In oligopolies, firms behave strategically and commit to actions that elicit favorable responses from rivals. Firm actions consequently are a function of the nature of these strategic interactions. In this paper, we develop a methodology for the empirical estimation of strategic interactions in product markets. We then apply our measure of strategic interactions to CEO compensation. We use quarterly data on profits and sales from Compustat to estimate the slope of firm’s reaction function. When the slope is negative and marginal profits decrease with an increase in the rival’s actions the firm is classified as a strategic substitute. When the slope is positive and marginal profits increase with an increase in the rival’s actions the firm is classified as a strategic complement. As predicted by theory, we find significant evidence that strategic substitutes decrease the pay for performance incentives of their CEOs. On the other hand, strategic complements significantly increase CEO pay for performance incentives. The empirical measure developed can be used to test a wide variety of strategic models.

Several recent papers document that the magnitude of potential gains from stock-based compensation is positively related to the likelihood of misreporting. In a sample of firms that announce restatements of their financial statements from 1997 to 2002, we examine whether managers realize these potential gains occurring from their accounting choices. After controlling for diversification needs and stock price impact, we find no significant evidence of higher option exercises by executives in the misreported years. However, for firms that are more likely to have made deliberate aggressive accounting choices, we find significant evidence of higher option exercises. For these firms, option exercises are higher by 20–60% in comparison to industry and size matched nonrestating firms. Options exercises by executives are also increasing in the magnitude of the restatement as captured by the effect of the restatement on net income. These higher option exercises tend to be more pervasive and are not just confined to the CEO and CFO of the firm.