Advisory of the Illegal Competitive Advantage (ICA) Economic Benefit (EB) Advisory Panel of the EPA Science Advisory Board (SAB)


         <    UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
         •                       WASHINGTON B.C. 20460
                                                                   OFFICE OF THE ADMINISTRATOR
                                                                    SCIENCE ADVISORY BOARD
                                   September 7, 2005

EPA-SAB-ADV-05-003

The Honorable Stephen L. Johnson
Administrator
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW
Washington, DC 20460

       Subject:       An Advisory of the Illegal Competitive Advantage (1C A) Economic
                    Benefit (EB) Advisory Panel of the EPA Science Advisory Board

Dear Administrator Johnson:

       The EPA Illegal Competitive Advantage (ICA) Economic Benefit (EB) Advisory Panel of
the Science Advisory Board has completed its review of Agency's Office of Enforcement and
Compliance Assurance (OECA) White Paper entitled "Identifying and Calculating Economic
Benefit That Goes Beyond Avoided and/or Delayed Costs, " dated May 25, 2003. The Panel
conducted its review in a public teleconference call on July 12 and a meeting August 5 & 6, 2004,
followed by three public conference calls on September 22, November 4, 2004 and January 19,
2005. The results of the Panel's efforts were administratively reviewed and approved by the
Board.

       The EPA has made the violator's economic benefit from violating the law the centerpiece
of its calculation of civil penalties.  The economic benefit from noncompliance consists of three
possible components: (a) the economic benefit from delayed costs associated with
noncompliance; (b) the economic benefit from avoided costs associated with noncompliance; and
(c) the economic benefit from an illegal competitive advantage generated by noncompliance.  The
Agency identifies four categories of cases in which the economic gain of noncompliance with an
environmental regulation will go beyond the benefit of delaying or avoiding compliance costs. It
refers to these as "Illegal Competitive Advantage" (ICA). The four categories of cases are:

       - violator gains additional market share;
       - violator sells products or services prohibited by law;
       - violator initiates construction or operation prior to government approval; and
       - violator operates at higher capacity than it should have.

The Agency has asked our advice regarding these categories and the proposed methods for
estimating economic benefit for each.

       The fundamental question for the determination of a penalty based on economic benefit is,
"How much did the profits of the firm increase (or losses decrease) as a result of its
noncompliance?"  Profits can be increased either by an increase in revenue or a decrease in the

-------
total cost of production (including abatement costs), or some combination of both. The Agency's
White Paper has essentially placed all of the factors influencing revenues in one of the four
categories under the heading of "benefit from illegal competitive advantage."

The Panel finds the Agency's use of the term "illegal competitive advantage" to be
unhelpful. It would be more transparent to have only two categories: (i) when economic
advantage is limited to delayed or avoided compliance costs; and (ii) when economic advantage
includes profits on increased sales. For all of those cases in which revenues increase, we
recommend that the Agency examine the facts of each case and use methods and data appropriate
to the case to estimate the changes in streams of revenue and/or production costs as well as
delayed or avoided compliance costs (if any). We suggest an approach to revising the White
Paper that is consistent with our recommendations.

The Panel also considered some broader issues relating to the determination of the
magnitude of penalties for noncompliance. We believe that one of these is of particular
importance to the Agency. This is the economic theory of optimal penalties. This theory makes
two points that are relevant to EPA's penalty policy. The first is based on the assumption that
potential offenders respond to both (a) the probability of detection and (b) the severity of
punishment, conditional upon being detected and punished. Thus, deterrence may be enhanced
by raising the penalty, by increasing monitoring activities to raise the likelihood that the offender
will be caught, or by changing legal rules to increase the probability of punishment. And second,
the economically optimal penalty balances the harm done by an offense against the cost of
deterring the offense in one or another of these ways. This balancing leads to the conclusion that
in those cases where the objective of regulation is to achieve economic efficiency, the appropriate
methodology for calculating a penalty is to charge an amount per offense equal to the (monetized)
harm done divided by the probability of punishment.

The Panel believes that the state-of-the-art in benefits estimation has progressed to the
point where EPA should seriously explore how it might incorporate "harm-based" measures into
its penalty formula, at least for some types of environmental harm. We also recommend that the
Agency explore ways to incorporate more explicitly the probability of detection and punishment
into its penalty policy as a way of attaining the full intended deterrent effects of its penalties.

Finally, if our recommendations regarding the penalty policy and the revisions to the
White Paper are accepted, it will be necessary for the Agency to provide economic input into
these processes. The necessary economic expertise could come either from the National Center
for Environmental Economics or by adding an economist to the staff of the OECA.

-------
       We are pleased to have participated in this process and are particularly interested in your
response to the points we raise in this advisory.

                                 Sincerely,


       /signed/                                 /signed/


Dr. M. Granger Morgan                   Dr. A. Myrick Freeman III
Chair                                   Chair
EPA Science Advisory Board              Illegal Competitive Advantage (ICA) Economic
                                        Benefit (EB) Advisory Panel
                                        EPA Science Advisory Board

-------
                                       NOTICE
       This report has been written as part of the activities of the EPA Science Advisory Board, a
public advisory committee providing extramural scientific information and advice to the
Administrator and other officials of the Environmental Protection Agency. The Board is
structured to provide balanced, expert assessment of scientific matters related to problems facing
the Agency. This report has not been reviewed for approval by the Agency and, hence, the
contents of this report do not necessarily represent the views and policies of the Environmental
Protection Agency, nor of other agencies in the Executive Branch of the Federal government, nor
does mention of trade names or commercial products constitute a recommendation for use.
Reports of the EPA Science Advisory Board are posted on the EPA website at
http ://www. epa. gov/sab.

-------
                                    ABSTRACT

       The U.S. Environmental Protection Agency's Illegal Competitive Advantage (ICA)
Economic Benefit (EB) Advisory Panel ("the Panel") provided advice on four charge questions
relating to an Agency White Paper entitled "Identifying and Calculating Economic Benefit That
Goes Beyond Avoided and/or Delayed Costs, " dated May 25, 2003.

       The EPA has made the recovery of a violator's economic benefit from violating the law
the basis of its calculation of civil penalties.  The Agency has asked the Panel for advice in
estimating economic benefits when a firm's noncompliance enables it to increase sales (which
EPA terms "illegal competitive advantage" or ICA), as opposed to simply avoiding or delaying
compliance costs.  The Panel suggests  that the four categories of cases identified by EPA as
falling under the term ICA and described in the White Paper are not helpful for several reasons.

       The Panel suggests that in all those cases in which revenues increase, the Agency should
examine the facts of each case and use methods and data appropriate to the case to estimate the
changes in streams of revenue and production costs, as well as any delayed or avoided
compliance costs.

       After a review of the economic theory of optimal penalties, the Panel  recommends that the
Agency explore ways to explicitly incorporate the probability of detection and punishment into its
penalty policy. The Panel also believes that the state-of-the-art in benefits estimation has
progressed to the point where EPA should seriously explore how it might incorporate "harm-
based" measures into its penalty formula, at least for some types of harm, and where conditional
deterrence is the objective.

Key Words: Compliance, Economic  Benefit, Economic Gain, Enforcement, Harm-Based
Measures, Illegal Competitive Advantage, Optimal Penalties

-------
                     U.S. Environmental Protection Agency
                             Science Advisory Board
Illegal Competitive Advantage (ICA) Economic Benefit (EB) Advisory Panel

CHAIR
Dr. A. Myrick Freeman III, Research  Professor, Department of Economics, Bowdoin College,
Brunswick, ME (Member, SAB Board)

MEMBERS
Dr. Dallas Burtraw, Senior Fellow, Resources for the Future, Washington, DC (Member, SAB/
EEAC)

Dr. Mark Cohen, Senior Associate Dean & Justin Potter Professor of American Competitive
Business, Owen Graduate School of Management, Vanderbilt University, Nashville, TN

Dr. Jane V. Hall, Professor, Department of Economics, California State University, Fullerton,
CA

Dr. W. Michael Hanemann, Professor, Department of Agricultural & Resource Economics &
Public Policy, University of California, Berkeley, CA (Member SAB/EEAC)

Dr. Catherine L. Kling, Professor, Department of Economics, Iowa State University, Ames, IA
(Member SAB/EEAC; SAB Board)

Dr. Arik Levinson, Associate Professor, Economics Department, Georgetown University,
Washington, DC (Member SAB/EEAC)

Dr. Clifford S. Russell, Professor of Economics Emeritus (Retired Former Director of the
Vanderbilt Institute for Public Policy Studies), Vanderbilt University, now residing in Alna, ME

Dr. Michael A. Salinger1, Professor of Economics, Finance & Economics Department, School of
Management,  Boston University, Boston, MA

Dr. David Sunding, Professor, Agricultural and Resource Economics, College of Natural
Resources and Boalt Hall School of Law, University of California, Berkeley, CA

SCIENCE ADVISORY BOARD STAFF
Dr. K. Jack Kooyoomjian, Designated Federal Officer, US EPA Science Advisory Board,
Washington, DC
1     Dr. Salinger currently is Director of the Bureau of Economics with the U. S. Federal Trade Commission in
      Washington, DC, and on leave as Professor of Economics at Boston University.  His position while serving
      on this Panel was as Professor of Economics at Boston University.

                                         iii

-------
                     U.S. Environmental Protection Agency
                             Science Advisory Board
                                     BOARD

CHAIR
Dr. M. Granger Morgan, Carnegie Mellon University, Pittsburgh, PA


VICE CHAIR
Dr. Domenico Grasso, The University of Vermont, Burlington , VT


SAB MEMBERS
Dr. Gregory Biddinger, ExxonMobil Biomedical Sciences, Inc, Houston, TX

Dr. James Bus, The Dow Chemical Company, Mildland, MI

Dr. Trudy Ann Cameron, University of Oregon, Eugene, OR
      Also Member: COUNCIL

Dr. Deborah Cory-Slechta, University of Medicine and Dentistry of New Jersey and Rutgers
State University, Piscataway, NJ

Dr. Maureen L. Cropper, University of Maryland, College Park, MD

Dr. Virginia Dale, Oak Ridge National Laboratory, Oak Ridge, TN

Dr. Kenneth Dickson, University of North Texas, Denton, TX

Dr. Baruch Fischhoff, Carnegie Mellon University, Pittsburgh, PA

Dr. A. Myrick Freeman, Bowdoin College, Brunswick, ME

Dr. James Galloway, University of Virginia, Charlottesville, VA

Dr. William H. Glaze, Oregon Health & Science University, Beaverton, OR

Dr. Lawrence Goulder,  Stanford University, Stanford, CA

Dr. Linda Greer, Natural Resources Defense Council, Washington, DC

Dr. Rogene Henderson,  Lovelace Respiratory Research Institute, Albuquerque, NM
      Also Member: CASAC

Dr. Philip Hopke, Clarkson University,  Potsdam, NY
      Also Member: CASAC

Dr. James H. Johnson, Howard University, Washington, DC

Dr. Meryl Karol, University of Pittsburgh, Pittsburgh, PA


                                        iv

-------
Dr. Roger E. Kasperson, Clark University, Worcester, MA

Dr. Catherine Kling, Iowa State University, Ames, IA

Dr. George Lambert, Robert Wood Johnson Medical School/ University of Medicine and
Dentistry of New Jersey, Piscataway, NJ

Dr. Jill Lipoti, New Jersey Department of Environmental Protection, Trenton, NJ

Dr. Genevieve Matanoski, Johns Hopkins University, Baltimore, MD

Dr. Michael J. McFarland, Utah State University, River Heights, UT

Dr. Rebecca Parkin,  The George Washington University, Washington, DC

Mr. David Rejeski, Woodrow Wilson International Center for Scholars, Washington, DC

Dr. Joan B. Rose, Michigan State University, E. Lansing, MI

Dr. Kristin Shrader-Frechette, University of Notre Dame, Notre Dame, IN

Dr. Robert Stavins, Harvard University, Cambridge, MA

Dr. Deborah Swackhamer, University of Minnesota,, Minneapolis, MN

Dr. Thomas L. Theis, University of Illinois at Chicago, Chicago, IL

Dr. R. Rhodes Trussell, Trussell Technologies, Inc., Pasadena, CA

Dr. Robert Twiss, University of California-Berkeley, Ross, C A

Dr. Terry F. Young, Environmental Defense, Oakland, CA

Dr. Lauren Zeise, California Environmental Protection Agency, Oakland, CA


SCIENCE ADVISORY BOARD STAFF
Mr. Thomas Miller, Designated Federal Officer, US EPA Science Advisory Board, Washington,
DC

-------
TABLE OF CONTENTS
1. EXECUTIVE SUMMARY 1
1.1. Current Civil Penalty Policy at the Agency 1
1.2. The Panel's Responses 2
1.3. Ex Ante vs. Ex Post Assessments 4
1.4. Toward an Optimal Penalty Policy 4

2. INTRODUCTION 6
2.1. Request for EPA Science Advisory Board (SAB) Review 6

3. CURRENT AGENCY PRACTICE AND QUESTIONS FOR THE PANEL 7
3.1. Statutory Provisions and the EPA Penalty Policy - Recapture Economic Gain ... 7
3.2. The Objectives of Penalties 8
3.3. Delayed and Avoided Compliance Costs and the BEN Model 10
3.4. The Four Categories of Illegal Competitive Advantage 11
3.5. The Charge Questions for The Panel 11

4. THE PANEL'S RESPONSES 13
4.1. The Economic Benefit is the Increase in Profits 13
4.2. Economic Benefit When Revenues Change Due to Noncompliance 13
4.3. The Four Categories of Illegal Competitive Advantage 17
4.4. Direct Responses to Charge Questions 19
4.5. Revising the White Paper 20

5. ADDITIONAL ISSUES 22
5.1. The Effect of Market Structure 22
5.2. Dynamic Effects 22
5.3. Ex Ante vs. Ex Post Assessments 23

6. TOWARD AN OPTIMAL PENALTY POLICY 25
6.1. Economic Theory of Optimal Penalties 25
6.2. Quantifying Harm 26
6.3. Probability of Detection and Punishment 27
6.4. Implications for Current EPA Policy 28

FIGURE 1 - Benefits from Noncompliance - Competition 14

FIGURE 2 - Benefits from Noncompliance - Monopoly 15

FIGURE 3 - Monopoly with Zero Output under Compliance 16

FIGURE 4 - Regulated Natural Monopoly 18
VI

-------
APPENDIX A - A MORE DETAILED DESCRIPTION OF THE SAB PROCESS
AND PANEL REVIEW PROCEDURES                                   A - 1
     A. 1.  Request for Review and Acceptance 	 A - 1
     A.2.  Panel Formation 	 A - 1
     A.3.  Panel Process and Review Documents	 A - 2

APPENDIX B - BRIEF BIOSKETCHES OF THE ILLEGAL COMPETITIVE
ADVANTAGE (ICA) ECONOMIC BENEFIT ( EB) ADVISORY PANEL          B - 1

APPENDIX - C ACRONYMS                                           C - 1

REFERENCES	 R - 1
                                  vn

-------
1. EXECUTIVE SUMMARY
The Illegal Competitive Advantage (1CA) Economic Benefit (EB) Advisory Panel of the
EPA Science Advisory Board (SAB) reviewed and evaluated a White Paper entitled "Identifying
and Calculating Economic Benefit That Goes Beyond Avoided and/or Delayed Costs, " dated May
25, 2003, as well as supplemental materials, along with a charge for the Panel. The Panel held a
conference call on July 12, 2004, met in Washington, DC, on August 5-6, 2004, and conducted
follow-up conference calls on September 22, November 4, 2004, and January 19, 2005 to
conclude its activity.

1.1 Current Civil Penalty Policy at the Agency

Since 1978, the EPA has made the violator's economic benefit from the violating the
Clean Air and Clean Water Acts the centerpiece of its calculation of civil penalties. In the
Agency's view, the economic benefit from noncompliance consists of three possible components:
(a) the economic benefit from delayed costs associated with noncompliance; (b) the economic
benefit from avoided costs associated with noncompliance; and (c) the economic benefit from an
illegal competitive advantage generated by noncompliance. The EPA's request to the SAB deals
with one aspect of just one of these three components of a penalty, the assessment of illegal
competitive advantage in the calculation of economic benefit.

The deterrence objective is clearly recognized in the EPA's penalty process. But one
consideration that plays a substantial role in the economic theory of deterrence appears to be
entirely missing from the current penalty assessment process; this is the probability of detection
and punishment associated with the violation in question.

An important aspect of fairness is the restoration of the status quo: the law has been
violated, and one objective of the penalty system is to return to the status quo before the violation
occurred. Requiring the polluter to surrender the profit he gained by not complying with the law
is one important aspect of restoration of the status quo. However, removing the economic benefit
is not the only action that might be required in order to restore the status quo. With a violation of
an environmental regulation, there is a loss resulting from the polluter's action in the form of
some harm to the natural environment. Whether the natural resource that is harmed belongs to a
private individual or the general public, restoration of the status quo can call for some appropriate
compensatory action, perhaps in the form of a penalty based on harm to the environment rather
than on gain to the polluter.

These two points raise issues that lie outside of the charge to the Panel. Nevertheless the
Panel believes that they deserve consideration in the continuing evolution of the Agency's civil
penalty policy. Further discussion is deferred to the concluding section of this report.

Regarding the calculation of economic gain, the Agency developed the BEN model to
estimate the economic benefits that result from cost-savings during the time that a facility is not in
compliance. Because BEN is presently limited to calculating the difference in discounted cash
flows that result from cost-savings during noncompliance, it is not now configured to support

-------
recapture of benefits that could result from higher revenues. There is, however, no inherent
reason that BEN could not be modified so that it could be used to estimate the benefits of higher
revenues.

In its White Paper the Agency identifies four categories of cases in which the economic
gain of noncompliance with an environmental regulation will go beyond the benefit of delaying or
avoiding compliance costs. It refers to these as instances of "Illegal Competitive Advantage"
(ICA). The four categories of cases are:

- violator gains additional market share;
- violator sells products or services prohibited by law;
- violator initiates construction or operation prior to government approval; and
- violator operates at higher capacity than it should have.

The Agency has asked our advice regarding these categories and the proposed methods for
estimating economic benefit for each.

1.2. The Panel's Responses

The fundamental question for the determination of the economic benefit component of the
penalty is how much the profits of the firm increased (or losses decreased) as a result of its
noncompliance. Profits can be increased either by an increase in revenue or a decrease in the
total cost of production (including abatement costs), or some combination of both. The BEN
model provides a reliable measure of the change in after-tax profit only if no other change would
have occurred that would have affected the firm's profit. There are several factors other than cost
that might influence the amount by which the violator's profit was increased by the violation.
The Agency's White Paper has essentially placed all of these factors in one of four categories
under the heading of "benefit from illegal competitive advantage."

For several reasons, the Panel finds that the Agency's use of the term "illegal competitive
advantage" and its identification of the four categories of ICA cases is unhelpful.

1. It is not clear what the modifier "competitive" is intended to convey.

2. Increases in market share will often be difficult to identify in terms of comparing the
noncompliance scenario with the counterfactual compliance scenario; and observed increases in
market share might be difficult to attribute exclusively to the noncompliance.

3. In any case, increases in market share are not inherently valuable to the firm; what
matters is the impact of changes in market share on profits.

4. The other categories of ICA appear to stem from unusual circumstances that are very
context dependent.

It would be more transparent to have only two categories: (i) when firms that experienced
no revenue increase so that profits were increased only by the amount of the delayed or avoided
compliance costs; and (ii) when firms gained profits from increased sales. The BEN model, as
currently configured, may be used as a calculation tool in the first category of cases. For the
other category, we recommend that the Agency examine the facts of each case and use methods
and data appropriate to the case to estimate the changes in streams of revenue and/or production

-------
costs as well as delayed or avoided compliance costs (if any). BEN may be reconfigured to assist
in the calculation once the facts are known and relevant shifts in supply curves have been
estimated.

When non-compliant firms sell more than they would have had they complied, their
economic benefit includes the profits they earn on the increased sales. We use a simple economic
model to identify the economic gain due to noncompliance. We show that when a firm is able to
increase sales, using avoided costs at the actual quantity produced overstates the true economic
benefits of noncompliance.

There are two situations in which a calculation of economic benefit based only on
avoided/delayed costs could still be justified. The first is if it can be assumed that the effect on
marginal cost and therefore output is sufficiently small that the error induced by ignoring output
effects is also small. The second is if compliance would affect fixed costs only. In that case,
compliance would leave marginal cost and, accordingly, output unchanged.

Before answering the charge questions, we consider each of the four categories of 1C A
described in the White Paper in more detail and offer comments on the appropriate methods for
estimating economic benefit for each.

Our answers to the four charge questions are as follows:

1. Are there categories of cases that would be useful for the Agency to consider in
calculating the ICA economic benefit, other than those that are identified in the White
Paper? Should any of these be combined?

We do not think that the categories offered in the White Paper are particularly useful. In
fact we believe that they should be combined into only one category - cases where profits increase
at least in part due to increases in revenue.

2. How can the Agency more accurately characterize the types of cases that are
described in the White Paper? Have any of the examples and counter-examples in the
White Paper been misidentified with regard to whether they are amenable to the BEN
model's simplifying paradigm?

As indicated above, we do not think that the categorization of cases in the White Paper is
useful. However, the White Paper is correct in its statements about whether specific cases can be
analyzed within the BEN framework as that model is currently configured.

3. Are there any suggestions for modifying the described analytical approach to
calculate the economic benefits and;

We believe that there is no substitute for a careful examination of the facts of each case
and the use of methods and data appropriate to each case to estimate the changes in streams of
revenue and/or production costs as well as delayed or avoided compliance costs (if any).

4 The Agency's proposed approach strives to avoid double-counting of the benefit
by laying out all relevant cash flows stemming from the violations, as opposed to simply
adding on the additional calculations to a BEN run. What additional measures (if any)
should the Agency put in place to avoid such potential double-counting?

-------
Every effort should be made to calculate economic advantage as avoided/delayed costs
(and therefore not to decompose the gain into separate components.) One should only resort to a
full-blown change-in-profit analysis when avoided/delayed costs leads to a clearly substantial
overestimate or underestimate of the economic benefit. If it is necessary to do change-in-profit
analysis, it is important that the estimate of costs under compliance reflect the lower level of
output the firm would have produced rather than the actual production of the polluter.

We recognize that if the recommendations that we make are accepted, it will be necessary
for the EPA to revise significantly the White Paper. We offer several suggestions about how this
can be done.

In order for the Office of Enforcement and Compliance Assurance (OECA) to implement
our recommendations, it will need to have access to the relevant expertise in economics. One
possible source of this expertise in the Agency is the National Center for Environmental
Economics (NCEE). But it might be more useful to OECA to have its own in-house economist.
This would be especially true if the agency accepts our recommendations in Section 6.4 for
rethinking the civil penalty policy.

1.3. Ex Ante vs. Ex Post Assessments

To the extent that a violator should pay a penalty based on its expected rather than its
realized economic benefit, there remains the practical issue of how that benefit is to be
determined. Without knowing exactly what information is available, it is hard to describe how to
perform an expected benefit calculation that would withstand judicial scrutiny. However, the
Panel believes that cases might arise in which the Agency should consider putting forward an
expected benefit calculation as an alternative measure of harm.

1.4. Toward an Optimal Penalty Policy

The economic theory of optimal penalties approaches the issue of deterrence from the
perspective of economic efficiency rather than that of fairness. This theory makes two points that
are relevant to EPA's penalty policy. The first is based on the assumption that potential offenders
respond to both (a) the probability of detection and (b) the severity of punishment, conditional
upon being detected and punished. Thus, deterrence may be enhanced by raising the penalty, by
increasing monitoring activities to raise the likelihood that the offender will be caught, or by
changing legal rules to increase the probability of punishment. And second, the economically
efficient penalty balances the harm done by an offense against the cost of deterring the offense
This balancing leads to the conclusion that the appropriate methodology for calculating a penalty
is to charge an amount per offense equal to the (monetized) harm done, divided by the probability
of punishment. We are aware that many of the statutes governing EPA appear not to make
economic efficiency the goal but rather imply a goal of absolute deterrence of polluting activities.

-------
For absolute deterrence, the penalty should be based on gain to the violator rather than harm to
the victim.

If an environmental violation results in emissions levels that are beyond a legal standard,
there may be some harm to natural resources or human health. Measuring people's value for non-
market items in monetary terms (e.g., measuring what they would be willing to pay to prevent a
specific harm to the natural environment) is inherently difficult, and in practice different
measurement techniques can produce different results. We also recognize that while some of the
methods used to value environmental harm can be employed with relatively little cost, others
require significant resources. Thus, in many cases, these methods may not be practical unless the
harm, and thus the expected penalty, is extremely large. Nevertheless, the Panel believes that the
state-of-the-art in benefits estimation has progressed to the point where EPA should seriously
explore how it might incorporate "harm-based" measures into its penalty formula, at least for
some types of environmental harm.

The probability of detection is likely to vary considerably by type of violation and even
across jurisdictions. An extremely harmful environmental violation is likely to have a probability
of detection and punishment of nearly one. If so, the optimal penalty for such a violation is likely
to be very close to the monetary equivalent of the harm. However, as the size of the harm
decreases, all else equal, we expect that the likelihood of detection also decreases. Other factors
that might influence the probability of detection and punishment are: (a) whether a violator is
subject to mandatory reporting that is available for the public to use in filing citizen lawsuits, (b)
the ratio of facilities to inspectors in an EPA region, (c) the strength of environmental activism in
a region/state, and (d) whether the violator has a history of violations and thus has been subject to
increased scrutiny or targeted enforcement.

Although not widely employed in the environmental literature to date, numerous
techniques are available to estimate the probability of detection and punishment. One widely
used method is the "time until capture" approach which is most appropriate for ongoing violations
that occur over a period of time. Another method - the "capture/recapture" approach is an
adaptation of methods used to estimate the number of animals in a given geographic area.

The current EPA penalty policy starts with the calculation of "gain" - i.e., estimating the
amount that the offender saved by not complying with environmental regulations, and then adds a
"gravity" component based in part on the harm from the offense. However, the policy does not
provide for quantifying the "harm" in monetary terms and also ignores any explicit consideration
of the probability of detection. In those cases where the goal of regulation is to achieve economic
efficiency, the penalty should be based on an estimate of the harm rather than the gain due to
noncompliance. If harm cannot be quantified, the base might either be "gain" or a "default" fine
level that is specified by type of offense. This base fine would then be divided by a factor that is
based on an estimate of the probability of detection. It should be emphasized that what is sought
here is an approximate estimate of the general probability of detection, not a highly elaborate
calculation tailored to all the specific details of the particular violation.

-------
                               2. INTRODUCTION
       2.1    Request for EPA Science Advisory Board (SAB) Review

       At the request of the EPA Office of Enforcement and Compliance Assurance (OECA), the
EPA Science Advisory Board convened a Panel to review and evaluate a White Paper entitled
"Identifying and Calculating Economic Benefit That Goes Beyond Avoided and/or Delayed
Costs, " dated May 25, 2003, as well as supplemental materials, along with a charge for the Panel
(U.S. EPA, 2003). The White Paper identifies four categories of cases in which the economic
gain of noncompliance with an environmental regulation will go beyond the benefit of delaying or
avoiding compliance costs, provides examples and counterexamples of each, and briefly describes
how the economic gain can be calculated.  The four categories of cases are:

       - violator gains additional market share;
       - violator sells products or services prohibited by law;
       - violator initiates construction or operation prior to government approval; and
       - violator operates at higher capacity than it should have.

       The proposed charge to the ICA EB Advisory Panel of the SAB was developed based on
discussions between the OECA and SAB Staff offices.  The specific charge questions are
presented in Section 3.5 below.

-------
3. CURRENT AGENCY PRACTICE AND QUESTIONS FOR THE PANEL


       3.1  Statutory Provisions and the EPA Penalty Policy - Recapture Economic Gain

       The U.S. Environmental Protection Agency (EPA) exercises primary enforcement
responsibility for many of the federal environmental protection laws, including the Clean Air Act
(CAA); the Clean Water Act (CWA); the Oil Pollution Act (OPA); the Safe Drinking Water Act
(SDWA); the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA); the Toxic Substances
Control Act (TSCA); the Resource Conservation and Recovery Act (RCRA); the Comprehensive
Environmental Response Compensation and Liability Act (CERCLA); and the Emergency
Planning and Community Right-to-Know Act (EPCRA). While each of the statutes is different in
its particulars, they generally provide for the assessment of civil penalties in the event of
noncompliance, and  they offer some guidance as to the factors that should be considered when
assessing a civil penalty.  For example, Section 7413(e)(l) of the CAA states:

       In determining the amount of any penalty to be assessed under this section or section
       7604(a) of this title, the Administrator or the court, as appropriate, shall take into
       consideration (in addition to such other factors as justice may require) the size of the
       business, the  economic impact of the penalty on the business, the violator's full
       compliance history and good faith efforts to comply, the duration of the violation as
       established by any credible evidence (including evidence other than the applicable test
       method), payment by the violator of penalties previously assessed for the same violation,
       the economic benefit of 'noncompliance, and the seriousness of the violation [emphasis
       added].

Section 7524(b) of the CAA, dealing with mobile sources, states:

       In determining the amount of any civil penalty to be assessed under this subsection, the
       court shall take into account the gravity of the violation, the economic benefit or savings
       (if any) resulting from the violation, the size of the violator's business, the violator's
       history of compliance with this title, action taken to remedy the violation, the effect of the
       penalty on the violator's ability to continue in business, and such other matters as justice
       may require [emphasis added].


       Since 1978, the EPA has based civil penalties under the CAA and CWA on the violator's
economic benefit from violating the law (U.S. EPA "Civil Penalty Policy" 1978). The monetary
estimate of the economic benefit from noncompliance becomes the starting point for establishing
a penalty, and this is then adjusted up or down based on a qualitative assessment of other
considerations such as the factors listed above. This approach was further formalized in February
1984 when the EPA  issued the Policy on Civil Penalties, EPA Enforcement Policy #GM-21 and
the accompanying Framework for Statute-Specific Approaches to Penalty Assessments, EPA
General Enforcement Policy #GM-22. As explained in the latter document: "The development of
a penalty figure is a two-step process. First the case development team must calculate a
preliminary deterrence figure.  This figure is composed of the economic benefit component
(where applicable) and the gravity component. The  second step is to adjust the preliminary
deterrence figure through a number of factors (U. S.  EPA, 1984b, p. 2)."

       According to the 1984 Guidelines, the economic benefit from noncompliance consists of
three possible components:  (a) the economic benefit from delayed costs associated with

                                           7

-------
noncompliance; (b) the economic benefit from avoided costs associated with noncompliance; and
(c) the economic benefit from an illegal competitive advantage generated by noncompliance.

Following the assessment of the economic benefit, the EPA then performs an assessment
of the gravity component. This involves ranking different types of violations according to the
seriousness of the act, considering (i) actual or possible harm, (ii) importance to the regulatory
scheme, and (iii) availability of data from other sources. In evaluating the actual or possible
harm, consideration should be given to (a) the amount of pollutant, (b) toxicity of pollutant, (c)
sensitivity of the environment, (d) length of time of a violation, and (e) size of the violator.
Having ranked the violations, according to the 1984 Guidelines one "then should assign
appropriate dollar amounts or ranges of amounts to the different ranked violations to constitute
the 'gravity component.' This amount, added to the amount reflecting benefit, constitutes the
preliminary deterrence figure (U. S. EPA, 1984b, p. 3)."

In the second step, the preliminary deterrence amount is adjusted "to ensure that penalties
also further Agency goals besides deterrence (i.e., equity and swift correction of environmental
problems). ... Adjustments (increases or decreases, as appropriate) that can be made to the
preliminary deterrence penalty to develop an initial penalty target to use at the outset of
negotiation include:
a. degree of willfulness and/or negligence;
b. cooperation/noncooperation through pre-settlement action;
c. history of noncompliance;
d. ability to pay; and
e. other unique factors (including strength of case, competing public
policy considerations) (U. S. EPA, 1984b, pp. 3-4)."

In summary, the dollar amount which the EPA calculates as its initial penalty target is
derived by calculating the economic benefit, adding a monetary amount which reflects the gravity
component, and adjusting the resulting total up or down based on the considerations listed
immediately above.

The EPA's request to the SAB deals with one aspect of just one of these three stages in the
development of a penalty target— the assessment of illegal competitive advantage in the
calculation of economic benefit. Nevertheless, before we address this question, it is useful to
situate the penalty procedure in the broader context of the economic and public policy
considerations that bear on the determination of a penalty for noncompliance with environmental
regulations.

3.2 The Objectives of Penalties

The EPA Policy on Civil Penalties establishes "a single set of goals for penalty
assessment in EPA administrative and judicial enforcement actions." These goals are
characterized as "deterrence, fair and equitable treatment of the regulated community, and swift
resolution of environmental problems (U. S. EPA, 1984a, p. 1)." In the context of our present
analysis, we see the last item as being more a constraint than an objective: whatever the formula
for assessing a civil penalty, it needs to be practical and amenable to implementation in a
reasonably timely manner. Accordingly we focus on the other two items -fairness and
deterrence - as primary objectives in the determination of a civil penalty; they are clearly evident
in the statutory provisions quoted above.

-------
       Deterrence and, especially, fairness have multiple possible interpretations depending on
both the philosophical position one adopts and how one interprets the violation of an
environmental law from a public policy perspective. In this section we note some issues that arise
in conceptualizing the objectives of fairness and deterrence.

       An important aspect of fairness is what might be called the restoration of the status quo:
the law has been violated and the restorative objective of a penalty system is to undo the violation
and return the situation to how it was before the violation occurred.  This is clearly the major
focus of the EPA's civil penalty policy since 1978. The assumption underlying this  policy is that
the noncompliance with environmental regulations was associated with, and perhaps motivated
by, some increase in profit to the responsible party (from now on, we will use "the polluter" as a
shorthand term to refer to this party). Whether or not the assumption is correct is obviously an
empirical question that depends on the particular circumstances of the  case; but, for now, we will
assume it is correct.  In that case, a key element of the restoration of the status quo is to compel
the polluter to surrender the profit he gained by not complying with the law.  This is essentially
what the EPA penalty policy focuses on by virtue of the prominent position it accords to the
calculation of economic benefit.

       It should be noted, however, that removing the economic  benefit is not the only action that
might be required in order to restore the status quo. This is because the failure to comply with a
federal regulation may entail not only an unwarranted  gain to the violator but also an unwarranted
loss to some other party. In the  case of violation  of an economic  regulation, for example, a
violation of anti-trust law may generate not only an unlawful gain to the seller but also an
unwarranted loss to the customers who purchase from this seller.  In that case, the restoration of
the status quo requires not only that the seller surrender his unlawful gain but also that the
customers be compensated for their unwarranted  loss.  With a violation of an environmental
regulation, while there may not be an unwarranted monetary loss to a third party, there is a non-
monetary loss resulting from the polluter's action in the form of some harm to the natural
environment, at least if the violation involves  releases  to the environment. Whether the natural
resource that is harmed belongs to a private individual or the general public, a loss has occurred,
and restoration of the status quo calls for some appropriate compensatory action.  Depending on
the circumstances, this action could include both  clean-up and some form of environmental
restoration.2  The costs of clean-up and environmental restoration are thus compensation that
should be paid by the polluter in order to restore the status quo.

      The popular name for what is being discussed here is "the polluter pays principle." Not
only is this called for by notions of fairness, but also it is supported by considerations of
economic efficiency. Ever since Pigou (1918), it has been recognized  that, in the presence of a
harmful externality such as that caused by pollution, a competitive market is generally unlikely to
lead to a socially optimal allocation of resources unless the polluter is required to bear the cost
that his pollution imposes on others.

      In summary, the restoration of the status quo would appear to be an important aspect of the
fairness objective in setting the penalty for a violation of an environmental regulation. This
       With respect to the latter, although the context is different, it strikes us as relevant to quote the language used
       by the Department of Interior (DOI) in its proposed regulations for natural resource damages under the
       Comprehensive Environmental Response Compensation and Liability Act (CERCLA). DOI describes the
       measure of damages as: "the cost of restoration, rehabilitation, replacement, and/or acquisition of the
       equivalent of the injured natural resources and the services those resources provide, plus the compensable
       value of the services lost to the public for the time period from the discharge or release until the attainment
       of the restoration, rehabilitation, replacement and/or acquisition of equivalent of the resources and their
       services to the baseline (italics added)." (56 Fed. Reg. at 19,769 (proposed 43  C.F.R. § 11.80(b)).
                                              9

-------
restorative goal can be seen to have two possible implications. If one focuses on the polluter's
unlawful gain, restoration of the status quo implies that he should give up this gain. If one
focuses on the unlawful harm to the environment, restoration of the status quo implies that he
should pay an amount covering the cost of cleanup and/or environmental restoration. In general,
there is no reason to expect that the two different approaches will lead to a similar assessment of a
monetary payment: the cost avoided by failing to control pollution need bear no relationship to
the damage caused by the pollution. This raises two questions: Which approach is presently
adopted by the EPA? Which approach seems preferable, or should they be combined in some
manner?

With regard to the first question, it must be recognized that the current EPA penalty policy
does contain some elements of both approaches, but they are combined in a manner that is
equivocal and perhaps somewhat muddled. The first step in the penalty assessment process, the
calculation of economic benefit, focuses on the unlawful gain to the polluter. The second step,
the assessment of the gravity component, contains elements that clearly relate to the unwarranted
loss to the environment, specifically item (i), the actual or possible harm. But, the
characterization of this item is somewhat confusing because, while it includes factors that relate
directly to the magnitude of the environmental damage — the amount of pollutant, the toxicity of
pollutant, the sensitivity of the environment, and the length of time of a violation - it also
includes a factor (the size of the violator) that has nothing to do with the amount of environmental
damage.3 We see the size of the violator as being relevant to the deterrence objective of a penalty
rather than the restoration of the status quo.

In short, the current EPA penalty process appears to focus overwhelmingly on the
calculation of the unlawful gain to the polluter, with no systematic consideration of the monetary
value of the environmental damage caused by the violation of the pollution control regulation.
We return to this issue in Section 6, below.

The deterrence objective is certainly recognized in the EPA's penalty process. In addition
to the considerations in the gravity component stage, noted above, the third stage of the process,
the adjustment stage, is heavily weighted toward factors that bear on deterrence, including the
degree of willfulness and/or negligence, the extent of cooperation through pre-settlement action,
the history of noncompliance, and the polluter's ability to pay. But one consideration that plays a
substantial role in the economic theory of deterrence appears to be entirely missing from the
current penalty assessment process; this is the probability of detection and punishment associated
with the violation in question. Economic theory indicates that, to obtain a given degree of
deterrence, the penalty should vary inversely with the probability of detection: given two possible
violations with the same economic benefit to the polluter but where one is much less likely to be
detected than the other, the first requires a larger penalty in order to provide the same degree of
deterrence. We also return to this question in Section 6, below.

3.3. Delayed and Avoided Compliance Costs and the BEN Model

Since 1978, a key EPA objective in assessing civil penalties has been to deter violators.
The "cornerstone" of achieving this goal is to recapture the economic benefit that accrues from
noncompliance. The BEN model, first issued in late 1984, was developed to calculate the
economic benefits that result from cost-savings during the time that a facility is not in
compliance. It can estimate savings from deferred capital investments in control equipment,
Size of firm may also enter into calculation of the gravity component if the firm is a small business and an
analogous penalty might jeopardize the survival of the firm.
10

-------
deferred one-time expenditures (such as establishing accounting/tracking systems), and reduced
recurring costs of maintaining and operating control systems.

The model is simple to run, requiring the user to provide a minimal amount of information
to estimate cost-savings. Standard values, for things such as tax rates, the cost of capital, and
equipment life are embedded in the model itself (although they can be modified by the user), and
are determined by the user's response to a set of "screening questions." Since the BEN model
became a central tool in the penalty assessment process, aggregate annual penalty assessments
have risen dramatically. It is not possible to entirely untangle the impact of BEN from the impact
of changes in EPA enforcement policies, but it seems apparent that BEN has been a factor in this
increase.

BEN is presently limited to calculating the difference in discounted cash flows that result
from cost-savings during noncompliance. Thus, it is not now configured to support recapture of
benefits that could result from higher revenues. Viewed as a calculator, however, there is no
inherent reason that BEN could not be used to estimate the benefits of higher revenues. This
would require construction of specific questions for the user, parallel to the present questions that
prompt the user to enter relevant information regarding differences in costs that result from
noncompliance. We suggest such questions in Section 4.5 below.

In cases where greater revenues might be a significant incentive to be non-compliant,
adding questions that would support estimation of differences in discounted net cash flows would
be useful and, in fact, critical to deterrence.

3.4 The Four Categories of Illegal Competitive Advantage

The White Paper identifies four categories of cases in which the economic gain of
noncompliance with an environmental regulation is said to go beyond the benefit of delaying or
avoiding compliance costs. It refers to these as "Illegal Competitive Advantage" (ICA). It also
provides examples and counterexamples of each category and briefly describes how the economic
gain can be calculated. The four categories of cases are:
- violator gains additional market share;
- violator sells products or services prohibited by law;
- violator initiates construction or operation prior to government approval; and
- violator operates at higher capacity than it should have.
3.5 The Charge Questions for The Panel

The specific charge questions are:

2. How can the Agency more accurately characterize the types of cases that are
described in the White Paper? Have any of the examples and counter-examples in the White
Paper been misidentified with regard to whether they are amenable to the BEN model's
simplifying paradigm?

-------
     3.       Are there any suggestions for modifying the described analytical approach to
calculate the economic benefits and;

     4.       The Agency's proposed approach strives to avoid double-counting of the benefit
by laying out all relevant cash flows stemming from the violations, as opposed to simply adding
on the additional calculations to a BEN run.  What additional measures (if any) should the
Agency put in place to avoid such potential double-counting?
                                            12

-------
4. THE PANEL'S RESPONSES
4.1 The Economic Benefit is the Increase in Profits

The fundamental question for the determination of the economic benefit component of the
penalty is ... how much the profits of the firm have increased or will increase as a result of its
noncompliance? Profits can be increased either by an increase in revenue or a decrease in the
total cost of production (including abatement costs), or some combination of both. The BEN
model provides a reliable measure of the change in after-tax profit only if no other change would
have occurred that would have affected the firm's profit. This is an empirical question that
should be explored and not assumed.

The Agency's White Paper has essentially placed all of the other factors that might
influence the amount by which the violator's profit was increased by the violation in one of the
four categories under the heading of "benefit from illegal competitive activity." For several
reasons, the Panel finds that the Agency's use of the term "illegal competitive advantage" and
also the identification of four categories of ICA cases to be unhelpful.

1. It is not clear what the modifier "competitive" is intended to convey.

2. Increases in market share will often be difficult to identify in terms of comparing the
noncompliance scenario with the unobserved counterfactual compliance scenario; and observed
increases in market share might be difficult to attribute exclusively to the noncompliance.

3. In any case, increases in market share are not inherently valuable to the firm; what
matters is the impact of changes in market share on profits.

4. The categories of ICA, other than increases in market share, appear to be unusual
circumstances that are very context dependent.

The Panel believes that it would be more transparent to have only two categories of benefit
from noncompliance: (i) firms experienced no revenue increase and violators' profits were
increased by the amount of the delayed or avoided compliance costs; and (ii) firms gained profits
from increased sales.5 The BEN model would be applicable for those cases that fit into the first
category. For all other cases, we recommend that the Agency examine the facts of each case and
use methods and data appropriate to the case to estimate the changes in streams of revenue and/or
production costs as well as delayed or avoided compliance costs (if any). As already noted, the
Panel believes that BEN can be modified to deal with estimates of increased revenues.
4.2 Economic Benefit When Revenues Change Due to Noncompliance

When non-compliant firms do sell more than they would have if they had complied, their
economic benefit includes the profits they earn on the increased sales. A key point of potential
confusion is whether (or when) profits on increased sales should be added to avoided/delayed
costs as opposed to being a substitute measure of economic benefit.
4 The Panel's responses to the specific charge questions are in Section 4.4 below.

5 It is conceivable that in the long run a noncomplying firm could gain sufficient market power to enable it to
increase profit by reducing output and raising price. But we think that this is a remote possibility.
13

-------
Figure 1, which shows the effect of noncompliance on the profits of a firm operating in a
competitive industry, illustrates the issues. The horizontal line at the level PM indicates the
competitive market price.6 The two upward-sloping lines represent the marginal cost when the
firm complies and when the firm does not comply with environmental regulations. The graph is
based on the assumption that noncompliance lowers marginal cost, which is why the marginal
cost curve for noncompliance is the lower of the two curves. When the firm complies, its profit-
maximizing output is QC, the output where the marginal cost (given compliance) equals the price
set in a competitive market. When it does not comply, its lower marginal cost curve induces it to
sell a higher output, QN.

Under compliance, profits are given by triangle A.7 Under noncompliance, they are the
sum of areas A, B, and C. The economic benefit is, therefore, the sum of areas B, and C. The
cost avoided by noncompliance of producing QN is the sum of areas B, C, and D, which exceeds
the increase in profits from noncompliance by Area D.

Figure 1

Benefits from Non-Compliance -Competition
PM
MC (compliant)
MC (non-compliant)
QC
QN
Quantity
6 The line is horizontal because the firm, by assumption, operates in a perfectly competitive industry, which
means that its output does not affect the market price.
7 This assertion assumes that there are no fixed costs. With fixed costs, the graph gets more complicated, but
the basic principle remains the same. Avoided cost at the level of output actually produced overstates the
gain from noncompliance.
14

-------
Figure 2
Benefits from Non-Compliance - Monopoly
MQAC (compliant)

MQAC (non-compliant)

Demand
QC
QN
Quantity
The principle that delayed and avoided cost overstates the change in profits is not
restricted to firms operating in perfectly competitive markets. As Figure 2 shows, it also applies
to monopolists and firms operating in monopolistically competitive industries. In Figure 2, the
two solid horizontal lines represent unit costs when the firm is and is not in compliance with EPA
regulations.8 QC and PC are the profit-maximizing quantity produced and price charged when
the firm is in compliance while QN and PN are the profit-maximizing quantity and price when the
firm is not in compliance. Again, noncompliance lowers marginal cost and therefore causes the
firm to produce more than it otherwise would.

When the firm complies with regulations, its profits are the sum of areas A and B. When
it does not comply, its profits are the sum of B, C, D, and E. The economic benefit is, therefore,
the difference between the two, or C + D + E - A. This benefit is difficult to calculate, because
all that is observed is the actual prices and quantities (QN and PN). Calculating the true
economic benefit requires counterfactual estimation of the quantities that would have been
produced, and the prices that would have been charged, if the firm had complied (QC and PC).9
The graph as drawn is based on the assumption of constant returns to scale both with and without
compliance. That assumption simplifies the graph because it implies that marginal and average cost are
equal to each other.
This analysis is partial equilibrium in nature. That is, it ignores the effects of price changes on other parties,
for example those who purchase the firm's products or supply factor inputs to it. There might also be effects
in markets for substitute or complimentary goods. For purposes of establishing penalties as incentives for
compliance, only those changes that affect the noncomplying firm need to be considered.
15

-------
If, instead of calculating the true economic benefits to the violator, the EPA used avoided
costs at the quantity actually produced, that measure in Figure 2 would be areas C + D. This
avoided cost measure differs from the true measure by the amount A - E. It is a general
proposition in economics that A is greater than E (so that E - A is negative). (If it were not, even
a compliant firm could make more profits by producing QN than QC.) Thus, using avoided costs
at the actual quantity produced (i.e., C + D) overstates the true economic benefits to the firm of
noncompliance (i.e., C + D + E - A), since E - A is negative.

There are two situations in which a calculation of economic benefit based on
avoided/delayed costs could still be justified. The first is if it can be assumed that the effect on
output is sufficiently small that the error induced by ignoring output effects is also small. This
might occur if a firm has quasi-fixed capital, meaning that it is operating at or near full capacity
utilization and cannot increase its level of economic activity, at least in the short run, without
adding new capital. The second is if compliance would affect fixed costs only. In that case,
compliance would leave marginal cost and, accordingly, output unchanged.

Figure 3 can be used to analyze cases in which output would be 0 under compliance,
which would be the case if the MC (= AC) for compliance lay above the vertical intercept of the
demand curve. The areas A + B + C, the area between the MC (= AC) for compliance and the
MC (= AC) for noncompliance out to the quantity QN, is the measure of cost savings at the
observed level of output. This is an overestimate of the economic gain, which is given by the area
C. This class of cases includes those when a firm sells illegal output. It also covers many cases
involving illegal development of wetlands, for example.
Figure 3
Monopoly with Zero Output under Compliance
Quantity
16

-------
4.3. The Four Categories of Illegal Competitive Advantage

In this section we consider each of the four categories of 1C A in more detail and offer
comments on the appropriate methods for estimating economic benefit.

A. Violator Gains Additional Market Share

In this case, a violator gains market share by offering a price to the market that compliant
competitors cannot match. This is possible because failure to comply lowers costs, allowing the
firm to under-cut the market price based on the costs of compliant firms. The presumption is that
a gain in market share then leads directly to higher net revenues. Profits might not increase,
however, even with higher revenues if the non-compliant firm also experienced unexpectedly
higher unit costs at a higher level of output. This could result from overtime payroll expenses or
a decline in quality control, for example. Further, compliance costs are typically a relatively
small share of operating costs and unlikely to support long-term under-cutting of the market price.
Consequently, a case that considers only changes in market share is not useful in determining
whether there was economic gain as a result of the violation.

Example #1 in the White Paper (a firm bidding on a cost-plus contract) is highly contrived
as it brings together elements that would not generally be observed in one case. As a result of a
cost advantage from noncompliance, a company that is subject, in effect, to minimum price
regulation charges a lower price than it otherwise would and obtains a contract it would not have
gotten.10 The experience gained from the contract helps it get future business. The set of facts
seems unlikely because most price regulation is maximum price regulation and because price
regulation tends to arise in monopoly markets.

The discussion in the White Paper through the bottom of page 14 of how to deal with the
profits from the contract in Example #1 is appropriate. The remainder of the discussion in that
section is highly speculative because of the problems in translating increases in market share into
increases in profits. It is not likely to form the basis for practical, defensible calculations of
economic benefit.

It is of interest to consider separately the issue of a firm subject to cost-based price
regulation. An electric utility would be a possible example. If it charged a lower price as a result
of not complying with environmental regulations, noncompliance would increase the quantity of
electricity sold. Again, avoided cost would tend to overstate the economic benefit the utility
gained because it would ignore the fact that the cost-based regulation would allow the utility to
pass the cost on to customers. This is shown with Figure 4.
10 "Minimum price regulation" is a price floor, meaning that a company could not charge less than the
regulated price even if it wanted to. With minimum price regulation, a firm often must demonstrate that the
prices it charges exceed its costs by a specified margin. A firm with lower demonstrable costs is able to
undercut rivals and increase output. Most price regulation, such as the regulation of public utilities, sets a
maximum or a ceiling on what price a company can charge. (Even when a regulatory agency sets an exact
price that is technically both a floor and a ceiling, the rationale is usually to prevent the company from
charging more.)
17

-------
Figure 4
Regulated Natural Monopoly
-I
tt
PC
PN
Demand
AC- (compliant)
AC- (non-co
npliant)
QC
QN
Quantity
With cost-based price regulation of natural monopoly, price is set at average cost (AC),
which includes a reasonable return to capital. Under non-compliance (and assuming the pricing
authority is unaware of the relevance of the environmental regulations), price is set where AC
equals demand, at PN.11 The firm provides quantity of QN, and revenue equals D + H, as does
total cost. Economic profits are zero. If the firm complies with environmental regulations that
increase its costs, the pricing authority will allow the firm to charge a price of PC. The firm
provides quantity QC, and revenue under compliance equals A + B+ C + D, as does total cost.
Economic profits are still zero. Note that areas B + C + E + F + G,the area between the AC for
compliance and the AC for noncompliance out to the quantity QN, is the measure of cost savings
at the observed level of output. This is an overestimate of the economic gain, which is zero.

B. Violator Sells Products or Services Prohibited by Law

Customers might prefer, based on correct or incorrect information, to use a product that
has been prohibited, such as leaded paint, Freon, or certain pesticides with limited legal
application. Non-compliant firms that produce or sell these products would then gain revenues by
selling products that compliant firms do not offer to their customers. Such products might well
cost more to provide, but customers might be willing to pay a higher price to obtain products that
they perceive will meet their needs better than compliant products. The economic benefit is the
profit on the sales.

C. Violator Initiates Construction or Operation Prior to Government Approval

This case involves premature sales, which are analogous to sales of an illegal product.
The sales are illegal in the period before the permit is obtained. The approach recommended in
the White Paper is theoretically correct but likely to be difficult to implement in full generality.
In practice, the period of time over which noncompliance has an effect should be limited. But it
need not be limited to the head-start period. For example, suppose a company begins operations 6
months before it is allowed to and that it typically takes 1 year to attain full market penetration.
11 In practice, prices under cost-based regulations do not necessarily adjust immediately to cost changes. One
exception, though, is fuel-price adjustment clauses, which could come into play if a utility used a lower-cost
but higher-polluting fuel source.
18

-------
The head start would affect the level of sales up to the point when full penetration would have
been obtained under compliance. The longer the duration of the hypothesized effect, the more
speculative the estimate becomes. Attempts to link permanent changes in market share to the
head start are likely to be too speculative to withstand scrutiny.

D. Violator Operates at Higher Capacity Than It Should Have

The case hypothesizes that the company installs durable capacity that is non-compliant but
that it is allowed to use the capacity. As with example 1, the hypothesis seems contrived.
However, if such a case were to arise, it would be useful to ask whether noncompliance resulted
in sales that the firm could not have made legally, or whether the firm could have generated the
same level of sales at higher cost. If the sales could not have been made legally, then the
economic benefit is the profits on the increased sales. If they could, then benefit could be
measured as avoided cost. The conceptual problem with doing so is that the higher level of sales
might have proven unprofitable if the firm had to incur the costs associated with compliance.

If the firm makes sales it could not have made legally under compliance, then the profits
on the illegal sales are part of economic benefit. The White Paper is correct that BEN can be used
if there would have been a legal but higher-cost way to make those sales. If an economic benefit
calculation does have these two pieces, then it is important that the avoided costs be limited to the
avoided costs of producing the output that would have been legal under compliance.

4.4. Direct Responses to Charge Questions

Our answers to the four charge questions are as follows:

3. Are there any suggestions for modifying the described analytical approach to
calculate the economic benefits?

4 The Agency's proposed approach strives to avoid double-counting of the benefit
by laying out all relevant cash flows stemming from the violations, as opposed to simply
19

-------
adding on the additional calculations to a BEN run. What additional measures (if any)
should the Agency put in place to avoid such potential double-counting?

Every effort should be made to calculate economic advantage as avoided/delayed costs
(and therefore not to decompose the gain into separate components.) One should only resort to a
full-blown change-in-profit analysis when using avoided/delayed costs leads to a clearly
substantial overestimate or underestimate of the economic benefit. If it is necessary to do
change-in-profit analysis, it is important that the estimate of costs under compliance reflect the
lower level of output the firm would have produced rather than the actual production of the
polluter.

4.5. Revising the White Paper

We recognize that if the foregoing recommendations are accepted, it will be necessary for
the EPA to revise significantly the White Paper. We suggest that this be done in the following
manner. The Paper should start with the observation that the fundamental question for the
determination of the economic benefit component of the penalty is ... How much did the profits of
the firm increase as a result of its noncompliance? Profits can be increased either by an increase
in revenue or a decrease in the total cost of production (including abatement costs), or some
combination of both.

To determine the nature of the economic gain, we propose the following screening
questions:

(a) Did the violation lead to an increase in sales volume and/or revenue that would not
otherwise have occurred!

If the answer is "No," then economic gain is limited to avoided/delayed costs, and the
BEN Model can be used. If the answer is "Yes," then:

(b) Was there an increase in revenue but not in volume! (The answer to this question is
likely "No." For the answer to be "Yes," the violator would have had to sell the same volume but
charged a higher price, perhaps because the violation was to add an illegal ingredient that made
the product more effective. But it is hard to imagine circumstances in which this outcome would
occur as a result of noncompliance.)

If the answer to (b) is "Yes," then the BEN model as presently configured is not
appropriate for computing economic gain. It is necessary to estimate the increase in revenues as
well as the avoided/delayed compliance cost.

If the answer to (b) is "No," then the firm must have sold units of output that it would not
have sold if it had complied with EPA regulations. As explained in Section 4.2, in such a setting
avoided/delayed compliance cost overstates the true economic benefit of noncompliance, at least
in competitive and monopolistic markets.

It might nonetheless be appropriate to estimate economic benefit as avoided/delayed costs
if there was nothing inherently illegal about the sales themselves. To ascertain whether that is
correct, a "Yes" to (a) and a "No" to (b) should be followed by:

(c) Could the firm have made these incremental sales legally and complied with
regulations! (If the firm sold an illegal item, the answer should be "No." If the firm simply
20

-------
chose a higher level of output because of its cost-savings from failing to comply, it should be
"Yes.")

       If the answer is "No," then the BEN model is not appropriate for computing economic
gain.

       If the answer to (c) is "Yes," then in principle, use of the BEN model is still inappropriate.
However, as explained in Section 4.2, if it can be assumed that the effect on marginal cost and
output is sufficiently small that the error induced by ignoring output effects is  small, then
avoided/delayed cost can be taken as a reasonable approximation of economic benefit.

       In order for the OECA to implement our recommendations, it will have to have access to
the relevant expertise in economics. One possible source of this expertise in the Agency is the
National Center for Environmental Economics (NCEE).  But it might be more useful to OECA to
have its own in-house economist. This would be especially true if the agency  accepts our
recommendations in Section 6.4 for rethinking the civil penalty policy.
                                           21

-------
5. ADDITIONAL ISSUES
5.1. The Effect of Market Structure

The analysis of Section 4.2 shows that measures of delayed and avoided cost overstate
economic benefit when output is increased because of lower cost. This result holds for both
monopoly and competitive market structures. Whether the point is true in oligopoly is less clear.
In the frequently-used Cournot model, the effect of avoided and delayed cost on the actual level
of output understates the gains companies get from not complying. However, there are other
oligopoly models, such as the Bertrand and Stackelberg models, in which avoided and delayed
costs overstate the economic benefit from noncompliance, as is the case with monopoly and
perfect competition.12 Cases might arise in which the Agency would want to compute profits
from increased sales based on an underlying model of oligopoly. As the appropriate choice
among competing models would likely depend on the details of the violator's industry, however,
the committee cannot recommend a standard approach. Any estimate of economic gain from
noncompliance based on an oligopoly model is likely to be controversial and harder to defend in
court than an estimate of avoided or delayed cost. Thus, the EPA should only attempt such
estimates when it believes that the profits on increased sales are substantial.

5.2. Dynamic Effects

To this point, we have implicitly assumed that economic benefit from noncompliance
arises during the period of noncompliance. There are a variety of reasons, however, why
noncompliance could have enduring effects. The violator might gain customers who remain
loyal. There might be "learning curve" effects that give it strategic advantages in future periods.
It might be involved in an industry in which market saturation takes time. If noncompliance
allows the firm to enter the market earlier than it otherwise would have, noncompliance might
move forward the entire diffusion path.

The presence of dynamic effects does not alter the point that avoided/delayed costs over-
estimate economic gains when the polluter increases sales because of lower marginal production
costs from the noncompliance. This point follows from the general logic of optimization.
Forcing the firm to pay what it would have cost to comply with regulations at its actual output
leaves the firm as well off as it would have been if it had chosen that output and complied.
However, the firm might have done still better by choosing a different (presumably lower) output.
Thus, the presence of dynamic effects does not cause avoided/delayed costs to understate
economic advantage.
12 Modeling oligopolistic markets raises fundamental issues of economic logic. In general, forcing a firm to
pay what it would have cost to comply, given its actual level of output, leaves the firm with the profits it
would have had if the firm complied and it chose that same level of output. If it had complied, however, the
firm would not have chosen that output because the profits it generates are lower than it could get with a
different output. This logic breaks down in oligopoly models in which firms make incorrect conjectures
about the responses of rivals. In the Cournot model, any one firm could make higher profits by increasing its
output. A reduction in marginal cost due to noncompliance then induces the firm to do what it should have
done anyway - expand output. The different result for the Bertrand model is because each firm starts by
producing too much rather than too little. A marginal cost reduction from noncompliance would cause the
firm to produce still more and move to even lower profit levels. For further discussion of these oligopoly
models see one of the standard treatises, for example Shapiro (1989) or Tirole (1988).
22

-------
Dynamic effects create more of a problem for profits on increased sales as a measure of
economic benefit. If the firm sells more by virtue of not complying and those sales increase
future profits, then the value of those future profits is part of the economic gain from
noncompliance. A case could arise, for example, in which a company gets an unexpectedly large
order from a valued customer. Had it anticipated the order, the company could have made the
investments needed to fill the order and comply with environmental regulations. Having not
anticipated the order, however, it must either violate environmental regulations or risk losing
subsequent business.13 One might compute the economic gain from the violation as profits on
increased sales, but the proper measure would include profits on future sales, the extent and
duration of which might be hard to measure. An easier approach might be to determine what it
would have cost to bring the plant into compliance for the level of activity that actually occurred.
(Even if the notice on the order was so short that it was not physically possible to comply prior to
filling the order, one might estimate the economic gain as what compliance would have cost if it
did have sufficient notice.)

5.3. Ex Ante vs. Ex Post Assessments

A conceptual issue is whether the economic benefit from noncompliance should be
measured as the benefit the violator expects at the time it decides not to comply or the benefit it
actually realizes. (In economic terminology, the former is referred to as the ex ante benefit
whereas the latter is the ex post benefit.) These can be quite different. For example, suppose a
company illegally develops a wetland to start a business that turns out to be unprofitable. This
would be an example of case 2 in the White Paper. If the benefit is computed as the ex post
profits actually earned, the economic benefit recapture portion of the penalty would be zero. Yet,
the company presumably developed the business because ex ante it expected it to be profitable, so
it did expect to get a benefit at the time it decided to violate the law. Of course, the ex ante
benefit may also be lower than the ex post benefit. In the wetland example, this case would occur
when the development earned higher profits than expected. Enforcement personnel should avoid
simply selecting the method that results in the largest or smallest penalty.

Panel members debated whether and when ex ante penalties would be more appropriate
than the ex post version. Most members could envision cases in which an ex ante penalty would
be more desirable, either for fairness or deterrence reasons, but the panel was unable to formulate
general rules that would arguably cover all possible decision situations for EPA. Therefore, the
panel considers its advice on this subject to be cautionary. The Agency should recognize that the
standard ex post approach will not fit every penalty context.

To the extent that a violator should pay a penalty based on its expected rather than its
realized economic benefit from a violation, the Panel recognizes the practical question of how to
estimate what that ex ante amount was. One possibility suggested was for EPA to base an
estimate on evidence from any business plan, if available, that justified the action taken to
executives and board. A second suggestion was to examine the average profits earned from
comparable ventures, whether or not these involved violations of environmental regulations (legal
wetland development activities, for example). Where the benefit from the violation was arguably
a reduction in the risk to the firm, it could be measured in the insurance market from premiums
13 In public comments, Jasbinder Singh, President of Policy, Planning & Evaluation, Inc. of Herndon, VA
(2004) recounted one such case to the Panel. In that case, an automobile parts paint company violated
environmental regulations while satisfying an unexpectedly large order from Chrysler. See also Singh
(1999, and 2000).
23

-------
avoided.  Without knowing in advance what information will be available for an assessment of ex
ante benefit, it is difficult to judge the adequacy of these suggestions.
                                           24

-------
6. TOWARD AN OPTIMAL PENALTY POLICY
6.1 Economic Theory of Optimal Penalties

As explained in Section 3.2, the EPA penalty policy sets the goals of fairness and
deterrence as primary objectives in the determination of a civil penalty. Here we wish to discuss
these objectives and the larger question of the approach to the determination of a civil penalty in
the light of the economic theory of "optimal" penalty, originally developed by Becker (1968) in
the context of criminal punishment, and subsequently elaborated in a large body of economic
literature applying the notion to civil penalties as well, including penalties for environmental
offenses (see e.g., Cohen, 1992 and 1999).

The economic theory of optimal penalties approaches the issue of deterrence from the
perspective of economic efficiency rather than that of fairness, that is, the goal is to achieve an
efficient allocation of resources. This theory makes two points that are relevant to EPA's penalty
policy. The first is based on the assumption that potential offenders respond to both (a) the
probability of detection, and (b) the severity of punishment conditional upon being detected and
punished. Thus, deterrence may be enhanced by raising the penalty, by increasing monitoring
activities to raise the likelihood that the offender will be caught, or by changing legal rules to
increase the probability of punishment. And second, the economically optimal penalty balances
the harm done by an offense against the cost of deterring the offense. This balancing leads to the
conclusion that the appropriate methodology for calculating a penalty is to charge an amount per
offense equal to the (monetized) harm done divided by the probability of punishment (see Becker,
1968). This makes the expected value of the penalty equal to the harm.

It is worth emphasizing that this optimal penalty is based on the "harm" caused by the
offense, not the "gain" to the offender. To take a simple criminal example, if a mugger obtained
$100 in a robbery and the victim ended up spending three days in the hospital, a penalty based on
the $100 gain to the offender would surely be too low - and would "under-deter" such offenses.
The appropriate penalty would compensate the victim for three days in the hospital and pain and
suffering. In the context of environmental offenses, suppose a firm fails to install a $100 safety
valve and as a result 10,000 gallons of crude oil spilled into a sensitive coastal area. The $100
"gain" to the offender would certainly not be an appropriate starting point for a penalty. In both
of these cases, the problem is the failure to take account of the harm done to the victim in setting
the penalty. On the other hand, if the gain due to noncompliance were large relative to the harm,
a harm-based penalty would not deter noncompliance. But since the gain from noncompliance
exceeds the harm, noncompliance is actually the overall socially efficient outcome.

Alternatively, if the goal is to deter every violation of the law whether or not the gain
exceeds the harm ("absolute deterrence"), then a gains-based penalty is appropriate. We could
impose a penalty equal to the gain to the offender divided by the probability of detection and
punishment. Then it would never be in the potential offender's interest to violate the law. Some
offenses - like violent assaults and rapes - are of this nature (economists sometimes refer to these
as "unconditionally deterred" offenses) - society would never condone these offenses regardless
of the private benefit to the offender. However, pollution is usually a byproduct of a socially
beneficial activity. In the jargon of the law and economics literature, pollution is a "conditionally
deterred" offense - one that we only want to prohibit when its overall social costs exceed its
overall social benefits. If the expected penalty greatly exceeds the expected benefit to the
offender and yet the harm from the offense is relatively minor, the result will likely be "over-

-------
deterrence." On the other hand, as suggested in the example in the previous paragraph
concerning the under-deterring of a mugging offense, and as Polinsky and Shavell (1994) show
more generally, if the enforcement agency underestimates the gain to the violator, it becomes
more profitable to violate the law. Thus, gain-based penalties are more susceptible to under-
deterrence than harm-based penalties, because, even if harm is underestimated, the offense is still
likely to be deterred if it is very harmful.

       Thus, conceptually, if the goal of environmental policy were economic efficiency, the
EPA enforcement office should start with an examination of both the harm and the probability of
punishment. To do so would require relatively good data on both these elements - which are
difficult and sometimes impossible to quantify.  We are aware that many of the statutes governing
EPA appear not to make economic efficiency the goal but rather imply a goal of absolute
deterrence of polluting activities. In these cases, continuing to base the penalty on the gains to the
violator is appropriate.

       The next two sections deal with each of the two components of an optimal penalty - harm
and probability of detection. Following that, we discuss the current EPA penalty policy that
focuses primarily on "gain" instead of "harm," and examine what features of that policy might be
improved upon.


       6.2. Quantifying Harm

       If an environmental violation results in emissions levels that are beyond a legal standard,
there may be some harm to natural resources or human health.  Over the past 40 years, economists
have developed a variety of techniques to measure these harms  in monetary terms - including
both revealed preference approaches (e.g., travel cost methodology) and stated preference
approaches (e.g., contingent valuation).  The field of non-market valuation has emerged as a
major branch of environmental economics and there is a very extensive literature on the subject.
Measuring people's  value for non-market items in monetary terms (e.g., measuring what they
would be willing to pay to prevent a specific harm to the natural environment or the compensation
they would require to accept that harm) is inherently difficult, and in practice different
measurement techniques can produce different results (this is also true of market valuation).
While the methodologies are now well developed and have been used extensively by government
agencies  for the cost-benefit assessment of public investment projects, the design of public
policies,  and the assessment of natural resource damages, the relevant methodologies do continue
to evolve and there is some continuing disagreement about the relative merits of alternative
approaches and their overall reliability.14 Nevertheless, the Panel believes that the state-of-the-art
in benefits estimation has progressed to the point where EPA should seriously explore how it
might incorporate "harm-based" measures into its penalty formula, at least for some types of
environmental harm.

       We recognize that while some of the  methods used to value environmental harm can be
employed with relatively little cost, others require significant resources.  Thus, in many (if not the
majority  of) cases, these methods may not be practical unless the harm (and thus expected
penalty) is extremely large. Harm-based measures might only be appropriate for a small number
of cases.  But these are likely to be the cases that result in very significant  and quantifiable harm.
Furthermore, since the EPA already makes extensive use of non-market valuation to assess the
14     For comprehensive presentations of the methods for valuing changes in environmental conditions, see
       Freeman (2003) and Champ, Boyle, and Brown (2003).
                                           26

-------
efficacy of its environmental protection programs and policies, it seems to us appropriate that the
Agency should in principle be prepared to apply these same techniques, at least in some cases, to
assess the value of the damage when the environmental laws are violated.

For some types of violations, for example, oil spills and accidental releases of toxic
chemicals, the nature of the violation will be the failure to comply with safety regulations,
perhaps on a regular or continuing basis. In such cases, the realized harm will likely depend on
conditions that will vary with the time and location of the spill or release. In these cases, the
expected value of the harm rather than the realization of the harm should be the basis of the
penalty, since neither the entity undertaking the risky activity nor the enforcement agency can
know in advance what these conditions and the realized harm will be.

A possible approach to establishing harm based penalties would be to allow for use of
"gain to the offender" in cases where harm is not easily quantified and the cost of estimating harm
with sufficient accuracy is too great. This approach is similar to that employed by the U.S.
Sentencing Commission in determining the default fine tables for organizations punished for
federal crimes (USSC, 2003: Chapter 8 - Sentencing of Organizations). However, they mandate
the larger of harm or gain and specifically indicate that if one is hard to estimate, the court may
use the other.
6.3. Probability of Detection and Punishment

The probability of detection is likely to vary considerably by type of violation and even
across jurisdictions. By definition, the probability of punishment is bounded between zero and
one. Using the optimal penalty formula, this means that the optimal penalty is bounded by harm
and an infinite multiple of harm. Taking the most simplistic case of a very large oil tanker
accident, the probability of detection and punishment is likely to be one. Hence, the optimal
penalty is simply equal to the harm. This suggests that the optimal penalty for an extremely
harmful environmental violation is likely to be the monetary equivalent of harm - without
inflating the harm by a multiple. However, as the size of the harm decreases, all else equal, we
expect that the likelihood of detection also decreases.

Other factors that might influence the probability of detection and punishment are: (a)
whether or not a violator is subject to mandatory reporting that is available to the public to
scrutinize and file citizen lawsuits, (b) the ratio of facilities to inspectors in an EPA region, (c) the
strength of environmental activism in a region/state, and (d) whether or not the violator had a
history of violations and thus was subject to increased scrutiny or targeted enforcement.

An additional consideration in penalty calculations is that the offender may take actions to
reduce the likelihood of detection. For example, an oil tanker might clean its tanks far at sea to
evade detection by the Coast Guard. A firm that fails to meet permit standards might falsify
mandatory reporting records. Inspectors might be bribed or their attention diverted with false
emergencies or false leads. While these hypothetical examples are not exhaustive, they illustrate
that the EPA (and/or the Court) might ultimately determine that evasive actions were taken to
reduce the chance of being caught or prosecuted. Those actions would lead to lower detection
probabilities and hence higher penalties under the optimal penalty framework.

Although not widely employed in the environmental literature to date, numerous
techniques are available to estimate the probability of detection and punishment - depending
upon the circumstances. For a detailed discussion of this issue, see Parker (1989: 578-81). One
27

-------
widely used method is the "time until capture" approach which is most appropriate for ongoing
violations that occur over a period of time. Nash (1991) used this approach to estimate the
probability of detection for four types of fraud violations enforced by the Federal Trade
Commission - violations of FTC orders, violations of FTC regulatory standards, Truth-in-
Lending case, and unfair business practices. Nash concluded that the appropriate multiple for
these types of regulatory violation is approximately 4.0, indicating that the penalty should be four
times the harm.

Another method - the "capture/recapture" approach has its foundation in estimating the
number of animals in a given geographic area. When there are multiple sources of detection (e.g.
government inspectors as well as private citizens monitoring self-report data), one can exploit the
fact that there is some overlap between these multiple sources. By examining how many different
offenses are observed between the two "inspectors" and how many are identical, one can estimate
the total number of offenders in the population. For example, Froehlich and Bellantoni (1981)
estimated that the probability of detection for oil spills greater than 10 000 gallons was 0.87,
based on the combination of two independent sources of information.1 Cohen (1987: 44-5)
combined this with Coast Guard data indicating that they can identify the source of about 70
percent of spills that are detected, to arrive at an overall probability of detection of 60 percent.

6.4. Implications for Current EPA Policy

As discussed earlier, the current EPA penalty policy starts with the calculation of "gain" -
i.e., estimating the amount that the offender saved by not complying with environmental
regulations, and then adds a "gravity" component based in part on the harm from the offense.
However, the policy does not provide for quantifying the "harm" in monetary terms and also
ignores any explicit consideration of the probability of detection.

At least in those cases where economic efficiency is the objective of regulation, an
alternative approach that might be explored by EPA would be to provide for a "base" fine that is
predicated on the harm. If harm cannot be quantified, the base might either be "gain" or a
"default" fine level that is specified by type of offense. For example, EPA might study average
awards for interim lost use value by type of pollutant natural resource damage cases to arrive at
an approximation of the harm per "gallon" or per "ton" of a particular pollutant or waste. This
could be incorporated into a default harms-based fine table. But where the objective of regulation
is some form of absolute deterrence, the base penalty should be formulated in terms of the gain to
the polluter due to noncompliance.

In either case, the base fine would then be multiplied by a factor that is based on the
probability of detection and a penalty being imposed.16 As discussed above, in several settings,
the appropriate probability is 1.0, or so close to that value that any difference could be ignored.
Examples include really massive oil spills, whether in coastal waters or on the open sea (where
they would most likely result from serious damage to the tanker hull), and wetland destruction for
development purposes, where the evidence is by definition permanently in place. Another
example would be self-reported violations where the violator explicitly comes forward and
15 When the sources of information are not independent, the analysis is more complicated; but the method can
still be used.
16 This is similar to the approach taken by the U.S. Sentencing Commission (2003). Also see U.S. Sentencing
Commission (1988) for draft guidelines for sentencing organizations that more explicitly identify harm and
probability of detection as the controlling factors.
28

-------
announces its violation, corrects any problems, and offers to pay the appropriate penalty. In fact,
under EPA's Audit Policy, violators who expeditiously self-report and remedy a violation are
penalized on the basis of the BEN model and do not have to pay any gravity component -
effectively yielding a multiple of one. For smaller oil spills and other sorts of discharges that are
not necessarily detected automatically, Section 6.3 contains several examples and citations to the
relevant estimation literature. These techniques are not too difficult to implement, and EPA
should be in a position to gather relevant data.

It must be acknowledged that if the probability of detection and imposition of a penalty
is small, say 0.1 to 0.5, the penalty will be several times larger than the economic gain due to
noncompliance. And this might raise concerns about the fairness of the penalty policy. But such
low probabilities indicate that the Agency should seek ways to increase the probability of
detection and imposition of the penalty by improved monitoring and/or changes in the legal rules
governing the imposition of penalties.

The more typical regulatory violations such as exceeding a point source pollution discharge
permit can be divided into two broad classes: self-monitoring/self-reporting sources (where firms
are required to submit periodic reports of measured discharges), and all other sources. For the
self-monitoring sources, if we assume honest reporting, the probability ought to be one. But it is
clear from the existence of citizen suits that state enforcement agencies lack the resources to find
and penalize many types of reported violations. (These self-reports are different from the type of
self-reported violations noted above where the offender essentially calls up the regulatory
authorities and turns himself in.) The applicable probability is therefore less than one - how much
less could be estimated by examination of the accumulated data and comparison with the data on
violations pursued, whether by the state or by an NGO or citizen suit. This analysis could be done
using a random sample of firms to reduce the burden of estimating the probability.17

For non-self-monitoring/reporting sources, the relevant probability can be derived from the
rate of EPA (or state EPA) inspections, assuming that the inspections detect all violations. Some
estimates along these lines have been made in the past (Russell, 1983). Estimates could be based
on existing EPA and state data on regulated sources, permits, and inspections. It is not a trivial
exercise and would require some further investigation and some informed assumptions about the
duration of a typical violation, etc. However, while the data are not perfect, neither is there a need
for 100% accuracy. Instead, the goal is to arrive at some realistic estimate of the probability for
various offenses that can be applied uniformly to those offense types. In any case, we anticipate
that the Agency would have to develop regulations and procedures for establishing the
probabilities used for penalties.

The probability of detecting RCRA offenses might be more difficult to estimate. However,
it might be possible to compare the number of known illegal dumpsites to the number of illegal
"midnight" dumpers who are convicted of those offenses. If there are two different sources (e.g.,
"informants" and those identified directly through other law enforcement surveillance), one might
17 If reporting is not honest, the enforcement problem becomes much harder, since "audits," in the usual sense
of the word are not possible due to the ephemeral nature of the discharges. To find a real violation when
there was reported compliance would require actual monitoring at a time coinciding with the reported
compliance. The act of the monitoring, if observable by the source, would, one expects, eliminate the
temptation to lie about the compliance state, and so, without an informant, catching lying would be
impossible, though catching violations would not. The applicable probability for a violation would, as
discussed just below, be based on the probability with which the discharges were subject to "surprise"
measurement.

-------
be able to use the "capture-recapture" method described above to estimate the probability of
detection. Alternatively, one might need to resort to a default multiple that is the equivalent of (or
higher than) other empirically derived multiples, based on the assumption that these violations are
the most difficult to detect.

Overall, for several situations that concern EPA, a probability close to or equal to one will
be appropriate. But this will not be true in general for routine point source discharge permit
violations because of the lack of effort going into monitoring, either of the discharges themselves
or of the self-reports. On the other hand, the research required to find reasonable values for the
probability for self-reporting sources by state ought to be straightforward, since the reports are
likely to be archived, and there should also be some record of enforcement actions undertaken.
For sources that do not self-monitor, the approach would be to attempt to estimate the probability
that a randomly chosen source is visited and has its discharges sampled in a manner that
corresponds to the terms of its permit.

It should be emphasized that what is sought here is an approximate estimate of the general
probability of detection, not a highly elaborate calculation tailored to all the specific details of the
particular violation. This could well be handled in a practical manner by identifying a small
number of different types of violation, each associated with a generic estimate of the probability of
detection.

A degree of generality in assigning probabilities of detection and successful prosecution is
desirable in that this generality may reduce the appearance of arbitrary or capricious assignments
of probabilities for different cases. The accuracy of the probability assessments must be traded off
against acceptance of the probability algorithm by regulated firms.

EPA's civil penalty policy currently incorporates a few features that might proxy for the
probability of detection and imposition of a penalty. Specific gravity components are (U.S. EPA,
1984a: pp. 14-15):

I. Importance of the regulatory scheme - The policy indicates that violations that are more
important to the regulatory scheme will receive higher penalties. The example
given suggests that more important violations will be harder to detect in many
situations. Thus, the fact that no warning label is contained on a product would be
more important than a warning label that was simply too small. The existence of
the small warning label makes detection easy - since the product has already been
identified as being hazardous. Whether this one example is illustrative, and
whether other cases are related to the detection probability is unclear.

II. Availability of data from other sources - If a record keeping or reporting requirement
is violated and that is the only source of information, the probability of detection is
much lower than if multiple sources of the same data are available elsewhere.
Thus, this gravity component appears to be consistent with increasing the penalty
when the likelihood of detection is smaller.

Importantly, the policy also contains a provision that addresses the "general
deterrent" effect of the calculated gravity component of the penalty (EPA, 1984a: 16). This
provision states that in some cases, "the normal gravity calculation may be insufficient to effect
general deterrence. This could happen if there was extensive noncompliance with certain
regulatory programs in specific areas of the United States. This would demonstrate that the

-------
normal penalty assessment had not been achieving general deterrence." Thus, even though there is
no guidance on a proper multiple, there appears to be some understanding that detection
probability needs to be taken into account. The Panel recommends that EPA begin to study the
feasibility of formalizing these concepts and providing more explicit guidance on how to calculate
penalties that take into account both the harm and probability of detection.18
18     One public commenter (Fuhrman, 2004 and 2004a) questioned whether EPA had the legal authority to
       consider probability in setting penalties. But as noted in Section 3.2 above, deterrence has long been one of
       the objectives of EPA penalty policy. And the probability of detection and imposition of a penalty is a key
       factor in the deterrent power of a penalty policy.
                                               31

-------
         APPENDIX A - A MORE DETAILED DESCRIPTION OF THE SAB
               PROCESS AND PANEL REVIEW PROCEDURES
       A.I   Request for Review and Acceptance

             In June 2002, the Office of Enforcement and Compliance Assurance (OECA)
requested that the Science Advisory Board review the OECA White Paper. After considering all
requests for 2004, the Science Advisory Board determined that the review should be conducted by
a specialized panel. The Director of the Science Advisory Board Staff Office, in consultation with
the Chairman of the Science Advisory Board,  selected SAB member Dr. A. Myrick Freeman of
Bowdoin College, as chair of the Illegal Competitive Advantage (1C A) Economic Benefit (EB)
Advisory Panel.

       A.2   Panel Formation

             The panel was formed in accordance with the principles set out in the 2002
commentary of the Science Advisory Board, Panel Formation Process: Immediate Steps to
Improve Policies and Procedures (EPA-SAB-EC-COM-02-003).  A notice offering the public the
opportunity to nominate qualified individuals for service on the panel was published in the Federal
Register on August 6, 2003 (68 FR 46604) soliciting nominations for Panel membership and can
be found on the SAB Web site at: http://www.epa.gov/sab. Eleven individuals were considered
for membership on the panel.  On the basis of candidates' qualifications, interest, and availability,
the SAB Staff Office made the decision to put 11 candidates on the "short list" for the panel.  On
March 26, 2004, the SAB Staff Office posted a notice on the SAB Web site inviting public
comments on the prospective candidates for the panel.

             The SAB Staff Office Director — in consultation with SAB Staff (including the
Designated Federal Officer (DFO) and the Acting SAB Ethics Advisor) and the Chair of the
Executive Committee — selected the final panel. Selection criteria included: excellent
qualifications in terms of scientific and technical expertise; the need to maintain a balance with
respect to qualifying expertise, background and perspectives; willingness to serve and availability
to meet during the proposed time periods; and the candidates prior involvement with the topic
under consideration.  The final panel includes persons with expertise in one on more of the
following areas:

       (a)    Financial Economics, which includes Corporate Finance,
       (b)    Economic Benefit Recapture Issues,
       (c)    Business/Commercial Damages, which includes Anti-trust Law, Torts, and
             Economics,
       (d)    Business Economics and Competitive Strategy, which includes aspects of Statistical
             Decision-Making and Game Theory, as well as Competitive Effects of Vertical
             Integration and Quantitative Economics, and
       (e)    Industrial Organization, in the context of environmental regulations, and their
             enforcement, as well as Environmental and Regulatory Economics, Environmental
             Ethics and Sustainability in this context.
                                         A-l

-------
       The Panel members include individuals who are SAB members or consultants familiar with
the Agency as well as first-time consultants.  The final panel determination memo was posted on
July 9, 2004.


       A.3   Panel Process and Review Documents

             The Panel first met via conference call on July 12, 2004. The purpose of this public
conference call meeting was to provide background information for the Panelists on the issues in
preparation for the advisory activity.  The Panelists a) discussed the charge, review and
background materials provided to the Panel,  b) discussed specific charge assignments for the
Panelists, and c) advised the Office of Enforcement and Compliance Assurance (OECA) of any
specific points that need clarification for the August 5 & 6 advisory meeting.  Two Panelists were
unable to attend this initial conference call meeting.

             August 5-6, 2004 face-to-face meeting was held in Washington, DC.  This also was
a public meeting, and as in the teleconference call, an opportunity was provided for public
comments pursuant to and consistent with the requirements of the Federal Advisory Committee
Act (Public Law 92-463).  All but one of the panelists were physically present at the August 5 & 6,
2004 meeting. The one unable to attend the Washington meeting was available via conference call
hookup.

             Follow-up  conference calls were held on September 22, and November 4, 2004 and
January 19, 2005 to prepare and complete edits to the draft Advisory.  At the September 22, 2004
public conference call, the Panel discussed in a public forum, the edits that  were needed on the
internal working draft advisory.  The first public draft Advisory dated October 22, 2004 was
shared with the interested  public, including the Agency and discussed at the November 4, 2004
public conference call.  The second public draft Advisory dated December  15, 2004 was shared
with the interested public, including the Agency for discussions to take place at the January 19,
2005 public conference call. Following the January 19, 2005 public conference call, a March 23,
2005 public draft was prepared and provided to the SAB's Quality Review  Committee (QRC),
which met in a public conference call session on April 29, 2005. Subsequent to this public
conference call session, a  June 15, 2005 draft was prepared for review by the Board in a public
conference call held on July 13, 2005. All the above drafts and supporting  information were
posted onto the SAB Web site (www.epa. gov/sab) for review by the interested public (including
the Agency).  The final  edits were incorporated into this version which is provided to the EPA
Administrator, The Honorable Stephen  L. Johnson.
                                          A-2

-------
APPENDIX B - BRIEF BIOSKETCHES OF THE ILLEGAL
COMPETITIVE ADVANTAGE (ICA) ECONOMIC BENEFIT ( EB)
ADVISORY PANEL

Dr. Dallas Burtraw:
Dr. Burtraw is a Senior Fellow at Resources for the Future. He recently served on the
National Research Council, Committee on Air Quality Management in the United States and
serves as a reviewer, National Energy Modeling System, Energy Information Administration,
(1992-present). Dr. Burtraw's areas of expertise include: air pollution, cost-benefit analysis,
electricity restructuring, regulatory design, and public finance. His research interests include the
restructuring of the electric utility market, the social costs of environmental pollution, benefit-cost
analyses of environmental regulation, and the design of incentive-based environmental policies.
His current projects include the study of integrated approaches to pollutant control in the
electricity sector and the valuation of natural resource improvements in the Adirondacks.
Recently, Dr. Burtraw analyzed the cost-effectiveness of various designs for NO2 emission trading
in the eastern states and of the design for a carbon emission trading program in the electricity
sector. He also investigated the effects on electric utilities of the sulfur dioxide emissions-permit
trading program legislated under the 1990 Amendments to the Clean Air Act, and evaluated the
benefits of emission reductions resulting from the 1990 Amendments. He holds a Ph.D. in
Economics and a Master in Public Policy from the University of Michigan.

Dr. Mark Cohen:
Professor Cohen is Senior Associate Dean and Justin Potter Professor of American
Competitive Business at the Owen Graduate School of Management at Vanderbilt University. He
also serves as Co-Director of the Vanderbilt Center for Environmental Management Studies, and
as Visiting Professor of Criminal Justice Economics at the University of York (UK). He recently
served as Chairman of the American Statistical Association's Committee on Law and Justice
Statistics and is currently a member of the Stakeholder Council of the Global Reporting Initiative.
Prior to his position at Vanderbilt, he had served as senior economist with the U.S. Sentencing
Commission. His work experiences include the Federal Trade Commission, the U.S.
Environmental Protection Agency, the U.S. Department of the Treasury, and the U.S. Senate
Banking Committee. He received his B.S.F.S. in International Economics from Georgetown
University, and his M.A. and Ph.D. in Economics from Carnegie-Mellon University. Professor
Cohen has published over 70 articles on diverse topics such as enforcement of government
regulation, law and economics, white-collar and corporate crime, and environmental management.
Some of his prior work related to the proposed panel include: the costs and benefits of oil spill
enforcement policies; analysis of EPA s penalty policy; optimal penalties for corporate crime
including environmental and antitrust offenses; the public's willingness-to-pay for crime control
policies; why firms comply (and overcomply) with environmental regulations; does it "pay" to be
green; and the effect of disclosure on environmental performance. Research grants over the past
few years include "Measuring Public Perception of Appropriate Prison Sentences" (National
Institute of Justice, 1999) and "Does It Pay to be Green? The Relationship between Environmental
and Financial Performance" (W. Alton Jones Foundation, 1996). In addition he has recently
served as a consultant to two different research projects on corporate environmental performance:
(1) University of Kansas, funded by EPA, and (2)University of Maryland, funded by NIJ.
B- 1

-------
Dr. A. Myrick Freeman III:
Dr. Freeman is Research Professor of Economics at Bowdoin College. In 2000 he retired
from teaching after 35 years. Dr. Freeman received his Ph.D. in economics from the University of
Washington in 1965. He has been on the faculty at Bowdoin since that time and has served as
chair of the Economics Department and Director of the Environmental Studies Program there. He
has also held appointments as Visiting College Professor at the University of Washington and
Robert M. La Follette Distinguished Visiting Professor at the University of Wisconsin-Madison
and as a Senior Fellow at Resources for the Future, a research organization in Washington, DC.

Dr. Freeman's principal research interests are in the areas of applied welfare economics,
benefit-cost analysis, and risk management as applied to environmental and resource management
issues. Much of his work has been devoted to the development of models and techniques for
estimating the welfare effects of environmental changes such as the benefits of controlling
pollution and the damages to natural resources due to releases of chemicals into the environment.
He has authored or co-authored eight books including Air and Water Pollution Control: A Benefit-
Cost Assessment, and The Measurement of Environmental and Resource Values: Theory and
Methods, now in its second edition. He has also published more than 70 articles and papers in
academic journals and edited collections. Dr. Freeman has been a member of the Board on
Toxicology and Environmental Health Hazards of the National Academy of Sciences and has
served as a member of the Advisory Council on Clean Air Compliance Analysis, the Clean Air
Science Advisory Committee and the Environmental Economics Advisory Committee of the U.S.
Environmental Protection Agency Science Advisory Board. Most recently, he chaired the EPA
SAB Review Panel on UST/RCRA Benefits, Costs, and Impacts Assessment.

Dr. Jane V. Hall
Dr. Jane V. Hall is Professor of Economics in the College of Business and Economics and
Co-Director of the Institute for Economic and Environmental Studies at California State
University, Fullerton. Her current research areas are assessing the value of environmental
protection, economics of air pollution policy, natural resource scarcity, and environmental
resource scarcity and conflict. She has lectured and conducted research on the topics of energy,
sustainability, resource scarcity and conflict, benefit assessment, economic performance and
environmental regulation, economic incentives for environmental management and related topics.
She has developed positions on air quality standards, fuel composition and taxation, energy policy
as an Associate Staff Scientist with the Environmental Defense Fund and as a Special Advisor to
the Chair of the California Air Resources Board, and Deputy Assistant for Environmental
Protection to the Governor of California. She has also served as an economist with Unocal (Union
Oil Company) to assess the impact of federal and state energy policies on the economy and the
energy industry. She has published over 100 articles, books or book chapters, working papers and
presentations on the above topics. She has served as a member of the Advisory Council on Clean
Air Compliance Analysis (COUNCIL), and its Health and Ecological Effects Subcommittee, the
EPA's Children's Health Protection Advisory Committee, and a number of other advisory and
scientific bodies. She has served as a reviewer for the National Science Foundation, California Air
Resources Board Research Division, and for the following publications: Contemporary Economics
Policy, Ecological Economics, Environmental Science and Technology, the Journal of Economics
and Environmental Management, the Journal of Environment and Development, and the National
Science Foundation's Science Journal. Dr. Hall received her B. A. in Economics from the
University of Washington, her M.S. in Agricultural and Resource Economics and her Ph.D. in
Energy and Resources from the University of California at Berkeley.

B-2

-------
During the past five years, Dr. Hall has had research funding from the California Air
Resources Board (A Pilot Study to Quantify Health Benefits of Incremental Improvements in Air
Quality; Economic Valuation of Ozone-Related School Absences in the South Coast Air Basin;
and Innovative Clean Air Technology Assessment), the W. Alton Jones Foundation (Growth for
Health: the Zero Emission Vehicle and California's Future Prosperity), Sea Grant/NOAA
(Economic Valuation of the Rocky Intertidal Zone), and the U.S. Environmental Protection
Agency and City of Houston (Valuation of Air Pollution and Health).

Dr. W. Michael Hanemann:
Dr. W. Michael Hanemann is Chancellor's Professor in the Department of Agricultural and
Resource Economics and Goldman School of Public Policy at the University of California,
Berkeley. He is Director of the California Climate Change Center at UC Berkeley. Dr.
Hanemann's research interests include non-market valuation, environmental economics and policy,
water pricing and management, demand modeling for market research and policy design, the
economics of climate change, the economics of irreversibility and adaptive management, and
welfare economics. Dr. Hanemann's recent publications have addressed the economic impact of
climate change on US agriculture, fishery management under multiple uncertainty, non-market
valuation using the contingent valuation method, the economic value of reducing asthma, and the
economic theory of willingness to pay and willingness to accept.

Dr. Hanemann was educated at Oxford University (B.A.), the London School of
Economics (M. Sc.), Harvard University (M.A. in Public Finance and Decision Theory, and
Harvard University (Ph.D. in Economics). He was awarded an Honorary Ph.D. by the Swedish
University of Agricultural Sciences. Dr. Hanemann is a member of the California Bay-Delta
Authority Drinking Water Advisory Committee. He served as Chair of the Organizing Committee
for the Second World Congress of Environmental and Resource Economists, held in Monterey CA
in June 2002. In the past 5 years, Dr. Hanemann has received research funding from the US EPA
STAR Grant Program (economic value of childhood asthma, embedding in contingent valuation);
NSF (price and non-price tools for water conservation), NOAA, MMS, the California State Water
Resources Control Board and The California Department of Fish & Game (economic value of
beach recreation in Southern California), and the California Energy Commission (climate change
policy in California).

Dr. Catherine L. Kling:
Dr. Kling is a Professor of Economics at Iowa State University (ISU) and Head of the
Resource and Environmental Policy Division of the Center for Agricultural and Rural
Development at ISU. Prior to coming to Iowa State University in 1993, she was an Associate and
Assistant Professor in the Department of Agricultural Economics at the University of California,
Davis. She has taught graduate and undergraduate courses in environmental economics,
microeconomic theory, and econometrics. Dr. Kling's research encompasses nonmarket valuation
issues in environmental economics and economic incentives for pollution control related especially
to agricultural problems. Her research has been published in a variety of economics journals
including The Review of Economics and Statistics, Journal of Public Economics, Journal of
Environmental Economics and Management, American Journal of Agricultural Economics, Land
Economics, Environmental and Resource Economics, and Ecological Economics.

Dr. Kling has also served the profession and the public sector in a variety of capacities
including her current membership on EPA's Environmental Economics Advisory Committee to the
Science Advisory Board. Current and past service includes as a member of the board of directors
and awards committee chair for the American Agricultural Economics Association, vice president

B-3

-------
and member of the board of directors of the Association of Environmental and Resource
Economists, associate editor for the American Journal of Agricultural Economics, and the Journal
of Environmental Economics and Management., as well as numerous ad hoc committees for the
AAEA, AERE, and other professional associations. Dr. Kling's research support has been
provided through grants from the Iowa Department of Natural Resources, the U.S. Environmental
Protection Agency, the U.S. Department of Agriculture, the California Institute for Energy
Efficiency, the Giannini Foundation, and the Sloan Foundation. Dr. Kling holds a B.A. in
Business and Economics from the University of Iowa, and a Ph.D. in Economics from the
University of Maryland.

Dr. Arik Levinson:
Dr. Levinson is an Associate Professor in the Economics Department of Georgetown
University, where he teaches environmental economics, public finance, and microeconomics, and
is Director of Undergraduate Economic Studies. He is a Faculty Research Fellow at the National
Bureau of Economic Research, is on the Editorial Council of the Journal of Environmental
Economics and Management, and is a member of the American Economic Association, the
Association of Environmental and Resource Economists, and the Association for Public Policy
Analysis and Management. Professor Levinson's research interests include the fields of public
finance and environmental economics. He has studied the theoretical welfare consequences of
states competing to attract manufacturers by enacting successively less stringent environmental
standards (a "race to the bottom"), and measured empirically the effects of interstate differences in
environmental standard stringency on manufacturer location decisions, trade, employment, and
foreign direct investment. Recently, he has written theoretical and empirical papers on the
relationship between countries' environmental quality and their incomes. He has studied the
energy efficiency consequences of apartment leases that include monthly utility costs, and he has
written about the relationship between individuals' willingness to pay for environmental quality,
household income, and national income. His research has in part been funded by the National
Science Foundation, and by the Association for Public Policy Analysis and Management. Dr.
Levinson holds a Ph.D. in Economics from Columbia University.

Dr. Clifford S. Russell:
Dr. Clifford S. Russell is Professor of Economics, Emeritus, Vanderbilt University; and
Research Associate, Bowdoin College. He joined the Vanderbilt faculty as professor of economics
and director of the Institute for Public Policy Studies in January, 1986. Before coming to
Vanderbilt, Dr. Russell was a Senior Fellow and Director of the Environmental Quality Research
Division at Resources for the Future in Washington, D.C. During his 17-year tenure there, he held
several other leadership positions. He is the author and editor of 16 books and author or co-author
of 68 articles in environmental economics. His major current interest is in the systematic
examination of environmental labeling as a tool of environmental policy. Dr. Russell has served
as a member of several National Academy of Science committees, and on the Environmental
Studies Board. In 1992/93 he chaired an NAS panel evaluating the U.S. Department of Energy's
proposed system for setting clean-up priorities at contaminated nuclear weapons and research
facilities. He was President of the Association of Environmental and Resource Economists in
1993 and 1994. From December, 1996, to August, 1997, he held the Valfrid Paulsson visiting
chair in environmental economics at the Beijer Institute, part of the Royal Swedish Academy of
Sciences in Stockholm. In 2003 he held the Thomas Sowell Distinguished Visiting Chair of
Economics at Bates College. In the 1970s and '80s Dr. Russell was on the Executive Committee
of the Board of the Environmental Defense Fund (now Environmental Defense). He also served
on the board of the Tennessee Environmental Council. Dr. Russell received his B.A. in
Mathematics from Dartmouth College and his Ph.D. from Harvard University, where he was a

B-4

-------
Harvard Graduate Prize Fellow in Economics. From 1960 through 1963, he served as a
commissioned officer in the U.S. Navy.

Dr. Michael A. Salinger:
Dr. Salinger is currently serving as Director of the Bureau of Economics with the U.S.
Federal Trade Commission, in Washington, DC and is on leave as Professor of Economics at the
Boston University School of Management. He served as an economist in the Bureau of Economics
in the Antitrust Division with the United States Federal Trade Commission while on leave from
Columbia University. At Columbia University, he served as Associate Professor of Economics
and Finance. He also was a Visiting Associate Professor of Economics at MIT's Sloan School of
Management. Dr. Salinger is on the Editorial Boards of the Journal of Industrial Economics, and
Review of Industrial Organization. He has published on such topics as the relationship between
market structure and corporate profitability, the competitive effects of business practices
(including vertical mergers and bundling), the statistical properties of firm growth, antitrust policy,
and the regulation of telecommunication prices. His recent research has been funded by the
National Science Foundation and by Microsoft. He has served as a peer reviewer of the BEN
model for the EPA. He received his B.A. in Economics form Yale University and his Ph.D. in
Economics from Massachusetts Institute of Technology.

Dr. David Sunding:
David Sunding is a professor at the University of California at Berkeley in both the College
of Natural Resources and the Boalt Hall School of Law. He received a B.A. in Economics from
Claremont McKenna College in 1983 and his Ph.D. in Agricultural and Resource Economics from
the University of California at Berkeley in 1989. He specializes in environmental policy, natural
resource economics, land use, and law and economics. Prior to his current position, Prof. Sunding
served as a senior economist at the President's Council of Economic Advisers where he had
responsibility for natural resource and environmental policy. He currently serves as member of the
Science Advisory Board of the National Center for Housing and the Environment and is the co-
director of UC Berkeley's Center for Sustainable Resource Development.

Professor Sunding is the author of over 50 journal articles and book chapters in the areas of
environmental economics, natural resource economics, and law and economics. He has been
commissioned to write over 30 technical reports and monographs for government and private
interests. Recently, Professor Sunding's research has focused on the measurement of
environmental compliance costs, environmental regulation and processes of urban growth and
development, and the diffusion of conservation technology. Dr. Sunding has had extensive
litigation experience in the areas of compliance cost measurement, environmental remediation and
cost allocation, antitrust and unfair competition, and agricultural and natural resource markets. He
has performed economic and financial analysis relating to damage calculations, market
determination, real property valuation, antitrust and price discrimination and has testified at
deposition and trial. He has recently received grants and/or research funding from the U.S.
Environmental Protection Agency, Food Systems Research Group, California Department of Food
and Agriculture, California Department of Water Resources and U.S. Department of the Interior.
B-5

-------
                       APPENDIX - C ACRONYMS
AAEA
AC
ACC
ADV
AERE
ALJ

BEN
BNA
CAA
CERCLA

CASAC
CFR
COM
COUNCIL

CWA
DC
DFO
DOT
EB
EC
EEAC

EPA
EPCRA
FIFRA
FR
FTC
GM
ICA
ISSN
ISU
LLC
MC
MIT
NAS
NCEE
NGO
NIJ
American Agricultural Economics Association
Average Cost
American Chemistry Council
Advisory
Association of Environmental Resource Economists
Administrative Law Judges (of the U.S. EPA; NOTE: There are
ALJs in other organizations)
Benefits Calculation Computer Model (to calculate the economic
benefit a violator derives from delaying and/or avoiding compliance
with environmental statutes)
Bureau of National Affairs
Clean Air Act
Comprehensive Environmental Response Compensation and
Liability Act
Clean Air Scientific Advisory Committee (of the U.S. EPA/SAB)
Code of Federal Regulations
Commentary (U.S. EPA/SAB)
Advisory Council on Clean Air Compliance Analysis (U.S.
EPA/SAB/COUNCIL)
Clean Water Act
District of Columbia
Designated Federal Officer
Department of the Interior (U.S. DOI)
Economic Benefit
Executive Committee (of the U.S. EPA/SAB)
Environmental Economics Advisory Committee (of the U.S.
EPA/SAB)
Environmental Protection Agency (U.S. EPA)
Emergency Planning and Community Right-to-Know Act
Federal Insecticide, Fungicide and Rodenticide Act
Federal Register
Federal Trade Commission
General Management
Illegal Competitive Advantage
International Standard Serial Number
Iowa State University
Limited Liability Corporation
Marginal Cost
Massachusetts Institute of Technology
National Academy of Science
National Center for Environmental Economics (U.S. EPA/NCEE)
Non-Government Organization
National Institute of Justice
                                 C-l

-------
NOAA             National Oceanic and Atmospheric Administration (U.S. NOAA)
NSF               National Science Foundation
OECA             Office of Enforcement and Compliance Assurance (U.S.
                   EPA/OECA)
OECM             Office of Enforcement and Compliance Monitoring (U.S.
                   EPA/OECM)
OPA               Oil Pollution Act
PC                 Price-Compliant
PM                Competitive Market Price
PN                 Price Non-Compliant
QC                 Quantity-Compliant
QN                Quantity Non-Compliant
QRC               Quality Review Committee (U. S. EPA/SAB)
RCRA             Resource Conservation and Recovery Act
SAB               Science Advisory Board (of the U.S. EPA/SAB)
SDWA             Safe Drinking Water Act
TSCA              Toxic Substances Control Act
UC                 University of California
UK                United Kingdom
USSC              United States Sentencing Commission
USC               United States Code
U.S.               United States
                                 C-2

-------
                                      REFERENCES

       WORKS CITED:

       Becker, Gary S. (1968). "Crime and Punishment:  An Economic Approach," Journal of
Political Economy 76. 169-217.

       Champ, Patricia A., Kevin J. Boyle, and Thomas C. Brown. (2003). A Primer on
Nonmarket Valuation. Norwell, MA: Kluwer Academic Publishers.

       Cohen, M.A. (1999). "Monitoring and Enforcement of Environmental Policy,"
International Yearbook of Environmental and Resource Economics 1999/2000. edited by Tom
Tietenberg and Henk Folmer; Edward Elgar publishers, pages 44-106.

       	. (1992). "Environmental Crime and Punishment: Legal/Economic Theory and
Empirical Evidence on Enforcement of Federal Environmental Statutes," Journal of Criminal Law
& Criminology 82(41 1054-1108.

                  . (1987). "Optimal Enforcement Strategy to Prevent Oil Spills: An Application
of a Principal-Agent Model with 'Moral Hazard'," Journal of Law and Economics. 30(1), 23-51.

       	. (1986). "The Costs and Benefits of Oil Spill Prevention and Enforcement,"
Journal of Environmental Economics and Management. 13, 167-88.

       Freeman, A. Myrick III.  (2003). The Measurement of Environmental and Resource
Values: Theory and Methods. 2nd edition, Washington, DC: Resources for the Future.

       Froehlich, M.A. and J. F. Bellantoni. (1981). "Oil Spill Rates in Four U.S. Coastal
Regions," in 1981 Oil Spill Conference Proceedings. American Petroleum Institute.

       Fuhrman, R.H. (2004). Comments of the Manufacturers Ad Hoc Group. [Prepared by
Robert H. Fuhrman, Seneca Economics and Environment, LLC], July 22, 2004

       Fuhrman, R.H. (2004a).  Comments on the Draft Advisory of the ICA EB Advisory Panel.
[Prepared and submitted by Robert H. Fuhrman, Seneca Economics and Environment, LLC on
Behalf of: American Chemistry Council, Corporate Environmental Enforcement Council],
November 3, 2004, 3 pages

       Nash, J. (1991). "To make the punishment fit the crime: The theory and statistical
estimation of a multi-period optimal deterrence model," International Review of Law and
Economics 11, 101-110.

       Parker, J. S. (1989). "Criminal Sentencing Policy for Organizations: The Unifying
Approach of Optimal Penalties." American Criminal Law Review. 26: 513-604.

       Pigou, Arthur. C. (1918). The Economics of Welfare. London: Macmillan.

       Polinsky, A.M. and S. Shavell. (1994).  "Should Liability Be Based on the Harm to the
Victim or the Gain to the Injurer?" Journal  of Law. Economics & Organization 10(2), 427-437.

                                         R- 1

-------
       Russell, Clifford S. (1983). Pollution Monitoring Survey Summary Report. Washington,
D.C.: Resources for the Future.

       Shapiro, Carl. (1989). "Theories of Oligopoly Behavior," in Richard L. Schmalensee and
Robert D. Willig, eds., Handbook of Industrial Organization. Vol. 1, Amsterdam: North Holland.

       Shefftz, Jonathan S. (2004). "Wrongful Profits: Setting the Record, and the Concept,
Straight," Environment Reporter. BNA, Inc. Vol. 35, No. 1, January 2, 2004

       Singh, Jasbinder. (1999). "Countering the Dean Dairy 'Hammer' With a Unified Theory of
Economic Benefit of Noncompliance," Environment Reporter. BNA, Inc., Vol. 29, No. 41,
February 19, 1999, pp. 2096-2101

       Singh, Jasbinder. (2000). "Making Business Sense of Environmental Compliance." MIT
Sloan Management Review. Vol. 41, No. 3, Reprint 4137, Spring, 2000

       Singh, Jasbinder. (2004). "Comments Before the Illegal Competitive Advantage Economic
Benefit Advisory Panel," Science Advisory Board, Environmental Protection Agency, August 5,
2004

       Tirole, Jean.  (1988). The Theory of Industrial Organization. Cambridge, MA: MIT Press.

       U.S. Environmental Protection Agency. (1978). "Civil Penalty Policy," EPA Office of
Enforcement, April 11, 1978

       U.S. Environmental Protection Agency. (1984a). Policy on Civil Penalties,  EPA General
Enforcement Policy #GM-21, February 16, 1984 (Reprinted in 17 Environmental Law Review
35083, October 1987)

       U.S. Environmental Protection Agency. (1984b). Framework for Statute-Specific
Approaches to Penalty Assessments, EPA General Enforcement Policy #GM-22.

       U.S. Environmental Protection Agency (EPA). (1999). BEN User's Manual.  September,
1999

       U.S. Environmental protection Agency (EPA). (1999). Appendix B, "Penalty Provisions
from Environmental  Statutes," 3 pages (An Appendix to Libber, Jonathan. (1999). "Making the
Polluter Pay: EPA's Experience in Recapturing A Violator's Economic Benefit from
Noncompliance," 1998 Proceedings from International Environmental Enforcement Conference,
Monterey, California);

       U.S. Environmental Protection Agency. (2003). "Identifying and Calculating Economic
Benefit That Goes Beyond Avoided and/or Delayed Costs, " May 25, 2003, Office of Enforcement
and Compliance Monitoring (OECM),  (The "White Paper")

       U.S. Sentencing Commission. (2003). Guidelines Manual (November).

       	. (1988). "Discussion Draft of Sentencing Guidelines and Policy Statements
for Organizations," reprinted in Whittier Law Review. 10(1): 7-75.

                                         R-2

-------
CASE STUDIES AND BACKGROUND MATERIALS:

Agency of Natural Resources v. Richard Demo (2001-532), Entry Order 2003 VT 36,
Supreme Court Docket No. 2001-532, Appealed from Environmental Court, January 2003;

Ballard, Andrew, M., "N.C. Court Orders Illegal Landfill to Close; Forfeiture of Profits
CalledNew State Tool, " BNA, Inc No. 84, Monday, May 3, 2004 ISSN 1521-9402;

Borden Ranch Partnership; Angelo K. Tsakopoulos, Plaintiffs-Appellants v. United States
Army Corps of Engineers, United States Environmental Protection Agency, No. 00-15700, United
States Court of Appeals for the Ninth Circuit Court, July 9, 2001 Argued and Submitted, San
Francisco, California, August 15, 2001, Filed;

Borden Ranch Partnership and Angelo K. Tsakopoulos, Plaintiffs, v. United States Army
Corps of Engineers and United States Environmental Protection Agency, Defendants. And
Related Counterclaim., CIV. S-97-0858 GEB JFM, United States District Court for the Eastern
District of California, Lexsee 12999 US Dist LEXIS 21389, November 8, 1999, Decided,
November 8,1999 Filed;

Garlow, Charles and Jay Ryan Article: A Brief Argument for the Inclusion of An
Assessment of Increased Market Share in the Determination of Civil Penalty Liability for
Environmental Violations: Letting Corporations Share the Regulatory Burden of Policing Their
Markets, 22 B.C. Envtl. Aff. L. Rev 27, Fall, 1994;

In the Matter of E.I. DuPont De Nemours & Co., Inc. Respondent, United States
Environmental Protection Agency, Office of Administrative Law Judges, 1998 EPA ALJ LEXIS
129, April 30, 1998;

In the United States District Court for the Western District of Wisconsin, United States of
America, Planitiff vs. Peter Thorson, Managed Investments Inc., Construction Management, Inc.,
and Gerke Excavating, Inc. Madison, Wisconsin, Case No. 03-C-0074-C, May 4, 2004;

Libber, Jonathan. (1999). "Making the Polluter Pay: EPA's Experience in Recapturing A
Violator's Economic Benefit from Noncompliance," 1998 Proceedings from International
Environmental Enforcement Conference, Monterey, California (Includes U.S. EPA, 1999,
Appendix B, "Penalty Provisions from Environmental Statutes," 3 pages);

Libber, Jonathan. (1996). "Impact of One Policy Change on EPA Enforcement Actions,"
Winter 1996, p. 11, Environmental Law. University of Maryland School of Law;

Summary of Significant 1C A Cases, Prepared by U.S. EPA/OECA Staff, July 26, 2004;

United States of America v. The Municipal Authority of Union Township, Dean Dairy
Products, Inc., d/b/a Fairmont Products, Appellant, No. 97-7115, United States Court of Appeals
for the Third Court, March 19, 1998, Argued, July 20, 1998, Filed
R-3

-------
       United States of America, Plaintiff, v. MAC's Muffler Shop, Inc., and Winston McKinney,
defendant, Civil Action No. C85-138R, United States District Court for the Northern District of
Georgia, Rome Division, 1986 U.S. Dist. LEXIS 18108; 25 ERC (BNA) 1369, November 4, 1986,
Decided and Filed;

       United States Environmental Protection Agency Before the Administrator, In the Matter of
Bretton Construction, Co., BIC Investments, Inc., and William and Mary Hammond, Respondents,
Docket No. CWA-III-096, 1994;

       United States Environmental Protection Agency Before the Administrator In the Matter of
Campeachy Corporation Respondent Docket No. 5-IFFRA-96-017, Initial Decision by Andrew S.
Pearlstein, Administrative Law Judge, February 25, 1999;

       United States of America, Plaintiff v. The Municipal Authority of Union Township; and
Dean Dairy Products Co., Inc. d/b/a Fairmont Products, Defendants, Civil Action No. 1 :CV-94-
0621, United States District Court for the Middle District of Pennsylvania, July 10, 1996, Decided,
July 10, 1996, Filed;

       United States Environmental Protection Agency Before the Administrator In the Matter of:
Lawrence John Crescio, III (also known as John Crescio) Respondent, Docket No. 5-CWA-98-
004, Initial Decision, May 17, 2001;

       Van Hollen, J.B., U.S. Attorney, Western District of Wisconsin, Press Release Pertaining
to Gerke Excavating, Inc. of Tomah, Wisconsin, May 5, 2004;
       PUBLIC COMMENTS:

       American Chemistry Council. (2004). Comments on the Draft Advisory of the 1C A EB
Advisory Panel. (2004). [Prepared and submitted by Robert H. Fuhrman, Seneca Economics and
Environment, LLC on Behalf of: American Chemistry Council (ACC), Corporate Environmental
Enforcement Council], November 3, 2004, 3 pages

       Manufacturers Ad Hoc Group. (2004). Comments of the Manufacturers Ad Hoc Group.
(2004). [Prepared by Robert H. Fuhrman, Seneca Economics and Environment, LLC], July 22,
2004
       Manufacturers Ad Hoc Group. (2005). Comments on the December 15. 2004 Draft
Advisory of the 1C A EB Advisory Panel. (2005). [Prepared and submitted by Robert H. Fuhrman,
Principal and CEO, Seneca Economics and Environment, LLC on Behalf of and with the
concurrence of the Manufacturers Ad Hoc Group], January 18, 2005

       Shefftz, Jonathan S. (2004).  "Wrongful Profits: Setting the Record, and the Concept,
Straight," Environment Reporter. BNA, Inc. Vol. 35, No. 1, January 2, 2004

       Singh, Jasbinder. (1999). "Countering the Dean Dairy 'Hammer' With a Unified Theory of
Economic Benefit ofNoncompliance, " Environment Reporter. BNA, Inc., Vol. 29, No. 41,
February 19, 1999, pp. 2096-2101


                                         R-4

-------
       Singh, Jasbinder. (2000). "Making Business Sense of Environmental Compliance, " MIT
Sloan Management Review, Vol. 41, No. 3, Reprint 4137, Spring, 2000

       Singh, Jasbinder. (2004). "Comments Before the Illegal Competitive Advantage Economic
Benefit Advisory Panel, " Science Advisory Board, Environmental Protection Agency, August 5,
2004
       (End of Text)
                                        R-5

-------