Excerpt of article which first appeared in The Antitrust Source, February, 2008. (www.abanet.org/antitrust/at-source/at-source.html)

Since 1993, the U.S. Department of Justice Antitrust Division’s Corporate Leniency Program has been instrumental in the Antitrust Division’s crusade against antitrust violators. In fact, antitrust violators have entered the Leniency Program at rates as high as two per month, resulting in the prosecution of some of the Antitrust Division’s biggest cases.

The Leniency Program’s success is due largely to its ability to offer an incentive that is simply too good to pass up. If accepted into the Leniency Program, a company can avoid all criminal penalties (both for itself and its officers and employees) for its antitrust violations, as long as the company complies with its obligations under its agreement with the Antitrust Division. This is a tremendous inducement for corporations that would otherwise face multi-million dollar fines and prison terms for their executives.

In 2004, the Leniency Program’s benefits were statutorily enhanced through the de-trebling of civil damages for successful leniency applicants, increasing the incentive for a company to come forward. But the recent completion of the Stolt-Nielsen saga—revocation of leniency, followed by indictment, and then dismissal of charges—raises questions about the value and requirements of the Leniency Program in the future, as well as the risks associated with participating in the program. Should companies still seek such leniency? If so, how should they make sure that their leniency is permanent?

Stolt-Nielsen: From Leniency to Indictment to Vindication

Background. Stolt-Nielsen S.A., a Luxembourg shipping company, participated in a customer-allocation, price-fixing, and bid-rigging conspiracy with two other shipping companies, Odfjell Seachem AS and Jo Tankers—a classic per se antitrust violation with serious exposure for criminal fines and jail time. When responsible officials at Stolt-Nielsen discovered the violation, the company sought and received protection under the Leniency Program.

Participation in the Leniency Program depends on the applicant’s ability to satisfy several conditions. Where an investigation has not yet begun (that is, the applicant’s self-reporting is truly the cause of a subsequent investigation), leniency is subject to these conditions: (1) the Division has not received information about the illegal activity from any other source; (2) on discovering the illegal activity, the company “took prompt and effective action to terminate its part in the activity”; (3) the company reports the wrongdoing “with candor and completeness” and continues to provide full and complete cooperation throughout the investigation; (4) the confession of wrongdoing is “truly a corporate act,” (as opposed to isolated confessions of individual executives or officials); (5) the company makes restitution to injured parties (where possible); and (6) the company was not the leader or originator and did not coerce another party to participate in the illegal activity. If the Division has already begun an investigation (or has received information about the activity at issue), a company can still obtain leniency if a three-pronged test is satisfied: (1) the company is the “the first one to come forward and qualify for leniency,” (2) the Division does not yet have evidence that is “likely to result in a sustainable conviction” of the company, and (3) granting leniency would not be “unfair to others.”

Stolt-Nielsen came forward after the Antitrust Division had already begun an investigation, but it met all of the conditions to the satisfaction of the Antitrust Division, at least initially. Stolt-Nielsen provided the Division with “volumes of highly incriminating evidence” concerning its role in the customer allocation conspiracy. This information allowed the Antitrust Division to prosecute Stolt-Nielsen’s co-conspirators: Odfjell was fined $42.5 million and two of its executives served prison terms and were fined personally; Jo Tankers was fined $19.5 million and one of its executives served a prison term and was fined personally. Indeed, according to the district court, these convictions would not have been possible without Stolt-Nielsen’s cooperation.

Stolt-Nielsen took extensive internal measures to comply with the obligation to take prompt and effective action to end the illegal activity, including:

  • Instituting a new antitrust policy and publishing an Antitrust Compliance Handbook;
  • Distributing the Compliance Handbook to all employees and competitors;
  • Holding mandatory seminars for all employees on antitrust compliance;
  • Requiring all employees to sign certifications that they would comply strictly with all terms the new Antitrust Compliance Policy; and,
  • Informing competitors of the new policy and of Stolt-Nielsen’s intent to comply with it.

In addition to informing its competitors of its new compliance policy, Stolt-Nielsen also began competing with its co-conspirators on at least some accounts—as the district court would later find.

Notwithstanding these steps, Stolt-Nielsen’s perceived lack of compliance with ending its antitrust violations gradually became a point of contention with the Antitrust Division. Specifically, the Antitrust Division did not believe that Stolt-Nielsen ended its illegal activities “promptly” but rather continued its anticompetitive conduct in subsequent meetings with its co-conspirators. The Antitrust Division’s suspicion arose largely from allegations from one of Stolt-Nielsen’s former coconspirators who claimed that Stolt-Nielsen did not end its anticompetitive activity. The Antitrust Division eventually found six other witnesses, all former conspirators, willing to corroborate that account.

From Leniency to Litigation. The Antitrust Division asserted that Stolt-Nielsen had violated its leniency agreement by failing to promptly withdraw from the antitrust conspiracy. As a result, on April 8, 2003, the Antitrust Division began the process of revoking Stolt-Nielsen’s leniency. The obligation to cooperate was suspended, an executive was arrested, and leniency was formally revoked.

By filing a suit to enjoin the Antitrust Division from indicting the company and its executives, Stolt-Nielsen preempted the Antitrust Division’s plan to obtain a grand jury indictment of the company. The District Court for the Eastern District of Pennsylvania found that Stolt-Nielsen had not breached the agreement and enjoined the Antitrust Division from revoking leniency. The Antitrust Division appealed, and the Third Circuit reversed on the grounds that the constitutional principle of separation of powers prohibited the district court from enjoining the prosecution. On appeal, the Third Circuit found that the non-prosecution agreement could not serve as a basis for enjoining an indictment, but the court made clear that the agreement could be asserted as a defense after indictment. Thus, on remand, when the company raised the non-prosecution agreement as a defense to an indictment, the district court would then be free to consider the agreement “anew,” and, among other things, consider whether the defendants fulfilled their obligations under the agreement.

Dismissal of the Indictment. Following the Third Circuit’s decision, Stolt-Nielsen and two of its executives were indicted. Before trial, the defendants moved for dismissal of the indictment upon the basis of a violation of the non-prosecution agreement. The motion was heard by a new judge, who found that the Antitrust Division violated the non-prosecution agreement, and dismissed the indictment.

The district court in Stolt-Nielsen III used a defense-friendly principle of interpretation for nonprosecution agreements. The court held that non-prosecution agreements are unique contracts that must be construed in light of the important constitutional rights at stake. A central question in adjudicating the dispute is whether the Antitrust Division’s “conduct comported with ‘what was reasonably understood by the defendant when entering’ the Agreement.” The court explained that the Antitrust Division may not rely on a “rigidly literal” construction of the agreement; rather, it “bears the burden of demonstrating that [the defendant] materially breached the Agreement.” In determining whether a breach is material, the most important factor is the incriminating nature of the evidence provided by the defendant—whether or not the government has received the benefit of its bargain. The court did not decide what quantum of proof was required to show such a material breach—whether “clear and convincing” evidence was necessary or whether a “preponderance of the evidence” was sufficient. The court found no need to reach this issue because here the Antitrust Division had not offered sufficient proof to meet the preponderance standard, much less the “clear and convincing” standard.

Why did the court find that a material breach was not established? The non-prosecution agreement required Stolt-Nielsen to take “prompt and effective action to terminate its part in the anticompetitive activity being reported upon discovery of the activity.” The Antitrust Division alleged that Stolt-Nielsen failed to live up to this obligation. Based on the testimony of seven executives at Odfjell and Jo Tankers, the Antitrust Division alleged that Stolt-Nielsen continued to collude on the allocation of three shipping contracts. The key to the court’s rejection of this assertion was that Stolt-Nielsen’s actions to end the antitrust violations were deemed “prompt and effective.” The court found that, “by its plain meaning, [prompt and effective action] requires a prompt and diligent process, and does not require immediate termination of all anti-competitive activity.”

The court said that this approach is what defendants would have reasonably understood.

The court readily found that Stolt-Nielsen had satisfied the requirement of “prompt and effective action” through its “large-scale effort” to “eliminate anticompetitive activity at all levels of the company, including senior management.” The court’s finding was supplemented by evidence that Stolt-Nielsen followed its Antitrust Compliance Policy by engaging in “genuine competition” on contracts previously allocated under the conspiracy. For example, a Stolt-Nielsen executive refused to withdraw a bid for a contract that was formerly allocated by conspiracy—defying the demands of a former co-conspirator. (The former co-conspirator did win that contract, but not because of anticompetitive activity by Stolt-Nielsen—the customer gave Stolt-Nielsen’s competitor a “last look” that allowed it to win the contract.) In another case, Stolt-Nielsen significantly reduced its rates to retain a contract.

The court found Stolt-Nielsen’s evidence to be more credible than that offered by the Antitrust Division. Stolt-Nielsen was able to provide corroborating testimonial and documentary evidence to support its position. In contrast, the Antitrust Division’s witnesses were discredited by their own contradictions as well as their incentives to be untruthful. The government’s witnesses were former co-conspirators agreeing to testify in exchange for reduced sentences, and they all withered under Stolt-Nielsen’s impeachment. Some of the government’s witnesses offered testimony that did not even support the argument that Stolt-Nielsen continued to engage in anti-competitive activity after obtaining its leniency agreement. The Antitrust Division alleged that Stolt-Nielsen had entered into a quid pro quo agreement to allocate some shipping contracts with Jo Tankers, only to have their star witness deny the existence of such an agreement. This put the Antitrust Division in the unenviable position of impeaching its own witness. The Antitrust Division fared only slightly better with its other witnesses. One witness claimed that he could not remember the details of a meeting in which Stolt-Nielsen had informed him of its intent to comply with its Antitrust Compliance Policy, but two other witnesses did remember the conversation. Another witness misstated basic facts about the contracts that were allegedly still allocated by conspiracy and then went on to state that he had “no clue” who drafted his grand jury declaration. Accordingly, the court rejected the Antitrust Division’s arguments and dismissed the indictment.

On December 21, 2007, three weeks after Stolt-Nielsen III was decided, the Antitrust Division announced that it would not appeal the dismissal of the indictment.

The Legacy of Stolt-Nielsen

So what is the practical effect of the Stolt-Nielsen saga? Two preliminary observations are certain and worth mention. First, corporations will continue to seek leniency under this program. Second, the Antitrust Division will continue to grant leniency. Corporations continue to have a tremendous incentive to cooperate. As before Stolt-Nielsen, the risk of leniency revocation will remain small: the Stolt-Nielsen revocation was the first since the current program’s debut in 1993, and the Antitrust Division stated that the action was “not take[n] lightly” and was “regrettabl[e]” but “necessary.” The benefits to both the companies seeking leniency, and the Antitrust Division, continue to be real and significant. It is improbable that companies will cease coming forward simply as the result of one attempted revocation. Instead, there likely will be much more oversight by the Antitrust Division into the actions of a company seeking leniency and its employees. In addition, it is likely that less ambiguous language will be used in future agreements so that the standards defining the company’s expected conduct are clearer. Clear standards carry with them two primary implications. First, clearer guidelines assist companies in their attempts at compliance and therefore may make it less likely that the Antitrust Division will be tempted to revoke their leniency. Second, because any revocation of leniency by the Antitrust Division will be made on a clearer record, it is more likely that, if the Antitrust Division chooses to revoke leniency in a future case, it will prevail.

Indeed, the Third Circuit’s decision in favor of the Government’s authority to indict did not seem to affect the steady stream of applicants for the Corporate Leniency Program. Even after the Antitrust Division’s revocation of leniency for Stolt-Nielsen and before the district court’s dismissal of the indictment, numerous companies continued to contact the Antitrust Division in an attempt to be the first in the door to qualify for the Leniency Program—including companies such as Virgin Atlantic and Lufthansa in Summer 2007. Simply put, the leniency-revocation litigation and the attendant uncertainty did not deter companies from seeking the benefits of the program.

Likewise, from the government’s viewpoint, the Leniency Program is far too valuable to permit it to fall into disuse. Significantly, the Antitrust Division has recognized that the Corporate Leniency Program needs to have a fairly high degree of certainty, predictability, and freedom from prosecutorial discretion, but it does not have be “risk free” in order to be an attractive option for antitrust violators. The Antitrust Division has described the program as its “most effective investigative tool” and “a model for similar corporate leniency programs . . . adopted by antitrust authorities around the world.” The program “has resulted in scores of convictions and nearly $4 billion in criminal fines” and has been a material source of information in “the majority of the Division’s major international investigations.” The Department of Justice cannot afford to turn off the spigot from which flows so much of its success in breaking up cartels.

Immediate reactions to Stolt-Nielsen III characterized the court’s decision as saving the Antitrust Division from its own error in judgment: the Antitrust Division never should have sought to revoke leniency in this particular case, much less indicted the company. This characterization may be true to an extent, but Stolt-Nielsen III is not as significant a decision as some might suggest. The district court did adopt a defense-friendly standard for reviewing compliance with agreements under the Leniency Program, but the case ultimately turned on the underlying facts. The Department of Justice’s decision not to appeal likely had more to do with not wanting to create bad law by appealing a case with bad facts or insufficient evidence. That is, the Department of Justice decided to limit Stolt-Nielsen III’s precedential impact and to confine the government’s loss.

Stolt-Nielsen III’s practical effect will be on the likelihood and imminence of a second leniency revocation. The Antitrust Division made clear that the revocation of Stolt-Nielsen’s leniency, though regrettable, was necessary to “to maintain the integrity of the program.” Certainly, at some point in the future, it is likely the Antitrust Division will again revoke a grant of leniency given to another corporation. Nevertheless, it is also true that the Antitrust Division is not going to risk bringing such further action unless it concludes that the conduct of the company granted leniency jeopardizes the integrity of its premier investigative tool and that the facts are strongly in its favor. Thus, the clarity, certainty, and severity of noncompliance will likely have to be significant before the Division will take action.

Accordingly, companies also should expect more scrutiny from the Antitrust Division regarding the details of leniency agreements. Stolt-Nielsen III turned in large measure on the court’s perception of the parties’ understanding and intention in the leniency agreement. To limit the effects of Stolt-Nielsen III, the Antitrust Division presumably will require greater specificity in leniency agreements—provisions that will demonstrate more clearly what the leniency applicant was bargaining away and what specifically will be required. Future participants in the Leniency Program may well be subject to more clearly defined agreements and may also be required to make a stronger showing that they meet the conditions of leniency. Consequently, Stolt-Nielsen III may actually mean that companies can have a marginally greater level of comfort about retaining the leniency after they are accepted into the Leniency Program,

But Stolt-Nielsen III does not mean that companies should test the limits of how far they can go before being deemed noncompliant. The risk of becoming the next revocation case may be small, but the cost is potentially staggering: Even though Stolt-Nielsen prevailed, it did so only after years of litigation—years in which the company and its cooperating officers and employees faced both expense and significant high-stakes risk. Additionally, had the district court denied the motion to dismiss, Stolt-Nielsen would have been in a far worse position than it would had it not cooperated in the first place.


Stolt-Nielsen III is unlikely to affect the fundamentals of the Leniency Program, because there is too much at stake both for the applicants and for the Department of Justice. Nevertheless, leniency applicants can expect greater clarity and specificity of requirements in their leniency agreements, and successful applicants must plan to live up fully to those commitments

Featured article in Dorsey's Spring 2008 White Collar Crime and Civil Fraud Update