Sunday, December 17, 2006

Why U.S. and European corporate scandals are different (hint: it has to do with stock options)

Columbia Law School professor John C. Coffee, Jr. has an interesting paper out there on why U.S. and European corporate scandals have been different. (See A Theory of Corporate Scandals: Why the U.S. and Europe Differ , March 2005.) It’s already more than a year and a half old, but the reason I bring it up is that what is says now has new pertinence given the recent stock option backdating scandals.

Coffee’s notes that, while Europe has had its share of financial scandals since the bursting of the stock market bubble in 2000 (Parmalat, Shell Oil, Royal Dutch Ahold, Vivendi, and now Siemens ), these scandals have differed in significant ways from those in the United States. In particular, Coffee notes that, while Enron, Worldcom and other big scandals have grabbed all the news, actual accounting restatements in the U.S. between 1997 and 2002 were far more numerous than is commonly understood (at least 10 percent of all listed companies). (Accounting restatements involve a company announcing that previous financial statements it made contained errors. Earnings restatements involve the company announcing that its earlier statements about how much money it made were wrong. Sometimes these involve mistakes where the company actually made more money than it thought it did, but most, as you can imagine, involve the company overstating how much it made. Obviously, a company that announces it is making a lot of money tends to see its stock price go up.)

Coffee also notes that it hasn’t always been this way. In 1990, for example, there were only 33 earnings restatements; in 1995 there were 50. By 1999, there were 233; 270 in 2001; 330 in 2002; 323 in 2003; and 414 in 2004.

While, comparatively speaking, only a handful of these accounting restatements have resulted in fraud charges brought against corporate leaders, stock markets have reacted to these restatements as if they were evidence of fraud — restating issuers lost on average 11 percent of their stock market value within 3 days of the restatement announcement, and on average 25 percent of their market value in the 120 days prior to the announcement. (Incidentally, if this isn’t a sign of rampant trading based on inside information, I don’t know what is. For all of those insider trading law haters out there (e.g. Stephen Bainbridge ), it seems clear that enforcement isn’t a plus-minus variable, but a continuum. And, at least in absolute terms, the prohibition on insider trading in the U.S. is honored more in the breach than in the observance.)

Why the difference? Coffee claims it comes down to executive compensation. Starting in the 1990s, U.S. companies shifted the way they pay their executives from cash to stock and (particularly) stock options. The reason for this shift was an attempt to align the interests of management with those of the shareholders — a problem that had been recognized since at least 1932 when Adolf Berle and Gardiner Means wrote about the separation of ownership and control in The Modern Corporation and Private Property. (Somewhat interesting side note: Coffee holds the Adolf A. Berle professorship at Columbia.) Since stock options mean the option holders make more money the higher the company’s stock price goes, the idea is that this would encourage executives to work hard to improve the company’s performance.

Of course, it hasn’t always worked that way in practice. CEO salaries have increased significantly over the past 15 years (from an average $1.2 million in 1990, with 92 percent of that cash, to over $6 million in 2001, with 66 percent of that in stock and options). As Coffee explains, this creates tremendous incentives for CEOs to manipulate a company’s financial statements, particularly when those stock options are in the money (originally issued at the then-current stock price). For example, say a CEO that holds options on two million shares of the company’s stock and that stock is trading at a price-to-earnings ratio of 30 to 1. If that CEO can cause the company’s financial statements to prematurely recognize future revenues such that the annual earnings increase by only $1 per share, the value of the CEO’s options increase by $60 million.

Further, Coffee cites studies showing that this temptation is powerfully effective. One study shows that in 2001-002, the factor with the most influence on the likelihood of a restatement was the presence of a substantial amount of in the money stock options in the hands of the CEO. If the CEO held options equalling or exceeding 20 times his or her annual salary (not something terribly unusual), the likelihood of a restatement was 55 percent!

By contrast, Coffee shows that, in Europe, the incentives are very different. Ownership of major companies in most of Continental Europe is much more concentrated than in the U.S. In the United States, for example, only about 15 percent of public companies are controlled by one shareholder or a small, interconnected group of shareholders (i.e. where these controlling shareholders own or vote 50 percent or more of the company’s shares). In Italy, 59.6 percent of public companies are controlled, while this figure is 64.6 percent in Germany and 64.8 percent in France. Under such a system, a company’s managers are much more easily monitored by the company’s owners. (In some cases, such as Parmalat, the majority owners were the managers.) Consequently, there is less need for the company to align the interests of the company’s executives with shareholders, since managers have less discretion to engage in opportunistic activities or loot the company. In addition, the controlling shareholder also has much less of an interest in the day-to-day price of the company’s stock, since these shareholders rarely, if ever, sell their shares to the public, as this would dilute their control over the company.

Rather than massage earnings disclosures to artificially boost stock prices, European corporate scandals such as Parmalat tend to involve controlling shareholders expropriating corporate assets for their own benefit. Of course, there is no incentive to do this where the controlling shareholders own all of a company; however, if they own only 51 percent and have sold 49 percent of the company’s shares to the public, that means (essentially) that 49 percent of the company’s assets are available for the taking. For example, as with Parmalat, the board of directors could vote to transfer company funds to a travel agency owned by the controlling shareholder’s daughter, in exchange for worthless or nonexistent services or products.

Coffee points out that some of Europe’s recent financial scandals, however, have involved earnings restatements — but that these tend to prove the rule, since most of these companies had evolved into American-style operations with diversified shareholding (i.e. Vivendi, Royal Dutch Ahold, Skandia Insurance and Adecco).

What I found most interesting about this paper is the focus apparent danger presented by in the money stock options. The recent options backdating scandals involved an attempt to make out of the money options into in the money options by changing the dates of the option issuances. In a sense, then, options backdating presents two dangers — outright fraud regarding disclosure of the value of the options given to executives, and an increase in the incentives for management to massage earnings.

For previous posts giving some background on the options backdating scandals, see: