Reputation Penalties for Corporations and the Federal Sentencing Guidelines REPUTATIONAL PENALTIES FOR CORPORATIONS AND THE FEDERAL SENTENCING GUIDELINES Dennis J. Recca* I. Introduction ............................................................... 879 II. Background: Corporate Sentencing Under the G uidelines ................................................................. 881 III. The Optim al Penalty ................................................. 883 IV. Over-Deterrence and Under-Deterrence .................. 884 V. M arket-Based Penalties ............................................ 888 VI. Potential Solutions .................................................... 893 V II. Conclusion ................................................................. 897 I. INTRODUCTION Corporations that commit crimes are subject to a variety of penalties. First, in many cases, a corporation is subject to civil penalties.1 An injured party may sue a corporation for damages, which may be in the form of either compensatory damages, intended to compensate the injured party for its loss, or punitive damages, which are intended to punish the corporation and act as a deterrent.2 Second, a corporation that commits an offense can generally expect to suffer reputational, or "marketplace" losses.' That is, consumers may be less likely to do business with offending corporations, either because they have been harmed by the corporation in the past, causing them to be wary, or because of a general desire among consumers not to transact business with an offending corporation based on ethical reasons or a desire to J.D. Candidate 2005, Columbia University School of Law; A.B. Economics 2002, Dartmouth College. 1 See Mark A. Cohen, Empirical Trends in Corporate Crime and Punishment, 3 FED. SENT. REP. 121, 122 (1990). 2 See BMW of N. Am. v. Gore, 517 U.S. 559, 568 (1996). ' See Cohen, supra note 1, at 123. COLUMBIA BUSINESS LAW REVIEW avoid a stigma attached to the corporation.4 This reaction by consumers may result in the loss of corporate revenue and profits. Finally, a corporation, as a legal person, is subject to the criminal law.5 Although certain criminal penalties, such as incarceration, are not applicable to corporate defendants, corporations are subject to criminal fines, probation, and other criminal sanctions.6 In recent years, Congress has greatly expanded the breadth of criminal offenses of which a corporation may be convicted, as well as the penalties attaching thereto.7 These disparate sanctions must be coordinated to arrive at the appropriate penalty if the socially optimal level of deterrence is to occur.' In the federal system, the form of criminal sanction and the precise amount of a fine, if any, that will be imposed on a corporation are determined by application of Chapter Eight of the Federal Sentencing Guidelines ("the Guidelines").9 Despite the presence of substantial civil and market-based sanctions in many cases, the Guidelines do not account for these additional penalties in determining an appropriate criminal sanction, aside from a brief policy statement that suggests that sentencing judges may, in their discretion, choose a fine, within a mandated range, that reflects "any collateral consequences of conviction, including civil obligations."" Substantial empirical evidence demonstrates 4 See Stephen A. Saltzburg, The Control of Criminal Conduct in Organizations, 71 B.U. L. REV. 421, 431 (1991). ' See U.S. DEPARTMENT OF JUSTICE MEMORANDUM REGARDING FEDERAL PROSECUTION OF CORPORATIONS, available at http://www.usdoj.gov/ criminal/fraud/policy/Chargingcorps.html. 6 See Shayne Kennedy, Note, Probation and the Failure to Optimally Deter Corporate Misconduct, 71 S. CAL. L. REV. 1075, 1079-83 (1998). 7 See Vikramaditya S. Khanna, Corporate Crime Legislation: A Political Economy Analysis, 82 WASH. U. L.Q. 95, 95-96 (2004). ' See Jeffrey S. Parker, The Economics of Mens Rea, 79 VA. L. REV. 741, 756 (1993). See Kennedy, supra note 6, at 1077. 'o See U.S. SENTENCING COMM'N, FEDERAL SENTENCING GUIDELINES MANUAL [hereinafter SENTENCING GUIDELINES], § 8C2.8(a)(3) (2003). [Vol. 2004 No. 3:879] REPUTATIONAL PENALTIES FOR CORPORATIONS that civil and market-based sanctions are not only significant but, in many cases, may be greater than the criminal sanction imposed on a corporation.1' The upshot of this is that a system of corporate sentencing guidelines that ignores civil and market-based sanctions in determining the criminal penalty will often fail to produce the socially optimal sanction, which will result in social costs of over-deterrence and under-deterrence. This Note focuses particularly on the effects of reputational, or market-based, penalties on corporations. In Part II, I provide a brief background on how the Guidelines operate to sentence corporate defendants. In Part III, I address the issue of the optimal penalty. In Part IV, I consider the problems of over-deterrence and under- deterrence, and the social costs of failing to apply the optimal penalty. In Part V, I discuss the reasons why corporations suffer market-based penalties for their crimes, and I explore a number of factors that affect their magnitude in specific cases. Finally, in Part VI, I suggest possible ways in which the Guidelines may be amended to account for these reputational effects. II. BACKGROUND: CORPORATE SENTENCING UNDER THE GUIDELINES Pursuant to the powers granted to it under the Sentencing Reform Act of 1984, the Federal Sentencing Commission, after concluding that criminal sanctions for corporations were too low, drafted Chapter Eight of the Guidelines.12 This chapter deals with the sentencing of organizational defendants, and its primary target is the publicly-held corporation." Since Congress did not act to See Cohen, supra note 1, at 123. 12 See Cindy R. Alexander et al., Regulating Corporate Criminal Sanctions: Federal Guidelines and the Sentencing of Public Firms, 42 J.L. & ECON. 393, 395 (1999). "' See Michael K. Block, Guest Editor's Observations, 3 FED. SENT. REP. 115 (1990). COLUMBIA BUSINESS IA W RE WE W [Vol. 2004 reject it, Chapter Eight became effective on November 1, 1991.14 Chapter Eight contemplates fines as the primary form of sanction for corporate defendants, but also provides for other penalties, including probation. 5 Under the Guidelines, the determination of the corporate fine begins with a minimum fine, which corresponds to the statutory offense for which the corporation was convicted. The "base fine" is then determined by taking the greatest of (1) this minimum fine, (2) the pecuniary gain to the organization, or (3) the pecuniary loss caused by the organization." A "culpability score" is then determined based on factors that are intended to reflect the organization's overall level of culpability. 7 The base fine is then multiplied by a multiplier corresponding to the culpability score.18 The result is a range of fines. 9 It is then within the sentencing judge's discretion to impose the precise fine he feels is appropriate within the mandated range. Section 8C2.8 includes a policy statement listing appropriate factors to be considered by a judge in setting a fine within the applicable range. One factor listed is "any collateral consequences of conviction, including civil obligations arising from the organization's conduct."" Finally, a sentencing judge has the authority to depart from the fine range and impose a fine that he feels is appropriate if "there exists an aggravating or mitigating circumstance of a kind, or to a degree, not adequately taken into consideration by the Sentencing Commission in formulating the guidelines that should result in a sentence different from that described."" Section 8C4 of the 14 See Jed S. Rakoff, The Corporation as Policeman: At What Price?, 207 N.Y.L.J. No. 6, at 3 (Jan. 9, 1992). " See SENTENCING GUIDELINES, supra note 10, at § 8. 16 See id. at § 8C2.4. 17 See id. at § 8C2.5. 18 See id. See id. at § 8C2.6. 21 See id. at § 8C2.8(3). 1 18 U.S.C. § 3553(b)(1) (2000). [Vol. 2004COLUMBIA BUSINESS LA W REVIEW No. 3:879] REPUTATIONAL PENALTIES FOR CORPORATIONS 883 Guidelines lists a number of factors that, according to the Commission, may serve as legitimate grounds for departure in the corporate context. This list is, however, not exhaustive, and does not preclude judges from departing in other cases in which the judge believes that a factor is not adequately taken into consideration by the Guidelines. III. THE OPTIMAL PENALTY Setting an optimal penalty for a given offense is a complicated task. Sanctions must be set at a level that induces the offender to internalize the total costs of his crimes.22 A potential offender, acting rationally, will compare the private gain which is expected to flow from the offense with the private loss that is expected to occur if the offender is caught, multiplied by the chances of being caught."8 The expected private loss in the corporate context includes civil and criminal sanctions and decreased profits resulting from reputational effects. If commission of the offense yields an expected net gain ex ante, the rational offender will commit the crime, while in the case of an expected net loss, the would-be criminal will be deterred. This leads to the natural result that, ceteris paribus, increased sanctions will lead to fewer crimes being committed, as fewer opportunities for net gains from commission of the crime will exist, and conversely, decreased sanctions will lead to an increased rate of commission. In order to induce the rational offender to internalize the total social costs of crime, the expected private loss to the criminal from committing the offense must be made equal to 22 See Bruce H. Kobayashi, Antitrust, Agency, and Amnesty: An Economic Analysis of the Criminal Enforcement of the Antitrust Laws Against Corporations, 69 GEO. WASH. L. REv. 715, 731 (2001). 23 See ANDREW ADSWORTH, PRINCIPLED SENTENCING: READINGS ON THEORY AND POLICY, at 45-46 (Andrew von Hirsch & Andrew Ashworth eds., 2d ed. 1998). COLUMBIA BUSINESS LA W REVIEW the expected loss to society." If a crime is certain to be detected, the optimal penalty will equal the harm caused by the offense. 5 Decreased probabilities of detection must be compensated by proportional increases in the total sanction. 6 Under this scheme, the rational criminal will be induced to internalize the costs of his crime. If the expected gain from commission is less than the expected total harm to society (measured by the expected private sanction), the crime will not be committed, whereas if the expected gain to the criminal is greater than the estimated total harm, the crime will be committed. This is exactly the result that should be desired. That is, socially detrimental conduct will be avoided, while socially beneficial conduct will not be discouraged. However, if the total penalty imposed on a corporation is not equal to the optimal penalty, problems of over-deterrence and under-deterrence occur.27 IV. OVER-DETERRENCE AND UNDER-DETERRENCE Between over-deterrence and under-deterrence, the problem of under-deterrence is more obvious. That is, if penalties are set too low, corporations will not fully internalize the costs of their crimes. This will result in social losses, as too many crimes will be committed, or more precisely, some crimes with a net social loss will not be deterred. The problem of over-deterrence is less obvious. In fact, the casual observer might remark that all crime is socially detrimental-is the goal of the criminal law not to eliminate all crime? This ignores several important considerations, 24 See Jennifer Arlen & Reinier Kraakman, Controlling Corporate Misconduct: An Analysis of Corporate Liability Regimes, 72 N.Y.U. L. REV. 687, 709-10 (1997). 2' This is not exactly true. To be precise, the costs of enforcement must also be considered. See Kennedy, supra note 6, at 1086. 26 See Kobayashi, supra note 22, at 732. 27 See Parker, supra note 8, at 757-58. [Vol. 2004 No. 3:879] REPUTATIONAL PENALTIES FOR CORPORATIONS including the possibility of socially beneficial crime, as well as the possibility that socially efficient legal behavior will be deterred because individuals (or corporate managers) fear that their conduct will be misinterpreted by the legal system as illegal conduct. Especially relevant in the corporate context is the possibility that excess costs will be expended to internally detect and prevent potentially criminal conduct. Over-deterrence occurs when socially beneficial behavior is deterred. This occurs when the total benefits of the deterred crime would have been greater than its costs.28 By way of illustration, let us consider the following hypothetical involving fines for illegal parking. Parking regulations and the mechanisms that enforce them are intended to induce vehicle operators to internalize the costs of parking. If the social costs of parking in front of a fire hydrant were, say, $50,29 but there were no laws prohibiting this activity (or the laws were not enforced), a driver who wished to park in front of a hydrant could choose to free ride on the public by not considering the social costs of his actions. If the use of the space were worth, say, $10 to him, he would choose to park in front of the hydrant, even though this would result in a net social loss of $40 (his $10 private gain minus the $50 public loss). In order to prevent this loss, the city might enact an ordinance prohibiting parking in front of fire hydrants. If the chances of being discovered and ticketed were estimated at, say, fifty percent, the optimal fine would be $100. Returning to our example, the estimated cost to the driver of violating the ordinance would now be $50. His private gain being $10, he would be faced with a net loss of $40 and would therefore choose not to park illegally. The $40 net social loss would be avoided. Notice that in this case, the driver's personal loss is equal to the net social loss. This is not a coincidence, but the result of an optimal penalty. 2' See Kobayashi, supra note 22, at 732. This figure might be calculated by the estimated increase in damage that a fire could be expected to cause due to the hindrance multiplied by the probability of a fire occurring within a given time. COLUMBIA BUSINESS LAW RE WE W [Vol. 2004 Now suppose that the city council, in an attempt to raise revenues, increases the fine for the same violation to $500. And now suppose our driver has a more valuable use for the parking space, making it now worth $100 to him.3" Under this scenario, the driver would choose not to park illegally (his expected personal gain is $100, but his expected personal loss is $250). However, his decision to avoid the possible sanction is socially inefficient because an opportunity to achieve a net social gain of $50 (our driver's $100 personal gain minus the $50 public loss) was eliminated. If the fine had remained at the socially optimal level, the driver would have chosen to park, and the social gain would have been achieved.3' The above hypothetical demonstrates one form of the social costs of over-deterrence by inappropriately high sanctions. However, in this situation, further costs will result from the impreciseness of the legal system. Since the legal system is not always correct in distinguishing legal from illegal behavior, a corporation might refrain from socially efficient legal behavior because of the possibility of legal error. Although this problem is always present in human institutions, it will be unnecessarily exacerbated if sanctions are increased beyond their appropriate levels, causing further losses in efficiency. In the corporate context, additional losses in efficiency will occur due to agency costs. Corporations themselves, unlike individuals, cannot commit criminal acts because they cannot possess the requisite mens rea.12 Instead, corporate criminal liability is based on vicarious liability for the acts of its officers and employees under the doctrine of respondeat 30 The other elements of our hypothetical remain constant, including the social costs of parking in front of a hydrant and the probability of detection. 31 In this hypothetical, we ignore for simplicity any indirect social gains from parking ticket revenues. To tighten the logic, we might assume that the revenues in both cases were just sufficient to fund enforcement. 32 See Daniel R. Fischel & Alan 0. Sykes, Corporate Crime, 25 J. LEGAL STUD. 319, 320 (1996). COLUMBIA BUSINESS LA W REVIEW [Vol. 2004 No. 3:879] REPUTATIONAL PENALTIES FOR CORPORATIONS 887 superior." Therefore, a corporation may be criminally liable if it fails to control the actions of its employees. The potential costs of criminal liability for corporations are therefore a form of agency costs. Faced with these potential costs, a corporation will spend money on programs that are designed to prevent employees from committing acts for which the corporation may be held criminally liable and to detect the commission of such acts, which may entitle the corporation to a reduction in penalties." In fact, since the corporation itself cannot commit a crime, the intent of the criminal law is to provide incentives to corporations to take such steps to prevent its agents from committing crimes. If sanctions are set at a level higher than the social costs of the crime, corporations will be induced to spend excessive amounts to monitor their employees and prevent crimes." These monetary costs will result in increased prices for consumers, as the increased costs of doing business will be passed on to them.36 Additionally, these costs will put domestic firms at a comparative disadvantage vis-h-vis foreign competitors not subject to American criminal law. Finally, by raising the cost of doing business in the United States, excessive criminal sanctions will discourage foreign investment. The determination of appropriate criminal penalties is therefore a very important task. If the penalties are set too low, excessive corporate crime and its attendant ills will result. If penalties are set too high, social costs will result, which in the corporate context are exacerbated by agency costs. Determining the optimal corporate sanction is a complicated task involving economic and empirical analysis. However, in setting the level of fines, there is no evidence that the Sentencing Commission considered the potential 3 See Kobayashi, supra note 22, at 735. ' See John C. Coffee, "Carrot and Stick" Sentencing: Structuring Incentives for Organizational Defendants, 3 FED. SENT. REP. 126 (1990). " See Kobayashi, supra note 22, at 735-36. 31 See Kennedy, supra note 6, at 1087. COLUMBIA BUSINESS L4WREWEW [Vol. 2004 penalties or undertook any such analysis. In fact, reputational penalties are not even mentioned in the Guidelines. Instead, the Commission based its fine schedule on the fines imposed by judges in pre-Guidelines cases, when judges had virtually complete discretion in setting fines.37 Even if it could be shown that the pre-Guidelines judges were, by and large, aware of the effects of reputational penalties and considered them in their assessments of the appropriate fines, the Commission had very little data on which to base its calculations. At the time Chapter Eight of the Guidelines was being drafted, there were only about 300 corporate convictions each year in federal courts." The fact that the Commission based its fines on such scant evidence is striking. Furthermore, the Commission, in drafting Chapter Eight, was acting largely under the presumption that corporate sanctions were too low.39 Empirical evidence has shown that criminal fines imposed on corporations have been substantially higher under the Guidelines than they had been previously. ° There is no reason to believe that the Commission in any way considered reputational sanctions when they determined that criminal sanctions should be increased. V. MARKET-BASED PENALTIES A substantial body of empirical evidence supports the proposition that corporations incur significant decline in reputation when news of impropriety is announced. Sam Peltzman found that publicly traded corporations suffered significant losses in goodwill after news of FTC investigations into false and misleading advertising practices " See Cohen, supra note 1, at 121. 38 Id. See Alexander et al., supra note 12, at 395. 40 Id. at 416. COLUMBIA BUSINESS LA W REVIEW [Vol. 2004 No. 3:879] REPUTATIONAL PENALTIES FOR CORPORATIONS 889 was released."1 Another study found that losses in shareholder wealth resulting from news of product recalls went beyond the direct costs of the recall.42 Mark Mitchell found that the pharmaceutical company Johnson & Johnson incurred substantial losses in goodwill following disclosure that their Tylenol medication had been tampered with in the 1980s. 3 Mitchell and Michael Maloney found that commercial airlines suffered significant losses in shareholder wealth resulting from decreased goodwill following crashes involving pilot error. Finally, Karpoff and Lott found substantial losses in shareholder wealth following revelations of corporate fraud.45 One explanation for the reputational losses experienced by corporations in these contexts is the loss of the ability of a corporation to charge a premium for quality. Consumers hold expectations about product quality, and when these expectations are not met, a reputational penalty is incurred.46 Another potential cause of reputational penalties is a stigma effect caused when consumers choose not to do business with a corporation that has committed some violation, either on ethical grounds or to avoid a social stigma that attaches to the corporation. 7 These two explanations for market-based reputational penalties 41 See Sam Peltzman, The Effects of FTC Advertising Regulation, 24 J.L. & ECON. 403 (1981). 42 See Gregg Jarrell & Sam Peltzman, The Impact of Product Recalls on the Wealth of Sellers, 93 J. POL. ECON. 512 (1985). ' See Mark L. Mitchell, The Impact of External Parties on Brand- Name Capital: The 1982 Tylenol Poisonings and Subsequent Cases, 27 ECON. INQUIRY 601 (1989). " See Mark L. Mitchell & Michael T. Maloney, Crisis in the Cockpit? The Role of Market Forces in Promoting Air Travel Safety, 32 J.L. & ECON. 329 (1989). 4 See Jonathan M. Karpoff & John R. Lott, Jr., The Reputational Penalty Firms Bear from Committing Criminal Fraud, 36 J.L. & ECON. 757, 796 (1993). 46 See Cindy R. Alexander, On the Nature of the Reputational Penalty for Corporate Crime: Evidence, 42 J.L. & ECON. 489, 493 (1999). 41 See Saltzburg, supra note 4, at 431. COLUMBIA BUSINESS L4WREWEW [Vol. 2004 suggest that we distinguish between offenses based on the identity of the injured party. Thus, offenses may be classified as either related-party or third-party offenses." Related-party offenses are those in which a party to the transaction in question has been directly harmed. These offenses include consumer fraud in its various forms, fraudulent overcharging, payment of bribes, receipt of kickbacks, and fraudulent bidding, such as where a contractor makes misrepresentations in order to win a contract. 9 In all these cases, the corporate misconduct results directly in harm to the injured party who is a party to the transaction. A second category of corporate crimes involves conduct that injures third parties. Environmental offenses, for example, do not harm related parties, but instead cause damage to a public good. Other examples of third-party offenses include various regulatory offenses, such as a bank that fails to report large cash transactions in violation of federal law, or an exporter of goods that disobeys an embargo, compromising national foreign policy.5" These offenses do not directly injure another party to the transaction, but instead cause a more indirect harm that is dispersed throughout some segment of the public, or the society as a whole. Because the harms caused by third party offenses are indirect and dispersed, it is unlikely that a corporation that commits such an offense would suffer a loss to its quality premium. In other words, a third party offense is unlikely to affect an individual consumer's perceptions about the quality of the corporation's product or service. However, this does not mean that the firm will escape all reputational losses. Consumers may be less likely to conduct business with an offending corporation because of ethical 48 See Alexander, supra note 46, at 497-500. 49 Id. at 498. 50 Id. at 500. COLUMBIA BUSINESS LA W REVIEW [Vol. 2004 No. 3:879] REPUTATIONAL PENALTIES FOR CORPORATIONS considerations or a stigma that may attach to the guilty firm. 51 A statistical analysis by Cindy Alexander demonstrates that the disparate impact of reputational penalties on corporations depends on the identity of the injured party. Alexander examines the impact of criminal convictions on seventy-eight public corporations occurring between 1984 and 1990.52 Alexander found that the corporations suffered substantial reputational penalties, measured by losses in shareholder wealth, after being convicted. Consistent with the theory of market-based penalties, she also found that corporations convicted of crimes involving related parties experienced significantly larger shareholder losses than those convicted of crimes involving third parties.5 ' These results persist even after controlling for the magnitude of the fine imposed. 4 The effect was most acute among smaller firms.55 The magnitude of reputational losses is also likely to be affected by certain characteristics of a firm's customers. In the normal case of atomistic consumers,56 no individual consumer can significantly influence the reputational penalty borne by a corporation. Reputational losses will therefore only be experienced if it is in the narrow self- interest of a significant number of consumers to terminate their relationship with the supplier. 7 This will depend on both the availability of information in the market and the ability of consumers to share this information. For example, if an instance of corporate misconduct receives widespread coverage in the national news media, a greater reputational s See Saltzburg, supra note 4, at 431. 52 Alexander, supra note 46, at 497-98. '3Id. at 504-05. 54Id. 5 Id. at 520-21. 6 These considerations are not present in the case of non-atomistic consumers. 17 See Alexander, supra note 46, at 494. COLUMBIA BUSINESS LAW REVIEW penalty can be expected because more consumers will be aware of the misconduct and may act upon it. On the other hand, if the event is not widely publicized, consumers will be less likely to be aware of the misconduct, and the reputational effects will be attenuated. Similarly, the ability of consumers to communicate with each other will affect the reputational losses incurred by a corporation, at least in the case of minimal publicity.58 This ability depends on several factors, including how widely dispersed consumers are. If a corporation operates a small store which draws the majority of its customers from a single community, it is likely that consumers will be able to communicate relatively cheaply. In this case, many customers may even know each other, and word-of-mouth can be expected to operate efficiently to inform consumers of misconduct. The operator of a large retail chain operating over a wide area may be expected to suffer a smaller reputational penalty, as customers will be more dispersed and information will not flow as readily. This will be especially true if the misconduct involves only one location. A company that operates a mail-order catalogue may expect an even smaller reputational penalty, as customers will tend to be even more dispersed, and customers will be unlikely to encounter each other. A company that sells its wares on the Internet may be in a similar situation, but the ability of its customers to identify each other and communicate relatively costlessly over the Internet will mitigate, if not eliminate, the effects of geographic dispersion. More generally, the impact of the Internet in substantially decreasing costs of communication may lead to greater overall reputational penalties for corporations, whether or not they transact business on the Web. ' If the event is well-publicized, a large number of consumers would likely become aware of the misconduct through the media, and the ability of consumers to communicate with each other would therefore be of reduced (and perhaps eliminated) importance. [Vol. 2004 No. 3:879] REPUTATIONAL PENALTIES FOR CORPORATIONS Despite the strong evidence that corporations bear substantial market-based sanctions for crime, the presence of these penalties was apparently not considered by the Sentencing Commission in formulating its scheme for corporate sentencing, as no mention of reputational penalties appears in the elaborate model that determines the criminal sanction. Of course, it is possible that the Commission recognized the existence of these sanctions and tailored the fine amounts sub silentio by mandating lesser fines than would have been appropriate given no reputational penalties. However, given the significant variation in the magnitude of reputational losses depending on the nature of the crime, the identity of the injured parties, and the nature of the firm's customers, the failure to account for these variations in setting criminal sanctions will necessarily result in either over-deterrence or under-deterrence. VI. POTENTIAL SOLUTIONS Although the Federal Sentencing Guidelines do not explicitly account for the presence of reputational penalties, sentencing judges can and should take these effects into consideration in determining the appropriate corporate sanction. In this way, individual judges may take significant steps toward optimal corporate sanctions under the current system. This is possible because the Guidelines include built-in safeguards that could perhaps make such an approach workable. The first, and less dramatic, is the ability of the sentencing judge to set a fine within a range determined by the Guidelines.59 A judge might therefore take into account the reputational factors in determining where within the range the fine should be set. In fact, the Guidelines suggest such an approach in Section 8C2.8, a policy statement listing appropriate factors to be considered by a judge in setting a fine within the applicable range. One factor listed is "any collateral consequences of conviction, " SENTENCING GUIDELINES, supra note 10, at § 8C2.7. COLUMBIA BUSINESS LAW REVIEW [vol. 2004 including civil obligations arising from the organization's conduct."6 ° The language "including civil obligations" does not limit the scope of this consideration to such, and this section could therefore be interpreted as applying also to reputational losses as "collateral consequences of conviction." The other, more dramatic, safeguard built in to the Guidelines to assure that appropriate sanctions are imposed is the ability of a sentencing judge to depart from the Guidelines-mandated range under certain circumstances. The requirement for a departure is "that there exists an aggravating or mitigating circumstance of a kind, or to a degree, not adequately taken into consideration by the Sentencing Commission in formulating the guidelines that should result in a sentence different from that described."6' 1 Section 8C4 of the Guidelines lists a number of factors that may not have been adequately taken into consideration and therefore may constitute appropriate grounds for departure. Although the list does not explicitly mention reputational penalties, it is not exhaustive. It is therefore possible that a sentencing judge, acting under the current formulation of the Guidelines, may depart based on reputational considerations. Despite the limited ability of judges to factor in reputational penalties under the current system, the Sentencing Commission should do more to account for the effects of these penalties within the Guidelines. Efforts to incorporate reputational penalties into the Guidelines may come in two forms. First, the Sentencing Commission may attempt to control for the presence of reputational penalties ex ante by adjusting the fine scheme to take these factors into consideration within the Guidelines. On the other hand, the Sentencing Commission may provide additional mechanisms to enable the courts to control for market-based penalties by performing ad hoc analyses in individual cases. 60 Id. at § 8C2.8(3). 61 18 U.S.C. § 3553(b) (2000). COLUMBIA BUSINESS LA W REVIEW [Vol. 2004 No. 3:879] REPUTATIONAL PENALTIES FOR CORPORATIONS In order to account for reputational penalties, the Sentencing Commission might amend the Guidelines to account for reputational effects in its scheme for determining the corporate fine. This could be achieved in a number of ways, but in any case, it would probably resemble a point system similar to that used throughout the Guidelines. Points might be added to or subtracted directly from the offense level based on reputational factors designed to predict the reputational penalty. Alternatively, points might be added to or subtracted from the culpability score, which affects the fine multiplier. Factors that would likely be accounted for are the size of the corporation, the nature of the offense, the identity of the injured party (whether a related party or third party), whether customers are atomistic or non-atomistic, the degree of dispersal of the corporation's customers, the ability of customers to communicate, the level of media coverage that the case has drawn, and the current state of the company's reputation. These factors are numerous and complex, and there is a significant degree of interplay between them. Furthermore, case-specific factors that would be nearly impossible to account for may render the calculus futile. For example, in the normal case of a crime involving the injury of a third party, such as a criminal violation of environmental regulations, we would expect the reputational penalty to be relatively low. On the other hand, consider the case of a company convicted of such a violation that is a manufacturer of consumer solar panels whose customers are primarily committed environmentalists. In this case, we would expect the market-based sanction to be high; at least higher than would normally be expected for an environmental violation involving unrelated injured parties. It is hard to imagine how the Guidelines could account for such factors ex ante, as they are likely to be numerous and unpredictable. An ex ante approach to the consideration of reputational effects within the Guidelines, combined with the judicial discretion to set a fine within a range and to depart in extraordinary circumstances, may achieve a satisfactory COLUMBIA BUSINESS LAW RE WE W [Vol. 2004 result in the determination of appropriate sanctions as well as provide for extraordinary cases. It would certainly be preferred to the current system, which does not account for reputational losses. However, such a system would still be imprecise, and would involve a significant amount of crude estimation and guesswork both on the part of the Sentencing Commission and the judge. A better system would therefore provide for a method of estimating the reputational losses that a convicted corporation may be expected to suffer in a particular case and applying this estimation to the final sanction imposed by the court. Toward this end, the Sentencing Commission might provide for a hearing within the sentencing phase of the trial. The purpose of this hearing would be to provide a reasonable estimate of the reputational losses that the convicted corporation can be expected to suffer as a result of the conviction. This ad hoc approach would have the advantage of being able to take into consideration all possible factors affecting the magnitude of the reputational loss, including extraordinary factors that cannot be contemplated ex ante (such as the solar panel example), and apply them to the unique considerations of the individual case. Experts from both sides could be called to testify to the amount of the expected reputational loss, and the judge would then weigh the testimony to determine the best estimate. This method of calculating the expected reputational penalty would resemble the provisions in the Guidelines for determining the "pecuniary gain to the organization from the offense" and "the pecuniary loss from the offense caused by the organization." 2 Similar to the provision that the court may forego these calculations if they would "unduly prolong or complicate the sentencing process,"63 the revised Guidelines might provide that in cases where the reputational sanction would likely be negligible or where the 62 SENTENCING GUIDELINES, supra note 10, at § 8C2.4(a). 6 Id. at § 8C2.4(c). COLUMBIA BUSINESS LA W REVIEW [Vol. 2004 No. 3:879] REPUTATIONAL PENALTIES FOR CORPORATIONS 897 costs of the process would likely outweigh the reputational losses, the estimation hearing would not be required, and the court may substitute a rough estimate. VII. CONCLUSION The reputational penalties suffered by corporations convicted of crimes and the civil remedies available to injured parties present issues for corporate criminal sentencing. In order to avoid the social costs of over- deterrence and under-deterrence, these independent sanctions must be coordinated in determining the socially optimal penalty. However, the Federal Sentencing Guidelines, as they currently stand, fail to sufficiently account for these factors. The evidence presented in this Note suggests that failure to achieve the optimal sanction results in social costs, which are exacerbated in the corporate context due to agency costs. Reputational losses incurred by corporations that are convicted of crimes may in some cases be substantial. The nature of the crime, the identity of the injured party, the extent of media coverage, and characteristics of the corporation's consumers, inter alia, account for differences in the magnitude of the market-based penalty a convicted corporation can be expected to experience. This suggests that the Guidelines should account for these factors in formulating an appropriate sentencing scheme. Nevertheless, the difficulties of estimating the effects of these factors ex ante and the many potentially unforeseeable factors that may affect the calculus in an individual case warrant an ad hoc approach. Although judges operating under the current formulation of the Guidelines may take significant steps toward the optimal sanction by considering the probable reputational losses in selecting a specific fine from within the Guidelines-mandated range and in deciding whether or not to depart from the range, the Sentencing Commission should nevertheless take affirmative steps to remedy this problem.