Are you in compliance?

Don't miss out! Sign up today for our weekly newsletters and stay abreast of important GRC-related information and news.

Behind the Myth of Insider Trading by SEC Employees

Bruce Carton | March 11, 2014

Last month, a new academic study led by an Emory University accounting professor on trading habits by employees of the Securities and Exchange Commission got lots of coverage in the financial press, with blaring, dramatic headlines such as:

  • “The Incredible Stock-Picking Ability of SEC Employees;”
  • “Study Finds SEC Staff Sold Shares Before Cases Made Public;”
  • “The SEC Should Really Start a Hedge Fund;” and
  • “Insider Trading Allegations Against U.S. Securities and Exchange Commission Employees”

A look at the study itself, however—which is still “preliminary and incomplete” and has yet to go through the academic review process—shows nothing that would justify the knee-jerk criticism that these headlines suggest.

The study looks at stock trading by SEC employees during the period between late 2009 and the end of 2011, and concludes that SEC employees traded profitably during this period, with trading profits of about 4 percent per year, which is “similar to [the profits] earned by corporate insiders.” In particular, the study suggests, SEC employees may have earned positive “abnormal returns” of 8.5 percent on their investments in U.S. common stocks and that these abnormal returns stemmed from the sale of these stocks.

The authors hypothesize that “not all the abnormal returns earned by SEC employees are attributable to skill or luck.” They assert that most of these abnormal returns are from the sale of stocks, and that this can be attributed to the fact that SEC employees are likely to know non-public information about sanctions against companies before the market as a whole. 

The authors also point to six specific cases in which “SEC employees appear to front-run the announcement that a firm is subject to costly SEC penalties (associated with the enforcement action).” The authors conclude that “at least some of these SEC employee trading profits are information based, as they tend to divest in the run-up to SEC enforcement actions ...” From that bold conclusion came the breathless media headlines mentioned before.

Do the assertions, assumptions, and conclusions in this study even make sense? I say no—let me count the ways:

1.       The “abnormal returns” earned by SEC employees by trading in U.S. common stock that underlie the study may not be so abnormal after all.

Notably, the authors do not claim that SEC employees actually made abnormal returns on their trades. Indeed, the authors acknowledge that they did not have individual information and had no way to know if SEC employees made or lost money on specific transactions. Rather, they claim that “a hedge portfolio that goes long on SEC employees' buys and short on SEC employees' sells earns positive and economically significant abnormal returns … of about 8.5 percent in U.S. common stocks.”

Notably, the authors do not claim that SEC employees actually made abnormal returns on their trades. Indeed, the authors acknowledge that they did not have individual information and had no way to know if SEC employees made or lost money on specific transactions.

Results from this hypothetical short-selling hedge fund, however, present an apples-to-oranges comparison because SEC staff are not permitted to conduct short sales of stocks and the results simply will not be the same. If an SEC employee buys a stock at $50 in July, for example, and sells it 6 months later at $40, he or she will take a 20 percent loss. If the study's hypothetical hedge fund shorts this stock at $40, however, and the stock further declines after the SEC employee's sale to $30, the hedge fund just made a 25 percent profit. The difference here is a loss in one case and a so-called “abnormal return” in the other.

2.       The assumption that the SEC employee sales analyzed in the study occurred because employees with inside information were front-running the “bad news” of upcoming enforcement actions.

There is quite a bit to unravel within this core assumption of the study, including the following:

  • First, as a threshold observation, SEC employees don't even seem to be particularly interested in buying or selling stock. As pointed out by Broc Romanek of, the study found that there were about 4,800 trades (counting both purchases and sales) of U.S. common stock made by the SEC's approximately 4,000 employees in the period at issue— an average of a whopping one whole trade per employee over a period of over two years.
  • Second, the sales in question took place beginning in the period between late 2009 and the end of 2011. In 2010, the SEC implemented a new policy prohibiting its employees from owning bank stocks. As a result, all SEC employees (and every new employee who joined the SEC thereafter) were required to sell all shares of any bank stocks. Many of the sales in the study period, then, were undoubtedly in response to this policy requiring SEC employees to divest themselves of bank stocks.
  • Third, of the SEC's 4,000 employees in 2010, less than a third of them were in the Enforcement Division where they might have access to information about an ongoing investigation. Of these roughly 1,200 Enforcement employees, only a subset would have even owned individual stocks in their portfolios—let's say 700 employees for argument's sake. With each of these 700 stock-owning Enforcement Unit employees handling a small number of enforcement investigations at any given time, how many of them would ever actually work (even momentarily, before being required to recuse themselves) on a matter involving a public company in which they owned stock? And how many of them would realistically stumble upon material, non-public information in the course of their employment about an ongoing SEC investigation into a company in which they owned stock?  Very few. Considering how closely guarded information is at the SEC, and the fact that the SEC brought a total of just 56 enforcement actions against publicly traded firms in 2010 and 2011 (according to the study), the chance of an employee learning in advance about an SEC enforcement action against a public company in which he or she owned stock is highly remote, at best.
  • Fourth, assume again for argument's sake that a small handful of SEC employees existed who learned in advance about an investigation in which an action was later filed against a public company in which they owned stock. How many of these employees would really be willing to risk their jobs and reputations to sell that stock in advance of the SEC's announcement of the case in violation of SEC's ethics rules prohibiting such sales without clearance from the agency? One? Two?
  • As Romanek states, SEC employees are far more attuned than most traders to just how easy it is for the SEC's Market Surveillance Unit to catch people who engage in insider trading. “There are very few [SEC employees] who would be dumb enough to engage in one of those abnormal trades when they are so aware how closely other SEC staffers are looking at them,” Romanek observes.
  • Finally, even if one or two SEC employees were dumb enough or brazen enough to seriously consider engaging in such unethical conduct and selling stock in advance of the SEC's announcement of a case against a company, would there be any monetary incentive whatsoever for them to do so? Raphael Kozolchyk of the SEC's Inspector General's office (an office that has hardly been an ally of the SEC in recent years) points out that the stock prices of the six public companies that the study found SEC employees sold shares in prior to the announcement of an enforcement action almost all went up, not down, after the SEC's announcement. Indeed, this is precisely what most close observers of SEC enforcement actions would expect to happen, but the opposite of what is assumed in the study. Because the existence of an SEC investigation is typically disclosed by the company in question months, if not years, before the SEC announces the filing of a case, the SEC's announcement of a settled lawsuit bringing closure to the matter is usually positive, not negative, news for such stocks.

There is an old saying that “perception is reality.” In this case, however, the perception being pushed on the public (“Insider trading by the SEC!”) appears to be far from reality.  For several years now the SEC has had a stringent trading policy (no bank stocks, no regulated entities, no IPOs, no shorting securities, no buying on margin, a 6-month holding requirement, all transactions reviewed and approved in advance by SEC Ethics Office, and so on) that is much tougher than policies at other agencies that handle market-moving information such as the Federal Trade Commission, Food and Drug Administration, Department of Justice, or the Pentagon. Based on this recent study, at least, there is no reason to believe this policy is not working as intended to prevent insider trading.