The Inside Story

Increased scrutiny makes interpreting their trades tougher


Over the past two decades that I have tracked the buying and selling of stocks by corporate insiders, I have come to feel a certain kinship with Dr. Heisenberg. The good professor, you'll recall, posited that the mere act of measuring a phenomenon can affect it. Back when few people paid any attention to insider transactions, they were relatively easy to track and interpret. But those days are long past.

In recent years, I've witnessed striking changes in the behavior of corporate insiders. Not the least of which has been the stepped-up spin control by companies as transactions by corporate insiders have gained increased attention. All this has made interpretation of their actions far trickier. As with so many aspects of the stock market, many of the old tools -- and rules -- simply don't work the same these days. Yet that has not been generally recognized.

Perhaps the first change resulted from the change in the behavior of the insiders themselves. In the early 'Eighties, many were barely aware of the rules for filing their trading activity with the SEC. Perhaps one-third of them filed late, or failed to file altogether. Little wonder, then, that reports of insider transactions were sparse. Prior to 1985, I would clip every media mention that appeared; today, I could fill a scrap book with a single month's insider reports. The SEC changed all that in the early 'Nineties by tightening the filing rules, which significantly aided the integrity of the insider data.

Even earlier, around 1987, the public and the media had discovered the value of tracking insider actions. After all, who would have a better feel for a company's prospects -- either improving or otherwise -- than management and employees? Not surprisingly, public-relations-savvy corporations have attempted to put the best face on their own insiders' actions in recent years. These companies, thoroughly cognizant of the implied messages the trades of their insiders may send, have heightened their efforts to orchestrate the "look" of their insider trades. These efforts at orchestration are among the main causes for the heightened complexity involved in the analysis of insider trading behavior today.

Most early studies of insider behavior find common ground in the assumption that, with the exception of new insiders who sometimes feel compelled to make token buys, most insiders are motivated by the same thing as any investor: profits. That assumption has been undercut by corporate campaigns designed to encourage insider purchases of the company's shares.

Indeed, some companies have enacted "executive ownership requirements," which mandate insiders acquire shares equal to some company-specified multiple of their salaries. To facilitate these purchases, some corporations have instituted loan programs. Even companies without executive ownership requirements may offer loan programs to induce insiders to buy. Whether motivated by carrot or stick, such insider purchases typically are heralded by company press releases. It never fails to amuse me, however, that these companies rarely issue similar releases when their insiders sell.

Although buying by corporate insiders under a company-induced program is not negative, a critical eye is needed to interpret such activity. Corporations try to "paint the tape" in this fashion, in the hope that the market will be impressed. When we spot these situations, we try to weight them accordingly.

One recent example involved Kent Electronics. A flurry of December purchases by 14 insiders in the $20-$24.50 range was inferred by some analysts as a screaming "buy" signal. Indeed, subsequent excitement, coupled with a frothy stock market, helped push Kent's shares quickly to the 29 level. Our analysis of the situation was not as bullish owing to an insider-loan program. Kent shares subsequently fell back, dropping 27% to 21 on February 23, when the company warned of a profits shortfall -- not the sort of thing that typically follows a surge of insider buying.

Conseco has one of the largest insider loan programs, helping insiders to buy literally millions of shares in the past two years, which has been trumpeted by company press releases. To be sure, Conseco's shares have performed well, but the importance of this insider buying is tempered by this inducement, in my opinion.

I don't want to create an impression that a majority of U.S. corporations are engaging in practices designed to paint the insider tape. The reality is that only a small handful of them seem to utilize the strategy. These are the very companies, however, whose insider buying seems to stand out the most, at times. When this happens, it's necessary to know what's actually behind the buying.

Another nontraditional type of insider buying -- possibly tied to the surge in shareholder class-action suits filed in the wake of insider selling ahead of bad news -- has emerged. In some cases, insiders, looking to avoid the appearance of having taken advantage of nonpublic information, will make token investments after significant drops in their companies" share prices.

Some of these "token" moves can look quite significant on the surface. Following last year's collapse of Oakley shares, Chairman James Jannard jumped in to buy over 1.9 million shares, at prices between $9 and $11.75, totaling over $20 million. The company immediately issued a press release asserting how the purchase reflected his confidence in the company. Prior to that purchase, however, Jannard had taken out almost $220 million through stock sales since 1995.

Investors follow insider transactions because they are, on the surface, straightforward objective, factual data. The spin put on them by corporate PR departments is something else, who can make a sale sound as if it were a buy.

For instance, a March 10 press release announced: "Staples Chairman Increases Stake in Company's Shares." The release states CEO Thomas Stemberg had exercised options to buy 1,300,000 shares and sold 1,143,000 of them. The release goes on to explain how, since he retained 157,000 of the shares, this "increased" the number of common shares he owned to 2.1 million. But when Stemberg's combined holdings of stock and options are taken together, his total holdings declined, to 4.3 million shares from 5.5 million shares before the options exercise.

The release also fails to mention that other key Staples insiders also disclosed their intention to sell shares, and lots of them. Ronald Sargent, president for North American sales, planned the sale of 200,000 shares, which would decrease his total stake (exercisable options plus common shares) by 33%, his largest disposition since initially filing in 1989. Joseph Vassalluzzo, Staples' president for realty and development, also announced plans to sell 200,000 shares, a 30% reduction in his position, and his largest sale since becoming a filer in 1991. In all, five Staples insiders have indicated their intention to sell a total of 1,646,940 shares in March -- a rather different story than the accumulation spin from the company's PR department.

Staples hardly has been alone in this tack. Here are two more examples: Walt Disney CEO Michael Eisner's December exercise of 7.3 million options, of which 5.7 million shares were subsequently sold, was heralded as a 1.6 million-share increase in his common stake. In actuality, the trade reduced his combined stock and options holdings by over 53%. Also last December, H.J. Heinz CEO Anthony O'Reilly exercised options for 1.12 million shares and liquidated all but 188,769, resulting in a reduction of his net stock and options holdings of about 13%. The company press release portrayed the trade as having increased his share holdings by 3%. Who says American corporations lack creativity?

Data on options, of which insiders now are required to provide a full picture of their holdings, give valuable clues about actual insider activity. This options information helped significantly in our analysis of insider behavior at Intel over the past few years.

Selling by Intel insiders had been decreasing in recent years as the share price rose. At the same time, Intel had been sitting on very large options positions, which moved further in the money as the stock ascended. All of which suggested that corporate insiders were happy to hold on to the Intel stakes rather than cash them in.

The Intel picture has changed in recent months, and it's not for the better. Late last year, insiders began to sell higher numbers of shares at successively lower prices -- always a warning sign to us that insiders think the top's been made.

During July and early October of 1997, for instance, as Intel shares traded at their all-time highs in the $90-$100 range, insiders had sold just 142,500 shares. Barely a blip on our radar screens. By late October, however, the stock had dropped sharply, as low as 69 3/4 . Sales actually increased on the way down as six insiders sold 685,236 shares, at month-end, at prices between $80.86 and $84.61 per share. By year-end, Intel shares again tested their lows, trading to as low as 67 3/8 before rallying back to the 95 range. On the rally, however, insiders were again quick to sell, and at even lower prices than they fetched in October -- over 916,000 shares in January at the again-lower prices of $75-$82. Even with the strong performance of technology in the first quarter, Intel last week remained in the mid-70s.

Signs of insider selling at other big-cap companies are beginning to emerge, perhaps indicating a shift in sentiment to the bearish side. That may be significant given insiders' track record. In 1994, when the Federal Reserve increased interest rates and seemed to threaten the bull market, insiders at large-cap companies were impressive buyers amid the market gloom. Typically, the number of sellers exceeds buyers by a 2-to-1 ratio. With each passing month in 1994, that ratio sank steadily below 2-to-1, and by year-end, it hit 1-to-1-something I've never seen in 20 years of tracking insider data. From there, of course, the market has gone on to three record-setting years.

Now, with the Dow closing in on 9000 and the Asian economic slowdown affecting an increasing number of U.S. corporate earnings reports, insiders seem to be going in the opposite direction from where they headed in 1994. We are sensing that a bearish trend we recognized late last year may be still intact. In August 1997, the insider sell/buy ratio moved above 3-to-1, a historically high mark, and remained there until year-end. But the market weakness of late December-early January apparently deterred some selling, and brought the sell/buy ratio back to its 2-to-1 norm.

There are several caveats in the January ratio improvement, however. We spotted many insider purchases associated with corporate loan programs rather than traditional cash investment buying. And although overall insider transactions seemed to be improving in January, we began to note an increase in insider selling in large-cap companies.

As the market improved in February, however, insiders resumed selling with a vengeance. Our sell/buy ratio has quickly moved back above the 3-to-1 level. If this represents the real trend of insider sentiment, the market could be in for some nasty weather ahead.

The analysis of insider sales at individual companies can be a bit more complex, however. Take, for instance, widely discussed sales at such big names as Microsoft, Dell Computer and Compaq. Insider sales at Microsoft, for instance, totaled 20 million shares in February, consisting mainly of four million by CEO Bill Gates and 14.8 million by co-founder Paul Allen. That, however, represented small portions of their Microsoft stakes, and left them with over 532 million shares and 166 million, respectively.

When splits are taken into account (there have been six of them since 1990), current sales by insiders are not high, especially compared with past years. Adjusted for splits, insiders sold 89 million shares in 1991, 55.8 million shares in 1992, 80.5 million in 1993, and 51.9 million in 1994. By 1995, insider sales had dwindled to eight million shares, 25.2 million in 1996, and 40 million in 1997.

Insider sales at Compaq and Dell also have attracted lots of attention, especially at the former computer maker, where sales occurred before a profit warning and subsequent drop in the shares. Notwithstanding the prescient appearance of the Compaq sales, insiders at the company are restricted to quarterly windows for selling, which may explain the timing of these recent sales. More evidence is needed to say more about them.

By contrast, insider selling at General Electric stands out. While relatively small in volume, the liquidation of 158,000 shares is anomalous given the rarity of selling by GE insiders.

Selling by insiders at electric utilities also appears to be on the increase. Over the years, they've been uncannily prescient regarding the direction of long-term interest rates, historically the main driver of this stock sector. They've tended to buy just ahead of or during downturns in bond yields, and vice-versa. Selling among electric-utility insiders emerged in mid-1993, a few months before the lows in bond yields, and may be surfacing once again. That doesn't jibe with the widespread bullishness in the bond market.

Where are we seeing buys?

They're still there, though they've become harder to spot. And there has not been much evidence of well-defined industry patterns in recent months.

Some of the accumulation is subtle but that's why we like it. Take, for instance, the accumulation at Vencor, where four insiders have acquired 258,911 shares since December 1. Only 40,000 of the shares were bought in the open market; the balance were acquired through options exercises where the insiders retained the shares. Two of the three involved cashing in some of their non-qualified options. This type of option results in an income-tax liability even though the gains have yet to be realized. Should the stock fall, the insiders would still be liable for the tax as if they had cashed in. That displays confidence on the part of the insiders, to say the least. Moreover, there was no press release from the company, which says more to us than PR hype.

Elsewhere, the beaten-down oil and gas sector not only has seen insiders holding on to their shares but accumulating at current price levels as well. Other companies where we have been monitoring recent insider buying appear in a table nearby.

While much has changed in tracking insider transactions, one verity remains: Nobody knows as much about a given company than the folks who run it. Not even Dr. Heisenberg can change that.

BOB GABELE is president of CDA/Investnet, a Fort Lauderdale company that specializes in analysis of insider data.


  1. Why might insider trading provide an indication of whether stocks are under or over valued?
  2. What are some of the changes in insider trading that the author notes have occurred in recent years, and what are the implications for using insider trading volume as a stock picking tool?
  3. Insider trading data is available much more easily accessible today than it was a decade ago. What implications does this have for the use of insider trading volume in investment strategies?