Do executive stock options induce excessive risk taking?

Abstract

We examine whether executive stock options can induce excessive risk taking by managers in firms’ security issue decisions. We find that CEOs whose wealth is more sensitive to stock return volatility due to their option holdings are more likely to choose debt over equity as a capital-raising vehicle. More importantly, the pattern holds not only in firms that are underlevered relative to their optimal capital structure but also in overlevered firms. This evidence is inconsistent with executive stock options aligning the interests of managers and shareholders; rather, it supports the hypothesis that stock options sometimes make managers take on too much risk and in the process pursue suboptimal capital structure policies.

Introduction

The 1990s witnessed an explosion in the use of restricted stock and stock options in executive pay packages (Murphy, 1999), and equity-based pay has represented a significant proportion of executive compensation ever since. One of the widely accepted economic justifications for this phenomenon is that equity ownership can potentially align the interests between managers and shareholders and thus mitigate the agency problems due to the separation of ownership and control characterizing large public corporations (Berle and Means, 1933; Jensen and Meckling, 1976).

However, greater equity ownership does not always lead to better managerial decisions and higher shareholder value. Large shareholdings may enable managers to effectively block control contests, become entrenched, and extract private benefits at the expense of shareholders (Morck et al., 1988, Stulz, 1988; McConnell and Servaes, 1990). Managers with significant stock and stock option ownership may also develop symptoms of managerial myopia and display certain perverse behaviors that are detrimental to shareholders. For example, the popular press and regulators hold stock options at least partially accountable for some of the high-profile corporate scandals early this decade. Academic studies emerge with evidence that stock and especially stock option holdings induce managers to manipulate corporate earnings and commit accounting fraud to boost market valuations and enrich themselves by selling shares or exercising options at inflated stock prices (Cheng and Warfield, 2005, Bergstresser and Philippon, 2006, Burns and Kedia, 2006, Burns and Kedia, 2008, Peng and Roell, 2008). Recent research has also uncovered rather widespread backdating of executive stock options by companies (Lie, 2005, Heron and Lie, 2007, Narayanan and Seyhun, 2008). This practice benefits executives receiving options and when exposed, causes tremendous damage to shareholder wealth (Narayanan et al., 2007, Bernile and Jarrell, 2009).¹

In this paper, we examine another potential concern about the incentives provided by stock and stock options, namely, whether they induce risk taking by managers beyond the level desirable for shareholders. It is well recognized that unlike diversified shareholders, managers are risk averse because of their organization-specific human capital and undiversified wealth portfolios (Amihud and Lev, 1981). As a result, they may forego risky but positive net present value (NPV) projects. To address this problem, firms resort to stock options and their convex payoff structure to encourage managers to take more risk.² Many studies provide evidence that managers do appear to respond to such incentives (Guay, 1999, Cohen et al., 2000, Knopf et al., 2002, Rajgopal and Shevlin, 2002, Chen et al., 2006, Coles et al., 2006, Brockman et al., forthcoming). However, what is left unaddressed is whether the induced risk taking by managers is beyond the optimal level. Lambert, 1986, Ju et al., 2003, and Raviv and Landskroner (2009) all show theoretically that under call option like contracts managers can take too much risk. This seems consistent with popular claims that companies where managers receive significant stock option compensation tend to borrow too much, a problem that hastened the demise of Enron in 2001,³ and that the compensation system at financial institutions was a contributing factor to the financial crisis of 2007–2009 by encouraging excessive risk taking and overleveraging.⁴

Despite the immediate relevance of the issue, systematic evidence on whether managerial equity incentives lead to undue risk taking is scarce, mainly because it is difficult to establish how much risk is too much. To overcome this obstacle, we examine the effect of equity-based compensation on managerial risk taking in the context of firms’ choices between debt and equity in security issue decisions. Corporate capital structure policies provide an ideal setting for our investigation since there is a well developed literature on optimal or target capital structure,⁵ which can help us identify what managerial actions represent risk taking beyond the optimal level.

Our analysis of 3734 security offerings by US public companies from 1993 to 2007 shows that managers are more likely to use debt than equity as a capital-raising vehicle when their wealth is more sensitive to stock return volatility, i.e., when they have a higher vega. This result is intuitive, since compared to equity financing, debt financing increases firm leverage and stock return volatility, and thus managers with a higher vega stand to gain more from a debt issue. We also find that managerial wealth sensitivity to stock price, i.e., delta, does not affect corporate financing decisions. Together, these two findings suggest that it is the stock option holdings instead of stock ownership that play a role in shaping managerial preference between debt and equity, since vega comes entirely from stock options while both stock and stock options contribute to delta.⁶

To see whether the preference of high-vega CEOs for debt reflects excessive risk taking, we follow the approach of Hovakimian et al. (2001) and separate firms into those that appear underlevered relative to their optimal leverage ratios and those that appear overlevered. We repeat our analysis of security issue decisions in the two subsamples, and find that CEOs with a higher vega are more likely to eschew equity and choose debt in both subsamples. While the evidence from the underlevered subsample is consistent with the argument in Berger et al. (1997) that stock option ownership aligns the interests of managers and shareholders and better-aligned managers increase leverage toward an optimal level, our finding of high-vega CEOs favoring debt financing even in overlevered firms is difficult for the incentive alignment-based hypothesis to explain. Instead, it supports the hypothesis that managers choose to issue debt rather than equity to maximize the value of their option portfolio even if doing so causes firms to deviate further from their optimal leverage and is not in the best interests of shareholders.

In light of the finding by Lemmon et al. (2008) that there is an unidentified time-invariant determinant of firms’ capital structure, we augment our optimal-leverage model by including firm fixed effects. We classify security issuers in our sample into overlevered and underlevered based on the prediction of the modified leverage model, and continue to observe high-vega CEOs preferring debt over equity in the overlevered subsample.

We also extend our analysis to examine how the overall change in a firm’s leverage is related to CEO vega and uncover evidence consistent with our security issue decision analysis. More specifically, CEOs with a higher vega are more aggressive in increasing leverage, even at firms that appear already overlevered compared to their optimal debt ratios.

Our study makes three contributions to the literature. First, we identify excessive risk taking as another unintended adverse consequence of giving large stock option grants to top executives, in addition to creating incentives to manage earnings, commit financial fraud, and manipulate the timing of grants. More specifically, we find that CEOs with a high vega from their stock option holdings sometimes pursue suboptimal capital structure policies by overleveraging their companies in order to increase the value of their stock option portfolios. Our evidence suggests that there may be merits in the call for and the practice of companies replacing executive stock options with restricted stock.⁷

Second, we add a new dimension to the agency theory proposed by Jung et al. (1996), who show that managerial discretion and private benefits play an important role in firms’ financing decisions. In our analysis, the potential value appreciation of executive stock options due to higher risk can be considered as a private benefit, which induces managers to take on excessive risk by shunning equity and choosing debt despite the fact that their firms are overlevered. Ironically, stock options, originally put in place by boards to align the interest of managers and shareholders, are the source of the agency conflict here.

Third, we extend the literature on the effect of stock options on managerial risk taking. Previous studies find that managers respond to stock option compensation by adopting riskier financial, investment, and hedging policies. We advance these findings one step further by showing that stock options sometimes drive managers to engage in excessive risk taking.

The remainder of the paper is organized as follows. Section 2 describes the sample and variable construction for the probit analysis of security issue decisions. Section 3 presents the results from the probit analysis in both the full sample and the overlevered and underlevered subsamples. Section 4 presents the results from (i) augmenting the optimal-leverage model by firm fixed effects, (ii) including preferred equity and convertible debt offerings as additional financing choices, (iii) examining the relation between overall leverage changes and CEO vega, and (iv) a variety of robustness checks. Section 5 concludes.