Anything but a “Racket”: Why Professor Richard Epstein’s Attack on the Nature and Function of Deferred and Non-Prosecution Agreements Misses the Mark
By Elizabeth R. Sheyn
December 26, 2007
In light of recent corporate scandals involving companies such as Enron and Arthur Andersen, the need to impose criminal penalties on corporations has been on the rise. More and more often, however, federal prosecutors have utilized deferred and non-prosecution agreements in lieu of indicting and criminally prosecuting business entities. Despite the advantages that such agreements create for both prosecutors and companies, Professor Richard A. Epstein has vigorously attacked this practice in his article, The Deferred Prosecution Racket, contending that deferred and non-prosecution agreements do not serve the public interest because they do not help to restore shareholder confidence in the firm and undermine the principle of separation of powers. Epstein’s arguments are ill-founded, however, as deferred and non-prosecution agreements rebuild shareholder and public confidence in corporations while allowing companies to avoid certain death by indictment. Moreover, it is only by imposing or by threatening to impose criminal liability on business entities, rather than by solely charging individuals, that the government can effectively stop and prevent corporate crime.Deferred prosecution agreements typically require prosecutors to file criminal charges against a corporation while agreeing that these charges will be dismissed within approximately one or two years, unless the company breaches any of the terms of the agreement. A majority of deferred prosecution agreements call for the corporation to cooperate “with the government in its investigation of culpable individuals . . . [and] accept . . . responsibility by acknowledging the acts of its employees.” Such agreements also require the business entity to take on “prospective internal reforms including effective compliance programs and independent monitors, retrospective review of particular financial transactions, and punitive measures, including penalties, restitution and surrender of ill-gotten financial gains.” Moreover, along with providing for corporate fines and penalties, deferred prosecution agreements “may require companies to disgorge any gains from the misconduct that ha[ve] not and will not be paid as restitution or by way of other remedial measures.”
A number of corporations have signed deferred prosecution agreements with the government within the last several years in lieu of facing criminal charges. For example, pursuant to its deferred prosecution agreement, Computer Associates International, Inc., a New York–based software producer, admitted that it committed accounting fraud and inflated earning reports, agreed to pay $225 million in restitution to its shareholders, waived attorney-client privilege, added independent directors to its Board, consented to furthering corporate governance reforms, and hired a compliance monitor in exchange for the government’s deferral of prosecution for eighteen months. Computer Associates also undertook a number of organizational reforms and agreed that it would not “make any public statement . . . contradicting its acceptance of responsibility or the allegations set forth in the [deferred prosecution agreement].” Companies such as America Online and Bristol-Meyers Squibb have also signed similar agreements.
Non-prosecution agreements, though usually encompassing many of the attributes of deferred prosecution agreements, do not involve the formal filing of criminal charges by the prosecutor against the corporation. Although non-prosecution agreements “allow the company to avoid any potential collateral consequences associated with the mere fact that the company has been charged with a crime, . . . they still require public acceptance of responsibility, restitution, and surrender of ill-gotten gains, full cooperation, and implementation of remedial measures.” In addition, as is true of deferred prosecution agreements, the government reserves the right to prosecute the business entity if it fails to comply with any part of its non-prosecution agreement. Corporations such as Merrill Lynch and the Bank of New York (BNY) have signed such agreements with the government.
In particular, under BNY’s non-prosecution agreement, BNY accepted responsibility for the misconduct of its employees, who “(a) aided and abetted the fraudulent activities of RW Professional Leasing Services Corp. (PLS) by executing sham escrow agreements . . .; and (b) willfully failed to . . . file a Suspicious Activity Report (SAR) . . . and failed to notify law enforcement authorities of suspicious activities relating to PLS and the escrow agreements.” The measures that BNY was required to take pursuant to its non-prosecution agreement in order to avoid being criminally charged by the government closely resemble those typically included in deferred prosecution agreements. BNY had to disclose all information relevant to the government’s investigation of its conduct. Additionally, BNY agreed to “pay a total of twelve million dollars . . . into a fund established to make restitution to any banking institutions that suffered losses in reliance on BNY Escrow Agreements,” forfeit its rights to twenty-six million dollars in illegal profits, undertake a number of remedial measures with respect to its procedures of reporting suspicious banking activities, and retain an independent examiner who would ensure BNY’s compliance with the agreement for a period of three years.
Although nearly all deferred and non-prosecution agreements have been found to be lawful by the courts, one federal district court judge has held that the deferred prosecution agreement of KPMG LLP, a major accounting firm, violated the rights of certain employees, whom KPMG refused to indemnify pursuant to the Thompson Memo in the hopes of demonstrating its compliance with the government’s investigation. Such prosecution tactics are unlikely to reoccur given the recent McNulty Memo, which alters the troubling aspects of the Thompson Memo’s charging policy. In particular, the McNulty memo negates the Thompson Memo’s stance that corporations that do not waive attorney-client and work product privileges and that indemnify their employees are not cooperating with the government.
Seizing upon the outcome of the KPMG litigation, however, Professor Richard Epstein attacks deferred prosecution agreements, contending that they no longer serve the public interest. Specifically, Epstein argues that such agreements do not work to restore the confidence of shareholders in the corporation and “erode[s] the most elementary protections of the criminal law, by turning the prosecutor into the judge and jury, thus undermining our principles of separation of powers.” Epstein also advocates for the complete elimination of corporate criminal liability, indicating that the government should prosecute human criminals, not fictional entities.
Although, in theory, it may be difficult for a corporation to function “with a government mole in its midst,” Steven R. Peikin argues in Deferred Prosecution Agreements: Standard for Corporate Probes that the utilization of independent (and not government-appointed) examiners to ensure that companies are complying with their deferred or non-prosecution agreements with the government is “frequently greeted with enthusiasm by investors and other stakeholders.” The institution of new governance structures and the need to obtain approval for every new measure, not to mention the hiring of an independent compliance examiner, is quite costly for a business entity, both in terms of time and money. However, once its criminal conduct has been discovered, a corporation can best reassure the public and its shareholders that it is going to survive by cooperating with the government and by demonstrating that its policies, which had previously permitted unlawful conduct to occur, will be drastically improved.
In addition, Epstein’s concern that deferred prosecutions undermine the principle of separation of powers is ill-founded. Federal prosecutors must weigh nine factors in deciding whether to impose criminal charges on a corporation; it is certainly not in the prosecutors’ best interest to bring baseless or frivolous charges, as they will be quickly dismissed. Although Epstein contends that, in negotiating for deferred or non-prosecution agreements, the “corporation’s sole remedy is to plead its case before the prosecutor in an environment wholly devoid of the most rudimentary procedural protections,” he does not appear to take into account the fact that companies negotiate with the government only after substantial evidence of criminal wrongdoing has been discovered and specifically because they desire to avoid imminent prosecution. Moreover, as Wray and Hur demonstrate, the government benefits from dealing as favorably as possible with cooperating business entities because “[i]f [corporations] perceive the government as overly aggressive or unpredictable in its corporate charging decisions, and if prosecutors fail to give appropriate weight to . . . [the] self-reporting of misconduct, then companies may decide that the risk is simply not worth taking.”
Furthermore, Epstein’s argument that the issue of the propriety or impropriety of deferred and non-prosecution agreements need not arise in the first place because the government should not impose criminal liability on corporations is misguided in several respects. First, the theory of respondeat superior has been applied to successfully root out criminal activity in corporations for approximately 100 years. In addition, it is quite difficult for the government, both financially and logistically, to “go after real . . . human criminals” in the corporate context. As Wray and Hur note, “[g]overnment resources are finite; because it benefits investigators enormously when a company’s often formidable resources work with the government instead of against it, prosecutors and regulators will continue to press companies to cooperate. To do otherwise would be both inefficient and ineffective.”
Even though Epstein contends that “[a]t bottom, corporations are just individuals tied together by an elaborate network of contracts,” it is quite difficult for the government to identify and to charge the particularly blameworthy individuals hiding behind (or within) the company. O’Sullivan explains that corporate liability is necessary, “less costly[,] and more fruitful” precisely because it is quite difficult to penetrate “the corporate ‘black box’ to locate the appropriate agent or agents to prosecute for a crime.” As a result, if the government attempts to only target individuals employed by business entities during its criminal investigations, it is unlikely to achieve the same level of success in prosecuting them as it has reached by using deferred and non-prosecution agreements. In addition, individual wrongdoing within the corporation is frequently a symptom of faulty organizational practices or bad corporate ethos. Only targeting particular offenders does nothing to alter the company’s “bad” or “criminally inclined” nature; instead, it will permit, if not encourage, the corporation’s future unlawful activity.
Epstein’s one legitimate concern, the Thompson Memo’s focus on corporate waiver of the attorney-client and work product privileges and on corporate refusal to indemnify individual employees, has also been recently assuaged by the McNulty Memo’s decisive alteration of these troubling policies. The government should, however, adhere to Epstein’s advice with respect to leaving out unnecessary and somewhat dubious requirements from its deferred and non-prosecution agreements, such as its demands for a corporation to endow a chair of business ethics at a particular law school.
Deferred and non-prosecution agreements provide substantial benefits to both business entities and the government. Indeed, such agreements are welcomed by the public and by shareholders because they allow corporations to avoid collapse, which would certainly follow their criminal indictments, while ensuring the improvement of companies’ organizational and operational policies. Furthermore, such agreements are advantageous for the government, as they “give [it] the salutary benefit of publicly airing its (negative) view of the corporation’s conduct, sending a clear message of the kind of corporate behavior it will not condone,” while allowing it to conserve its limited investigational resources. Thus, deferred and non-prosecution agreements are the best means of rooting out and preventing unlawful behavior in the corporate context, particularly since the government, with the promulgation of the McNulty Memo, has negated its previous requirement that corporations demonstrate their cooperation by waiving attorney-client and work product privileges and by denying indemnification to individual employees. As a result, Professor Richard Epstein’s attack on such agreements, and on the notion of corporate criminal liability in general, misses its mark.
 Richard A. Epstein, Commentary, The Deferred Prosecution Racket, Wall St. J., Nov. 28, 2006, at A14, available at http://www.ilta.org/WhatsNew/December2006/0612%20DeferredProsecution.pdf. Epstein’s solution to the problem of corporate crime is, in effect, to completely eliminate corporate criminal liability. Id.
 Christopher A. Wray & Robert K. Hur, Corporate Criminal Prosecution in a Post-Enron World: the Thompson Memo in Theory and Practice, 43 Am. Crim. L. Rev. 1095, 1104 (2006).
 Id. at 1105.
 See id. at 1140 (citing Deferred Prosecution Agreement, United States v. Computer Assocs. Int’l, Inc., Cr. No. 04-837 (E.D.N.Y. Sept. 22, 2004)).
 Computer Associates Deferred Prosecution Agreement, supra note 6, ¶ 27.
 See Wray & Hur, Corporate Criminal Prosecution, supra note 2, at 1139-1141 (describing the deferred prosecution agreements signed by a number of high-profile corporations).
 See id. at 1105.
 See id.
 See id. at 1142 (surveying several corporations’ non-prosecution agreements with the government).
 It is likely that BNY’s high level of cooperation with the government’s investigation and its willingness to acknowledge wrongdoing, combined with the fact that a seemingly isolated subset of employees was engaged in the unlawful conduct, led to this agreement being a non-prosecution agreement, as opposed to a deferred prosecution agreement.
 BNY Non-Prosecution Agreement, supra note 13, ¶ 7.
 Id. ¶ 8.
 Id. ¶ 9.
 Id. ¶ 10–12.
 In 2003, then-Deputy Attorney General Larry Thompson promulgated a revised set of guidelines designed to help federal prosecutors decide whether to charge a corporation, rather than or in addition to individuals within the corporation, with criminal offenses. The charging factors listed in the Thompson Memo include: (1) “the nature and seriousness of the offense;” (2) “the pervasiveness of wrongdoing within the corporation;” (3) “the corporation’s history of similar conduct;” (4) “the corporation’s timely and voluntary disclosure of wrongdoing and its willingness to cooperate in the investigation of its agents, including, if necessary, the waiver of corporate attorney-client and work product protection;” (5) “the existence and adequacy of the corporation’s compliance program;” (6) “the corporation’s remedial actions;” (7) “collateral consequences;” (8) “the adequacy of the prosecution of individuals responsible for the corporation’s malfeasance;” and (9) “the adequacy of remedies such as civil or regulatory enforcement actions.” Memorandum from the Deputy Attorney Gen. Larry D. Thompson to U.S. Attorneys (Jan. 20, 2003), in Julie R. O’Sullivan, Federal White Collar Crime: Cases and Materials 236 (2d ed. 2003).
 Memorandum from the Deputy Attorney Gen. Paul J. McNulty to U.S. Attorneys 8 (Dec. 12, 2006) (on file with author). The McNulty Memo supplanted the Thompson Memo, thereby altering several troublesome charging factors included in the Thompson Memo. Under the McNulty Memo, whether prosecutors have a legitimate need for a given corporation’s privileged information depends upon: “(1) the likelihood and degree to which the privileged information will benefit the government’s investigation; (2) whether the information sought can be obtained in a timely and complete fashion by using alternative means that do not require waiver; (3) the completeness of the voluntary disclosure already provided; and (4) the collateral consequences to a corporation of a waiver.” Id. at 9.
 Moreover, the McNulty Memo expressly provides safeguards to ensure that prosecutors do not abuse their power and continue to require corporations to waive attorney-client and work product privileges and to abstain from indemnifying their employees.
 Epstein, supra note 1.
 Id. (“The government has a vital role in criminal enforcement. So let it go after real, i.e., human, criminals the old-fashioned way, by careful investigation and skilled prosecution.”).
 Steven R. Peikin, Deferred Prosecution Agreements: Standard for Corporate Probes, N.Y.L.J., Jan. 31, 2005, at 6.
 Epstein, supra note 1.
 Wray & Hur, supra note 2, at 1171. Wray and Hur also note that “[i]f industry comes to view the possibility of a criminal conviction as only marginally less attractive than the certainty of harsh treatment at the hands of the government after self-reporting its misconduct, then fewer and fewer companies will be willing to step forward, voluntarily disclose violations and cooperate with ensuing government investigations.” Id.
 Epstein, supra note 1.
 Wray & Hur, supra note 2, at 1171. Corporate cooperation alone has led to an increase in the number of government “investigations that proceed at a dramatically quicker pace.” Id.
 Epstein, supra note 1.
 Julie Rose O’Sullivan, Professional Discipline for Law Firms? A Response to Professor Schneyer’s Proposal, in O’Sullivan, Federal White Collar Crime, supra note 19, at 198.
 See id. at 200 (“There are two potentially overlapping but analytically distinct ways in which an entity may be deemed ‘responsible’ in a causal sense for its agents’. . . misconduct. First, [an entity] may fail to put in place organizational policies or practices sufficient to prevent certain types of . . .misconduct. Second, [an entity] may possess a bad ‘culture’ or ‘ethos,’ which may. . . [actually encourage employee wrongdoing].”).
 See Epstein, supra note 1 (describing a provision of a deferred prosecution agreement which required the corporation to “endow a chair at Seton Hall University School of Law,” the prosecuting attorney’s alma mater).
 See supra note 25 and accompanying text.
 See supra note 25 and accompanying text.