Frank Peake

Frank Peake and the challenge of sending antitrust defendants to prison

In early April 2013, some three months after a federal jury found former shipping executive Frank Peake guilty of price fixing, the foreman of that jury told the court to go easy on Peake.

In a letter to Judge Daniel Dominguez, juror Luis Torres Cordero stressed that everyone on the jury continued to believe that the former president of Sea Star Lines had broken the law. But despite Peake’s executive status and his alleged oversight of the cartel, Cordero told Judge Dominguez he and other jurors harboured significant doubts about how much Peake was actually involved in the plot, and suggested that the court give Peake far less prison time than the Department of Justice had recommended.

What the government asked for was the longest sentence for an antitrust crime in US history. If the court agreed, Peake would serve more than seven years in custody, twice as long as any previous antitrust defendant. Cordero reminded the judge that the jury had deadlocked twice as to Peake’s guilt during the two-plus days of deliberations, as jurors raised serious doubts as to how much Peake actually knew about the years-long price-fixing plot. “There was a concern among the jury about the defendant receiving an equal or even more severe punishment than those who decided not to go through the trial process, but seemed to have a bigger role on the conspiracy” according to the evidence, Cordero wrote. Whatever the judge decided, the foreman wrote, Peake should not be given a longer sentence than his co-conspirators who became government witnesses in the case – specifically Peake’s subordinate at Sea Star, Peter Baci.

Peake is now inmate number 38588-069, serving out the longest antitrust sentence in history inside FCI Cumberland, a medium-security prison with an adjacent minimum-security camp in suburban Maryland. He’s scheduled to be released in May 2021, although he’ll likely be freed early for good behaviour. The five-year sentence Judge Dominguez gave Peake stands as one of the Obama era’s most notable criminal antitrust enforcement victories, along with its successful prosecution of AU Optronics for its role in the LCD cartel.

But in both cases, the government did not get everything it wanted; indeed, in prosecuting some executives for their roles in the crime, far less. Both cases and other Obama-era prosecutions suggest that, for a variety of reasons, judges are unwilling to adhere to federal sentencing guidelines that would saddle the guilty with a decade in prison. What’s more, despite the federal government’s 93% conviction rate in criminal cases overall, its success rate in criminal antitrust trials is far lower. The AUO trial and the record-breaking prison sentences handed down to Peake and others sit alongside more than two dozen acquittals, hung juries and mistrials during the Obama administration that saw those who forced the government to prove its accusations walk away free. According to the DOJ’s statistics, 47 criminal defendants were found guilty after trial, while juries or judges acquitted 25 and otherwise declined to convict another six.

The government’s record in trials won and lost sheds light on a paradox in criminal antitrust enforcement on open display during the Obama era. The government has been extraordinarily efficient at securing guilty pleas from both corporate and individual participants in price-fixing and bid-rigging schemes. But when put to their evidence in court, prosecutors have at times struggled to secure a conviction or to convince judges to adhere to the country’s sentencing guidelines when deciding the amount of time former executives should spend in prison. If a guilty plea will result in at least some jail time – but taking the government to trial gives executives a better than one-in-three chance of going free – what choice will indicted executives make in the future? And if the punishment for a guilty verdict is only marginally higher than that for a guilty plea, why not take the risk?

Yet so far that doesn’t appear to be the thinking of most executives accused of antitrust crimes who take the government to trial. Taking a plea deal with the government early remains the norm for top executives. Peake was an exception; he was in charge at Sea Star and had direct oversight of the unit of the company that had established and maintained the price-fixing plot. Most antitrust defendants who take the government to court rest far lower on the corporate ladder. They are often the “last man standing”, as it is called – the final defendant in an antitrust investigation who, at that point, has little ability to help the government and earn a discount on their sentence. When no sweetheart deal is available, taking the government to trial and making it prove its case is the best and only option.

The type of executives who do put the government’s evidence to the test in court partially explains prosecutors’ less-than-stellar record in securing guilty verdicts from juries, and judges’ reluctance to sentence them to the lengthy prison terms recommended by the government and the federal sentencing guidelines.

Mark Rosman, a former Antitrust Division prosecutor and current partner at Wilson Sonsini Goodrich & Rosati, says that in most cases the most obvious culprits involved in a price-fixing scheme confess to their participation and take whatever deal their lawyers can get for them. Executives are often left as the “last man standing” when the evidence of their guilt is less than compelling, or when their role in the conspiracy was so minor, they are convinced that a jury would never convict them. “It’s harder to convict those lower-level people,” Rosman says. He notes that in another kind of criminal case – a drug crime, for example – the government could offer to charge the defendant with a lesser crime in exchange for his or her cooperation. But in antitrust law, there’s only one crime and it comes with a 10-year maximum sentence. Bit players in a conspiracy would often rather fight.

* * *

The AUO case was significant for multiple reasons: securing the first-ever guilty verdicts against foreign executives is chief among them. The two AUO executives the jury found guilty – HB Chen and Hui Hsiung –were two of the primary organisers of the LCD cartel and, the government alleged, deeply culpable for the company’s actions. But those two convictions came alongside three other executives the jury declined to convict – two by acquittal and one due to a persistently hung jury (the DOJ did eventually secure a guilty verdict in that case). LJ Chen and Hubert Lee, the two defendants acquitted of the crime, had far less to do with the actual crime of fixing prices than their bosses did, the government admitted. Their decision to take the government to trial paid off, and future antitrust defendants will certainly take heed. Meanwhile, Rosman points out that the jury deadlocked as to the guilt of Steven Leung, the alleged “note-taker” of the cartel’s now-infamous Crystal Meetings, despite prosecutors equating Leung’s role to that of a getaway driver during a bank robbery. That Leung was ultimately convicted of the crime matters, but the initial question of his guilt suggests jurors remain reluctant to convict lesser players in cartel schemes – even the metaphorical getaway driver.

Convincing a jury to convict any executive of an antitrust crime is inherently difficult, Rosman says, even when compared to other white-collar crimes such as fraud. In fraud cases, there’s a clear victim and a clear wrong; someone was essentially robbed. Cartel crimes are different. In the marine hose cartel case, for example, the primary victims were the sovereign wealth funds of dictator-led countries and the world’s largest oil companies. Such victims aren’t exactly compelling when trying to convince a jury to send a man to prison, which might explain the government’s mixed conviction record in that prosecution. Jurors, Rosman says, “might have some vague intuition that it’s wrong, but it’s a tougher case, and you actually have to educate a jury that this is a crime to start with.” In the same shipping cartel case that ensnared Peake, another defendant, Thomas Farmer, was acquitted entirely. Former DOJ official Renata Hesse said at the time that the government felt good about the prosecution, but the jury simply didn’t think Farmer deserved jail time. “He was an elderly man and people just didn’t want to send him to jail,” Hesse said then.

Even when juries do convict, judges have largely rejected the government’s requests to sentence the guilty to maximum or near-maximum prison sentences for their crimes. In the AUO case, the jury convicted HB Chen and Hui Hsiung based on prosecutors’ charges that they ran the Crystal Meetings where the conspiracy was hatched, and ordered their subordinates to put the fixed LCD screen prices in place. The economic effect of the cartel was thought to be enormous: computer and phone makers spent billions on LCD screens for their products. Coupled with the duration of the conspiracy and other factors, the guidelines recommended prison sentences above even the statutory 10-year maximum. Prosecutors asked for all the time the guidelines recommended – it was, DOJ attorneys insisted, the most egregious cartel the US had ever prosecuted. When it came time for sentencing, the court was decisive: three years in custody for each defendant.

The AUO executives could have faced much worse. They were guilty well beyond a reasonable doubt, Judge Susan Illston said during sentencing. They led the conspiracy and organised the meetings. They had every opportunity to come into the DOJ, plead guilty and cooperate with the prosecution. Instead, they went to trial and lost, for which they should be punished.

But Judge Illston struggled to impose the government’s requested maximum sentence. Instead, she appeared to find some sympathy for the defendants and their situation. “There was enormously bad judgment exercised by this corporation, these defendants, and the other corporations engaged in this conduct,” she said. But Chen, a Taiwanese national, was the 60-year-old president of a major Taiwanese company, and Hsiung, 58, was the vice president. They had no criminal record; they were smart, rich, good family men and generous in the community. Plus, they were working in a fledgling industry in a different culture – an important consideration as they were the first foreign executives to go to prison for cartel crimes – and trying to help their company and industry. Those features count, Judge Illston said. “They don’t make it not a crime. They don’t excuse it, but they go a long way to explain it.”

Just as a jury might, Judge Illston compared the AUO executives’ actions with those of a fraud defendant who took other people’s money so he could have it for himself. That’s not what Chen and Hsiung did. They were fighting for their industry, and although they likely knew what they were doing was illegal, the judge seemed to find it understandable, if not commendable.

One theory to explain the persistently short antitrust crime sentences is that judges, whether consciously or not, view the executives before them as peers. As Judge Illston and other jurists have pointed out, cartel defendants work important jobs, make good money, care for their families and communities, and so on. Compared to sovereign wealth funds and oil companies, the individual perpetrator can be far more sympathetic than the victim; another LCD defendant, Tony Cheng from Chunghwa, had his mother send a note to the judge asking for the government to leave him alone.

Researchers have found this to be generally true of white-collar defendants. A 1980 study of judges’ views on defendants found there was a “fundamental tension many judges feel between the aims of general deterrence on the one hand, and the particular attributes of white-collar offenders on the other.”1 That empathy continues; a recent study showed that judges tend to see white-­collar defendants differently from other criminal suspects, specifically because they see much of themselves in white-collar defendants.2

There have been obvious exceptions to traditionally short fraud sentences. Bernie Madoff, the Ponzi scheme orchestrator, is serving 150 years in prison for his fraud – effectively a death sentence for a septuagenarian. Of course, Madoff’s sentence reflects the scale of the harm he inflicted, and the difference in the limits of the sentencing guidelines. In fraud cases, the average guideline minimum for losses of more than $100 million is 324 months.3 In antitrust cases, economic losses of $100 million might get the offending executive two years in prison. The difference is in the metrics – who the victims were and how directly they were harmed.4 The direct victims of antitrust crimes – often themselves large multi­national corporations – simply aren’t as compelling as those of fraud crimes, and the individual defendants remain generally upstanding citizens in the eyes of the courts.

Are these prosecutorial barriers unique to the Obama administration? Certainly not: antitrust conviction rates have sat below the DOJ’s overall criminal conviction rate since the government began prosecuting international price-fixing cases in the 1990s. And while prison sentences for both those convicted by juries and those who plead guilty have grown in number as the government has more actively pursued cartels, they have grown in length only incrementally. Critics of the government’s push for longer sentences say judges have generally issued short sentences for good reason. Two years in federal prison is more than enough to dissuade otherwise law-abiding executives from fixing prices. A felony conviction alone will likely end a career. Five, six or seven years in prison is piling it on, critics say. The message has already been delivered.

Yet the quest for long sentences persists. During Peake’s sentencing, Antitrust Division prosecutor Brent Snyder told the court that the DOJ doesn’t ask for record-breaking prison terms because it wants to secure such terms; it requests long sentences because the federal sentencing guidelines demand it. Indeed, the department press release announcing Peake’s five-year sentence never boasts about its record-breaking nature; at the most, the DOJ said the sentence should signal to others that fixing prices is a bad idea.

But Rosman and others suggest the Peake sentence will matter when defendants found guilty of antitrust crimes stand for sentencing in the future. Prosecutors can now point to Peake’s five-year prison term as a high-water mark. In future cases where an executive defendant’s behaviour is deemed more severe – for a conspiracy that lasted longer or affected a larger chunk of the US economy, for example – judges could be willing to issue prison terms even longer than five years. If that happens, the Peake sentence may indeed be seen as part of the lasting legacy of Obama-era antitrust enforcement.


  1. Kenneth Mann et al, “Sentencing the White-Collar Offender”, 17 Am. Crim. L. Rev. 479, 481 (1980).
  2. Daniel Richman, “Federal White-Collar Sentencing in the United States: A Work in Progress”, 76 L. & Contemp. Probs. 53, 55 (2013).
  3. Mark W Bennett et al, “Judging Federal White-Collar Fraud Sentencing: An Empirical Study Revealing the Need for Further Reform”, Iowa L. Rev., 961 (2017).
  4. This is not to say that all white-collar criminals were punished harshly for their frauds. As Bennett et al point out, under the guidelines the average prison sentence for fraud in 2014 was 24 months. But the average loss from those frauds was far less than the amounts at issue in antitrust crimes – tens of thousands of dollars lost through fraud compared to, at a minimum, tens of millions of dollars of commerce affected for most international cartels. Again, the sentencing guidelines and nature of the crimes are very different but the study, in which 240 judges participated, found the same lenient treatment of white-collar defendants regardless of the crime.

Unlock unlimited access to all Global Competition Review content