Accounting for Employee Stock Options

Transcription

1 Accounting for Employee Stock Options By Wayne Guay, S.P. Kothari and Richard Sloan* Accounting for employee stock options (ESOs) is controversial, with many arguing that it has substantial economic consequences. Such arguments rely on the assumption that one or more interested parties fixate on accounting numbers and fail to understand the real costs and benefits of ESOs. We review the various accounting issues and economic consequence arguments created by ESOs. We conclude that the accounting should facilitate a clear and consistent understanding of the costs of doing business, and that expensing ESOs best achieves this objective. Employee stock options (ESOs) are a ubiquitous form of compensation in corporate America. By the late 1990s, ESOs outstanding at large corporations averaged 7% of total outstanding shares, with top executives holding approximately one-third of total ESOs (John Core and Wayne Guay, 2001). Empirical evidence suggests that firms use ESOs to align employees and shareholders' interests, attract and retain employees, and compensate employees for their labor while simultaneously raising capital from employees (Kevin Murphy, 1999; Core and Guay, 1999 and 2001). There is currently an intense debate nationally and internationally among standard setters, politicians, investors, corporate executives, and academics about whether to require corporations to deduct the estimated value of ESO grants as a business expense in reported income. Existing accounting standards require firms to expense most forms of pay, such as salaries, cash bonuses, and the value of stock grants, but allow firms to choose whether to expense the value of ESO grants. Until very recently, nearly all firms chose not to expense ESOs. However, firms that do 1

2 not expense ESOs must publicly disclose in the financial statement footnotes what reported income would have been if the ESOs were expensed. In a recent sample of large growth firms, Christine Botosan and Marlene Plumlee (2001) find that mandatory expensing of ESOs would have resulted in a 14% median reduction in firms' earnings per share. Firms are also required to disclose details of top executive ESO compensation in the annual proxy statement. Underlying the ESO debate is the concern that the choice among alternative financial accounting treatments have real economic consequences. A large literature beginning with Ross Watts and Jerold Zimmerman (1978) provides evidence that accounting choice can impose economic costs on firms when contracts (e.g., debt and executive compensation contracts) or influential external parties (e.g., tax authorities) rely on reported accounting numbers (see Thomas Fields, Thomas Lys and Linda Vincent, 2001, for a survey of this literature). Accounting choice can also have economic consequences if investors fixate on particular numbers, such as reported earnings, resulting in security mispricing and misallocation of capital. Proponents of mandatory expensing argue that ESOs reflect a cost of acquiring employee labor, and that expensing ESOs conveys this information to outsiders consistently with other labor costs. Some argue that the absence of ESO expense results in stock mispricings because investors fixate on reported earnings and fail to understand or utilize supplemental footnote disclosures about the true economic cost of ESO grants. Others argue that when investors and boards of directors fixate on accounting earnings, the absence of ESO expense exacerbates ineffective corporate governance, and allows management to use ESOs to extract excessive compensation. Proponents of this view argue that expensing ESOs will reign in management compensation by putting it under a brighter light. 2

3 Opponents of expensing ESOs argue that deducting ESOs' cost from earnings conveys an impression of weaker financial results to investors and, under the assumption that investors fixate on reported earnings, could raise the firms cost of financing and stifle corporate investment and innovation. There is also a concern that external parties, such as taxing authorities, might use changes in financial accounting treatment as a cue to alter regulatory and tax policy. In section I, we discuss ESO accounting issues, including our rationale for expensing ESOs. Section II explores economic consequences of expensing ESOs. Section III concludes. I. Accounting Issues The ESO transaction involves the exchange of labor inputs for a contingent equity claim on the firm, and raises several complex accounting issues. We consider three such issues below. A. ESO Issuance Combines Operating and Financing Activities Granting ESOs is economically equivalent to two separate transactions. In the first transaction, the firm sells warrants to the employee for cash. This is a pure financing transaction, resulting in the generation of cash and an increase in the firm's equity capital. In the second transaction, the firm pays the cash to the employee as compensation for services rendered. This is a pure operating transaction, resulting in the subsequent use of resources and a corresponding charge to earnings. Thus, consistent with the existing accounting for stock grants to employees, the proper accounting treatment for an ESO grant is an entry to increase contributed equity capital and an entry to deduct the value of ESOs from reported earnings. To see this point, consider two economically equivalent firms: Firm A issues common stock warrants to investors for cash and then uses cash to pay for all production inputs. Firm B uses ESOs to pay for all production inputs. Firm A computes earnings as revenue received from the sale of the inputs less the purchase price of the inputs. Assuming no expense for ESOs, Firm 3

4 B s earnings equal revenues. Thus, earnings are very different across the two firms, yet both firms raise the same amount of capital and generate the same economic earnings. Many prominent business leaders, such as Harvey Golub (former CEO, American Express) and Andrew Grove (CEO, Intel), argue passionately against expensing ESOs. Their argument is that the economic impact of ESOs manifests itself through shareholder dilution, and that the denominator in the computation of earnings per share (EPS) already adequately captures the dilution cost of ESOs. As a result, expensing ESOs essentially double-counts ESOs' cost. As background, for a firm with common stock and ESOs outstanding, the denominator in the computation of EPS is the sum of (i) the average number of outstanding common shares, and (ii) an adjustment for outstanding ESOs based on a formula that converts the number of outstanding ESOs into an equivalent number of common shares. However, the above argument is flawed. It makes the mistake of confusing the financing implications of raising equity capital with the operating costs of paying for operating resources. To clarify this point, consider the following two firms: Firm A initially holds $200 in assets funded by 20 shares of common stock issued at $10 per share. Firm A sells the assets for $220, recognizes $220 of revenue and $200 of expense. Earnings are $20 and EPS is $1 ($20/20 shares), reflecting a 10% return on equity capital. Firm B holds $100 in assets funded by 10 shares of common stock issued at $10 per share, and also grants $100 of stock to a new employee in return for labor worth $100. Firm B then sells the assets and labor for $220. If the cost of the new labor is not expensed, Firm B recognizes $220 of revenues and $100 of expense. Earnings are $120 and EPS is $6 ($120/20 shares), reflecting a 60% return on equity capital. Although both firms generate a 10% economic return on capital, Firm B s EPS reflects a misleading accounting return of 60% on capital. Note that because the denominator of Firm B's 4

5 EPS includes an adjustment for the 10 shares of stock issued to acquire labor, the problem with Firm B's EPS stems solely from the failure to deduct labor expense from reported earnings in the numerator. This example illustrates the need for an EPS measure where (i) the numerator of EPS includes a deduction for ESO grants to reflect the cost of the labor resources received by the firm upon granting the option, and (ii) the denominator includes an adjustment for outstanding shares and ESOs to reflect the fact that existing shareholders and option holders have an economic claim on the firm s earnings performance. 1 Although the above example assumes stock, not ESOs, are granted to acquire labor, the economic intuition is identical when ESOs are granted to acquire labor (see Dieter Hess and Erik Lueders, 2001; Core, Guay, and Kothari, 2002). B. ESO Grants are a Barter Transaction Granting ESOs is a barter transaction involving the exchange of labor services for a contingent equity claim. As with all barter transactions, determination of the fair value of the exchange is an accounting issue. Although option-pricing techniques (e.g., Black Scholes) are well developed, ESOs have features, such as vesting provisions, nontransferability, and accelerated maturity when the holder terminates employment, that deviate from the assumptions underlying standard option pricing models for publicly traded options. As a result, managers exercise policies likely deviate from the assumptions underlying the Black-Scholes framework (e.g., Steven Huddart, 1994; and Charles Cuny and Phillipe Jorion, 1995). Reasonable solutions to this problem have been proposed in which ESOs are valued using the expected time until exercise (e.g., Thomas Hemmer, Steven Matsunaga, and Terry Shevlin). Accounting standard setters are often reluctant to recognize numbers in the financial statements that cannot be measured reliably (e.g., research and development expenditures are not recognized as assets because it is argued that the future benefits of these expenditures cannot be 5

6 reliably estimated). Some argue that ESOs should not be expensed for this reason. Further, empirical evidence suggests that some firms use discretion in the assumptions underlying ESO valuation to manipulate ESO expense (Mary Barth, David Aboody, and Ron Kasznik, 2002). However, we believe that ESO grant valuation is no more complicated than the estimation of many other common corporate expenses (e.g., the annual expense recognized for pensions and postretirement benefits is based on estimates of the present value of future retirement benefits earned by current employees during the year). A related argument is that because employees are risk averse and are not allowed to construct the type of risk free hedges that form the basis for option pricing models, traditional option pricing models may overvalue ESO grants from the employee s perspective (e.g., Brian Hall and Murphy, 2002). Some firms use these arguments to justify either a lower option expense or to bolster the case against the reliable measurement of ESO value. However, regardless of employees' valuation of ESOs, it is the cost of the ESO grant to the firm s existing owners rather than the employees valuation that matters when determining the appropriate amount to expense. To illustrate, companies sometimes reward employees with fringe benefits for incentive or reward purposes, such as first class air tickets or country club memberships. The fact that some employees value these perquisites at less than company cost does not mean the company should expense these benefits at less than cost. C. ESOs as a Contingent Financial Obligation Through Time A third issue arising in expensing ESOs is the treatment of changes in the fair value of the contingent equity claim between the grant and exercise dates. The amount of value the option holder ultimately receives depends on the exercise value (i.e., share price) at the exercise 6

7 date. Thus, because option holders bear risk associated with changes in equity value over time, ESOs provide existing shareholders with a form of insurance against future firm performance. An interesting accounting issue is whether the ex post realization of the risk borne by the option holders should be reflected in firm earnings. That is, should changes in ESO value between the grant and exercise dates be included as a component of earnings to reflect that fact that option holders receive a portion of any change in the net asset value of the firm? Such an approach is consistent with an accounting objective where earnings appropriately reflect any change in the common stockholders net assets (i.e., assets less liabilities). A problem with this approach is that the change ESO portfolio value reflects changes in the firm's stock price, and stock price reflects the capitalized change in the present value of expected firm earnings. Because accounting earnings generally do not reflect changes in the present value of most assets, marking-to-market some components of equity, but not assets, may result in an earnings stream that hinders market participants in assessing the amounts and risk of firms future cash flows. 2 II. Economic Consequences of Expensing ESOs Accounting for ESOs is one of the most controversial accounting issues in recent history. Corporate America and the major accounting firms lobbied aggressively against the mandated expensing of stock options during the 1990s. Their efforts forced the Financial Accounting Standards Board (FASB) to back down and recommend, but not require the expensing of ESOs. The FASB summarizes the most commonly voiced concern with the mandatory expensing of ESOs as follows: Respondents who objected to the proposed accounting on the basis of public policy concerns asserted that the recognition of compensation costs for fixed stock options will result in lower stock prices and higher costs of capital and will therefore cause many 7

8 companies to eliminate or significantly curtail their stock-price based compensation programs 3 Firms publicly disclose details of ESO plans in their financial statements, including the estimated cost of option grants and the total number of ESOs outstanding. Details about the level and composition of top executives' compensation, including the estimated value of the ESO grants, are disclosed publicly in the annual proxy statement. Thus, the concern described above relies on a form of market inefficiency where marginal investors fixate on reported earnings and ignore information about business expenses not explicitly recognized in contemporaneous earnings. Accumulated empirical research over the past four decades contradicts this extreme form of market inefficiency (for evidence that investors do not ignore ESO disclosures see Aboody (1996) and Timothy Bell, Wayne Landsman, Bruce Miller and Shu Yeh (2002)). But even if markets were characterized by such inefficiency, it would seem to strengthen the case for recognizing these very real costs in earnings. One hypothesis for executives' opposition to expensing ESOs, is that it would influence contracting arrangements by making ESO compensation to top executives more visible. This, in turn, would make it more difficult for top executives in firms characterized by poor corporate governance to justify awarding themselves excessively lucrative pay packages. Contracting and transactions costs render corporate governance an imperfect process, and unlike stock prices which evidence suggests are influenced primarily by marginal investors, the effectiveness of corporate governance depends on the actions of all voting shareholders. Individual shareholders often do not have strong incentives to expend effort on governance activities, and without a transparent and potentially contractible number associated with ESO grants, some top executives can use ESOs to transfer wealth from shareholders. Supporting empirical evidence finds firms that most actively lobbied against the expensing of ESOs are characterized by top executives 8

9 who receive a greater proportion of their compensation from options, receive higher total compensation, and use ESOs more aggressively for themselves versus other employees (Patricia Dechow, Amy Hutton and Richard Sloan, 1996). This hypothesis generates an empirical prediction that ESO awards, especially to top executives, will decline following the mandatory expensing of ESOs, particularly in firms with ineffective corporate governance. Recently many firms have voluntarily expensed the cost of ESO grants. The hypothesis also predicts that the governance of expensing firms is more effective than that of the non-expensers. 4 A final possibility to explain firms' opposition, is that expensing ESOs imposes real economic costs on firms related to contracting or influential external parties. For example, lower publicly reported profits due to ESO expense might result in the loss of customer confidence or more binding debt covenants. Alternatively, young growth firms that rely heavily on ESOs may fear that the FASB's endorsement of ESO expense will prompt the IRS to tax ESO grants to employees as regular compensation, instead of deferring employee tax until exercise. III. Conclusion Corporate and political pressures should not determine ESO accounting rules. ESOs are a key component of top executive compensation that serve useful contracting functions. However, the goal of accounting is not to distort financial performance to subsidize particular business activities. Accounting should reflect the true costs of doing business, and labor acquired through ESO grants is a real economic cost that firms should deduct from earnings as an expense. Armed with clear information on operating costs, investors, creditors, boards of directors and regulators should be left to determine business practice. 9

10 Footnotes * Guay: The Wharton School, University of Pennsylvania, Philadelphia, PA, 19104; Kothari: Sloan School of Management, Massachusetts Institute of Technology, Cambridge, MA 02142; Sloan: University of Michigan Business School, Ann Arbor, MI We acknowledge helpful comments of John Core, Rich Frankel and Joe Weber. 1. Core, Guay, and Kothari (2002) document that although an ESO adjustment to the denominator of EPS is indeed necessary, the current accounting rules require an adjustment that understates the economic dilution of outstanding ESOs by about 50%, on average. 2. For example, consider a technology firm with substantial outstanding ESOs whose stock price rises substantially due to good news about future sales from a newly patented product. If the firm expenses the increase in ESO value that results from the increased stock price, the firm will experience a sharp decline in reported earnings in a period where good news occurs. The potential informational problem here stems from the fact that reported accounting earnings in general do not capture large portions of the information reflected in current stock returns. 3. From FASB Preliminary Summary of Responses to Exposure Draft, issued June 30, We note however, that compensation contracts will only be written efficiently if boards and investors fully understand both the cost of ESOs and the incentive effects of options, stock, cash, etc. Reaching agreement about the appropriate ESO expense, and conveying this information to boards and investors is likely to be much easier than the difficult tasks of ensuring that boards and investors understand the incentive effects of various types of compensation contracts, and understand which contracts are most efficient in certain settings. 10

11 References Aboody, David. Market Valuation of Employee Stock Options. Journal of Accounting and Economics, August-December 1996, 22(1-3), pp Aboody, David; Barth, Mary and Kasznik, Ron. Do Firms Manage Stock-Based Compensation Expense Disclosed Under SFAS 123? Working paper, Stanford University, Bell, Timothy; Landsman, Wayne; Miller, Bruce and Yeh, Shu. The Valuation Implications of Employee Stock-Option Accounting for Profitable Computer Software Firms. The Accounting Review, October 2002, 77(4), pp Botosan, Christine and Plumlee, Marlene. Stock Option Expense: The Sword of Damocles Revealed. Accounting Horizons, December 2001, 15(4), pp Core, John and Guay, Wayne. The Use of Equity Grants to Manage Optimal Equity Incentive Levels. Journal of Accounting & Economics, December 1999, 28(2), pp Stock Option Plans for Non-Executive Employees. Journal of Financial Economics, August 2001, 61(2), pp Core, John; Guay, Wayne and Kothari, S. The Economic Dilution of Employee Stock Options: Diluted EPS for Valuation and Financial Reporting. The Accounting Review, July 2002, 77(3), pp Cuny, Charles and Jorion, Philippe. Valuing Executive Stock Options with a Departure Decision. Journal of Accounting and Economics, September 1995, 20(2), pp Dechow, Patricia; Hutton, Amy and Sloan, Richard. Economic Consequences of Accounting for Stock-Based Compensation. Journal of Accounting Research 1996, 34(Supplement), pp

12 Fields, Thomas; Lys, Thomas and Vincent, Linda. Empirical Research on Accounting Choice. Journal of Accounting and Economics, September 2001, 31(1-3), pp Hall, Brian and Murphy, Kevin. Stock Options for Undiversified Executives. Journal of Accounting & Economics, February 2002, 33(1), pp Hemmer, Thomas; Matsunaga, Steven and Shevlin, Terry. Estimating the Fair Value of Employee Stock Options with Expected Early Exercise. Accounting Horizons, 1994, 8, pp Hess, Dieter and Lueders, Erik. Accounting for Stock-Based Compensation: An Extended Clean Surplus Relation. Working paper, University of Konstanz, Huddart, Steven. Employee Stock Options. Journal of Accounting and Economics, September 1994, 18(2), pp Murphy, Kevin. Executive Compensation. Handbook of Labor Economics, Vol. 3, North Holland, Watts, Ross and Zimmerman, Jerold. Towards a Positive Theory of the Determination of Accounting Standards. The Accounting Review, January 1978, 53(1), pp