Home > Stock Options > Articles Online >

Is Expensing the Y2K of Equity Compensation?

Corey Rosen, NCEO Executive Director

Barring unlikely Congressional action, in 2005, public companies will have to start showing an accounting expense for stock options. Many companies object to this, saying they will have to cut back on their plans to avoid a decline in their stock prices. However, the research indicates otherwise: expensing does not affect stock prices. Corporate leaders would be well-advised to base decisions on real, rather than perceived, costs and benefits.

Barring unlikely Congressional action, in 2005, public companies will have to start showing an accounting expense for stock options. Other equity pay is already expensed, but new rules will make some changes in how this is done. Many surveys, as well as CEO testimonials, indicate that expensing will force companies to cut back on options, usually by reducing the number of employees eligible for them. Overall, about 45% of companies say they will follow this course, both on stock options and their close cousin, employee stock purchase plans. But few companies say they will reduce total equity pay for executives, albeit the form may change.

It is important to understand that the accounting rules will not, as widely reported, "cost" companies any more money. Accounting changes simply require a new way to report what options already cost; companies and their shareholders won't pay a nickel more or less for options after this change.

An underlying assumption of those concerned about being forced to expense stock options is that companies with significant options programs would see their stock prices decline in response. Of course, this is just an assumption. Others have argued that there won't be much effect at all, because this information already exists in the income statement's footnotes. Markets, they say, already factor options costs into a company's stock price. Indeed, previous reforms to accounting conventions, such as in health care costs in the 1980s, have had no impact on stock prices. A more nuanced argument contends that companies with options programs that do not seem well targeted or well justified in terms of compensation strategies would suffer, but others would not, or that only companies with options programs well outside their industry norms would suffer.

If options expensing really will not affect share prices, then the move by companies to change their plans in response to accounting changes seems to be a case of mass delusion. Even more delusional is the decision to reduce options for rank-and-file employees when reams of data show that concentrating equity compensation in top executives is unrelated or negatively related to corporate performance and shareholder return while broad-based programs are positively related.

Based on the survey and the empirical research reported below, it appears that options expensing will have very little impact, if any, on stock prices. Expensing could well turn out to be the Y2K problem of equity compensation-much feared, but with little impact. Or perhaps we should say little impact on stock prices-but potentially a major and costly impact on corporate performance. Rarely have business decisions of such moment been made by so many people on such bad information.

To assess these arguments, in January 2003 we conducted an email survey of 180 finance professors at 30 leading graduate schools of business. We received 37 responses. Of these, 14 respondents said expensing would either have no impact, 7 said it would have a small impact, and 10 said it would have a small impact on companies with expenses larger than industry norms. Only 2 thought it would have a large impact across the board, and only 4 thought it would have a large impact on companies with unusually large options expense.

A large number of empirical studies have now put specific data to these theories. None suggest that there will be a significant impact. The largest and most significant of these are summarized briefly below. In general, they show that the market already incorporates expensing data into stock prices. This does not mean that investors do not care about what options expenses are. In fact, the studies generally show that higher expenses are associated with lower stock prices. It does mean that a company that already has high expensing costs in its footnotes is not likely to see a hit to its stock price when it has to move these costs to its income statement once expensing is required.

The Towers Perrin 2004 Study

In 2004, Towers Perrin analysts looked at 335 companies that voluntarily chose to expense options between April 2, 2001 and August 14, 2003 (Michael Grund and Richard Ericson, "Options Expensing Announcement Has No Impact on Share Prices," March 31, 2004). Share prices were tracked from the day of the company's earnings declarations and the 150 days before and after. The results were then compared to the S&P 500 and MidCap 400 indices. Adjusted for general market movement, the average price of the announcing companies over the period showed no significant variation from the 900 companies in the companies index. The report concluded that options expensing does not affect share prices.

"Expensing Stock Options: What Are the Markets Saying?", Ashish Garg and William T. Wilson, 2003

This study of whether expensing will affect share price was published in The Journal of Employee Ownership Law and Finance 15, no. 4 (winter 2003). Garg is a senior manager in the economics and business analytics practice of Ernst & Young, while Wilson is a senior economist at the firm. Both hold doctorates in economics. The study is of particular importance because it was able to hold constant the impact of options expensing on earnings per share in calculating market reactions to expensing announcements. It found that expensing does not cause a change in stock prices.

The authors used an event study methodology. Event studies measure stock market reaction to events one or more days before and after the event. The movement is adjusted for movement in stock prices generally, as well as by sector or other variables. The study looked at 54 companies announcing options expensing in the last half of 2002 and early 2003 that that did not have simultaneous other factors that could affect the results. The study looked at the three-day period surrounding the announcement.

The results present a very clear picture. The average abnormal share reaction to expensing over the three-day window being studied was only +0.02% (in other words, stock prices rose .02% more than would be expected). The median was -0.52%. Neither result was statistically significant (not likely to occur at random), however. Industry, market capitalization and other company factors did not make any difference. More important, companies with larger expenses actually had a somewhat higher than normal shareholder return.

The Towers Perrin Study, 2002

In a study of the market's reaction to announcements that companies were going to expense options, Towers Perrin found that in the 120 days surrounding 103 companies' summer 2002 announcements that they were going to expense options, stock prices at these companies did not go up or down more than would have been expected based on normalized comparative results for the market in general ("Announcement of Option Expensing Has No Impact on Share Price, Towers Perrin Study Finds," Nov. 21, 2002). Richard Ericson and Michael Grund of Towers Perrin performed the study. Like the Garg and Wilson study described above, the Towers Perrin analysis found no correlation between the size of the impact expensing would have had on earnings in 2001 and stock prices.

The Haidan Li Study, 2002

In "Employee Stock Options, Residual Income Valuation and Stock Price Reaction to SFAS Footnote Disclosures" (ERN Labor Journals, December 19, 2002) Haidan Li at the University of Iowa looked at the three-day periods surrounding both the announcements of 10-K filings (which provide details on corporate earnings) and the three-day period surrounding earnings announcements to the public (these are typically released some time before 10-K filings). The study included data from 1,500 S&P companies. Li found that the market does react negatively (albeit in a relatively minor way) to "unexpected" stock options expenses in the period around 10-K filings, but there is "no significant association between unexpected stock option expense and unexpected stock returns around earnings announcements." What this means, she says, is that the market already incorporates data from footnote disclosures, indicating that formal expensing would not add much new information.

Conclusion

A number of previous studies also assessed this problem, but these were performed before very many companies started to expense options. The analyses were thus more indirect, but they came to the same conclusion: expensing won't affect share prices. It is not difficult to understand why corporate executives and many pundits on this issue are ignoring the research. After all, it seems like recording the expense should affect share prices, and for many years companies used options instead of sometimes more effective appropriate strategies because they did not have to record an expense for options. But many things seem like they should be true, only to be disproven by research. It may well be that once companies start to expense, if the market, as these studies indicate, just yawns, companies will take a more rational approach to how they decide about equity pay. But herd instincts can be powerful, even among sophisticated corporations. All of this is not to say that expensing is a good idea on its own merits; the NCEO does not take a position on that issue (we do not take lobbying positions). It would be wise, however, for corporate leaders to step back and reassess decisions they are planning on equity pay and base them instead on real, rather than perceived, costs and benefits.

For more details on these studies, as well as on surveys of corporate intentions and research on equity compensation and corporate performance, see the NCEO's issue brief, The Future of Broad-Based Stock Options. Corey Rosen, the author of this article, can be contacted at crosen@nceo.org.

^^Top of Page