FASB fumbles stock options

Alan Reynolds
|
Posted: Apr 15, 2004 12:00 AM

 The Financial Accounting Standards Board (FASB) recently proposed requiring U.S. companies to "expense" all employee stock options starting next year. Public companies would be required to first estimate the "fair value" of such options at the time they are granted, and then subtract that estimate from revenues as if it were a known and current expense (rather than an unknown and future expense). Even though the estimated value of those options changes every day, the estimate at the time they were granted would nonetheless remain the amount deducted over several years of vesting.
 
The business press has generally reported about this issue in a naive and moralistic way. Crusaders for expensing are said to be advocating an obviously more honest measure of earnings, while opponents are said to be niggling over trivial estimating difficulties. A recent Congressional Budget Office (CBO) paper on the topic is not much better.

 Critics of expensing often get bogged down in the difficulty of estimating the value of options when granted. But a much more critical point follows from the fact that, as the CBO inadvertently remarks, "An option's value at the end of the exercise period is almost always different from its value when it was granted." If the correct definition of cost is the actual value when options are exercised, then deducting the estimated value when options are granted must be incorrect.

 The fair value of options to an employee on the day they are granted is an entirely different concept (and dollar amount) from the actual cost to an employer if and when those options are later exercised. So, what is the cost of employee stock options to corporate shareholders, and when does that cost occur?

 The cost of options has heretofore been thought to occur after they are vested and exercised, typically three to five years after they were granted. That is, for example, the official IRS definition of the options' cost -- the date when employees receive taxable income from exercising options and the employer receives a matching deduction. Assuming the stock price has risen, the firm must sell shares to employee at the lower strike price and either buy or issue shares at the higher exercise price. Companies do that by issuing more shares or by using cash or debt to repurchase existing shares. And, contrary to numerous erroneous news reports, the cost of exercised options is certainly not hidden from shareholders.

 Issuing more shares obviously reduces earnings per share, which is the only measure that matters for pricing stock. Using cash to back shares reduces earnings directly, because that cash would otherwise have generated investment income. And using debt to buy back shares reduces earnings through added interest expense.

 According to sophisticated proponents of expensing, however, neither dilution nor share repurchases represent the "true" cost of granting stock options. That cost, they argue, occurs at the moment options are granted. The true cost is said to be the "opportunity cost" of giving these options to employees rather than selling them to investors.

 Before there can be an opportunity cost, however, there must be an opportunity. Firms have no opportunity to sell options comparable to those granted to employees -- options which cannot be traded and whose value depends on whether or not employees may quit or be fired.

 The CBO claims exercised options are "irrelevant to a firm's income statement because they affect shareholders directly, not the firm itself." To treat the firm as something separate from its owners is a bizarre definition of irrelevance, one that highlights how conjectural accounting is now taking precedence over good economics. The only reason income statements matter at all is to assist shareholders in pricing the stock. Yet the CBO somehow claims exercised options "have no effect on the value of the firm." That is quite impossible, since "the value of the firm" depends entirely on how the market values the firm's stock. And the stock's value must be affected by issuing new shares, or by using cash or debt to repurchase old shares.

 Because dilution and buybacks clearly do reduce earnings per share when options are exercised, to also expense fair value at the time options are granted is to double-count two contradictory concepts of cost, rather than choosing between them. Indeed, such deliberate exaggeration of costs is implicit in the least honorable argument on behalf of expensing -- that it will discourage employers from offering stock options even when that form of compensation would otherwise be in the mutual interests of employees, stockholders and the national economy.

 In essence, FASB is proposing to treat a possible future expense if it were as an actual actual expense. This proposed blurring of the critical distinction between actual costs today and possible costs tomorrow matters most to cash-starved younger firms whose earnings are typically reinvested in expanding the firm. There are two reasons: First, deferred expenses are particularly preferable to immediate expenses whenever revenue is expected to be higher in the future. Second, expenses that are contingent on both a higher stock price and employee retention are always preferable to larger fixed salaries from a stockholder's point of view. The fashionable trend of switching from options to restricted stock, by contrast, transfers risk from executives to stockholders -- dilution is immediate and restricted stock retains value even if the stock falls.

 If the unique benefits of stock options in linking risk and reward are artificially discouraged by mandating an artificial redefinition of costs, that will reduce the information and comparability of reported earnings by increasing the portion of earnings that depends on inherently crude estimates constructed with assorted subjective techniques. China, Taiwan and others are likely to retain a more neutral bookkeeping policy with respect to letting employers and employees negotiate their own pay packages. Such countries will therefore be able to make greater use of employee stock options to attract and motivate skilled employees and managers, and to thereby gain an unnatural advantage over the United States in the creation and survival of entrepreneurial ventures.

 The FASB proposal to require companies to treat the estimated fair value of stock options as an actual and current expense rests on a dubious conjecture about what the cost of stock options to stockholders really is and when it occurs. In reality, the estimated value of options to employees at the moment they are granted is not at all the same as the cost to shareholders if and when those options are later exercised. The FASB scheme looks like a risky way to repair some problem that has yet to be seriously defined.