The recently completed trial of James Fleishman, the salesperson for Primary Global Research (PGR), was a stretch case for the prosecution. Unlike previous Galleon-related trials it was difficult to link the defendant directly to material non-public information. According to sources close to the trial, the case seemed to be going against the prosecution, yet the jury delivered in the end. Informed observers believe this case will embolden investigators to prosecute marginal cases that heretofore they might have avoided. Most troubling, however, is whether mosaic theory is still valid based on the definitions of material non-public information used in the case.
A Tough Case
From the outset, it looked like the Fleishman trial would be a difficult one for the prosecution. From the initial complaint, there did not seem to be any clear-cut incriminating evidence like earnings numbers provided by Fleishman to clients. Unlike other PGR employees charged in the investigations, he was not directly involved in working with the experts, making it harder to show that he was privy to illegal information.
Fleishman was charged on two counts: conspiracy to commit securities fraud (passing material non-public information) and conspiracy to commit wire fraud (passing confidential information). The government’s case was built on wire taps collected by Karl Motey, a cooperating witness who signed up as a PGR client at the FBI’s request. By posing as a prospective client, Motey tried to incriminate Fleishman.
On cross-examination, Motey admitted that the information he received from PGR consultants like Daniel DeVore, a Dell purchasing manager who was a PGR expert, would not be sufficient by itself to base a buy/sell decision on Dell.
Emphasis on Confidential Information
It seemed that the prosecution was focusing its efforts on proving that Fleishman was aware that confidential information was being passed (i.e., wire fraud). Prosecutors argued, for example, that Fleishman knew Dell was on clients’ ‘do not call’ lists, and still facilitated meetings with Dell employees. In summarizing the case during closing arguments, the prosecution described it revolving around confidential information: “Ladies and gentlemen, this case at the end is going to be about whether James Fleishman agreed with others to pass confidential information.”
The prosecution ultimately put PGR on trial, to characterize the expert network as a criminal enterprise. The prosecution called the rogues gallery of PGR employees and experts as witnesses during the trial. Besides Motey, the prosecution took testimony from Daniel DeVore, Tony Langoria, another PGR expert who was a former supply chain manager for Advanced Micro Devices, and Bob Nguyen, a former PGR employee who was the semiconductor sector specialist. In a potential sign of desperation, prosecutors granted immunity to a new witness, Suk-Joo Hwang, a former Samsung employee who testified about passing information about competitor Apple’s iPad.
Fleishman’s defense was that, as a salesperson, he was not in a position to determine whether information was material or not. The information linked to Fleishman was not the clear-cut earnings or revenue numbers in most of the other PGR related cases. Is channel check information on the pricing of hard drives material information? Defense argued that there was no evidence that Fleishman was proactively peddling inside information. Motey repeatedly tried to get illegal information from Fleishman, but the evidence he obtained did not include clear-cut examples of Fleishman’s complicity.
Nguyen, the PGR semiconductor specialist, generated channel check emails containing the title “Handle With Care.” Fleishman passed these emails on to clients, but his defense argued that the “Handle With Care” title reflected business concerns not illegal content. Nguyen obtained the information from experts who had other client relationships and did not want the emails circulated to retainer clients.
The trial took nearly three weeks, most of which was the prosecution presenting material about Fleishman and PGR. The expectation was that it would take the jury a long time to go through the materials. As one of the prosecutors put it, “There’s been a lot of paper in this case, a lot of recordings in this case. We know that you will look at those diligently.”
The jury in Fleishman’s trial took 6 hours to return guilty verdicts on both charges. Given the complexity of the case, the short jury deliberation was surprising. Also interesting was the fact that the jury returned a guilty verdict on securities fraud, even though the case focused mainly on confidential information.
The judge trying the case, Jed Rakoff, told the jury that ANY non-public information is off limits: “As I told you earlier in the trial, every employee of a company has a legal duty not to disclose to anyone outside the company ‘non-public information,’ such as financial or other confidential information about the company that the company has not yet made public or authorized to be made public and that has not otherwise been disclosed to the public.”
This interpretation challenges mosaic theory which holds that non-public information is ok provided it is not material (i.e. would not move the stock.) As the Big Lots situation made clear, researchers must be mindful of company policies on confidentiality, and careful not to induce employees to breach any fiduciary obligations to their employer not to disclose confidential information. Judge Rakoff’s interpretation is much broader, and holds that all employees are under a fiduciary obligation for any non-public information, material or not.
What It Means
The Fleishman case has important implications for the investment research community. First, we can expect the prosecution to be energized by the verdict. The pre-trial materials from the Fleishman case showed that the government has gathered information on at least 50 hedge funds and affiliated people, 300 hours of wiretaps containing conversations with some 250 consultants, and turned over 11,000 pages of instant messages between hedge funds and Primary Global employees, and 6,500 pages of email. One knowledgeable source expects that the government has enough material for another three to four years of litigation.
Second, we can expect a wider range of cases to be prosecuted. When considering whether to bring cases based on evidence they’ve collected, prosecutors are more likely to go for it. A case built around confidential information can also yield a securities fraud verdict.
Which brings us to the most troubling aspect: does mosaic theory still hold? The definition of material non-public information as defined in this case has broadened to include any non-public information. This seems to invalidate mosaic theory, which relies on the test that the non-public components of the mosaic should not be market moving. If all non-public information is off limits to researchers, good investment research has become much harder.