The Effect of CEO Inside Debt Compensation on Risk-taking Behavior

(1)

Master Thesis MSc BA Organizational & Management Control

The Effect of CEO Inside Debt Compensation on

Risk-taking Behavior

by

Yanran Fan

University of Groningen

Faculty of Economics and Business

June 24, 2013 Jozef Israelsstraat 21 9718GB Groningen (06)19067658 y.fan@student.rug.nl s2096927 Supervisor: dr. Bo Qin

Co-assessor: drs. Abdul Rehman Abbasi

(2)

1

ABSTRACT

This master thesis primarily examines the influence of Chief Executive Officer (CEO) inside debt compensation on CEO’s risk-taking behavior. Besides, this study simultaneously compares the impacts of CEO inside debt, CEO equity-linked and CEO bonuses compensation on risk-taking related investment and financial behavior. In general, CEO inside debt compensation is considered as unsecured and unfunded liabilities of the firm, which makes CEOs’ attitudes to default risk similar to outside creditors (Cassell et al., 2012). Jensen and Meckling (1976) state that CEOs with large inside debt holdings will display lower levels of risk-seeking behavior. The conflicts between shareholders and outside creditors align with different CEO compensation components on risk-taking behavior stimulate the development of the theories and the empirical research in this study. My main findings indicate that CEO inside debt compensation has a significantly negative influence on R&D investment policy and a significantly positive impact on firm working capital and leverage with regard to financial policy. With respect to CEO equity-linked compensation, I find it has a positive and significant influence on firm diversification investment policy. Meanwhile, CEO equity-linked compensation has a significantly negative influence on working capital. Moreover, CEO bonuses compensation also has a positive and significant effect on firm diversification investment policy. My results provide empirical evidence showing the CEO with larger inside debt compensation tend to make investment policy on R&D and financial policy on working capital less risky. However, with regard to financial policy – firm leverage, the empirical results are complicated than conservative thoughts which need to be further investigated.

Keywords:

Inside debt compensation; Pensions; CEO incentives; Bonuses; Risk-taking behavior

Acknowledgement:

(3)

2

INTRODUCTION

The compensation contract of CEOs is structured to align the interests of top management with the owners of the firm (Bebchuk & Jolls, 1999). Prior researches emphasize the role of equity-linked compensation in encouraging CEOs to maximize shareholders value by making risky decisions and implementing risky policies. However, in the context of recent financial crisis, Cassell et al. (2012) state that the financial crisis highlighted the vulnerability of certain components of firm-specific executive wealth during times of financial distress as several prominent CEOs surrendered significant portions of their inside debt compensation (pension benefits and deferred compensation) when their firms failed during the crisis. Take two recent incidents as examples, for Rick Wagner, who was the former CEO of GM, the supplemental pension benefits at GM are not pre-funded and are paid out of the corporation’s general assets. Therefore, there is nothing to back up these obligations at GM and the benefits will be wiped out. Since Rick Wagoner was let go in March 2009, it appears that he will lose virtually off of this pension benefits (Chasan, 2009). Another examples also occurred in 2009, Lee Iacocca, who was the former Chrysler car chief executive, was about to lose a huge amount of his pension due to the U.S. automaker’s bankruptcy (Sterne, 2009). In general, inside debt compensation represent unsecured and unfunded liabilities of the firm, rendering these executive holdings sensitive to default risk similar to that faced by other outside creditors (Sundaram & Yermack, 2007; Edmans & Liu, 2011). Thus, this research thesis primarily intends to investigate the effects of inside debt compensation on CEOs’ risk-taking behavior.

After the CEO is hired, he or she contributes to making the firm’s investment and financial decisions. Risk aversion may cause the CEO to abandon some risky but positive net-present-value projects. The lower the risk aversion the CEO has, the more likely he or she is to invest in the risky project (Niu, 2010). Jensen and Meckling (1976) indicate that agency problems arise due to the separation of ownership and control characterizing large public corporations. More specifically, there are differences in the incentive structures of principals (shareholders and debt holders) and agents (CEOs).

(4)

3

does not have adequate incentives to expend optimal effort (Sundaram & Yermack, 2007). More specifically, agency costs of equity occur because CEO wealth is generally less diversified than that of shareholders, CEOs tend to have lower risk-appetite than what shareholders would prefer. A rational and risk-averse CEO may correctly estimates the residual risk of the risky projects, but may not invest in it due to risk aversion. To increase CEOs’ risk-taking, firms offer him or her compensation contract to align with shareholders value (Niu, 2010).

On the other hand, Sundaram and Yermack (2007) also note that outside debt creates risk-shifting problems: CEO, as a holder of a convex residual claim on the firm, has an incentive to suboptimally increase the riskiness of the firm’s cash flows. However, agency conflicts between CEOs and debt holders arise when CEOs increase firm risk (e.g., through firm investment and financial policies) in ways that benefit shareholders at the expense of debt holders (Jensen & Meckling, 1976; Dewatripont & Tirole, 1994). Debt holders prefer firms to be more conservative because debt holdings are characterized by an asymmetric payoff function with respect to the firm’s net assets (Watts, 2003). While payoffs are fixed when firm performance is good, debt holders face substantial risk if firm performance is poor (Cassell et al., 2012).

The value of CEO’s inside debt compensation depends on the ability of the company to make future payments to CEOs. Jensen and Meckling (1976) argue that pension plans have the potential of mitigating the risk-shifting problem of managers since CEOs who own inside debt are concerned not only about shareholders’ interests, but also about firm default risk. A research by Edmans and Liu (2011) indicates that inside debt compensation is probably an efficient remedy for the asset substitution problem.

(5)

4

Inside debt compensation are prevalent and often substantial (Sundaram & Yermack, 2007). Wei and Yermack (2011) find that 84% of CEOs in their sample hold some form of inside debt and that average inside debt compensation are approximately $10 million for sample CEOs. They note that to the extent that managers have large unfunded deferred compensation claims against their firms, outside investors expect them to manage their firms conservatively, implying low-risk investment strategies that would tend to make debt safer and equity less attractive. Nevertheless, there is limited empirical evidence to support where there is a relationship between CEO inside debt compensation and investment and financial risk-taking behavior. Cassell et al. (2012) examine the association between the CEO inside debt holdings and a risk-taking policy choice, which is that CEO with large inside debt holdings prefer investment and financial policies that are less risky. However, there is still limited empirical evidence to support their findings. Due to the data availability, Cassell et al. (2012) test the CEO inside debt holdings instead of CEO inside debt new grants annually. In order to fill this gap, this master thesis therefore intends to use UK company data to investigate whether CEO inside debt new grants affect current investment and financial policies. Furthermore, I attempt to simultaneously compare the different CEO compensation components (CEO inside debt compensation, CEO equity-linked compensation and CEO bonuses).

I expect that the empirical results will stimulate researchers’ interest who investigate the influences of CEO compensation components, and contribute to both shareholders and debt holders who wish to evaluate the extent to which CEO preferences are aligned with their own, and regulators who seek to the impact that CEO compensation packages can have on CEO risk-taking behavior.

(6)

5

RELATED LITERATURE AND HYPOTHESIS DEVELOPMENT

Related Literature

The reason of agency theory for existence lies in the separation of ownership and control of the public corporation and the problems might bring about. Apparently it is not always possible to have the owner as the manager of the company because firms grow beyond the financial means of a single owner (Davis et al., 1997). The main theoretical principle behind agency theory is that the firm is composed of a nexus of contracts. As a consequence, agency theory is applicable to all contractual relationships in the firm (Grabke-Rundell & Gomez-Mejia, 2002). However, it focuses mainly on top management since they are at the strategic apex of the firm because they are responsible for resources allocation decisions, new market entries, acquisitions and divestitures, etc. (Sanders & Carpenter, 1998).

The conflicts of interest between shareholders and debt holders arise from the different pay-off structures that the two parties have. Debt holders have a fixed claim on the firm’s cash flows. However, shareholders have a residual claim on the firm’s cash flows, while enjoying limited liability. Hence, debt holders are interested in having the firm operated with a low risk of default, while shareholders prefer to maximize their value – sometimes at the expense of debt holders (Belkhir & Boubaker, 2012). Galai and Masulis (1976) argue that limited liability shareholders can extract wealth from debt holders by selecting riskier projects. Myers (1977) indicates that the presence of debt in a firm’s capital structure is commonly called the agency costs of debt. Edmans and Liu (2011) interpret the terms “asset substitution” and “risk-shifting” as any action that benefits shareholders but reduces total firm value.

(7)

6

provide empirical evidence that compensation contracts are designed not only to align managers’ interests with those of shareholders, but also to mitigate the conflict of interest between shareholders and debt holders.

Jensen and Meckling (1976) refer to debt-like compensation components as “inside-debt”. In general, executive pension plans are defined benefit plans and pay an employee a fixed amount per year after retirement. Deferred compensation refers to defined contribution plans in which specific contributions are made to a retirement plan in the form of either deferred employee compensation or employer contributions. Cassell et al. (2012) adopt the terminology of Jensen and Meckling (1976) and refer to the sum of cumulative defined pension benefits (including supplemental executive retirement plans) and deferred compensation as total inside debt.

Edmans and Liu (2011) consider a set of standard securities in their research: debt, equity and a fixed bonus that pays off only in solvency, and initially assume that the manager holds an exogenous equity stake to create risk-shifting incentives. They show that inside debt is a superior remedy to the agency costs of debt than the bonuses. They state that bonuses are effective in encouraging the manager to avoid bankruptcy, because they are only received in solvency. However, creditors are concerned with not only the probability of default, but also recovery values in default. Thus, optimal contracts should depend on the value of assets in bankruptcy, as well as the occurrence of bankruptcy. This implies the critical difference between inside debt and bonuses: inside debt yields a positive payoff in bankruptcy, proportional to the liquidation value. Therefore, it renders the manager sensitive to the firm’s value in bankruptcy. In contrast, bonuses have zero bankruptcy payoffs, regardless of the liquidation value, and represent binary options rather than debt.

(8)

7

Hill & Hansen, 1989; Hill et al., 1988, 1992; Morck et al., 1988). This “risk differential” (Beatty & Zajac, 1994) between agents and principals creates a “moral hazard” problem in the principal-agent relationship.

There is a stream of literatures which study the influence of managerial compensation on the alignment between managers’ and equity holders’ interests. These studies mainly emphasize the role of equity-based compensation to incentive managers to take actions that maximize shareholder value (McConnel & Servaes, 1990; Murphy, 1985). However, there is limited literatures which investigate the association between managers’ and debt holder’s interests.

Sundaram and Yermark (2007) show the evidence for the role of inside debt in mitigating the agency problem. They conduct the pension value of 237 large firms during the period of 1996-2002 as a proxy of inside debt to investigate the determinants of inside debt and its relation with CEO turnover and firm default risk. The results of their research show that pension values are higher when firm leverage is higher. Besides, CEOs prefer to take conservative investment policies when their personal debt-to-equity ratio is higher than the firm leverage ratio.

However, Sundaram and Yermark only use the pension obligations to the CEO as the proxy in their research; they could not apply another significant component of inside debt – deferred compensation plans. According to the new disclosure rule of SEC in 2006, firms are required to disclose accumulated actuarial present value of each executive officer’s pension plan, as well as the contributions, earnings and balances of each executive officer’s nonqualified deferred compensation account after fiscal year 2006. Cen (2013) expands Sundaram and Yermack’s work by examining both pension obligations and deferred compensation plans. Cen documents the new disclosed information of CEO’s inside debt and examines its determinants and implications to agency theory. According to the results of Cen’s research, deferred compensation is in the same order of magnitude as pension. The average deferred compensation represents about 6.2% of a CEO’s total compensation, whereas pension contribution represents about 5.5% of overall CEO compensation.

(9)

8

is more complicated than prior studies have indicated. To be more specific, Wei and Yermack examine the influence of firm’s first required disclosures of their CEO’s deferred compensation and other inside debt holdings in early 2007. Their results indicate a reduction in firm risk, a transfer of value from equity toward debt, and an overall destruction of firm value when CEOs’ inside debt holdings are large. However, Cen does not observe the linear positive association between firm leverage and CEO pension. Instead, Cen finds an inverted U-shape relation: inside debt initially increases with the firm leverage, but when firm leverage reaches a certain level, CEO inside debt holdings are negatively associated with the firm leverage. This U-shape association is also inconsistent with the argumentation of Sundaram and Yermark.

Hypothesis Development

(10)

9

investment in high-risk securities. Benett et al. (2012) show evidence that banks where CEOs whose incentives are more aligned with the debt holders face less default risk.

With respect to CEOs’ risk-taking behavior, it can be measured as investment policy and financial policy. In terms of investment policy, R&D expenditures have been the focus of research investigating the association between executive compensation incentive effects and firm investment choice. Compared to other investment vehicles, R&D expenditures tend to be more risky given the high degree of uncertainty relating to their future economic benefits (Coles et al., 2006). According to the empirical finds of Cassell et al. (2012), there is a negative association between CEO inside debt holdings and R&D expenditures. Besides, CEOs can reduce the riskiness of their firms’ operations by increasing the diversity of their firms’ operations among different industry segments (Coles et al., 2006). Managerial risk aversion as a motive for diversification is noted by Amihud and Lev (1981), May (1995), and Tufano (1996). Cassell et al. state that an undiversified firm may collapse if the market for the firm’s sole product or service deteriorates. On the contrary, a diversified firm could absorb the impact of such an event if the health of the firm’s other business segments is not affected. Hence, a diversified firm faces a reduced exposure to bankruptcy. In their findings, they show that there is a positive association between CEO inside debt holdings and the diversification of firm operations.

CEOs’ risk-taking behavior can be also measured as financial policy. Inside debt compensation is generally unsecured and unfunded, features which put the value of these holdings at risk in the event of bankruptcy. CEOs with large inside debt holdings can decrease their risk exposure by holding more liquid assets (Ohlson, 1980) and decreasing the debt leverage of the firm. Cassell et al. (2012) measure asset liquidity as working capital. According to their empirical research, there is a positive relationship between CEO inside debt compensation and working capital.

(11)

10

a negative association between CEO inside debt holdings and financial leverage. They also note that since the firm’s leverage appears in the denominator of relative CEO debt-to-equity ratio, any documented association between CEO inside debt holdings and financial leverage could be driven (in part) by a mechanical relationship. Furthermore, unlike the work of Sundaram and Yermack, Cen (2013) does not observe the linear positive association between firm leverage and CEO pension. Instead, Cen find an invereted U-shape relation: CEOs in the middle-leverage firms have higher inside debt holdings than both in low-leverage firms and high-leverage firms. Besides, Cen suggest that the underlying reason is related to firm financial distress and CEO risk aversion. Therefore, the influence of CEO inside debt compensation on firm financial leverage is unpredictable.

According to the empirical findings of Cassell et al. (2012), they only test the association between CEO inside debt holdings and risk-taking policies. Nevertheless, they do not investigate the differences between the effect of CEO inside debt compensation, CEO equity-linked compensation, and bonuses on investment and financial policies. Besides, due to the data availability, prior studies use CEO inside debt holdings instead of inside debt new annual reward. Therefore, based on the data differences, empirical results and theory gaps above, this paper primarily intends to test the following hypotheses:

H1a: Larger executive inside debt compensation has a negative influence on R&D expenditures. H1b: Larger executive inside debt compensation has a positive influence on firm diversification. H1c: Larger executive inside debt compensation has a positive influence on working capital. H1d: Larger executive inside debt compensation has a positive (or negative) influence on financial leverage.

(12)

11

value (Rost & Osterloh, 2008). In agency-based models incentive alignment as a control mechanism is achieved by making some portion of agent compensation contingent upon satisfying performance targets specified in the contract (Welbourne et al., 1995). However, prior research has examined the relation between CEO equity-linked incentives and firm policies, such as leverage, but with mixed results on how equity-linked compensation affects managerial risk-taking behavior (Berger et al., 1997; Geczy et al., 1997; Rejgopal & Shevlin, 2002; Tufano, 1996). Thus, according to the theoretical predictions and empirical results, this study attempts to examine the following hypotheses:

H2a: Larger executive equity-linked incentives have positive influences on R&D expenditures. H2b: Larger executive equity-linked incentives have positive (or negative) influences on firm diversification.

H2c: Larger executive equity-linked incentives have negative influences on working capital. H2d: Larger executive equity-linked incentives have positive (or negative) influences on financial leverage.

(13)

12

H3a: Larger executive bonuses compensation has a positive (or negative) influence on R&D expenditures.

H3b: Larger executive bonuses compensation has a positive (or negative) influence on firm diversification.

H3c: Larger executive bonuses compensation has a positive (or negative) influence on working capital.

H3d: Larger executive bonuses compensation has a positive (or negative) influence on financial leverage.

RESEARCH DESIGN

This study conducted with a sample of CEOs of UK companies between 2002 and 2009. Over this period, it can be seen that the pension new annually grants and the rapid growth in the use of equity-linked executive compensation, as well as CEO bonuses that are in the use of CEO compensation package. To be included in the sample, a firm has to have available data of investment and financial policies. According to the previous literatures, this paper eliminates finance firms and utilities. As a consequence, the financial firms are excluded of the data sample. The final data sample consists of 257 firms, with a total of 1840 firm-years. In other words, the sample consists of 1840 observations of UK companies. This study executes all control variables at the 1% level in both tails and for each year.

To test the hypotheses above, I follow the previous study and develop models to examine the influence of CEO debt to total ratio and CEO equity to total ratio on riskiness of investment and financial policies. Hence, I estimate the following regressions:

R&D Exp i,t=β0+β1 CEO debt/total ratio i,t+Controls i,t+ε i,t (1a)

R&D Exp i,t+1=β0+β1 CEO debt/total ratio i,t+Controls i,t+ε i,t (1b)

Divers i,t=β0+β1 CEO debt/total ratio i,t+Controls i,t+ε i,t (2a)

(14)

13

Working Capital i,t=β0+β1 CEO debt/total ratio i,t+Controls i,t+ε i,t (3a)

Working Capital i,t+1=β0+β1 CEO debt/total ratio i,t+Controls i,t+ε i,t (3b)

Leverage i,t=β0+β1 CEO debt/total ratio i,t+Controls i,t+ε i,t (4a)

Leverage i,t+1=β0+β1 CEO debt/total ratio i,t+Controls i,t+ε i,t (4b)

R&D Exp i,t=β0+β1 CEO equity/total ratio i,t+β2 CEO Delta i,t+Controls i,t+ε i,t (5a)

R&D Exp i,t+1=β0+β1 CEO equity/total ratio i,t+β2 CEO Delta i,t+Controls i,t+ε i,t (5b)

Divers i,t=β0+β1 CEO equity/total ratio i,t+β2 CEO Delta i,t+Controls i,t+ε i,t (6a)

Divers i,t+1=β0+β1 CEO equity/total ratio i,t+β2 CEO Delta i,t+Controls i,t+ε i,t (6b)

Working Capital i,t=β0+β1 CEO equity/total ratio i,t +β2 CEO Delta i,t+Controls i,t+ε i,t (7a)

Working Capital i,t+1=β0+β1 CEO equity/total ratio i,t+β2 CEO Delta i,t+Controls i,t+ε i,t (7b)

Leverage i,t=β0+β1 CEO equity/total ratio i,t+β2 CEO Delta i,t+Controls i,t+ε i,t (8a)

Leverage i,t+1=β0+β1 CEO equity/total ratio i,t+β2 CEO Delta i,t+Controls i,t+ε i,t (8b)

This study also intends to investigate the impact of CEO bonuses to total compensation ratio on R&D expenditures, firm diversification, working capital and financial leverage. In order to do so, I estimate the following regressions:

R&D Exp i,t=β0+β1 CEO bonus/total ratio i,t+Controls i,t+ε i,t (9a)

R&D Exp i,t+1=β0+β1 CEO bonus/total ratio i,t+Controls i,t+ε i,t (9b)

Divers i,t=β0+β1 CEO bonus/total ratio i,t+Controls i,t+ε i,t (10a)

(15)

14

Working Capital i,t=β0+β1 CEO bonus/total ratio i,t+Controls i,t+ε i,t (11a)

Working Capital i,t+1=β0+β1 CEO bonus/total ratio i,t+Controls i,t+ε i,t (11b)

Leverage i,t=β0+β1 CEO bonus/total ratio i,t+Controls i,t+ε i,t (12a)

Leverage i,t+1=β0+β1 CEO bonus/total ratio i,t+Controls i,t+ε i,t (12b)

where ε i,t is the error term. All the variables are measured at the end of fiscal year t and year t+1.

Controls i,t means control variables as determinants of the policy measures and incentives based on prior literature. An important reason to include control variables is to represent forces that drive CEO inside debt compensation and other compensation components together with investment and financial policy. Table 1 shows the definition of variables used in data analysis.

TABLE 1 Variable definition Variables Definition

Dependent variables

R&D Expenditures The ratio of R&D expenditures to total sales

Diversification Acquisition refers to the purchase of one business or company

Working Capital Current assets minus current liabilities scaled by total assets

Debt leverage ratio The ratio of total long term debt to total assets

Independent variables

CEO debt/total ratio The ratio of CEO inside debt compensation to total compensation

CEO equity/total ratio The ratio of CEO equity-linked compensation to total compensation

CEO bonus/total ratio The ratio of CEO bonus to total compensation

Control variables

Wealth Delta The change in a CEO’s wealth in the company for each 1% change in the stock price Total LTIPs Sum of all cash, equity, equity matched and option plans awarded or held

Direct Compensation Sum of all cash based compensation (bonuses and salary)

CEO Age (Yrs) The age of the CEO in years

(16)

15

Variable Description

Dependent variables:

According to previous literatures, managers can change firm risk through R&D expenditures, firm focus, capital expenditures, and leverage (Coles et al., 2006). Hence, I measure CEO’s risk-taking behavior with regard to both investment and financial policies. The first proxy of investment policy is R&D expenditures which measured as the ratio of research and development expenditures to total sales at the end of fiscal year t and t+1. I measure the firm’s diversification using acquisitions. Companies often implement corporate-level acquisition strategies to achieve product diversification that can build core competencies (Open Learning World, 2011). In fact, acquisition strategy is the most common means of implementing diversification. This variable also measures over two windows: t and t+1.

With respect to financial policies, I use working capital as the first proxy calculated by current assets minus current liabilities scaled by total assets measured at the end of fiscal year t and t+1, respectively. The second measurement of financial policy is firm leverage. Financial policies determine the probability distribution of cash flows and stock returns of the firm. CEOs can increase firm risk by altering financial policy, in this case, by increasing leverage (Coles et al., 2006). I calculate firm leverage by dividing total long term debt by total assets. The greater the

In role (Yrs) CEO in role in years

On board (Yrs) CEO on board in years

In organization (Yrs) CEO in organization in years

Total assets Total value of assets of the company

MtB The market value of assets divided by the book value

ROA The rate of return on assets

Sales growth The ratio of total sales in year t to total sales in year t-1

Instrumental variables

New CEO An indicator variable set equal to one if the firm has a new CEO (within one year), and zero otherwise

(17)

16

ratio is, the higher the share of long term debt within the balance sheet, and the riskier for the operation of the firm. This variable also measured at the end of fiscal year t and t+1.

Independent variables:

Since the regression models intend to examine the effects of CEO inside debt compensation (CEO pension annual reward), CEO equity-linked compensation (annual reward) and CEO bonuses (annual reward) simultaneously, I use CEO total compensation (annual reward) as denominator. I calculate CEO inside debt to total compensation ratio, CEO equity-linked to total compensation ratio, and CEO bonuses to total compensation ratio by dividing these three components by total compensation.

Other control variables:

The control variables which are used as determinants of the investment and financial policy measures are all based on prior literature. An important reason to include control variables is to represent forces that drive CEO compensation components together with investment and financial policy.

According to prior work, stock option holdings and percentage stock ownership are noisy measures of managerial incentives because they do not explicitly measure the relation between CEO portfolio wealth and stock returns and risk (Low, 2009). However, Wealth Delta is estimated as the change in a CEO’s wealth in the company for each 1% change in the stock price (Core & Guay, 2002). They indicate that Wealth Delta gives the sensitivity of director’s wealth in the company (including stock and stock option holdings) to stock price. I measure Wealth Delta by using the “one-year approximation” method outlined in Core and Guay (2002). I take natural logarithmic transformation of this variable. Previous research has demonstrated a strong link between this measurement and the risk-taking policies of the firm (Coles et al., 2006; Guay, 1999; Knopf et al., 2002). In this case, I attempt to consider CEO Wealth Delta as an alternative measurement of executive risk-taking incentives along with CEO equity-linked to total compensation ratio.

(18)

17

typically emphasize the role of such plans in aligning the interests of top management with those of the firm’s owners. LTIPs typically comprise one or more of the following vehicles: stock option plans, stock appreciation rights (SARs), restricted stock, and performance plans. Stock options provide executives the right to purchase a certain number of shares at a predetermined price-usually the market value at the time they are granted-within a given time period. Stock appreciation rights are typically attached to option grants and permit executives to exchange options. Restricted stock refers to stock of a company which is not fully transferable until certain conditions have been met. Upon satisfaction of those conditions, the stock becomes transferable by the person holding the award. I take a natural logarithmic transformation of this variable so as to control for outliers and skewness.

CEO’s Direct Compensation is the sum of bonus and salary for the period, which is all cash-based compensation. On the one hand, Berger et al. (1997) argue that CEOs with higher cash compensation are more likely to be entrenched and choose less risky projects. On the other hand, Guay (1999) argues that CEOs with higher cash compensation are likely to choose more risky projects. Hence, the sign of the coefficient on CEO’s direct compensation is difficult to predict (Niu, 2010). I also take a natural logarithmic transformation of this variable so as to control for outliers and skewness.

I include CEO Age into the regression as control variable. CEO Age is the CEO’s age in years. This proxy has been applied in related studies such as Guay (1999). I expect a negative association between CEO Age and risk-seeking behavior. Besides, consistent with the existing literature, I use CEO Tenure (to retirement, in role, on board, and in organization) to proxy for the CEO’s level of risk aversion. For instance, Berger et al. (1997) indicate that CEOs with longer tenures are more likely to be entrenched and will seek to avoid risk. Therefore, I expect the negative association between in role, on board, in organization and risk-seeking behavior. However, I expect the positive association between to retirement and risk-seeking behavior, since the more years to retirement for CEOs, the more likely CEOs are willing to take risks.

(19)

18

financing are significantly lower for larger firms (Archer & Faerber, 1966). Easy access to external funds may allow large firms to finance risky investment projects, such as R&D projects, and/or take an aggressive position in their financial policies such as holding less liquid assets (Cohen et al., 1987; Opler et al., 1999). The next control variable is the Market-to-Book ratio (MtB), defined as market value of assets to book value of assets, as a proxy for investment opportunities and growth of the firm. I include Return on Assets (ROA) defined as EBITDA scaled by assets. In order to control the investment and growth opportunities (Coles et al., 2006), I calculate sales growth as the ratio of total sales in year t to total sales in year t-1, because high-growth firms may be inclined to take on additional risk. However, Hymer and Pshigian (1962) argue that high-growth companies are normally young firms so that they may have more difficulty to access capital need to finance risky investment projects.

Finally, since my data sample includes multiple years and multiple industries, I take account of year and industry fixed effects to control industry characteristics and overall macroeconomic factors over time. Therefore, I add year dummies and industry dummies in my regression model as well.

(20)

19

Sample Distribution and Summary Statistics

(21)

20

statistics of firm characteristics variables. It can be noticed that the total sales has a mean of 5816.3949 and a stand deviation of 22,706.0183 in £millions. The mean of MtB and ROA are 1.1675 and 0.0562, respectively. Finally, the sales growth which measured as the ratio of total sales in year t to total sales in year t-1 has a mean of 2.3572 and a standard deviation of 25.4324. Overall, Table 3, Table 4 and Table 5 provide descriptive statistics for dependent, independent and control variables. All numbers are similar to values reported in previous related researches, such as Barclay et al. (2006), Cassell et al. (2012) and Guay (1999).

TABLE 2

Sample distribution by year

Year Frequency Percent

(22)

21

Table 3 Summary statistics of risk-taking policy related variables. This table presents the summary statistics for the variables used in my analysis over the period 2002 to 2009. R&D exp/sales ratio is the ratio of R&D expenditures to total sales. Diversification is the acquisition of businesses or companies by the firm. Working capital is current assets minus current liabilities scaled by total assets. Leverage is the ratio of total long term debt to total assets. All variables are winsorized at the 1st and 99th percentile levels.

TABLE 3

Summary statistics of risk-taking policy related variables

Obs. Mean Std. Dev. 1％ 25％ Median 75％ 99％

Risk-taking policy

R&D exp/sales ratio t 696 0.2669 2.8993 0.0001 0.0022 0.0174 0.0592 1.5181

R&D exp/sales ratio t+1 507 0.1958 1.7924 0.0001 0.0021 0.0172 0.0594 1.4291

(23)

22

Table 4 Summary statistics of CEO characteristics related variables. This table shows the descriptive statistics for the variables over the period 2002 to 2009. All variables are winsorized at the 1st and 99th percentile levels.

TABLE 4

Summary statistics of CEO characteristics variables

CEO characteristics

CEO inside debt pay 939 105,000 106,800 3,000 29,000 68,000 145,000 485,000

CEO equity-linked pay 1511 1,287,388 2,470,315 3,000 250,000 555,000 1,300,000 11,498,440

CEO bonus 1577 397,000 446,800 19,200 131,000 262,000 492,000 2,280,000

CEO total pay 1836 1,860,412 2,679,326 102,590 600,000 1,064,000 2,001,000 13,283,490

CEO debt/total ratio 939 0.0814 0.0752 0.0022 0.0341 0.0603 0.1051 0.3964

CEO equity/total ratio 1512 0.4560 0.2047 0.0046 0.3195 0.4584 0.6023 0.9016

CEO bonus/total ratio 1557 0.2349 0.1356 0.0275 0.1413 0.2160 0.3006 0.7169

Wealth Delta 1795 190,093 652,170 2,000 15,000 41,000 104,000 2,964,520

LTIPs 1456 1,250,000 2,285,000 21,570 267,000 564,500 1,300,000 11,800,000

Direct Compensation (salary + bonus) 1834 801,780 605,840 82,350 410,000 629.500 990,000 3,118,200

CEO age (years) 1839 51.45 6.750 37.00 47.00 51.00 56.00 67.00

To retirement 1839 12.36 7.047 -2.30 7.30 12.30 17.50 27.62

In role 1840 5.07 5.372 0.00 1.60 3.60 6.60 27.20

On board 1840 7.89 7.000 0.20 2.90 6.00 10.60 33.93

(24)

23

Table 5 Summary statistics of firm characteristic related variables. This table indicates the descriptive statistics for the variables over the period 2002 to 2009. Market-to-Book ratio (MtB) is the ratio of the market value of equity to the book value of assets. Sales growth is the ratio of total sales in year t to total sales in year t-1.All variables are winsorized at the 1st and 99th percentile levels.

TABLE 5

Summary statistics of firm characteristic variables

Firm characteristics

Total assets (£millions) 1829 5816.3949 22,706.0183 36.1758 301.9490 890.9000 2792.4500 131,915.200

MtB 1732 1.1675 1.1146 0.0467 0.5383 0.8655 1.3895 5.8831

ROA 1732 0.0562 0.0819 -0.3537 0.0291 0.0561 0.0919 0.2632

(25)

24

(26)

25 TABLE 6

Correlations among key variables

1 2 3 4 5 6 7 8

1 R&D exp/sales ratio t 1

2 Ln(Diversification t) -0.029 (0.603) 1 3 Working capital t -0.007 (0.851) -0.086* (0.015) 1 4 Leverage t -0.093* (0.014) 0.130** (0.000) -0.052* (0.027) 1

5 CEO debt/total ratio -0.021

(0.696) -0.104* (0.031) 0.035 (0.290) 0.100** (0.002) 1

6 CEO equity/total ratio -0.128**

(0.002) 0.236** (0.000) 0.017 (0.514) 0.047 (0.066) -0.382** (0.000) 1

7 CEO bonus/total ratio -0.016

(0.690) -0.054 (0.154) -0.017 (0.503) -0.072** (0.005) 0.040 (0.251) -0.616** (0.000) 1 8 Ln(Wealth Delta) -0.065 (0.090) 0.267** (0.000) 0.037 (0.114) -0.061** (0.010) -0.154** (0.000) 0.384** (0.000) 0.011 (0.680) 1

**. Correlation is significant at the 0.01 level (2-tailed).

(27)

26

DISCUSSIONS OF EMPIRICAL RESULTS

Table 7 below demonstrates the regression results intend to test hypotheses H1a and H1b. The primary explanatory variable is CEO inside debt to total compensation ratio. I include year and industry fixed effects in all specifications. Consistent with existing studies on investment policies, I control the natural logarithm of CEO’s Wealth Delta, CEO’s total LTIPs, and CEO’s direct compensation in my regression. Besides, I also take CEO’s age and tenure (including the years to retirement, in role, on board and in organization) into account. Besides, the natural logarithm of firm’s total assets, market-to-book ratio, ROA and sales growth are also included as control variables.

The first and the second OLS regressions (column 1 and column 2) in Table 7 show that the coefficient on CEO inside debt to total ratio is negative and significant at statistical significance 5% level (p<0.05) at year t and year t+1, suggesting that R&D expenditures depend negatively on CEO debt to total ratio. Thus, it can be concluded that CEO inside debt to total ratio reduce the R&D expenditures, which implies that CEOs who have higher inside debt to total compensation ratio, tends to not take larger investment on R&D related projects. This result is consistent with the hypothesis H1a, which shows that the larger CEO inside debt to total compensation ratio, the less riskily that CEOs conduct R&D investment policy choices. Meanwhile, the regression coefficient on Wealth Delta is positive and significant at statistical significance 5% level (p<0.05) at year t and year t+1. However, the OLS regressions also indicate that there is a negative and significant coefficient on LTIPs (p<0.01) at year t and year t+1. Besides, I find out that the coefficient on MtB is positive and significant at 5% level. On contrary, the coefficient on ROA is negative and statistically significant at 1% level.

(28)

27

acquisition in year t+1 (p<0.05). Moreover, the regression coefficient on years of CEOs on board is positive and statistically significant at 5% in year t and 10% in year t+1. Furthermore, it can be seen that the coefficient on sales growth in year t is positive and statistically significant at 1% in year t. However, I find no significant results between sales growth and diversification in year t+1.

Table 7 OLS regression results of risk-taking investment policies on CEO inside debt to total compensation ratio. The dependent variables are R&D expenditures to sales ratio in year t and year t+1, and the firm diversification in year t and year t+1, respectively. All control variables are described in Table 1. Intercepts are not reported, p-values are two-tailed within parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

TABLE 7

The impact of CEO inside debt to total compensation on investment policy

R&D exp/sales ratio t R&D exp/sales ratio t+1 Diversification t Diversification t+1

(1) (2) (3) (4)

CEO debt/total ratio -0.154**

(29)

28 Ln(Total assets) -0.118 (0.175) -0.080 (0.418) 0.641*** (0.000) 0.490*** (0.000) MtB 0.144** (0.012) 0.152** (0.018) 0.104** (0.036) 0.062 (0.245) ROA -0.425*** (0.000) -0.396*** (0.000) 0.016 (0.751) 0.113** (0.032) Sales growth 0.026 (0.626) -0.006 (0.918) 0.143*** (0.001) 0.049 (0.289)

Year Dummies YES YES YES YES

Industry Dummies YES YES YES YES

N 283 235 358 329

Adjusted R2 0.247 0.230 0.352 0.313

Table 8 demonstrates the impact of CEO inside debt to total compensation ratio on financial policies so as to examine hypotheses H1c and H1d. Firstly, the results in column 1 show that the coefficient on CEO inside debt to total ratio is not significant in year t. Nevertheless, I notice that CEO inside debt to total ratio has positive and statistically significant impact on working capital in year t+1 (p<0.1). This result is consistent with hypothesis H1c. Besides, the regression coefficient on direct compensation is positive and significant at statistical significance 10% level. The coefficient on CEO age and years to retirement are also positive and statistically significant (p<0.01) in year t and year t+1. Moreover, I also find out that the years of CEOs in organization has positive and significant at 5% level in year t and year t+1.

(30)

29

Table 8 OLS regression results of risk-taking financial policies on CEO inside debt to total compensation ratio. The dependent variables are firm’s working capital in year t and year t+1, and the firm leverage in year t and year t+1, respectively. Working capital is measured as current assets minus current liabilities scaled by total assets. Firm leverage is defined as the ratio of total long-term debt to total assets. All control variables are described in Table 1. Intercepts are not reported, p-values are two-tailed within parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

TABLE 8

The impact of CEO inside debt to total compensation on financial policy

Working capital t Working capital t+1 Leverage t Leverage t+1

(1) (2) (3) (4)

(31)

30 (0.155) (0.023) (0.187) (0.305) Sales growth 0.012 (0.740) 0.042 (0.297) -0.003 (0.929) 0.006 (0.872)

N 718 599 718 599

Adjusted R2 _0.044 _0.061 _0.134 _0.139

Table 9 includes the regression results which attempt to test hypotheses H2a and H2b. The independent variable is CEO equity-linked to total compensation ratio. Year and industry fixed effects are included in all specifications. Based on prior research with respect to investment policies, I use the natural logarithm of CEO’s Wealth Delta, CEO’s total LTIPs, and CEO’s direct compensation as control variables in my regression. Moreover, I also control CEO’s age and tenure (including the years to retirement, in role, on board and in organization), the natural logarithm of firm’s total assets, market-to-book ratio, ROA and sales growth.

According to table 9 (column 1 and column 2), it can be noticed that the coefficient on CEO equity-linked to total ratio is not statistically significant in the first and the second OLS regression. Nevertheless, we can see that the regression coefficient on Wealth Delta is positive and significant at statistical significance 1% level both in year t and in year t+1. CEO Wealth Delta is the dollar change in the CEO’s wealth for a 1% change in stock price. This variable is often used to measure CEO incentives to increase CEO risk-taking behavior. In this case, it can be concluded that CEO with larger Wealth Delta increase the R&D expenditures, which also implies that larger Wealth Delta has an impact for CEO to take more risks. Since I take Wealth Delta as alternative measure of risk-taking incentive apart from CEO equity to total ratio, this result shows consistency with hypothesis H2a.

(32)

31

The regression coefficient on sales growth is positive and significant at statistically significance 1% level in year t.

Table 9 OLS regression results of risk-taking investment policies on CEO equity-linked to total compensation ratio. The dependent variables are R&D expenditures to sales ratio in year t and year t+1, and the firm diversification in year t and year t+1, respectively. All control variables are described in Table 1. Intercepts are not reported, p-values are two-tailed within parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

TABLE 9

The impact of CEO equity-linked to total compensation on investment policy

(1) (2) (3) (4)

(33)

32 (0.000) (0.001) (0.097) (0.031) ROA -0.381*** (0.000) -0.367*** (0.000) -0.003 (0.928) 0.097** (0.015) Sales growth 0.015 (0.701) 0.016 (0.714) 0.173*** (0.000) 0.024 (0.489)

N 561 482 645 626

Adjusted R2 0.158 0.150 0.337 0.273

Table 10 indicates the effects of CEO equity-linked to total compensation ratio on risk-taking financial policies. In order to test hypotheses H2c and H2d, I examine the coefficient on CEO equity to total ratio of firm working capital. The results show that the coefficient on CEO equity to total ratio is negative and statistically significant at 10% in year t+1, which means that the firm working capital is negatively depends on CEO equity-linked to total ratio. This result is consistent with hypothesis H2c. Moreover, I also find out that CEO age and the years to retirement has positive and significant influence on working capital at 1% level both in year t and in year t+1. However, the coefficient on years of CEO on board is negatively significant at 5% level in year t and year t+1.

(34)

33

Table 10 OLS regression results of risk-taking financial policies on CEO equity-linked to total compensation ratio. The dependent variables are firm’s working capital in year t and year t+1, and the firm leverage in year t and year t+1, respectively. Firm leverage is defined as the ratio of total long-term debt to total assets. All control variables are described in Table 1. Intercepts are not reported, p-values are two-tailed within parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

TABLE 10

The impact of CEO equity-linked to total compensation on financial policy

(1) (2) (3) (4)

(35)

34 (0.011) (0.406) (0.001) (0.002) ROA 0.096*** (0.001) 0.108*** (0.000) -0.066** (0.021) -0.033 (0.278) Sales growth 0.145*** (0.000) 0.249*** (0.000) -0.066*** (0.010) -0.069** (0.013)

N 1372 1178 1372 1178

Adjusted R2 0.070 0.109 0.122 0.118

(36)

35

Table 11 OLS regression results of risk-taking investment policies on CEO bonuses to total compensation ratio. The dependent variables are R&D expenditures to sales ratio in year t and year t+1, and the firm diversification in year t and year t+1, respectively. All control variables are described in Table 1. Intercepts are not reported, p-values are two-tailed within parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

TABLE 11

The impact of CEO bonuses to total compensation on investment policy

(1) (2) (3) (4)

(37)

36 (0.000) (0.000) (0.499) (0.021) Sales growth 0.016 (0.701) 0.016 (0.718) 0.165*** (0.000) 0.031 (0.397)

N 500 437 574 570

Adjusted R2 _0.192 _0.186 _0.340 _0.258

(38)

37

Table 12 OLS regression results of risk-taking investment policies on CEO bonuses to total compensation ratio. The dependent variables are firm working capital (current assets minus current liabilities scaled by total assets) in year t and year t+1, and the firm leverage in year t and year t+1, respectively. Firm leverage ratio is defined as total long-term debt divided by total assets. All control variables are described in Table 1. Intercepts are not reported, p-values are two-tailed within parentheses. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

TABLE 12

The impact of CEO bonuses to total compensation on financial policy

(1) (2) (3) (4)

(39)

38 (0.019) (0.722) (0.006) (0.008) ROA 0.108*** (0.000) 0.115*** (0.000) -0.087*** (0.004) -0.056* (0.089) Sales growth 0.158*** (0.000) 0.267*** (0.000) -0.066** (0.016) -0.072** (0.015)

N 1204 1035 1204 1036

Adjusted R2 0.103 0.132 0.124 0.116

To be concluded, Table 13 show the comparison results between CEO inside debt to total compensation ratio, CEO equity-linked to total compensation ratio, and CEO bonuses to total compensation ratio. It is shown that R&D exp/sales ratio is negatively depends on CEO debt to total ratio both in year t and year t+1. However, R&D investment policy is positively depends on CEO Wealth Delta. CEO bonuses/total ratio has no significant influence on R&D investment policy. With respect to firm diversification, the results indicate CEO equity to total ratio, CEO Wealth Delta, and CEO bonuses to total ratio have positive influence on firm diversification. In terms of financial policies, I find CEO debt to total ratio has positive influence on working capital in year t+1, and positive impact on firm leverage in year t+1, respectively. On the contrary, CEO equity to total ratio has negatively effects on working capital and leverage in year t+1. Meanwhile, CEO Wealth Delta and CEO bonuses to total ratio have no significant influence on firm working capital and leverage.

TABLE 13

(40)

39

Robustness Test

The robustness test intends to control the potential endogeneity between CEO compensation components and investment and financial risk-taking policies. Hence, a further sensitivity test is conducted by using a two-stage least-squares framework. Cassell et al. (2012) indicate that research testing the effects of inside debt on investment and financial policies is scant. Hence, it is relatively difficult to find appropriate instrumental variables for CEO inside debt to total ratio, as well as CEO equity-linked to total ratio and CEO bonuses to total ratio. In this case, all instrumental variables should be valid, which means that all instrumental variables should be correlated with independent variables and uncorrelated with the error term in the second stage. According to Sundaram and Yermack (2007), there are two instrumental variables which can be taken into account: a categorical variable coded one if the firm has a New CEO, and zero otherwise; and another categorical variable equals to one if the company faces a Liquidity constraint (the company generates negative cash flow), and zero otherwise. To be more specific, the indicator variable of New CEO equals to one if the CEO is in role within one year, otherwise this variable equals to zero.

(41)

40

Table 14 illustrates the results of 2SLS analysis. I include year and industry fixed effects in 2SLS analysis. p-values in parentheses are based on robust standard errors that are adjusted for heteroskedasticity and clustered by firm. p-values are two-tailed. ***, **, and * represent significance at the 1%, 5%, and 10% levels, respectively.

TABLE 14

The association between CEO inside debt to total ratio and risk-taking behavior

Two-stage least-squares (2SLS) results

R&D exp/sales ratio t+1 Diversification t+1 Working capital t+1 Leverage t+1 CEO debt/total ratio

(42)

41 ROA -0.403*** (0.000) 0.178** (0.024) -0.112** (0.040) -0.037 (0.386) Sales growth 0.000 (0.994) -0.005 (0.945) -0.032 (0.549) 0.002 (0.966)

N 234 328 540 598

Adjusted R2 0.217 0.240 0.098 0.107

Table 15 indicates the 2SLS analysis with regard to CEO equity-linked to total ratio and risk-taking investment and financial policies (OLS regression 5b, 6b, 7b and 8b). The positive correlation between CEO equity/total ratio and lagged R&D exp/sales ratio is robust to controlling for other factors. It is also robust to adding firm fixed effects and incorporating the endogenous choice of CEO equity/total ratio. Table 15 presents the results for the second-stage model that are consistent with the results based on OLS when I apply the endogenous values of the independent variables.

Table 15 illustrates the results of 2SLS analysis. I include year and industry fixed effects in 2SLS analysis. p-values in parentheses are based on robust standard errors that are adjusted for heteroskedasticity and clustered by firm. p-values are two-tailed. ***, **, and * represent significance at the 1%, 5%, and 10% levels, respectively.

TABLE 15

The association between CEO equity-linked to total ratio and risk-taking behavior

R&D exp/sales ratio t+1 Diversification t+1 Working capital t+1 Leverage t+1 CEO equity/total ratio

(43)

42 (0.335) (0.235) (0.080) (0.154) CEO Age -0.052 (0.727) 0.110 (0.377) 0.017 (0.930) 0.067 (0.614) To retirement -0.098 (0.523) 0.108 (0.391) -0.060 (0.767) 0.079 (0.563) In role 0.061 (0.364) -0.020 (0.681) 0.005 (0.945) -0.054 (0.268) On board -0.108 (0.200) 0.052 (0.494) -0.393*** (0.001) 0.062 (0.486) In organization 0.018 (0.797) -0.081 (0.190) 0.248*** (0.001) -0.172*** (0.002) Ln(Total assets) -0.099 (0.241) 0.411*** (0.000) -0.082 (0.534) 0.527*** (0.000) MtB 0.152*** (0.002) 0.086** (0.033) 0.201*** (0.006) -0.046 (0.421) ROA -0.358*** (0.000) 0.097** (0.015) -0.168** (0.015) -0.068 (0.138) Sales growth 0.012 (0.776) 0.024 (0.491) 0.037 (0.400) -0.051 (0.149)

N 481 625 1015 1177

Adjusted R2 0.096 0.272 0.059 0.088

(44)

43

Table 16 illustrates the results of 2SLS analysis regarding CEO bonus/total ratio and risk-taking behavior. I include year and industry fixed effects in second-stage analysis. p-values in parentheses are based on robust standard errors that are adjusted for heteroskedasticity and clustered by firm. p-values are two-tailed. ***, **, and * represent significance at the 1%, 5%, and 10% levels, respectively.

TABLE 16

The association between CEO bonuses to total ratio and risk-taking behavior

R&D exp/sales ratio t+1 Diversification t Working capital t+1 Leverage t+1 CEO bonus/total ratio

(45)

44 Sales growth 0.017 (0.712) 0.202*** (0.000) 0.269*** (0.000) -0.070** (0.028)

N 436 573 1034 1035

(46)

45

CONCLUSIONS

(47)

46

independent variable to test if this ratio has an influence on risk-taking behavior. The results demonstrate that CEO bonus to total ratio has a significantly positive impact on firm diversification with regard to investment policy.

To sum up, my primary contribution is to show how CEO inside debt compensation (new annually grants) affect investment and financial risk-taking behavior. Meanwhile, I also compare the influences of executive equity-linked incentives and bonuses on risk-taking behavior simultaneously. The main findings show that CEOs with larger inside debt to total compensation ratio tend to avoid aggressive R&D investment choices and has a higher financial policy on working capital. However, the results show no significant association between executive inside debt compensation and firm diversification. More importantly, I find the executive inside debt compensation has a positive influence on firm leverage. The potential reason will be discussed in the following future research section. Furthermore, I also find that CEO equity-linked incentives have a significantly positive impact on R&D investment policy and a negative effect on working capital financial policy, which also implies that CEO equity-linked incentives have contrary impacts on R&D investment policy and working capital financial policy compared with CEO inside debt compensation. In general, it is indicated that executive bonuses compensation has no significant association with risk-taking behavior except firm diversification.

(48)

47

across firms (Guay, 1999), and both affect risk-taking behavior (Coles et al., 2006). Moreover, in the robustness sensitivity tests, I do not consider tax status of the firm, the tax rate on individual income, industry median and cash surplus (net cash flow from operations less depreciation expense plus R&D expenditures) due to the lack of data variables. Taking into account of the variables which are not used in my analysis could posses interesting possibilities for future research.

(49)

48

REFERENCES

Amihud, Y., & Lev, B. (1981). Risk reduction as a managerial motive for conglomerate mergers. The Bell Journal of Economics, 12 (2), 605-617.

Anantharaman, D., Fang, V. W., & Guong, G. (2011). Inside debt and the design of corporate debt contracts. Working Paper, available on www.ssrn.com.

Archer, S. H., & Faerber, L. G. (1966). Firm size and the cost of externally secured equity capital. Journal of Finance, 21 (1), 69-83.

Barclay, M. J., Smith, C.W., Jr., & Morellec, E. (2006). On the debt capacity of growth options. Journal of Business, 79 (1), 37-59.

Baysinger, B., Kosnik, R., & Turk, T. (1991). Effects of board and ownership structure on corporate R&D strategy. Academy of Management Journal, 34 (1), 205-214.

Beatty, R. P., & Zajac, E. J. (1994). Managerial incentives, monitoring, and risk bearing: A study of executive compensation, ownership, and board structure in initial public offerings. Administrative Science Quarterly, 39 (2), 313-335.

Belkhir, M., & Boubaker, S. (2012). CEO inside debt and hedging decisions: lessons from the U.S. Banking Industry. Journal of International Financial Markets, Institutions & Money. doi: 10.1016/j.intfin.2012.11.009

Benchuk, L. A., & Jolls, C. (1999). Managerial value diversion and shareholder wealth. Journal of Law, Economcis, & Organization, 15 (2), 487-502.

Bennett, R., Guntay, L., & Unal, H. (2012). Inside debt and bank default risk during the crisis. Working Paper, University of Maryland.

(50)

49

Bryan, S., Lee, H. S., & Lilien, S. (2000). CEO stock-based compensation: An empirical analysis of incentive-intensity, relative mix, and economic determinants. Journal of Business, 73 (4), 661-693.

Cassell, C. A., Huang, S. X., Sanchez, J. M., & Stuart, M. D. (2012). Seeking safety: The relation between CEO inside debt holdings and the riskiness of firm investment and financial policies. Journal of Financial Economics, 103 (3), 588-610.

Cen, W. (2013). The determinants of CEO inside debt and its components. Working paper, European Accounting Association.

Chasan, E. (2009). Iacocca losing pension, car in Chrysler bankruptcy. Reuters, May 29, 2009 from

http://www.reuters.com/article/2009/05/29/us-chrysler-iacocca-idUSTRE54S0RY2009 0529

Chava, S., Kumar, P., & Warga, A. (2010). Managerial agency and bond covenants. Review of Financial Studies, 23 (3), 1120-1148.

Cohen, R., Hall, B., & Viceira, L. (2000). Do executive stock options encourage risk-taking?. Working paper, Harvard Business School.

Cohen, W. M., Levin & R. C., & Mowery, D. C. (1987). Firm size and R&D intensity: A re-examination. Journal of Industrial Economics, 35 (4), 543-565.

Coles, J. L., Daniel, N. D., & Naveen, L. (2006). Managerial incentives and risk-taking. Journal of Financial Economics, 79 (2), 431-468.

Core, J., & Guay, W. (2002). Estimating the value of employee stock option portfolios and their sensitivities to price and volatility. Journal of Accounting Research, 40 (3), 613-630.

(51)

50

Dewatripont, M., & Tirole, J. (1994). A theory of debt and equity: Diversity of securities and manager-shareholder congruence. The Quarterly Journal of Economics, 109 (4), 1027-1054.

Donaldson, E. F. (1958). Corporation Finance: Principles and Problems. Journal of Finance, 12 (4), 567-568.

Edmans, A., & Liu, Q. (2011). Inside debt. Review of Finance, 15 (1), 75-102.

Galai, D., & Masulis, R. W. (1976). The option pricing model and the risk factor of stock. Journal of Financial Economics, 3 (1/2), 53-81.

Garen, J. E. (1994). Executive compensation and principal-agent theory. Journal of Political Economy, 102 (6), 1175-1200.

Geczy, C., Minton, B. A., & Schrand, C. (1997). Why firms use currency derivatives. Journal of Finance, 52 (4), 1323-1354.

Gerakos, J. J. (2010). CEO pensions: Disclosure, managerial power, and optimal contracting. Working Paper, University of Chicago, Booth School of Business.

Gianluca, P., & Speciale, B. (2010). Financial leverage and managerial compensation: Evidence from UK, Research in Economics, forthcoming.

Grabke-Rundell, A., & Gomez-Mejia, L. R. (2002). Power as a determinant of executive compensation. Human Resource Management Review, 12 (1), 3-23.

Guay, W. (1999). The sensitivity of CEO wealth to equity risk: an analysis of the magnitude and determinants. Journal of Financial Economics, 53 (1), 43-71.

Hill, C. W. L., Hitt, M. A., & Hoskisson, R. E. (1988). Declining U.S. Competitiveness: Reflections on a crisis. Academey of Management Executive, 2 (1), 51-60.

(52)

51

Hill, C. W. L., Hitt, M. A., & Hoskisson, R. E. (1992). Cooperative versus competitive structures in related and unrelated diversified firms. Organization Science, 3 (4), 501-521.

Hymer, S., & Pashigian, P. (1962). Firm size and rate of growth. Journal of Political Economy, 70 (6), 556-569.

Jacquemin, A., & Berry, C. (1979). Entropy measure of diversification and corporate growth. Journal of Industrial Economics, 27 (4), 359-369.

Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3 (4), 305-360.

John, T., & John, K. (1993). Top-management compensation and capital structure. Journal of Finance, 48 (3), 949-974.

Knopf, J. D., Ham, J., & Thornton, J. H., Jr. (2002). The volatility and price sensitivities of managerial stock option portfolios and corporate hedging. Journal of Finance, 57 (2), 801-813.

Low, J. (2009). Managerial risk-taking behavior and equity-based compensation. Journal of Financial Economics, 92 (3), 470-490.

May, D. (1995). Do managerial motives influence firm risk reduction strategies? Journal of Finance, 50 (4), 1291-1308.

McConnell, J. J., & Servaes, H. (1990). Additional evidence on equity ownership and corporate value. Journal of Financial Economics, 27 (2), 595-612.

Morck, R., Shleifer, A., & Vishny, R. W. (1988). Management ownership and market valuation: An empirical analysis. Journal of Financial Economics, 20 (1/2), 293-315.

Murphy, K. (1985). Corporate performance and managerial remuneration: An empirical investigation. Journal of Accounting and Economics, 7 (1-3), 11-42.