No abstract

We examine the influence of CEO equity-based compensation on strategic risk taking by the firm. Building off the Behavioral Agency Model, Agency Theory, and Prospect Theory, we develop arguments about when equity-based compensation elements will increase or decrease executive risk propensity and, in turn, strategic risk taking. Incorporating a behavioral perspective into our models of incentive alignment provides us with new and potentially more accurate predictions about how individual elements of CEO pay will influence risk selection, as well as how equity compensation interacts with cash compensation and with other factors to influence risk preferences. In general, this study provides evidence that CEO equity-based compensation significantly influences strategic risk, but that this influence is more nuanced and complex than conventional treatments of executive compensation assume. In particular, we find that different forms of equity-based pay exhibit dissimilar influences on strategic risk and that their influence changes as their value and vesting status change. Second, we find that cash-based forms of pay moderate the incentive properties of equity-based pay, indicating that cash-based pay may affect how executives perceive risks associated with equity pay. Finally, we find that stock price volatility and board actions each also moderate the incentive effects of equity-based pay. In sum, our results argue for increased recognition of a behavioral perspective on executive compensation and greater precision in how we measure and model the incentive alignment properties of CEO compensation.

Scholars who study organizational decline have argued that declining organizations reduce or eliminate their riskier activities such as innovation. Further, they cite reduced risk-taking as a primary contributor to further decline. Scholars with an interest in risk per se come to the opposite conclusion: low performing firms often take more risks than other firms and such risks reduce subsequent performance. This study attempts to resolve these conflicting views by examining the risk of firms in decline. Our model, based on Cyert & March's (Cyert, R. M., J. G. March. 1963. A behavioral theory of the firm. Prentice-Hall, Englewood Cliffs, NJ.) behavioral theory of the firm, includes six basic variables: (1) performance, (2) slack, (3) aspirations, (4) expectations, (5) risk (income stream uncertainty), and (6) organization size as a measure of decline. The estimated model includes prior levels of risk and performance in the risk and performance equations respectively as controls. This study uses data on 344 low-performing firms in 19 manufacturing industries to estimate a time-series model that addresses both decline's influence on risk and risk's influence on performance while controlling for firm slack resources and industry factors. The results suggest (1) organizational decline and potential slack (debt/equity) positively influence risk whereas recoverable slack (SGA/Sales), and the difference between performance aspirations and expectations negatively influence risk; and (2) recoverable slack and risk negatively influence performance. We also examined the influences of performance and slack on organizational decline and found that performance associates negatively whereas recoverable and potential slack associate positively with decline. These results challenge the argument that declining firms reduce risk and so hurt their subsequent performance. Instead, the results suggest a cyclical process with positive feedback in which decline and the loss of certain slack resources increases risk which in turn reduces performance and results in further organizational shrinkage. Thus firms facing decline fall into a trap of taking unprofitable risks that ultimately exacerbates the decline. In addition, results relating slack to risk and performance exhibited inconsistent relations across the measures raising questions about the role slack resources play in risk and decline. Previous discussions of this slack implicitly assume all forms of slack have similar influences but our results demonstrate different forms of slack have very different effects.

No abstract

No abstract