Printer Friendly

Corporate governance and dividend payout policy: a test using antitakeover legislation.

Managers strongly prefer not to pay dividends as dividend payouts reduce the amount of cash subject to managerial discretion (Easterbrook, 1984; Jensen, 1986). Previous empirical tests of the relationship between corporate governance and dividend payout policy employ endogenous measures of this agency problem. Using a relatively exogenous measure that incorporates state antitakeover laws and the differences-in-differences approach, our analysis indicates that dividend payout ratios and propensities fall when managers are insulated from takeovers. The impact of antitakeover laws on dividend payouts is more pronounced for firms with poor corporate governance and small firms.

**********

Finance academics have long wondered why firms pay dividends even though cash distributions (such as dividends) are less advantageous from a tax perspective than cash retention or stock repurchases (Black, 1976). Despite extensive literature regarding dividend payout policy, researchers have not yet reached a consensus about what determines dividend payout ratios. Theories based on the signaling hypothesis (Aharony and Swary, 1980; Khang and King, 2006), tax clienteles (Michaely, Thaler, and Womack, 1995), or catering (Baker and Wurgler, 2004) are not always consistent with the existing empirical results, though Allen and Michaely (2003) provide detailed surveys of the existing work on payout policy.

Another theory, which researchers have given more attention to recently (Easterbrook, 1984; Jensen, 1986; Hart and Moore, 1994; Zwiebel, 1996; Fluck, 1998, 1999; Myers, 2000), is that agency problems between corporate insiders and outside shareholders have an impact on dividend payout policy. According to this theory, managers prefer to reduce or even cut dividend payouts since paying dividends reduces the amount of cash at the managers' disposal and this, in turn, exposes companies to potential external financial needs to encourage capital market monitoring and discourages inefficient investments.

However, empirical evidence regarding the relationship between corporate governance and dividend payout policy is mixed. (1) The possible problem that arises when analyzing the effect of corporate governance on dividend payout policy is endogeneity. Specifically, firms with higher dividend payouts may have stronger firm-level corporate governance because their managers do not have enough cash flow to make inefficient investments and they have to go through monitoring in capital markets to obtain external financing. Moreover, firms with various corporate governance approaches probably have differences in other unobservable dimensions that cannot be easily controlled for (Bertrand and Mullainathan, 2003). This may explain why it is difficult for previous studies to test the relationship between corporate governance and dividend payout policy.

In this paper, we attempt to examine the impact of corporate governance on dividend payout policy using an exogenous measure of variations in takeover pressure: the passage of antitakeover laws by various states in the 1980s. The key advantage of this natural experimental setting is that we can avoid the drawback of endogenous measures of corporate governance in the existing dividend payout policy literature. Additionally, we can detect not only the correlation between corporate governance and dividend payout policy but also the causal relation between the two.

In the mid to late 1980s, a number of states in the United States passed second-generation antitakeover legislation that raised the barriers to launching takeover bids on firms incorporated in those states. The enactment of these laws is widely viewed as an exogenous shock that weakens corporate governance as managers worry less about the possibility of being replaced by successful bidders and become more entrenched (Garvey and Hanka, 1999; Bertrand and Mullainathan, 2003; Cheng, Nagar, and Rajan, 2005). Similarly, under the antitakeover laws, shareholders might not be able to sell shares to potential hostile raiders who then gain control over non-dividend-paying companies. As a result of increased managerial entrenchment following the enactment of these laws, we speculate that managers of firms incorporated in these states may reduce or even cut dividend payouts to serve their own interests.

We test this prediction on a large number of US industrial firms in from 1981 to 1993. First, univariate analysis indicates that, on average, dividend payout ratios and dividend propensities decrease following the passage of antitakeover laws. Next, using a differences-in-differences approach, we find that dividend payout ratios also significantly decrease by 1.7% after the enactment of antitakeover legislation. Additionally, we observe that the likelihood of dividend payouts significantly declines by 8.9% in response to antitakeover laws. Finally, the effect of passing antitakeover legislation is more pronounced for firms with poor internal corporate governance and small firms.

We conduct a series of robustness tests on the results. First, our analyses indicate that total payout ratios (including dividends and share repurchases) decline after managers are insulated from takeovers. This suggests that our main results are robust even if we consider the effect of stock repurchases as substitutes for dividends.

Another possible explanation for our results is that firms intentionally pay too much in dividends prior to the passage of antitakeover legislation. They may do this due to takeover pressures induced by a market with a short-term orientation. As such, they reduce or even cut their dividend payouts after the passage of these laws for the sake of their shareholders. To address this concern, we demonstrate that the effects of passing antitakeover laws are more pronounced for firms with high cash flows and low growth opportunities. It is less likely that such firms were underinvesting before the law changes. Thus, this alternative explanation cannot drive our results.

Additionally, we further examine where managers employ the cash that is not distributed to shareholders. The existing literature reveals that one possible destination is increased chief executive officer (CEO) and employee compensation (Bertrand and Mullainathan, 1999a, 1999b, 2003). Another possible use of funds is reduction in leverage (Garvey and Hanka, 1999). There is no evidence that managers use the unpaid cash for investments and acquisitions (Hackl and Testani, 1988; Bertrand and Mullainathan, 2003). Given that, we further demonstrate how firms reserve additional cash after the law changes. Taken together, we conclude that managers do not use leftover cash for investments or to maximize shareholder wealth. They use it to increase their own compensation and that of their employees, to reduce their firms' leverage, and reserve the excess cash for use under their discretion.

Finally, we argue that our results are not driven by dividend payout restrictions imposed by state laws. We also confirm that our results still hold for firms with less than the median value of leverage, whose debt holders are less likely to impose restrictive covenants. Therefore, the additional covenants imposed by debt holders after a change in legislation are not responsible for our main results. We also confirm that our results are not driven by a specific sample and variable construction method. Additionally, by examining the time trend in our estimations, we determine that the endogenous timing of regulatory reform does not drive our results.

Collectively, our results demonstrate the significant role that corporate governance plays in managers' dividend payout policy decisions, confirming the theoretical analyses of Easterbrook (1984) and Jensen (1986). Our results also suggest a departure from the frictionless world based on the assumptions of the Modigliani-Miller (1958) theorem. Specifically, insiders may prefer to keep cash flow inside firms and get some private benefit from it instead of allocating all the profits to shareholders.

This research contributes to the literature on dividend payout policy and corporate governance in several important ways. First, the existing literature only focuses on the effects of several dimensions of firms' corporate governance mechanisms such as board structure and institutional ownership on dividend payout policy. We extend this stream of literature by examining the effect of one of the firms' external governance mechanisms, the corporate control market on dividend payout decisions. This gives us greater insight regarding the role of corporate governance in firms' dividend payout policy. Additionally, we complement these studies by introducing an exogenous measure, the passage of antitakeover laws. By doing so, we can avoid the drawback of endogenous measures of corporate governance in the existing dividend payout policy literature. Moreover, we can detect not only the correlation between corporate governance and dividend payout policy but also the causal relationship between the two. Finally, the existing literature indicates that when managers are insulated from takeovers, they begin to enjoy a "quiet life" (Bertrand and Mullainathan, 2003) and increase their own compensation (Bertrand and Mullainathan, 1999b). We complement this stream of literature by documenting the effect of takeover pressure on firms' dividend payout policy. This gives us a comprehensive picture as to how the corporate control market influences management's behavior.

The remainder of this paper is structured as follows. Section I describes the institutional aspects of antitakeover laws and develops our hypotheses. Section II describes our sample, variable definitions, and descriptive statistics. The empirical results are presented in Section III, while Section IV reviews a series of robustness tests. We discuss our conclusions in Section V.

I. Related Literature and Hypothesis Development

A. Description of Antitakeover Laws

In this paper, the main proxy for change in corporate governance is the passage of antitakeover legislation. Thus, we begin with a brief description of antitakeover laws. More detailed discussions can be found in several prior studies (Karpoff and Malatesta, 1989; Bertrand and Mullainathan, 2003).

Although federal intervention in mergers and acquisitions was based largely on various antitrust laws and on the 1968 Williams Act intended to protect investors from unannounced takeovers, the first-generation antitakeover laws were passed in many states but deemed unconstitutional in 1982. To address these concerns, states passed the second-generation laws in the 1980s. The third-generation laws of the 1990s test current limits, attempting to make antitakeover legislation more stringent.

In our study, we focus on second-generation state antitakeover laws. Three types of laws in the second-generation legislation exist: 1) control share acquisition laws, 2) fair price laws, and 3) business combination laws. Because the Supreme Court declared the first-generation statutes unconstitutional, constitutionality was an important concern for the second-generation legislation. Thus, many states imitated each others' existing laws leading to legislation, although the laws differ in some dimensions in various states. This allows us to consider these laws as almost homogenous for the purposes of conducting our empirical tests.

B. Impact of Laws

Antitakeover laws have played an important role in deterring takeovers. However, antitakeover laws have also increased agency costs (Jensen, 1986). The existing literature (Karpoff and Malatesta, 1989; Morck, Shleifer, and Vishny, 1990; Francis et al., 2010; Bertrand and Mullainathan, 1999a, 1999b, 2003) contains outstanding research using the passage of these laws as a testing ground. They argue that the laws increase the power of the existing managers. These are managers who do not maintain operational efficiency, who lose their outside market discipline, and who increase agency costs. Although some scholars, such as Stein (1988), have questioned whether agency cost increases after antitakeover laws are enacted, Cheng et al. (2005) contend that Stein (1988) arguments are debatable as they do not consider the firm's possible loss of control to a raider. Therefore, the passage of these laws provides a natural exogenous event that is associated with weakened corporate governance for the protected firms.

C. Main Hypotheses

The existing literature demonstrates the negative effects of the passage of antitakeover laws matters when examining firms' investment and compensation policies. For example, when managers are insulated from takeovers, they begin to enjoy a "quiet life" (Bertrand and Mullainathan, 2003) and increase their own compensation (Bertrand and Mullainathan, 1999b). In terms of their dividend payout policy, entrenched managers prefer to reduce or even cut dividend payouts since paying dividends is a return of corporate earnings to investors. These become earnings that are no longer available for the managers' own benefit. Additionally, paying dividends exposes companies to potential external financial needs and capital market monitoring in the future. This gives investors an opportunity to discourage inefficient investments (Easterbrook, 1984; Jensen, 1986). Thus, when managers are insulated from takeovers and become more entrenched, they may choose to reduce dividend payouts. We summarize these predictions in Hypothesis 1.

Hypothesis 1: All else being equal, after managers are insulated from takeovers, they reduce their firms' dividend payout ratios.

Managers typically decompose the dividend payout decision into two steps (Lintner, 1956; Kumar and Lee, 2001). First, they choose whether to pay. If they decide to pay a dividend, they then decide the level of the dividend payout. Thus, when managers are insulated from takeovers and become more entrenched, they may also choose not to pay dividends (although omitting dividend payouts altogether is a very strong action). We summarize these predictions in Hypothesis 2.

Hypothesis 2: All else being equal, after managers are insulated from takeovers, they are less inclined to pay dividends.

There are a variety of governance mechanisms in addition to takeover pressures such as cumulative voting and secret ballots. If a firm's internal governance mechanisms are poor, the passage of antitakeover laws should have a greater effect on dividend payout policy. We summarize these predictions in Hypothesis 3.

Hypothesis 3: All else being equal, the effect of the passage of antitakeover legislation on dividend payout is more pronounced when the firm has poor internal corporate governance.

The passage of these laws may have different impacts on different firms. Obviously, small firms have more exposure to takeover threats because size itself seems to be a natural or incidental deterrent. For example, Fields and Todd (1995) find that small firms display a larger abnormal return in the presence of state antitakeover legislation than large firms. In line with this thought, our hypothesis is formalized as below:

Hypothesis 4: All else being equal, the effect of the passage of antitakeover legislation on dividend payout policy is more pronounced for small firms.

II. Data

A. Sample Selection and Variable Definition

Our sampling procedure is following Fama and French (2001). Specifically, we include only NYSE, NASDAQ, and AMEX firms that are incorporated in the United States and that have securities with CRSP share code 10 or 11. We exclude financial and utility firms, defined as firms with SIC codes outside the intervals 4900-4949 and 6000-6999, and firms with book equity below $250,000 or total assets below $500,000. We focus on our sample firms for calendar year t (1981-1993) and include those firms with fiscal year ends in t.

Firms' states of incorporation come from Compustat. (2) Bertrand and Mullainathan (2003) employ the passage of business combination laws to represent the passage of antitakeover laws over time. They argue that business combination laws have the most impact in terms of defending companies against outside hostile raiders, as Karpoff and Malatesta (1989) find significant negative reactions to the passage of business combination laws, less negative and insignificant responses to fair price laws, and no reaction to the adoption of control share acquisition laws. Therefore, we define the dummy variable Treat as 1 if a firm is incorporated in a state that passed business combination laws and 0 otherwise. We also define the dummy variable PostLaw as 1 for certain companies in the years in which business combination laws exist and 0 otherwise. (3) We multiply these two dummy variables and construct the interaction term Treat * PostLaw, which is the key variable and captures the effect of the passage of antitakeover laws on firms' dividend payout policy. Table I summarizes the passage year of the business combination laws in various states.

We derive the dependent and control variables based on Compustat data, and the item number is in parentheses. Our dependent variables include Dividend Dummy, Dividend Payout Ratio, and Total Payout Ratio constructed on firm-year bases. Dividend Dummy is equal to 1 if a firm has positive dividends (21) in the fiscal year that ends in year t and 0 otherwise. Dividend Payout Ratio is defined as dividends (21) scaled by earnings (18). Total Payout Ratio is defined as the sum of dividends (21) and share repurchases (115-56), scaled by earnings (18). To construct these variables, we only include firms with positive earnings. To mitigate the effect of outliers, we eliminate observations with dividend payout ratios greater than 1.

Because only half of our firm-year observations have positive earnings, we define the variable Dividend Payout Ratio (II) as dividends as a percentage of the book value of total assets and the variable Dividend Payout Ratio (Ill) as dividends as a percentage of the market value of total assets to conduct the robustness tests. We also define the variable Dividend Payout Propensity as the difference between Dividend Dummy and expected probability of dividend payout. The expected probability of dividend payout is calculated by the formula: Pr (Dividend Dummy = 1) = logit [-0.14 + 4.26 (Firm Size) -0.81 (Market-to-Book Ratio)--1.07 (Growth Rate of Assets) + 15.57 (Profitability)] (Baker and Wurgler, 2002). A higher value means that firms are more likely to pay dividends.

In terms of control variables, Fama and French (2001) identify four firm characteristics that are important determinants of a firm's likelihood to pay dividends: 1) firm size, 2) investment opportunity, 3) growing speed, and 4) profitability. Firm size (the Firm Size variable) is measured as the NYSE market capitalization percentile (i.e., the fraction of NYSE firms having equal or smaller capitalization than firm i in year t). Investment opportunity (the Market-to-Book Ratio variable) is measured as total assets (6) minus book equity plus market equity (199 * 25) divided by total assets (6). Growing speed (the Growth Rate of Assets variable) is measured as the percentage growth in assets (6) from year t-1 to year t. Profitability (the Profitability variable) is measured as earnings before extraordinary items (18) plus interest expense (15) plus income statement deferred taxes (50) divided by assets (6).

DeAngelo, DeAngelo, and Skinner (2004) and Bulan, Subramanian, and Tanlu (2007) find that firms are more likely to pay dividends in the mature stage of their lifecycles. The ratio of retained earnings (36) to total capital (199 * 25), called the Retained Earnings to Total Capital variable, is a proxy for a firm's lifecycle stage and has strong explanatory power when estimating a firm's likelihood to pay dividends. (4) Hoberg and Prabhala (2009) argue that idiosyncratic risk is a more important variable than the dividend premium mentioned in Baker and Wurgler (2004) in determining firms' dividend payout policy. Thus, we construct and control for the variable Idiosyncratic Risk, which is measured as the standard deviation of residuals from a regression of its excess returns (raw returns less the riskless rate) on the Fama and French (200l) factors high minus low (HML), small minus big (SMB), and MKT. (5) The variable Idiosyncratic Risk is expected to be negatively related to dividend payout since managers pay out only what can be sustained by future earnings with a high degree of certainty.

We also control for a series of other variables. The ratio of book value of current and long-term debt (9 + 34) to total assets (6) (the Leverage variable) and the ratio of capital expenditures (128) to total assets (6) (the Investment variable) are expected to be negatively correlated with dividend payout. Firms faced with a higher share of income taxes in earnings before interest and tax (EBIT) are also expected to have lower dividends. We construct the ratio of total income taxes (16) to EBIT (178) (the Taxes variable) to control for this effect. The variable Analyst Forecasts (the natural log of the number of one-year-ahead analyst forecasts of earnings per share (EPS) for the current fiscal year obtained from the I/B/E/S database) captures the level of information asymmetry problems of the firm. (6) Li and Zhao (2008) find that firms that are more subject to information asymmetry are less likely to pay dividends. The firms that reserve more cash are more likely to pay dividends. We use the ratio of cash and marketable securities (1) to total assets (6) (the Cash Holding variable) to control for this.

The existing literature demonstrates that dividend payout policy is related to ownership structure (Zeckhauser and Pound, 1990; Fenn and Liang, 2001; Short, Zhang, and Keasey, 2002). Thus, we obtain institutional ownership data from Thomson Financial (previously known as CDA Spectrum). The variable Institutional Ownership is measured as total ownership of all institutions as a percentage of the firms' total shares outstanding. We obtain managerial ownership data from Compact Disclosure. The variable Managerial Ownership is measured as total ownership by officers and directors as a percentage of the firms' total shares outstanding. We construct the variable Internal Governance to control for the firms' existing internal corporate governance. This variable is measured as 24 minus the G index reported by the IRRC in 1990 after removing the provisions related to state laws. (7) Firms with greater internal governance are expected to pay higher dividends.

To remain in the final sample, each firm-year observation must have the information available to derive the following variables: 1) Dividend Dummy, 2) Firm Size, 3) Market-to-Book Ratio, 4) Growth Rate of Assets, 5) Profitability, 6) Retained Earnings to Total Capital, 7) Idiosyncratic Risk, and 8) Cash Holding. We impose additional Compustat data availability conditions when conducting some of our tests and related analyses. The final sample contains 1,469 firms and an overall total of 11,473 observations over the period 1981-1993. Our sample reduces to 6,127 observations when we examine the impact of law changes on dividend payout ratios. (8)

B. Descriptive Statistics

Table II provides several descriptive statistics. Variable construction details are provided in the Appendix. All independent variables except Treat and Treat * PostLaw are winsorized at the 1% level to address the potential problems with extreme observations.

The mean value of Dividend Dummy is 0.524. This suggests that about half of the firms in our sample are dividend payers. The Dividend Payout Ratio and the Total Payout Ratio variables are censored at 0 from below and at 1 from above with the mean value of 0.284 and 0.342, respectively. The variable Treat has a mean value of 0.858, implying that most of

firms are incorporated in the law-protected states. The variable Treat * PostLaw has a mean of 0.504, indicating the data is relatively evenly spread pre- and post-passage. In Table II, we also report summary statistics of the firm characteristics that are controlled for in our estimations. The results are consistent with similar studies.

III. Empirical Results

In the analysis that follows, we control for firm characteristics and examine whether our proxy for change of corporate governance (passage of antitakeover legislation) has any additional power to explain firms' dividend payout policy. We start by examining dividend payout patterns before and after the passage of antitakeover legislation. Then, we use multivariate analysis to estimate the impact of law change on dividend payout ratios. Additionally, we run logit regressions to estimate the change in the firms' propensity to pay dividends in response to the passage of antitakeover laws. Finally, we repeat our dividend payout ratio regressions for firms with superior and poor corporate governance as well as the subsample with large firms and small firms.

In the multivariate analysis, we use the same differences-in-differences methodology as Bertrand and Mullainathan (2003). Our control group for any given year consists of the firms incorporating in states that did not enact antitakeover laws at that time (even if they passed the laws later). This gives our control group enough firm-year observations, although most of the companies in our sample are incorporated in the states passing laws. The differences-in-differences methodology helps us rule out the influence of factors other than corporate governance that may be associated with governance. The key variable in the multivariate test is Treat * PostLaw. This variable captures the pure effect on firms' dividend payout policy via the enactment of antitakeover legislation.

A. Dividend Payout Pattern around the Adoption of Laws

Table III presents the change in the dividend payout pattern for firms incorporated in states that passed antitakeover legislation for three years prior to and three years after passage of the law. The year of the laws' passage is considered time 0. We observe that the percentage of dividend payers declines after the passage of the legislation. Additionally, our two-sided, two-sample t-test indicates that Dividend Payout Ratio and Dividend Propensity significantly decrease in mean value. These results suggest that passage of antitakeover legislation has a strong impact on firms' dividend payout policy. Typically, dividend payouts decline when managers are insulated from takeovers.

B. Effect of the Passage of Laws on the Dividend Payout Ratio

Although the univariate results provide some evidence regarding the impact of the enactment of antitakeover laws on firms' dividend payout policy, they do not control for other firm characteristics that might be correlated with corporate governance, as well as dividend payout policy. To address the issue, we employ the differences-in-differences approach using a set of firm-level controls. The dependent variable Dividend Payout Ratio equals total dividends scaled by earnings. (9) Firm and year fixed effects are controlled in the estimations.

As reported in Table IV, the coefficient of Treat * PostLaw is always negative and statistically significant. This result is also economically significant. Based on the results reported in the first column of Table IV, Dividend Payout Ratio is reduced by 1.7% when firms are insulated from takeovers. This is comparable to the sample mean value of 28%. Dividend Payout Ratio increases with Market-to-Book Ratio, Retained Earning to Total Capital, Cash HoMing, Analyst Forecasts, Institutional Ownership, and Internal Governance. It decreases with Growth Rate of Assets, Idiosyncratic Risk, Taxes, and Managerial Ownership. The significant negative coefficient of Profitability implies that dividend payments increase in lower proportion as compared to given increases in earnings. The explanatory effect of Firm Size, Leverage, and Investment on Dividend Payout Ratio is not always significant. We also employ Tobit regressions for Dividend Payout Ratio to avoid biases from using ordinary least squares (OLS) estimation with censored variables. As reported in the last column of Table IV, the results have the same sign and similar level of magnitude and significance.

C. Effect of the Passage of Laws on the Dividend Payout Probability

When we examine the effect of antitakeover laws on firms' dividend payout ratios, we eliminate firms with negative earnings from our sample. To make sure that this does not bias our results, we replace the dependent variable Dividend Payout Ratio with Dividend Dummy and conduct the estimations in a larger sample. We use the logit model to predict the likelihood of being a dividend payer. All logit regressions are performed in the panel data format. We control for industry effects (two-digit SIC codes) rather than firm effects since there is no large variation in these propensities among firms during the sample period. (10,11) Table V presents the regression results.

We observe that firms are less likely to issue dividends when managers are more entrenched after the passage of antitakeover legislation. Based on the results reported in the first column of Table V, the decreased probability of dividend payout is around 8.9%, which is economically significant when compared to the 52% mean value of the sample. The result is robust even with more control variables in the regression equation. Dividend payout probability increases with Profitability and Firm Size. It decreases with Market-to-Book Ratio and Growth Rate of Assets. These results are consistent with Fama and French (2001). Dividend payout probability increases with Retained Earnings to Total Capital, similar to DeAngelo, DeAngelo, and Stulz (2006), and decreases with Idiosyncratic Risk, akin to Hoberg and Prabhala (2009). Additionally, dividend payout probability increases with Cash Holding, Institutional Ownership, and Internal Governance. It decreases with Leverage, Investment, and Managerial Ownership. As demonstrated in the last column of Table V, we obtain significant results with the same sign and similar level of magnitude and significance in terms of coefficients when we employ probit regressions as a robustness check.

O. Effect of the Passage of Laws on Firms with Superior and Poor Governance

There are a variety of governance mechanisms besides the state adoption of antitakeover laws. In this subsection, we want to test whether the effect of the passage of antitakeover laws on dividend payout policy are different in firms with superior and poor internal governance. Specifically, we add an interaction term of Treat * PostLaw and a dummy variable, Poor Internal Governance, which is equal to 1 if the firm has lower than the sample median value of Internal Governance and 0 otherwise, into the regression and run it in the full sample. As presented in Columns (1) and (2) of Table VI, the coefficient of this interaction term is significantly negative. The results indicate that the coefficients of Treat * PostLaw in the poor internal governance subsample are significantly larger than that in the subsamples with superior governance. In unreported results, when the variable Internal Governance is omitted to avoid biased estimates, our results still hold with the same sign, similar magnitude, and significance. We conclude that the passage of

antitakeover laws has a more significant effect on cash distributions (in particular dividends) when firms' internal corporate governance is poor.

E. Effect of the Passage of Laws on Large and Small Firms

Small firms are the most attractive acquisition targets and, as such, have greater exposure to takeover threats. To test whether antitakeover laws have a greater impact on small firms' dividend payouts, we add an interaction term of Treat * PostLaw and a dummy variable Small Firm (which is equal to 1 if the firm's total assets are lower than the median value of the sample in a specific year and 0 otherwise) into the regression and run it in the full sample. As reported in Columns (3) and (4) of Table VI, the coefficient of this interaction term is significantly negative. In unreported results, when the variable Firm Size is omitted to avoid biased estimates, our results still hold with the same sign, similar magnitude, and significance. These results indicate that the coefficients of Treat * PostLaw are significantly larger in the small firm subsample, than in the large firm subsample, suggesting that antitakeover laws have a more significant negative impact on small firms' dividend payout policy.

IV. Robustness Tests

To investigate whether our results are driven by alternative hypotheses, sample selection methods, and regression specifications, we examine a number of alternative models.

A. Substitution Effect from Repurchases

Grullon and Michaely (2002) argue that firms have gradually substituted share repurchases for dividend payouts. Thus, the indication that dividend payout ratios decrease after states adopt antitakeover legislation may be driven by more share repurchases. However, Fama and French (2001) argue that repurchases are not a substitute for dividends as companies often reissue that repurchased stock to employee stock ownership plans, firms acquired in mergers, or as executive stock options.

Nevertheless, to formally address this concern, we replace our dependent variable Dividend Payout Ratio with Total Payout Ratio (including both dividend and repurchase) and repeat the regression. As presented in Table VII, we find that Total Payout Ratio also significantly declines after the passage of antitakeover legislation. These results imply that the repurchase substitution hypothesis cannot explain the dividend payout decline after the passage of antitakeover laws, at least in our sample period.

B. Reduce Dividend Payout for the Sake of Shareholders

Another possible explanation for our results is that firms intentionally pay too much in dividends prior to the passage of antitakeover legislation due to takeover pressures stimulated by a market with a short-term orientation. Thus, firms reduce or even cut their dividend payouts after the passage of the legislation for the sake of their shareholders.

To address this concern, we divide the full sample into two subsamples using our proxy for the severity of the free cash flow problem. Jensen (1986) suggests that agency conflicts are more likely to occur in firms with few growth opportunities and large amounts of cash. In line with this thought, we add an interaction term of Treat * PostLaw and a dummy variable, Low Free Cash Flow Problem, which is equal to 1 if a firm has above the sample median value of Market-to-Book Ratio and below the sample median value of Cash Flow in a specific year and 0 otherwise, into the regression. (12) As demonstrated in Table VIII, the coefficient of this interaction term is significantly negative. The results indicate that the regression coefficients of Treat * PostLaw do indeed significantly differ across these two subsamples. We conclude that the passage of antitakeover laws has a more significant effect on cash distributions, dividends in

particular, when the free cash flow problem is more severe. It is less likely that such firms were underinvesting prior to the law changes. Therefore, these results imply that firms do not reduce dividend payouts to serve the interests of shareholders.

To further address this concern, we examine how managers dispose of funds not disbursed. One possible destination is increased CEO and employee compensation. Bertrand and Mullainathan (1999b) confirm that CEO compensation increased in law-protected firms. Bertrand and Mullainathan ( 1999a, 2003) demonstrate that when managers are insulated from takeovers, they pay more to workers (especially white collar workers). Another possible use is reducing firms' leverage as in Garvey and Hanka (1999). There is no evidence that managers use the undistributed funds to make acquisitions. In fact, Hackl and Testani (1988) document that antitakeover laws actually diminish takeover activities. Additionally, there is no evidence that managers use the undistributed cash to make capital investments. Bertrand and Mullainathan (2003) determine that the creation of new plants actually falls. Therefore, the passage of antitakeover laws does not have an effect on capital expenditures.

In this paper, we further test whether managers increase their cash holdings after they reduce dividend payouts. The results are presented in Table IX. The coefficients of Treat * PostLaw are positive and statistically significant. These results suggest that managers reserve more cash after they are insulated from takeover. Taken together, we conclude that managers do not use the undistributed funds to invest to maximize shareholder wealth. Instead, they increase their employees' compensation, as well as their own, reduce leverage, and keep the cash for use at their discretion.

C. Dividend Payout Restrictions Imposed by State Laws and Debt Holders

In addition to antitakeover legislation, there are several other variations in state laws that may influence firms' dividend payout decisions. State laws related to the minimum asset-to-debt ratio necessary for a payout (Qi and Wald, 2008) are one example. An omitted variable problem would occur if we do not control for it. To deal with these time-invariant state effects, we put a state incorporation dummy into the regression. Because the firm fixed effect in the Dividend Payout Ratio regression dominates the state incorporation effect, our results are the same after controlling for state incorporation effects.

Firms may face similar dividend payout constraints from debt covenants. Qi and Wald (2008) determine that debt holders use more debt covenants to minimize agency costs when borrowers are incorporated in states with stronger antitakeover statutes. Thus, our results may be driven by more dividend payout constraints caused by increased debt covenants after the enactment of antitakeover laws. An omitted variable problem would occur if we do not control for it. Because restrictions on distributions are typically a function of a firm's capital structure (Qi and Wald, 2008), our control variable Leverage can, in some sense, capture this effect. We find that there is negative relationship between firms' dividend payouts and leverage ratios, although it is not always significant. Nonetheless, to address this concern, we repeat our regressions in the subsample including firms with below-median debt ratios in a specific year as these firms are less likely to suffer this problem. Our main results still hold as reported in Table X.

O. Robustness Checks in Alternative Samples and Specifications

The results related to dividend payouts and governance continue to hold after a number of robustness checks. As demonstrated in Table XI, when we use the variable Dividend Payout Ratio (II), measured as dividends as a percentage of the book value of total assets, and the variable Dividend Payout Ratio (III), measured as dividends as a percentage of the market value of total assets as the dependent variables, we still get similar results and the same conclusion. This ensures that our results are not biased by a specific dividend payout ratio definition. In Column (1) of Table XII, our results still hold in the sample that excludes Delaware firms. The coefficients of independent variables are of the same sign and of similar magnitude and significance. This ensures that our results are not driven by the fact that Delaware firms dominate our sample. Column (2) presents similar results based on the 1982-1993 sample period when compared to our main results based on the 1981-1993 period. This confirms that our results are not biased by a certain time period. In Column (3), we run regressions in the sample that only contains firms incorporated in states that passed antitakeover legislation. Therefore, our results are not influenced by the possible differences among firms in states that do and do not pass antitakeover statutes.

Our empirical test may be not appropriate for firms that opt out of state law coverage. For example, Wahal, Miles, and Zenner (1995) document that several large institutional shareholders pressured some firms to opt out of the Pennsylvania law. Fortunately, Gompers, Ishii, and Metrick (2003) argue that "a small minority of firms elect to opt out of some laws" after they checked the data reported in IRRC. To formally address this concern, we repeat our regressions in the sample without the firms that opt out of these three laws reported in the IRRC data in 1993. We obtain the same results, which are presented in Column (4) of Table XII. The coefficients of independent variables have the same sign, similar magnitude, and significance. Thus, we conclude that only a small portion of firms opt out of the statutes in our sample, which does not bias our results.

Serial correlation is a concern in the differences-in-differences approach as the PostLaw dummy changes little over time, as it is 0 prior to and 1 after the passage of antitakeover legislation. Not

correcting for either problem can lead to a serious understatement of standard errors. To account for the presence of cross-sectional and serial correlation of the error terms, we use the block bootstrapping method that constitutes a reliable solution in Bertrand, Duflo, and Mullainathan (2004). We obtain the same outcome based on the results presented in Column (5) of Table XII.

E. Endogenous Timing of Regulatory Reform

It is also possible that states passed antitakeover laws as a result of underlying changes in the economic environment, which may be associated with firms' dividend policy, in states that passed the legislation during the sample period. This means that the passage of antitakeover statutes might be endogenous and there may be some preexisting trends in firms' dividend policy. Therefore, our main results would overstate the impact of antitakeover laws on corporate dividend payouts.

To test these hypotheses, we replace the PostLaw dummy variable with five dummy variables: 1) [Law_Before.sub.it] (-2), 2) [Law_Before.sub.it] (-1), 3) [Law_After.sub.it] (0), 4) [Law_After.sub.it] (1), and 5) [Law_After.sub.it] (2+) in our regressions. [Law_Before.sub.it] (-2, -1) indicates whether an observation for firm i in year t takes place (two years, one year) before antitakeover legislation is passed in the state of incorporation for firm i. [Law_After.sub.it] (0, 1, 2+) is an indicator of whether an observation for firm i in year t takes place (zero year, one year, two or more years) after antitakeover statutes are passed in the state of incorporation for firm i.

If there are underlying trends in the firms' corporate payout policy before the laws are passed, then the coefficients of the terms [Law_Before.sub.it] (-2) and [Law_Before.sub.it] (-1) should be negative and significant. However, as depicted in Table XIII, the data contradict this hypothesis. Across all specifications, the coefficients for [Law_Before.sub.it] (-2) and [Law_Before.sub.it] (-1) are statistically insignificant, but the coefficients for [Law_After.sub.it] (0), [Law_After.sub.it] ( 1 ), and [Law_After.sub.it] (2+) are negative and statistically significant. The findings strongly support the underlying assumption that the passage of antitakeover laws is exogenous and not determined by preexisting trends in corporate policy.

V. Conclusions

The evidence in this paper reveals the strong effects of corporate governance on firms' dividend payout policy. We employ a relatively exogenous measure of the degree of corporate governance, state adoption of antitakeover legislation, instead of endogenous measures adopted in the existing literature. Thus, we can detect not only the correlation between corporate governance and dividend payout policy, but also the causal relation between the two, specifically the role of corporate governance on dividend payout policy.

Consistent with the prediction by agency theory, we find that managers pay fewer dividends and even cut dividend payouts after firms are insulated from takeover threats. The evidence also indicates that the passage of antitakeover statutes has a more significant impact on firms with poor internal corporate governance and small firms. These results are not driven by sample selection and alternative explanations.

Our results are consistent with recent findings in studies that use country-level investor protection as a proxy for corporate governance, which is also an exogenous measure (La Porta et al. 2000; Dittmar, Mahrt-Smith, and Servaes, 2003). They find that dividend payouts are lower and cash holdings are greater in countries with weak investor protection.
Appendix: Variable Definitions

Dependent variables

Dividend Dummy Defined as 1 if dividend (21) is greater
 than zero, and 0 otherwise. Source:
 Compustat

Dividend Payout Ratio Ratio of dividends (21) to earnings
 (18). Source: Compustat

Total Payout Ratio Ratio of total payout (2l + 115--56) to
 earnings (18). Source: Compustat

Dividend Propensity Difference between the Dividend Dummy
 and the expected probability of dividend
 payout. The expected probability of
 dividend payout is calculated by the
 formula: Pr (Dividend

 Dummy = 1) = logit [-0.14 + 4.26 (Firm
 Size)--0.81

 (Market-to-Book Ratio) - 1.07 (Growth
 Rate of Assets) + 15.57

 (Profitability)] (Baker and Wurgler,
 2002). Source: Compustat

Dividend Payout Ratio (II) Ratio of dividends (21) to total assets
 (6) multiplied by 100. Source: Compustat

Dividend Payout Ratio (III) Ratio of dividends (21) to market value
 of assets [Total asset (6) - book equity
 + fiscal year end market price (199) *
Independent variables common shares outstanding (25)]
 multiplied by 100. Source: Compustat

Treat Treat takes a value of 1 if a firm is
 incorporated in a state that passed
 antitakeover laws, and 0 otherwise.
 Source: Compustat

Treat * PostLaw Treat times PostLaw. PostLaw takes a
 value of unity in the year when and
 after the corresponding state's
 antitakeover law is passed, and zero
 otherwise. Source: Compustat

Profitability [Income before extraordinary items (18)
 + interest expense (15) + deferred taxes
 (50) if available]/total asset (6).
 Source: Compustat

Book Equity Stockholder's equity (216) minus
 preferred stock plus balance sheet
 deferred taxes and investment tax credit
 (35) minus post retirement asset (330).
 If Data Item 216 is not available, it is
 replaced by either common equity (60)
 plus preferred stock par value (130) or
 assets (6)-liabilities (181). Preferred
 stock is preferred stock liquidating
 value (10) [or preferred stock
 redemption value (56), or preferred
 stock par value (130)]. Source:
 compustat

Market-to-Book Ratio [Total asset (6)--book equity + fiscal
 year end market price (199) common
 shares outstanding (25)]/total assets
 (6). Source: Compustat

Firm Size Percent of NYSE firms with the same or
 lower market capitalization. Source:
 Compustat

Growth Rate of Assets [Total assets, (6)--total assets,_,
 (6)]/total assetst (6). Source:
 Compustat

Retained Earnings to Total Ratio of retained earnings (36) to total
Capital capital (199 * 25). Source: Compustat

Retained Earnings to Total Ratio of retained earnings (36) to total
Assets assets (6). Source: Compustat

Idiosyncratic Risk Standard deviation of residuals from a
 regression of the excess returns (raw
 returns less the riskless rate) on the
 Fama and French (1993) factors HML, SMB,
 and MKT. Data on the Fama and French
 (1993) factors are from
 http://mba.tuck.dartmouth.edu/pages/
 faculty/ken.french/data_library.html

Leverage Ratio of book value of current and
 long-term debt (9 + 34) to total assets
 (6). Source: Compustat

Investment Ratio of capital expenditures (128) to
 total assets (6). Source: Compustat

Taxes Ratio of total income taxes (16) to EBIT
 (178). Source: Compustat

Anahst Forecasts Natural log of the number of
 one-year-ahead analyst forecasts of EPS
 for the current fiscal year. Source:
 I/B/E/S

Cash Holding Ratio of cash and marketable securities
 (1) to total assets (6) Source:
 Compustat

Institutional Ownership Total ownership by officers and
 directors as a percentage of firms'
 total share outstanding. Source: Thomson
 Financial

Managerial Ownership Total ownership of all institutions as a
 percentage of firms' total share
 outstanding. Source: Compact Disclosure

Internal Governance 24 minus G index reported by IRRC in the
 year 1990 after removing the provisions
 related to state laws. Source: IRRC


References

Aharony, J. and I. Swary, 1980, "Quarterly Dividend and Earnings Announcements and Stockholders' Returns: An Empirical Analysis," Journal of Finance 35, 1-12.

Allen, F. and R. Michaely, 2003, "Payout Policy," in G. Constantinides, M. Harris, and R. Stulz, Eds. Handbook of the Economics of Finance, Amsterdam, North-Holland, 337-430.

Baker, M.P and J.A. Wurgler, 2002, "Why Are Dividends Disappearing? An Empirical Analysis," New York University Working Paper.

Baker, M.P. and J.A. Wurgler, 2004, "A Catering Theory of Dividends," Journal of Finance 59, 1125-1165.

Bertrand, M., E. Duflo, and S. Mullainathan, 2004, "How Much Should We Trust Differences-in-Differences Estimates?" Quarterly Journal of Economics 119, 249-275.

Bertrand, M. and S. Mullainathan, 1999a, "Is There Discretion in Wage Setting? A Test Using Takeover Legislation," RAND Journal of Economics 30, 535-554.

Bertrand, M. and S. Mullainathan, 1999b, "Corporate Governance and Executive Pay: Evidence from Takeover Legislation," MIT Working Paper.

Bertrand, M. and S. Mullainathan, 2003, "Enjoying the Quiet Life? Managerial Behavior Following Antitakeover Legislation," Journal of Political Economy 111, 1043-1075.

Black, F., 1976, "The Dividend Puzzle," Journal of Portfolio Management 2, 5-8.

Bulan, L., N. Subramanian, and L. Tanlu, 2007, "On the Timing of Dividend Initiations," Financial Management 36, 31-65.

Cheng, S., V. Nagar, and M.V. Rajah, 2005, "Identifying Control Motives in Managerial Ownership: Evidence from Anti-takeover Legislation," Review of Financial Studies 18, 637-672.

DeAngelo, H., L. DeAngelo, and D.J. Skinner, 2004, "Are Dividends Disappearing? Dividend Concentration and the Consolidation of Earnings," Journal of Financial Economics 72, 425-456.

DeAngelo, H., L. DeAngelo, and R.M. Stulz, 2006, "Dividend Policy and the Earned/Contributed Capital Mix: A Test of the Life-Cycle Theory," Journal of Financial Economics 81, 227-254.

Dittmar, A., J. Mahrt-Smith, and H. Servaes, 2003, "International Corporate Governance and Corporate Cash Holdings," Journal of Financial and Quantitative Analysis 38, 111-133.

Easterbrook, F.H., 1984, "Two Agency-Cost Explanations of Dividends," American Economic Review 74, 650-659.

Fama, E.F. and K.R. French, 1993, "Common Risk Factors in the Returns on Stocks and Bonds," Journal of Financial Economics 33, 3-56.

Fama, E.F. and K.R. French, 2001, "Disappearing Dividends: Changing Firm Characteristics or Lower Propensity to Pay?" Journal of Financial Economics 60, 3-43.

Fenn, G.W. and N. Liang, 2001, "Corporate Payout Policy and Managerial Stock Incentives," Journal of Financial Economics 60, 45-72.

Fields, M.A. and J.M. Todd, 1995, "Firm Size, Anti-takeover Charter Amendments, and the Effect of State Anti-takeover Legislation," Managerial Finance 21, 35-51.

Fluck, Z., 1998, "Optimal Financial Contracts: Debt versus Outside Equity," Review of Financial Studies 11, 383-418.

Fluck, Z., 1999, "The Dynamics of the Management Shareholder Conflict," Review of Financial Studies 12, 347-377.

Francis, B., I. Hasan, K. John, and M. Waisman, 2010, "The Effect of State Antitakeover Laws on the Firm's Bondholders," Journal of Financial Economics 96, 127-154.

Garvey, G.T. and G. Hanka, 1999, "Capital Structure and Corporate Control: The Effect of Anti-takeover Statutes on Firm Leverage," Journal of Finance 54, 519-546.

Gompers, P., J. Ishii, and A. Metrick, 2003, "Corporate Governance and Equity Prices," Quarterly Journal of Economics 118, 107-155.

Grinstein, Y. and R. Michaely, 2005, "Institutional Holdings and Payout Policy," Journal of Finance 60, 1389-1426.

Grullon, G. and R. Michaely, 2002, "Dividends, Share Repurchases, and the Substitution Hypothesis," Journal of Finance 57, 1649-1684.

Hackl, J.W. and R.A. Testani, 1988, "Second Generation State Takeover Statutes and Shareholder Wealth: An Empirical Study," Yale Law Journal 97, 1193-1231.

Hart, O. and J. Moore, 1994, "A Theory of Debt Based on Inalienability of Human Capital," Quarterly Journal of Economics 109, 841-880.

Hoberg, G. and N. Prabhala, 2009, "Disappearing Dividends, Catering, and Risk," Review of Financial Studies 22, 79-116.

Hu, A. and P. Kumar, 2004, "Managerial Entrenchment and Payout Policy," Journal of Financial and Quantitative Analysis 39, 759-790.

Jensen, M.C., 1986, "Agency Cost of Free Cash Flow, Corporate Finance, and Takeovers," American Economic Review 76, 323-329.

Karpoff, J.M. and P.H. Malatesta, 1989, "The Wealth Effects of Second Generation State Takeover Legislation," Journal of Financial Economics 25, 291-322.

Khang, K. and T.H.D. King, 2006, "Does Dividend Policy Relate to Cross-Sectional Variation in Information Asymmetry? Evidence from Returns to Insider Trades," Financial Management 35, 71-94.

Kumar, P and B. Lee, 2001 "Discrete Dividend Policy with Permanent Earnings," Financial Management 30, 55-76.

La Porta, R., F. Lopez-de-Silanes, A. Shleifer, and R.W. Vishny, 2000, "Agency Problems and Dividend Policies around the World," Journal of Finance 55, 1-33.

Li, K. and K.L. Zhao, 2008, "Asymmetric Information and Dividend Policy," Financial Management 37, 673-694.

Lintner, J., 1956, "Distribution of Incomes of Corporations among Dividends, Retained Earnings, and Taxes," American Economic Review 46, 97-113.

Michaely, R., R.H. Thaler, and K. Womack, 1995, "Price Reactions to Dividend Initiations and Omissions: Overreaction or Drift?" Journal of Finance 50, 573-608.

Modigliani, F. and M. Miller, 1958, "The Cost of Capital, Corporation Finance and the Theory of Investment," American Economic Review 48, 261-297.

Morck, R., A. Shleifer, and R.W. Vishny, 1990, "Do Managerial Objectives Drive Bad Acquisitions?" Journal of Finance 45, 31-48.

Myers, S., 2000, "Outside Equity Financing," Journal of Finance 55, 1005-1037.

Qi, Y. and J. Wald, 2008, "State Laws and Debt Covenants," Journal of Law and Economics 51, 180-207.

Rozeff, M.S., 1982, "Growth, Beta and Agency Costs as Determinants of Dividend Payout Ratios," Journal of Financial Research 3, 249-259.

Short, H., H. Zhang, and K. Keasey, 2002, "The Link between Dividend Policy and Institutional Ownership," Journal of Corporate Finance 8, 105-122.

Stein, J.C., 1988, "Takeover Threats and Managerial Myopia," Journal of Political Economy 96, 61-80.

Wahal, S., K. Miles, and M. Zenner, 1995, "Who Opts Out of State Anti-takeover Protection? The Case of Pennsylvania's SB 1310," Financial Management 24, 22-34.

Zeckhauser, R. and J. Pound, 1990, "Are Large Shareholders Effective Monitors? An Investigation on Share Ownership and Corporate Performance," in G. Hubbard, Ed., Asymmetric Information, Corporate Finance, and Investment, Chicago, IL, University of Chicago Press, 149-180.

Zwiebel, J., 1996, "Dynamic Capital Structure under Managerial Entrenchment," American Economic Review 86, 1197-1215.

(1) For example, La Porta et al. (2000) find that dividend payouts are lower in countries with weaker investor protection. Rozeff (1982) and Hu and Kumar (2004) find that firms with low insider ownership indeed pay more dividends. Short, Zhang, and Keasey (2002) report a strong positive relationship between firms' institutional ownership and dividend payouts. In contrast, Zeckhauser and Pound (1990) argue that institutional investors and dividend payments are substitutes in terms of the monitoring and signaling rote. Controlling for the simultaneous determination of dividends and institutional ownership, Grinstein and Michaely (2005) find no significant effect of institutional ownership on payouts. Further, Fenn and Liang (2001) find that the likelihood and level of dividend payouts increase when factors such as manager-owned stock options, block ownership by managers and outsiders, CEO compensation, and board independence indicate a high likelihood of managerial entrenchment and a corresponding high agency cost.

(2) Bertrand and Mullainathan (2003) argue that few firms change their states of incorporation, especially in a very large sample.

(3) Cheng, Nagar, and Rajah (2005) use the year of the enactment of the first antitakeover law to measure the adoption of the laws. They argue that investors project the passage of new laws by reflecting on the passage of previous laws, and the impact of new laws is not as strong as the impact of those first antitakeover laws. Thus, the dummy variable PostLaw for each firm takes a value of unity in the years in which the state has its first antitakeover law. When we use this approach to define the variable PostLaw to do robustness tests, our results still hold. The coefficients of independent variables are the same sign and of similar magnitude and significance. The results are available upon request though not reported.

(4) When we replace Retained Earnings to Total Capital with Retained Earnings to Total Assets, our results still hold. The coefficients of independent variables are the same sign and of similar magnitude and significance.

(5) We obtain these factors from http://mba.tuck.dartlnouth.edu/pages/faculty/ken.french/data_library.html.

(6) When we use analyst forecast dispersion, our results still hold. The coefficients of independent variables are the same sign and of similar magnitude and significance.

(7) IRRC only reports G index data for the years 1990 and 1993 in our sample period. When we use G index data in the year 1993 to construct our internal governance proxy and repeat the estimations, we get the same conclusions.

(8) There are two reasons for so many observations drop out. The first reason is that the variable earnings are missing values for some observations as our sample period is 1980s. The second reason is that the variable earnings are negative values for some observations.

(9) In unreported results, we eliminate firms with zero dividend payments and rerun our estimations. Our results still hold. The coefficients of independent variables are the same sign and of similar magnitude and significance.

(10) When we examine the effects of the passage of antitakeover laws on dividend payout ratios, the firm fixed effect makes Treat redundant (Bertrand and Mullainathan, 2003). When we examine the effects of the passage of antitakeover laws on Dividend Dummy, we only control for industry effects as there are no big variations in these propensities within firms. Thus, we put Treat into the regression equation along with Treat * PostLaw.

(11) When we use firm fixed effects in the regression, all the firms that constantly pay or not pay dividends in our sample period are dropped because a firm dummy perfectly predicts the outcome variable, which is the Dividend Dummy. Thus, controlling for firm fixed effect, our sample reduces from 14,473 observations to a substantially smaller 2,800 observations. However, our results still hold in the reduced sample. The coefficients of independent variables have the same sign, similar magnitude, and equally strong statistical significance level.

(12) Cash Flow is measured as ratio of EBITDA (13) to total assets (6).

We thank Bill Chrisite (Editor), two anonymous referees, and seminar participants at Rensselaer Polytechnic Institute and the 2007 FMA annual conference in Orlando, FL for valuable comments.

Bill B. Francis, Iftekhar Hasan, Kose John, and Liang Song *

* Bill B. Francis is at the Lally School of Management and Technology, Rensselaer Polytechnic Institute, Troy, NY. Iftekhar Hasan is at the Lally School of Management and Technology, Rensselaer Polytechnic Institute, Troy, NY, and at the Bank of Finland, Helsinki, Finland. Kose John is at the Stern School of Business, New York University, New York, NY. Liang Song is at the School of Business and Economics, Michigan Technological University, Houghton, MI.
Table I. State Antitakeover Legislation

Table I describes business combination law passage times in
various states. (Source: Bertrand and Mullainathan, 2003).

State Year

Arizona 1987
Connecticut 1989
Delaware 1988
Georgia 1988
Idaho 1988
Illinois 1989
Indiana 1986
Kansas 1989
Kentucky 1987
Maine 1988
Maryland 1989
Massachusetts 1989
Michigan 1989
Minnesota 1987
Missouri 1986
Nebraska 1988
Nevada 1991
New Jersey 1986
New York 1985
Oklahoma 1991
Ohio 1990
Pennsylvania 1989
Rhode Island 1990
South Carolina 1988
South Dakota 1990
Tennessee 1988
Virginia 1988
Washington 1990
Wisconsin 1987
Wyoming 1989

Table II. Descriptive Statistics

Table II presents descriptive statistics for the sample of 1,469
firms and an overall total of 11,473 firm-year observations over
the period 1981-1993. All the independent variables except Treat
and Treat * PostLaw have been winsorized at the 1 st and 99th
percentiles. The details of definitions and sources of all the
variables are reported in the Appendix.

 Obs Mean Std. Min Max
 Dev.

Dependent variables
 Dividend Dummy 11,473 0.524 0.499 0.000 1.000
 Dividend Payout Ratio 6,127 0.284 0.216 0.000 0.996
 Total Payout Ratio 5,574 0.342 0.252 0.000 0.962
 Dividend Payout Ratio 11,473 1.273 1.798 0.000 3.046
 (II)
 Dividend Payout Ratio 11,473 0.874 1.088 0.000 2.514
 (III)
Independent variables
 Treat 11,473 0.858 0.349 0.000 1.000
 Treat * PostLaw 11,473 0.504 0.500 0.000 1.000
 Profitability 11,473 0.052 0.125 -0.629 0.246
 Market-to-Book Ratio 11,473 1.764 1.392 0.614 9.512
 Firm Size 11,473 0.333 0.302 0.050 1.000
 Growth Rate of Assets 11,473 0.084 0.194 -0.610 0.686
 Retained Earnings to 11,473 0.323 0.332 -0.233 0.877
 Total Capital
 Idiosyncratic Risk 11,473 0.030 0.019 0.009 0.109
 Cash Holding 11,473 0.170 0.090 0.001 0.293
 Leverage 10,263 0.212 0.165 0.000 0.676
 Investment 10,154 0.073 0.060 0.002 0.331
 Taxes 8,658 0.315 0.136 0.060 0.502
 Analyst Forecasts 6,962 1.699 1.074 0.000 3.584
 Institutional Ownership 6,876 0.370 0.123 0.006 0.504
 Managerial Ownership 3,567 0.069 0.039 0.002 0.123
 Internal Governance 2,606 15.000 2.600 12.000 22.000

Table III. Dividend Payout Pattern versus Time

Table III presents the change in firms' dividend payout pattern
three years before and three years after the passage of
antitakeover legislation. The year in which the law is passed is
considered time 0. Dividend Payout Ratio is defined as the total
dividend scaled by earnings. To construct this variable, only
firms with positive earnings are included. Dividend Propensity
measures the difference between firms' real dividend payout
status and expected dividend payout status. The details of
definitions and sources of all the variables are reported in the
Appendix. The significance levels of the differences in the mean
values between three years after and three years before the
passage of antitakeover legislation are based on a two-tailed
t-test and t-statistics are reported.

Relative Year -3 -2 -1 0

Total Number of Firms 595 648 721 1234

Number of Dividend 364 373 386 488
Payers

Percentage of Dividend 61.18 57.56 53.54 39.55
Payer (%)

Relative Year -3 -2 -1 1

Dividend Propensity -0.074 -0.076 -0.115 -0.173

Dividend Payout Ratio 0.216 0.208 0.191 0.166

Relative Year 1 2 3

Total Number of Firms 1065 958 887

Number of Dividend 478 468 452
Payers

Percentage of Dividend 44.88 48.85 50.96
Payer (%)

Relative Year 2 3 Difference

Dividend Propensity -0.159 -0.134 -5.62 **

Dividend Payout Ratio 0.194 0.208 -2.27 *

** Significant at the 0.05 level.

* Significant at the 0.10 level.

Table IV. Effects of Passage of Antitakeover Laws on Dividend
Payout Ratio

The sample covers the period 1981-1993. Variable definitions are
reported in the Appendix. Coefficients and t-statistics are
reported.

 Dependent Variable =
 Dividend Payout Ratio

 (1) (2)

Treat * PostLaw -0.017 ** -0.0019 ***
 (-2.455) (-2.678)

Market-to-Book Ratio 0.011 *** 0.012 ***
 (3.206) (3.078)

Growth Rate of Assets -0.124 *** -0.109 ***
 (-9.133) (-8.122)

Profitability -1.953 *** -1.902 ***
 (-31.823) (-30.012)

Firm Size 0.047 * 0.025
 (1.952) (0.093)

Retained Earning to 0.045 *** 0.049 ***
Total Capital (6.037) (5.245)

Idiosyncratic Risk -2.300 *** -2.417 ***
 (-8.966) (-9.345)

Cash Holding 0.024 *** 0.025 ***
 (5.367) (5.567)

Leverage -- -0.026
 -- (-1.012)

Investment -- --
 -- --

Taxes -- --
 -- --

Analyst Forecasts -- --
 -- --

Institutional Ownership -- --
 -- --

Managerial Ownership -- --
 -- --

Internal Governance -- --
 -- --

Estimation Methods OLS

Firm and year effects Yes Yes

Observations 6,127 5,704

[R.sup.2]/Likelihood ratio 0.709 0.716

Prob. > F/Prob.> Chi-square 0.000 0.000

 Dependent Variable =
 Dividend Payout Ratio

 (3) (4)

Treat * PostLaw -0.021 *** -0.020 ***
 (-3.157) (-3.123)

Market-to-Book Ratio 0.012 *** 0.015 ***
 (3.134) (3.724)

Growth Rate of Assets -0.121 *** -0.126 ***
 (-8.123) (-8.525)

Profitability -1.913 *** -1.955 ***
 (-29.253) (-30.012)

Firm Size 0.019 0.003
 (0.546) (0.009)

Retained Earning to 0.045 *** 0.047 ***
Total Capital (5.163) (5.252)

Idiosyncratic Risk -2.455 *** -2.475 ***
 (-9.143) (-8.134)

Cash Holding 0.024 *** 0.026 ***
 (5.346) (5.289)

Leverage -0.029 -0.038 *
 (-1.034) (-1.781)

Investment 0.046 0.040
 (0.812) (0.607)

Taxes -- -0.005 ***
 -- (-4.220)

Analyst Forecasts -- --
 -- --

Institutional Ownership -- --
 -- --

Managerial Ownership -- --
 -- --

Internal Governance -- --
 -- --

Estimation Methods OLS

Firm and year effects Yes Yes

Observations 5,638 5,515

[R.sup.2]/Likelihood ratio 0.721 0.739

Prob. > F/Prob.> Chi-square 0.000 0.000

 Dependent Variable =
 Dividend Payout Ratio

 (5) (6)

Treat * PostLaw -0.017 ** -0.018 ***
 (-2.058) (-4.123)

Market-to-Book Ratio 0.023 *** 0.024 ***
 (5.123) (5.256)

Growth Rate of Assets -0.125 *** -0.127 ***
 (-7.456) (-7.789)

Profitability -2.223 *** -2.226 ***
 (-28.478) (-28.283)

Firm Size 0.044 *** 0.045 ***
 (4.976) (4.532)

Retained Earning to 0.086 *** 0.087 ***
Total Capital (7.250) (7.367)

Idiosyncratic Risk -1.309 *** -1.311 ***
 (-5.156) (-5.367)

Cash Holding 0.027 *** 0.026 ***
 (5.256) (5.467)

Leverage -0.012 -0.013
 (-0.506) (-0.647)

Investment 0.071 0.070
 (0.882) (0.875)

Taxes -0.025 *** -0.026 ***
 (-4.123) (-4.478)

Analyst Forecasts -0.011 * 0.012 *
 -1.942 (1.943)

Institutional Ownership -- 0.019 ***
 -- (5.234)

Managerial Ownership -- --
 -- --

Internal Governance -- --
 -- --

Estimation Methods OLS

Firm and year effects Yes Yes

Observations 3,981 3,556

[R.sup.2]/Likelihood ratio 0.773 0.771

Prob. > F/Prob.> Chi-square 0.000 0.000

 Dependent Variable =
 Dividend Payout Ratio

 (7) (8)

Treat * PostLaw -0.017 *** -0.016 ***
 (-4.023) (-4.021)

Market-to-Book Ratio 0.022 *** 0.021 ***
 (5.356) (5.312)

Growth Rate of Assets -0.126 *** -0.125 ***
 (-7.789) (-7.712)

Profitability -2.224 *** -2.223 ***
 (-28.265) (-28.212)

Firm Size 0.043 *** 0.042 ***
 (4.956) (4.912)

Retained Earning to 0.085 *** 0.084 ***
Total Capital (7.251) (7.223)

Idiosyncratic Risk -1.310 *** -1.311 ***
 (-5.345) (-5.323)

Cash Holding 0.028 *** 0.027 ***
 (5.145) (5.112)

Leverage -0.014 -0.013
 (-0.578) (-0.523)

Investment 0.072 0.071
 (0.882) (0.881)

Taxes -0.023 *** -0.022 ***
 (-4.434) (-4.422)

Analyst Forecasts 0.012 * 0.011 ***
 (1.944) (3.912)

Institutional Ownership 0.018 *** 0.017 ***
 (5.289) (5.378)

Managerial Ownership -0.017 *** -0.018 ***
 (-6.333) (-6.123)

Internal Governance -- 0.011 ***
 -- (4.789)

Estimation Methods OLS

Firm and year effects Yes Yes

Observations 3,190 2,004

[R.sup.2]/Likelihood ratio 0.772 0.742

Prob. > F/Prob.> Chi-square 0.000 0.000

 Dependent Variable =
 Dividend Payout Ratio

 (9)

Treat * PostLaw -0.016 ***
 (-4.848)

Market-to-Book Ratio 0.027 ***
 (5.456)

Growth Rate of Assets -0.164 ***
 (-8.457)

Profitability -2.167 ***
 (-25.922)

Firm Size 0.1 18
 (4.220)

Retained Earning to 0.140 ***
Total Capital (10.467)

Idiosyncratic Risk -2.768 ***
 (-5.400)

Cash Holding 0.024 ***
 (5.389)

Leverage -0.023
 (-0.911)

Investment 0.018
 (0.268)

Taxes -0.021 ***
 (-2.846)

Analyst Forecasts 0.017 ***
 (3.299)

Institutional Ownership 0.018 ***
 (3.444)

Managerial Ownership -0.016 ***
 (-6.123)

Internal Governance 0.010 ***
 (4.712)

Estimation Methods Tobit

Firm and year effects Yes

Observations 2,004

[R.sup.2]/Likelihood ratio 1,452.12

Prob. > F/Prob.> Chi-square 0.000

*** Significant at the 0.01 level.

** Significant at the 0.05 level.

* Significant at the 0.10 level.

Table V. Effects of Passage of Laws on Dividend Payout Probability

The sample covers the period 1981-1993. Variable definitions are
reported in the Appendix. Marginal effects and z-statistics are
reported.

 Dependent Variable =
 Dividend Dummy (1 If
 Firm Is Paying
 Dividend)

 (1) (2)

Treat * PostLaw -0.089 *** -0.086 ***
 (-3.368) (-3.221)

Treat 0.038 0.033
 (1.535) (1.223)

Market-to-Book Ratio -0.103 *** -0.013 ***
 (-10.284) (-9.345)

Growth Rate of Assets -0.266 *** -0.342 ***
 (-3.439) (-4.777)

Profitability 1.314 *** 1.380 ***
 (9.986) (9.114)

Firm Size 0.863 *** 0.889 ***
 (23.927) (22.045)

Retained Earning to 0.274 *** 0.255 ***
Total Capital (8.894) (7.221)

Idiosyncratic Risk -16.983 *** -19.224 ***
 (-16.462) (-15.234)

Cash Holding 0.225 *** 0.224 ***
 (4.777) (4.334)

Leverage -- -0.244 ***
 -- (-4.334)

Investment -- --
 -- --

Taxes -- --
 -- --

Analyst Forecasts -- --
 -- --

Institutional Ownership -- --
 -- --

Managerial Ownership -- --
 -- --

Internal Governance -- --
 -- --

 Logit

Industry and year effect Yes Yes

Observations 11,473 10,263

Likelihood ratio -4,140.46 -3,588.66

Prob. > Chi-square 0.000 0.000

Pseudo-[R.sup.2] 0.478 0.497

 Dependent Variable =
 Dividend Dummy (1 If
 Firm Is Paying
 Dividend)

 (3) (4)

Treat * PostLaw -0.091 *** -0.079 ***
 (-3.378) (-3.675)

Treat 0.031 0.013
 (1.134) (0.234)

Market-to-Book Ratio -0.120 *** -0.089 ***
 (-9.776) (-7.647)

Growth Rate of Assets -0.322 *** -0.336 ***
 (-4.274) (-5.874)

Profitability 1.376 *** 0.893 ***
 (9.344) (5.467)

Firm Size 0.912 *** 0.787 ***
 (22.034) (21.375)

Retained Earning to 0.251 *** 0.346 ***
Total Capital (7.123) (10.723)

Idiosyncratic Risk -19.345 *** -17.246 ***
 (-15.578) (-14.736)

Cash Holding 0.225 *** 0.226 ***
 (4.573) (4.567)

Leverage -0.253 *** -0.264 ***
 (-4.367) (-5.456)

Investment -0.511 *** -0.479 ***
 (-3.364) (-3.567)

Taxes -- -0.004
 -- (-0.840)

Analyst Forecasts -- --
 -- --

Institutional Ownership -- --
 -- --

Managerial Ownership -- --
 -- --

Internal Governance -- --
 -- --

 Logit

Industry and year effect Yes Yes

Observations 10,141 8,550

Likelihood ratio -3,356.62 -2,977.67

Prob. > Chi-square 0.000 0.000

Pseudo-[R.sup.2] 0.499 0.476

 Dependent Variable =
 Dividend Dummy (1 If
 Firm Is Paying
 Dividend)

 (5) (6)

Treat * PostLaw -0.078 *** -0.076 ***
 (-2.879) (-3.445)

Treat 0.002 0.003
 (0.045) (0.044)

Market-to-Book Ratio -0.033 *** -0.034 ***
 (-4.567) (-4.334)

Growth Rate of Assets -0.354 *** -0.353 ***
 (-4.845) (-4.123)

Profitability 0.476 *** 0.478 ***
 (3.456) (3.879)

Firm Size 0.590 *** 0.591 ***
 (13.000) (13.888)

Retained Earning to 0.359 *** 0.360 ***
Total Capital (8.925) (8.898)

Idiosyncratic Risk -14.777 *** -14.765 ***
 (-10.856) (-10.456)

Cash Holding 0.227 *** 0.228 ***
 (4.789) (4.745)

Leverage -0.254 *** -0.256 ***
 (-4.024) (-4.345)

Investment -0.378 *** -0.379 ***
 (-3.999) (-3.989)

Taxes -0.004 -0.003
 (-0.665) (-0.055)

Analyst Forecasts -0.013 -0.014
 (-0.985) (-0.675)

Institutional Ownership -- 0.118 ***
 -- (4.367)

Managerial Ownership -- --
 -- --

Internal Governance -- --
 -- --

 Logit

Industry and year effect Yes Yes

Observations 5,738 5,435

Likelihood ratio -1,952.34 -2,345.56

Prob. > Chi-square 0.000 0.000

Pseudo-[R.sup.2] 0.467 0.488

 Dependent Variable =
 Dividend Dummy (1 If
 Firm Is Paying
 Dividend)

 (7) (8)

Treat * PostLaw -0.077 *** -0.075 ***
 (-3.855) (-3.889)

Treat 0.003 0.002
 (0.042) (0.022)

Market-to-Book Ratio -0.034 *** -0.033 ***
 (-4.566) (-4.578)

Growth Rate of Assets -0.356 *** -0.355 ***
 (-4.555) (-4.512)

Profitability 0.477 *** 0.476 ***
 (3.444) (3.111)

Firm Size 0.594 *** 0.593 ***
 (13.234) (13.214)

Retained Earning to 0.356 *** 0.350 ***
Total Capital (8.914) (8.987)

Idiosyncratic Risk -14.765 *** -14.787 ***
 (-10.458) (-10.498)

Cash Holding 0.229 *** 0.200 ***
 (4.754) (4.700)

Leverage -0.255 *** -0.250 ***
 (-4.453) (-4.400)

Investment -0.377 *** -0.370 ***
 (-4.134) (-4.100)

Taxes -0.004 -0.004
 (-0.643) (-0.640)

Analyst Forecasts -0.013 -0.010
 (-0.954) (-0.954)

Institutional Ownership 0.119 *** 0.117 ***
 (4.563) (4.512)

Managerial Ownership -0.145 *** -0.144 ***
 (-4.253) (-4.212)

Internal Governance -- 0.010 ***
 -- (4.566)

 Logit

Industry and year effect Yes Yes

Observations 3,446 2,500

Likelihood ratio -2,674.67 -2,612.44

Prob. > Chi-square 0.000 0.000

Pseudo-[R.sup.2] 0.498 0.497

 Dependent Variable =
 Dividend Dummy (1 If
 Firm Is Paying
 Dividend)

 (9)

Treat * PostLaw -0.067 **
 (-3.556)

Treat -0.003
 (-0.132)

Market-to-Book Ratio -0.038 ***
 (-4.564)

Growth Rate of Assets -0.339 ***
 (-4.954)

Prgritability 0.416 ***
 (3.087)

Firm Size 0.653 ***
 (12.543)

Retained Earning to 0.314 ***
Total Capital (7.543)

Idiosyncratic Risk -12.654 ***
 (-9.543)

Cash Holding 0.224 ***
 (4.678)

Leverage -0.045 ***
 (-4.456)

Investment -0.413 ***
 (-3.678)

Taxes -0.005 ***
 (-4.234)

Analyst Forecasts -0.018
 (-0.735)

Institutional Ownership 0.117 ***
 (4.525)

Managerial Ownership -0.143 ***
 (-4.458)

Internal Governance 0.011 ***
 (4.578)

 Probit

Industry and year effect Yes

Observations 2,500

Likelihood ratio -2,456.87

Prob. > Chi-square 0.000

Pseudo-[R.sup.2] 0.499

*** Significant at the 0.01 level.

** Significant at the 0.05 level.

Table VI. Effects of Passage of Laws on Dividend Payout Ratio for
Firms with Different Characteristics

The sample covers the period 1981-1993. Variable definitions are
reported in the Appendix. The dummy variable Poor Internal
Governance is equal to 1 if the firm has lower than the sample
median value of Internal Governance and 0 otherwise. The dummy
variable Small Firm is equal to 1 if the firm's total assets are
lower than the median value of the sample in a specific year and
zero otherwise. Coefficients and t-statistics are reported.

Dependent Variable Dividend Payout Ratio

 (1) (2)

Treat * PostLaw -0.007 *** -0.008 ***
 (-4.893) (-4.945)

Treat * PostLaw * Poor Internal -0.016 *** -0.017 ***
Governance (-5.013) (-5.560)

Treat * PostLaw * Small Firm -- --
 -- --

Market-to-Book Ratio 0.021 *** 0.035 ***
 (3.782) (5.145)

Growth Rate of Assets -0.134 *** -0.167 ***
 (-9.002) (-7.356)

Profitability -1.902 *** -2.145 ***
 (-30.003) (-28.003)

Firm Size 0.049 *** 0.056 ***
 (4.873) (4.823)

Retained Earning to Total Capital 0.034 *** 0.089 ***
 (6.345) (7.003)

Idiosyncratic Risk -2.567 *** -1.453 ***
 (-8.003) (-5.003)

Cash Holding 0.026 *** 0.030 ***
 (5.004) (5.004)

Leverage -- -0.009
 -- (-0.782)

Investment -- 0.065
 -- (0.783)

Taxes -- -0.028 ***
 -- (-4.592)

Analyst Forecasts -- 0.015 ***
 -- (3.874)

Institutional Ownership -- 0.019 ***
 -- (5.156)

Managerial Ownership -- -0.016 ***
 -- (-6.352)

Internal Governance -- 0.015 ***
 -- (4.914)

Firm and year effect Yes Yes

Observations 6,127 2,004

[R.sup.2] 0.71 0.744

Prob. > F 0.000 0.000

Dependent Variable Dividend Payout Ratio

 (3) (4)

Treat * PostLaw -0.008 *** -0.009
 (-4.902) (-4.867)

Treat * PostLaw * Poor Internal -- --
Governance -- --

Treat * PostLaw * Small Firm -0.014 *** -0.013 ***
 (-4.698) (-4.670)

Market-to-Book Ratio 0.014 *** 0.027
 (3.986) (5.025)

Growth Rate of Assets -0.129 *** -0.132 ***
 (-9.005) (-7.463)

Profitability -1.935 *** -2.278 ***
 (-28.004) (-28.135)

Firm Size 0.050 *** 0.047 ***
 (5.425) (4.673)

Retained Earning to Total Capital 0.041 *** 0.087
 (6.673) (7.003)

Idiosyncratic Risk -2.674 *** -1.517 ***
 (-8.104) (-5.156)

Cash Holding 0.025 *** 0.029 ***
 (5.241) (5.226)

Leverage -- -0.017
 -- (-0.892)

Investment -- 0.076
 -- (0.735)

Taxes -- -0.026
 -- (-4.892)

Analyst Forecasts -- 0.019 ***
 -- (3.874)

Institutional Ownership -- 0.015
 -- (5.156)

Managerial Ownership -- -0.021 ***
 -- (-6.005)

Internal Governance -- 0.010
 -- (4.578)

Firm and year effect Yes Yes

Observations 6,127 2,004

[R.sup.2] 0.711 0.743

Prob. > F 0.000 0.000

*** Significant at the 0.01 level.

Table VII. Effects of Passage of Laws on Total Payout Ratio

The sample covers the period 1981-1993. Variable definitions are
reported in the Appendix. Coefficients and t-statistics are
reported.

Dependent Variable Total Payout Ratio

 (1) (2)

Treat * PostLaw -0.072 *** -0.090
 (-4.633) (-4.373)

Market-to-Book Ratio -0.105 *** -0.099 ***
 (-6.838) (-3.798)

Growth Rate of Assets -0.067 *** -0.110 ***
 (-5.012) (-5.383)

Profitability -0.690 *** -1.022 ***
 (-4.567) (-4.737)

Firm Size 0.765 *** 0.750
 (5.293) (4.384)

Retained Earning to Total Capital 0.036 *** 0.059
 (6.838) (4.939)

Idiosyncratic Risk -2.722 *** -2.385 ***
 (-8.384) (-8.372)

Cash Holding 0.036 *** 0.029
 (4.123) (4.282)

Leverage -- -0.211 ***
 -- (-4.838)

Investment -- 0.239
 -- (0.089)

Taxes -- -0.003
 -- (-0.001)

Analyst Forecasts -- 0.020
 -- (0.036)

Institutional Ownership -- 0.020
 -- (5.838)

Managerial Ownership -- -0.023
 -- (-6.292)

Internal Governance -- 0.015 ***
 -- (4.756)

Firm and year effect Yes Yes

Observations 5,574 1,996

[R.sup.2] 0.522 0.573

Prob. > F 0.000 0.000

*** Significant at the 0.01 level.

Table VIII. Effects of Passage of Laws on Dividend Payout Ratio for
Firms with Low and High Cash Flow Problems

The sample covers the period 1981-1993. Variable definitions are
reported in the Appendix. The dummy variable Low Free Cash Flow
Problem is equal to 1 if a firm has above the sample median value
of Market-to-Book Ratio and below the sample median value of Cash
Flow in a specific year and 0 otherwise. Coefficients and
t-statistics are reported.

Dependent Variable Dividend Payout Ratio

 (1) (2)

Treat * PostLaw -0.009 *** -0.010
 (-4.678) (-4.892)

Treat * PostLaw * Low Free Cash Flow -0.016 *** -0.015 ***
Problem (-5.301) (-5.356)

Market-to-Book Ratio 0.018 *** 0.025
 (3.892) (5.679)

Growth Rate of Assets -0.135 *** -0.130 ***
 (-9.009) (-7.004)

Profitability -1.875 *** -2.245 ***
 (-33.003) (-28.458)

Firm Size 0.049 *** 0.045
 (5.003) (4.825)

Retained Earning to Total Capital 0.049 *** 0.089
 (6.356) (7.000)

Idiosyncratic Risk -2.569 *** -1.368 ***
 (-8.004) (-5.004)

Cash Holding 0.029 *** 0.023 ***
 (5.784) (5.256)

Leverage -- -0.017
 -- (-0.892)

Investment -- 0.089
 -- (0.898)

Taxes -- -0.025
 -- (-4.673)

Analyst Forecasts -- 0.017 ***
 -- (5.005)

Institutional Ownership -- 0.019 *
 -- (5.007)

Managerial Ownership -- -0.021
 -- (-6.004)

Internal Governance -- 0.017 ***
 -- (4.703)

Firm and year effect Yes Yes

Observations 6,127 2,004

[R.sup.2] 0.712 0.746

Prob. > F 0.000 0.000

*** Significant at the 0.01 level.

Table IX. Effects of Passage of Laws on Cash Holding

The sample covers the period 1981-1993. Variable definitions are
reported in the Appendix. Coefficients and t-statistics are reported.

Dependent Variable Cash Holding

 (1) (2)

Treat * PostLaw 0.008 *** 0.007
 (4.372) (4.654)

Market-to-Book Ratio -0.007 *** -0.022
 (-3.984) (-5.356)

Growth Rate of Assets -0.023 *** -0.024
 (-9.372) (-7.362)

Profitability 0.843 *** 0.729
 (31.123) (28.372)

Firm Size 0.047 *** 0.044
 (4.123) (4.832)

Retained Earning to Total Capital 0.016 *** 0.015 ***
 (6.281) (7.233)

Idiosyncratic Risk 1.800 *** 1.111
 (8.382) (5.035)

Leverage -- 0.011
 -- (0.008)

Investment -- 0.042
 -- (0.002)

Taxes -- -0.013
 -- (-0.004)

Analyst Forecasts -- 0.011
 -- (0.044)

Institutional Ownership -- -0.014
 -- (-5.123)

Managerial Ownership 0.015
 (6.737)

Internal Governance -- -0.010 ***
 -- (-4.722)

Firm and year effect Yes Yes

Observations 11,473 2,500

[R.sup.2] 0.422 0.453

Prob. > F 0.000 0.000

*** Significant at the 0.01 level.

Table X. Effects of Passage of Laws on Dividend Payout Ratio for
Firms with Low Leverage

The sample covers the period 1981-1993. Only firms with below the
median value of Leverage in a specific year are included.
Variable definitions are reported in the Appendix. Coefficients
and t-statistics are reported.

Dependent Variable Dividend Payout Ratio (2)

 (1) (1)

Treat * PostLaw -0.018 *** -0.017
 (-4.283) (-4.382)

Market-to-Book Ratio 0.012 *** 0.024
 (3.983) (5.123)

Growth Rate of Assets -0.125 *** -0.123
 (-9.012) (-7.012)

Profitability -1.954 *** -2.225 ***
 (-31.283) (-28.123)

Firm Size 0.048 *** 0.044
 (4.983) (4.839)

Retained Earning to Total Capital 0.044 *** 0.084
 (6.031) (7.123)

Idiosyncratic Risk -2.312 *** -1.319 ***
 (-8.012) (-5.372)

Cash Holding 0.023 *** 0.029 ***
 (5.123) (5.382)

Leverage -- -0.013
 -- (-0.061)

Investment -- 0.071
 -- (0.92)

Taxes -- -0.024 ***
 -- (-4.438)

Analyst Forecasts -- 0.013 ***
 -- (4.911)

Institutional Ownership -- 0.019 ***
 -- (5.123)

Managerial Ownership -- -0.018 *
 -- (-6.123)

Internal Governance -- 0.011 ***
 -- (4.767)

Firm and year effect Yes Yes

Observations 3063 1002

[R.sup.2] 0.634 0.627
t
Prob. > F 0.000 0.000

*** Significant at the 0.01 level.

Table XI. Effects of Passage of Laws on Dividend Payout Ratio
Using Different Samples and Specifications

The sample covers the period 1981-1993. In Column (1), we exclude
Delaware firms. In Column (2), the sample period covers 1982-1993
instead of 1981-1993. In Column (3), we exclude firms in states
without antitakeover laws. Column (4) excludes firms that opt out
of these three laws reported in IRRC data in the year 1993. In
Column (5), we use a block bootstrapping method to deal with
serial correlation. Variable definitions are reported in the
Appendix. Coefficients and t-statistics are reported.

Dependent Variable Dividend Payout Ratio

 (1) (2) (3)

Treat * PostLaw -0.021 *** -0.017 *** -0.018 ***
 (-4.123) (-4.038) (-4.083)

Market-to-Book Ratio 0.030 *** 0.023 *** 0.025 ***
 (5.172) (5.172) (5.834)

Growth Rate of Assets -0.134 *** -0.123 *** -0.124 ***
 (-5.273) (-7.021) (-7.039)

Profitability -2.190 *** -2.137 *** -2.136 ***
 (-21.145) (-28.162) (-28.314)

Firm Size 0.133 *** 0.068 *** 0.067 ***
 (3.173) (4.932) (4.939)

Retained Earning to 0.095 *** 0.087 *** 0.086 ***
Total Capital (6.012) (7.382) (6.956)

Idiosyncratic Risk -2.335 *** -1.314 *** -1.269 ***
 (-4.109) (-4.893) (-4.839)

Cash Holding 0.031 *** 0.036 *** 0.041 ***
 (4.231) (4.382) (4.374)

Leverage -0.011 -0.012 -0.011 ***
 (-0.004) (-0.005) (-4.834)

Investment 0.021 0.051 0.080
 (0.005) (0.002) (0.004)

Taxes -0.025 *** -0.024 *** -0.023 ***
 (-4.103) (-3.983) (-3.614)

Analyst Forecasts 0.014 *** 0.011 *** 0.013 ***
 (4.832) (4.834) (4.145)

Institutional Ownership 0.017 *** 0.019 *** 0.017 ***
 (5.382) (5.384) (5.382)

Managerial Ownership -0.016 *** -0.018 *** -0.016 ***
 (-6.382) (-6.839) (-6.382)

Internal Governance 0.010 *** 0.012 *** 0.011 ***
 (4.712) (4.654) (4.897)

Firm and year effect Yes Yes Yes

Observations 1,278 1,800 1,912

[R.sup.2] 0.634 0.621 0.616

Prob. > F 0.000 0.000 0.000

Dependent Variable Dividend Payout Ratio

 (4) (5)

Treat * PostLaw -0.019 *** -0.018
 (-4.281) (-4.392)

Market-to-Book Ratio 0.026 *** 0.023 ***
 (5.281) (5.382)

Growth Rate of Assets -0.125 *** -0.127 ***
 (-7.283) (-7.382)

Profitability -2.138 *** -2.232 ***
 (-28.838) (-28.232)

Firm Size 0.068 *** 0.048 ***
 (4.382) (4.973)

Retained Earning to 0.090 *** 0.086 ***
Total Capital (6.001) (7.382)

Idiosyncratic Risk -1.345 *** -1.317 ***
 (-4.792) (-5.328)

Cash Holding 0.044 *** 0.029
 (4.123) (5.382)

Leverage -0.014 *** -0.014
 (-4.987) (-0.009)


Investment -0.070 *** 0.079
 (-5.008) (0.069)

Taxes -0.027 *** -0.024 ***
 (-3.938) (-4.838)

Analyst Forecasts 0.015 *** 0.013 ***
 (4.463) (4.838)

Institutional Ownership 0.019 *** 0.017 ***
 (5.384) (5.182)

Managerial Ownership -0.019 *** -0.018
 (-6.182) (-6.111)

Internal Governance 0.012 *** 0.013
 (4.912) (4.876)

Firm and year effect Yes Yes

Observations 1,833 2,004

[R.sup.2] 0.621 0.793

Prob. > F 0.000 0.000

*** Significant at the 0.01 level.

Table XII. Effects of Passage of Laws on Dividend Payout Ratio
Using Alternative Dependent Variable Definitions

The sample covers the period 1981-1993. In Columns (1) and (2),
the dependent variable is Dividend Payout Ratio (11) measured as
dividends scaled by the book value of total assets. In Columns
(3) and (4), the dependent variable is Dividend Payout Ratio
(III) measured as dividends scaled by market value of total
assets. Variable definitions are reported in the Appendix.
Coefficients and t-statistics are reported.

Dependent Variable Dividend Payout Ratio (II)

 (1) (2)

Treat * PostLaw -0.151 *** -0.140 ***
 (-4.892) (-4.762)

Market-to-Book Ratio 0.124 *** 0.130 ***
 (5.002) (5.134)

Growth Rate of Assets -0.883 *** -0.901 ***
 (-5.782) (-6.003)

Profitability -3.004 *** -3.124 ***
 (-17.001) (-17.245)

Firm Size 0.345 *** 0.363 ***
 (4.645) (4.782)

Retained Earning to Total Capital 0.497 *** 0.485 ***
 (5.006) (5.134)

Idiosyncratic Risk -3.205 *** -3.126 ***
 (-5.007) (-5.291)

Cash Holding 0.153 *** 0.143 ***
 (5.784) (5.241)

Leverage -- -0.076
 -- (-0.627)

Investment -- -0.426
 -- (-0.572)

Taxes -- -0.077 ***
 -- (-0.592)

Analyst Forecasts -- 0.074
 -- (0.782)

Institutional Ownership -- 0.078
 -- (0.416)

Managerial Ownership -- -0.099
 -- (-0.224)

Internal Governance -- 0.086
 -- (0.817)

Firm and year effect Yes Yes

Observations 11,473 2,500

[R.sup.2] 0.596 0.601

Prob. > F 0.000 0.000

Dependent Variable Dividend Payout Ratio (III)

 (3) (4)

Treat * PostLaw -0.071 *** -0.064 ***
 (-4.672) (-4.573)

Market-to-Book Ratio 0.079 *** 0.084
 (5.115) (5.024)

Growth Rate of Assets -0.505 *** -0.521 ***
 (-6.005) (-6.552)

Profitability -3.135 *** -3.035 ***
 (-20.655) (-20.003)

Firm Size 0.179 *** 0.163 ***
 (4.913) (4.892)

Retained Earning to Total Capital 0.366 *** 0.335
 (6.002) (6.102)

Idiosyncratic Risk -2.988 *** -2.964 ***
 (-5.557) (-5.638)

Cash Holding 0.115 *** 0.104 ***
 (5.116) (5.372)

Leverage -- -0.051
 -- (-0.415)

Investment -- -0.293
 -- (-0.674)

Taxes -- -0.080
 -- (-0.435)

Analyst Forecasts -- 0.050
 -- (0.816)

Institutional Ownership -- 0.071
 -- (0.355)

Managerial Ownership -- -0.073
 -- (-0.126)

Internal Governance -- 0.050
 -- (0.712)

Firm and year effect Yes Yes

Observations 11,473 2,500

[R.sup.2] 0.589 0.636

Prob. > F 0.000 0.000

*** Significant at the 0.01 level.

Table XIII. Analysis of Time Trends in Regression Estimates

The sample covers the period 1981-1993. Variable definitions are
reported in the Appendix. Law Before;, (-2, -1) is an indicator
for whether an observation for firm i in year t takes place (two
years, one year) before an antitakeover law is passed in the
state of incorporation for firm i. Law After;, (0, 1, 2+) is an
indicator for whether an observation for firm i in year t takes
place (zero year, one year, two or more years) after an
antitakeover law is passed in the state of incorporation for firm

i. Coefficients and t-statistics are reported.

Dependent Variable Dividend Payout Ratio

Treat * Law Before (-2) 0.011 0.005
 (1.543) (0.309)

Treat * Law Before (-1) 0.011 0.008
 (1.158) (0.131)

Treat * Lax After (0) -0.010 *** -0.008
 (-3.899) (-4.456)

Treat * Law After (1) -0.023 *** -0.038 ***
 (-4.111) (-4.374)

Treat * Law After (2+) -0.012 *** -0.008 ***
 (3.374) (-4.678)

Market-to-Book Ratio 0.007 *** 0.024
 (4.983) (5.012)

Growth Rate of Assets -0.125 *** -0.124
 (-9.291) (-7.011)

Profitability -2.001- * -2.012
 (-32.632) (-28.001)

Firm Size 0.066 *** 0.067
 (2.332) (4.832)

Retained Earning to Total Capital 0.092 *** 0.088
 (6.001) (7.234)

Idiosyncratic Risk -2.312 *** -1.317 ***
 (-4.001) (-4.832)

Cash Holding 0.030 *** 0.034
 (4.111) (4.234)

Leverage -- -0.011
 -- (-0.006)

Investment -- 0.052
 -- (0.003)

Taxes -- -0.025
 -- (-3.989)

Analyst Forecasts -- 0.012
 -- (4.812)

Institutional Ownership -- 0.020 ***
 -- (5.983)

Managerial Ownership -- -0.019
 -- (-6.998)

Internal Governance -- 0.010 ***
 -- (4.987)

Firm and year effect Yes Yes

Observations 6,127 2,004

[R.sup.2] 0.683 0.699

Prob. > F 0.000 0.000

*** Significant at the 0.01 level.
COPYRIGHT 2011 Financial Management Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Francis, Bill B.; Hasan, Iftekhar; John, Kose; Song, Liang
Publication:Financial Management
Date:Mar 22, 2011
Words:14131
Previous Article:Deviations from expected stakeholder management, firm value, and corporate governance.
Next Article:Does corporate headquarters location matter for firm capital structure?
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters