declare the ban. However, there may arise decisions regarding product
bans that are close calls. The availability of Indemnification funds
could result in not banning the chemical because of the need to pay
out a large amount of compensation to the industry. The availability
of indemnification funds could also have the opposite effect in
influencing the government to ban the chemical because the economic
impact to the industry would be softened, though this does not seem
likely.
2. Manufacturers
The decision to invest in the production of a new chemical
involves consideration of the expected returns from that investment
and the uncertainty associated with those returns. An Indemnification
provision which would cover the cost of an Inventory loss as a result
of a product ban under TSCA would make, investment in new chemicals
more attractive.
The degree to which an indemnification provision would
increase the attractiveness of investment in new chemicals is a func-
tion of the ratio of the amount of losses covered by indemnification
to the amount of total losses. In our analysis of the FIFRA § IS
indemnification provisions, we found that, for the pesticide we
studied, the indemnification program should not significantly alter
the initial investment decision. This same conclusion applies to
chenicals similar to the pesticide we studied. It is likely that with
product bans an uncommon regulatory approach, manufacturers' invest-
ment behavior in new chemicals should not change significantly.
E-9
-------
INDEMNIFICATION STUDY
Appendix E
3. Distributors
As we have pointed out, distributors handle only a small part
of chemical sales. Their losses due to a product ban would be mostly
their inventory cost. If this is indemnified, their losses would be
small. Without indemnification, they would tend to hold smaller
inventories and perhaps diversify more in the products they handle.
However, as product bans do not occur often, distributors are unlikely
to significantly alter their behavior.
A. Users
The primary users of chemicals are chemical manufacturers
themselves. Without indemnification, these users would tend to hold
smaller inventories and might be less likely to use chemicals that
have a high probability of being banned. However, the predominant
factor in choosing between chemicals would relate to how well they
serve the users' purposes.
*
L
E-10
-------
INDEMNIFICATION STUDY
Appendix E
III. SOME INDEMNIFICATION IMPACTS OF FIFRA PROGRAM
The direct benefit of indemnification to qualifying manufacturers,
distributors, and users is the payment of the cost of finished pesti-
cide inventory in stock at the time of cancellation that cannot be
sold or used. Cost is not defined in the statute; however it may not
exceed the fair market value of the inventory. The cost of such
inventory is different for each part of the pesticide distribution
chain. In the Silvex situation, the cost of the manufacturer's inven-
tory was set at 70% of the invoice price, and the cost of the users'
and distributors' inventory was set at 115% of the Invoice price. We
have not been able to determine the basis for these percentages, how-
ever they appear consistent with price markups in the pesticide
industry.
A. Manufacturers
The impacts of pesticide registration cancellations on the manu-
facturers are determined by the timing of the cancellation. A manu-
facturer must typically invest large amounts of funds in research and
development, production, and promotion over a period of years before a
pesticide begins to be profitable. For many years, the cash flow
from the investment is negative. One estimate is that sales begin for
the average pesticide eight years after it is first synthesized.
Therefore, if a pesticide is cancelled soon after registration, the
manufacturer will take a heavier loss than if the cancellation occurs
later in the pesticide's life cycle.
Also, a manufacturer's inventory of a pesticide varies over the
year because while production is relatively constant, usage is
seasonal. Generally speaking, inventories are highest immediately
before the growing season and lowest immediately afterwards. Hence, a
registration cancellation immediately before the season will result in
higher potential inventory losses to manufacturers than would a
registration cancellation after the season.
In making investment decisions regarding the production of a new
pesticide, a manufacturer considers the expected return from the
investment and the uncertainty (risk) associated with that return.
Because returns and the risks associated with those returns are posi-
tively correlated (i.e., high returns are usually associated with high
risks), a manufacturer must accept some risk to achieve a desired
level of expected return.
To a manufacturer of pesticides, the possibility that EPA might
cancel a registration at some future time translates into increased
uncertainty (risk) associated with investment in a particular
pesticide. The fact that Indemnification payments might be available
E-ll
-------
INDEMNIFICATION STUDY
Appendix E
(under certain conditions) for some cancellations theoretically could
provide some incentive to a manufacturer to invest more in pesticide
research and development and production than would otherwise be
invested. This is because some of the investment risk is removed by
the availability of the indemnification payments.
The degree to which the availability of indemnification payments
for inventory losses may affect manufacturers' investment decisions
can be assessed to some extent through analysis of the expected cash
flows associated with a typical pesticide. Framing the analysis in
this way can provide important Insights in this regard. However, a
full assessment of the impact of indemnficiation payments on manufac-
turers' investment decisions would involve detailed study of a repre-
sentative sample of pesticides. Perhaps some particular category of
pesticides (e.g., moderate volume pesticides involving small research
and development costs) would be more (less) sensitive to the avail-
ability of indemnification payments than would other categories. Such
an approach was beyond the scope of our available time or budget.
Exhibit 3 provides cash flow information of the insecticide
TH 6040. TH 6040 was discovered in 1969, and extensive research and
development expenditures were necessary before the first registration
was obtained at the end of 1975. Presumedly the pesticide will be a
viable, commercial product until the end of the century. We can apply
the net present value (NPV) method, commonly used in business, to
determine the viability and profitability of an investment, to analyze
alternative registration cancellation scenarios and the Impact of
indemnification payments on the initial Investment decision. With
NPV, the timing of the cash flow and the discount rate assumptions can
be as Important as the magnitude of the case flows themselves.
The relative riskiness of an investment project is reflected in
the discount rate assumptions; in essence, the higher the risk, the
higher the discount rate. Intuitively, this means that future
dollars, being more uncertain, are valued less; the more uncertainty
associated with the future, the higher the discount rate. For the
development of commodity type pesticide chemicals, the discount rates
often range from 10 to 15 percent.8 For new innovative high-risk
pesticides, they can range from 15 to 25 percent. In the NPV analysis
of TH 6040, we bound the analysis by using discount rates of both 10
and 20 percent.9 The discounted cash flows from the pesticide, given
no cancellations, are shown in Exhibits 4 and 5* If no cancellations
occur, TH 6040 is expected to be very profitable.
The effects of cancellation one, five, and 10 years after regis-
tration are shown in Exhibits 6 and 7. Inventory losses are not shown
because they depend on the inventory level assumptions. Each pesti-
cide manufacturing plant is assumed to be worth 402 of its original
cost as salvage value.^ This is a very rough estimate, subject to a
large variance. Clearly, a plant could be worth a great deal more
E-12
-------
BIT 3
EXPECTED CASH FLOWS OF TYPICAL PESTICIDE - TH 60HO
T5in thousands)
ni
i
Is)
p>
Year
1969
1970
1971
1972
1973
197U
1975
1976
1977
1978
1979
1980
1981
1982
1983
1981
1985
1986
1987
1988
1989
1990
1991
1992
1993
199U
1995
1996
1997
1998
1999
Production
( Ibs)
0
0
0
0
0
0
0
3,675
79,163
223.99*4
370,356
682,106
1,155,788
1,557,338
1,889,013
2,220,7l4l4
2,388,700
2,392,881
2,395,288
2,355.39'*
2.2714,256
2,192,200
2.018,181
1,752.188
1,1485.206
1,2014,725
9*48.225
679,850
1455,075
273,0l4«4
91,013
RAD Costs
-875.0
-875.0
-26U.O
-5*40.0
-1,060.0
-1,230.0
1.U50.0
-6U2.0
-359.0
-30.0
-30.0
-30.0
Sales
at $16/lb
58.8
1.266.6
3,583.9
5.925.7
10.913.7
18.1492.6
214,917.14
30,22«4.2
35,531.9
38,219.2
38,286.1
38,3214.6
37,686.3
36,388.1
35,075.2
32,290.9
28,035.0
23,763.3
19,275.6
15.171.6
10.877.7
7.281.2
14,368.7
1.1456.2
Product ion
and
Di stribut ion
Costs
-191.1
-6014.2
-1.396.8
-2,197.7
-14,0714.7
-6,666.8
-9,035.2
-11.021.3
-12,836.6
-13,755.7
-13.778.6
-13,791.0
-13,573.5
-13,129.5
-12,680.14
-11,728. 1
-10,101.5
-8,6140.5
-7,003.1
-5,531.0
-14,062.14
-2.661.3
-1,665.2
-668.1
Capi taI
Costs I/
-1,125.0
-1,125.0
0
0
0
-1,125.0
1,125.0
0
-1.125.0
-1,125.0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2,700.0
Pre-Tax
Cash Flow
-875.0
Cross
After-Tax
Cash Flow
-1437.5
-875.0
-264. 0
-5140.0
-1,060.0
-2.355.0
-2,575.0
-77U.3
303.14
2,157.1
2,573.0
5, 6814.0
11,825.8
114,757.2
18,077.9
22,695.3
214,1463.5
2M.507. 5
214,533.0
214,113.0
23,259.0
22,395.0
20,563.0
17,9314.0
15,123.0
12.273.0
9,6141.0
6,815.0
U.620.0
2,7014.0
3,«487.1
-1437.5
-132.0
-270.0
-503.0
-1,177.5
-1.287.5
-387.2
151.7
1,078.6
1,286.5
2,81(2.0
5,912.9
7,378.6
9,039.0
11.3147.7
12,232.0
12,25U.O
12,266.0
12,056.0
11,629.0
11, 197.0
10,281.0
8,967.0
7,561.14
6,136.0
14.820.0
3.1408.0
2310.0
1,352.0
1,7143.6
Source: SRI, New Innovative Pesticides. 1977, with additions cited in the footnotes below.
I/Three $2.25 million plants was assumed to have been required to manufacture the pesticides.
was assumed to be UO percent of the original cost.
2/Ten year straight line depreciation, assuming a 50 percent tax rate.
After-Tax
Deprec i ;i t ion
Benefits 2/
0
0
0
0
0
56.3
112.5
112.5
112.5
112.5
168.8
225.0
225.0
281 .3
337.5
281.3
225.0
225.0
225.0
225.0
168.8
112.5
112.5
56.3
0
0
0
0
0
0
0
Net
After-Tax
Cash Flow
-1437. 5
-1437.5
-13?.0
-270.0
-503.0
-1,121.3
-1, 175.0
-27<4.7
26'4. 2
1,119.1
1.M55.3
3,067.0
6,137.9
7,659.9
9,376.5
11,629.0
12,1457.0
12,1479.0
12,1491 .0
12,281.0
11,798.0
11,310.0
10,3914.0
9,023.0
7,561.14
6.136.0
14,820.0
3,1408.0
2,310.0
1,352.0
1,7143.6
Salvage value of the plants
-------
EXHIBIT 6
EFFECTS OF CANCELLATION ON THE NPV
OF CASH FLOWS. BEFORE VALUE OF INVENTORY IS CONSIDERED*
($ in thousands)
No Cancellation
Cancellation After 1 Year
Cancel- Cancel-
lation After lation After
5 Years 10 Years
Cumulative Cash
Flows (less
salvage value
of plant)
Salvage Value
41,833
-4,619
1,930
^Assumes 10 percent discount rate.
EXHIBIT 7
EFFECTS OF CANCELLATION ON THE NPV
OF CASH FLOWS. BEFORE VALUE OF INVENTORY IS CONSIDERED*
($ in thousands)
22,317
of Plant(s)
NPV
78
41,911
677
-3,942
923
2,853
860
23,177
Cumulative Cash
Flows (less
salvage value
of plant)
Salvage Value
of Plant(s)
NPV
No Cancellation
Cancellation After 1 Year
11,989
9
11,998
-5,368
520
-4,848
Cancel- Cancel-
lation After lation After
5 Years 10 Years
-1,497
503
-994
7,056
303
7,359
*Assumes 20 percent discount rate.
E-12d
-------
INDEMNIFICATION STUDY
Appendix E
than 40X of its original cost if it can serve, with minimal change-
over, to produce another pesticide or chemical. A plant could be
worth much less if the machinery, building and location are unsuitable
for production of an alternative. The results would not be materially
affected for TH 6040 if a 0 or 100 percent salvage value were
assumed. *•
Research and development expenditures for one pesticide often can
benefit the development of other pesticides. In an ideal and most
certainly non-existent cost accounting system, the percentage of
research and development expenditures that would benefit other pesti-
cides would be allocated to these pesticides. We assume that this has
been done in the case of TH 6040, although it would not materially
affect the analysis if the proper allocations had not been made.
Exhibits 6 and 7, before considering inventory effects, show the
significant impacts of a cancellation one year after registration.
Even a cancellation 10 years after registration reduces the NPV signi-
ficantly. However, the effects of finished inventory in stock at the
time of suspension can be indemnified. To determine those effects, we
must estimate the levels of inventory that would exist at the time of
suspension.
Immediately before the Northern Hemisphere pesticide application
season, pesticide inventories are at a peak. Industry representatives
have stated that inventory could amount to as much as 100Z of a year's
production.^ Immediately after the Northern Hemisphere application
season, the pesticide inventories are apt to be at the lowest levels.
We will assume 10Z of a year's production to be a safe lower bound,
502 as a seasonable average inventory level, and 100Z as an upper
bound.
Exhibit 8 shows the after tax value of the pesticide inventory,
assuming the manufacturer receives the full wholesale price of $10 a
pound (62.5Z of retail) in Indemnity payments. We assume that the
TH 60AO manufacturer Is in effect "selling" its inventory at the
wholesale price, therefore they make a profit on the transaction.^
'Without indemnification, the manufacturer loses the variable costs of
producing the inventory (remember the fixed costs have already been
counted). Depending on what assumptions are chosen, the indemnifica-
tion program would raise the after-tax NPV of TH 6040 from $24,000 to
$4,810,000.
These are not insubstantial sums, but the crucial question is
their impact on the investment decisions of the firm. In other words,
how much is the expected NPV of an investment changed by the indemni-
fication program. Cancellation of a pesticide regulation, after all,
Is a rare event; only nine of any significance have occurred since
1970.14
E-13
-------
EXHIBIT 8
NET PRESENT VALUE OF INVENTORY GAINS/LOSSES:
WITH AND WITHOUT INDEMNIFICATION
($ in thousands)
Inventory Levels as a Percentage of Production
10% Discount
Rate
1
Years After 5
Registration 10
WITH INDEMNIFICATION |
i
100%
249
3032
3902
50%
125
1516
1951
10% |
25 |
303 |
390 j
i
j )
| WITHOUT INDEMNIFICATION j
i
I 100%
1
1 -71
| -706
| -907
50%
-35
-353
-454
10% |
-7 |
-71 |
-91 |
1
20% Discount
Rate
Years After
Registration
1
h~
1 1
5 |
10 I
i
100%
191
1652
1374
50%
96
826
687
10%
19
165
137
i
100%
50%
10%
-54 | -27 | -5 |
-384 I -192 I -38 |
-319 I -160 I -32
1 1 i
E-13a
-------
INDEMNIFICATION STUDY
Appendix E
We will assume a high expected cancellation rate for TH 60AO to
magnify the effects of the indemnification program on the NPV. We
will assume that the probability of cancellation increases as the
pesticide becomes older, its use becomes more widespread, and any
chronic health effects would have had time to develop. We assume a IX
chance of cancellation a year after registration, a 5% chance of can-
cellation five years after registration, and a 10% chance of cancella-
tion ten years after registration. For purposes of illustration, we
assume a 10% discount rate and an inventory level of SOX of the pro-
duction of the year following cancellation.
Exhibit 9 shows that, even with these very high cancellation
probabilities, the expected NPV of TH 6040 is only increased $336.00
by the indemnification. This amount, less than IX of the expected
NPV, is not large enough to significantly influence decision making
for most companies.
The conclusions of this analysis agree with anectodal evidence
gathered from industry representatives, both during the original leg-
islative hearings and in recent Interviews.^ The current indemnifi-
cation program should have had little effect on the amount of money
invested in the research and development, production, or marketing of
pesticides of the type represented by the pesticide we have analyzed.
For pesticides with lower research and development or capital costs,
overall losses associated with cancellation would be more senitive to
inventory losses. On the other hand, pesticides with expected low
research and development costs are likely to be more similar to chemi-
cals currently registered and thus are probably less likely to be more
similar to chemicals currently registered and thus are probably less
likely to be cancelled than a truly innovative pesticide such as
TH 6040. Furthermore, a pesticide would have to have very much lower
research and development and capital costs than TH 6040, even assuming
identical expected cancellation probabilities, for the availability of
indemnification payments to make much difference in the investment
decision.
B. Distributors
The distributors of pesticides typically invest very little except
the cost of inventory in any given pesticide. If a pesticide regis-
tration is cancelled, they can usually sell a substitute product. The
substitute product may be more or less profitable and the routine of
business may be disrupted, but on the whole the majority of the cost
of cancellation to a distributor is captured in their cost of inven-
tory. The § IS indemnification program is aimed at reimbursing these
costs. Without indemnification, distributors would have incentives to
keep inventories low and to diversify into handling more different
pesticides. Any higher risk of loss would be translated into higher
retail prices, i.e., they would require a higher return to accept the
higher risk.
E-14
-------
EXHIBIT 9
IMPACT OF INDEMNIFICATION ON THE NET
PRESENT VALUE OF A PESTICIDE
($ in thousands)
With Indemnification
Without Indemnification
Amount Gained by
Indemnification
Pro'bab-
ility
0.84
0.10
0.05
0.01
x
x
X
X
Amount
41,911 = 34,205
25,128 = 2,513
4,368 = 218
-3,817 = -36
NPV $36,898
Probab-
ility
0.84 x
0.10 x
0.05 x
0.01 x
Amount
41,911 =
22,723 =
2,500 =
-3,977 =
34,205
2,272
125
-40
$36,562
$336 (Amount Saved with Indemnification)
= 0.9% of original NPV
$36,562 (Original NPV without Indemnification) is indemnity
Assume: 10% discount factor
50% of production in inventory
10% probability of imminent hazard cancellation after 10 years
5% probability of imminent hazard cancellation after 5 years
1% probability of imminent hazard cancellation after 1 year
E-14a
-------
INDEMNIFICATION STUDY
Appendix E
C. Users
The users of pesticides invest relatively little in pesticides.
Farmers are major users and less than 102 of their annual variable
costs per acre are pesticide purchases. " However, if a pesticide of
choice is not available, the farmer will have to depend on an alter-
native, probably inferior pesticide until an adequate substitute is
available. Therefore, the losses to a farmer from a registration can-
cellation are predominantly dependent on the annual losses sustained
by using an inferior alternative and the number of years the alter-
native must be employed. One would expect that the amount of money
invested in the cancelled pesticide itself to be a small percentage of
the total losses. Indemnification against these losses would not sig-
nificantly alter a farmer's choice of pesticides. The absence of such
indemnification payments might Influence some farmers to hold less
pesticides in storage for future needs.
An illustrative example of the registration cancellation losses to
a farmer is provided by the case of Sllvex use on sugar cane in
Florida.^ For three years after the cancellation of Silvex, the
pesticide 2,4-D would have to be substituted on Florida sugar cane
resulting in average yield loss.
E-15
-------
EXHIBIT 10
CANCELLATION AND INDEMNIFICATION EFFECTS ON A USER:
Pesticide
Silvex
2,4,0
Amount
of Pes-
ticide
Used
(lb/
acre)
1
2
SILVEX VS
Pes-
ticide
Unit
Cost
($/lb)
5.50
2.13
. 2,4-D USE ON FLORIDA SUGAR CANE
Pes-
ticide
Cost
($/
acre)
5.50
4.26
Private
Appli-
cation
Cost
($)
2.00
2.00
Total
Pes-
ticide
Cost
Per
Acre
($)
7.50
6.26
Tons of
Cane
Yield
Per
- Acre
30.4
21.3
Value
of
Cane
($/ton)
20.06
20.06
Pre-Tax
Revenue
($/
acre)
609.82
427.28
Pretax Revenue per Acre:
Less difference in pesticide cost:
Difference in gross revenue per acre:
$609.82
-427.28
$182.54
-1.24
Silvex
2,4-D
$181.30
Assumptions:
For 3 years after cancellation, no substitute except 2,4-D.
After 3 years a perfect substitute.
Transition costs are zero.
Prices hold constant.
No shifting of cane growth from current average.
1979 dollars.
E-15a
-------
INDEMNIFICATION STUDY
Appendix E
FOOTNOTES
^•"Economic Analysis of Final Section 5 Notice Requirements,
Appendix F: Comparison of TSCA Premanufacture Notification
Procedures, the Pre-Market Procedures under FFDCA, and the
Registration Process under FIFRA," ICF Incorporated, July, 1981.
^"Analysis of TSCA Section 8(a) Small Manufacturers Exemption,"
ICF Incorporated, May 1981. *'
^"Preserving Innovation Under the Toxic Substances Control Act,"
Chemical Manufacturers Association, January 20, 1982, p. 26. ;
^Kline Guide to the Chemical Industry, Fourth Edition, Charles H. !
Kline and Co., Inc., Fairfleld, N.J., Susan Curry, Editor, 1980, p. ?
46. j
i
5Id. at 66. \
6Id. at 67. i
~~~~~ i
, }
'J.D. Riggleraan, Presidential Address at the 33rd Annual Meeting »
of Northeastern Weed Science Society, Boston, January 3, 1979. v
^Based on interviews with new product development personnel at i.
several major pesticide manufacturers. I
v
^ These discount rates are net of inflation. The effect of infla- .,'
tion has been eliminated from the analysis by using constant dollars. |
*y
leased on interviews with new product develoment personnel at |
several major pesticide manufacturers. (
\
**A pesticide formulation plant would probably be worth con* .-
siderably more than 40Z of original cost, because of the ease of tran- ]
sition from one formula to another. We do not specifically consider •
the case of the formulation plant in this example. }
12Teatimony on H.R. 4152 by Richard Wellman, Vice President and
General Manager, Union Carbide Company before the House Committee on
Agriculture, March 1971.
could have alternatively assumed the manufacturer would only
receive their variable costs of producing the product as inventory
payments. If this were the case, the manufacturer might be able to
sell the inventory to customers in countries that do not recognize EPA
cancellation actions.
r
Pesticide Cancellations/Suspensions; A Survey of Economic
Impact, Office of Pesticide Programs, March 1980.
E-16 |
-------
INDEMNIFICATION STUDY
Appendix E
on testimony before Congress and recent conversations with
personnel of two major pesticide producers.
^Economics of the Pesticide Industry, ICF Incorporated, for EPA
Office of Pesticide Programs, August 1980, page 3.
taken from The Biologic and Economic Assessment of Silvex.
A report of the USDA-States-EPA Silvex RPAR team, Draft Final Report,
March 2, 1981.
i *'
•u.s. GOVI;IINM,>T FIIINTINC OKFICM
E-17
------- |