|
21 ELR 10246 | Environmental Law Reporter | copyright © 1991 | All rights reserved The Role of EPA's BEN Model in Establishing Civil PenaltiesRobert H. FuhrmanMr. Fuhrman is a Principal in the Washington, D.C., office of Putnam, Hayes & Bartlett, Inc. (PHB), an economic and management consulting firm. He received a B.A. from Columbia University and an M.B.A. from Harvard Business School. From 1977 to 1983, he worked as an economist at the U.S. Environmental Protection Agency, primarily in the Office of Policy, Planning, and Evaluation. The federal government and citizen groups are attempting to increase enforcement of environmental laws and regulations. Under the major environmental laws, government agencies and other groups may sue alleged violators for injunctive relief and civil penalties. Several of these laws authorize plaintiffs to recover civil penalties that (1) deprive violators of the economic savings they may have obtained through noncompliance and (2) account for harm to human health and the environment. The Office of Enforcement of the U.S. Environmental Protection Agency (EPA) generally assumes that corporations and municipalities have reaped substantial profits as a result of their failure to install, operate properly, and maintain pollution control equipment. Within the past year, EPA and the Department of Justice (DOJ) have demonstrated their intention to impose record-setting penalties in such cases.1 The plaintiffs often use financial analysis as part of their efforts to establish monetary penalties to punish such violators and to deter others from violating environmental requirements. EPA's BEN model2 is the most frequently used analytical tool for calculation of the economic savings (or "benefit") a firm or municipality may have obtained through noncompliance. This Dialogue discusses the general context of EPA's civil penalty assessment and then examines the BEN model to determine whether it adequately estimates economic savings due to noncompliance. The most importantpart of this discussion focuses on the government's use of the cost of equity capital (a rate associated with long-term returns on common stock) as the basis for discounting cash flows and for calculating prejudgment interest.3 The use of this rate for these purposes conflicts with generally accepted financial theory and can lead to a gross overestimation of the value of a company's savings due to noncompliance. Other problems discussed in this Dialogue include the model's treatment of investment tax credits and low-interest financing, and the absence of guidance in the BEN User's Manual4 about the appropriate treatment of precompliance expenditures. The methodological deficiencies of the BEN model heavily bias the calculations against the defendants. Confronted with the results of such calculations, defendants sometimes consider paying fines that they do not feel are justified rather than incurring the large costs associated with litigation and the negative publicity that it may provoke. Important aspects of a defense may include the ability to question the integrity of the plaintiff's economic benefit calculations. To defend against unreasonable or incorrectly calculated penalties, it is crucial for defendant's counsel to understand how the BEN model works and to be able to rebut the plaintiff's penalty calculations where appropriate. Several environmental statutes authorize the federal government and the courts to require violators of environmental laws to pay fines that reflect, among other things, the amount of money they saved through noncompliance. For example, Federal Water Pollution Control Act (FWPCA) § 309(d) states: In determining the amount of a civil penalty, the court shall consider the seriousness of the violation or violations, the economic benefit (if any) resulting from the violation, any history of such violations, any good-faith efforts to comply with the applicable requirements, the economic impact of the penalty on the violator, and such other matters as justice may require.5 Although § 309(d) lists each of these factors as "considerations," [21 ELR 10247] in settlement negotiations, EPA and DOJ view them as additive. Citizen groups may also sue alleged violators for civil penalties6 under Clean Air Act § 304, FWPCA § 505, and Resource Conservation and Recovery Act (RCRA) § 7002. Citizen plaintiffs rely on EPA's civil penalty policies in settlement negotiations. Although courts have consistently stated that penalties assessed pursuant to federal environmental laws must be paid to the U.S. Treasury,7 this has not always been the practice.8 To promote settlements, EPA and DOJ attorneys (and their citizen suit counterparts) take three factors into account in negotiations over the appropriate civil penalty: the recalcitrant behavior of the violator, the gravity (i.e., seriousness) of the violations, and the economic benefit that may have been obtained through noncompliance. Of the three issues, recalcitrance is the one that is most obviously subjective and arbitrary, since it is not supported by any detailed framework of analysis. It is also frequently the first to drop away in settlement negotiations. Through EPA's issuance of general policy statements on civil penalties9 and statute-specific penalty policies,10 EPA has attempted to provide a basis for placing a monetary value on the gravity component. However, the relevant methodology is often subjective. For example, under the FWPCA civil penalty policy, a monetary value is placed on gravity through the use of a scoring system in which each point adds $ 1,000 to the penalty. Points are awarded based on four seemingly plausible criteria: the significance of the violation, harm to health and the environment, the number of violations, and the duration of noncompliance. For each month in which a violation occurred, one additional point is added. Given the subjective nature of some of the criteria, two entities in identical circumstances may receive quite different assessments. The third issue, and focus of this Dialogue, is the economic benefit of noncompliance. EPA guidance states that the "existence and extent of the economic benefit is a factual matter which may be objectively measured in dollar terms."11 The BEN model is designed to compute those benefits. However, in litigation, EPA may rely on the model, or it may hire an "independent" economic expert. Financial analyses provided by such experts, "while consistent with the principles of the BEN model, may not necessarily be identical to that set forth in the BEN User's Manual."12 According to the FWPCA civil penalty policy, in developing an initial penalty amount, government analysts should add the economic benefit calculated by the BEN model to the figure developed for gravity.13 An additional amount of money, up to 150 percent of this subtotal, is added for recalcitrance. (EPA policy allows enforcement attorneys to increase this factor during the period of the negotiations.) The resulting subtotal should then be adjusted downward for litigation considerations such as equities that favor the defendant, the difficulty the government may face in succeeding in the litigation, whether the defendant relied to its detriment on representations made by a federal or state governmental agency, and other factors. Further reductions may be made for other reasons, including the defendant's agreement to undertake a "mitigation project." Regardless of the monetary value of any proposed mitigation projects, it is EPA policy to never settle a case for less than the economic savings obtained through noncompliance. Because many mitigation projects may provide benefits to the defendants, EPA reserves the right to allow [21 ELR 10248] less than a dollar-for-dollar reduction for mitigation in any calculation of the value of the gravity component and recalcitrance. EPA's civil penalty policies do not prevent the Agency from arguing in court for the maxiumum civil penalty authorized by a statute. EPA guidance states that: [G]overnment litigators shall not argue before a judge or neutral decision-maker for a civil penalty based upon the specific methodology set out in the CWA penalty policy, nor should they offer evidence, including expert testimony, as to how specific CWA penalty policy gravity component calculations apply to a given case.14 argue in litigation for the highest civil penalty appropriate under the law, considering the applicable statutory factors, our ability to prove the allegations in the complaint, and whatever financial burdens may be placed upon the government by the continuing litigation.15 Under the FWPCA § 309(d), maximum statutory penalties are set at $ 10,000 per day of violation through early February 1987 and at $ 25,000 per day thereafter. Under RCRA § 3008(g) and Clean Air Act § 113(b), maximum statutory penalties are set at $ 25,000 per day. In summary, EPA's penalty policies are designed to provide alleged violators a strong incentive to settle rather than litigate, and BEN model calculations are treated as the objective component of many penalty calculations. Indeed, lawyers for EPA, DOJ, and citizen groups frequently describe the BEN model in just this manner. However, the model has never been subject to public review and comment,16 and various aspects of its methodology are unsound. Although EPA and DOJ officials may recognize the unreasonableness of certain aspects of the BEN methodology, they generally avoid discussions of this subject on the basis that, as a matter of Agency policy, they do not have the authority to deviate from the standard methodology of the BEN model and the BEN User's Manual. Attorneys for citizen groups have also attempted to apply the BEN methodology as if they lacked the legal authority to deviate from it. Although enforcement plaintiffs claim large amounts of money for gravity and recalcitrance, they are typically unwilling to provide analytic support for these numbers. Instead, they focus settlement discussions on the costs of achieving compliance that are critical to their calculations of the economic benefits. The BEN model focuses on two types of economic savings: those that a firm may have obtained by delaying capital expenditures necessary for environmental compliance, and those that may have resulted from not operating and not maintaining pollution control equipment during the period of noncompliance.17 The model does not attempt to quantify the economic benefit that a firm may have obtained by improving its competitive position due to obtaining these savings.18 To estimate the total economic savings of noncompliance, the model calculates the costs that the firm would have incurred assuming "on-time" compliance, and subtracts from that value the costs that result from "delayed" (or actual) compliance.19 These costs include the capital investment to purchase the required apparatus, the cost of operating and maintaining the equipment throughout its useful life, and the present value of the cost of replacing and operating the equipment into the future. To perform the necessary computations, the BEN model requires at least the following inputs: the dates of noncompliance, compliance, and penalty payment; avoided costs, such as operations and maintenance expenses, that would have been incurred had the investment been made "on time" (i.e., prior to the onset of noncompliance); and delayed costs, such as capital investment(s) and any one-time expenditures (for example, establishing a computerized monitoring system). Model inputs may also include the useful life of the pollution control equipment, an inflation rate, a discount rate, and information on debt financing (e.g., a corporate debt rate and/or a rate for low-interest financing). If values are not specified for these inputs, the model applies certain standard (or default) values and assumes that low-interest financing was not available. Before calculating the economic benefits of noncompliance, the model accounts for investment tax credits and corporate taxation, and restates each cash flow on an after-tax basis. The adjustment for taxation relies on the use of [21 ELR 10249] different tax rates before and after 1986, when Congress enacted sweeping tax reform legislation. To account for the differences in timing of the various cash flows, the model first discounts all the cash flows back to the time when compliance was originally required (i.e., the date of noncompliance) to determine the total value of each set of cash flows (i.e., the set for the "on-time" case and the set for the "delay" case) as of that date. The difference in the discounted present values is said to be the economic benefit of noncompliance as viewed from the date of noncompliance. This difference is brought forward to the penalty payment date by using an interest factor, which adjusts the past value to its present value. Appendix 2 provides an example of a BEN model calculation. The lower portion of the example shows the input assumptions. The upper portion displays the results of the calculation. Such information is frequently provided to defendant's counsel in the course of settlement negotiations. The Uses of the Equity Cost of Capital According to the BEN User's Manual: To calculate the economic benefit of delay as of the noncompliance date, BEN uses the equity cost of capital to discount the relevant cash flows. The equity cost of capital represents the return a company needs to earn to compensate stockholders for the risk associated with its equity securities. BEN also uses the equity cost of capital to bring the initial economic benefit forward to the penalty payment date.20 The BEN User's Manual does not provide any further justification for the use of the cost of equity capital for these purposes.21 The government's apparent goal is to deprive the alleged violator of both the savings due to noncompliance and the return on the funds that might have been earned if they had been invested in common stock or some other investment opportunity open to the company. The Appropriate Rate for Discounting. According to a leading text on corporate finance, "[e]ach project should be evaluated at its own opportunity cost of capital …. The true cost of capital depends on the use to which the capital is put."22 The clear implication is that the net present value of a project (that is, its discounted future value) does not change based on who owns it or how it was financed. As stated in an expert report prepared on behalf of a defendant in an environmental civil penalty case: The fundamental principle for discount rate and [prejudgment] interest rate selection is that the rate used to bring a cash flow back or forward to the present must reflect the riskiness of the cash flow in question. The risk premium has nothing to do with who supplied the money. Both the riskiest electronics firm and the safest utility invest at the same rate when buying the same bond, because that rate depends on the bond's risk, not the risk of the company who buys it.23 Textbooks and articles on corporate finance typically advocate the use of the weighted average cost of capital (WACC, i.e., a rate based on the firm's mix of stock and debt financing) as the appropriate discount rate for the "average" project a firm undertakes. However, as Brealey and Myers warn: [This] formula gives the right discount rate for projects that are just like the firm undertaking them. The formula is incorrect for projects that are safer or riskier than the average of the firm's existing assets.24 While it is difficult to assess the "riskiness" of an investment in pollution control (which is not usually undertaken to increase a firm's revenues), such an investment is likely to be no more risky than the average for-profit investment undertaken by the firm. In fact, such a project may well be less risky because its cash flows involve only cost streams, part of which are fixed and do not vary with production levels. The cash flows resulting from profit-oriented investments are typically more variable because they include both cost and revenue streams. Ideally, an individual risk assessment would be made to derive the appropriate discount rate for any given pollution control project. Where this is technically or practically infeasible, the firm's WACC would constitute a reasonable upper limit for the discount rate. There is no support in the main stream of the finance literature for discounting anticipated project cash flows solely on the basis of the cost of equity capital when a corporation is financed with both debt and equity.25 Use of the WACC is consistent with modern finance theory and practice.26 The Appropriate Rate for Calculating Prejudgment Interest. Relevant federal environmental statutes are silent on the question of prejudgment interest — how to adjust the calculated economic savings from the date of noncompliance to the date of assumed or actual penalty payment. However, this issue has been extensively explored in other contexts. At least two careful and authoritative articles have been written on the subject. In a typical damages case, it might seem reasonable at first blush to compensate the plaintiff for the returnsit failed to earn because certain assets were destroyed or damaged by the defendant. However, Franklin M. Fisher and R. Craig Romaine argue: The fallacy here … has to do with risk. The plaintiff's opportunity cost of capital includes a return that compensates the plaintiff for the average risk it bears. But, in depriving the plaintiff of an asset worth Y at time O, the defendant also relieved it of the risks associated with investment in that asset. The plaintiff is thus entitled to interest compensating it for the time value of money, but it is not also entitled to compensation for the risks it did not bear. Hence prejudgment interest should be awarded at the risk-free rate ….27 Later in the same article, Fisher and Romaine argue that the risk-free rate should be adjusted to reflect appropriate tax treatment. Thus, they conclude that it is appropriate to use the after-tax risk-free rate to adjust damages for prejudgment interest.28 The same reasoning applies in the context of the BEN model.29 The second article was written in 1982 by Drs. James M. Patell, Roman L. Weil, and Mark A. Wolfson. According to this article, prejudgment interest should be calculated based on the defendant's debt rate.30 The idea here is that the plaintiff in a damages case is entitled to compensation for the loss of interest calculated at the risk-free rate plus a premium for the "nonzero probability of default by the defendant."31 Of course, the defendant corporation's debt rate is usually substantially lower than its cost of equity capital. For many corporations, this debt rate is close to the risk-free rate associated with Treasury bills. Further support for the use of the risk-free rate in calculations of prejudgment interest may be found in Judge A. David Mazzone's recent decision in Polaroid Corp. v. Eastman Kodak Co., a landmark patent infringement case. Judge Mazzone wrote: After consideration of all the factors and having in mind that the goal to be achieved is fair and adequate compensation, I conclude that an award of prejudgment interest should be based on the Treasury bill rate for the period [from the first date of infringement until the date of the judgment], compounded annually.32 This discussion shows the sharp contrast between what several prominent academics and one judge have had to say about the appropriate treatment of prejudgment interest and the related practice on which EPA relies in BEN model calculations. As discussed in Appendix 2, in revising the BEN User's Manual, EPA recently increased its standard value for the cost of equity capital from 17.5 to 18.1 percent. By comparison, the after-tax risk-free rate associated with three-month Treasury bills issued between January 1980 and July 1990 was only about 5 percent compounded annually. The Sensitivity of Assumed Rates for Discounting and Prejudgment Interest. In economic benefit calculations involving BEN, the results are extremely sensitive to the rate assumed for calculating prejudgment interest and less sensitive to the selection of the discount rate.33 As noted above, between January 1980 and July 1990, the after-tax risk-free rate was close to 5 percent per annum. The difference between compounding calculated economic savings at approximately 5 percent and at 18 percent is obvious, but a numerical example may highlight the importance of this issue. If a firm saved $ 1 million as of January 1, 1980, and if on July 1, 1990, it had to repay the principal with interest compounded at 18 percent (assuming monthly compounding), it would pay approximately $ 6.5 million. If the interest rate had been 5 percent, it would have had to pay approximately $ 1.7 million. As shown in this example, the rate selected for adjusting for prejudgment interest may have a direct and substantial bearing on the outcome of an economic benefit calculation. Treatment of These Issues in Negotiations and Litigation with EPA. EPA and DOJ attorneys generally do not agree that using the cost of equity capital for discounting and for calculating prejudgment interest is unreasonable.34 Instead, they usually maintain that, as a matter of policy, the use of the equity cost of capital is correct and not open to discussion. However, EPA and DOJ attorneys seem to acknowledge implicitly that there is a high level of litigation risk associated with this aspect of the methodology.35 As discussed in court testimony in Louisiana-Pacific36 and Powell Duffryn Terminals, Inc.,37 and in deposition testimony in United States v. Menominee Paper Co.,38 it is clear that some economists believe that the equity cost of capital (whether it is a company-specific rate or some general standard value) is too high a rate to use for discounting cash flows related to pollution control equipment and for computing prejudgment interest in these cases. Unfortunately for current defendants in civil penalty cases, the cases in which this issue was litigated were decided on other grounds.39 The appropriateness of the government's methodology waits for judicial consideration in some other case. As stated above, there are good theoretical reasons for believing that the use of the cost of equity capital in the BEN methodology is incorrect. Yet the BEN methodology has not been subject to formal rulemaking, and EPA has provided no evidence that its methodology would receive substantial support among financial theorists. Many civil penalty cases involve firms that have at least tried to comply with environmental requirements. In settlement negotiations, however, government attorneys have been reluctant to allow credits for capital expenditures that failed to lead to full compliance. That does not seem reasonable. If a firm saved money by not making an investment in a particular piece of hardware but spent some or all of the money on another investment in pollution control, it is clear that the firm did not obtain the full benefit of delayed compliance. By failing to discuss this issue directly, the BEN User's Manual allows government attorneys to argue that the economic benefit component of the civil penalty policy excludes the granting of credit for unsuccessful capital investments. Such a policy conflicts with the instructions Congress provided to EPA in Clean Air Act § 120.40 Such a policy also seems unduly punitive, particularly when a company has made good-faith efforts to reduce or eliminate pollution problems. To provide correct tax treatment, the BEN model provides a 10 percent investment tax credit (ITC) for capital investments that were or should have been made before the end of 1985. The BEN model does not provide ITC for investments made after that date. However, under the 1986 tax reform legislation, the full amount of ITC was still available for projects satisfying certain transition rules until June 30, 1987. After that time, qualifying transition property would be subject to a 35 percent reduction in ITC. According to the transition rules, investments qualify for ITC if the property was constructed, reconstructed, or acquired under a written contract binding on December 31, 1985; if the construction was begun by that date and the lesser of $ 1 million or 5 percent of the cost had been incurred; or if by that date the equipped building or plant facility construction was begun and more than half of the costs were committed. Considering the lead time involved in ordering and installing pollution controls, it would seem reasonable to allow the BEN model to assume that the ITC was available in 1986. The BEN model provides for the use of low-interest financing of pollution control equipment as one of the variables in both the "on-time" and the "delay" cases and in replacement cycle calculations regardless of the timing of these cash flows. In enacting the 1986 tax reform legislation, Congress eliminated new issues of low-interest financing for pollution control after December 31, 1986. To model more accurately the economics of "on-time" versus "delayed" compliance, the BEN model should allow the use of low-interest financing prior to the end of 1986 if the subject firm can provide reasonable evidence that it could have obtained such financing when it was available. Except under unusual circumstances that may justify different treatment, the BEN model should not assume that if such financing was available in the on-time case, the firm would also be able to obtain it after the end of 1986. In short, to calculate fairly the economic savings due to noncompliance, the model should recognize that by not complying on time, companies may have failed to obtain the economic benefits of low-interest financing that were not available to them under delayed compliance. Correcting the treatment of low-interest financing for pollution control in the BEN model might lead to a reduction in the penalties the government can claim. However, such a correction would be consistent with the goal of trying to measure the amount of money a firm actually saved through noncompliance. The Availability of Grants for Municipalities The BEN model allows nonprofit organizations to avoid [21 ELR 10252] being penalized by the full cost of pollution control equipment if some of the investment would have been subsidized through federal or state grants. As a case in point, during the 1970s and 1980s, the federal government provided municipalities with billions of dollars in grants to offset some of the costs of sewage treatment construction. BEN assumes that the level of grant money obtained in the delay case is the same percentage of the capital costs that would have been subsidized in the on-time case. This is not always correct. The federal grant share of sewage treatment systems was originally set at 75 percent, but was reduced to 55 percent as of October 1, 1984, per FWPCA § 202(a)(1).41 By the late 1980s, this federal grant program was phased out. Thus, in some circumstances, the on-time case should assume 75 percent grant money and the delay case should assume 55 percent or possibly less. BEN is not sufficiently flexible to perform this calculation. Given EPA's civil penalty policy and the deficiencies of the BEN model, it is not surprising that the settlement process has become Byzantine. From the government's perspective, it is important to force violators to comply fully with environmental laws and to deprive them of the economic advantages obtained through noncompliance. In so doing, the government sends a loud message to potential violators. The government is clearly not required to accept as a penalty only the amount of money that a firm has saved through noncompliance. From the perspective of the defendants, it is important to correct existing problems and to alleviate the government's legitimate concerns. However, corporate executives also feel responsible for protecting shareholders' assets against what they perceive as an unjustified confiscation by the government. While they may not be eager to engage in costly litigation, they are reluctant to pay any amount of money demanded unless it is backed up by good economics and sound public policy. As in a Turkish bazaar, where both sides only slowly reveal their willingness to reach an agreement, the settlement process can be quite drawn out. The current process ties up human and financial resources that both the government and private industry can surely apply to more productive purposes. To increase the fairness of the settlement process and to avoid unnecessary litigation, the government should subject its noncompliance penalty policy, including its economic benefit methodology, to public review and comment. That the government has chosen not to do so does not mean that lawmakers or judges should ignore the deficiencies in the BEN model. By correcting the model's deficiencies and by encouraging financial economists to comment on the relevant issues as part of a public review process, EPA could put its civil penalty policy on more defensible footing. Meanwhile, it will be crucial for defendants to fully understand the weaknesses in the BEN methodology and to pursue their legitimate interests accordingly. Appendix 1 - Changes in the Revised BEN Model (BEN90) In July 1990, EPA revised the BEN model and issued a new version of the BEN User's Manual, along with a "litigation-sensitive" document entitled the BEN User's Guide. This appendix briefly highlights the differences between the revised model (BEN90) and the earlier version (BEN87). BEN90 eliminates one of the major problems that existed in the previous version of the model (BEN87) — the assumption that all capital equipment necessary for pollution control would be replaced at the end of its useful life and operated for an indefinite number of years into the future. It is obvious that identical equipment will not necessarily be installed whenever a production or pollution control technology is changed. It is even more obvious that replacement of equipment will not occur when a plant or process is closed down. Prior to the recent revision, whenever such a decision was analytically justified, it was necessary to perform offline calculations to eliminate the effects of the replacement cycle. This is no longer necessary. BEN90 allows designation of specified capital investments as nonrecurring. Although more could be said about the continued assumption of replacement cycles, this is an unqualified improvement. Depending on the facts in specific cases, this change will usually lead to lower calculations of the economic savings. The previous version of the BEN model treated corporate taxation as if the same rates applied in all states. BEN87 applied a 50 percent marginal tax rate to expenditures made prior to 1986 and a 38.5 percent rate to expenditures made thereafter. BEN90, on the other hand, provides pre-1986 and post-1986 tax rates that vary by state. This change should improve the accuracy of the calculations. However, it should be noted that in BEN90, the default values for pre- and post-1986 marginal taxes are 49.6 and 38.4 percent, respectively, which are similar to the previous standard values. Additionally, unless state tax rates were very different before and after 1987, the change in results due to this one factor will be relatively insignificant. As noted above, if the user does not specify certain inputs, BEN90 will provide standard assumptions. BEN90 assumes an inflation rate of 4.1 percent, in contrast to the 3.4 percent rate assumed in BEN87. Depending on the length of the period of noncompliance, whether the capital will be replaced, and the relative magnitude of the delayed capital and the avoided operations and maintenance costs, this change may slightly increase or slightly decrease the results of a penalty calculation. Based on what is stated in the revised BEN User's Manual, the model's treatment of depreciation has changed. BEN87 recognized that the tax shields associated with depreciation were relatively fixed and therefore not subject to a high degree of risk. Accordingly, it discounted [21 ELR 10253] ciation tax shields at only two-thirds of the after-tax discount rate. In BEN90, however, depreciation tax shields are discounted at the full cost of equity capital. This change may be expected to increase economic benefit calculations by 5 to 10 percent in some cases. Discounting Rates and Prejudgment Interest After an apparently long period of debate within the Agency, EPA's Office of Enforcement staff decided to retain the use of the cost of equity capital as the appropriate discount rate for use in BEN model calculations. In BEN87, the standard value for this important rate was 17.5 percent. In BEN90, this value has been increased to 18.1 percent. (Any similarity with the interest rates charged by Mastercard or VISA is purely coincidental!) At a 38 percent marginal tax bracket, this rate corresponds to a 29.2 percent pre-tax rate of return, which is much higher than most companies actually obtain. The revised BEN User's Manual no longer provides a clear discussion of how to select different discount rates for firms in different industries. Instead, it provides the standard formula for what financial experts will recognize as the "capital asset pricing model," a formula that most users will find difficult to apply in specific cases without further guidance. For this reason, defendants and their counsel may have difficulty knowing whether the default value for cost of equity capital is the appropriate rate to use in BEN calculations, even assuming that this methodology is correct. The only obvious change in the way the revised BEN User's Manual deals with municipalities is that it provides flexibility in the selection of the discount rate for such entities. BEN87 required the use of a specific default value for this purpose. According to page B-I-5 of the revised manual, the cost of municipal debt should be used to select the proper discount rate. Thus, an analyst need only obtain the rating for municipal bonds issued during the period of noncompliance and determine the applicable interest rates by referring to a bond yield chart included in the document. This is an improvement over the previous method of calculation. According to EPA staff, this change will lead to penalties for municipalities approximately 16 percent lower than those calculated with BEN87. As noted in the July 12, 1990, cover letter that announced the release of the new BEN User's Manual, EPA has also distributed to government officials a new, litigation-sensitive document entitled the BEN User's Guide. According to page I-2 of the Manual, the BEN User's Guide covers computer software log-on procedures and "discusses some issues that typically arise in running BEN…." Since this document "is nonreleasable in its entirety," one can only guess about its contents. It is hard to understand how a document that is not case-specific can be described as "litigation-sensitive," particularly if it provides information about only technical computer matters, such as log-on procedures. (There are many other ways to provide computer security.) This situation is inconsistent with open government and is bound to result in Freedom of Information Act requests and appeals for a review of government decisions to withhold the document. A little glasnost on this side of the Atlantic would be helpful. Appendix 2 — Example of a BEN Model Calculation
1. For example, during the 1980s, no penalty in a Federal Water Pollution Control Act (FWPCA) case exceeded $ 2 million. However, in July 1990, a $ 2.1 million settlement was reached in United States v. Menominee Paper Co., No. M88-108 (W.D. Mich.). This figure was surpassed in October 1990, when British Petroleum agreed to a $ 2.3 million settlement in United States v. British Petroleum, CA 86-0792 (E.D. Pa. consent decree entered Dec. 17, 1990.) DOJ is seeking substantially higher settlements in several ongoing cases. 2. BEN is short for "economic benefit." TCS Management Group, Inc., of Nashville, Tennessee, developed the BEN model for EPA in 1985. TCS revised the model and the BEN USER'S MANUAL in 1987 to reflect changes in federal tax law. Under EPA's supervision, in July 1990, Industrial Economics, Inc., revised the model and the manual. Also, in July 1990, EPA issued a "litigation-sensitive" document entitled the BEN USER'S GUIDE, which is "nonreleasable in its entirety." BEN USER'S MANUAL, at I-2. See infra Appendix 1 for a discussion of these documents. 3. "Discounting" is a technique used in financial analysis to adjust a stream of monetary payments or costs for the time value of money. The concept underlying this adjustment is that a dollar received today is worth more than a dollar received one year from now, and the further the payment is in the future, the less it is worth today. For example, using a 10 percent discount rate, a dollar received one year from now is only worth 91 cents today. One dollar received two years from now is worth 83 cents today. (Discount factors may be found in a "discount table" published in any introductory finance textbook.) 4. INDUSTRIAL ECONOMICS, INC., BEN USER'S MANUAL (July 1990). 5. 33 U.S.C. § 1319, ELR STAT.FWPCA 037. See also Clean Air Act § 120, 42 U.S.C. § 7420, ELR STAT.CAA 019. Several administrative penalty provisions require EPA to account for economic benefits from noncompliance in assessing civil penalties. See, e.g., Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) § 109(a)(3), 42 U.S.C. § 9609 (a)(3), ELR STAT.CERCLA 031; Clean Air Act § 113(e), 42 U.S.C. § 7413(3), ELR STAT.CAA 014; FWPCA § 309(g)(3), 33 U.S.C. § 1319(g)(3), ELR STAT.FWPCA 037; Emergency Planning and Community Right-to-Know Act § 325(b)(1)(C), 42 U.S.C. § 325(b)(1)(C), ELR STAT.EPCPA 011. The phrase "economic benefit," or similar language, does not appear in the penalty provisions of the Resource Conservation and Recovery Act (RCRA), the Toxic Substances Control Act (TSCA) or the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA). See, e.g., RCRA § 3008(a)(3), 42 U.S.C. § 6928(a)(3), ELR STAT. RCRA 021; TSCA § 16(a)(2)(B), 15 U.S.C. § 2615(a)(2)(B), ELR STAT. TSCA 026; FIFRA § 14(a)(4), 7 U.S.C. § 1361, ELR STAT. FIFRA 024. However, according to the RCRA Civil Penalty Policy (Oct. 1990), ELR ADMIN. MATERIALS 35273, it is clear that EPA uses the BEN computer model to calculate the economic benefit of noncompliance with RCRA. Id. at 35280-81. It is not clear whether the Agency uses the BEN model in cases involving violations of TSCA or FIFRA. 6. Clean Air Act § 304, 42 U.S.C. § 7604, ELR STAT.CAA 043; FWPCA § 505, 33 U.S.C. § 1365, ELR STAT. FWPCA 062; RCRA § 7002, 42 U.S.C. § 6972, ELR STAT. RCRA 034. 7. See, e.g., Gwaltney of Smithfield, Ltd. v. Chesapeake Bay Foundation, Inc., 484 U.S. 49, 53, 18 ELR 20142, 20145 (1987); Middlesex County Sewage Authority v. National Sea Clammers Association, 453 U.S. 1, 14 n.25, 11 ELR 20684, 20687 n.25 (1981); Public Interest Research Group of New Jersey, Inc. v. Powell Duffryn Terminals, Inc., 913 F.2d 64, 81, 21 ELR 21216, 21224 (3d Cir. 1990); Atlantic States Legal Foundation, Inc. v. Tysons Foods, Inc., 897 F.2d 1128, 1131 n.5, 20 ELR 20788, 20789 n.5 (11th Cir. 1990); and Sierra Club v. Chevron U.S.A., Inc., 834 F.2d 1517, 1522, 18 ELR 20237, 20240 (9th Cir. 1987). The Third Circuit, in Public Interest Research Group of New Jersey v. Powell Duffryn Terminals, Inc., noted that a court in an FWPCA case may use its equitable powers to "fashion injunctive relief requiring a defendant to pay monies into a remedial fund, if there is a nexus between the harm and the remedy. But here, once the court labeled the money as civil penalties it could only be paid into the Treasury." 913 F.2d at 82, 20 ELR at 21225. 8. For example, in February 1990, the defendant in Sierra Club v. Union Oil Company of California, No. C 84-3435 SC (N.D. Cal. consent decree entered Feb. 22, 1990) agreed to the largest settlement in an FWPCA citizen suit to date. Union Oil agreed to pay $ 4.22 million in civil penalties and $ 1.33 million for the plaintiff's legal expenses. Of the total amount assessed for civil penalties, $ 2.72 million was to be paid to the Trust for Public Land; $ 1 million to the California Department of Justice for deposit in the Ocean Site Designation Fund; and $ 500,000 to the State of California Water Pollution Control Cleanup and Abatement Account, State Water Quality Control Fund. 9. See EPA, A FRAMEWORK FOR STATUTE-SPECIFIC APPROACHES TO PENALTY ASSESSMENTS: IMPLEMENTING EPA'S POLICY ON CIVIL PENALTIES (Feb. 16, 1984), ELR ADMIN. MATERIALS 35073, 35077; EPA, POLICY ON CIVIL PENALTIES (Feb. 16, 1984), ELR ADMIN. MATERIALS 35083. 10. See, e.g., EPA, RCRA CIVIL PENALTY POLICY (Oct. 1990), ELR ADMIN. MATERIALS 35273, 35279; EPA, CLEAN WATER ACT PENALTY POLICY FOR CIVIL SETTLEMENT NEGOTIATIONS (Feb. 11, 1986), at 3-5; EPA, CLEAN AIR ACT STATIONARY SOURCE CIVIL PENALTY POLICY (Mar. 25, 1987), at 12-18; EPA, TSCA CIVIL PENALTY POLICY, 45 Fed. Reg. 59770 (Sept. 10, 1980); EPA, FINAL PENALTY POLICY FOR EPCRA (EMERGENCY PLANNING AND COMMUNITY RIGHT-TO-KNOW) §§ 302, 303, 304, 311, and 312 and CERCLA § 312 (June 13, 1990), ELR ADMIN. MATERIALS 35261, 35268; EPA, PENALTY GUIDANCE FOR VIOLATIONS OF UNDERGROUND STORAGE TANK REGULATIONS (Nov. 1990), ELR ADMIN. MATERIALS 35303, 35312. 11. Memorandum from Edward E. Reich, Deputy Assistant Administrator of EPA for Civil Enforcement, et al., Guidance on the Distinctions Among Pleading, Negotiating, and Litigating Civil Penalties for Enforcement Cases Under the Clean Water Act (Jan. 19, 1989) at 8. 12. BEN USER'S MANUAL, supra note 4, at I-1 and I-2. In United States v. Louisiana-Pacific Corp., No. 86-A-1880 (D. Colo. 1988), Dr. Katherine McElroy, the government's economic witness, provided testimony that relied on BEN model calculations. However, in several other federal cases, an expert witness for the plaintiff has relied on computerized spreadsheets compatible with the BEN methodology to perform the necessary calculations. By retaining the right to use an "independent" economic expert in court who may deviate from the BEN methodology, EPA and DOJ reserve the opportunity to claim that information concerning the reasonableness of the model or the BEN methodology is outside the scope of discovery. The sometimes articulated view is that obtaining such information "could not lead to the discovery of admissible evidence." 13. Although the penalty policies established under the different statutes vary in specific details, they typically add the value calculated for the economic benefit to the value added for gravity and then adjust that sum for other considerations. 14. Reich memorandum, supra note 11 (emphasis in original). 15. Id. at 6-7. 16. When Congress passed Clean Air Act § 120 in 1977, it required the Administrator of EPA "to promulgate regulations requiring the assessment and collection of a noncompliance penalty … after notice and opportunity for a public hearing." 42 U.S.C. § 7420, ELR STAT. CAA 019. Section 120 penalty methodology is codified at 40 C.F.R. pts. 66 and 67. Unfortunately, Congress did not require review and comment on implementation of the FWPCA penalty provisions. The § 120 model differs from the BEN model in several respects. Most importantly, it grants the alleged violator credit in penalty calculations for measures it has taken to achieve compliance regardless of whether they were successful. The BEN model does not do this. (See infra text accompanying notes 34-40 for a discussion on treatment of precompliance expenditures.) In recent years, the § 120 model has fallen into general disuse. Unlike the BEN model, the § 120 model has not been revised in light of the 1986 tax reform legislation. According to a senior EPA enforcement attorney, EPA does not intend to update the § 120 model due to budgetary reasons and EPA's desire to avoid a formal rulemaking on its penalty methodology. 17. It is not clear that the true economic advantage obtained through noncompliance is always greater than or equal to the costs associated with the delayed investment and the avoided operations and maintenance expenses. For example, if an operation was known to be unprofitable at the time of noncompliance, it may have obtained little or no economic advantage by remaining in business through its failure to comply with environmental requirements. However, the owner of such a firm might still be penalized for harming human health and the environment and for willfully disregarding the law. 18. Calculation of the competitive advantage is fraught with difficulties. It does not necessarily follow that if a firm saved money through noncompliance, it was able to expand its market share. 19. The BEN model intersperses information that is supposedly known prior to the date of noncompliance (such as the firm's expected cost of equity capital) with information that is only known afterwards (such as changes in tax treatment). It is thus difficult to know whether the BEN model is supposed to calculate the firm's actual or its expected savings (both adjusted for prejudgment interest). 20. BEN USER'S MANUAL, supra note 4, at II-29. The adjustment of the calculated savings for "interest forward" is essentially compounding for prejudgment interest. 21. The BEN methodology for selection of the discount rate (i.e., the equity cost of capital) is based on returns that were obtained in the stock market from 1926 to 1989, not the returns actually obtained during the period of noncompliance. Id. at III-30. The model assumes that all investments in pollution control are financed through the sale of corporate bonds financed at the corporate debt rate except when low-interest financing is involved. Id. at III-32. Despite this exclusive emphasis on debt financing, the BEN model discounts all cash flows and adjusts discounted values for prejudgment interest on the basis of the equity cost of capital. The reasoning behind these assumptions is not self-evident. It is also not clear that these assumptions are mutually compatible. 22. R. BREALEY & S. MYERS, PRINCIPLES OF CORPORATE FINANCE 173 (3d ed. 1988) (emphasis in original). 23. J. Speyer & A. Kolbe, Economic Analysis Related to the Powell Duffryn Facility in Bayonne, New Jersey, at 2 (Aug. 15, 1988). This report was prepared for use in Public Interest Research Group v. Powell Duffryn Terminals, Inc., 720 F. Supp. 1158, 20 ELR 20152 (D.N.J. 1989), aff'd in part 913 F.2d 64, 20 ELR 21216 (3d Cir. 1990). 24. R. BREALEY & S. MYERS, supra note 22, at 451. 25. None of the three leading graduate textbooks on corporate finance recommend discounting at the equity rate. R. BREALEY & S. MYERS, supra note 22; T. COPELAND & J. WESTON, FINANCIAL THEORY AND CORPORATE POLICY (3d ed. 1988); and E. BRIGHAM & L. GAPENSKI, INTERMEDIATE FINANCIAL MANAGEMENT (3d ed. 1990). 26. As noted in footnote 19, the BEN methodology intersperses information that is known as of the date of noncompliance with information that is only known later through hindsight. An alternative methodology would use ex post information (i.e., hindsight) for dealing with savings in the past and ex ante information (i.e., what is believed today about the future) for calculating savings from the future. In this methodology, past cash flows would be adjusted to the present by compounding at an appropriate risk-free rate and future cash flows would be adjusted to a present value through discounting at an appropriate risk-adjusted rate. 27. Fisher & Romaine, Janis Joplin's Yearbook and the Theory of Damages, 5 J. OF ACCT., AUDITING & FIN. 146 (Winter 1990). 28. Id. at 148-149. 29. Had a firm's economic savings due to noncompliance been invested in a risk-free instrument such as short-term Treasury bills, the firm would have earned interest on which it would have had to pay tax. Therefore, to deprive the alleged violator of the calculated gains due to noncompliance, it is necessary to take away from the violator the principal amount of the savings and the after-tax interest. This is the argument for use of the after-tax risk-free rate for adjusting for prejudgment interest in civil penalty calculations. 30. Patell, Weil & Wolfson, Accumulating Damages in Litigation: the Roles of Uncertainty and Interest Rates, II J. LEGAL STUD. 362 (June 1982). 31. Id. at 343. 32. Polaroid Corp. v. Eastman Kodak Co., No. 76-1634-MA, 16 U.S.P.Q. 2d 1481, 1541 (D. Mass. 1990). 33. The prejudgment interest rate has a direct and clear-cut effect on the calculation of the penalty. However, the end result of a change in the discount rate is less obvious. An increase in the rate of prejudgment interest will always increase the magnitude of the penalty. But an increase in the discount rate could either increase or decrease the magnitude of the penalty depending on the relative magnitudes of the capital investment and the annual operations and maintenance expenses, the period of noncompliance and the useful life of the equipment, as well as other factors. 34. In one settlement negotiation, an attorney for EPA argued that even if the BEN methodology is wrong, to change it for one firm would be to treat other corporations unfairly. In another case, an EPA attorney suggested averaging the plaintiff's and the defendant's proposed rates for prejudgment interest. 35. According to Thomas Echikson of the law firm Sidley & Austin, courts dealing with penalty calculations in citizen suits have rejected the use of theoretical interest rates such as those suggested by EPA. See Student Public Interest Research Group of N.J. v. Hercules, Inc., 19 ELR 20902 (D.N.J. Apr. 6, 1989) (rejecting the use of the BEN model and speculative interest rates to compute economic benefit); Proffitt v. Lower Bucks County Joint Municipal Authority, No. 86-7220 (E.D.Pa. May 12, 1988) (court adopts interest rates "prevalent at the relevant time rather than simply relying on standard rates fixed in a computer program"); Student Public Interest Research Group of N.J. v. Monsanto Co., 18 ELR 20999, 21005 (D.N.J. Mar. 24, 1988) (rejecting use of interest rates that "were demonstrated to be inaccurate or at least in serious doubt"); State v. Dayton Malleable, Inc., 10 ELR 20677, 20679 (Ohio Ct. C.P., Vol. 10, Oct. 10, 1979) (in calculating the economic benefit of noncompliance, the "Court believes that DMI's actual rate of return during this critical period should be the measure, not some outside standard"). Mr. Echikson could not find any citizen suits or FWPCA cases accepting the EPA/DOJ position concerning prejudgment interest rates. 36. In United States v. Louisiana-Pacific Corp., No. 86-A-1880 (D. Colo. 1988), James M. Speyer and Dr. A. Lawrence Kolbe testified on behalf of the defendant. 37. Mr. Speyer and Dr. Kolbe also testified in Public Interest Research Group of New Jersey v. Powell Duffryn Terminals, Inc., 720 F. Supp. 1158, 20 ELR 21052 (D.N.J. 1989), aff'd in part 913 F.2d 64, 20 ELR 21216 (3d Cir. 1990). 38. In United States v. Menominee Paper Co., No. M88-108C2 (W.D. Mich.), Robert H. Fuhrman provided deposition testimony on April 23, 1990. Stewart C. Myers's deposition was taken on May 10, 1990. (Dr. Myers is the Billard Professor of Finance at the Sloan School of Management at the Massachusetts Institute of Technology, as well as co-author of PRINCIPLES OF CORPORATE FINANCE, supra note 22, and past president of the American Finance Association.) 39. See supra cases cited in notes 36 and 37. 40. See supra note 16. 41. 33 U.S.C. 1282(a)(1), ELR STAT. FWPCA 014. In some states, the reduction occurred earlier with the concurrence of the governor and the EPA Administrator. 21 ELR 10246 | Environmental Law Reporter | copyright © 1991 | All rights reserved |