کارشناس مالیCEOها: سابقه کار CEO و سیاست های مالی شرکت

We study CEOs with a career background in finance. Firms with financial expert CEOs hold less cash, more debt, and engage in more share repurchases. Financial expert CEOs are more financially sophisticated: they are less likely to use one companywide discount rate instead of a project-specific one, they manage financial policies more actively, and their firm investments are less sensitive to cash flows. Financial expert CEOs are able to raise external funds even when credit conditions are tight, and they were more responsive to the dividend and capital gains tax cuts in 2003. Analyzing CEO-firm matching based on financial experience, we find that financial expert CEOs tend to be hired by more mature firms. Our results are consistent with employment histories of CEOs being relevant for corporate policies. However, we cannot formally rule out that our findings are partly explained by endogenous CEO-firm matching.

There is evidence that chief executive officers (CEOs) have particular management styles and that they matter for corporate performance (see Adams et al., 2005 and Bennedsen et al., 2008). Most evidence on CEO-specific heterogeneity examines education, personal characteristics, or personality traits. Malmendier and Tate (2005) study education, Kaplan, Klebanov, and Sorensen (2012) personal characteristics, and Malmendier and Tate, 2005 and Malmendier and Tate, 2008, Graham, Harvey, and Puri (2013), Hirshleifer, Low, and Teoh (2012), Malmendier, Tate, and Yan (2011), and Lin, Ma, Officer, and Zou (2011) analyze personal traits. Less is known about the work experience of CEOs. An exception is Custódio and Metzger (2013) who study CEOs׳ industry expertise in the context of diversifying mergers and acquisitions (M&A) activity. We contribute to the literature on CEO style in three ways. First, we show that previous financial expertise of CEOs is correlated with firms׳ financial policies. Second, we provide micro-level evidence on how financial experience operates. Third, we explicitly discuss the possibility that endogenous matching between CEOs and firms based on time-varying characteristics biases our results. Even though it is very difficult in general to establish a clean causal link between financial expertise and firm policies because exogenous variation in the financial expertise of the CEO is simply not observed, we provide a coherent and vast set of results that support the view that financial expertise matters. We also make use of both exogenous variations in the business conditions in which the CEO operates and CEO turnovers that are likely to be exogenous. We first show that an important fraction of CEOs are financial expert CEOs. That is, 41% of the CEOs have previous experience in the financial industry or in a financial role. When we look at their firm׳s financial policies, we find that nonfinancial firms headed by financial expert CEOs hold less cash on average. Furthermore, they are more leveraged and they have a higher propensity than other firms to pay out money to shareholders. These differences are economically meaningful. Regression results indicate that cash holdings are about 12% lower and average leverage ratios are 6% higher (evaluated at the mean). These firms are also 7% more likely to repurchase shares. Evaluating financial expert CEOs׳ actions and decisions at a micro-level, we provide evidence that they are more financially sophisticated. Typical firms use companywide discount rates to evaluate investment projects rather than a project-specific one. This has been called the weighted average cost of capital (WACC) fallacy (Graham and Harvey, 2001 and Krüger et al., 2011). Following Krüger, Landier, and Thesmar (2011), we use segment-level investment policy in diversified firms to find that only firms without a financial expert CEO fall into the WACC fallacy trap. We then exploit exogenous changes to business conditions to test whether financial experts react in a way that is consistent with higher levels of financial sophistication. We use the 2003 cuts to dividend and capital gains taxes, known as the “Bush tax cuts,” and find that financial experts were more responsive as measured by increases in total payout to shareholders. The tax cuts increased the dollar value of dividends and capital gains after 2003, therefore, increasing payout would be more beneficial to shareholders. We also show that financial expert CEOs are better able to raise external financing, even under tighter credit market conditions, suggesting that they have better access to capital markets. These results are consistent with the idea that financial experts can follow more aggressive financial policies (holding less cash and more debt) because they can access financial markets more easily. Guner, Malmendier, and Tate (2008) show in their research on board of directors׳ composition that financial experts have better access to capital markets. They show that firms with bankers on their boards use more external financing and have lower investment-cash flow sensitivity. However, the presence of bankers on a board is associated with policies that tend to benefit those financial institutions, but not necessarily the lending firms׳ shareholders. In fact, banks seem to provide funding to firms with good credit but poor investment opportunities.1 We also show lower investment-cash flow sensitivity for financial expert CEOs, but interestingly, the benefits of CEO financial expertise do not come at the cost of benefiting financial institutions. We find no evidence suggesting that financial experts overinvest just because they have better access to financing. This result seems plausible, as CEOs with financial expertise do not have a conflict of interest as bankers on the board do. Additional evidence that financial expertise matters is that newly hired financial expert CEOs are more likely to replace an incumbent chief financial officer (CFO) within the first year of their hire. We interpret this result as indication that financial expert CEOs are more involved with firm financial policies. It is likely to be easier to implement changes in a firm by changing a management team, in this case the CFO. Of course, we cannot rule out the alternative hypothesis that CFOs resign voluntarily more often if a financial expert CEO is hired. Financial expert CEOs might also directly benefit from their personal links to the financial sector. We test this hypothesis by examining whether former employers of the CEO are providing loans to his or her firm. There is no evidence that these direct links are driving our results. We use data on syndicated bank loans from Dealscan to show that only a very small fraction (2.1%) of loan facilities is provided by former employers of the CEOs. We also do not find evidence that financial expert CEOs are more likely to use indirect links beyond their previous employers, as they are no more likely to borrow from a new lending institution (a bank that was not providing loans to the company before the CEO was appointed). Overall, our findings also contribute to the literature that tries to understand cross-sectional variation in cash holdings, leverage, and payout policy. Early survey evidence by Graham and Harvey (2001) and more recently by Lins, Servaes, and Tufano (2010) suggest that the managerial impact on financial policies is likely to be big. They show that managers largely differ in their criterions and application of financial theories when it comes to capital structure. A common limitation in the literature on CEO characteristics is interpretation of the results. There are at least two ways that selection can bias the estimation of any potential effect of a CEO characteristic such as financial expertise on firm decision making. First, there might be omitted variables on the CEO level. For instance, financial expertise might be correlated with other CEO characteristics such as talent or education that drive our results. We address this concern in several ways. We control for education, general expertise, and talent using detailed biographical data. We also show direct evidence of financial sophistication by financial experts. This increases our confidence that our measure of financial expertise is not a proxy for something else. There is also a concern that endogenous CEO-firm matching could bias the results. There might be unobserved firm heterogeneity that simultaneously explains the matching between firms and financial expert CEOs as well as firm financial policies. For instance, financial expert CEOs might have preferences for certain firms that also happen to have low cash and high leverage ratios. Researchers are well aware of this problem and have addressed it by exploiting within-firm variation.2 In our analysis, we also control for firm fixed effects. The fixed effects approach, however, addresses endogenous matching based only on time-invariant characteristics. Although most authors do not explicitly address it, this concern is not new. It is recognized in the seminal work of Bertrand and Schoar (2003), which shows that CEO fixed effects are significant across several firm policies. Bertrand and Schoar acknowledge that their results are consistent with two alternative interpretations: (1) CEOs impose their idiosyncratic styles on companies, and (2) boards choose CEOs because of their attributes, in case firms׳ optimal strategies change over time. One contribution of our work is direct evidence on the matching between firms and CEOs based on both observable CEO and observable firm characteristics. 3 We show that financial expert CEOs tend to be appointed by mature firms, while non-experts tend to be appointed by firms in the growth stage of their life cycle. This is consistent with assortative CEO-firm matching based on the life cycle of the firm and the career background of the CEO. 4 Endogenous CEO-firm matching is not necessarily inconsistent with a causal interpretation of the documented effects of financial policies. Indeed, even though the life cycle of a firm might determine its optimal financial policies, the financial expertise of the CEO might still be crucial for implementing these strategies (“financial skills-based view”). That is, without the financial expertise of the CEO, the firm could not implement these optimal policies. Under an alternative hypothesis, firms are able to implement any financial policy, whatever the identity of the CEO. Thus, the effects observed originate purely from the preferences, or other unobserved characteristics of financial expert CEOs who self-select firms that implement specific financial policies. Under this “no financial skills view,” appointing a financial expert CEO could be the cheapest way to implement a given strategy, either because they have a preference to run a firm under those policies, or because for some other reason-other skills for instance-it is easier for the CEO to run the firm in such a way. For example, assuming that for some reason financial expert CEOs have a preference for running mature firms, companies in that stage of their life cycle might hire them because they might be willing to work at a lower salary. We directly test whether compensation for financial expert CEOs and non-experts is different in level or structure. We do not find any differences. Consistent with the notion that financial skills are crucial and valuable (in contrast to a “no financial skills” view), we also show that there is a positive valuation effect of financial expert CEOs in mature companies. We argue that our findings are consistent mostly with a financial skills-based interpretation of the estimated effects of financial expertise, even though we cannot formally reject the alternative view. As already noted, we provide direct evidence that financial experts have skills and are financially more sophisticated in application of finance theory. Moreover, exploiting exogenous variation in the business environment, we find that financial expert CEOs are able to access credit markets in unexpectedly tight conditions and react quickly to personal tax changes in a way that is beneficial to shareholders. We intend to control for endogenous matching based on the life cycle of the firms by directly including further firm-level controls and by using a matching estimator. We also analyze likely exogenous timing of a turnover. We look at a subsample of firms replacing a retiring CEO and the results hold in this subsample. Even though we are aware that the choice of the new CEO is endogenous, the timing of the turnover is likely exogenous, and this can help us to identify the effect of financial expertise when these firms switch from a nonfinancial expert CEO to a financial expert CEO. Finally, we analyze whether financial expertise of the CEO comes at the cost of other skills relevant to the firm. We focus on innovation, marketing, and labor policies. We find that firms headed by financial experts tend to invest less in research and development (R&D) and generate less innovation as measured by the number of filed patents and patent citations. When we analyze advertising expenses, the sensitivity of sales to advertising expenses, and labor costs, we do not find significant differences associated with financial expertise. The paper proceeds as follows. The next section describes the data. Section 3 analyses firms׳ financial policies by financial expert CEOs. Section 4 provides more micro-evidence on how financial expertise operates. Section 5 discusses alternative interpretations with emphasis on the matching between firms and CEOs. In Section 6 we analyze nonfinancial policies. Section 7 concludes.

Our analysis of the role of financial expert CEOs in nonfinancial firms indicates that firms headed by financial expert CEOs tend to hold less cash, more debt, and engage more in share repurchases. Analysis of actions and behavior of these financial expert CEOs at a micro-level yields evidence consistent with a conclusion that financial expert CEOs are more financially sophisticated and are more likely to follow academic theory. We provide evidence that financial experts are less likely to be trapped by the WACC fallacy (using a companywide discount rate instead of a project-specific one). They more dynamically manage their cash and leverage. They are able to increase both cash and debt levels in periods when it is difficult to do so, when default spreads are high, and credit conditions are tight. We find this practice to be advantageous for shareholders. Firms run by financial expert CEOs are less exposed to the riskiness of internal cash flows when they need to invest. They were also more responsive to the dividend and capital gains tax cuts in 2003 and paid out more to shareholders. We find no evidence that access to better personal networks in the financial sector drive these results. To implement changes in the company, financial expert CEOs also replace incumbent chief financial officers (CFOs) more often. We interpret this result as consistent with the idea that changing management helps to overcome inertia, which makes the implementation of new strategies either feasible or easier. These results have several implications. We show that previous work experience of the CEOs, namely, their financial expertise, affects firms׳ financial policies. Researchers to date have focused mostly on (partly unobservable) top manager characteristics or on age and education. We also explicitly discuss the concern that endogenous matching between CEOs and firms based on time-varying characteristics biases the results. We analyze the matching process itself, and we show some evidence of endogenous CEO-firm matching based on the financial expertise of the CEO and the life cycle of the firm. Results in a battery of additional tests provide evidence that is mostly consistent with a “financial skills view” interpretation of the effects of financial expertise. In addition, we show that financial expertise seems to come at the cost of other skills relevant to the firm, as suggested by the evidence that financial experts produce less innovation. Concerns about endogenous CEO-firm matching based on time-varying characteristics are likely to be relevant for other research on CEO characteristics as well, a subject for future research.