Equity or Debt Financing: Does Good Corporate Governance Matter?

Vivek Mande White, Nelson Professor of Accounting California State University, Fullerton

Young K. Park Professor of Finance Sungkyunkwan University

Myungsoo Son* Associate Professor of Accounting California State University, Fullerton

January 2010

*4350 Mihaylo Hall, California State University, Fullerton, Fullerton, CA 92834. TEL/FAX/EMAIL: (657)278-2732/ (657)278-4518/[email protected].

Equity or Debt Financing: Does Good Corporate Governance Matter? SUMMARY

We examine whether good corporate governance plays a role in influencing a firm’s financing policy. Specifically, we hypothesize that firms with strong governance tend to prefer stocks over long-term debt. We argue that this is because strong governance reduces information asymmetry between providers of capital (equity investors) and users of capital (managers). While the reduction of information asymmetry has positive connotations for both equity and debt financing, we argue that there is a more significant effect on equity financing. Our test results are consistent with this prediction. We find that equity is preferred to debt financing in firms with high quality governance. This finding contradicts an alternative theory, the pecking order hypothesis, which suggests that firms will issue equity as their last resort. Our results suggest, instead, that strong governance can lower agency costs to a level where equity becomes more attractive to debt as a method of long-term financing. Keywords: corporate governance; equity finance; agency theory JEL Classification: G3

1

Equity or Debt Financing: Does Good Corporate Governance Matter? 1. Introduction This paper examines whether the quality of a firm’s corporate governance plays a role in the firm’s choice of equity versus debt financing. Agency problems arise due to the separation of ownership of a company from its management, and the resultant asymmetry of information between the two groups (Jensen and Meckling 1976). La Porta et al. (1998) argue that owners facing information asymmetry seek a mechanism to protect their interests.

Investing in high quality

corporate governance provides a potential remedy to many agency problems. Good governance, for example, ensures that the quality of the financial reporting process is high (Cohen et al. 2004).1 Fund providers (investors) monitor fund users (managers) by evaluating their performance using reported financial measures. Without high quality financing reporting, investors would have to expend additional resources to monitor the proper use of their funds by managers. Prior studies find that investors are more willing to provide financing to firms with high quality governance. For example, Shleifer and Vishny (1997) consider corporate governance as “the way in which suppliers of finance to corporations assure themselves of getting a return on their investment”. La Porta et al. (1998) also argue that a higher level of investor protection increases investors’ willingness to provide financing as reflected in a lower cost of capital and a greater availability of external financing. Coombes and Watson (2000) report survey results showing that three-quarters of institutional investors view board governance being at least as important as financial

1

Firms with strong corporate governance are also less likely to manipulate financial information (Wu et al. 2007).

2

performance when they evaluate companies for investment. Their survey also shows that institutional investors are willing to pay more for shares of well-governed companies. A popular asymmetric information theory of external financing (i.e., capital structure) examined by the literature is the pecking order hypothesis (POH) (see Klein et al. 2002). This theory argues that in the face of information asymmetry, firms will prefer debt over equity, because investors will significantly discount the price of equity taking into account agency costs, and make an equity offering unattractive for the firm. Our paper, however, is motivated by anecdotal empirical observation that some firms with effective governance appear to choose equity over debt. We offer an alternative hypothesis to the POH, proposing that effective governance can reduce information asymmetry problems facing some firms to a level where equity becomes more desirable to debt. Using a comprehensive set of measures of corporate governance, we study whether governance quality is positively related with the probability of issuing equity. In addition, we examine whether the amount of equity financing is positively related with corporate governance. Our univariate analysis and multivariate regression results are generally consistent with our hypothesis, namely that we find a positive association between our governance proxies and the likelihood and amount of equity financing. Additional analysis shows that board independence and board size are significantly related to equity financing, confirming that the quality of board governance is important to investors providing capital to firms. We also find that the impact of corporate governance with regard to equity financing is more pronounced for small firms who face bigger information asymmetry problems than large firms.

3

The organization of this study consists of four remaining parts. Section 2 discusses the literature and presents the hypotheses, section 3 discusses our research design, section 4 reports the empirical results, and the last section concludes the study. 2. Literature Review and Hypotheses Development There are three main ways for a firm to finance an investment opportunity: internal funds, debt financing, and equity financing. The pecking order hypothesis (POH) argues that because managers have more information about a firm’s prospects than outside investors, markets will undervalue a stock offering (Baker and Wurgler 2002). That is, the information asymmetry facing investors creates a potential moral hazard problem where managers are free to pursue their own interests at the expense of external investors. Investors predictably will take actions to protect their interests by discounting the firm’s stock price (Myers and Majluf 1984). Given that debt securities are less sensitive to information asymmetry problems, and consequently are less affected by underpricing, the POH argues that equity financing will be used as a last resort, i.e., after firms first pursue financing using internal funds and debt financing. Another genre of studies argues that entrenched managers tend to avoid debt financing. These papers argue that managers generally like lower leverage because this provides them with greater financial flexibility and brings higher prestige to the firm in the form of higher bond ratings (Hovakimian et al. 2001). Entrenched managers also dislike being under pressure to perform because with debt financing, managers are “bonded” to make fixed payments of interest and principal (Jensen 1986). Jensen argues that high debt levels constrain managers from diverting free cash flow to pursue their personal goals at the expense of value maximization. In support, Berger et al. (1997) find

4

that firms with entrenched managers (low quality governance) tend to issue equity rather than debt. This paper, however, proposes an alternative hypothesis which predicts that firms with effective governance will prefer to raise capital externally using equity instead of debt. Following Bhojraj and Sengupta (2003), we argue that effective corporate governance reduces both agency and information risk. Agency risk is defined as the risk that managers as agents take actions2 that deviate from the maximization of firm value (Bhojraj and Sengupta 2003). Effective corporate governance makes it possible to actively monitor management’s actions and for boards to take actions to protect shareholders’ interest (e.g., Weisbach 1988; Borokhovich et al. 1996). Information risk is the risk that information reported by management is not credible for evaluating a firm’s performance by outside investors. Effective corporate governance reduces the information risk by ensuring timely release of credible financial information (Ajinkya et al. 2005). We posit that a reduction in agency and information risks can make equity financing a desirable choice for some firms. Prior empirical studies support the idea that effective corporate governance reduces the cost of equity. Ashbaugh et al. (2004) find a negative relation between the cost of equity and a number of governance attributes, including earnings quality, the independence of the board and audit committees, the percentage of institutional ownership, and the percentage of the board that owns stock. Gompers et al. (2003) also find that firms with stronger shareholder rights have higher share values and enjoy a lower cost of equity capital. Although these results provide a link between governance and the cost of equity, they do not provide direct evidence on whether the quality of 2

Examples include expropriation of assets and empire-building activities.

5

governance affects a firm’s decision to finance its investment opportunities using equity instead of debt. We consider the possibility that effective governance also reduces the cost of debt financing. For example, Bhojraj and Sengupta (2003) find that firms with a higher percentage of outside directors and those with greater institutional ownership enjoy lower bond yields and higher credit ratings on their new debt issues. They argue that effective governance reduces default risk by mitigating agency costs and through better monitoring of managerial performance as a result of a reduced level of information asymmetry between the firm and its lenders. Using a broader set of governance variables, AshbaughSkaife et al. (2006) also document that firms with strong governance benefit from higher credit ratings on debt issuances. Given that good corporate governance has positive connotations for both debt and equity financing, which financing policy will be chosen is an empirical issue. We argue, however, that firms with high quality governance are more likely to consider issuing equity than debt for the following reasons. First, it is a generally well accepted fact that equity, compared to debt, is more sensitive to the reduction of information asymmetry (Boot and Thakor 1993; Fulghieri and Lukin 2001). Chang et al. (2006), for example, find that as the number of analyst-following (a proxy for lower information asymmetry) increases, firms are more likely to issue equity than debt. Similarly, Change et al. (2009) find that firms using Big N auditors (a proxy for external monitoring) show a preference for equity over debt financing. Second, the POH relies on the logic that firms prefer debt financing to equity because the impact of information asymmetry problems is greatest on equity financing. Finally, some bondholders view good corporate governance differently

6

than do shareholders. Klock et al. (2005) find that weak shareholder rights due to antitakeover provisions (weak governance) are associated with a lower cost of debt financing. That is, bondholders view a takeover as a “negative” event because it significantly increases the financial risk of the firm and results in wealth transfers from bondholders to stockholders (Warga and Welch 1993). This suggests that some components of good governance such as strong shareholder rights that reduce the cost of equity financing have the opposite effect on the cost of debt financing. For these reasons, we expect the positive impact of reduced information asymmetry from quality corporate governance to be greater for equity financing.3 Based on the discussion above, we state two related hypotheses in the alternative form as follows: H1A: Firms with good corporate governance are more likely to issue equity than long-term debt. H1B: Firms with good corporate governance issue greater amounts of equity than other firms.

3. Research Design 3.1. Sample Our initial sample consists of 3,195 firm-years (900 distinct firms) that: (1) issued equity or debt (but not both), and (2) had corporate governance data available on Investor Responsibility Research Center’s (IRRC’s) databases during 1998-2006. 4 We then exclude 592 observations due to the lack of firm-specific financial and stock return data. 3

Several studies show that equity issuances are “information sensitive”. For example, it is rare for firms to issue equity prior to the release of their annual reports because earnings releases mitigate asymmetric information problems (Klein et al. 2002). Further, unusually positive releases precede equity issues in contrast to earnings releases in the years following equity issues (Korajezyk et al. 1991). 4 We chose 1998 as the first year in our sample period because IRRC has provided a complete set of corporate governance information since 1998.

7

Following other studies, we also exclude firms (554 observations) belonging to the financial services industry (SIC codes 6000-6900), and utilities (SIC codes 4900-4999) (for example, Chang et al. 2009). This leaves us with a final sample of 2,049 observations consisting of 288 observations with equity issuances and 1,761 observations with debt issuances. Panel A of Table 1 provides the distribution of our sample by year. The firms are generally evenly distributed across years with a slight concentration in the year 2001 (276 observations). On average, about 14% of the firms (288 / 2,049) chose equity financing, which is consistent with debt being a more popular choice for long-term financing (Bhojraj and Sengupta 2003). However, the fact that a non-trivial percentage (i.e., 14%) of the sample chose equity financing over debt suggests that the POH provides only an incomplete theory of corporate financing policy. This becomes more evident when we examine the trend in Panel B of Table 1. The panel presents the sample distribution by “Governance Score”, which is our main proxy for the quality of a firm’s corporate governance. As discussed in more detail in the next section, the possible values of the score range from 0 (minimum) to 6 (maximum), with high scores representing a higher quality of corporate governance. The panel shows that the percentage of firms issuing stock monotonically increases with the governance score. It is also noteworthy that a significantly large number (41%) of the sampled firms with the highest governance score (six) chose equity over debt financing. The pattern seen in Panel B is not consistent with the prediction of the POH. Rather, this preliminary result is more consistent with our hypothesis that the probability of issuing equity increases with the quality of a firm’s

8

corporate governance.5 A Chi-square test of association indicates that equity financing is positively and significantly associated with governance score at the 1 % level. (Insert TABLE 1 Here) 3.2. Corporate Governance Measures From the IRRC databases, we select variables representing the various governance characteristics of a firm. In addition, we obtain institutional ownership from Thomson Reuters’ (CDA/Spectrum) databases. Following DeFond et al. (2005), we assign a score of one or zero for each governance variable, sum these scores across the different variables, and obtain a composite index for the quality of a firm’s corporate governance. 6 This index is our “Governance Score” which is our main proxy for the quality of governance: 1) Board size (DNUMBD) – Previous studies find that small boards are associated with good governance (Yermack 1996; Cheng 2008). As board size increases, coordination and communication problems increase, which outweighs any benefits from having more expertise on the board (Jensen 1993; Lipton and Lorsch 1992). Larger boards are also less efficient because they take more time to arrive at a consensus (Cheng 2008). Therefore, we score 1 (for good governance) if the size of a firm’s board is less than the sample median size, and 0 otherwise. 2) Board independence (DRINDBD) – A higher proportion of outside directors is associated with stronger governance (Denis et al. 1997; DeFond et al. 2005). Thus, we

5

We find a similar pattern when we replaced the Governance Score with the lagged Governance Score. Specifically, 15.57% of firms with higher scores (4-6) issued equity, while only 8.42% of firms with lower scores (0-3) use the equity financing. 6 We include all firms available in IRRC to compute the Governance Score during 1998-2006.

9

score 1 (for good governance) if 60% or more of the directors are independent, and 0 otherwise (see DeFond et al. 2005). 3) Audit committee size (DRACBD) – Large audit committees foster stronger governance. NYSE and NASDAQ, for example, require listed companies to have at least three directors on the audit committee. When more members of the board are involved in monitoring the financial reporting process of the firm, we expect the quality of the governance to increase (Anderson et al. 2004; Blue Ribbon Committee 1999). We score 1 (for good governance) if the proportion of a firm’s audit committee size to the full board size is greater than the sample median value, and 0 otherwise. 4) Audit committee independence (DRINDAC) – Fully independent audit committees are associated with stronger governance. We score 1 (for good governance) if the audit committee is fully independent, and 0 otherwise.7 5) Institutional ownership (D%INOWN) – Institutional owners improve corporate governance by providing external monitoring (Bhojraj and Sengupta 2003). Thus, we score 1 (for good governance) if the percentage of institutional ownership is greater than the sample median ownership, and 0 otherwise. 6) G index (DGINDEX) – We score 1 if a firm’s G index is below the sample median value, and 0 otherwise. The G index, developed by Gompers et al. (2003), is a composite index of 24 provisions that represents the level of shareholder protection or conversely the level of managerial entrenchment.8 High values of the G index represent a 7

Although stock exchange and SEC rules require audit committees to be fully independent for years after 2002, exemptions can be granted to firms by the SEC. See, for example, Zhang et al. (2007) and Goh (2009) who use our proxy for audit committee independence for years following 2002. 8 The G-Index is due to Gompers et al. (2003) who divide IRRC’s firm-level governance provisions first into five categories: 1) tactics for delaying hostile bidders 2) shareholder voting rights 3) directors/officer protections 4) other takeover defenses and 5) state laws. From these, the authors select 24 unique governance and calculate the G index by giving one point when a provision exists and 0 otherwise.

10

high level of management entrenchment, for example protection to management from takeovers (thus weak governance). 3.2.1. Alternative Governance Measures: G-Index and E-Index In addition to our main proxy, we consider two alternative proxies for the quality of governance. A number of previous studies (e.g., Ashbaugh et al. 2004; Klock et al. 2005; Cremers and Nair 2005) have utilized the continuous values of the G index as a measure of the quality of governance. Another governance proxy that has some appeal is one developed by Bebchuk et al. (2009a) whose entrenchment index (E Index) uses a sub-set of 6 of the 24 governance provisions included in the G index.9 Bebchuk et al. argue that the six provisions that they choose are sufficient to capture how the broad powers to govern are distributed between managers and shareholders. The main merit of the E index is its parsimony.10 We, however, use the DeFond et al.’s (2005) comprehensive index as our main proxy because it encompasses both the G and E indices, as well as includes additional proxies for board governance and external monitoring. Specifically, it includes proxies for the independence and effectiveness of the full board and its audit committee, as well as a proxy for monitoring by institutional investors. These additional governance variables have been shown by numerous studies to be significant for reducing information asymmetry and increased monitoring of management’s actions (DeFond et al. 2005). Importantly, prior studies show that investors take into account the quality of

9

The six provisions consists of four that relate to constitutional limits on shareholder voting power (staggered boards, limits on shareholder amendments of the by-laws, supermajority requirements for mergers, and supermajority requirements for charter amendments). The remaining two provisions relate to poison pills and golden parachute arrangements. 10 Studies using E index include Dittmar and Mahrt-Smith (2007), Bates,et al. (2008), and Bauguess et al. (2008).

11

board governance in deciding whether to provide external financing to firms (Coombes and Watson 2000). We examine and compare the association of all three governance indices with the choice of a firm’s financing policy. As discussed, the POH does not offer any prediction regarding the nature of the association between the quality of a firm’s governance and its capital structure. In contrast, our hypothesis predicts a positive relationship between quality of governance and equity financing because we argue that good governance will reduce information asymmetry problems to a level where issuing equity becomes more attractive to issuing debt. As discussed, a competing theory in the literature is that entrenched managers will prefer to issue equity (Berger et al. 1997; Jensen 1986) because the entrenched managers do not want to be bonded to making fixed cash outlays each period. This theory, however, predicts a negative relationship between the quality of governance and the likelihood of issuing equity. 3.3. Regression Model Our main hypothesis is that firms with higher quality corporate governance will choose equity over debt financing; specifically that the likelihood and the amount of equity issuance will increase with higher governance scores. To test this, we run logistic and ordinary least squares (OLS) regressions of equity issuances on corporate governance quality and a set of control variables identified by prior studies: EISSUE (or ESIZE) = f (Governance Index, Determinants)

Variable EISSUE

Expected Sign

(1)

Definition 1 if a firm sells common and/or preferred stocks (Compustat #108) greater than 5 percent of beginning total assets, and 0 if a firm issues long-term debt (#111) greater than 5 percent of 12

beginning total assets; ESIZE GOV EINDEX GINDEX ROA

+ -

RETURN PPE DEVTGT

+ +

GROWTH

+

MB CREDIT

+ -

TAXRATE BIG N FINDEF

+ -

SIZE

-

Amount of issuance of common and preferred stocks (#108) divided by beginning total assets; Governance score suggested by DeFond et al. (2005); E index developed by Bebchuk et al. (2009a); G index developed by Gompers et al. (2003); Net income before extraordinary items (#18), deflated by total assets at the beginning of the fiscal year. Buy-and-hold annual stock returns; The ratio of fixed assets to total assets; Difference between a company’s leverage ratio (lagged one year) and the industry target debt ratio measured using the median values of leverage for each two-digit industry; Changes in sales, deflated by sales at the beginning of the fiscal year; Market value of equity deflated by book value of equity; 1 if the firm has been assigned a S&P credit rating (#280), and 0 otherwise. Marginal tax rates provided by Graham (1996);11 1 if the firm is audited by a Big N auditor, 0 otherwise; and Total financing needs of a firm measured using the total amount of financing divided by total assets. Natural logarithm of beginning total assets;

In the first model, the dependent variable, EISSUE, is a binomial variable which takes a value of 1 if a firm issues equity greater than five percent of the beginning total assets of the firm and takes a value of 0 if a firm issues debt greater than five percent of the beginning total assets. In obtaining this proxy we use the cash proceeds received (i.e., we follow the cash flow approach as opposed to the balance sheet approach which compares ending balances of debt and capital on the balance sheet).12 We use the gross amount of equity and debt issues rather than their net amounts (i.e., share issues are not netted with share repurchases) because different incentives are present for firms 11

See http://faculty.fuqua.duke.edu/~jgraham/taxform.html. Studies using the cash flow approach include Change et al. (2009, 2006), Frank and Goyal (2003) and Shyam-Sunder and Myers (1999), while those using the balance sheet approach include Baker and Wurgler (2002). We adopt a cash flow approach as cash flow accounts yield less noisy constructs (Kayhan and Titman 2007).

12

13

increasing versus decreasing the level of financing. For example, firms tend to issue equity following increases in their stock prices and tend to repurchase shares following stock price declines (Hovakimian et al. 2001).13 We also run an OLS regression using as a dependent variable ESIZE, which is the dollar amount of equity issued deflated by the beginning total assets of a firm. This test is to demonstrate that a firm with good quality governance can not only access equity more frequently, but also raise greater amounts. The principal test variable is the governance score (GOV), which is expected to be positively related to the dependent variables. In order to assess the relationship between the dependent variables and the test variable, we must control, however, for other variables that may have a significant impact on the choice and amount of financing. 1) Past Profitability: Firms with greater profits in the past increase a firm’s capacity to take more debt, and, therefore, are more likely to issue debt (Elliott et al. 2008). Hovakimian et al. (2001) also argue that firms with high earnings tend to issue more debt than equity because these firms want to readjust their capital structure to compensate for the effect of higher past accumulated earnings. We include returns on assets (ROA) lagged by one year to measure firms’ profitability, and expect a negative coefficient on this variable. 2) Market condition: We include a firm’s stock performance (RETURN) as a control variable because according to the market timing theory, firms are likely to issue additional equity when their stocks are performing well in the market (Baker and Wurgler 13

As a test for sensitivity, we measure equity issue as net amount, that is, the sale of common and preferred stock (Compustat #108) minus the purchase of common and preferred stock (Compustat #115). Our inferences remain the same, although the number of firms issuing stocks based on the new definition decreases from 288 to 234.

14

2002). Thus, the coefficient on stock market returns (RETURN) lagged by one period is expected to be positive. 3) Tangibility: Firms with large tangible assets are more likely to issue debt over equity because tangible assets can be used as collateral for debt (Chang et al. 2009). The coefficient on PPE is expected to be negative. 4) Target leverage: Firms are expected to adjust their leverage to converge to the industry median debt ratio. This suggests that changes in debt ratios will be related to whether a company’s debt ratio is above or below the industry target ratio (Flannery and Rangan 2006). DEVTGT is measured by difference between the leverage ratio of a company and the median value of leverage ratios of companies in the same two-digit SIC industry. We expect a positive coefficient of DEVTGT if a firm adjusts its leverage toward its target ratio. 5) Future Growth Opportunities: Firms with more opportunities for future growth tend to issue more equity than debt because they want to preserve their borrowing capacity for the future, and preserve their financial flexibility (Kayhan and Titman 2007). The change in sales is used to represent a firm’s growth prospects (GROWTH). The market to book value ratio (MB) is included as well, since firms with higher market-tobook ratios are also regarded as having greater growth opportunities (Hovakimian et al. 2001). We expect GROWTH and MB to have positive coefficients. 6) Credit quality: Firms rated by a credit rating agency have greater access to lower cost debt and therefore, are more likely to issue debt (Elliot et al. 2008; Kisgen 2006). We, therefore, expect a negative coefficient of CREDIT.

15

7) Tax benefits: Firms with high marginal tax rates benefit more from the deductibility of interest payments, and therefore are more likely to issue debt over equity. Thus, we expect a negative coefficient on TAXRATE. 8) Auditor assurance: Gillan (2006) argues that independent auditors form a type of external corporate governance structure14 monitoring corporate financial reporting and internal control processes. We expect a positive coefficient on BIGN because the higher level of assurance should help firms to raise equity financing. 9) Financial deficit: We define financial deficit as total financing needs of a company divided by total assets ([debt issued + equity issued] / total assets), following Chang et al (2009). Shyam-Sunder and Myers (1999) find that firms with higher financial deficits (i.e., firms that raise more external capital) tend to issue more debt than equity. 10) Firm size: Prior research finds that larger firms which are viewed as less risky are able to and tend to issue more debt than equity (Chang et al. 2006). The natural logarithm of total asset (SIZE) is used to proxy for firm size, and we expect the coefficient on this variable to have a negative sign. 4. Empirical Results 4.1. Univariate Tests Table 2 presents a comparison of firms issuing stocks with those issuing debt in terms of means, medians, and standard deviations of the variables used in our tests.15 The main test variable, GOV, is statistically significantly bigger for firms issuing stocks (median value=4) when compared to those issuing debt (median value=3). This supports

14

Other mechanisms of external corporate governance include law/regulation, capital markets, and other forms of external oversight such as from the media (Gillan 2006). 15 All continuous variables are winsorized at both 1 percent and 99 percent levels to reduce the effects of extreme values.

16

our hypothesis that firms with equity issuances tend to have better quality corporate governance. We find similar results using the alternative measures of governance: E index (median values are 2 versus 3) and G index (median values are 8 versus 10).16 There are also substantial differences in board governance measures and other firm characteristics. Using the median values again, companies issuing stocks, compared to those issuing debt, tend to have a smaller boards (6 versus 8 members), a greater number of board members serving on the audit committee (43% versus 36%), and a higher ratio of institutional ownership (79% versus 72%). These comparisons provide results that are generally consistent with our expectations. Firms issuing debt have on average (based on median values) more than twice the percentage of total assets consisting of PPE (27% versus 13%), are almost always rated by a credit rating agency (100% versus 0%), and face higher tax rates (35% versus 32%). Other results worth noting are that firms issuing equity are smaller; have higher growth prospects and higher market returns; and have lower leverage. All of the above results are consistent with our predictions with an exception of DEVTGT. Interestingly, firms issuing equity are less likely to use a Big N auditor which is a surprising result because we should expect Big N auditors to provide a higher level of assurance and improved governance. In the multivariate analysis discussed later, however, the coefficient of BIGN loses its statistical significance. (Insert TABLE 2 Here) 4.2. Regression Results In Table 3, we present results of four logistic models, each of which uses a different proxy for governance: 1) composite measure 2) E index, 3) G index, and 4) 16

Note with G and E indices lower values mean better governance.

17

individual components that make up the composite measure, respectively. All models are statistically significant as seen from the Wald Chi-square values; the Pseudo-R2 s for these models are similar, about 24 to 25%.17 In the first column, results show that the coefficient on the main test variable, GOV, is positive and statistically significant at the 1% level. Similar significant results are obtained using the E and G indices as governance proxies. These findings suggest that firms with a higher quality of governance are more likely to choose equity over debt. Importantly, the results using the E and G indices are not consistent with studies suggesting that entrenched managers will prefer issuing equity over debt to avoid being bonded to pay fixed charges each period (e.g., Hovakimian et al. 2001; Jensen 1986; Berger et al. 1997). The last column of Table 3 shows the results when individual components of our corporate governance proxy (GOV) are separately included. We find that board size (DNUMBD), the ratio of independent members on boards (DRINBD), and the G index (DGINDEX) are significantly related to the dependent level at least at the 5 % level, while the coefficient on institutional ownership (D%INOWN) approaches statistical significance. These results support prior research showing that board governance is important to investors in providing external financing to firms.18

17

We do not believe multicollinearity to be a problem because we find that the highest variance inflation factor in OLS regression formats is only 2.42, which is well below 10.00, the level where concern arises (Belsley et al. 1980). 18 Because the role of board governance is to reduce information asymmetry and provide monitoring of management’s actions, finding statistical significance on the board variables increases our confidence that a reduction in information asymmetry is why firms with strong governance choose equity over debt financing. In contrast to board governance, the G (and E) index only proxies for the broad distribution of powers between managers and shareholders, which albeit indirectly reduces information asymmetry.

18

We also compute the marginal effect of GOV (2.26), which indicates that a change in one unit of GOV increases the likelihood of issuing equity by 2.26%.19 To put this number in perspective, the unconditional probability of issuing equity in our sample is about 14% as shown Panel B of Table 1. Therefore, a one unit increase in GOV accounts for about 16% of the unconditional probability that a firm will issue equity than debt. With respect to the control variables, the probability of issuing equity is higher for firms with growth opportunities, smaller firms and those with lesser tangible assets, firms with lower leverage relative to its peers in the same industry, those with lower tax rates and lower financial deficits, and firms not rated by a credit rating agency. These variables are significantly related at the 1 % level, and the signs on these variables are generally consistent with those found by prior studies. An exception is the coefficient of DEVTGT which has a negative sign, contrary to our prediction that firms with higher leverage compared with the median leverage of their industry tend to issue more equity in order to move towards their optimal level of leverage. However, this result is consistent with Lemmon et al. (2008) who suggest that high (low) leveraged firms maintain a high (low) level of leverage for a long time, spanning over 20 years. Because the adjustment towards the median is not immediate, a negative sign can be expected. All of the above results generally hold, regardless of which governance proxy is used. (Insert TABLE 3 Here) 4.3. Small versus Large Firms

19

The marginal effect is defined as the change in the estimated probability of issuing equity corresponding to a unit change in a variable, holding all other variables constant at their sample mean values.

19

We also test whether the relationship of equity issuance and corporate governance is different for firms that differ by size. We conjecture that small firms should benefit most from investing in high quality governance systems because these firms have a higher degree of information asymmetry and bigger agency problems. To test this we interact each of our test variables (GOV, E index, G index, and individual components of GOV) with a dummy variable SSIZE which is coded 1 if a firm is smaller than the median size of the sampled firms, and 0 otherwise. The results of these tests are shown in Table 4. As the first column shows, we find an insignificant coefficient on our main test variable GOV, while the coefficient on the interaction variable between SSIZE and GOV is positive and statistically significant at the 1 % level. As seen in the second and third column, we find similar results using the different proxies of corporate governance. The last column shows that as far as the governance variables go, the only coefficients that are statistically significant at the 5% level are S_DNUMBD and S_DGINDEX. This confirms our conjecture that the impact of high quality corporate governance on financing policy is more pronounced in small firms. (Insert TABLE 4 Here) 4.4. Magnitude of Equity Issues Next, we document OLS regression results using the magnitude of equity issuances on the corporate governance and control variables.20 We define issue size (the dependent variable) as the amount of equity issued as a percentage of the total assets at the beginning of the year. All models are statistically significant and explain more than 30 % of the variation in the dependent variable. As shown in the first column, the

20

For this test, we restrict our sample to firms having a positive number for Compustat item # 108, the sale proceeds from issuances of common and preferred stocks.

20

coefficient on main test variable GOV is statistically significant at the 1% level. Using E and G indices we find similar results, although the significance on the test coefficients, EINDEX and GINDEX is lower. In the last column, we find the coefficients on the three governance variables, DNUMBD, D%INOWN, and DGINDEX, are statistically significant at least at the 10% level. These results suggest that firms issue more stocks as the size of board decreases, percentage of institutional ownership increases, and managers are less entrenched. The results for the control variables are generally similar to those documented previously with the exception of FINDEF, which has positive coefficient. This result is plausible because firms with higher needs for external financing are likely to raise funds in larger amounts regardless of the type of security. (Insert TABLE 5 Here) 4.5. Additional Analysis 4.5.1. CEO variables The power wielded by the CEO impacts the effectiveness of a firm’s corporate governance.21 The effectiveness of a firm’s corporate governance is diminished when the current or former CEO is also the chairman of the board (CEOCHAIR), when CEOs have long tenure (CEOTEN), when CEOs have high equity ownership (CEOOWN), and when CEOs are old (CEOAGE).22 We include these four CEO power variables to test whether there is any incremental effect of these variables on the choice of equity versus debt financing. If CEO power negatively affects the quality of a firm’s corporate governance, we should expect to find negative coefficients on these power variables.

21 22

We obtain CEO variables from the Investor Responsibility Research Center (IRRC) database. See Imhoff (2003); Bebchuk et al. (2009b); Bizjak et al. (2009).

21

Untabulated univariate analyses show that equity financing is positively correlated with CEOOWN, negatively correlated with CEOCHAIR and CEOAGE, and insignificantly correlated with CEOTEN. Thus, only two variables, CEOCHAIR and CEOAGE, have the expected signs in our univariate analysis. In our multivariate analysis (untabulated), we find that only CEOAGE is negatively and significantly related with the dependent variable at the 10% level. When we include individual components of GOV, we find that none of CEO variables are significant. The results on the variables of main interest are unchanged in these regressions. Therefore, we conclude that the role of the CEO is marginal in a firm’s choice of debt versus equity financing. 5. Conclusion This paper examines whether good corporate governance plays a role in influencing a firm’s choice of financing policy, specifically long-term debt versus equity. We hypothesize that as corporate governance becomes stronger, firms tend to issue equity more frequently and in greater amounts. This is because strong governance reduces information asymmetry between fund users and fund providers. While this is true for both equity and debt financing, we argue that a more pronounced effect of reduced information asymmetry occurs with equity financing. Our findings are consistent with this prediction. We find in univariate and multivariate tests that firms with strong governance show a preference for equity when compared to debt. As the corporate governance score increases from 0 to 6, the percentage of firms choosing equity over debt increases monotonically from 4% to 41%. This strong result is confirmed in logistic regressions which show that the likelihood of equity issuances increases with the quality of corporate governance. These regressions

22

also show that in addition to shareholder rights, board governance (independence and board size) plays an important role in reducing information asymmetry for investors providing firms with equity financing. As a check for robustness, we examine whether there is any differential effect of corporate governance on financing policy according to different levels of information asymmetry. In support of our hypothesis, we find that the link between governance and preference for equity financing is more pronounced in small firms, which we expect to benefit the most from a reduction in information asymmetry. We also find that the amount of equity financing increases with corporate governance quality. Interestingly, our results do not show that CEO power affects a company’s financing policy. Prior evidence showing the ability of cross-sectional variation in governance structures for explaining firm value and/or firm decision making has been relatively weak (Richardson 2006). By demonstrating a significant relationship between governance and financing policy, our study makes a positive contribution to this area of the literature. Our results represent an interesting contrast to a popular theory in the literature, the pecking order hypothesis, which suggests that managers prefer debt to equity financing in the face of information asymmetry. Klein et al. (2002) state the pecking order theory is the only asymmetric information theory of capital structure that has consistently been studied by the literature. As discussed in section 2, however, there is mounting evidence that the pecking order theory provides only a partial explanation for a firm’s financing policy. Our findings are that a non-trivial percentage of firms (14%) choose equity over debt, and further that the number of firms choosing equity increases as the quality of corporate governance also increases. This positive relationship is not a prediction that can

23

be attributed to the pecking order theory. We provide an alternative theory, namely that strong governance lowers information asymmetry down to a level where issuing equity can become more attractive to debt financing for some firms. We believe the results of this study shed new light on the important topic of corporate financing policy and capital structure.

24

REFERENCES

Ajinkya, B., S. Bhojraj, and P. Sengupta. 2005. The association between outside directors, institutional investors and the properties of management earnings forecasts. Journal of Accounting Research 43(3): 343 - 376 Anderson, R. C., S. A. Mansi, and D. M. Reeb. 2004. Board characteristics, accounting report integrity, and the cost of debt. Journal of Accounting and Economics 37: 315–342. Ashbaugh, H., D. W. Collins, and R. LaFond. 2004. Corporate governance and the cost of equity capital. Working paper, University of Wisconsin-Madison and University of Iowa. Ashbaugh-Skaife , H., D. W. Collins, and R. LaFond. 2006. The effects of corporate governance on firms’ credit ratings. Journal of Accounting and Economics: 203-243. Baker, M., and J. Wurgler. 2002. Market timing and capital structure. Journal of Finance 57: 1-32. Bates, T. W., D. A. Becher, and M. L. Lemmon. 2008. Board classification and managerial entrenchment: Evidence from the market for corporate control. Journal of Financial Economics 87: 656–677. Bauguess, S. W., and M. Stegemoller. 2008. Protective governance choices and the value of acquisition activity. Journal of Corporate Finance 14: 550-566. Bebchuk, L., A. Cohen, and A. Ferrell. 2009a. What matters in corporate governance? The Review of Financial Studies 22(2): 783-827. Bebchuk, L., Y. Grinstein, and U. Peyer. 2009b. Lucky CEOs and lucky directors. Journal of Finance, forthcoming. Belsley, D. A., E. Kuh., and R. E. Welsch. 1980. Regression Diagnostics: Identifying Influential Data and Sources of Collinearity. New York, NY: John Wiley. Berger, P. G., E. Oferk, and D. L. Yermack. 1997. Managerial entrenchment and capital structure decisions. The Journal of Finance 52 (4): 1411-1438. Bhojraj, S., and P. Sengupta. 2003. Effect of corporate governance on bond ratings and yields: the role of institutional investors and the outside directors. The Journal of Business 76: 455-475. Bizjak, J., M. Lemmon, and R. Whitby. 2009. Option backdating and board interlocks. 25

Review of Financial Studies 22(11): 4821-4847. Blue Ribbon Committee. 1999. Report and recommendations of the Blue Ribbon Committee on improving the effectiveness of corporate audit committees. New York: New York Stock Exchange and National Association of Securities Dealers. Boot, A. W., and A. V. Thakor. 1993. Security design. Journal of Finance 48: 1349-1378. Borokhovich, K., R. Parrino, and T. Trapani. 1996. Outside directors and CEO selection. Journal of Financial and Quantitative Analysis 31 (September): 337-355. Chang, X, S. Dasgupta, and G. Hilary. 2006. Analyst coverage and financing decisions. The Journal of Finance 61 (6): 3009-3048. Chang, X., S. Dasgupta, and G. Hilary. 2009. The effect of auditor quality on financing decisions. The Accounting Review 84(4): 1085-1117. Cheng, S. 2008. Board size and the variability of corporate performance. Journal of Financial Economics 87: 157-176. Cohen, J., G. Krishnamoorthy, and A. Wright. 2004. The corporate governance mosaic and financial reporting quality. Journal of Accounting Literature 23: 87-152. Coombes, P., and M. Watson. 2000. Three surveys on corporate governance. The McKinsey Quarterly 4: 74-77. Cremers, M., and V. Nair. 2005. Governance mechanisms and equity prices. Journal of Finance 60: 2859-2894. Dechow, P. M., R. G. Sloan, and A. P. Sweeney. 1996. Causes and consequences of earnings manipulation: An analysis of firms subject to enforcement actions by the SEC. Contemporary Accounting Research 13(1): 1-36. DeFond, M. L., R. Hann, and X. Hu. 2005. Does the market value financial expertise on audit committees of boards of directors? Journal of Accounting Research 43(2): 153-193. Denis, D. J., D. K. Denis, and A. Sarin. 1997. Ownership structure and top executive turnover. Journal of Financial Economics 45: 193-222. Dittmar, A., and J. Mahrt-Smith. 2007. Corporate governance and the value of cash holdings. Journal of Financial Economics 83: 599-634. Elliott, W. B., J. Koeter-Kant, and R. S. Warr. 2008. Market timing and the debt-equity choice. Journal of Financial Intermediation 17: 175-197. Flannery, M. J., and K. P. Rangan. 2006. Partial adjustment toward target capital structures. Journal of Financial Economics 79: 469-506. 26

Frank, M., and V. Goyal. 2000. Testing the pecking order theory of capital structure, Working Paper, University of British Columbia and Hong Kong University of Science and Technology. Fulghieri, P., and D. Lukin. 2001. Information production, dilution costs, and optimal security design. Journal of Financial Economics 60: 187-243. Gillan, S. L. 2006. Recent developments in corporate governance: An overview. Journal of Corporate Finance 12: 381-402. Goh, B. 2009. Audit committees, boards of directors, and remediation of material weaknesses in internal control. Contemporary Accounting Research 26(2): 549-579. Gompers, P. A., J. L. Ishii, and A. Metrick. 2003. Corporate governance and equity prices. Quarterly Journal of Economics 118: 107-155. Graham, J. R. 1996. Proxies for the corporate marginal tax rate. Journal of Financial Economics 79: 469-506. Hovakimian, A., T. Opler, and S. Titman. 2001. The debt-equity choice. Journal of Financial & Quantitative Analysis 36: 1-24. Imhoff, E. A. 2003. Accounting quality, auditing, and corporate governance. Accounting Horizons 17(Supplement): 117-128. Jensen, M. C. 1986. Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76: 323-339. Jensen, M. C. 1993. The modern industrial revolution, exit and the failure of internal control systems. Journal of Finance 48: 831–880. Jensen, M. C., and W. H. Meckling. 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3: 305-360. John, K., and L. Litov. 2009. Corporate governance and financing policy: New evidence. Working paper, New York University and Washington University. Kayhan, A., and S. Titman. 2007. Firms’ histories and their capital structures. Journal of Financial Economics 83: 1-32. Kisgen, D. 2006. Credit ratings and capital structure. Journal of Finance 41: 1035-1072. Klein, L. S., T. J. O’Brien, and S. R. Peters. 2002. Debt vs. equity and asymmetric information: A review. The Financial Review 37: 317-350.

27

Klock, M., W. Maxwell, and S. Mansi. 2005. Does corporate governance matter to bondholders? Journal of Financial and Quantitative Analysis 40: 693–719. Korajezyk, R. A., D. J. Lucas, and R. L. McDonald. 1991. Firm value and seasoned equity issues: Price pressure, wealth redistribution, or negative information. Review of Financial Studies 4: 685-708. La Porta, R., F. Lopez-de-Silanes, and A. Shleifer. 1998. Law and finance. Journal of Political Economy 106(6): 1113-1155. Lemmon, M. L., M. R. Roberts, and J. F. Zender. 2008. Back to the beginning: Persistence and the cross-section of corporate capital structure. The Journal of Finance 63 (4): 1575-1608. Lipton, M., and J. Lorsch. 1992. A modest proposal for improved corporate governance. Business Lawyer 48: 59–77. Myers, S. C., and N. S. Majluf. 1984. Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics 13: 187-221. Richardson, S. 2006. Over-investment of free cash flow. Review of Accounting Studies 11: 159-189. Shleifer, A., and R. Vishny. 1997. A survey of corporate governance. Journal of Finance 52: 737-775. Shyam-Sunder, L. and S. C. Myers. 1999. Testing static tradeoff against pecking order odels of capital structure. Journal of Financial Economics 51: 219-244. Warga, A., and I. Welch. 1993. Bondholder losses in leveraged buyouts. Review of Financial Studies 6: 959-982. Weisbach, M. 1988. Outside directors and CEO turnover. Journal of Financial Economics 20(January-March): 431-460. Wu, Z., J. H. Chua, and J. J. Chrisman. 2007. Effects of family ownership and management on small business equity financing. Journal of Business Venturing 22: 875895. Yermack, D.1996. Higher market valuation for firms with a small board for director. Journal of Financial Economics 40: 185-211. Zhang, Y., J. Zhou, and N. Zhou. 2007. Audit committee quality, auditor independence, and internal control weaknesses. Journal of Accounting and Public Policy 26: 300-327.

28

TABLE 1 Sample Distribution Panel A: Sample distribution by year Year 1998 1999 2000 2001 2002 2003 2004 2005 2006 Total

# Issuing Equity 22 32 39 30 23 27 40 39 36 288

# Issuing Debt 234 221 191 246 167 168 155 178 201 1,761

Total Firms 256 253 230 276 190 195 195 217 237 2,049

Panel B: Sample distribution by governance index

Governance Index 0 1 2 3 4 5 6 Total

Total Firms 27 69 286 626 591 318 132 2,049

# Issuing Equity 1 3 20 66 70 74 54 288

29

Percentage 0.037 0.043 0.069 0.105 0.118 0.233 0.409 0.141

TABLE 2 Univariate Comparisons of Variables Used in our Tests

Variables GOV GINDEX EINDEX NUMBD RINDBD RACBD RINDAC %INOWN SIZE GROWTH MB ROA RETURN PPE DEVTGT CREDIT TAXRATE BIGN FINDEF

Companies Issuing Equity (288) Mean 4.2188 8.2292 2.1215 6.5174 0.7683 0.4392 0.9523 0.7469 6.7627 0.2552 6.0324 0.0528 0.0163 0.1854 -0.0515 0.2292 0.1965 0.9514 0.0689

Median 4.0000 8.0000 2.0000 6.0000 0.7778 0.4286 1.0000 0.7873 6.6233 0.1995 4.6156 0.0759 0.0165 0.1281 -0.0834 0.0000 0.3202 1.0000 0.0506

Companies Issuing Debt (1,761)

Std 1.2732 2.4643 1.1548 2.4707 0.1808 0.1905 0.1384 0.1794 1.2714 0.2808 5.0890 0.1341 0.0435 0.1602 0.2188 0.4210 0.1617 0.2154 0.2477

Mean 3.4355 9.5014 2.6292 7.8325 0.7793 0.3660 0.9469 0.7058 7.6784 0.1061 3.1737 0.0535 0.0049 0.3368 0.0383 0.6689 0.2290 0.9852 0.0522

•

Median 3.0000 10.0000 3.0000 8.0000 0.8000 0.3636 1.0000 0.7221 7.5170 0.0789 2.2452 0.0560 0.0065 0.2709 0.0195 1.0000 0.3498 1.0000 0.0196

Std 1.1916 2.6140 1.1796 2.5915 0.1916 0.1709 0.1343 0.1730 1.3247 0.2030 3.5608 0.0657 0.0328 0.2345 0.1861 0.4707 0.1485 0.1206 0.1979

Test of differences of means t-statistic 10.02*** -7.72*** -6.79*** -8.04*** 0.91 6.63*** 0.63 3.72*** -10.94*** 10.89*** 11.80*** -0.13 5.24*** -10.56*** -7.39*** -14.91*** -3.40*** -3.86*** 1.28

t- values are based on two-tailed tests. *, **, and *** represent statistical significance at the 10%, 5%, and 1% levels, respectively. GOV Corporate governance index developed by DeFond et al. (2005); GINDEX G index developed by Gompers et al. (2003); EINDEX E index developed by Bebchuk et al. (2009a); Number of board members; NUMBD Ratio of independent to full board members; RINDBD Ratio of audit committee to full board members; RACBD Ratio of independent audit committee members to total audit committee members; RINDAC Institutional owner ownership; %INOWN SIZE Natural logarithm of total assets; GROWTH Changes in sales, deflated by sales at the beginning of the fiscal year; MB Market value of equity deflated by book value of equity; ROA Net income before extraordinary items deflated by total assets at the beginning of the fiscal year. RETURN Buy-and-hold annual stock return; PPE Ratio of fixed assets to total assets; DEVTGT Difference between the leverage ratio and the target debt ratio measured by the median values of two-digit industry; CREDIT 1 if the firm has been assigned a credit rating by S&P, and 0 otherwise. TAXRATE Marginal tax rates provided by John Graham (Graham 1996); BIGN 1 if the firm is audited by a Big N auditor, and 0 otherwise; and FINDEF Total financing needs (net debt issued + net equity issued) divided by total assets.

30

TABLE 3 Results of Logistic Regressions: Equity versus Debt Financing

Variable

INTERCEPT GOV EINDEX GINDEX DNUMBD DRINDBD DRACBD DRINDAC D%INOWN DGINDEX SIZE GROWTH MB ROA RETURN PPE DEVTGT CREDIT TAXRATE BIGN FINDEF Wald ChiSquare Pseudo-R2 N

•

Expected Sign

+/+ + + + + + + + + + + + -

Logistic Model Using Governance Index Coeffici ent -0.1943 0.3125

-0.2941 2.3718 0.1978 -1.6166 2.9955 -3.5981 -2.2691 -1.1118 -1.9530 -0.5907 -1.0890 319.33 0.2437 2,049

p-value 0.8131