Oil in the Joints or Monkey Wrench in the Gears: Deferred and Non-Prosecution Agreements in Antitrust Cases

by Jay L. Himes

November 03, 2014

We observe the possible effects of a shift from both, a government enforcer and defense perspective, and discuss what this means for future antitrust enforcement.

I. Introduction 

In February of 2013, the Department of Justice Antitrust Division, together with the Criminal Division (collectively, DOJ), announced its first-ever Deferred Prosecution Agreement (DPA), made with the Royal Bank of Scotland plc (RBS) in connection with the bank's role in benchmark interest rate (LIBOR) manipulation. 1 Under the DPA, the bank agreed "to admit and accept responsibility for its misconduct[,]" to cooperate with the DOJ's ongoing investigation, and to pay a $100 million penalty. 2 

The Antitrust Division has traditionally avoided using DPAs and Non-Prosecution Agreements ("NPAs") in criminal antitrust cases, favoring instead its Corporate Leniency Program. However, in recent years the Antitrust Division has offered NPAs in a few cases. Thus, the DPA with RBS caused the antitrust bar and legal analysts to wonder whether this resolution might foreshadow a shift away from the Antitrust Division's predilection for its Corporate Leniency Program. 

Here, we examine the potential effects of such a shift from both a government enforcer and defense perspective. Specifically, we discuss whether DPAs and NPAs can act as companions to the Corporate Leniency Program by enhancing corporate compliance and averting significant collateral consequences such as financial ruin, or whether the agreements undermine the very core of the Corporate Leniency Program and its success in furthering criminal antitrust enforcement. We also (1) note the potential effects that increased use of DPAs and NPAs could have on related civil antitrust litigation, (2) examine whether there are guiding principles found in those cases where the Antitrust Division has offered criminal defendants DPAs and NPAs in lieu of prosecution, and (3) discuss what all of this means for future antitrust enforcement. 

II. Background 

A. What Are DPAs and NPAs and Where Do They Come From?

DPAs and NPAs are flexible agreements that federal prosecutors and companies under criminal investigation can tailor to meet the circumstances. The agreements offer the DOJ a middle ground between pursuing a potentially harmful criminal process, and forgoing criminal prosecution outright. Under an NPA, the government agrees not to prosecute a company, so long as it satisfies its obligations under the agreement. 3 The agreement may never even become public. Under a DPA, the government files formal criminal charges, but refrains from prosecuting the case on the condition that the company performs its obligations under the agreement. 4 DPAs are typically public documents filed on the court's docket. 5 

Generally speaking, both DPAs and NPAs require the contracting company to: 

• admit to its criminal conduct; 
• cooperate fully with the government's investigation; 
• implement (or strengthen) a corporate compliance program or other remedial measures; 
• expand and strengthen existing internal controls; 
• terminate employment of responsible individuals; 
• refrain from making any public statements contradicting the terms of the agreement or the company's admissions; 
• pay a criminal penalty; and 
• face prosecution if the agreement is breached. 6 

These agreements with corporations have their roots in pretrial diversion programs for individual defendants, which DOJ has used where it has determined that the societal benefits of prosecution are outweighed by the costs to the defendant. Typically, such programs are limited to first-time offenders who have committed relatively minor offenses. Similarly, corporate DPAs and NPAs are designed to prevent or mitigate collateral harm that could result from a corporate conviction, while preserving the deterrent, punitive, and rehabilitative goals of criminal prosecution. 

Although used intermittently over the years, corporate prosecution agreements became more common following the DOJ's prosecution and conviction of the auditing firm Arthur Andersen in 2002, 7 which led to the firm's demise and to 28,000 employees losing their jobs. Criticism that the conviction resulted in the "corporate death penalty" led the DOJ to revise its policy on the prosecution of corporations in 2003. In what is referred to as the "Thompson Memo," the DOJ stated that cooperation and voluntary disclosure could merit "granting a corporation immunity or amnesty or pretrial diversion." 8 The memo also referred prosecutors to the section of the United States Attorneys' Manual that sets forth principles governing non-prosecution agreements for individuals. 9 This was the first time the DOJ suggested using corporate prosecution agreements in exchange for cooperation by companies involved in criminal activity. 

The DOJ further revised the U.S. Attorneys' Manual in 2008, expressly endorsing the use of prosecution agreements for corporations: 

[W]here the collateral consequences of a corporate conviction for innocent third parties would be significant, it may be appropriate to consider a non-prosecution or deferred prosecution agreement with conditions designed, among other things, to promote compliance with applicable law and to prevent recidivism. Such agreements are a third option, besides a criminal indictment, on the one hand, and a declination, on the other....Under appropriate circumstances, a deferred prosecution or non-prosecution agreement can help restore the integrity of a company's operations and preserve the financial viability of a corporation that has engaged in criminal conduct, while preserving the government's ability to prosecute a recalcitrant corporation that materially breaches the agreement. Such agreements achieve other important objectives as well, like prompt restitution for victims. 10 

The U.S. Attorneys' Manual also sets forth factors relevant to the government's consideration whether to offer a prosecution agreement to a corporation "due to the nature of the corporate 'person.'" 11 For example, "prosecutors should consider...the pervasiveness of wrong-doing within the corporation" and the "existence and effectiveness of the corporation's pre-existing compliance program." 12 Other factors to consider are the "collateral consequences" of prosecution, such as "disproportionate harm to shareholders, pension holders, employees, and others not proven personally culpable, as well as the...adequacy of remedies such as civil or regulatory enforcement actions." 13 

In the years that followed, several branches of the DOJ, which unlike the Antitrust Division, do not have leniency programs, used DPAs and NPAs as powerful enforcement tools. For example, the DOJ has used DPAs and NPAs in cases relating to accounting fraud, environmental pollution, money laundering, health care fraud, and the Foreign Corrupt Practices Act. Indeed, the Criminal Division now enters into more prosecution agreements than plea agreements. In 2012, the Department entered into at least 35 DPAs and NPAs, securing $9 billion in fines. 14Similarly, in 2013, the Department entered into 27 DPAs and NPAs, making 2013 "the fifth consecutive year in which at least 20 agreements have been executed." 15 

Although the Antitrust Division entered into a few NPAs prior to 2011, in recent years the Division has used these agreements more frequently, particularly in cases involving financial institutions accused of price fixing or bid rigging. In 2011, for example, the Antitrust Division entered into NPAs with four financial institutions that were accused of bid rigging relating to municipal bonds.16 Most recently, the Antitrust Division entered into its first DPAs with Coöperatieve Centrale Raiffeisen-Boerenleen-bank B.A. ("Rabobank") 17 and RBS, 18 after filing a criminal information against each company charging them with manipulation of LIBOR rates. Historically, however, the Antitrust Division has relied on its Corporate Leniency Program to enforce criminal antitrust violations. 

B. The Antitrust Division's Corporate Leniency Program 

The Antitrust Division's Corporate Leniency Program (the Program) is the cornerstone of antitrust criminal enforcement in the United States. Announced in 1978, the Program was significantly revised in 1993 and again in 2008 to further encourage cartel members to self-report their illegal activity. 19 The Program, which affords criminal leniency to self-reporting companies, has successfully assisted the Division to uncover and prosecute antitrust cartels (and other criminal antitrust violations). 

Under the Program, the first cartel member to report illegal activity and turn in the other members can, by cooperating with the Division in investigating and prosecuting the antitrust violation, avoid all criminal convictions, fines and prison terms for the company and its cooperating individual officers and employees. 20 This immunity, however, is available to only one company in any particular matter. 21 Accordingly, the Program incentivizes cartel members to avoid criminal exposure by getting to the government first, and thereby instigates a race among co-conspirators to be the "first-in." 

The Antitrust Division offers two types of leniency: Type A, granted before an investigation has begun, and Type B, granted after the Division has received information about illegal cartel activity, regardless of whether it has opened a formal investigation. 22 For Type A leniency, the company-referred to as the "leniency applicant"-must meet six conditions: 

(1) At the time the corporation comes forward, the Division has not received information about the activity from any other source. 
(2) Upon the corporation's discovery of the activity, the corporation took prompt and effective action to terminate its participation in the activity. 
(3) The corporation reports the wrongdoing with candor and completeness and provides full, continuing, and complete cooperation to the Division throughout the investigation. 
(4) The confession of wrongdoing is truly a corporate act, as opposed to isolated confessions of individual executives or officials. 
(5) Where possible, the corporation makes restitution to injured parties. 
(6) The corporation did not coerce another party to participate in the activity and clearly was not the leader in, or the originator of, the activity. 23 

An applicant that cannot meet these six conditions may nevertheless receive Type B leniency where the applicant is (1) "the first to come forward and qualify for leniency with respect to the activity[,]" and (2) at the time, the Division does not have evidence against the applicant that is likely to result in a conviction. 24 For Type B leniency, the applicant also must satisfy Type A conditions 2 through 5. 25 In addition, the Division must be able to determine that leniency "would not be unfair to others, considering the nature of the activity, the confessing corporation's role in the activity, and when the corporation comes forward." 26 

Where the applicant satisfies the necessary conditions, the Division then issues what is called a "conditional leniency letter." 27 The leniency is conditional until the applicant provides full and complete cooperation to the Division's investigation and prosecution, including document production and witness interviews and testimony. 28Once the leniency applicant satisfies the conditions of the letter-something that often means a multi-year commitment-leniency becomes final. 29 

While the "top prize" is reserved for the leniency applicant, a "second-in" company that provides timely and valuable cooperation may also receive substantial benefits, such as reduced fines and more favorable treatment of its culpable executives. 30 Additionally, a company that loses the race, but offers DOJ evidence of a criminal antitrust violation involving another product or geographic area can qualify for leniency concerning the newly uncovered conspiracy, and also receive more favorable treatment where it lost the race. 31 This is called "amnesty plus," and is another way that DOJ seeks to mitigate the "winner take all" approach to criminal immunity. 32 

While there are benefits available to a "second-in" company, the Division seeks to calibrate them to the timing and value of the cooperation offered. However, a second-in company, as well as any of its culpable employees, always remains subject to full prosecution and almost always pleads guilty to criminal antitrust violations. 

By publicly announcing and regularly discussing its Corporate Leniency Program, the Antitrust Division intends that its terms be transparent to cartel members. 33 Thus, the Program is designed not only to sow the seeds of mistrust among cartel members, thus destabilizing the cartel, but also to assure those seeking to enter the Program predictable treatment by the Antitrust Division. 34 Central to the Program, however, is the notion that the first-in company will be treated fundamentally better than any other cartel member. 

To illustrate, take the case of Chien Chung Chen aka Andrew Chen, former executive vice president of a U.S. company that participated in an international conspiracy to fix the prices of aftermarket auto lights sold in the U.S. In 2008, Chen's company sued the other cartel members for trying to drive the company out of business, thereby effectively "blowing the whistle" on the cartel. 35 However, neither Chen nor his company sought amnesty under the Corporate Leniency Program because of erroneous legal advice. Meanwhile, another cartel member-one of those whom Chen's company had sued-did, and the DOJ granted that cartel member conditional leniency. 36 

Aided by its leniency applicant, the DOJ launched a criminal investigation and indicted Chen and his company, among others. 37 Eventually, Chen and the company agreed to plead guilty for participating in the conspiracy. 38 

At Chen's sentencing hearing, the DOJ argued adamantly that Chen should serve a six-month prison term even though he was responsible for exposing the conspiracy. 39 Indeed, the DOJ distinguished Chen's situation from that in another case specifically on the ground that Mr. Chen was not the first in the door. 40 In the DOJ's view, because Chen was not the "first-in," he was not entitled to the same benefits enjoyed by the leniency applicant, even though Chen also cooperated with the government, and might have secured immunity for himself and his company but for erroneous legal advice. The court disagreed with the DOJ's position on sentencing, and credited Chen's cooperation while imposing one year of probation, six months of house arrest, and a fine of $25,000. 41 Chen's company, which also pleaded guilty, received a fine of $200,000. 42 

Notwithstanding that result, the DOJ's position in Chen's case illustrates an important principle of the DOJ's criminal antitrust enforcement approach: only the "first-in" cartel member escapes criminal prosecution in antitrust cases under, the Corporate Leniency Program. 

Thus, the Corporate Leniency Program reflects a "one size fits all" approach. In contrast, DPAs and NPAs are, by their nature, fashioned to resolve criminal matters on an individual, circumstance-by-circumstance basis. Accordingly, the question becomes whether DPAs and NPAs can act as companions to the Corporate Leniency Program, or whether these sorts of prosecution agreements, if they should become prevalent and well publicized, would undermine the Program's core purpose. 

III. DPAs and NPAs: Friend or Foe? 

A. An Enforcer's Perspective: DPAs and NPAs "Water Down" the Corporate Leniency Program 

Proponents of NPAs and DPAs tout them as a middle ground approach between criminal prosecution and out-right declination. Indeed, as noted above, many branches of the DOJ have successfully used NPAs and DPAs as effective enforcement tools. However, as also noted, those same branches do not have the alternative of a corporate leniency program, which tends to spare the Antitrust Division the dilemma of whether to indict or "walk away." 

In the antitrust context, the so-called "middle ground" already consists, historically, of cooperation benefits that the Antitrust Division can offer to companies that lose the race for criminal immunity. Generally, the "second-in" company and those that follow can obtain proportionate benefits measured by the value and timing of their cooperation. But, those companies almost always still face criminal prosecution. By denying criminal immunity to anyone but the first-in company, the Antitrust Division sends a clear message: the Corporate Leniency Program will reward the first-in company profoundly better than any later company that self-reports its crimes. 

DPAs and NPAs, of course, offer prosecutors much of the same relief as a guilty plea, including an admission of guilt, large monetary penalties, cooperation, and promises of corporate compliance. Critics are concerned, however, that DPAs and NPAs will not only dilute the Corporate Leniency Program's core message, but also afford those companies who are not first-in benefits that are significantly better than those that would be available to those companies under the Corporate Leniency Program: non-prosecution and, correspondingly, relief from consequences that often flow from criminal conviction, plus an increased opportunity to negotiate benefits. 

1. Could the "Second-in" Get a Better Deal Under a Prosecution Agreement than Under the Corporate Leniency Program? 

Today, a company that approaches the government after the leniency applicant simply cannot predict how it will fare. As noted above, second-in (or later) benefits tend to be proportional, depending on the timing and value of the cooperation received. For example, a second-in company can either significantly advance an investigation or find that its cooperation is duplicative and unnecessary. Thus, second-in benefits depend on a variety of unknowns-typically, the state of the investigation at the time of the cooperation and the nature and extent to which the cooperation advanced the investigation.43 The Antitrust Division seeks to measure the value of the company's cooperation carefully to ensure proportional treatment of cooperating parties. 

While second-in companies may receive some benefits, the Antitrust Division's traditional approach has been to prosecute these companies notwithstanding their cooperation. The purpose of this approach is to incentivize companies to "race" to be the first in the door. Some argue that if DPAs and NPAs become prevalent and highly publicized, companies might begin to believe that they can escape prosecution even without leniency, and thus, may be less interested in winning the race to the door. The availability of DPAs and NPAs could also diminish the interest of second-in (or later) companies to avail themselves of "amnesty plus" by disclosing new conspiracies. Indeed, a "second-in" company might attempt to secure a DPA or NPA on the known violation, instead of disclosing the new violation, if the company determines that the risks-particularly, inevitable civil liability-weigh against disclosure. 

Consequently, by allowing second-in companies to escape prosecution, the use of DPAs or NPAs on a regular basis would not only depart from the Antitrust Division's historical practice; it also could undermine incentives created by DOJ's existing "winner-take-all" and "amnesty plus" practices. 

2. What's the Rush? 

In a speech before the International Workshop on Cartels in 2000, the Antitrust Division's former head of criminal enforcement Scott Hammond said, "The first element common to both deter-ring cartel activity and creating a successful Leniency Program is the threat of severe sanctions.... 
[C]artel activity will not be deterred if the potential penalties are perceived by firms and their executives as outweighed by the potential rewards. If the potential sentences are not sufficiently punitive, then the potential sanctions will merely be seen as a cost of doing business." 44

Essentially, if cartel members do not fear detection, they will not be inclined to self-report their wrongdoing to authorities in exchange for leniency. 45 

DPAs and NPAs create opportunities currently unavailable for cartel members who do not win the race for leniency, and therefore, could arguably cheapen the value of first-in benefits now offered, particularly if DPAs and NPAs are used on a frequent, predictable basis and are well publicized. Under the current regime, the "winner takes all" approach creates a race, which, in turn, breeds tension and mistrust among cartel members, 46 thus creating incentives to be the first in the door-not the second (or worse). 

If DPAs and NPAs became prevalent in antitrust cases, the concern is that leniency would become less attractive because second-in cartel members could avoid criminal prosecution via these alternate routes. If cartel members have more options available, will there be such an intense race to the government? Perhaps not. Similarly, the race to beat the others just to be second or third or so forth-in order to increase the company's chances of receiving proportional benefits-could also be diminished if the proportionality aspect of the Corporate Leniency Program were to become diluted. If a cartel member is better able to predict its penance, what is the rush? 

Accordingly, decelerating the race is not without consequences. The Corporate Leniency Program has been described as the single greatest investigative tool available to anti-cartel enforcement, having been "responsible for detecting and cracking more international cartels than all of our search warrants, secret audio or videotapes and FBI interrogations combined." 47 At the heart of the Program is the race. Getting to the government first, turning in co-conspirators, providing extensive cooperation, shining light on new conspiracies-these are all products of the race that assists the Antitrust Division to successfully detect and prosecute cartels. DPAs and NPAs turn the "winner takes all" approach on its head. If cartel members are no longer incentivized to participate in the race, DOJ investigations, and subsequently, cartel enforcement, will suffer. 

B. Defense Perspective: Prosecution Agreements Complement the Antitrust Leniency Program and Fill the Gap Between Immunity and Prosecution 

DPAs and NPAs offer the DOJ both a carrot and a stick. On the one hand, they complement the Corporate Leniency Program by enhancing second-in benefits, thereby increasing the incentive to cooperate, even where the subject company is ineligible for criminal immunity. On the other hand, these agreements also deter antitrust violations, because they typically impose hefty penalties and structural remedies on wrongdoers, without causing unwarranted collateral consequences. 

1. Prosecution Agreements Incentivize Cooperation 

According to DPA and NPA proponents, prosecution agreements do not "water down" the Corporate Leniency Program because they can incentivize cooperation from more cartel members than the Corporate Leniency Program can on its own. That is particularly true if prosecution agreements (1) become more widespread, (2) are offered in a predictable manner, and (3) are well-publicized. Under these circumstances, businesses that are ineligible for leniency are incentivized to come forward anyway. 

The requirements for Type A leniency, noted above, are strict and may preclude leniency for some companies-for instance, a cartel member that coerced another party to participate in the activity is ineligible for Type A leniency. To obtain Type B leniency, the Division must "determine that granting leniency would not be unfair to others, considering the nature of the illegal activity, the confessing corporation's role in it, and when the corporation comes forward." 48 Because Type B leniency is discretionary, companies cannot be confident they will obtain it. Thus, if a company knows it cannot secure Type A leniency, and believes that circumstances are such that it is unlikely to obtain Type B leniency, the company has little incentive to come forward. 

If the perceived inability to receive either Type A or Type B leniency is in fact currently deterring self-reporting by cartel members, then a policy that offers limited benefits to these companies in exchange for cooperation could strengthen the Antitrust Division's investigatory and enforcement efforts. 

Equally important, from a policy perspective, a form of settlement-without the severe consequences of a guilty plea-can appropriately mitigate the arguable harshness that the Corporate Leniency Program's first-in principle can produce. As the DOJ has acknowledged, "on a number of occasions, the second company to inquire about a leniency application has been beaten by a prior applicant by only a matter of hours." 49 The Antitrust Division has also stated that "there have been dramatic differences in the disposition of the criminal liability of corporations whose respective leniency applications to the Division were very close in time." 50 Such a rigid and formalistic distinction between the first-in and the second-in should not determine whether the company secures immunity or will be subjected to potentially crippling criminal prosecution. 
By offering a prosecution agreement to RBS, the Division implicitly acknowledged that this formalistic approach is not always appropriate. Although RBS was not eligible for leniency, the DOJ "credit[ed] RBS' cooperation in disclosing LIBOR misconduct" 51 and acknowledged that RBS had "provided highly valuable information that expanded and advanced the criminal investigation." 52 

2. A Prosecution Agreement is Not a Free Pass 

Prosecution agreements serve as a "stick" because they allow the government to impose penalties and obtain remedies that might not otherwise result from the Corporate Leniency Program. For example, in its DPA with RBS, the DOJ required RBS to admit the misconduct charged in the information in an extensive and detailed statement of facts. 53 Further, if the government ever pursues criminal prosecution, RBS agreed not to "contest the admissibility of, nor contradict, the Statement of Facts." 54 RBS paid a $150 million penalty, including the $50 million penalty imposed by the Court upon RBS Securities Japan Limited in connection with its guilty plea. 55 Finally, RBS had to expand and strengthen its compliance program and internal controls in order to prevent future instances of LIBOR-related misconduct, fire the individuals who were responsible, agree to continued cooperation in the government's ongoing investigation, 56 and represent that it would not publicly contradict RBS' acceptance of responsibility. 57Thus, the competing policy goals of rewarding cooperation and punishing wrongdoers intersected quite nicely in the RBS case, making it a prime candidate for a prosecution agreement. 

When it comes to the individuals who actually engaged in criminal conduct, prosecution agreements continue to act as a "stick." As Judge Rakoff recently wrote, "Companies do not commit crimes; only their agents do." 58 When a company has no pervasive culture of criminality, and a select few individuals have acted on behalf of the corporation in a criminal fashion, the Corporate Leniency Program's immunity, available to the first-in company, nevertheless allows culpable individuals to go free (so long as they cooperate). 59 A DPA or NPA, however, enables the government to prosecute the individuals who actually committed the crime (so long, of course, as the company is willing to enter the agreement on this basis). 

For example, after the Antitrust Division entered into an NPA with UBS arising out of an investigation into municipal bond bid rigging, the government was able to prosecute various UBS executives, which might not have happened if UBS had received leniency. Again, here, prosecution agreements provide an intermediate position between letting a company and its wrongdoing employees avoid all criminal sanctions, and prosecuting the company while potentially harming the global or national economy in the process. 

IV. Collateral Consequences and Other Considerations 

Notably, prosecution agreements allow the wrong-doing company to avoid or mitigate unintended and harmful collateral consequences of a corporate conviction. These collateral consequences, though arguably rare, can be detrimental when they do occur-not only to the company itself, but also to its innocent shareholders, employees and pension beneficiaries, as well as to the victims of the crime, the company's upstream and downstream partners, the economy more broadly, and, ultimately, consumers themselves. These concerns can be particularly salient in the antitrust context, where, as the Assistant Chief of the Antitrust Division's San Francisco field office Peter Huston put it, "the last thing [antitrust enforcers] want to do is punish someone out of business if that is going to mean a reduction in competition." 60 For example, if a felony conviction resulted in a company being barred from government contracts or required a company to surrender licenses (as in the case of Arthur Andersen), the felony conviction could act as the "corporate death penalty," leading the company to its demise. If several large financial institutions were to be charged with cartel conduct, multiple convictions could seriously destabilize the global economy. 

Although the collateral consequences of a corporate conviction can be potentially disastrous, actual financial ruin has been rare in practice. Moreover, as some might argue, antitrust violations tend to be pervasive and systematic throughout companies, and it is not uncommon for one antitrust investigation to unearth another involving those very same defendants, or overlapping groups of companies. 61 These violations often go to the heart of a business and affect competition in entire industries. Further, antitrust conspiracies often involve multiple companies, and thus, it is hardly clear how the DOJ can decide who among cartel members gets a DPA or NPA, and who gets prosecuted. How can such a distinction be made? And, perhaps most importantly, what message does it send when very serious economic crimes are committed and not all members of the conspiracy are prosecuted accordingly? 

Proponents of these agreements respond that DPAs and NPAs serve the three policy goals of criminal enforcement: they are punitive, rehabilitative and deterrent." 62 Prosecution agreements are punitive because they impose significant financial penalties and can require restitution to victims. 63 They facilitate rehabilitation by imposing structural remedies designed to prevent any similar violations from happening again. 64 Lastly, DPAs and NPAs serve as general deterrents for the obvious reason that they can impose severe penalties and obligations, as well as become admissible (in certain cases) as party-opponent admissions in follow-on civil litigation. 65 They also serve as specific deterrents because the government can resume or initiate prosecution if the company breaches the agreement. 66 

One might argue, however, that prosecution agreements do not act as a deterrent to other companies because they suggest a willingness on the government's part to entertain the "too big to jail" defense. In other words, to the extent prosecution agreements become more common in antitrust cases, individuals engaged in antitrust misconduct might believe they have a better chance of escaping prosecution than they currently have under the Corporate Leniency Program. Nonetheless, prosecution agreements are not guaranteed, can impose severe penalties, and generally require the company to terminate the employment of those involved. Furthermore, those individuals can still be prosecuted, even after the company enters into the agreement, if they are "carved out." (The carve out feature, however, is also available and used today by the Antitrust Division with second-in cartel members.) 

V. The Effects of DPAs and NPAs on Follow-on Civil Litigation 

Major criminal antitrust investigations typically spark almost immediate civil treble damage litigation by private plaintiffs. When a company under investigation for criminal antitrust charges enters into a plea agreement with the Antitrust Division, that agreement can negatively affect the defendant's position in the related civil litigation and present important advantages for civil plaintiffs. While DPAs and NPAs can require an admission of guilt, courts vary in their treatment of such admissions in follow-on civil litigation. 

For their part, guilty pleas are prima facie evidence of an antitrust violation in follow-on civil litigation under Section 5(a) of the Clayton Act, with some exceptions. 67 Thus, the guilty plea "create[s] a rebuttable presumption in favor of the plaintiff and against each defendant against whom the government judgment was entered." 68 Further, Federal Rule of Evidence 803(22) allows plaintiffs to use a guilty plea to establish facts related to a claim. The rule provides that a final judgment entered after a guilty plea for sufficiently serious crimes is admissible over a hearsay objection to prove "any fact essential to the judgment." 69 

Additionally, information in one defendant's guilty plea can be used as evidence against other defendants to the extent that the others are specifically implicated in the conduct underlying the plea. 70 Thus, the factual and legal allegations in the plea, and statements made during the court hearing at which the plea is entered, present potential challenges for an antitrust defendant in follow-on civil litigation. At the same time, they afford opportunities for plaintiffs in civil litigation to benefit significantly, not only at trial, but also in opposing motions to dismiss, and even in seeking at least partial summary judgment. 

On the one hand, plaintiffs might argue that admissions made in connection with a non- or deferred prosecution agreement are no different than admissions made in connection with a guilty plea. Accordingly, they reason, courts should treat admissions in prosecution agreements as admissible prima facie evidence of liability. 71 If courts do not give prosecution agreements as much weight as guilty pleas, DPAs and NPAs could arguably hinder the civil process by allowing antitrust offenders to escape the civil consequences they would have faced had they pleaded guilty. 

On the other hand, defendants argue that, unlike guilty pleas, prosecution agreements do not reflect adjudications of facts, and are thus immaterial or irrelevant. Accordingly, they argue, allegations based on prosecution agreements should be struck pursuant to Rule 12(f) of the Federal Rules of Civil Procedure. 72 An example of unadjudicated facts being used in this manner is playing out in In re LIBOR-Based Financial Instruments Antitrust Litigation, where plaintiffs have used RBS' various agreements with the DOJ, which expressly cover the alleged manipulation of LIBOR as to certain currencies-not including the U.S. dollar-as a factual basis for claims that are based on the U.S. dollar and other currencies that were not covered by the prosecution agreement. 73 It is controversial whether "facts" like these, where the nature and extent of RBS' role in LIBOR misconduct cannot be gleaned from the agreement with the DOJ, should-as a policy matter-be allowed to support a civil claim against RBS. 

Finally, defendants in other cases have also argued that prosecution agreements are inadmissible hearsay 74 or are inadmissible under Federal Rule of Evidence 410(a)(4), which prohibits the use of "statement[s] made during plea discussions...if the discussions did not result in a guilty plea." 75 Again, some fear that suppressing evidence of prosecution agreements could lead to some second-in companies-that would otherwise face civil consequences for a guilty plea being able to avoid civil liability. On the other hand, defendants argue that civil plaintiffs should, and do, bear the burden of proving their claims without relying only on unadjudicated allegations. 

VI. Looking Forward: In Which Cases Might the Division Negotiate a Prosecution Agreement? 

The Antitrust Division apparently does not have immediate plans to increase its use of prosecution agreements over plea agreements in criminal cases. In fact, the former head of the Antitrust Division's criminal enforcement, Scott Hammond, distinguished the RBS case from other antitrust investigations, explaining that a DPA was appropriate there, at least in part because the banking industry is already heavily regulated and the Antitrust Division was working together with the Criminal Division. 76 Hammond went on to state that this "resolution is not an indication that we've changed our policy, [it's] an indication that just the facts of this particular case merited that result." 77 Yet, given the potential collateral consequences of prosecuting large financial institutions, it could be asserted that prosecution agreements are appropriate under similar circumstances. 

Going forward, one could speculate that prosecution agreements may also be appropriate in unregulated industries with large-scale global impact, where compliance programs are critical, as well as situations where the Antitrust Division is working with another DOJ division. Indeed, the former head of the Antitrust Division's National Criminal Enforcement Section, Tony Nanni, commented that the Division had "wisely modified [the practice of avoiding DPAs and NPAs] when the alleged conduct at issue is properly seen as primarily a criminal fraud rather than an antitrust violation." 78 

VII. Conclusion

DPAs and NPAs are still infrequent in antitrust cases. However, the recent uptick of prosecution agreements for financial institutions involved in price fixing and bid rigging suggests that the Antitrust Division could be open to exploring these agreements in future similar cases. While prosecution agreements offer benefits to criminal enforcement beyond those available under the Corporate Leniency Program, it is not clear whether more frequent use of these agreements would "chill" the willingness of companies to come forward under the Leniency Program. 


1. See Press Release, U.S. Departmetn of Justice, RBS Securities Japan Limited Agrees to Plead Guilty in Connection with Long-Running Manipulation of Libor Benchmark Interest Rates, Feb. 6, 2013, available at (last visited Feb. 5, 2015) [hereinafter RBS Press Release]; see also Deferred Prosecution Agreement, United States of America v. The Royal Bank of Scotland PLC, No. 3:13CR73 (MPS) 
(D.C. Conn. Feb. 5, 2013), available at (last visited Feb. 5, 2015) [hereinafter RBS Deferred Prosecution Agreement]; Plea Agreement, United States of America v. The Royal Bank of Scotland PLC, No. 3:13CR73 (MPS) (D.C. Conn. Feb. 5, 2013), available at (last visited Feb. 5, 2015). 
2. See id. 
3. See Accountability, Transparency, and Uniformity in Corporate Deferred and Non-Prosecution Agreements Before the House Comm. on the Judiciary, 4 (June 25, 2009) (statement of Gary G. Grindler, Deputy Assistant Attorney General, Criminal Division, U.S. DOJ), available at (last visited Feb. 5, 2015) [hereinafter Grindler Statement]; see also, e.g., Non-Prosecution Agreement, United States of America v. UBS AG (Dec. 18, 2012), available at (last visited Feb. 5, 2015). 
4. See Grindler Statement, supra note 3, at 3-4. 
5. See Eugene Illovsky, Corporate Deferred Prosecution Agreements, The Brewing Debate, 2006 A.B.A. Sec. Crim. Just. 36, at 36, available at (last visited Feb. 5, 2015). 
6. See id. 
7. See id. at 36. 
8. Memorandum from Larry D. Thompson, Deputy Attorney General, to Heads of Department Components, United States Attorneys 6 (Jan. 20, 2003), available at (last visited Feb. 5, 2015) [hereinafter Thompson Memorandum] (emphasis added). 
9. See id. at 3. 
10. 9 U.S.A.M. § 28.1000(B) (Aug. 2008), available at (last visited Feb. 5, 2015) [hereinafter U.S. Attorneys' Manual, Title 9]; see also U.S. Attorneys' Manual, Title 9, at §§ 28.000-28.1300. 
11. U.S. Attorneys' Manual, Title 9, supra note 10, at § 28.300(A). 
12. Id. at (A)(2), (A)(5) (citations omitted). 
13. Id. at (A)(7), (A)(8) (citations omitted). 
14. 2012 Year-End Update on Corporate Deferred Prosecution Agreements (DPAs) and Non-Prosecution Agreements (NPAs), Gibson Dunn & Crutcher LLP, Jan. 3, 2013, at 2, available at (last visited Feb. 5, 2015). 
15. 2013 Year-End Update on Corporate Non-Prosecution Agreements (NPAs) and Deferred Prosecution Agreements (DPAs), Gibson, Dunn & Crutcher LLP, Jan. 7, 2014, at 2, available at (last visited Feb. 5, 2015). 
16. 2011 Year-End Update on Corporate Deferred Prosecution and Non-Prosecution Agreements, Gibson, Dunn & Crutcher LLP, Jan. 4, 2012, at 4, available at (last visited Feb. 5, 2015). 
17. Deferred Prosecution Agreement, United States of America v. Coöperatieve Centrale Raiffeisen-Boerenleenbank B.A., No. 3:13CR200 (AWT) (D.C. Conn. Oct. 29, 2013), available at (last visited Feb. 5, 2015); Information,United States of America v. Coöperatieve Centrale Raiffeisen-Boerenleenbank B.A., 3:13CR200(AWT) (D.C. Conn. Oct. 29, 2013), available at (last visited Feb. 5, 2015). 
18. RBS Deferred Prosecution Agreement, supra note 1; Information, United States of America v. The Royal Bank of Scotland PLC, No. 3:13CR73 (D.C. Conn. Apr. 12, 2013), available at (last visited Feb. 5, 2015). 
19. See U.S. DOJ, Antitrust Division, Corporate Leniency Policy (1993), available at (last visited Feb. 5, 2015) [hereinafter Leniency Policy]; see generally Scott D. Hammond, The 24th Annual National Institute on White Collar Crime: The Evolution of Criminal Antitrust Enforcement over the Last Two Decades, U.S. DOJ, Feb. 25, 2010, available at (last visited Feb. 5, 2015). 
20. See Scott D. Hammond et al., Frequently Asked Questions Regarding the Antitrust Division's Leniency Program and Model Leniency Letters, U.S. DOJ, Nov. 9, 2008, at 1, available at (last visited Feb. 5, 2015) [hereinafter FAQs]. 
21. See id. 
22. Id. at 4. 
23. Id. 
24. Id. at 5. 
25. Id. 
26. Id. 
27. Id. 
28. See id. at 23-24. 
29. See id. at 24. 
30. See Scott D. Hammond, Measuring the Value of Second-In Cooperation in Corporate Plea Negotiations, U.S. DOJ, The 54th Annual American Bar Association Section of Antitrust Law Spring Meeting, Mar. 29, 2006, at 1, available at (last visited Feb. 5, 2015) [hereinafter Second-In Cooperation]. 
31. See FAQs, supra note 20, at 8-9. 
32. See id. 
33. Second-In Cooperation, supra note 30, at 1. 
34. Id. 
35. See In re Aftermarket Auto. Lighting Products Antitrust Litigation, No. 09 MDL 2007-GW(PJWx), 2013 U.S. Dist. LEXIS 126308, at *8-9 n.5 (C.D. Cal. Aug. 26, 2013). 
36. See id. 
37. See Information, United States v. Sabry Lee (U.S.A.) Inc., No. CR 11-0599 WHA (N.D. Cal. Aug. 30, 2011), available at (last visited Feb. 5, 2015); Information, United States v. Chen, No. CR 11-0166 RS (N.D. Cal. Mar. 22, 2011), available at (last visited Feb. 5, 2015). 
38. See Plea Agreement, United States v. Sabry Lee (U.S.A.), Inc., No. CR 11-0599 RS, at *2-3 (N.D. Cal. Oct. 5, 2011), available at (last visited Feb. 5, 2015); Plea Agreement,United States v. Chen, No. CR 11-0166 RS, at *2-3 (N.D. Cal. May 25, 2011), available at (last visited Feb. 5, 2015). 
39. See Tr. of Proceedings, Sentencing Hearing at 6-12, United States v. Chen, No. 11-cr-166 (N.D. Cal. Feb. 5. 2013). 
40. Id. at 8-9. 
41. Id. at 26-28. 
42. Amended Judgment at 2, United States v. Sabry Lee (U.S.A.), Inc., No. 11-cr-599 (N.D. Cal. Oct. 6, 2011). 
43. See Second-In Cooperation, supra note 30, at 2. 
44. Scott D. Hammond, Detecting and Deterring Cartel Activity through an Effective Leniency Program, U.S. DOJ International Workshop on Cartels, Nov. 21-22, 2000, at 2, 3, available at (last visited Feb. 5, 2015). 
45. See id. at 4. 
46. Id. at 5.' 
47. Id. at 1. 
48. Leniency Policy, supra note 19, at 3; FAQs, supra note 20, at 15. 
49. FAQs, supra note 20, at 2. 
50. Id. at 5-6. 
51. RBS Press Release, supra note 1. 
52. RBS Deferred Prosecution Agreement, supra note 1, at 5. 
53. Id. at 2-3, att. A. 
54. Id. at 2-3. 
55. Id. at 8-9, 10-11. 
56. Id. at 4-5. 
57. Id. at 16-17. 
58. Jed S. Rakoff, The Financial Crisis: Why Have No High-Level Executives Been Prosecuted?, N.Y. Review of Books, Jan. 9, 2014, available at (last visited Feb. 5, 2015). 
59. Leniency Policy, supra note 19, at 4 (providing that individuals who admit to their involvement in the misconduct will receive leniency "if they admit their wrongdoing with candor and completeness and continue to assist the Division throughout the investigation"). 
60. 2013 Mid-Year Update on Corporate Deferred Prosecution Agreements (DPAs) and Non-Prosecution Agreements (NPAs), Gibson, Dunn & Crutcher LLP, July 9, 2013, at 8, available at (last visited Feb. 5, 2015) (citing Ron Knox, DOJ Official: In Some Industries, NPAs Needed to Protect Competition, Global Competition Review, Feb. 25, 2013). 
61. As the Antitrust Division recently reported regarding its massive criminal auto parts investigation. 
To date, the investigation has resulted in charges against 26 companies and 29 individuals and more than $2 billion in criminal fines for participation in conspiracies to fix prices of and rig bids on automobile parts, including safety systems such as seat belts, air bags, steering wheels, and antilock brake systems, and critical parts such as antivibration rubber, instrument panel clusters, starter motors, and wire harnesses. Division Update Spring 2014, Criminal Program, U.S. DOJ, (last visited Feb. 5, 2015). 
62. See Assistant Attorney General Lanny Breuer,Address to the New York City Bar Association, The Role of Deferred Prosecution Agreements in White Collar Criminal Law Enforcement (Sept. 13, 2012), available at (last visited Feb. 5, 2015). 
63. See id. 
64. See id. 
65. See id. 
66. See id. 
67. See 15 U.S.C. § 16(a); see also Jonathan I. Gleklen, Antitrust Law Developments, 2012 ABA Section of Antitrust Law 933. 
68. Id. at 934 (citation omitted). 
69. Fed. R. Evid. 803(22); see also Criminal Antitrust Litigation Handbook, 2006 ABA Section of Antitrust Law 56. Note also, that the Antitrust Division has a long-standing policy against accepting a nolo contendere plea, which would not be admissible in a civil litigation. See Scott D. Hammond, The U.S. Model of Negotiated Plea Agreements: A Good Deal with Benefits for All, U.S. Department of Justice, 23 n.18, Oct. 17, 2006, available at (last visited Feb. 5, 2015); see generally U.S. Attorneys' Manual, Title 9, supra note 10, at § 16.010; See U.S. Department of Justice, Grand Jury Manual, at IX-71 (1st ed. 1991), available at (last visited Feb. 5, 2015). 
70. United States ex rel. Miller v. Bill Harbert Int'l Constr., Inc., 608 F.3d 871, 891-92 (D.C. Cir. 2010) (admitting proof of one defendant's prior guilty plea to a Sherman Act violation as "relevant pieces of evidence that are admissible against all defendants"). 
71. Some courts have cited prosecution agreements as a factor in denying motions to dismiss, acknowledging that they are not final adjudications of the facts, but allowing plaintiffs to rely on them in order to get over the pleading hurdles of Twombly and Iqubal. See, e.g., Thompson's Gas & Elec. Serv., Inc. v. BP Am. Inc., 691 F. Supp. 2d 860 (N.D. Ill. 2010) (denying, in part, motion to dismiss market manipulation claims that were the subject of a DPA with the CFTC, and noting that the DPA was an "alternative to adjudication"); Davis v. Beazer Homes, U.S.A. Inc., No. 1:08CV247, 2009 U.S. Dist. LEXIS 107253, at *23-25 (M.D.N.C. Nov. 17, 2009) (considering a prosecution agreement as a "significant factor in assessing the 'plausibility' of one of the plaintiff's claims, and recommending denial of a motion to dismiss that claim, but noting that the DPA was not a "final adjudication of the matters addressed therein"). 
72. See Lipsky v. Commonwealth United Corp., 551 F.2d 887, 893-94 (2d Cir. 1976) (holding that a consent decree "can have no possible bearing" on the dispute before the court because it was not a "true adjudication[ ]of the underlying issues," and that it could "only be introduced in a later trial for collateral estoppel purposes if the issues sought to be precluded were actually adjudicated in the prior trial"); Footbridge Ltd. Trust v. Countrywide Home Loans, Inc., No. 09 Civ. 4050 (PKC), 2010 U.S. Dist. LEXIS 102134, at *1445 (S.D.N.Y. Sept. 28, 2010) (striking allegations in a complaint that were "based on pleadings, settlements, and government investigations in other cases"). 
73. [Corrected] Second Amended Consolidated Class Action Complaint [224, In re Libor-Based Fin. Instruments Antitrust Litigation, Nos. 1:11-md-2262, 11-cv-2613 (S.D.N.Y. Sept. 30, 2013), ECF No. 438. 
74. In re Oil Spill by the Oil Rig "Deepwater Horizon," MDL No. 2179 Section: J(1), 2012 U.S. Dist. LEXIS 15902, at *13 (E.D. La. Feb. 9, 2012) (granting motion in limine to exclude DPA as inadmissible hearsay). 
75. Fed. R. Evid. 410(a)(4); see Escue v. Sequent, Inc., No. 2:09-cv-765, 2012 U.S. Dist. LEXIS 9949, at *25-31 (S.D. Ohio Jan. 25, 2012) (denying motion in limine to exclude plea discussions and DPA on the grounds that defendants had provided no information demonstrating their expectation that they were negotiating a plea agreement and that that expectation was reasonable). 
76. Melissa Lipman, DOJ Not Wavering on Antitrust Guilty Pleas, Official Says, Law360, Apr. 10, 2013, available at (last visited Feb. 5, 2014). 
77. Id. 
78. Charles F. (Rick) Rule et al., Antitrust Division Enters into First Deferred Prosecution Agreement, Lexology, Feb. 14, 2013, available at (last visited Aug. 1, 2014). 

Reprinted with permission from: NY Litigator, Fall 2014, Vol. 19, No. 2, published by the New York State Bar Association, One Elk Street, Albany, NY 12207. 

Insights & Analyses

We're attuned to (often ahead of) the key issues facing the market. Our archive is an arsenal. It's about arming investors with the information they need to make sound decisions.