Not As Random As It Looks?

Some executives seem to be making big gains in automatic trading plans

John B. Blystone appeared bullish on the prospects of SPX Corp. (SPW ) back in late 2003 and early 2004. In his role as chief executive of the Charlotte (N.C.) industrial goods maker, he prepped investors to expect strong earnings and a sharp rise in free cash flow. And by emphasizing the sustainability of those gains, he helped send the stock rocketing from 48 to nearly 62, before it settled back to 53 by late February.

Blystone also told investors that for "personal financial and estate planning purposes" he had set up a prearranged trading plan in January to diversify his holdings. By Feb. 25, according to a shareholder lawsuit filed later that year, Blystone had sold 800,000 shares valued at $45.3 million.

Then came the bombshell: On Feb. 26, SPX announced its results. Furious investors fled the stock when they realized that one-time gains accounted for much of the jump in earnings and cash flow. SPX shares dropped 21%, to 42, in a day. Before long shareholders filed suit alleging that Blystone knew things weren't as rosy as projected when he set up his automatic trading plan.

In theory, that isn't supposed to happen. Blystone did his trading through what's known as a 10b5-1 plan. The Securities & Exchange Commission created such prearranged trading plans in 2000 to clarify the rules under which executives can legally sell shares they've accumulated. As long as they initiate the plan at a time when they don't know of any significant nonpublic information, lay out in advance the dates or prices at which sales will be made, and don't control the trades, they are legally protected from insider trading charges.

But questions are growing as to whether executives such as Blystone are finding ways around those restrictions to boost their returns. According to recent research by Alan D. Jagolinzer, an assistant professor at the Stanford University Graduate School of Business, some executives appear to be making better-than-expected gains while trading within 10b5-1 plans. Not only do plan participants tend to beat the market, they also appear to do significantly better, on average, than other insiders who do not set up automatic trading plans. One reason is that they sell shares in advance of negative news twice as frequently as they sell ahead of good news.

All of which leads to an obvious question: Are executives using inside dope when they set up such plans? "What this suggests is that when executives know negative news is coming, they contract out a plan and direct it to sell," says H. Nejat Seyhun, a finance professor at the University of Michigan's Ross School of Business. "They are making an end run around the insider trading laws."

Until now, little study has been done on trading within 10b5-1 plans, in part because comprehensive data are hard to come by. Executives simply set up a trading schedule with their lawyer and their broker; SEC rules don't even require them to publicly disclose that they've created one. But the plans have been more widely adopted and disclosed in the last couple of years, thanks to the legal protections they offer. For his study, which is under peer review, Jagolinzer compiled data from available disclosures on nearly 100,000 trades -- the vast majority of them sales -- made by 2,995 insiders at 1,016 companies between 2003 and 2005. In a preliminary analysis, he found that trades inside 10b5-1 plans beat the market by 5.6% over six months.

That appears to confirm results Jagolinzer came up with last year based on a sample of 191 companies. Back then, he found that participants in 10b5-1 plans outperformed the market by 4.8%, while insiders at the same companies who did not enter prearranged plans lagged the market by 2.9%.

Moreover, he found, sales by plan participants take place ahead of price declines more often than not. Executives initiated 10.4% of their sales ahead of negative earnings news, on average, while only 5.2% came in advance of positive earnings news. "If executives were initiating plans without any inside information, you would expect those sales to even out," Jagolinzer says. "It's like Vegas: Any individual might win, but average everyone out, and you should see winners and losers split 50-50."

So what explains the pattern showing up in the Stanford study? Jagolinzer emphasizes that it's difficult to tell from the numbers alone, and none of it necessarily implies anything illegal. But the data suggest several possible explanations. One is that insiders may sense that a company's fortunes are turning well ahead of when weaker earnings are announced. A key product may not be panning out, or early signs of a sales slowdown may be starting to show up. "People hear things aren't going well; they pick up on a bad vibe," says Jesse M. Fried, co-director of the Center for Law, Business & the Economy at the University of California at Berkeley School of Law. "There's a lot of valuable information floating around a company to trade on that doesn't rise to the standard of being 'material' legally."

Others suggest executives, watching their stock run up, may worry that it is getting overvalued, or they may simply want to take some money off the table. Mark LoPresti, who tracks insider trades for Thomson Financial (TOC ), says many execs, particularly in tech, seem to enact trading plans when stocks return to the levels not seen since the boom years. "The insiders know the previous high, and they may be fearful that the stock may back off again," he says. Whatever the reason, the result is a pattern that can lead to relatively favorable results: Stock plans are enacted as a stock runs up toward a peak, and often the trading slows or stops altogether when the stock declines.


Take, for example, stock sales made by executives of Barr Pharmaceuticals Inc. (BRL ) last year. Several executives, including CFO William T. McKee and President Paul M. Bisaro, set up 10b5-1 plans in the summer and fall. After bouncing between 35 and 50 for the previous 18 months, Barr shares took off in September on strong earnings and new generics. As the stock headed up to $68.70 in February, 2006, the executives -- along with others -- sold heavily, according to company filings. When patent problems caused the shares to tumble to 45 in July, much of the selling stopped. Barr declined to comment.

Sales made by Kenneth E. Keiser, president of soda distributor PepsiAmericas Inc. (PAS ), show a similar pattern. From a low of 11.19 in early 2003, its shares rose to peak above 26 by mid-2005. In February, of that year, Keiser initiated a plan to sell 15,000 shares a month through December. But he ended it in September, as the stock headed back to around 22, where it still trades.

Keiser says he set up the plan to diversify his holdings, having never sold a share in the previous five years. And he says he stopped the plan because the company was engaged in discussions that may have had an impact on the stock. "I felt it wasn't appropriate, so I canceled the plan," he says.

The patterns suggested by Jagolinzer's study also raise the possibility that executives, knowing the dates on which their shares will be sold, may sometimes be able to delay the release of bad news until after the sale. That essentially is what shareholders in SPX alleged. Both the company and Blystone, who resigned under pressure later in 2004, have said the allegations are without merit, and have sought to have the suit dismissed. Blystone could not be reached for comment.

There's one other factor that appears to show insiders don't leave the timing of their trades entirely to chance: plan terminations. The data here are even more sparse, as many execs don't disclose when they stop their pre-arranged plans. But those who do show a discernible pattern: Executives have a tendency to stop their planned sales just before the stock takes off. Often they resume selling when prices are higher. While there is nothing illegal in that -- after all, no one can be forced to continue selling shares -- it belies the argument that executives wishing to diversify simply put their trades on autopilot.

All of which suggests investors ought to pay more attention to trades done within 10b5-1 plans and the signals they send. Thomson's LoPresti says investors tend to ignore such trades, assuming they are out of the control of executives. But that, he argues, is a mistake. "The fact that a trade is being done through a plan doesn't negate its impact," he says. Indeed, Thomson is putting together its own database of 10b5-1 plans to track the trades more closely. The SEC, too, has looked at the possibility that such trades can be manipulated. Jagolinzer may be among the first to look closely at their impact, but he won't be the last.

By Jane Sasseen, with Susan Zegel and John Cady

    Before it's here, it's on the Bloomberg Terminal.