02 Does Board Gender Diversity Increase Dividend Payouts Analysis of Global Evidence

Journal of Corporate Finance 58 (2019) 1–26
Contents lists available at ScienceDirect
Journal of Corporate Finance

journal homepage: www.elsevier.com/locate/jcorpfin
Does board gender diversity increase dividend payouts? Analysis of

T
global evidence
⁎
Dezhu Yea, Jie Denga, Yi Liub, Samuel H. Szewczykc, Xiao Chend,
a
Department of Finance, Research Institute of Finance, Jinan University, Guangzhou 510630, China
b
Department of Finance, Insurance, Real Estate, and Law. G. Brint Ryan College of Business, University of North Texas, 1155 Union Circle #305339
Denton, TX 76203-5017, USA
c
Department of Finance, LeBow College of Business, Drexel University, 3220 Market Street, Phiadelphia, PA 19104, USA
d
School of Management and Economics, The Chinese University of Hong Kong, Shenzhen, 2001 Longxiang Road, Longgang District, Shenzhen 518172,
China
A R T IC LE I N F O ABS TRA CT
JEL classifications: Employing 63,464 firm-year observations of 8876 companies in 22 countries from 2000 to 2013,
G15 we conduct a series of multiple regression analyses that reveal a significantly positive relation-
G30 ship between board gender diversity and dividend payouts. The empirical results confirm that
G35 board gender diversity facilitates corporate governance and consequently promotes dividend
payouts. We also show that a good institutional environment may weaken the effect of board
Keywords:
gender diversity on dividend payouts. Institutional ownership is positively associated with board
Corporate governance
gender diversity and that corporate dividend payouts increase when female senior executives
Agency problem
Board gender diversity have shareholdings. The findings of our analysis are robust after controlling for potential en-
Dividend payout dogeneity concerns.
Female board directors
1. Introduction
Most studies attribute dividend payout policy to corporate investment opportunities, ownership structure, cash flow uncertainty,
and country-level institutional environment (Chay and Suh, 2009; Denis and Osobov, 2008; Fama and French, 2001; La Porta et al.,
2000). Chen et al. (2017) recently filled an empirical gap in the literature by examining how the gender composition of corporate
boards influences dividend payout policy. They find that female independent directors were likely to promote dividends in U.S. listed
companies during the period of their study, 1997 to 2011. However, their study uses a relatively small sample of U.S. listed com-
panies and mainly discusses the association between female independent directors and dividend payouts. To the best of our
knowledge, there are no studies examining the relationship between board gender diversity and dividend payouts across listed
companies worldwide. Consequently, this paper employs a larger, internationally diverse sample over a longer time span to explore
whether board gender diversity, across all directors, influences the likelihood and ratio of dividend payouts worldwide.
According to statistics for globally listed companies in 2016, approximately 11.7% of the directors serving on the average cor-
porate board of globally listed companies and 4.6% of their CEOs are female.1 There is clearly much room for improvement in both
⁎
Corresponding author.
E-mail addresses: tyedezhu@jnu.edu.cn (D. Ye), Jiedeng0905@163.com (J. Deng), Ian.Liu@unt.edu (Y. Liu),
szewczsh@drexel.edu (S.H. Szewczyk), chenxiao427@163.com (X. Chen).
1
http://www.gender-map.com/
https://doi.org/10.1016/j.jcorpfin.2019.04.002
Received 12 January 2018; Received in revised form 1 April 2019; Accepted 5 April 2019
Available online 08 April 2019
0929-1199/ © 2019 Elsevier B.V. All rights reserved.
D. Ye, et al. Journal of Corporate Finance 58 (2019) 1–26
areas. In this century, financial crises and corporate scandals have affected policies relating to the composition of corporate boards
and alerted authorities to the importance of board gender diversity (Adams and Funk, 2012; Liu et al., 2014). In 2003, the Norwegian
parliament enacted legislation requiring all listed companies to achieve at least 40% female board representation by the end of 2008.
Other European countries, such as Spain and Sweden, followed suit in setting a female quota for their domestically listed companies
(Ahern and Dittmar, 2012; Gul et al., 2011). More recently, on August 29, 2018, California legislators passed a bill requiring major
listed companies in the state to have at least one female on their boards by the end of 2019; companies that fail to do so face financial
penalties.2
Boards of directors authorize the payment of dividends and set dividend payout policy. The quality of corporate governance can
affect dividend payout policy as dividend payments are influenced, in part, by the conflicts of interest between corporate insiders and
outside shareholders mediated by the board (Easterbrook, 1984; Jensen et al., 1992; La Porta et al., 2000). Paying dividends would
reduce free cash flow available to finance investment, requiring managers to raise external capital; this exposes management to
greater regulation by outside investors, thereby mitigating agency problems (DeAngelo and DeAngelo, 2006). By removing free cash
flow from the firm, dividend payouts also frustrate managers who would misuse free cash flow left in their control to pursue their
private interests at the expense of shareholders (Jensen, 1986). La Porta et al. (2000) argue that, in countries with better institutional
environments, greater shareholder protection rendered by these institutional environments reduces agency problems, enhances
corporate governance, and raises the dividend payout ratio. Chae et al. (2009) suggest that, when corporate financing constraints are
not severe, improving the dividend payout ratio can mitigate agency problems. In an analysis of companies paying dividends for the
first time, Kale et al. (2012) find that firms with greater agency problems are more likely to pay dividends. This implies that more
effective corporate governance will promote higher dividend payout policies to address agency problems.
Prior studies suggest that gender-diverse boards promote more effective corporate boards (Adams and Ferreira, 2009; Gul et al.,
2011; Cumming et al., 2015). Studies have shown that females can differ significantly from males with respect to work ethic, personal
morality, risk aversion, and decision-making behavior. Adams and Ferreira (2009) note that female board directors tend to be more
diligent and have a higher meeting attendance rate relative to males. Other studies suggest that females are more inclined to comply
with rules and laws (Bernardi and Arnold, 1997) and more sensitive to ethical issues (Cumming et al., 2015). Price (2012) finds that
female managers are more risk averse. Eckel and Fullbrunn (2015) argue that, being more cautious and risk-averse, females are less
likely to go to extremes and exhibit overconfidence. Other studies note that females easily develop cooperative relations which
promotes balanced corporate decisions. Adams and Kirchmaier (2016) also suggests that the better communication skills of female
directors promote board efficiency. Additionally, gender-diverse boards tend to be inclusive of various points of view, and this can
strengthen the board's ability to resolve complex problems. Collectively, these studies suggest that gender diverse boards are more
likely to make well-reasoned decisions and more inclined to promote shareholder's interest by addressing agency problems, and,
consequently, are more inclined to initiate dividends and set higher dividend payout ratios.
This paper empirically analyzes the association between corporate gender diversity and corporate payout policy using a global
dataset. We construct a large international sample comprising observations from 8876 companies in 22 countries over the period
2000 to 2013. Our corporate finance data are mainly collected from the Worldscope database. Data for the corporate boards are
sourced from the BoardEx database. We find that the effect of board gender diversity on dividend payouts is significantly positive.
Specifically, both the number and proportion of female board directors affect the likelihood and the level of dividend payouts. We
further find evidence that a country's institutional environment moderates the effects of board gender diversity on dividend payouts.
Statistically, the effect of board gender diversity on dividend payouts is smaller in good institutional environments. The empirical
results also reveal a significantly positive relationship between gender-diverse boards and dividend initiations and increases. The
findings are robust to the considerations of macro factors and share repurchases, as well as to alternative measures of board gender
diversity. In further analysis, the paper investigates the impact of financial crisis (specifically the global financial crisis), national
culture, family-controlled ownership, and state-controlled ownership on the association between board gender diversity and dividend
payouts. Our collective results are consistent with arguments that board gender diversity promotes effective corporate governance,
and this promotes the use of dividend payout policy as a tool to reduce agency costs. Our results indicate that this occurs worldwide
across countries.
We adopt a variety of methods to address potential endogeneity concerns. To address omitted variables, we follow Lin et al.
(2013), Chen (2015), and An et al. (2016) by defining the average number of industry–year–country female directors (IndMeanFDIR)
as the instrumental variable (IV), using the double-digit industry code. This method of constructing the IV is also commonly used by
the most recent and relevant literature (Adhikari and Agrawal, 2018; Chen, 2015; Eom, 2018; Hasan and Cheung, 2018; Huang and
Mazouz, 2018; Jiang and Yuan, 2018; Kim et al., 2017; Ward et al., 2018; Zhang et al., 2016). To consider reverse causality, we adopt
two methods: the first-order lag of independent variables and a forecasting model of the number (proportion) of female directors.
Specifically, we first examine the impact of the number (proportion) of unexplained female directors on dividend policy and then test
for reverse causality between the two, theoretically based on a linear relationship between the number (proportion) of female
directors and corporate characteristics. To address selective bias problems, we utilize two methods: (1) excluding U.S. companies; and
(2) logit propensity score matching (PSM) for boards with at least one female director (FD = 1) and boards with at least three female
directors (FD3 = 1), where FD and FD3 are the respective treatment groups. Finally, we employ PSM and difference-in-difference
(DID) to examine how the exogenous shock of board gender quotas being introduced in several European countries affected the
dividend policy. Our results empirically show that introducing a board gender quota has a significantly positive influence on
2
http://www.ntmarkets.com/
2
corporate dividend payments across different countries.

This paper contributes to several strands of the finance literature. Our research is the first to clarify the association between board
gender diversity and dividend payouts in an international setting. In doing so, it provides abundant robust evidence that board gender
diversity increases corporate payouts. Our findings further corroborate the results of Chen et al. (2017) by considering all (rather than
only independent) female directors and using data covering a longer time span and wider sample range. For instance, after con-
sidering differences in the legal and governance environment across countries and over time, as well as the exogenous shock of board
gender quotas being introduced in European countries, this study uses comparison tests and PSM-DID to examine the characteristics
of and variances in payout policy across different countries. It also explores the impact of board gender diversity on dividend
initiations and increases and considers other characteristics of firms, such as share repurchases and whether they are state- or family-
controlled. By exploring the impact of board gender quotas being introduced in European countries, we provide a new explanation for
the effect of institutional environment on firm-level payout policy. To reinforce our empirical findings, our exploratory analysis uses
six other popular board gender diversity indicators and other measures of payout policy. We expect our results to inform the im-
portance of gender diversity among policymakers, firms, and individuals.
The paper is structured as follows: Section 2 develops the research hypothesis, and then Section 3 details the research design.
Section 4 presents the empirical analysis, followed by additional robustness tests in Section 5. Finally, Section 6 concludes the paper.
2. Research hypotheses
Monitoring and resolving agency problems originating from conflicts of interest between corporate insiders and outside share-
holders is a fundamental function of corporate governance (Jensen and Meckling, 1976; Fama, 1980; Fama and Jensen, 1983). In
countries where corporate ownership is widely dispersed such as in the United States, Canada, and Great Britain, the principal
insiders are managers who are delegated decision-making authority to be used on behalf of the firm's shareholders; however,
managers can abuse their authority by devoting time and the firm's resources in pursuing their own interests rather than working to
maximize shareholder value (Jensen and Meckling, 1976). In most of the world's countries, corporate ownership is concentrated, and
principal owners are controlling shareholders who effectively direct the decisions of managers or may manage the company
themselves (La Porta and Lopez-de-Silanes, 1999). Controlling shareholders can use their control to benefit themselves at the expense
of outside shareholders who are most likely minority shareholders.
Agency problems associated with corporate conflicts of interest can include excessive salaries and consumption of perquisites by
insiders; empire building in which insiders pursue expansionary policies that increase the size of the firm even though the growth
does not benefit outside shareholders; and the tendency of insiders to invest earnings left in the firm in suboptimal marginal in-
vestments (Jensen and Meckling, 1976; Grosssman and Hart, 1988; Jensen, 1986). Where ownership is concentrated, this can even
include outright expropriation of minority shareholder value through abuses such as the sale of corporate assets at below value to
other firms controlled by inside shareholders (La Porta et al., 1998). Given these agency conflicts, outside shareholders have pre-
ference for disgorging cash flow though payments of dividends as the expectation is that cash left in the firm will be diverted to yield
personal benefits of control to corporate insiders or invested inefficiently (La Porta et al., 2000).
Dividend payout policy can be an important tool for addressing agency problems (Brav et al., 2005; Easterbrook, 1984; La Porta
et al., 2000). Dividend payments reduces corporate free cash flow which forces managers to raise more external capital. This in-
creases the regulation of managers by outside investors and works to increase external monitoring of corporate agency problems
(DeAngelo and DeAngelo, 2006). Moreover, disgorging corporate earnings to shareholders prevents managers from using them to
pursue their private interests and precludes managers from investing free cash flow in suboptimal projects, especially in cases where
the firm has few good investment opportunities (Fama, 1980). Chae et al. (2009) show that when corporate financing constraints are
not severe, improving the dividend payout ratio can mitigate agency problems. In an analysis of U.S. companies paying dividends for
the first time, Kale et al. (2012) find that firms with greater agency problems are prone to pay corporate dividends.
Previous studies suggest that gender-diversity on boards contributes to more effective corporate governance (Adams and Ferreira,
2009; Gul et al., 2011; Hillman et al., 2007). Ahern and Dittmar (2012) and Kim and Starks (2016) report that increasing the number
of female directors promotes higher corporate valuations. Adams and Kirchmaier (2016) demonstrate the positive effect of female
directors in science and technology-listed corporations. By improving corporate governance and the protection of shareholder in-
terests, board gender diversity can encourage the payment of corporate dividends and higher payout ratios (Levit and Malenko, 2016;
Li and Srinivasan, 2011; Wintoki et al., 2012). Chen et al.'s (2017) study of U.S. listed companies finds that firms with more female
independent directors issue larger dividends.
Board gender diversity can influence board efficiency at both individual and team levels. At the individual level, researchers find
that females can differ from males in ways that can improve board efficiency. For example, Adams and Ferreira (2009) find that
female board directors are more diligent and have higher meeting attendance rates relative to male directors. Bernardi and Arnold
(1997) find that males pay more attention to earnings and promotion and are more inclined to bend rules to achieve success. In
contrast, they find that females prefer harmony and are more inclined to abide by rules and laws. According to Gul et al. (2011),
female directors are more sensitive to ethical issues than males. Females are also more risk-averse (Price, 2012). Consistent with these
findings, Cumming et al. (2015) reports that increased gender diversity at the board level is associated with significantly fewer
incidents of corporate fraud and insider trading. Collectively, these studies imply that, on the individual level, as female directors are
more inclined to abide by rules and laws, are more sensitive to ethical issues, and more risk adverse, that they will be more concerned
with agency problems and the reputation of the firm for good corporate governance and generally more inclined to promote the
interests of all shareholders. Firms with more female directors on the boards appear to face lower litigation risk, which perhaps in
3
part explains higher payouts (Liu, 2018). Therefore, from the perspective of the individual level, in general, we expect female
directors to be more supportive of higher dividend payout.
At the team level, many studies show that teams with female participants are more effective at solving complex problems. The
literature on leadership traits suggest that woman employ a leadership style characterized by trust and cooperation (Agarwal et al.,
2016; Croson and Gneezy, 2009; Niederle and Vesterlund, 2007). On corporate boards, trust can encourage a greater exchange of
information among board directors, and cooperation, a better synthesis of information, both contributing to balanced and appro-
priate decisions. Adams and Kirchmaier (2016) notes that female directors bring their better communication skills to the board which
increases board efficiency. Gender diverse boards have greater diversity in points of view and inclusion of diverse perspectives avoids
‘group think’ and leads to better corporate board decision making (Gul et al., 2011). These studies suggest that gender-diversity on
boards promote extensive communication and discussion at the board level, incorporating various perspectives which results in better
informed and appropriate decisions, and these would include decisions concerning agency problems and promotion of all share-
holders' interests. Therefore, from the team level perspective, we expect greater gender diversity on corporate boards to be associated
with higher dividend payout.
Based on the above discussion, our primary hypothesis is as follows:
Hypothesis 1. Board gender diversity significantly promotes higher corporate dividend payments.
The 22 countries in our dataset have different institutional environments, providing a unique opportunity to study the effect of
institutional environment on the relationship between board gender diversity and dividend payout.
A country's institutional environment, particularly the content and enforcement of the legal protections afforded to outside
investors, can influence corporate dividend policy (Brockman and Unlu, 2009; Cumming et al., 2011; Dang et al., 2015). La Porta
et al. (2000) propose two models by which agency and the institutional environment affect dividend payouts. In the first, the outcome
model, dividends are the outcome of the country's legal protection of shareholders. To prevent an unacceptable proportion of
earnings being diverted to provide private benefits to insiders, outside shareholders will pressure corporate insiders to disgorge cash.
The better the legal protection given to shareholders, the more effective the pressure outside shareholders can exert on insiders, and
the larger will be dividend payouts, especially for firms with few profitable investment opportunities. Consistent with this argument,
La Porta et al. (2000) find firms in countries with better minority shareholder protection pay higher dividends. In the second, the
substitution model, dividends substitute for weak legal investor protection in countries with poor institutional environments. Divi-
dend payouts are means by which firms can establish reputation for fair treatment of outside shareholders in the financial market.
Such reputation is especially valuable for firms needing external financing in countries with poor investor protection. The weaker the
legal protection given investors, the greater is the incentive for firms to build reputation with potential buyers of their common stock,
and the larger will be dividend payouts, especially for firms with good investment opportunities (La Porta et al., 2000). Jiraporn and
Ning (2006) find an inverse association between dividend payouts and shareholder rights, indicating that firms pay higher dividends
where shareholder rights are more suppressed.
Regardless which mechanism dominates given the country's institutional environment, we expect that the effect of board gender
diversity on dividend payout to be inversely related to the quality of the country's institutional environment. In countries with weak
minority shareholder protection, agency problems are larger and outside shareholders are in a weaker legal position to prevent
insiders from expropriating their wealth (La Porta et al., 2000; Brockman and Unlu, 2009). In such environments, dividend policy can
be called upon to play a greater role in addressing agency problems given the lesser role played by the law. Therefore, based on our
previous discussion, to the extent that female directors have greater concern for agency problems, the promotion of all shareholders'
interests, and for the reputation of the firm regarding treatment of shareholders, the effect of board gender diversity on dividend
payout will be greater in weaker institutional environments.
In countries with strong institutional environments, agency problems are smaller as stronger legal protection for shareholders can
supplement or substitute for corporate governance protections on the firm level. Given strong legal protections, dividend policy can
be relegated to a smaller role in addressing agency problems and other factors determining dividend payout be given greater con-
sideration. With lower agency costs and agency playing a smaller role in determining dividend payout, the focus of female directors'
concern for agency problems may be directed to other areas of corporate governance and to other tools addressing agency problems.
Consequently, the effect of board gender diversity on dividend payout will be smaller the stronger the institutional environment.
Based on the above discussion, we propose the following second hypothesis:
Hypothesis 2. Good institutional environment moderates the effect of board gender diversity on dividend payouts.
3. Research design
We utilize corporate financial data from the Worldscope database and board data from the BoardEx database for the sample
period from 2000 to 2013. We include only data for firms marked as “Major Security” or “Primary Quote” to avoid those that have
issued securities in several stock markets simultaneously. Second, we exclude firms with null values for board gender or the corporate
International Securities Identification Number (ISIN) code. We merge corporate financial data with board data using the ISIN code.
Following the literature, we process the final data as follows. (1) We exclude financial and utility companies based on the Industry
Classification Benchmark code. (2) We exclude firms with a null or negative value for cash dividend. (3) We exclude firms with net
profits that exceed their sales. (4) We exclude countries with fewer than 40 firms. (5) Based on La Porta et al. (2000), we exclude
Luxembourg and mandatory dividend countries: Brazil, Chile, Colombia, Greece, and Venezuela. (6) We winsorize corporate financial
4
Table 1
Country and firm sample distribution.
Country Number of firms Percentage of firms Number of samples Percentage of samples
Australia 621 7.00 3180 5.01

Belgium 40 0.45 371 0.58
Canada 639 7.20 3227 5.08
China 81 0.91 329 0.52
Finland 42 0.47 252 0.40
France 276 3.11 2182 3.44
Germany 230 2.59 1438 2.27
Hong Kong 229 2.58 823 1.30
India 214 2.41 1083 1.71
Ireland 48 0.54 407 0.64
Israel 59 0.66 421 0.66
Italy 50 0.56 399 0.63
Japan 227 2.56 444 0.70
New Zealand 53 0.60 448 0.71
Norway 92 1.04 802 1.26
Singapore 186 2.10 639 1.01
South Africa 141 1.59 564 0.89
Spain 70 0.79 472 0.74
Sweden 102 1.15 948 1.49
Switzerland 56 0.63 476 0.75
United Kingdom 1493 16.82 11,658 18.37
United States 3927 44.24 32,901 51.84
Total 8876 100.00 63,464 100.00
continuous variables at the 2% level. Following these steps, our sample comprises 63,631 firm-year observations from 8897 com-
panies in 22 countries from 2000 to 2013. Table 1 gives a detailed distribution of the sample firms by country.
Table 2 presents the definitions of the indicators. First, Panel A defines the relevant indicators for dividend payment. DIV refers to
whether the firm pays dividends, DIVEARN is the cash dividend-to-net profits ratio, and DIVTA is the cash dividend-to-total asset
ratio. Based on Chen et al. (2017) and Chay and Suh (2009), this paper primarily considers these three measures of dividend payout
policy. TOTALEARN (Cash dividends + share repurchases) / net income) and TOTALTA (Cash dividends + share repurchases) / total
assets), regarded as the total dividend payout ratio after considering share repurchases, are then used in robustness tests. Second,
Panel B defines the indicators of gender diversity. Following Gul et al. (2011), we measure board gender diversity by: (1) the number
of female directors and the number of female non-executive directors; and (2) female directors and female independent non-executive
directors, respectively, as a proportion of total directors. Additionally, we add several dummy variables, such as the presence of at
least one female director and three female directors on the board. Regarding board gender diversity, the critical point theory proposes
that “one is a representative, two is an existence, three is a kind of voice” (Liu et al., 2014). With three or more females on the board,
females are no longer regarded as “outsiders” and can affect a wider range of board discussions (Apesteguia et al., 2012). FCEO and
FCHAIR are defined to identify whether the CEO or the chairperson is female. We also define the dummy variable FCEODUAL by
whether the CEO and the chairperson are the same female.
Table 3 reports each variable's statistics, including the number of observations, mean value, standard deviation, and the 25th,
50th, and 75th quartiles. As Table 3 shows, 45.6% of the sample firms paid cash dividends, but the number of female directors on
each firm's board is, on average, below 1. Only 1.7% of CEOs and 2.2% of chairpersons were female, thus evidencing that the
proportion of females serving in senior positions is very low worldwide.
The results in Appendix A present the distribution of corporate payout policy and board gender diversity by firm-year. In absolute
terms, there was an increase in the number of dividend-paying firms and firms with gender-diverse boards from 2000 to 2013.
However, in relative terms, dividend-paying firms and firms with gender-diverse boards show a decreasing-increasing V shape, and
values for both were lowest in 2006. Using data from U.S. listed companies from 1978 to 1999, Fama and French (2001) highlight a
declining trend in dividend payments. Denis and Osobov (2008) elaborate that this declining trend is due to the reduction of new
listed companies' dividend payment, with dividend payments overall not diminishing. The results presented in Appendix A illustrate a
decreasing trend in dividend payments among listed companies globally from 2000 to 2006, reinforcing Fama and French's (2001)
conclusion; however, after the global financial crisis in 2007, the number of listed companies paying dividends began to increase,
which is consistent with the findings of Floyd et al. (2015).
Fig. 1 shows the yearly average number of female directors over time (FDIR is the yearly average number of female board
directors for the whole sample; FD means the yearly average value of the female dummy variable, which equals 1 when a board has at
least one female, and 0 otherwise). Fig. 2 presents the national average percentage of female board directors worldwide, while the
average number of female directors for each country is shown in Fig. 3.
Overall, the number of female board directors from 2000 to 2013 presents an increasing trend worldwide, illustrating that
companies around the world are beginning to realize that females are an important driver of economic development (Alesina et al.,
2013; Fernandez, 2013). However, among the 22 countries in our sample, Japanese firms have the lowest figures for both the number
and proportion of female board directors. This may reflect the social status of Japanese females (Jin et al., 2015). By contrast, the
5
Table 2
Variable source and definitions.
Variable Names Definition Source
Panel A Dividend payment
DIV Dummy variable, one if the firms pay cash dividends, zero otherwise WorldScope
DIVEARN Dividend payout ratio, cash dividends/net income WorldScope
DIVTA Dividend payout ratio, cash dividends/total assets WorldScope
TOTALEARN Dividend payout ratio after considering share repurchase,(Cash dividends+share repurchase)/net income WorldScope
TOTALTA Dividend payout ratio after considering share repurchase,(Cash dividends+share repurchase)/total assets WorldScope
STR Dummy variable, one if a firm repurchases stock, zero otherwise WorldScope
PRIORPAYER Dummy variable, which equals 1 in year t if the firm has continuously paid dividends in year t − 1 and year t, zero WorldScope
otherwise
INDIVEA Dummy variable, which equals 1 when a firm increases dividend, and 0 otherwise and is calculated by DIVEARN WorldScope
INDIVTA Dummy variable, which equals 1 when a firm increases dividend, and 0 otherwise, and is calculated by DIVTA WorldScope
Panel B Gender diversity index
FDIR Number of female directors on board BoardEx

LFDIR Natural logarithm of number of female directors plus one BoardEx
FSD Number of female independent non-executive directors on board BoardEx
LFSD Natural logarithm of number of female independent non-executive directors plus one BoardEx
P_FDIR Female directors as a percentage of all directors on board BoardEx
P_FSD Female independent non-executive directors as a percentage of non-executive directors on board BoardEx
FCEO Dummy variable: one if the CEO is female, zero otherwise BoardEx
FCHAIR Dummy variable: one if the chairperson of the board is female, zero otherwise BoardEx
FCEODUAL Dummy variable: one if CEO and Chairman of the board are the same person and that person is female, zero otherwise BoardEx
FD Dummy variable: one if there is at least one female director on board, zero otherwise BoardEx
FD3 Dummy variable: one if there are three or more female directors on board, zero otherwise BoardEx
IndMeanFDIR Industry (double-digit industrial code)-country-year average of female directors BoardEx
L.P_FDIR Natural logarithm of P_FDIR plus one BoardEx
L.FSD Natural logarithm of FSD plus one BoardEx
L.P_FSD Natural logarithm of P_FSD plus one BoardEx
L.FD Natural logarithm of FD plus one BoardEx
Panel C Corporate governance index
Own Percentage of management holdings WorldScope

INST Institutional investor holdings: strategic investors hold at least 5% of shares WorldScope
LegCom Whether the country applies common law: one if so, zero otherwise La Porta et al. (1998)
WEF_invPro The strength of investor protection, which is scaled from 1 (worst) to 10 (best) WEF 2010
Panel D Control variables
LOGTA Natural logarithm of total assets, in dollar terms WorldScope

LEV Financial leverage, total debts/total assets, in dollar terms WorldScope
TOBINSQ Tabin's Q (total assets-ordinary shares + market value of ordinary shares)/total assets, in dollar terms WorldScope
ROA Net income/total assets, in dollar terms WorldScope
RETE Lifecycle (retained earnings/total common equity)*100, in dollar terms WorldScope
Cash Cash holdings (cash + current investment)/total assets, in dollar terms WorldScope
STDRET Standard deviation of previous two years' weekly stock returns Datastream
MB Market-to-book: natural logarithm of total market value/total assets, in dollar terms Datastream
LDIR Logarithm of the total number of board directors BoardEx
LPSD Logarithm of the number of independent directors scaled by board size BoardEx
L.LOGTA Natural logarithm of LOGTA plus one WorldScope
L.LEV Natural logarithm of LEV plus one WorldScope
L.TOBINSQ Natural logarithm of TOBINSQ plus one WorldScope
L.ROA Natural logarithm of ROA plus one WorldScope
L.RETE Natural logarithm of RETE plus one WorldScope
L.Cash Natural logarithm of Cash plus one WorldScope
L.STDRET Natural logarithm of STDRET plus one Datastream
L.MB Natural logarithm of MB plus one Datastream
CRISIS Dummy variable, which equals 1 when the sample companies existed during 2007 and 2008, and zero otherwise
FAM Dummy variable, which equals 1 when a company is controlled by families, zero otherwise BvD
Panel E Variables of female directors' forecasting model
RESID_FDIR Residual number of female directors on board

(continued on next page)
6
Table 2 (continued)
Panel E Variables of female directors' forecasting model
RESID_P_FDIR Residual percentage of female directors

RESID_FSD Residual number of female independent directors
RESID_P_FSD Residual percentage of female independent directors
RESID_FD Residual gender diversity
ROE Return on equity, net profits/book value of end-of-term equity, in dollar terms WorldScope
LOGTA Natural logarithm of total assets, in dollar terms WorldScope
FIRMAGE Firm age: natural logarithm of difference between recent two years WorldScope
SALESGROWTH Sale growth: average sales growth over prior three fiscal year Corporate social network: number of firms with an average of WorldScope
one director acting as external non-executive director
DIRECTORSHIPS Total risks: standard deviation of current year's yearly stock returns BoardEx
TOTALRISK Tobin's Q (total assets-ordinary shares + market value of ordinary shares)/total assets, in dollar terms WorldScope
TOBINSQ Ratio of total assets minus book value of equity plus market value to total WorldScope
RET Annual stock return during fiscal year WorldScope
VWRETD Value-weighted market return during fiscal year WorldScope
Table 3
Summary statistics.
Variable Mean SD P25 P50 P75
DIV 0.456 0.498 0.000 0.000 1.000

DIVEARN 0.184 0.402 0.000 0.000 0.268
DIVTA 0.013 0.025 0.000 0.000 0.017
TotalEarn 0.331 0.727 0.000 0.000 0.468
TotalTA 0.026 0.047 0.000 0.005 0.030
FDIR 0.724 1.070 0.000 0.000 1.000
LFDIR 0.407 0.493 0.000 0.000 0.693
FSD 0.624 0.987 0.000 0.000 1.000
LFSD 0.357 0.472 0.000 0.000 0.693
P_FDIR 0.083 0.117 0.000 0.000 0.143
P_FSD 0.089 0.136 0.000 0.000 0.166
FCEO 0.017 0.128 0.000 0.000 0.000
FCHAIR 0.022 0.148 0.000 0.000 0.000
CEODUAL 0.495 0.500 0.000 0.000 1.000
FCEODUAL 0.008 0.088 0.000 0.000 0.000
FD 0.453 0.498 0.000 0.000 1.000
FD3 0.055 0.228 0.000 0.000 0.000
LPSD 0.346 0.179 0.201 0.336 0.470
LDIR 2.007 0.341 1.792 1.946 2.197
LOGTA 12.780 2.156 11.300 12.820 14.290
LEV 0.145 0.172 0.000 0.083 0.240
TOBINSQ 2.051 1.593 1.104 1.503 2.321
ROA −0.758 20.550 −2.650 4.730 9.270
RETE −52.520 308.700 −35.930 27.230 69.530
Cash 0.204 0.216 0.045 0.123 0.290
STDRET 0.142 0.079 0.086 0.122 0.176
MB 0.282 0.712 −0.216 0.196 0.709
LegCom 0.870 0.337 1.000 1.000 1.000
INST 33.120 23.480 14.000 29.000 51.000
Own 27.670 25.170 3.390 21.840 46.500
Variables are as defined in Table 2.
northern European countries of Norway, Finland, and Sweden were the top-three ranked for the percentage of female directors. For
the average number of female directors, Norway and Sweden ranked second and third, respectively. European society pays more
attention to gender equality. For example, the Norwegian government promulgated a decree for female board directors in 2003,
followed by a legislative gender quota for female board directors in 2005. Other European countries, such as Spain, have followed suit
with corresponding measures (Bohren and Staubo, 2014).
To further analyze the characteristics of corporate payout policy and board gender diversity, Table 4 presents the results of the
difference test according to whether firms pay dividends. As Table 4 shows, dividend-paying firms have 0.96 female directors on
average, while the average in non-dividend-paying firms is only 0.53 female directors. The difference between the two is statistically
significant. Moreover, except for female CEOs, female board chairpersons, and the CEO and board chairperson being the same female,
other gender diversity indicators in the dividend-paying firms are significantly greater than those in the non-dividend-paying firms.
In line with previous literature, there are significant differences in other indicators between firms that pay and do not pay
dividends. Larger firms and those with greater profitability, more cash, less risky cash flow, and fewer investment opportunities are
more likely to pay dividends (Chay and Suh, 2009; DeAngelo and DeAngelo, 2006; Denis and Osobov, 2008; Fama and French, 2001).
7
1
0.9
0.8
0.7
0.6
Average
0.5 FDIR
0.4
FD
0.3
0.2
0.1
0
Fig. 1. Trend in Gender Diversity.

Data Source: Based on BoardEx database and the calculations in this paper.
Women on Board: FDIR%

30.00%
25.00%
20.00%
15.00%
10.00%
5.00% FDIR%
0.00% Hong Kong
Netherlands
South Afric
Norway
Germany
Ireland
Switzerland
Australia
Italy
UK
Canada
Singapore
Belgium
France
Israel
US
China
Sweden
Japan
India
Spain
Finland
Fig. 2. Average Percentage of Female Directors on the Board Worldwide.
Women on Board: FDIR

2
1.5
1
0.5
0 FDIR
Hong Kong
Germany
Norway
South Afric
UK
Italy
France
Switzerland
Australia
Netherlands
Singapore
Canada
Ireland
US
Israel
Belgium
China
Spain
Finland
Sweden
Japan
India
Fig. 3. Average Number of Each Country's Female Directors Worldwide.

4. Empirical results
Following prior studies, we measure board gender diversity mainly using the following:
• FDIR: the number of female directors;

• P_FDIR: the proportion of total directors who are female;
• LFDIR: ln (number of female directors +1) (it is necessary to add 1 to avoid LFDIR being 0 when a firm has no female directors);
• FSD: the number of female independent non-executive directors;
• P_FSD: the proportion of independent non-executive directors who are female;
• LFSD: ln (number of female independent non-executive directors +1);
• FD: whether there is at least one female director on a board;
• FD3: whether there are at least three female directors on a board;
• FCEO: whether the CEO is female;
8
Table 4
Difference test (dividend payers and dividend non-payers).
Variables DIV = 0 Mean DIV = 1 Mean Diff
FDIR 33,680 0.530 28,263 0.960 −0.43***

LFDIR 33,680 0.320 28,263 0.520 −0.20***
FSD 33,680 0.450 28,263 0.840 −0.39***
LFSD 33,680 0.270 28,263 0.470 −0.19***
P FDIR 33,680 0.070 28,263 0.100 −0.03***
P FSD 33,680 0.070 28,263 0.110 −0.04***
FCEO 33,680 0.020 28,263 0.020 0
FCHAIR 33,680 0.020 28,263 0.020 0.01***
CEODUAL 33,680 0.480 28,263 0.510 −0.03***
FCEODUAL 33,680 0.010 28,263 0.010 0.00***
FD 33,680 0.370 28,263 0.550 −0.18***
FD3 33,680 0.030 28,263 0.090 −0.06***
LPSD 33,680 0.330 28,263 0.360 −0.03***
LDIR 33,680 1.910 28,263 2.130 −0.22***
LOGTA 33,668 11.950 28,263 13.780 −1.83***
LEV 33,565 0.130 28,227 0.170 −0.04***
TOBINSQ 33,423 2.250 28,180 1.810 0.44***
ROA 33,135 −7.490 27,987 6.980 −14.48***
RETE 33,339 −130.4 26,789 44.340 −174.71***
Cash 33,556 0.260 28,184 0.130 0.13***
STDRET 31,728 0.170 27,207 0.110 0.07***
MB 33,410 0.370 28,180 0.180 0.19***
LegCom 33,617 0.930 28,008 0.790 0.14***
INST 25,832 31.190 21,722 35.520 −4.32***
Own 33,680 26.840 28,263 28.660 −1.83***
Variables are as defined in Table 2. ***, ** and * represent significance at the l%, 5%, and 10% levels, respectively.
• FCHAIR: whether the board chairperson is female;

• FCEODUAL: whether the CEO and chairperson are the same female.
The rest of this section details the control variables. We use natural logarithm of corporate total assets (LOGTA) to measure
corporate size. Firm size was previously identified in the studies of U.S. listed companies by Fama and French (2001) and Denis and
Osobov (2008) to be significantly positively associated with payout policy. There are two main reasons for this. First, large firms have
more net profits relative to small firms, thus creating conditions for dividend payment. Second, large firms are more willing to reduce
the agency cost, and so tend to adopt dividend payments as a means to mitigate the agency problem. Following prior literature, we
expect a positive association between corporate size and dividend payouts.
We use LEV to represent corporate financial leverage. Previous studies show that highly leveraged firms tend to pay smaller
dividends (Jensen et al., 1992). This is because they require internal cash flow to promptly repay debts to creditors. Specifically, since
highly leveraged firms face quite expensive transaction costs for external refinancing, they must ensure internal funds are adequate
for due performance. Moreover, such firms are undoubtedly subject to closer external supervision from creditors. Thus, we assume
that the association between corporate financial leverage and payout policy is negative.
TOBINSQ represents Tobin's Q, and MB is defined as market-to-book ratio. Fama and French (2001) propose corporate growth
opportunity as a proxy for investment opportunity. In their empirical study of U.S. listed companies during 1978–1999, high growth
opportunities led to an increase in the number of firms paying dividends. DeAngelo and DeAngelo (2006) argue that U.S. listed firms
with higher MB pay substantially smaller dividends. Officer (2011) also suggests that firms with a lower Tobin's Q are less likely to
pay dividends. While Chay and Suh (2009) found no significant relationship between corporate investment opportunity and payout
policy worldwide, Brav et al. (2005) highlight that the dividend payout ratio is significantly positively correlated with corporate
future earnings, while the corporate earnings level is determined by growth (investment) opportunity. On balance, we assume a
positive association between corporate investment opportunity and dividend payout policy.
We used ROA (net income/total assets) to measure current profitability. Many theoretical and empirical studies confirm the
smoothness of corporate payout policy. In Lintner's (1956) empirical study of U.S. listed companies, corporate size, the historical
payout policy, and stable corporate profitability were found to collectively determine the corporate dividend payment. Management
is less willing to increase (decrease) the dividend payment unless the firm has a stable increase (decrease) in surplus; therefore, firms
only pay dividends when they have adequate profits. Fama and Babiak (1968) and Baker and Wurgler (2004) evidence the re-
lationship between corporate profitability and payout policy among U.S. listed companies; their findings are supported by the
worldwide empirical analysis of Chay and Suh (2009). In view of these, we assume a positive relation between corporate profitability
and payout policy.
RETE measures the life cycle profitability of firms, calculated by the retained earnings-to-total common equity ratio. Numerous
empirical studies find that mature firms tend to have stronger profitability but limited investment opportunities, making them more
likely to pay dividends. By contrast, growing firms have comparatively more investment opportunities but limited resources;
9
consequently, they are reluctant to pay dividends (DeAngelo and DeAngelo, 2006; Fama and French, 2001). DeAngelo and DeAngelo
(2006) point out that firms with a lower ratio of retained earnings to total equity (RE/TE) are generally in the capital accumulation
stage and thus tend to pay little or even no dividends; by contrast, firms with higher RE/TE are mostly in the mature stage and
accumulate adequate profits, enabling them to pay larger dividends. Denis and Osobov (2008) concur that firms with higher RE/TE
are more inclined to pay dividends. Therefore, we expect a positive association between corporate life cycle and payout policy.
Cash represents cash holdings. Agency theory posits that management behavior is not always consistent with shareholder interests
(Fama and Jensen, 1983; Jensen and Meckling, 1976; Shleifer and Vishny, 1997). A higher dividend payment will diminish the cash
available to management, thereby preventing the misuse of corporate funds (Easterbrook, 1984). Therefore, for firms with larger cash
holdings, investors are more likely to force the management to pay dividends, which is supported by many empirical studies (Grullon
and Michaely, 2002; Jensen, 1986). However, Chay and Suh (2009) argue that the empirical results are not in line with the expected
relation between payout policy and cash holdings, since firms planning to invest in a large project also need to hold abundant cash,
rather than paying dividends. DeAngelo and DeAngelo (2006) also contend that the source of cash has a stronger effect on corporate
dividend payment compared to cash holdings. Dividend-paying firms are likely to hold less cash. In view of this, we assume a
negative relationship between corporate cash holdings and dividend payout policy.
Following Chay and Suh (2009), we use stock return volatility (STDRET) to measure the uncertainty risk of corporate cash flow.
Lintner (1956) suggests that management hesitates to pay a dividend due to possible future risk. Brav et al. (2005) further note that,
for management, corporate risk is a determinant of payout policy. Non-dividend-paying firms may suffer greater cash flow un-
certainty risk than dividend-paying firms (Jagannathan et al., 2000). Hoberg and Prabhala (2009) suggest that low-dividend-paying
firms tend to suffer higher uncertainty, which can explain the “dividend puzzle” of Fama and French (2001). Chay and Suh's (2009)
empirical analysis of 33 countries directly proves that the uncertainty of cash flow can significantly reduce dividend payments.
Therefore, we expect a negative association between corporate cash flow uncertainty and payout policy.
Hu and Kumar (2004) and Chen et al. (2017) propose that managerial entrenchment affects corporate payout policy, while earlier
studies suggest that smaller and more independent boards are better monitors of firms. Therefore, we include two specific board
variables, board size and the proportion of independent directors, to capture the effect of board independence and the quality of
corporate governance from the variance in the number of directors and a board's overall independence (Chen et al., 2017). We use
LDIR as the natural logarithm of the total number of a board's directors plus one, and LPSD as the natural logarithm of the number of
independent directors scaled by board size plus one.
4.1. Board gender diversity and dividend payout
In this section, we test Hypothesis 1 (that board gender diversity can significantly promote corporate dividend payment). Our
approach differs partly from that of Chen et al. (2017): whereas they discuss the relation between female independent directors and
dividend payouts based on U.S. environment, we investigate board gender diversity by considering all female directors, using data
from 22 countries.
Tables 5 and 6 present the regression results for board gender diversity and payout policy, respectively. In Panel A of Table 5, we
test the effect of board gender diversity on whether firms pay dividends. In column 1, we include only control variables in the
regression. In columns 2–6, we separately add five main board gender diversity indicators, including FDIR, the number of female
directors (column 2), P_FDIR, the proportion of female directors in the boardroom (column 3), FSD, the number of female in-
dependent non-executive directors (column 4), P_FSD, the proportion of female independent non-executive directors in the board-
room (column 5), and FD, the dummy variable of whether the board has at least one female director (column 6). We find that all
pseudo R2 values are improved to different degrees, indicating the incremental explanatory power of female directors on corporate
dividend policy. All five indicators of board gender diversity are positively significant at the 1% level in explaining dividend payout
policy. The results suggest that firms with a more gender-diversified board are more likely to pay cash dividends.
Panels B and C of Table 5 examine the impact of board gender diversity on two measures of the level of dividend payment:
DIVEARN and DIVTA, respectively. In column 1, we include only control variables in the regression. In columns 2–6, we add,
respectively, the following indicators of board gender diversity: FDIR (column 2), P_FDIR (column 3), FSD (column 4), P_FSD (column
5), and FD (column 6). We find that all R2-adjusted values are improved to varying degrees, indicating the incremental explanatory
power of female directors on dividend payment level. Furthermore, all indicators of board gender diversity tend to significantly
impact on the level of the dividend payment. Specifically, the magnitude of the significant coefficient of FDIR indicates that a one
standard deviation increase in FDIR increases DIVEARN by 4.35% (0.008/0.184) in Panel B and DIVTA by 76.9% (0.01/0.013) in
Panel C, relative to their average values. The magnitude of the significant coefficient of P_FDIR indicates that a one standard de-
viation increase in P_FDIR increases DIVEARN by 31.5% (0.058/0.184) and DIVTA by 38.5% (0.005/0.013), relative to their average
values. Overall, these results indicate that the firms with a more gender diversified board are more likely to increase the cash
dividend payout ratio.
We test the effects of alternative board gender diversity indicators on whether firms pay dividends in Panel A of Table 6. In
columns 1–6, we add, respectively, LFDIR, the natural logarithm of female directors plus 1 (column 1), LFSD, the natural logarithm of
female independent non-executive directors plus 1 (column 2), FCEO, female CEO (column 3), FCHAIR, female board chairperson
(column 4), FCEODUAL, the CEO and board chairperson being the same female (column 5), and FD3, the dummy variable of whether
the board has at least three female directors (column 6). The coefficients of LFDIR (column 1), LFSD (column 2), and FD3 (column 6)
are significantly positive; by contrast, FCEO (column 3), FCHAIR (column 4), and FCEODUAL (column 5) have insignificant effects on
corporate payout policy. As mentioned above, there are very few female CEOs and chairpersons in our sample, so these are not our
10
Table 5
Board gender diversity and dividend payout policy.
(1) (2) (3) (4) (5) (6)
Panel A DIV DIV DIV DIV DIV DIV
FDIR 0.104***
(4.17)
P_FDIR 0.718***
(3.51)
FSD 0.109***
(3.85)
P_FSD 0.595***
(3.41)
FD 0.192***
(3.73)
LDIR 0.817*** 0.711*** 0.790*** 0.719*** 0.797*** 0.727***
(8.47) (7.28) (8.23) (7.37) (8.29) (7.47)
LPSD −0.279 −0.238 −0.240 −0.210 −0.242 −0.225
(−1.47) (−1.25) (−1.27) (−1.11) (−1.28) (−1.18)
LOGTA 0.226*** 0.215*** 0.218*** 0.214*** 0.217*** 0.220***
(11.00) (10.42) (10.59) (10.32) (10.50) (10.68)
LEV −0.252 −0.222 −0.232 −0.221 −0.229 −0.237
(−1.49) (−1.31) (−1.37) (−1.30) (−1.35) (−1.40)
TOBINSQ 0.033 0.028 0.029 0.028 0.029 0.029
(0.90) (0.78) (0.81) (0.77) (0.81) (0.80)
ROA 0.037*** 0.037*** 0.037*** 0.037*** 0.037*** 0.037***
(15.60) (15.66) (15.63) (15.68) (15.63) (15.64)
RETE 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001***
(7.74) (7.74) (7.74) (7.75) (7.74) (7.74)
Cash −1.136*** −1.135*** −1.136*** −1.127*** −1.128*** −1.130***
(−6.98) (−6.98) (−6.98) (−6.94) (−6.94) (−6.94)
STDRET −11.292*** −11.259*** −11.269*** −11.257*** −11.280*** −11.249***
(−26.20) (−26.18) (−26.18) (−26.16) (−26.19) (−26.13)
MB −0.065 −0.066 −0.065 −0.065 −0.066 −0.063
(−0.93) (−0.94) (−0.93) (−0.92) (−0.93) (−0.90)
_cons −4.381*** −4.125*** −4.307*** −4.129*** −4.299*** −4.241***
(−15.43) (−14.37) (−15.18) (−14.34) (−15.14) (−14.88)
YEAR Yes Yes Yes Yes Yes Yes
IND Yes Yes Yes Yes Yes Yes
COUNTRY Yes Yes Yes Yes Yes Yes
N 56,161 56,161 56,161 56,161 56,161 56,161
r2_p 0.366 0.367 0.366 0.366 0.366 0.366
Panel B DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN
FDIR 0.008***
(3.06)
P_FDIR 0.058***
(2.62)
FSD 0.008***
(2.93)
P_FSD 0.059***
(3.16)
FD 0.017***
(3.09)
LDIR 0.058*** 0.051*** 0.057*** 0.052*** 0.057*** 0.051***
(5.65) (4.86) (5.46) (4.93) (5.49) (4.86)
LPSD −0.030 −0.027 −0.027 −0.026 −0.027 −0.026
(−1.44) (−1.33) (−1.32) (−1.24) (−1.30) (−1.26)
LOGTA 0.007*** 0.006*** 0.006*** 0.006*** 0.006*** 0.006***
(3.39) (2.99) (3.10) (2.93) (2.97) (3.13)
LEV 0.034** 0.036** 0.036** 0.036** 0.036** 0.035**
(1.97) (2.11) (2.06) (2.11) (2.09) (2.06)
TOBINSQ 0.005* 0.005* 0.005* 0.005* 0.005* 0.005*
(1.91) (1.82) (1.83) (1.81) (1.81) (1.83)
ROA 0.002*** 0.002*** 0.002*** 0.002*** 0.002*** 0.002***
(15.57) (15.67) (15.59) (15.71) (15.62) (15.59)
RETE 0.000* 0.000* 0.000* 0.000* 0.000* 0.000*
(1.81) (1.81) (1.81) (1.82) (1.82) (1.78)
11
Table 5 (continued)
Cash −0.011 −0.011 −0.011 −0.010 −0.011 −0.011

(−0.93) (−0.85) (−0.88) (−0.83) (−0.85) (−0.85)
STDRET −0.869*** −0.867*** −0.867*** −0.867*** −0.867*** −0.866***
(−26.90) (−26.89) (−26.89) (−26.87) (−26.89) (−26.90)
MB 0.022*** 0.021*** 0.022*** 0.021*** 0.022*** 0.022***
(3.26) (3.22) (3.25) (3.22) (3.24) (3.27)
_cons 0.538*** 0.552*** 0.539*** 0.552*** 0.542*** 0.546***
(7.50) (7.63) (7.49) (7.62) (7.52) (7.59)
N 56,158 56,158 56,158 56,158 56,158 56,158
r2_a 0.138 0.138 0.138 0.138 0.138 0.138
Panel C DIVTA DIVTA DIVTA DIVTA DIVTA DIVTA
FDIR 0.001***
(3.79)
P_FDIR 0.005***
(3.04)
FSD 0.001***
(3.63)
P_FSD 0.004***
(3.06)
FD 0.001**
(2.49)
LDIR 0.004*** 0.004*** 0.004*** 0.004*** 0.004*** 0.004***
(5.61) (4.63) (5.41) (4.69) (5.46) (5.00)
LPSD −0.003** −0.003* −0.003* −0.003* −0.003* −0.003*
(−2.02) (−1.88) (−1.88) (−1.75) (−1.88) (−1.89)
LOGTA −0.000 −0.000** −0.000* −0.000** −0.000* −0.000*
(−1.49) (−2.00) (−1.81) (−2.07) (−1.87) (−1.67)
LEV −0.005*** −0.005*** −0.005*** −0.005*** −0.005*** −0.005***
(−3.75) (−3.55) (−3.62) (−3.55) (−3.61) (−3.67)
TOBINSQ 0.000 0.000 0.000 0.000 0.000 0.000
(1.57) (1.49) (1.50) (1.48) (1.50) (1.53)
ROA 0.000*** 0.000*** 0.000*** 0.000*** 0.000*** 0.000***
(19.09) (19.17) (19.10) (19.21) (19.12) (19.10)
RETE 0.000*** 0.000*** 0.000*** 0.000*** 0.000*** 0.000***
(2.83) (2.83) (2.82) (2.84) (2.84) (2.80)
Cash −0.001 −0.001 −0.001 −0.001 −0.001 −0.001
(−0.70) (−0.62) (−0.65) (−0.59) (−0.64) (−0.65)
STDRET −0.060*** −0.059*** −0.059*** −0.059*** −0.059*** −0.059***
(−25.99) (−25.95) (−25.95) (−25.94) (−25.93) (−25.95)
MB 0.008*** 0.008*** 0.008*** 0.008*** 0.008*** 0.008***
(14.53) (14.48) (14.52) (14.49) (14.51) (14.55)
_cons 0.024*** 0.026*** 0.024*** 0.026*** 0.024*** 0.024***
(6.00) (6.56) (6.20) (6.58) (6.22) (6.20)
N 56,161 56,161 56,161 56,161 56,161 56,161
r2_a 0.246 0.247 0.247 0.247 0.247 0.246
Variables are as defined in Table 2. Z/T statistics are based on robust corporate clustering standard error. ***, ** and * represent significance at the l
%, 5% and 10% levels, respectively.
main research variables; we list them to enhance the results' integrity.

We also test the effects of these alternative board gender diversity indicators on DIVEARN and DIVTA as measures of the level of
corporate dividend payment (see Panels B and C of Table 6). In columns 1–6, we add LFDIR (column 1), LFSD (column 2), FCEO
(column 3), FCHAIR (column 4), FCEODUAL (column 5), and FD3 (column 6). All of those variables were found to have a sig-
nificantly positive impact on DIVTA, while all except FCEO (column 3) had a significant impact on DIVEARN. The magnitudes of the
coefficients suggest that a one standard deviation increase in LFDIR increases DIVEARN by 10.9% (0.020/0.184) and DIVTA by
15.4% (0.002/0.013) relative to their average values. Furthermore, a one standard deviation increase in LFSD increases DIVEARN by
11.4% (0.021/0.184) and DIVTA by 15.4% (0.002/0.013) relative to their average values. Overall, board gender diversity is sig-
nificantly positively related to the corporate dividend payment level.
Using various measures of board gender diversity, our results generally support the proposition that a gender-diverse board is
12
Table 6
Alternative board gender diversity index and dividend payout policy.
(1) (2) (3) (4) (5) (6)
Panel A DIV DIV DIV DIV DIV DIV
LFDIR 0.243***
(4.47)
LFSD 0.251***
(4.25)
FCEO 0.098
(0.60)
FCHAIR 0.212
(1.30)
FCEODUAL 0.288
(1.13)
FD3 0.253***
(2.79)
_cons −4.126*** −4.118*** −4.382*** −4.386*** −4.385*** −4.277***
(−14.39) (−14.31) (−15.43) (−15.44) (−15.44) (−14.96)
CONTROLS Yes Yes Yes Yes Yes Yes
N 56,161 56,161 56,161 56,161 56,161 56,161
r2_p 0.367 0.367 0.366 0.366 0.366 0.366
LFDIR 0.020***
(3.44)
LFSD 0.021***
(3.38)
FCEO 0.012
(0.68)
FCHAIR 0.030*
(1.89)
FCEODUAL 0.049*
(1.95)
FD3 0.013
(1.26)
_cons 0.553*** 0.553*** 0.538*** 0.538*** 0.538*** 0.542***
(7.64) (7.65) (7.50) (7.49) (7.49) (7.49)
N 56,158 56,158 56,158 56,158 56,158 56,158
r2_a 0.139 0.139 0.138 0.138 0.138 0.138
LFDIR 0.002***
(3.61)
LFSD 0.002***
(3.48)
FCEO 0.003**
(1.99)
FCHAIR 0.002*
(1.79)
FCEODUAL 0.005**
(2.27)
FD3 0.003***
(3.32)
_cons 0.025*** 0.026*** 0.024*** 0.024*** 0.023*** 0.025***
(6.46) (6.48) (6.01) (6.00) (5.99) (6.34)
13
Table 6 (continued)
N 56,161 56,161 56,161 56,161 56,161 56,161

r2_a 0.247 0.247 0.246 0.246 0.246 0.247
Variables are as defined in Table 2. ***, ** and * represent significance at the l%, 5%, and 10% levels, respectively.
more likely to pay a dividend compared to a single-gender (all-male) board; moreover, firms with higher board gender diversity tend
to pay larger dividends. The empirical results thus support Hypothesis 1.
4.2. Institutional environment and dividend payout
In this section, we test Hypothesis 2 (that the influence of board gender diversity on dividend payouts may be smaller in good
institutional environments). La Porta et al. (1998) provide effective tools to compare the institutional basis across countries and
investigate the basic effect of the legal environment. For the indicator of institutional environments, we first use the dummy variable
of national law system defined by La Porta et al. (1998) to measure a country's corporate governance level: LegCom equals 1 if the
country applies common law, and 0 otherwise La Porta et al. (1998) show that common law countries have the best legal protection
of shareholders and civil law countries, the worse.
We then test the effect of board gender diversity on corporate payout policy after considering institutional environments mea-
sured by LegCom in Appendix B. The dependent variables are DIV, DIVEARN, and DIVTA. We measure board gender diversity by
FDIR, P_FDIR, FSD, P_FSD, and FD.
As shown in Appendix B, the dependent variable of Panel A is DIV, measuring whether firms pay dividends, while the dependent
variables in Panels B and C are DIVEARN and DIVTA, respectively, measuring the level of dividend payment. The results indicate
that, after controlling for institutional environments, the indicators of board gender diversity still exert a significant effect on cor-
porate payout policy; namely, firms with higher board gender diversity are more likely to pay dividends and have higher levels of
dividend payments. The magnitudes of the coefficients imply that a one standard deviation increase in P_FDIR increases DIVEARN by
31% (0.057/0.184) and DIVTA by 38.5% (0.005/0.013) relative to their average values. A one standard deviation increase in FSD
causes a 4.34% (0.008/0.184) increase in DIVEARN and a 7.7% (0.001/0.013) increase in DIVTA relative to their average values.
Additionally, we find that the institutional environments indicator LegCom is significantly and negatively related to corporate payout
policy. We test the interaction between institutional environment and gender diversity later in this section.
4.3. Further tests of institutional environment
Following Cumming et al. (2011) and Dang et al. (2015), we conduct a further array of tests related to different institutional
environments. The effect of the legal system on firm-level payout policy is mainly reflected in investor protection; thus, we divide the
sample into good legal environment countries (Good IE) and bad legal environment countries (Bad IE), assigning those in which the
strength of investor protection (WEF_InvPro) is above the median value of all countries to the former category and others to the latter.
Appendix C presents the results of comparison tests of dividend payout policy for different legal environments. The mean values of
DIV, DIVEARN, and DIVTA for the WEF_InvPro indicators differ significantly in Panels A, B and C, respectively. On average, DIV,
DIVEARN, and DIVTA are smaller in Good IE countries than in Bad IE countries. For example, a firm in a weak investor-protection
country (WEF_InvPro ≤8) has greater values of DIV, DIVEARN, and DIVTA. This demonstrates, in accordance with La Porta et al.'s
(2000) “substitute model,” that in a poor system environment, paying dividends is an alternative way to build corporate reputation.
The mean values of DIV, DIVEARN, and DIVTA between subsamples differ significantly. The results indicate that firms in
countries with a weak institutional environment tend to pay dividends. This may evidence a substitute effect between dividend policy
and investor protection law: that is, firms in countries with weak investor protection tend to pay dividends. This is in line with the
adoption of mandatory dividends in civil law countries with generally weaker investor protection (La Porta et al., 2000), as reinforced
by Fatemi and Bildik's (2012) finding that 65% of firms in civil law countries paid dividends in 2006, compared to 40% of firms in
common law countries.
We then reran the regressions to consider the possible mediation effect of institutional environment on dividend payout policy via
board gender diversity. La Porta et al. (2000) provide effective methods to compare institutional foundations between countries and
research the basic effects of the legal environment. A good institutional environment promotes corporate governance, thus improving
investor protection and mitigating agency problems (La Porta et al., 2000). In terms of gender equality, national regulations can
significantly improve opportunities for females to participate in corporate management. In a study on labor laws in 85 countries,
Botero et al. (2004) propose that the female labor participation rate is higher than that of males in countries with better employment
protection law, which allows females to find an optimal balance between career and family. We can, thus, assume that institutional
environment may exert a mediation effect on dividend payout policy by affecting board gender diversity. If macro corporate gov-
ernance affects this relation, LegCom can explain country fixed effects on firm-level payout policy.
As shown in Table 7, the estimates of gender diversity indicators are mostly significant and positive. A one standard deviation
increase in FDIR increases DIVEARN by 13.6% (0.025/0.184) relative to its average value, while a one standard deviation increase in
P_FDIR increases DIVEARN by 117.9% (0.217/0.184) relative to its average value. The interactive items of gender diversity
14
Table 7
Interactions between macroscopic corporate governance and board gender diversity.
(1) (2) (3) (4) (5)
Panel A DIV DIV DIV DIV DIV
FDIR 0.152***
(2.86)
fdir_legcom −0.055
(−0.94)
P_FDIR 1.206***
(2.61)
pfdir_legcom −0.559
(−1.10)
FSD 0.126**
(2.27)
fsd_legcom −0.019
(−0.31)
P_FSD 0.512
(1.47)
pfsd_legcom 0.103
(0.26)
FD 0.372***
(3.44)
fd_legcom −0.200*
(−1.68)
LegCom −1.718*** −1.704*** −1.721*** −1.737*** −1.661***
(−6.79) (−6.69) (−6.81) (−6.83) (−6.50)
_cons −2.413*** −2.618*** −2.416*** −2.573*** −2.582***
(−6.15) (−6.69) (−6.15) (−6.57) (−6.58)
CONTROLS Yes Yes Yes Yes Yes
YEAR Yes Yes Yes Yes Yes
IND Yes Yes Yes Yes Yes
COUNTRY Yes Yes Yes Yes Yes
N 55,861 55,861 55,861 55,861 55,861
r2_p 0.366 0.366 0.366 0.366 0.366
Panel B DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN
FDIR 0.025***
(4.46)
fdir_legcom −0.021***
(−3.50)
P_FDIR 0.217***
(3.51)
pfdir_legcom −0.184***
(−2.82)
P_FSD 0.107**
(2.56)
pfsd_legcom −0.057
(−1.24)
FD 0.051***
(3.81)
fd_legcom −0.039***
(−2.73)
FSD 0.024***
(3.97)
fsd_legcom −0.019***
(−2.98)
LegCom −0.298*** −0.295*** −0.317*** −0.296*** −0.301***
(−5.07) (−5.01) (−5.43) (−5.01) (−5.13)
_cons 0.381*** 0.361*** 0.385*** 0.370*** 0.383***
(5.78) (5.49) (5.89) (5.64) (5.82)
N 55,858 55,858 55,858 55,858 55,858
r2 0.140 0.140 0.140 0.140 0.140
15
Table 7 (continued)
Panel C DIVTA DIVTA DIVTA DIVTA DIVTA
FDIR 0.001***
(2.64)
fdir_legcom −0.001
(−1.38)
P_FDIR 0.015***
(2.87)
pfdir_legcom −0.011**
(−2.06)
FSD 0.001**
(2.51)
fsd_legcom −0.001
(−1.27)
P_FSD 0.009**
(2.43)
pfsd_legcom −0.005
(−1.31)
FD 0.002**
(2.10)
fd_legcom −0.001
(−1.23)
LegCom −0.019*** −0.018*** −0.019*** −0.019*** −0.019***
(−5.50) (−5.21) (−5.47) (−5.67) (−5.63)
_cons 0.034*** 0.032*** 0.034*** 0.033*** 0.033***
(8.65) (7.87) (8.65) (8.29) (8.31)
N 55,861 55,861 55,861 55,861 55,861
r2_a 0.247 0.247 0.247 0.247 0.247
Definitions of all variables are given in Table 2. Z/T statistics are based on robust corporate clustering standard error. ***, ** and * represent
significance at the l%, 5% and 10% levels, respectively.
indicators and LegCom are negative with unstable significance, whereas the coefficients of gender diversity indicators are positive
and significant. This indicates that there should be a large difference in the effects of board gender diversity on corporate payout
policy among countries with different institutional environments. Our results thus support Hypothesis 2.
4.4. Institutional ownership and dividend payout
In a perfect market, ownership structure should not affect corporate governance (Miller and Modigliani, 1961). However, in-
stitutional investors have obvious advantages of corporate supervision in the complicated real world. They are prone to use pro-
fessional knowledge and control rights to play the role of external supervisor, thereby reducing the agency cost (An and Zhang, 2013;
Jensen and Meckling, 1976). For instance, institutional investors can directly communicate with the management team and vote in
general shareholder meetings (Firth et al., 2016). They are also likely to force the management to pay dividends by threatening to
withdraw their investment (Edmans and Manso, 2011). Since the sale of a large number of the firm's shares by institutional investors
will likely harm the share price, management will usually endeavor to prevent such a scenario. Smith (1996) argues that institutional
investors with a larger shareholding are more motivated to supervise the management, and thereby improve corporate value. Both
internal and external governance mechanisms constrain management behavior and decision-making (Armstrong et al., 2010).
Many studies argue that institutional ownership can affect the corporate dividend policy. In their study of Canadian listed
companies, Espen Eckbo and Verma (1994) find that firms with stronger voting rights for institutional investors are likely to pay
higher dividends. Allen et al. (2000) find that firms with higher institutional ownership are more willing to pay dividends to retain
good shareholders. Short et al. (2002) also prove the existence of this relation among U.K. listed companies. In an empirical study of
U.S. listed companies, Grinstein and Michaely (2005) observe that institutional investors avoid investing in firms that do not pay any
dividends but are prone to invest in lower-paying firms. Crane et al. (2016) also suggest that firms with higher institutional ownership
tend to pay larger dividends.
Appendix D reports the results on how board gender diversity affects corporate dividend payouts after controlling for the pro-
portion of institutional ownership. The separate dependent variables are DIV, DIVEARN, and DIVTA. Based on the median of in-
stitutional ownership, we classify the sample into two groups (low and high institutional ownership) and separately investigate the
influence of board gender diversity on corporate dividend policy in each. In the three panels, INST is defined as the percentage of
total issued shares of 5% or more held strategically and not available to ordinary investors. We also measure board gender diversity
by: FDIR (columns 1 and 2), P_FDIR (columns 3 and 4), FSD (columns 5 and 6), P_FSD (columns 7 and 8), and FD (columns 9 and 10).
The results show that, after controlling for institutional ownership, all indicators of board gender diversity are significantly
16
positively related to dividend payout policy in firms with low institutional ownership. In firms with high institutional ownership,
most board gender diversity indicators have a significant positive influence on payout policy. Specifically, when DIVEARN is used as
the dependent variable, the coefficients of P_FSD (column 4 in Panel B) are positive but not significant. The magnitudes of the
coefficients imply that a one standard deviation increase in FDIR increases DIVEARN by 4.89% (0.009/0.184) in the low group and
2.17% (0.004/0.184) in the high group, relative to their respective mean values. A one standard deviation increase in P_FDIR causes a
29.3% (0.054/0.184) increase in DIVEARN in the low group and a 20.1% (0.037/0.184) increase in DIVEARN in the high group,
relative to their respective mean values. We also notice that the regression coefficients are greater for the low institutional ownership
subsample than for the high institutional ownership subsample, meaning that the effects of board gender diversity tend to be greater
with lower institutional investor ownership. This illustrates that, to a certain degree, there is a substitute effect between a gender-
diverse board and institutional ownership.
4.5. Insider ownership and dividend payout
Equity structure plays a fundamental role in corporate governance (Shleifer and Vishny, 1997). Healy and Palepu (2001) explore
the optimal contracts mitigating the agency problem between shareholders and insiders. They find that optimal contracts help to
motivate managers to align their interests with shareholders, thereby further diminishing corporate agency costs. A strong board can
also act as a good supervisor by forcing management to make more decisions aligned with shareholder interests. The possible threat
of an external hostile takeover as corporate performance declines may improve capital utilization efficiency. Chetty and Saez (2005)
suggest that when management holds higher ownership and takes several methods of stock option incentive, the firm's dividend
payment level tends to improve. Using the same method, Brown et al. (2007) research the association between management own-
ership and corporate dividend payment; they find an obvious increase in total dividend payments if the firm's top five executives hold
numerous equities, following the dividend tax reduction in 2003. Therefore, we assume that if the management owns corporate
shares, agency conflicts are likely to be reduced. Because managers are restricted from selling their shares, paying dividends offers
another possible way to obtain returns. However, Chay and Suh (2009) find no strong evidence that management ownership in-
fluences dividend payout policy.
Appendix E shows the relationship between board gender diversity and dividend payouts after controlling for the management
holding shares. The separate dependent variables are DIV, DIVEARN, and DIVTA. Based on the median of management ownership,
we classify the sample into two groups and separately investigate the influence of board gender diversity on corporate dividend
policy in each. Own is defined as the proportion of management ownership. We measure board gender diversity by: FDIR, P_FDIR,
FSD, P_FSD, and FD.
The results suggest that, after controlling for management ownership, the five indicators of board gender diversity are all sig-
nificantly positively correlated with dividend policy. The coefficients of board gender diversity are slightly greater in Panel B (Own
value higher than the median) than in Panel C (Own value less than or equal to the median). This suggests that firms with a larger
number of female directors are more inclined to pay dividends when management investors hold a larger ownership stake. This
further supports our hypothesis and illustrates that, in firms whose female senior executives have substantial shareholdings, the
governance effect of management ownership on agency problems tends to be stronger.
When DIV is the dependent variable (Panel A), the coefficients of Own are positive but not significant. In Panels B and C, the
regression coefficients of Own on DIVEARN and DIVTA are negative for median values of Own or less but positive for above-median
values of Own. This suggests that where the management holds relatively few shares, they have strong motivation to reduce the
dividend payout ratio, further increase corporate free cash flow, and consequently increase agency problems. Once management
ownership reaches a certain quota, the interests of the management and shareholders tend to be consistent; at this point, agency
conflicts are correspondingly diminished and the management is more likely to obtain returns through paying dividends (Brown
et al., 2007).
4.6. Consideration of share repurchases
Many studies have shown that an increasing number of firms substitute dividends by repurchasing shares (Bond and Zhong, 2016;
Dittmar and Field, 2015; Fama and French, 2001; Grullon and Michaely, 2002; Hillert et al., 2016; Jagannathan et al., 2000; Young
and Yang, 2011). Fama and French (2001) suggest that the dividend payout ratio in U.S. listed companies is gradually diminishing,
but Grullon and Michaely (2002) counter that the total payout level is not actually decreasing as many cash dividends are replaced by
share repurchases. Other studies indicate a tendency in European listed companies to reduce the dividend payout ratio (Von Eije and
Megginson, 2008). Share repurchases have smaller risk sensitivity and allow firms greater flexibility in their dividend policy (Brav
et al., 2005).
Investors prefer firms with the characteristics of safety, such as dividend payout, when the dividend premium is high, but favor
firms that pursue maximum capital appreciation when the dividend premium is low (Baker and Wurgler, 2004). Li and Lie (2006)
confirm that firms with a low dividend premium prefer repurchasing shares to paying dividends. Firms with high cash-flow un-
certainty are more likely to choose share repurchases instead of dividends as a payout to investors (Chay and Suh, 2009). Therefore,
following the methods of Chay and Suh (2009), Floyd et al. (2015), and Evgeniou and Vermaelen (2017), this paper adopts STR as a
dummy variable that equals 1 if a firm repurchases shares (and 0 otherwise). STR is substituted for DIV, and the sum of cash
dividends and share repurchases substituted for dividend payout ratio. The results presented in Table 8 show that there is a sig-
nificantly positive association between board gender diversity and corporate dividend payouts after controlling for share repurchases.
17
Table 8
Stock repurchase and dividend payout.
(1) (2) (3) (4) (5)
Panel A STR STR STR STR STR
FDIR 0.076***
(4.21)
P_FDIR 0.619***
(4.07)
FSD 0.077***
(3.85)
P_FSD 0.525***
(4.15)
FD 0.153***
(4.04)
_cons −2.400*** −2.536*** −2.407*** −2.532*** −2.485***
(−11.48) (−12.42) (−11.47) (−12.39) (−12.07)
N 56,161 56,161 56,161 56,161 56,161
r2_p 0.201 0.201 0.201 0.201 0.201
Panel B TotalEarn TotalEarn TotalEarn TotalEarn TotalEarn
FDIR 0.031***
(6.84)
P_FDIR 0.242***
(6.35)
FSD 0.035***
(6.98)
P_FSD 0.205***
(6.44)
FD 0.061***
(6.93)
_cons 0.319*** 0.266** 0.322*** 0.274*** 0.288***
(3.06) (2.56) (3.08) (2.65) (2.80)
N 56,158 56,158 56,158 56,158 56,158
r2_a 0.109 0.109 0.109 0.109 0.109
Panel C TotalTA TotalTA TotalTA TotalTA TotalTA
FDIR 0.003***
(7.61)
P_FDIR 0.021***
(6.68)
FSD 0.003***
(7.64)
P_FSD 0.016***
(6.39)
FD 0.004***
(6.16)
_cons 0.009 0.004 0.010* 0.004 0.004
(1.53) (0.62) (1.67) (0.64) (0.68)
N 56,161 56,161 56,161 56,161 56,161
r2_a 0.235 0.234 0.235 0.234 0.233
Definitions of all variables are given in Table 2. Z/T statistics are based on robust corporate clustering standard error. ***, ** and * represent
significance at the l%, 5%, and 10% levels, respectively.
18
The magnitude of the significant coefficient of FDIR indicates that a one standard deviation increase in FDIR causes a 9.37% (0.031/
0.331) increase in TOTALEARN (Panel B) and an 11.5% (0.003/0.026) increase in TOTALTA (Panel C) relative to their respective
average values. The magnitude of P_FDIR's significant coefficient indicates that a one standard deviation increase in P_FDIR increases
TOTALEARN by 73.1% (0.242/0.331) and TOTALTA by 80.8% (0.021/0.026) relative to their average values.
We also conducted a difference test of SHR, finding that mean values of board gender diversity indicators are greater in firms with
share repurchases than in those without (see Appendix F). For example, the average FDIR is 0.600 for non-SHR firms and 1 for SHR
firms (t-stat = −0.4 for the mean difference).
4.7. Dividend initiations and increases
Besides the level of dividends, we also study the impact of female directors on dividend initiations and increases. Several studies
suggest that once firms start paying dividends, they rarely stop. Smoothing dividends over time is the most common dividend policy;
therefore, we carefully verify the characteristics of our dividend data again. We find that some companies may suddenly stop paying
dividends: for instance, during the dividend payment period of 1985 to 2013, Globeride Inc. in Japan stopped paying dividends from
2001 to 2007. However, such cases are very rare in the full sample and do not affect the overall measurement. Thus, to measure the
effect of gender-diverse boards on dividend initiations, we classify a firm as a PRIORPAYER in year t if it has continuously paid
dividends in year t − 1 and year t, and then substitute PRIORPAYER for DIV. For example, if Company A continuously paid dividends
from 2004 to 2008, and Company B continuously paid dividends from 2006 to 2008, we calculate that there were six priorpayers and
four non-payers during this period. Our empirical results in Panel A of Table 10 report that all estimates of board gender diversity
indicators are significantly positive, indicating a significantly positive relation between a gender-diverse board and dividend in-
itiations.
To investigate the effect of board gender diversity on dividend increases, we create two dummy variables of whether a firm
increases its dividend: INDIVEA and INDIVTA. INDIVEA equals 1 when DIVEARN increases relative to last year's value (and 0
otherwise), and INDIVTA equals 1 when DIVTA increases relative to last year's value (and 0 otherwise). Substituting INDIVEA for
DIVEARN and INDVTA for DIVTA, we rerun the logit regression. Panels B and C of Table 9 show that board gender diversity is
significantly positively correlated with INDIVEA and INDIVTA, respectively.
Overall, our empirical results evidence a significantly positive relationship between board gender diversity and dividend in-
itiations and increases.
4.8. Endogeneity problems
4.8.1. Omitted variables

Our findings are consistent with the results of Chen et al. (2017) for U.S. listed companies. However, our conclusions still face the
challenge of endogeneity problems, the first being omitted variables. Though we use industry, year, and state dummy variables to
control for the potential determinants of dividend payment policy, our conclusions may still be affected by unobserved variables. To
mitigate the effect of omitted variables on our conclusions, we must first find a suitable IV not directly correlated with corporate
payout policy but able to directly influence board gender diversity. For this purpose, we follow Lin et al. (2013), Chen (2015), and An
et al. (2016) by using IndMeanFDIR as the IV. Using data for 82,644 listed companies between 1965 and 2008 from the Center for
Research in Security Prices (CRSP), Leary and Roberts (2014) find that listed firms imitated one another's financial decisions and
were greatly affected by the financial decisions of peers in the same industry. Furthermore, many corporate finance studies have used
the industrial average to construct an exogenous instrument variable (Adhikari and Agrawal, 2018; Chen, 2015; Eom, 2018; Hasan
and Cheung, 2018; Huang and Mazouz, 2018; Jiang and Yuan, 2018; Kim et al., 2017; Ward et al., 2018; Zhang et al., 2016). For
example, Faccio et al. (2016) take the percentage of a firm's peers with a female CEO as the instrument variable for firm CEOs. Huang
and Mazouz (2018) propose that firms in the same industry tend to adopt similar corporate policies, and so use the natural logarithm
of industry average excess cash as the instrument variable of the firm's excess cash. Eom (2018) uses the natural logarithm of industry
average oversubscription in the five most recent IPOs as the IV of oversubscription. Meanwhile, Hasan and Cheung (2018) regard the
industry-level mean organization capital in each year as an instrument variable. Following those studies, we believe that In-
dMeanFDIR is suitable as this paper's IV: board gender diversity of firms in the same industry may be similar and closely associated
with the industry-level board gender diversity. The industry-level board gender diversity directly affects the board gender diversity of
each firm in this industry but cannot directly affect corporate dividend policy. Table 10 shows that IndMeanFDIR has a significantly
positive impact on board gender diversity indicators, including FDIR, P_FDIR, FSD, P_FSD, and FD. The results indicate that In-
dMeanFDIR should be a valid instrument. Although this method is unlikely to fully solve endogeneity problems, it can partially
alleviate the effect of omitted variables, which do not vary with time according to our conclusions.
Selecting DIVEARN as the dependent variable and the board gender diversity indicators FDIR, P_FDIR, FSD, P_FSD, and FD as
endogenous variables, we use the two-stage least squares (2SLS) and the generalized method of moment estimation (GMM) to run the
regression. The results presented in Table 10 and Appendix G indicate that board gender diversity indicators exert significant effects
on the level of dividend payouts. We also find that with both the 2SLS and GMM, all coefficients of the board gender diversity
indicators are greater than the OLS estimates after controlling for the effects of omitted variables, implying that the OLS probably
underestimates the effects of board gender diversity on dividend policy. Therefore, this study's results may be affected by omitted
variables, but our conclusions remain valid.
We also use PSM and DID to test the effect of the exogenous shock of board gender quota law being introduced (Li et al., 2017).
19
Table 9
Consideration of dividend initiations and dividend increases.
(1) (2) (3) (4) (5)
Panel A PRIORPAYER PRIORPAYER PRIORPAYER PRIORPAYER PRIORPAYER
FDIR 0.111***
(4.38)
P_FDIR 0.800***
(3.85)
FSD 0.116***
(4.05)
P_FSD 0.658***
(3.72)
FD 0.218***
(4.12)
LDIR 0.769*** 0.852*** 0.778*** 0.861*** 0.781***
(7.70) (8.69) (7.79) (8.76) (7.85)
LPSD −0.318 −0.319 −0.289 −0.322* −0.301
(−1.63) (−1.64) (−1.48) (−1.65) (−1.54)
LOGTA 0.219*** 0.222*** 0.217*** 0.220*** 0.223***
(10.33) (10.48) (10.22) (10.39) (10.58)
LEV −0.263 −0.272 −0.262 −0.270 −0.278
(−1.51) (−1.57) (−1.50) (−1.55) (−1.60)
TOBINSQ 0.032 0.033 0.031 0.033 0.033
(0.86) (0.88) (0.84) (0.88) (0.88)
ROA 0.037*** 0.037*** 0.037*** 0.037*** 0.037***
(15.54) (15.51) (15.58) (15.53) (15.53)
RETE 0.002*** 0.002*** 0.002*** 0.002*** 0.002***
(7.97) (7.96) (7.97) (7.96) (7.96)
Cash −1.286*** −1.287*** −1.276*** −1.278*** −1.280***
(−7.49) (−7.49) (−7.45) (−7.45) (−7.45)
STDRET −11.816*** −11.827*** −11.814*** −11.839*** −11.804***
(−26.26) (−26.26) (−26.24) (−26.27) (−26.21)
MB −0.103 −0.103 −0.102 −0.103 −0.100
(−1.43) (−1.42) (−1.41) (−1.43) (−1.39)
_cons −4.312*** −4.506*** −4.317*** −4.497*** −4.430***
(−14.64) (−15.46) (−14.61) (−15.42) (−15.13)
N 56,161 56,161 56,161 56,161 56,161
r2_p 0.377 0.376 0.376 0.376 0.376
Panel B INDIVEA INDIVEA INDIVEA INDIVEA INDIVEA
FDIR 0.056***
(4.03)
P_FDIR 0.552***
(4.36)
FSD 0.054***
(3.49)
P_FSD 0.455***
(4.24)
FD 0.136***
(4.19)
_cons −2.089*** −2.189*** −2.101*** −2.183*** −2.151***
(−12.24) (−13.05) (−12.26) (−13.01) (−12.80)
N 47,904 47,904 47,904 47,904 47,904
r2_p 0.151 0.151 0.151 0.151 0.151
Panel C INDIVTA INDIVTA INDIVTA INDIVTA INDIVTA
FDIR 0.049***
(3.63)
P_FDIR 0.426***
20
Table 9 (continued)
Panel C INDIVTA INDIVTA INDIVTA INDIVTA INDIVTA
(3.41)
FSD 0.051***
(3.35)
P_FSD 0.345***
(3.30)
FD 0.103***
(3.08)
_cons −2.190*** −2.284*** −2.191*** −2.280*** −2.256***
(−12.38) (−13.15) (−12.35) (−13.13) (−12.95)
N 47,904 47,904 47,904 47,904 47,904
r2_p 0.153 0.153 0.153 0.153 0.153
%, 5%, and 10% levels, respectively.
Since Norway was the first European country to legislate a gender quota for female board directors, we take Norway as the treatment
group and all other countries in the sample as the control group in Panel A of Table 11, where t is 2003 (when Norway passed its
board gender quota bill). In Panel B, Norway and Spain are the treatment group and all other countries are the control group.
Table 11 shows the results of PSM-DID analysis of payout policy across countries. Panel A reports statistically significant and positive
estimates of DIV, DIVEARN, and DIVTA for all countries. The PSM-DID estimates (p-values) are 0.135 (0.000), 0.105 (0.000), and
0.006 (0.000) for DIV, DIVEARN, and DIVTA, respectively. Panel B reports PSM-DID estimates (p-values) of 0.129 (0.000), 0.113
(0.000), and 0.005 (0.000) for DIV, DIVEARN, and DIVTA, respectively. The consistent results in Panels A and B show that in-
troducing a board gender quota has a significantly positive influence on corporate dividend payment across countries.
4.8.2. Reverse causality

A possible alternative explanation is that improving the dividend payment level is likely to prompt board gender diversity.
Following An and Zhang (2013), Byun et al. (2013), and Gul et al. (2011), we employ two methods to test for reverse causality. First,
we use the first-order lag of independent variables. Second, we devise a forecasting model of the number of female directors,
investigate the effect of the number (proportion) of unexplained female directors on dividend policy and then test for reverse
causality between the two, which is theoretically based on the number (proportion) of female directors and corporate characteristics
like ROE, FIRMAGE, and TOBINSQ having a linear relationship. According to Gul et al. (2011), a firm's characteristics that predict the
number (proportion) of female directors can be used to explain the majority of payout policy changes, meaning that the number
(proportion) of female directors can be regarded as a proxy for those characteristics. If the unexplained part of the number (pro-
portion) of female directors mostly explains payout policy changes, then that unexplained part should be causally linked with the
number (proportion) of female directors.
As presented in Appendices H–J, the results of the two methods show that all board gender diversity indicators have significantly
positive impacts on corporate dividend policy. This further illustrates that our conclusion is robust.
4.8.3. Selective bias

Our conclusions are likely to be affected by sample selection bias. The study's sample contains many U.S. companies due to the
limitation of the original data; thus, our conclusions are heavily influenced by U.S companies. Furthermore, the level of dividend
payment tends to be higher in firms with higher board gender diversity. In view of these considerations, we use two main methods to
test for sample selection bias. First, we exclude U.S. companies from the sample and rerun the regression. Second, we use PSM,
adopting the 1:2 nearest neighbor method and female director variables in the forecasting model (logit) for FD and FD3, respectively:
the treatment groups include boards with at least one female director (FD = 1) and boards with at least three female directors
(FD3 = 1). The results reported in Appendices K–L prove that our basic conclusion is robust.
5. Robustness tests
5.1. Financial crisis and dividend payout
Our sample period is from 2000 to 2013, and thus includes the global financial crisis commencing in 2007. While the exact start of
the crisis continues to be disputed by academics (Fahlenbrach and Stulz, 2011; Gonzalez, 2016; Ho et al., 2016), it undoubtedly had a
catastrophic impact on most companies (Bliss et al., 2015; Fahlenbrach and Stulz, 2011). Our focus here is on whether board gender
diversity still exerts a positive effect on corporate dividend policy after controlling for the influence of the financial crisis.
Following An and Zhang (2013), we define a dummy variable CRISIS as a proxy for financial crisis, which equals 1 when the
sample companies existed in 2007 and 2008, and 0 otherwise. Presented in Appendix M, our result is consistent with the research
21
Table 10
Endogeneity problems caused by omitted variables: 2SLS.
(1) (2) (3) (4) (5)
Panel A FDIR P_FDIR FSD P_FSD FD
IndMeanFDIR 0.873*** 0.091*** 0.804*** 0.096*** 0.318***

(34.33) (34.01) (32.59) (31.64) (29.79)
ROE −0.000*** 0.000 −0.001*** 0.000 −0.000***
(−2.73) (0.15) (−4.43) (0.09) (−3.07)
LOGTA 0.144*** 0.008*** 0.151*** 0.011*** 0.058***
(21.55) (11.45) (25.48) (14.90) (22.87)
FIRMAGE 0.076*** 0.005*** 0.084*** 0.007*** 0.031***
(7.53) (4.43) (9.37) (5.71) (6.31)
SALESGROWTH −0.010 −0.002** −0.010* −0.002* −0.017***
(−1.62) (−2.08) (−1.86) (−1.74) (−4.52)
DIRECTORSHIPS −0.000 −0.000 −0.004 −0.001 0.000
(−0.08) (−0.05) (−1.47) (−1.30) (0.01)
TOTALRISK −1.869*** −0.195*** −1.171*** −0.191*** −1.183***
(−4.67) (−3.79) (−3.37) (−3.17) (−5.54)
TOBINSQ 0.054*** 0.004*** 0.055*** 0.006*** 0.019***
(8.71) (6.17) (9.91) (6.83) (6.88)
RET −0.706*** −0.053*** −0.665*** −0.060*** −0.146***
(−7.25) (−4.92) (−7.40) (−4.67) (−3.32)
VWRETD 403.598** 0.031 457.631*** 6.525 −101.003
(2.26) (0.00) (2.65) (0.32) (−1.64)
_cons −1.966*** −0.098*** −2.147*** −0.145*** −0.586***
(−20.82) (−9.65) (−25.78) (−13.19) (−16.22)
N 55,084 55,084 55,084 55,084 55,084
r2_a 0.265 0.178 0.290 0.169 0.189
Panel B DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN
FDIR 0.516***
(7.03)
P_FDIR 6.951***
(5.66)
FSD 0.534***
(7.85)
P_FSD 4.831***
(6.35)
FD 1.550***
(5.76)
_cons 1.468*** 0.898*** 1.433*** 0.773*** 1.213***
(6.99) (4.97) (7.75) (5.47) (5.31)
N 52,439 52,439 52,439 52,439 52,439
FDIR 0.043***
(7.31)
P_FDIR 0.547***
(5.73)
FSD 0.042***
(8.12)
P_FSD 0.363***
(6.38)
FD 0.124***
(5.78)
_cons 0.130*** 0.080*** 0.121*** 0.067*** 0.106***
(7.75) (5.65) (8.58) (6.33) (5.82)
N 52,442 52,442 52,442 52,442
22
Table 10 (continued)
52,442
%, 5% and 10%, levels, respectively.
Table 11
PSM-DID analysis of payout policy across countries.
Panel A PSM-DID(DIV) PSM-DID(DIVEARN) PSM-DID(DIVTA)
N 56,686 55,683 55,058

Mean DID 0.135*** 0.105*** 0.006***
(p-value) (0.000) (0.000) (0.000)
[Standard error] 0.013 0.012 0.001
Panel B PSM-DID(DIV) PSM-DID(DIVEARN) PSM-DID(DIVTA)
N 56,214 55,160 56,213

Mean DID 0.129*** 0.113*** 0.005***
(p-value) (0.000) (0.000) (0.000)
[Standard error] 0.013 0.012 0.001
Variables are as defined in Table 2. ***, ** and * represent significance at the l%, 5% and 10%, levels, respectively.
finding of Floyd et al. (2015) on banking payout policy: firms still paid dividends, but the dividend payout ratio tended to gradually
diminish during the financial crisis. We also find that the board gender diversity indicators FDIR, P_FDIR, FSD, P_FSD, and FD are still
significantly positively correlated with corporate dividend policy: that is, our conclusions are unaffected by the financial crisis.
5.2. National culture and dividend payout
National culture is likely to influence corporate behavior (Chen et al., 2015; Eun et al., 2015; Lievenbrück and Schmid, 2014). For
instance, Boubakri and Saffar (2016) find that firms in countries with higher uncertainty avoidance tend to have lower financing
constraint. Through empirical analysis of listed companies in 41 countries, Chen et al. (2015) show that firms in countries with higher
uncertainty avoidance tend to retain more cash flow compared to those in countries with lower uncertainty avoidance.
The uncertainty avoidance index (UAI) measures the degree to which individuals feel comfortable with unknown or uncertain
situations in a particular society. Strong UAI societies exhibit more difficulty dealing with uncertainty: most individuals in these
societies are intolerant of change and do not want to break the structured environment. Frijns et al. (2013) suggest that firms' equity
premiums tend to be higher in strong UAI societies.
We use UAI to define national culture (Beugelsdijk and Frijns, 2010; Srite and Karahanna, 2006), with a greater value indicating a
higher degree of national uncertainty avoidance. Based on the median of UAI, we classify the full sample into two groups (Low UAI
and High UAI), and separately investigate the effect of board gender diversity on dividend payouts in each group. We measure board
gender diversity using FDIR, P_FDIR, FSD, P_FSD, and FD.
The results reported in Appendix N show that, after controlling for national culture, board gender diversity indicators are sig-
nificantly positively related to DIV, DIVEARN, and DIVTA in firms with low UAI. However, in firms with high UAI, board gender
diversity indicators have an insignificant positive influence on DIVTA. Thus, the effect of board gender diversity on dividend payouts
tends to be more significant in firms with low UAI. Our results are almost consistent with those of Breuer et al. (2014). Board gender
diversity indicators exert an evidently positive impact on corporate dividend policy, supporting the robustness of our conclusion after
considering national culture.
5.3. Corporate ownership and dividend payout
Recent literature notes that corporate ownership and even control is concentrated in the hands of a few owners in most countries
(La Porta and Lopez-de-Silanes, 1999). Lam et al. (2012) prove that state-controlled firms tend to have larger cash dividend payouts
and smaller share dividends, but both foreign ownership and cross-listing are significantly and negatively correlated with cash
dividends. Reluctance to cut dividends varies among different controlling owners: specifically, family-controlled firms are least
reluctant to cut dividends and have lower target payouts, while state-controlled firms are most reluctant to cut dividends and have
higher target payouts (Gugler, 2003). Gugler (2003) proposes that a double principal-agent problem might exist in state-controlled
firms but is rarely found in family-controlled firms, whose shareholders and managers are the same family members.
Family firms differ characteristically from other firm types. Some characteristics of family firms, particularly their long-term
perspective and political connectedness, are likely to promote corporate economic efficiency; by contrast, such characteristics as
family legacy and inheritance norms may negatively affect corporate performance. The number of family members participating in
23
the firm influences corporate policies (Mullins and Schoar, 2016; Neckebrouck et al., 2018; Sun and Yin, 2017; Tsoutsoura, 2015).
Isakov and Weisskopf (2015) confirm that family members involved in firms are more likely to meet their higher income needs
(relative to other staff) via dividend payments.
We define FAM as a dummy variable that equals 1 if a firm is family-controlled, and 0 otherwise, and STATE as a dummy variable
that equals 1 if a firm is state-controlled, and 0 otherwise. We then add these two control variables into the logit regression.
Consistent with the aforementioned literature, Appendix O reports that the family-controlled variable is significantly positively
correlated with dividend payout policy. Besides, gender diversity indicators including FDIR, P_FDIR, FSD, P_FSD and FD still exert a
significantly positive effect on dividend payout after controlling for family-controlled firms and state-controlled firms; that is, our
main conclusion remains valid.
5.4. Further tests
5.4.1. Sample selection

We rerun the regressions with all financial and utility companies, all sample countries with > 10 companies, and all sample
countries with > 20 companies, respectively. The results remain robust. For the sake of brevity, we do not show the corresponding
results.
5.4.2. Changing the estimation method

In the main analyses, DIV estimation is mainly based on logit regression. However, probit regression is also a suitable binary
choice model. Furthermore, Chay and Suh (2009) use Tobit regression to estimate the dividend payout ratio. Therefore, to eliminate
the selective bias of measuring methods, we use probit and Tobit models and separately estimate DIV and DIVEARN as two dependent
variables. Changing the estimation model does not change our main conclusions. For the sake of brevity, we do not show the
corresponding results.
5.4.3. Selection of standard error estimation method

To control for heteroskedasticity, all estimates in this paper are based on robust corporate clustering standard error. Therefore, to
eliminate the selective bias of standard error estimation methods, we separately use the methods of standard error, robust standard
error, robust industrial (double-digit industry code) clustering standard error, robust national clustering standard error, and the
bootstrap (100 times) standard error to run the regression. Our conclusions are robust and the corresponding results are not shown
for brevity.
6. Conclusion
The current era is witnessing increases in the number of female directors (Adams and Kirchmaier, 2016). This paper is primarily
motivated by three considerations. First, the issue of board gender diversity is of considerable importance worldwide. Second, many
countries are implementing corporate governance reforms, especially to improve board gender diversity. Third, most early studies of
the governance effect of board gender diversity focus only on a single country, so the literature lacks a global and holistic perspective.
To address the limitations in Chen et al.'s (2017) groundbreaking research on how board gender composition impacts corporate
dividend payouts, our study empirically analyzes this relationship using 63,631 firm-year observations from 8897 companies in 22
countries from 2000 to 2013. We find that board gender diversity can significantly alleviate agency problems and improve the
dividend payout ratio. We also find that the influence of board gender diversity on dividend payouts may be smaller in good
institutional environments. After conducting an array of robustness tests, the main conclusions remain valid.
The policy implications of this paper are mainly manifested in two aspects. The first concerns board gender diversity. European
countries have continually introduced relevant laws, such as gender quotas applicable to listed companies (Adams and Kirchmaier,
2016). Our empirical research evidences that female directors can reflect their values in corporate governance. This provides gui-
dance for national policymakers. The second concerns female career development. Gender-diverse boards tend to provide a variety of
perspectives on such challenges as solving the agency problem in corporate governance. Therefore, policymakers should establish a
professional skills training mechanism and create a fair competitive environment for female senior management to nurture female
career development.
Acknowledgments
We are grateful to Douglas Cumming(the editor) and the anonymous referees for their very insightful comments. We would also
like to thank the seminar participants at Nanyang Technological University, University of North Texas, and Jinan University, and
conference participants at China Economics Annual Conference 2016, IFABS Asia 2017 China Conference, China Economics Annual
Conference 2017, Chinese Economists Society 2017 North America Conference, Chinese Economists Society 2018 North America
Conference, and China Finance Annual Conference 2018. Xiao Chen would like to thank Professor Yew-Kwang Ng, Professor Donghui
Li, Dr. Bing Qi, Dr. Bihong Huang and Dr. Zhe An for their careful guidance and insights. Xiao Chen also grateful to Nanyang
Technological University and Zichun Huang from Singapore Management University for their support in this paper. Jie Deng would
like to thank Dr. Mantecon and Allyson Jones for bringing many valuable suggestions. All remaining errors are ours.
We also gratefully acknowledge research funding from national social sciences founding project （No.18BJY242), China.
24
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.jcorpfin.2019.04.002.
References
Adams, R.B., Ferreira, D., 2009. Women in the boardroom and their impact on governance and performance. J. Financ. Econ. 94, 291–309.
Adams, R.B., Funk, P., 2012. Beyond the glass ceiling: does gender matter? Manag. Sci. 58, 219–235.
Adams, R.B., Kirchmaier, T., 2016. Women on boards in finance and STEM industries. Am. Econ. Rev. 106, 277–281.
Adhikari, B.K., Agrawal, A., 2018. Peer influence on payout policies. J. Corp. Financ. 48, 615–637.
Agarwal, S., Qian, W.L., Reeb, D.M., Sing, T.F., 2016. Playing the boys game: golf buddies and board diversity. Am. Econ. Rev. 106, 272–276.
Ahern, K.R., Dittmar, A.K., 2012. The changing of the boards: the impact on firm valuation of mandated female board representation. Q. J. Econ. 127, 137–197.
Alesina, A., Giuliano, P., Nunn, N., 2013. On the origins of gender roles: women and the plough. Q. J. Econ. 128, 469–530.
Allen, F., Bernardo, A.E., Welch, I., 2000. A theory of dividends based on tax clienteles. J. Financ. 55, 2499–2536.
An, H., Zhang, T., 2013. Stock price synchronicity, crash risk, and institutional investors. J. Corp. Financ. 21, 1–15.
An, Z., Li, D.H., Yu, J., 2016. Earnings management, capital structure, and the role of institutional environments. J. Banking Financ. 68, 131–152.
Apesteguia, J., Azmat, G., Iriberri, N., 2012. The impact of gender composition on team performance and decision making: evidence from the field. Mang. Sci. 58,
78–93.
Armstrong, C.S., Guay, W.R., Weber, J.P., 2010. The role of information and financial reporting in corporate governance and debt contracting. J. Account. Econ. 50,
179–234.
Baker, M., Wurgler, J., 2004. A catering theory of dividends. J. Financ. 59, 1125–1165.
Bernardi, R.A., Arnold, D.F., 1997. An examination of moral development within public accounting by gender, staff level, and firm. Contemp. Account. Res. 14,
653–668.
Beugelsdijk, S., Frijns, B., 2010. A cultural explanation of the foreign bias in international asset allocation. J. Banking Financ. 34, 2121–2131.
Bliss, B.A., Cheng, Y.M., Denis, D.J., 2015. Corporate payout, cash retention, and the supply of credit: evidence from the 2008–2009 credit crisis. J. Financ. Econ. 115,
521–540.
Bohren, O., Staubo, S., 2014. Does mandatory gender balance work? Changing organizational form to avoid board upheaval. J. Corp. Financ. 28, 152–168.
Bond, P., Zhong, H.D., 2016. Buying high and selling low: stock repurchases and persistent asymmetric information. Rev. Financ. Stud. 29, 1409–1452.
Botero, J.C., Djankov, S., La Porta, R., Lopez-de-Silanes, F., Shleifer, A., 2004. The regulation of labor. Q. J. Econ. 119, 1339–1382.
Boubakri, N., Saffar, W., 2016. Culture and externally financed firm growth. J. Corp. 41, 502–520.
Brav, A., Graham, J.R., Harvey, C.R., Michaely, R., 2005. Payout policy in the 21st century. J. Financ. Econ. 77, 483–527.
Breuer, W., Rieger, M.O., Soypak, K.C., 2014. The behavioral foundations of corporate dividend policy a cross-country analysis. J. Banking Financ. 42, 247–265.
Brockman, P., Unlu, E., 2009. Dividend policy, creditor rights, and the agency costs of debt. J. Financ. Econ. 92, 276–299.
Brown, J.R., Liang, N., Weisbenner, S., 2007. Executive financial incentives and payout policy: firm responses to the 2003 dividend tax cut. J. Financ. 62, 1935–1965.
Byun, H.Y., Choi, S., Hwang, L.S., Kim, R.G., 2013. Business group affiliation, ownership structure, and the cost of debt. J. Corp. Financ. 23, 311–331.
Chae, J., Kim, S., Lee, E.J., 2009. How corporate governance affects payout policy under agency problems and external financing constraints. J. Banking Financ. 33,
2093–2101.
Chay, J.B., Suh, J., 2009. Payout policy and cash-flow uncertainty. J. Financ. Econ. 93, 88–107.
Chen, T., 2015. Institutions, board structure, and corporate performance: evidence from Chinese firms. J. Corp. Financ. 32, 217–237.
Chen, Y.Y., Dou, P.Y., Rhee, S.G., Truong, C., Veeraraghavan, M., 2015. National culture and corporate cash holdings around the world. J. Banking Financ. 50, 1–18.
Chen, J., Leung, W.S., Goergen, M., 2017. The impact of board gender composition on dividend payouts. J. Corp. Financ. 43, 86–105.
Chetty, R., Saez, E., 2005. Dividend taxes and corporate behavior: evidence from the 2003 dividend tax cut. Q. J. Econ. 120, 791–833.
Crane, A.D., Michenaud, S., Weston, J.P., 2016. The effect of institutional ownership on payout policy: evidence from index thresholds. Rev. Financ. Stud. 29,
1377–1408.
Croson, R., Gneezy, U., 2009. Gender differences in preferences. J. Econ. Lit. 47, 448–474.
Cumming, D., Johan, S., Li, D., 2011. Exchange trading rules and stock market liquidity. J. Financ. Econ. 99, 651–671.
Cumming, D., Leung, T.Y., Rui, O., 2015. Gender diversity and securities fraud. Acad. Mange. J. 58, 1572–1593.
Dang, T.L., Moshirian, F., Zhang, B.H., 2015. Commonality in news around the world. J. Financ. Econ. 116, 82–110.
DeAngelo, H., DeAngelo, L., 2006. Dividend policy and the earned/contributed capital mix: a test of the life-cycle theory. J. Financ. Econ. 81, 227–254.
Denis, D.J., Osobov, I., 2008. Why do firms pay dividends? International evidence on the determinants of dividend policy. J. Financ. Econ. 89, 62–82.
Dittmar, A., Field, L.C., 2015. Can managers time the market? Evidence using repurchase price data. J. Financ. Econ. 115, 261–282.
Easterbrook, F.H., 1984. Two agency-cost explanations of dividends. Am. Econ. Rev. 74, 650–659.
Eckel, C.C., Fullbrunn, S.C., 2015. Thar she blows? Gender, competition, and bubbles in experimental asset markets. Am. Econ. Rev. 105, 906–920.
Edmans, A., Manso, G., 2011. Governance through trading and intervention: a theory of multiple blockholders. Rev. Financ. Stud. 24, 2395–2428.
Eom, C., 2018. Institutional bidding behaviors during IPO bookbuilding: evidence from Korea. J. Corp. Financ. 48, 413–427.
Espen Eckbo, B., Verma, S., 1994. Managerial shareownership, voting power, and cash dividend policy. J. Corp. Financ. 1, 33–62.
Eun, C.S., Wang, L.L., Xiao, S.C., 2015. Culture and R-2. J. Financ. Econ. 115, 283–303.
Evgeniou, T., Vermaelen, T., 2017. Share buybacks and gender diversity. J. Corp. Financ. 45, 669–686.
Faccio, M., Marchica, M.T., Mura, R., 2016. CEO gender, corporate risk-taking, and the efficiency of capital allocation. J. Corp. Financ. 39, 193–209.
Fahlenbrach, R., Stulz, R.M., 2011. Bank CEO incentives and the credit crisis. J. Financ. Econ. 99, 11–26.
Fama, E.F., 1980. Agency problems and the theory of the firm. J. Polit. Econ. 88, 288–307.
Fama, E.F., Babiak, H., 1968. Dividend policy: An empirical analysis. J. Am. Stat. Assoc. 63, 1132–1161.
Fama, E.F., French, K.R., 2001. Disappearing dividends: changing firm characteristics or lower propensity to pay? J. Financ. Econ. 60, 3–43.
Fama, E.F., Jensen, M.C., 1983. Agency problems and residual claims. J. Law Econ. 26, 327–349.
Fernandez, R., 2013. Cultural change as learning: the evolution of female labor force participation over a century. Am. Econ. Rev. 103, 472–500.
Firth, M., Gao, J., Shen, J.H., Zhang, Y.Y., 2016. Institutional stock ownership and firms' cash dividend policies: evidence from China. J. Banking Financ. 65, 91–107.
Floyd, E., Li, N., Skinner, D.J., 2015. Payout policy through the financial crisis: the growth of repurchases and the resilience of dividends. J. Financ. Econ. 118,
299–316.
Frijns, B., Gilbert, A., Lehnert, T., Tourani-Rad, A., 2013. Uncertainty avoidance, risk tolerance and corporate takeover decisions. J. Banking Financ. 37, 2457–2471.
Gonzalez, F., 2016. Creditor rights, bank competition, and corporate investment during the global financial crisis. J. Corp. Financ. 37, 249–270.
Grinstein, Y., Michaely, R., 2005. Institutional holdings and payout policy. J. Financ. 60, 1389–1426.
Grullon, G., Michaely, R., 2002. Dividends, share repurchases, and the substitution hypothesis. J. Financ. 57, 1649–1684.
Gugler, K., 2003. Corporate governance, dividend payout policy, and the interrelation between dividends, R&D, and capital investment. J. Banking Financ. 27,
1297–1321.
Gul, F.A., Srinidhi, B., Ng, A.C., 2011. Does board gender diversity improve the informativeness of stock prices? J. Account. Econ. 51, 314–338.
Hasan, M.M., Cheung, A., 2018. Organization capital and firm life cycle. J. Corp. Financ. 48, 556–578.
Healy, P.M., Palepu, K.G., 2001. Information asymmetry, corporate disclosure, and the capital markets: a review of the empirical disclosure literature. J. Account.
25
Econ. 31, 405–440.

Hillert, A., Maug, E., Obernberger, S., 2016. Stock repurchases and liquidity. J. Financ. Econ. 119, 186–209.
Hillman, A.J., Shropshire, C., Cannella, A.A., 2007. Organizational predictors of women on corporate boards. Acad. Mange. J. 50, 941–952.
Ho, P.H., Huang, C.W., Lin, C.Y., Yen, J.F., 2016. CEO overconfidence and financial crisis: evidence from bank lending and leverage. J. Financ. Econ. 120, 194–209.
Hoberg, G., Prabhala, N.R., 2009. Disappearing dividends, catering, and risk. Rev. Financ. Stud. 22, 79–116.
Hu, A.D., Kumar, P., 2004. Managerial entrenchment and payout policy. J. Financ. Quant. Anal. 39, 759–790.
Huang, W., Mazouz, K., 2018. Excess cash, trading continuity, and liquidity risk. J. Corp. Financ. 48, 275–291.
Isakov, D., Weisskopf, J.P., 2015. Pay-out policies in founding family firms. J. Corp. Financ. 33, 330–344.
Jagannathan, M., Stephens, C.P., Weisbach, M.S., 2000. Financial flexibility and the choice between dividends and stock repurchases. J. Financ. Econ. 57, 355–384.
Jensen, M.C., 1986. Agency costs of free cash flow, corporate finance, and takeovers. Am. Econ. Rev. 76, 323–329.
Jensen, M.C., Meckling, W.H., 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. J. Financ. Econ. 3, 305–360.
Jensen, G.R., Solberg, D.P., Zorn, T.S., 1992. Simultaneous determination of insider ownership, debt, and dividend policies. J. Financ. Quant. Anal. 27, 247–263.
Jiang, X.Y., Yuan, Q.B., 2018. Institutional investors' corporate site visits and corporate innovation. J. Corp. Financ. 48, 148–168.
Jin, Z., Shiomura, K., Jiang, L.Z., 2015. Assessing implicit mate preferences among Chinese and Japanese women by providing love, sex, or money cues. Psychol. Rep.
116, 195–206.
Jiraporn, P., Ning, Y., 2006. Dividend Policy, Shareholder Rights, and Corporate Governance. Working Paper. Pennsylvania State University.
Kale, J.R., Kini, O., Payne, J.D., 2012. The dividend initiation decision of newly public firms: some evidence on signaling with dividends. J. Financ. Quant. Anal. 47,
365–396.
Kim, D., Starks, L.T., 2016. Gender diversity on corporate boards: do women contribute unique skills? Am. Econ. Rev. 106, 267–271.
Kim, K., Patro, S., Pereira, R., 2017. Option incentives, leverage, and risk-taking. J. Corp. Financ. 43, 1–18.
La Porta, R., Lopez-de-Silanes, F., 1999. The benefits of privatization: evidence from Mexico. Q. J. Econ. 114, 1193–1242.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R.W., 1998. Law and finance. J. Polit. Econ. 106, 1113–1155.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R.W., 2000. Agency problems and dividend policies around the world. J. Financ. 55, 1–33.
Lam, K.C., Sami, H., Zhou, H., 2012. The role of cross-listing, foreign ownership and state ownership in dividend policy in an emerging market. China J. Account. Res.
5, 199–216.
Leary, M.T., Roberts, M.R., 2014. Do peer firms affect corporate financial policy? J. Financ. Econ. 69, 139–178.
Levit, D., Malenko, N., 2016. The labor market for directors and externalities in corporate governance. J. Financ. 71, 775–808.
Li, W., Lie, E., 2006. Dividend changes and catering incentives. J. Financ. Econ. 80, 293–308.
Li, F., Srinivasan, S., 2011. Corporate governance when founders are directors. J. Financ. Econ. 102, 454–469.
Li, D.H., Chen, Z.A., An, Z., Murong, M., 2017. Do financial analysts play a role in shaping the rival response of target firms? International evidence. J. Corp. Financ.
23, 84–103.
Lievenbrück, M., Schmid, T., 2014. Why do firms (not) hedge? – Novel evidence on cultural influence. J. Corp. Financ. 25, 92–106.
Lin, C., Ma, Y., Malatesta, P., Xuan, Y.H., 2013. Corporate ownership structure and the choice between bank debt and public debt. J. Financ. Econ. 109, 517–534.
Lintner, J., 1956. Distribution of incomes of corporations among dividends, retained earnings, and taxes. Am. Econ. Rev. 46, 97–113.
Liu, Y., Wei, Z.B., Xie, F.X., 2014. Do women directors improve firm performance in China? J. Corp. Financ. 28, 169–184.
Liu, C., 2018. Are women greener? Corporate gender diversity and environmental violations. J. Corp. Financ. 52, 118–142.
Miller, M.H., Modigliani, F., 1961. Dividend policy, growth, and the valuation of shares. J. Bus. 34, 411–433.
Mullins, W., Schoar, A., 2016. How do CEOs see their roles? Management philosophies and styles in family and non-family firms. J. Financ. Econ. 119, 24–43.
Neckebrouck, J., Schulze, W., Zellweger, T., 2018. Are family firms good employers? Acad. Mange. J. 61, 553–585.
Niederle, M., Vesterlund, L., 2007. Do women shy away from competition? Do men compete too much? Q. J. Econ. 122, 1067–1101.
Officer, M.S., 2011. Overinvestment, corporate governance, and dividend initiations. J. Corp. Financ. 17, 710–724.
Price, C.R., 2012. Gender, competition, and managerial decisions. Manag. Sci. 58, 114–122.
Shleifer, A., Vishny, R.W., 1997. A survey of corporate governance. J. Financ. 52, 737–783.
Short, H., Zhang, H., Keasey, K., 2002. The link between dividend policy and institutional ownership. J. Corp. Financ. 8, 105–122.
Smith, M.P., 1996. Shareholder activism by institutional investors: evidence from CalPERS. J. Financ. 51, 227–252.
Srite, M., Karahanna, E., 2006. The role of espoused national cultural values in technology acceptance. Mis. Quart. 30, 679–704.
Sun, H.W., Yin, S.X., 2017. Information leakage in family firms: evidence from short selling around insider sales. J. Corp. Financ. 47, 72–87.
Tsoutsoura, M., 2015. The effect of succession taxes on family firm investment: evidence from a natural experiment. J. Financ. 70, 649–688.
Von Eije, H., Megginson, W.L., 2008. Dividends and share repurchases in the European Union. J. Financ. Econ. 89, 347–374.
Ward, C., Yin, C., Zeng, Y.Q., 2018. Institutional investor monitoring motivation and the marginal value of cash. J. Corp. Financ. 48, 49–75.
Wintoki, M.B., Linck, J.S., Netter, J.M., 2012. Endogeneity and the dynamics of internal corporate governance. J. Financ. Econ. 105, 581–606.
Young, S., Yang, J., 2011. Stock repurchases and executive compensation contract design: the role of earnings per share performance conditions. Account. Rev. 86,
703–733.
Zhang, M., M, L.J., Zhang, B., Yi, Z.H., 2016. Pyramidal structure, political intervention and firms' tax burden: evidence from China's local SOEs. J. Corp. Financ. 36,
15–25.
26

02 Does Board Gender Diversity Increase Dividend Payouts Analysis of Global Evidence

Uploaded by

Copyright:

Available Formats

You might also like

02 Does Board Gender Diversity Increase Dividend Payouts Analysis of Global Evidence

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

02 Does Board Gender Diversity Increase Dividend Payouts Analysis of Global Evidence

Uploaded by

Copyright:

Available Formats

Journal of Corporate Finance 58 (2019) 1–26

Contents lists available at ScienceDirect

Journal of Corporate Finance

Does board gender diversity increase dividend payouts? Analysis of

corporate dividend payments across diﬀerent countries.

Australia 621 7.00 3180 5.01

Panel A Dividend payment

Panel B Gender diversity index

FDIR Number of female directors on board BoardEx

Panel C Corporate governance index

Own Percentage of management holdings WorldScope

Panel D Control variables

LOGTA Natural logarithm of total assets, in dollar terms WorldScope

Panel E Variables of female directors' forecasting model

RESID_FDIR Residual number of female directors on board

Panel E Variables of female directors' forecasting model

RESID_P_FDIR Residual percentage of female directors

DIV 0.456 0.498 0.000 0.000 1.000

Variables are as deﬁned in Table 2.

Fig. 1. Trend in Gender Diversity.

Women on Board: FDIR%

Women on Board: FDIR

Fig. 3. Average Number of Each Country's Female Directors Worldwide.

• FDIR: the number of female directors;

FDIR 33,680 0.530 28,263 0.960 −0.43***

• FCHAIR: whether the board chairperson is female;

4.1. Board gender diversity and dividend payout

Panel A DIV DIV DIV DIV DIV DIV

Panel B DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN

Panel B DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN

Cash −0.011 −0.011 −0.011 −0.010 −0.011 −0.011

Panel C DIVTA DIVTA DIVTA DIVTA DIVTA DIVTA

main research variables; we list them to enhance the results' integrity.

Panel A DIV DIV DIV DIV DIV DIV

Panel B DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN

Panel C DIVTA DIVTA DIVTA DIVTA DIVTA DIVTA

Panel C DIVTA DIVTA DIVTA DIVTA DIVTA DIVTA

N 56,161 56,161 56,161 56,161 56,161 56,161

4.2. Institutional environment and dividend payout

4.3. Further tests of institutional environment

Panel A DIV DIV DIV DIV DIV

Panel B DIVEARN DIVEARN DIVEARN DIVEARN DIVEARN

(continued on next page)

Panel C DIVTA DIVTA DIVTA DIVTA DIVTA

4.4. Institutional ownership and dividend payout

4.5. Insider ownership and dividend payout

4.6. Consideration of share repurchases

Panel A STR STR STR STR STR

Panel B TotalEarn TotalEarn TotalEarn TotalEarn TotalEarn

Panel C TotalTA TotalTA TotalTA TotalTA TotalTA

4.7. Dividend initiations and increases

4.8. Endogeneity problems

4.8.1. Omitted variables

Panel A PRIORPAYER PRIORPAYER PRIORPAYER PRIORPAYER PRIORPAYER

Panel B INDIVEA INDIVEA INDIVEA INDIVEA INDIVEA

Panel C INDIVTA INDIVTA INDIVTA INDIVTA INDIVTA

Panel C INDIVTA INDIVTA INDIVTA INDIVTA INDIVTA

IndMeanFDIR 0.873* 0.091* 0.804* 0.096* 0.318***