In 2006, Reps. Brian Baird (D-Wash.) and Louise Slaughter (D-N.Y.) introduced a bill into Congress. Titled "Stop Trading on Congressional Knowledge" -- a "STOCK" Act -- the bill aimed to prohibit lawmakers from trading stocks on insider information. But why did they do it? Logically, Baird and Slaughter must have thought that there was insider trading going on in Congress. But what made them think so?

The answer to this question is contained in a series of three academic research papers. I know -- boring. But stick around for a moment. I'll make it worth your while.

2004: Pulling back the curtain on congressional insider trading
In 2004, Professor Alan Ziobrowski of Georgia State University published a paper titled "Abnormal Returns From the Common Stock Investments of the U.S. Senate." The title might just as easily have read: "Wake Up! Congress Is Robbing You."

The culmination of years spent collating, transcribing, and digitizing mountains of congressional annual financial disclosure forms, this report detailed how U.S. senators somehow managed to outperform the stock market by 12 percentage points in their trading. As one commentator quickly pointed out, such results "strongly imply that some members of Congress are using nonpublic information to make trading decisions." He was right... but the story is a bit more complex than that.

Elsewhere in our STOCK Act coverage, my colleague Matt Koppenheffer will tell you about all sorts of problems he's discovered in the congressional financial disclosure process. But for now, let's keep this simple: The disclosure process stinks. It's disorganized and vague. As Ziobrowski lamented: "The available data do not permit us to measure the magnitude of profits [actually] earned by individual Senators."

That said, as Professor Ziobrowski told me, "We did the best with what we had." Working from 6,052 transactions reported during the five-year period examined, he began with the dates on which transactions were reported, and tracked the stocks' performance over the next 12 months. Over the years 1993 to 1998, Ziobrowski concluded:

  • A stock purchased by a U.S. senator outperformed the S&P 500 by 0.85% per month over the ensuing 12 months.
  • A stock sold (or sold short) by a U.S. senator underperformed the S&P 500 by 0.12% per month over the ensuing 12 months.

Combined, these returns add up to approximately 12 percentage points' worth of annual outperformance. Mind you -- this is not just "12% annual profits" the senators would have earned. This is the profits everyone else was earning plus 12 points of additional profit on top of that. Ziobrowski calls such performance "both economically large and statistically significant." Indeed, these kinds of numbers represent "some of the highest excess returns ever recorded over a long period of time, significantly outperforming even hedge fund managers," wrote Ziobrowski. Even Warren Buffett only outperformed the market by 10 percentage points annually over that same time frame, or 11 points annually over his entire tenure at Berkshire Hathaway. In short, Ziobrowski's study strongly suggests senators "trade with a substantial informational advantage."

2011: Confirmation of criminality?
As you can imagine, Ziobrowski's 2004 report caused quite a stir on Capitol Hill. It did, however, contain one glaring omission: the House of Representatives. Were senators alone in trading on insider information, or were their colleagues in the House making out like bandits, too? Ziobrowski set out to find the answer, which he laid out in the 2011 study: "Abnormal Returns From the Common Stock Investments of Members of the U.S. House of Representatives."

Ziobrowski found that U.S. representatives were profiting handsomely. Reps, it seems, outperformed the S&P 500 by a significant margin.

In the 2011 study, Ziobrowski again constructed a "synthetic" portfolio. He again assumed that a stock, once bought, was held for one year before being sold and the profit taken. But this second study differed from the 2004 report in a few important respects. For one thing, it was more comprehensive, covering transactions reported from all the way back in 1985 up through 2001 -- 10,075 reported transactions in all (of which, 8,294 were ultimately deemed reliable enough for analysis). For another, Ziobrowski appeared to have second thoughts about the usefulness of sales data -- and while reporting it, did not weigh it in calculating excess profits earned by members of the House.

In the end, though, the results were analogous to those found in the Senate study: The U.S. representatives managed to outperform market returns by 6 percentage points.

2011: Meanwhile, in Massachusetts...
So an open-and-shut case, right? Lock 'em up and throw away the key? Not so fast -- it's a fundamental law of academia that every study, once published by one professor, must spawn an equal and opposite study by another professor.

No sooner had Ziobrowski's 2011 report rolled off the presses than up in Massachusetts, a pair of professors from M.I.T. and the London School of Economics published a study of their own that cast doubt upon Ziobrowski's assertions. Challenging Ziobrowski's findings, professors Andrew Eggers and Jens Hainmueller argued that their review of four years' worth of disclosed trading activity in the Senate and House showed not significant outperformance of the market -- but rather "mediocre performance" instead.

Criticizing the findings of Ziobrowski's synthetic portfolios as "fragile and inconclusive," Eggers and Hainmueller argued that while it's conceivable that the lawmakers examined in the previous studies outperformed the market, this conclusion is far from certain: "The margin of error is big enough that it's impossible to say these results did not come about by chance. You cannot reject the null hypothesis."

Eggers and Hainmueller therefore took a different approach, and set out to calculate the actual profits that lawmakers earned from their trading, over a period running from 2004 to 2008. Their ability to do this was facilitated by the recent creation of the website OpenSecrets.org, which now posts congressional financial disclosures on the Web, making the data easier to assemble and process.

Eggers and Hainmueller were accordingly able to reconstruct detailed stock portfolios for each sitting lawmaker during the period covered, and to track these portfolios' performance on a daily basis. What they found was that if you examine lawmakers' actual performance, from stock purchase to stock sale, they actually underperformed the stock market by about 2 to 3 percentage points.

A tale of three studies
So who's got the right side of this debate? Is Ziobrowski right that Congress is clearly trading on insider information? Or are Eggers and Hainmueller correct to say that the data simply doesn't support this contention -- and may actually refute it?

It's hard to say, not least because the three studies don't really cover the same data at all, but rather review different groups of lawmakers trading stocks during different periods of time. Perhaps crucially, Ziobrowski points out that the Eggers/Hainmueller study reviewed only stock transactions that took place after his own report clued Congress in to the fact that they were under the microscope. As he told me, lawmakers "would have had to have been deaf and blind to not notice what was going on" -- and may have changed their behavior accordingly.

But when you get right down to it, it doesn't even matter whether members of Congress who possess insider information "display unusually good trading acumen." Regardless of how well these lawmakers trade, what matters is that they appear to trade on insider information. It's the appearance of impropriety that's truly toxic to the integrity of the stock market. As Eggers and Hainmueller wrote in The Boston Globe:

It is hard to see any reason why members of Congress should oppose legislation limiting their ability to play the stock market while in office. Political insiders face clear temptations to profit from their privileged information; regardless of whether they successfully indulge in those temptations, citizens will suspect them of doing so.

Indeed, while these professors may differ on their interpretations of the numbers, Ziobrowski, Eggers, and Hainmueller are all in agreement on one fundamental point: If there is insider trading going on in Congress, it must stop. If there's uncertainty as to whether insider trading laws apply to Congress, passing the STOCK Act would make these restrictions clear.

We agree.