Speech by SEC Staff:
Increased Importance of Models: Disclosure, Fair Value and Accounting

by

Chester S. Spatt

Chief Economist and Director, Office of Economic Analysis
U.S. Securities and Exchange Commission

"Model Governance and Model Validation" Roundtable
Wharton School of the University of Pennsylvania
Philadelphia,Pennsylvania
June 1, 2007

This was prepared for presentation at a panel discussion on "Increased Importance of Models: Disclosure, Fair Value and Accounting" at the "Model Governance and Model Validation" Roundtable at the Wharton School of the University of Pennsylvania on June 1, 2007. The Securities and Exchange Commission disclaims responsibility for any private publication or statement of any SEC employee or Commissioner. This presentation expresses the author's views and does not necessarily reflect those of the Commission, the Commissioners, or other members of the staff.

I am pleased to participate in the panel discussion on the "Increased Importance of Models: Disclosure, Fair Value, and Accounting." I am especially delighted to speak about this here at the University of Pennsylvania, since this is where I began thinking like a professional economist, having received my doctoral degree in economics here. The subject of my presentation is one that I have had considerable opportunity to reflect upon in recent years, particularly due to (a) my work on option expensing, while serving as Chief Economist of the Securities and Exchange Commission, and (b) my understanding of financial theory including derivative security valuation broadly. At the onset of my remarks, I want to emphasize that, of course, the views and perspectives that I am expressing today are my own, and not those of the Commission or my colleagues on the SEC staff.

In recent years there has been considerable focus upon the use of fair value estimates and economic models in accounting, such as for employee stock option expensing. For example, the SEC's Office of Economic Analysis, which I direct, helped to develop the implementation guidance with respect to both models and markets prepared by the SEC's staff for options expensing.1 While the appropriateness of expensing is widely agreed upon by economists as illustrated by the analysis by Bodie, Kaplan and Merton in the Harvard Business Review (2003), it had been a very controversial subject in Washington, D.C. over an extended period of time.

Curiously, some of the companies for whom options grants represented a significant portion of their compensation program were among those that argued the most strongly that they didn't know how to value the options. Furthermore, such companies seem particularly uneasy and uncomfortable about adapting conventional models for use in evaluating the cost of employee stock options. It seems surprising that companies that apparently didn't understand the cost of a particular compensation tool would be inclined to use that instrument to an especially large extent. Alternatively, some critics have argued that under FAS 123(R) that the amount required to be expensed is too high and in excess of the benefit derived by the employees. This does not conform to my own understanding of the accounting standard, which emphasizes the importance of making reasonable parameter choices in order to apply the accounting standard. However, economic principles suggest that under the compensation plan selected by the firm the benefits derived by the firm must ultimately be even larger than the costs incurred by the firm. In any case, corporate accounting ultimately concerns the costs incurred by the firm and its stockholders rather than the employees' benefits. This was precisely the measurement objective suggested by the FAS123(R) accounting standard.

As a result of their training and experiences financial economists intrinsically appreciate that financial options are valuable in a rich array of contexts; however, this is not as widely recognized in the broader population. Because of the efficiency of our financial markets, I would not expect options expensing to lead to substantial changes in the valuation of companies that have significant option programs. However, one potential impact from the use of models for options expensing may be to reduce the potential overuse of these grants in compensation programs by educating boards of directors as to the potential ex ante cost to these grants and removing the natural bias in favor of a compensation tool that was not reflected previously on the profit and loss statement.2

Critics of options expensing have also criticized the modeling of employee stock options, so I also think that it would be helpful to make a few observations about this. While the employee's stock option is not readily hedgeable and most employees are risk averse, the valuation cost to the employer of the resulting liability can potentially be assessed. One method of valuing employee stock options is clear from the history of the market for mortgage-backed securities. This analogy is instructive because of the lack of transferability of the mortgage obligation and the importance of the mortgage borrower's risk preferences. Interesting predictions about exercise and forfeiture behavior can be obtained from the mortgage-backed securities perspective and the use of arbitrage principles3 and the valuation tools of modern financial economics can be adapted to the employee stock option context. The tools for developing the valuation of mortgage-backed securities were developed decades ago4 and in recent years a number of interesting papers have explored the valuation of employee stock options.5 Just as these modeling approaches have been very successful in the context of mortgage-backed securities, I would expect that analogous tools for employee stock option valuation that take into account the relevant frictions would be similarly fruitful. More broadly, the nature of the development of our capital markets and its heavy reliance upon derivative securities provides evidence of the successful application of these tools.

However, firms need not rely explicitly upon models for the purposes of determining the expenses for employee stock valuation. Indeed, FAS 123(R) points to the possibility of using valuations from liquid markets if these were to arise. A few alternatives have been suggested to develop instruments that attempt to replicate the valuation of these options from the perspective of a market instrument. The potential advantages of such market-based instruments would include facilitating the ability of firms to hedge their compensation exposure, making feasible new alternative forms of employee option compensation or even potentially reducing the out-of-pocket costs of measuring the fair value. Of course, the underlying measurement goal of the accounting standard is to obtain a valuation that reflects what a willing buyer and seller would pay with respect to the cost of the firm's exposure. Consequently, there could be tension between the language of the accounting standard and the desires and claims by product innovators that a particular design of a market instrument will lead to valuations that are a fraction of model prices. It's also worth emphasizing that model prices are not hypothetical, but have a strong influence upon market valuations; after all, professionals who trade derivatives rely heavily upon model-based valuations making model and market prices closely intertwined. Not surprisingly, bidders participating in early manifestations of these markets have made clear that they rely heavily upon model valuations in establishing their own willingness to pay and their bids, so the market prices are heavily dependent upon models and indirectly reflect competition about suitable model parameters.

Of course, I'd like to note that I applaud the development of new designs and innovations in this important space, especially in light of the centrality of option grants to the cost structure of many firms. At the same time there are a variety of challenges in doing so. To what extent can parties or even exchanges develop designs such that the resulting valuations reflect underlying liquid markets either at the point of issuance or through deep markets in which the options are re-trading? This may be related to whether the capital markets are being used to directly share large-scale aspects of the risks of the underlying exposure rather than simply an attempt by an interested party to obtain an alternative valuation.

An interesting challenge with respect to fair value accounting is how can accounting policy reward innovative approaches? To the extent that the goal of accounting policy is to replicate a particular measurement objective, how can innovative ways to define or reach that objective be encouraged? Indeed, from the perspective of a registrant it is plausible that the goal is to minimize the expense being measured rather than measuring the expenses more accurately. This suggests potential tension between designs that are of interest to registrants and those that measure the expense more accurately, especially if the out-of-pocket cost of implementing that market-based approach exceeds that of implementing the model alternative.

References

Bettis, J.C., J. Bizjak, and M. Lemmon, 2005, "Exercise Behavior, Valuation, and the Incentive Effects of Employee Stock Options," Journal of Financial Economics 76, 445-470.

Bodie, Z., R. Kaplan and R. Merton, 2003, "For the Last Time: Stock Options Are an Expense," Harvard Business Review 81, 63-71.

Carpenter, J., 1998, "The Exercise and Valuation of Executive Stock Options," Journal of Financial Economics 48, 127-158.

Dunn, K. and J. McConnell, 1981, "Valuation of GNMA Mortgage-Backed Securities," Journal of Finance 36, 599-616.

Dunn, K. and C. Spatt, 1999, "Call Options, Points and Dominance Restrictions on Debt Contracts," Journal of Finance 54, 2317-2337.

Hall, B. and K. Murphy, 2003, "The Trouble with Stock Options," Journal of Economic Perspectives 17, 49-70.

Marquardt, C., 2002, "The Cost of Employee Stock Option Grants: An Empirical Analysis," Journal of Accounting Research 40, 1191-1217.

Merton, R., 1973, "Theory of Rational Option Pricing," Bell Journal of Economics and Management Science, 4, 141-183.

Richard, S. and R. Roll, 1989, "Prepayments on Fixed Rate Mortgage-Backed Securities," Journal of Portfolio Management 15, 73-82.

Spatt, C., C. R. Alexander, D. Dubofsky, M. Nimalendran and G. Oldfield, "Economic Perspective on Employee Option Expensing: Valuation and Implementation of FAS 123(R)," Office of Economic Analysis Memo to Donald Nicolaisen, Chief Accountant, March 18, 2005. http://www.sec.gov/interps/account/secoeamemo032905.pdf

Spatt, C., C. R. Alexander, M. Nimalendran and G. Oldfield, 2005, "Economic Evaluation of Alternative Market Instrument Designs: Toward a Market-Based Approach to Estimating the Fair Value of Employee Stock Options," Office of Economic Analysis of the Securities and Exchange Commission Memorandum. http://www.sec.gov/news/extra/memo083105.htm


Endnotes