Corporate Crime, Prosecutorial Discretion, and Separation of Powers: U.S. v. Fokker Services

Many criminal investigations of corporations are resolved by a Deferred Prosecution Agreement (DPA) or its less frequently used cousin, a Non-Prosecution Agreement (NPA). These are essentially negotiated settlements between the Department of Justice and the defendant, in which the company agrees to certain sanctions and changes in behavior in exchange for avoiding criminal penalties.

Typically the terms and conditions of such agreements are set by the prosecutors. Decisions about whether to charge at all, what charges to bring, and the terms of any resolution are at the core of prosecutorial discretion. But in 2015, in an unprecedented decision, U.S. District Judge Richard J. Leon rejected a DPA between prosecutors and a company called Fokker Services because he thought the company had gotten too sweet a deal. The U.S. Court of Appeals for the District of Columbia Circuit recently reversed that decision, chiding the judge for overstepping his bounds.

As I wrote in this earlier post, I have concerns about the increased use of DPAs and the resulting effect on the criminal justice system. But regardless of how one feels about the merits of DPAs, the D.C. Circuit’s decision is a welcome reaffirmation of the importance of separation of powers and the proper role of the judiciary.

DPAs, NPAs, and Corporate Crime

The use of DPAs has been on the rise over the past decade ever since the Arthur Andersen debacle. The huge accounting company was driven out of business – and tens of thousands of people lost their jobs – as a result of a criminal prosecution that was later thrown out by the Supreme Court. To avoid such a potential “corporate death penalty,” companies have a tremendous incentive to enter into a DPA and avoid a criminal conviction.

In a DPA, the government files criminal charges but agrees to put the prosecution of those charges on hold. Under the agreement, the company generally admits to the charges and may agree to pay fines or restitution, undergo internal reforms, cooperate in the prosecution of individual employees, and take other remedial steps to atone for its misbehavior. In many DPAs the government requires the company to hire a monitor, at the company’s expense, to oversee its compliance with the agreement. The government agrees that when the DPA ends, usually in a few years, it will dismiss the criminal charges if the company has fully complied.

A NPA may contain the same kinds of terms as a DPA. The biggest difference is that in a NPA there are no charges filed with the court – it’s just a private agreement between the company and the prosecution. A NPA thus lacks the imprimatur of a court proceeding and the gravity of charges actually being filed. And because there is no public court filing, a NPA could potentially take place without the public even knowing about it.

Fokker Services logo

United States v. Fokker Services B.V.

Fokker Services is a Dutch aerospace services company. In 2010 Fokker self-reported to the government that it may have violated federal export control laws in its dealings with Burma, Sudan and Iran. Over the next four years, Fokker cooperated with the government in a wide-ranging investigation of this misconduct. The investigation uncovered more than 1,000 illegal transactions, from which Fokker earned about $21 million.

As part of its cooperation, Fokker instituted a number of internal reforms. It also fired its president and demoted or reassigned a number of other employees involved. The government described Fokker’s remedial efforts as “a model to be followed by other corporations.” In light of this cooperation, the government negotiated a DPA with Fokker under which the company would continue its compliance and remediation efforts for another 18 months and pay a $21 million fine.

Pursuant to the DPA, the government filed a one-count criminal information against Fokker on June 5, 2014. Once federal charges are filed, the Speedy Trial Act normally requires that trial begin within seventy days. But the Act excuses any period of delay “during which prosecution is deferred by the attorney for the Government pursuant to written agreement with the defendant, with the approval of the court, for the purpose of allowing the defendant to demonstrate his good conduct.” It’s that “approval of the court” language that led to the dispute in this case.

After the charging document and DPA were filed, Judge Leon indicated he might withhold his approval. That would have the effect of torpedoing the DPA, because the Speedy Trial Act clock would keep running and the government would be forced either to take the case to trial or dismiss it long before the eighteen-month term of the DPA had expired. At a series of hearings Judge Leon said he thought the DPA was “too good a deal for the defendant.” He expressed concern about why no individual employees had been charged and why the government had not required Fokker to hire a corporate monitor.

After months of consideration, on February 5, 2015 Judge Leon denied the motion to exclude time under the Speedy Trial Act. He wrote it was not in the interest of justice to approve the DPA when Fokker had engaged in such egregious conduct and was being punished only “anemically.” He concluded the DPA was “grossly disproportionate to the gravity of Fokker Services’ conduct in a post-9/11 world” and “[did] not constitute an appropriate exercise of prosecutorial discretion.”

This marked the first time a district judge had denied a motion to exclude time under the Speedy Trial Act due to the judge’s disagreement with the terms of a DPA. Both the United States and Fokker appealed the judge’s order.

The D.C. Circuit’s Decision

The D.C. Circuit reversed Judge Leon’s ruling in an unanimous opinion written by Judge Sri Srinivasan (who reportedly was on the very short list for President Obama’s most recent Supreme Court nomination). The Court went out of its way to note it was not agreeing or disagreeing with Judge Leon’s views about the merits of the DPA, and that it had no occasion to do so. The point was that a judge has no business making such a judgment.

As the Court of Appeals noted, it has long been settled that criminal charging decisions – including what kind of charges to bring, when to dismiss charges, and whether to bring charges at all — are almost exclusively an Executive Branch prerogative. These decisions involve many considerations including the strength of the evidence, deterrence value of a prosecution, allocation of scarce resources, and law enforcement priorities. The Judiciary is ill equipped to make these judgments, and absent some kind of abuse or a constitutional violation a court will almost never second- guess such decisions.

Judge Leon, however, rejected the Fokker DPA because he thought the government had not been tough enough on Fokker and its employees. The Court of Appeals made it clear that was not the judge’s call. Whether he personally agreed with the terms of the agreement was irrelevant; Judge Leon should not have “assume[d] the role of Attorney General” by questioning the prosecutors’ decisions.

The approval of the court under the Speedy Trial Act, the Court of Appeals held, should simply be to ensure that the proposed DPA is actually for the legitimate purpose of allowing the defendant to “demonstrate his good conduct.” If it is, the judge’s inquiry is at an end. In questioning the terms of the deal and the government’s charging decisions, the Court of Appeals held, Judge Leon “significantly overstepped [his] authority.”

Consequences of the Court’s Decision

The D.C. Circuit’s decision is a welcome reaffirmation of the importance of separation of powers. Although judicial scrutiny of DPAs might have some facial appeal, it would actually raise a host of problems. Not only do judges simply lack the necessary information to make such decisions, but judicial intervention would have a number of other negative consequences.

When criticizing the lack of a corporate monitor in the Fokker DPA, Judge Leon remarked, “One can only imagine how a company with such a long track record of deceit and illegal behavior ever convinced the Department of Justice to agree to that!” But that’s precisely the point – the judge can only imagine. He doesn’t have the information necessary to make an informed judgment about the terms of the deal or why the government might have made that decision.

There may be many explanations for a DPA that looks lenient to the outside world. The alternative to a deal is a trial, where the government must prove its case beyond a reasonable doubt. Prosecutors may have information about possible difficulties in meeting that burden — such as problems with particular witnesses or the admissibility of certain evidence – unavailable to those not involved in the investigation. These considerations will always influence the government when deciding what kind of a deal it should make, whether it’s a plea agreement or a DPA, but likely will be unknown to the judge.

The government also must make judgments about the allocation of prosecution resources, law enforcement priorities, and deterring other wrongdoing while incentivizing cooperation. As the Court of Appeals noted, these are core functions of the Executive Branch, charged with faithfully executing the laws. A judge generally lacks much of the information necessary to make such judgments, and even if the judge had some relevant information, it is not the court’s place to intervene.

Apart from the structural separations of powers concerns, judicial review of the terms of DPAs would also have a number of practical negative consequences. Uncertainty is never a plus for parties trying to negotiate a resolution. Defendants would rightly be more cautious and reluctant to cooperate if they knew that whatever the government is offering them is not really the last word and that a judge might second guess whatever agreement they make.

Another likely, and undesirable, consequence of judicial scrutiny of the terms of DPAs would be a shift to the use of NPAs instead. Non-prosecution agreements do not require any charges to be filed with the court, and so do not require any court approval. Judge Leon himself noted that, “this Court would have no role here if the Government had chosen not to charge Fokker Services with any criminal conduct – even if that decision was the result of a non-prosecution agreement.”

Prosecutors can require as part of a NPA that defendants waive the statute of limitations, so they could still preserve the right to file charges later if the defendant failed to live up to the agreement. And the other terms of the agreement may be largely the same, allowing the prosecution to achieve the same goals as a DPA.

The primary difference is that a DPA generally is filed with the court and made public. A NPA could remain entirely private if the parties so chose. If DPAs were routinely second-guessed by trial judges, the logical response would be for prosecutors and defendants to shift to NPAs in order to avoid any such judicial interference. This could result in more secret deals and in less information being available to the public about any resolutions. The irony, therefore, would be that by purporting to subject the terms of these agreements to greater scrutiny, judicial review likely would instead drive such agreements underground and out of public view entirely.

There are legitimate concerns about the increasing use of DPAs. Companies face tremendous pressure to resolve criminal investigations short of a trial, which gives prosecutors enormous leverage. DPAs risk transforming the criminal justice system into a sort of regulatory, administrative regime run by prosecutors relieved of their burden of proving criminal conduct beyond a reasonable doubt.

But having judges play Monday morning quarterback concerning prosecutors’ charging decisions is not the answer. It’s fortunate the D.C. Circuit agreed.

Click here to join the Sidebars mailing list and receive e-mail notification of future posts.

The Crisis in Corporate Criminal Liability

Last week in my White Collar Crime class we discussed corporate criminal liability. The presence of corporations as potential defendants is one of the distinguishing features of white collar crime. The way corporate criminal liability has been trending for the past decade, however, the topic is starting to feel more appropriate for a course in legal history.

Indicting and convicting big corporations has fallen out of favor, increasingly replaced by a system of negotiated Deferred Prosecution Agreements and Non-Prosecution Agreements. It’s a troubling trend and a sign that something in the criminal justice system is out of whack.


No Soul to be Damned, and No Body to be Kicked 

At common law, corporations could not be charged with a crime.  With “no soul to be damned, and no body to be kicked,” companies were not considered an appropriate target of criminal sanctions. A corporation could not be deterred from criminal conduct through fear of moral condemnation in this life or the next. Nor could a company be deterred through fear of corporal punishment or be locked up to protect the rest of us. Criminal law as applied to an artificial entity just didn’t seem to make sense.

The Supreme Court tossed this common law rule aside more than 100 years ago in New York Central & Hudson River Railroad Co. v. United States (1909). The railroad and its freight manager were convicted of providing illegal rebates to sugar producers to encourage them to ship their product by rail rather than by water.

The company argued it could not be convicted because to do so would harm innocent shareholders without due process. Shareholders had not been charged and had no opportunity to defend themselves, but they were the ones who would suffer if the company were convicted. The railroad also argued that a criminal act could not truly be considered an act of the corporation unless formally authorized by the board of directors, and the freight manager had been acting on his own.

The Supreme Court rejected both arguments. Adopting the theory of respondeat superior (“let the master answer”) from tort law, the Court concluded the same doctrine applies to criminal liability and makes the corporation liable for the criminal acts of its agents. The Court reasoned that criminal sanctions are essential to controlling corporate misbehavior and that it is appropriate to hold the company responsible for the crimes of those acting on its behalf.

Under respondeat superior, in federal cases a corporation is liable for the criminal acts of its agents if those acts were taken at least in part to benefit the corporation and if the agent was acting within the actual or apparent scope of his or her employment.   This is true even if the actions were those of a “rogue employee” acting without authorization.

Throughout most of the twentieth century, corporate criminal liability was an important weapon in the battle against white-collar crime. In the past decade, however, there has been a marked decrease in the Department of Justice’s appetite for prosecuting big corporations. Many people trace this development to the Arthur Andersen debacle.

The Legacy of Arthur Andersen

Arthur Andersen, one of the so-called “big five” accounting firms, was indicted in 2002 for obstruction of justice. Andersen was Enron’s auditor, and a few employees in its Texas office had shredded millions of pages of Enron-related documents in anticipation of an SEC investigation.

Andersen was convicted at trial and appealed all the way to the Supreme Court. In 2005 the Supreme Court reversed the conviction, finding that the jury instructions in the case had been flawed. But it was a hollow victory; during the appeals the company went out of business as clients fled from an accounting firm operating under a criminal cloud. More than 28,000 employees in the U.S., and 85,000 worldwide, lost their jobs.

The Department of Justice was roundly criticized for the Andersen prosecution, which caused tremendous collateral damage to innocent parties based on the actions of a handful of employees. The fallout seemed to make DOJ gun-shy about indicting companies. In the wake of the 2008 financial crisis, some administration officials even expressed concern that certain companies may be “too big to jail” and should not be prosecuted because if they were to collapse the damage to the economy would be too great.

Deferred Prosecution Agreements and Non-Prosecution Agreements

This growing reluctance to indict and prosecute companies has coincided with another phenomenon: the rise of the Deferred Prosecution Agreement. (Non-Prosecution Agreements are basically the same thing; for simplicity here I will just refer to DPAs.)

A DPA is a negotiated deal to resolve a criminal investigation. In a DPA, the prosecution agrees it will not proceed with criminal charges if the corporation complies with the term of the agreement. Typically these terms include paying fines or restitution, undertaking internal corporate reforms, cooperating in ongoing investigations, and hiring (and paying for) a monitor to oversee the company’s compliance for a period of years. If the company fully complies with the agreement, any potential criminal charges are dropped.

DPAs have seen an explosive rise in popularity. In the early 2000s before the Andersen case, there was an average of only 2 or 3 per year. In recent years the average number of cases resolved through DPAs has been in the thirties. The real number is probably higher because not all such agreements are made public (particularly NPAs).

Both sides find much to like in a DPA. The prosecution obtains many of the same remedies (including substantial fines) that it could obtain from an indictment and conviction, and saves the time and expense of a lengthy grand jury investigation and trial. Avoiding criminal prosecution also avoids the collateral damage (and subsequent criticism) that might result from an Andersen-like collapse of a company.

For its part, the corporation avoids the risk of a criminal conviction and potentially crippling penalties. It is able to resolve the investigation and put the matter behind it, providing the certainty that Wall Street craves. Fearing they could be the next Andersen, executives may jump at the chance to reach an agreement and avoid even a slight possibility of criminal sanctions.

But although DPAs are popular and seem like a win-win, I believe their rise is a troubling trend.

The Problems with DPAs

Criminal sanctions are meant to be reserved for the most egregious conduct, in cases in which the government can prove guilt beyond a reasonable doubt to a judge or jury. As more and more criminal cases are resolved through DPAs, these principles erode. Prosecutors no longer must assemble a strong enough case to prove guilt beyond a reasonable doubt to a neutral factfinder; they need only assemble enough evidence to convince the corporation that it should cut a deal to get the matter resolved.

This almost certainly results in more marginal cases being pursued. The discipline of conducting a grand jury investigation and assembling the evidence necessary to prepare for trial has a way of weeding out cases that don’t really merit criminal sanctions or that can’t be proven beyond a reasonable doubt. That discipline is lost if prosecutors can simply pressure a company into a DPA in a borderline case.

A criminal prosecution also contains checks and balances on prosecutorial power. A judge rules on the law, and a judge or jury decides guilt or innocence. If there is a conviction a judge, not the prosecutor, determines the sentence. With a DPA, all of those checks and balances are gone. The prosecutor is also judge and jury, deciding not only that there has been a violation but what the appropriate penalty should be.

This also means the government’s legal theories are seldom tested. In a prosecution, new or aggressive legal interpretations would be evaluated by a judge; with a DPA, the law is whatever the prosecutor says it is. This has been particularly true in areas such as the Foreign Corrupt Practices Act, where so many cases are resolved by DPAs that there is very little decided caselaw. The law comes to be defined through the “common law” development of the prosecution’s interpretations as embodied in the DPAs, with little or no judicial oversight.

All of this places enormous additional power in the hands of prosecutors. It also transforms the criminal justice system into a quasi-regulatory regime, with prosecutors making decisions and demands about appropriate corporate behavior. The DPA may mandate changes in Board membership, internal compliance programs, and other business practices that affect corporate governance. Prosecutors end up shifting their focus from punishing past criminal acts to shaping future corporate behavior. This is not a criminal prosecutor’s area of expertise.

But perhaps the most damaging result of this trend is the erosion of the moral force of the criminal law. Criminal sanctions are a unique deterrent due to the moral condemnation of the community that accompanies them. Individuals do not want to be labeled a criminal in the eyes of the community; it carries a stigma not associated with any other penalty. Most defendants, particularly white-collar defendants (who typically are not career criminals), will work hard to avoid that stigma.

Although corporations have no “soul to be damned,” this same moral stigma still applies to corporate criminal sanctions. One simple reason is the bottom line: companies do not want their customers to see them labeled as a criminal organization. It’s bad for business to be thought of as a crook. What’s more, companies are run by individuals for whom this moral condemnation also holds sway – the CEO of a company does not want her friends and neighbors to see in the news that the company she leads is a criminal.

One need only see how mightily a corporation will strive to avoid a criminal conviction to know that this deterrent effect of criminal sanctions is real.

If large companies start to believe that criminal sanctions are basically off the table, the criminal law will lose this unique deterrent capability. Corporations will know that, no matter how bad the misconduct, they can likely buy their way out of a criminal case by entering into a DPA. This may encourage corporate officers to skate closer to the criminal line – or to jump over it.

 The Future of Corporate Criminal Prosecutions

It’s hard to see a good solution to this trend. Corporations are under tremendous pressure to cut a deal and avoid prosecution. Resolving an investigation short of criminal sanctions usually will be rewarded by shareholders and the market. And as long as prosecutors know they have this tremendous leverage, their incentive will be to force as many companies as possible into a DPA rather than pursuing a full-blown criminal case.

My friend and former colleague Mike Volkov, who writes the Corruption, Crime and Compliance blog, wrote a great post a few weeks ago called “Corporations Need to Say the Words – ‘Let’s Go to Trial!’” He argues that, in appropriate cases, more companies need to stand up to DOJ and not simply roll over during a criminal investigation. If more corporations resisted the pressure to cut a deal, they might find many of these investigations actually going away or being referred to other agencies for civil enforcement. Call the prosecutors’ bluff, and they may decide that a potential criminal case is not really strong enough to merit the investment of the next two years of their lives.

Even when there is an indictment, it most likely will not mean the end of the company. Arthur Andersen was in a unique situation given the nature of its business and the general post-Enron hysteria that was raging at the time. Historically, though, a criminal indictment usually does not result in a corporate “death penalty.” General Electric, Exxon, Tyson Foods, Genentech — all have been criminally indicted at one time or another and are still going strong.

Companies need to realize that even if they are indicted it will be bad but need not be fatal. If necessary the criminal case may be resolved through a plea agreement with terms probably no more severe than those in a DPA, and with the added benefit that the terms will be reviewed by a judge. Or the case may be taken to trial and defended, with the chance that the company will prevail.

Prosecutors, too, need to recognize that they do not need to be gun-shy about indicting companies based on the Arthur Andersen experience. In appropriate cases, when criminal sanctions against a company truly are called for, criminal charges should be pursued to fulfill the special moral mission of the criminal law.

We already have enough regulatory agencies policing corporate behavior. Prosecutors should get back to indicting and trying criminal cases – and more corporations should develop the backbone to say, “see you in court.”

Click here to subscribe to the Sidebars mailing list and receive email notifications of future posts.