In our forthcoming Journal of Financial Economics paper, Stock Options and Managerial Incentives for Risk Taking, we exploit the change in the accounting treatment of stock-based compensation under FAS 123R, which was issued by the Financial Accounting Standards Board (FASB) and took effect in December 2005, to provide new evidence on the role that convexity in compensation contracts plays in providing incentives for risk taking by managers. An additional rationale that is often stated for the dramatic rise in option-based compensation over time revolves around how stock options were treated for accounting purposes. Prior to the implementation of FAS 123R, firms were allowed to expense stock options at their intrinsic value. Because nearly all firms granted stock options at-the-money, no expenses for option-based compensation were generally reported on the income statement.
Hall and Murphy (2003) argue that, due to their favorable accounting treatment and the fact that there is no cash outlay at the time of the grant, firms act as though the perceived cost of options is lower than their true economic cost. If firms make decisions based on the perceived costs instead of the economic costs, they grant more options than they would otherwise, and options with their favorable accounting treatment are preferred to possibly better incentive plans with less favorable accounting treatment. Consistent with this view, Carter, Lynch, and Tuna (2007) provide evidence that the accounting treatment of stock options affected their use, showing that a comprehensive proxy for financial reporting concerns was positively related to the use of stock options prior to FAS 123R. The implementation of FAS 123R eliminated the ability to expense options at their intrinsic value and instead required firms to begin expensing stock-based compensation at its fair value, effectively eliminating any accounting advantages associated with stock options.
This change in accounting policy represents an exogenous shock to the accounting benefits of using option-based compensation with no impact on the underlying economic benefits of stock options. If firms care about the perceived accounting costs of options, we expect them to reduce their use of options post 123R. Moreover, to the extent that convexity in the wealth-performance relation has a causal effect on managers’ risk-taking behavior, we expect that any significant changes in convexity associated with changes in the usage of stock options after FAS 123R will be correlated with changes in firm financial and investment policies related to risk taking and ultimately with changes in firm risk in a systematic manner.
Our findings do not generally support the view that providing incentives for risk taking is a primary rationale for the use of stock options. We show that firms dramatically decrease their usage of option-based compensation for the chief executive officer (CEO) following the implementation of FAS 123R. Based on our estimates, the value of stock options as a proportion of total compensation decreased by about 17 percentage points on average after FAS 123R became effective. More important, we find strong evidence that firms that would face higher accounting charges under FAS 123R reduced their reliance on stock options the most and that the reduction in option usage is generally unrelated to differences in firm characteristics that prior research has argued are associated with greater benefits from promoting risk-taking behavior—the market-to-book ratio, R&D expense, and new economy industry membership (Murphy, 2003; and Ittner, Lambert, and Larcker, 2003).
In response to the decline in option usage, we find that firms increase their reliance on bonuses, restricted stock, and long-term incentive awards (LTIAs). These forms of compensation are useful in increasing the pay-performance sensitivity of the manager’s compensation, but they do not generally share the convexity inherent in option-based compensation, suggesting that firms appear to be concerned with maintaining incentives to increase stock price but are less concerned with maintaining convexity in the compensation contract.
To explore this issue further, we directly measure how pay-performance sensitivity (delta) and convexity (vega) in compensation change around FAS 123R. Consistent with the results above, we find that convexity decreases significantly following the adoption of FAS 123R. In contrast, changes in pay-performance sensitivity are relatively modest around the adoption of FAS 123R as firms substitute away from stock options toward other forms of performance-based pay. This latter result is consistent with Hall and Murphy (2002), who argue that restricted stock represents a more efficient (less costly) mechanism compared with stock options for providing incentives to risk-averse managers to increase the firm’s stock price.
Finally, we explore whether the changes in convexity that are driven by the change in the accounting rules are accompanied by corresponding changes in investment and financing policies and firm risk that are consistent with the decline in incentives for risk taking. Our findings provide little evidence that the decline in convexity in compensation contracts after FAS 123R has been accompanied by a similar decline in risky financial and investment policies or firm risk.
Overall, our results are difficult to reconcile with the view that, at least on average, a primary motivation for the use of option-based compensation was to increase the convexity of the manager’s wealth-performance relation to provide incentives for risk taking. Instead, our findings suggest that accounting benefits were an important driver of the use of stock options prior to the implementation of FAS 123R. Interestingly, our results differ from the findings of two recent papers that also exploit exogenous changes in firm’s operating environments to examine the effects of option-based compensation on risk-taking behavior. Low (2009) shows that firms with low convexity decreased volatility following a court ruling that decreased the threat of takeovers. She also finds that firms appear to increase the convexity in managerial compensation contracts following the legal change. Gormley, Matsa, and Milbourn (2010) find evidence that managers whose compensation contracts have high sensitivity to stock price, low sensitivity to volatility, and options that are more in-the-money appear to reduce risk taking in response to an exogenous increase in downside risk.
In addition, although option usage declines significantly following the implementation of FAS 123R, stock options have not completely disappeared. While the conventional wisdom that convexity in compensation contracts provides incentives for risk taking appears to be too simplistic, it remains a challenge to understand the conditions under which convexity in compensation contracts affects managerial behavior and the role that options play relative to other forms of compensation as an efficient mechanism for paying managers.
The full paper is available for download here.