CORPORATE GOVERNANCE AND DIVIDEND PAYOUT POLICY: BEYOND COUNTRY‐LEVEL GOVERNANCE

• DOI: http://doi.org/10.1111/jfir.12159
• Author: Bin Chang,PengCheng Zhu,Samir Saadi,Shantanu Dutta
• Date: 01 December 2018
• Published date: 01 December 2018

CORPORATE GOVERNANCE AND DIVIDEND PAYOUT POLICY:

BEYOND COUNTRY-LEVEL GOVERNANCE

Bin Chang

University of Ontario Institute of Technology

Shantanu Dutta and Samir Saadi

University of Ottawa

PengCheng Zhu

University of San Diego

Abstract

We address the mixed empirical findings on how corporate governance affects dividend

payout policy by analyzing a large sample of firms from 30 countries. Our results

indicate that firms with better firm-level governance pay more dividends, even after

controlling for country-level governance. However, this relation is pronounced only in

countries with low shareholder rights. In addition, we find that when the shareholder

rights index is high, firm-level governance is unrelated to dividend payout in the full

sample period. Finally, we show that in high-shareholder-rights countries, firm-level

governance changes its role from before to after the 2008–2009 financial crisis.

JEL Classification: G34, G35

I. Introduction

Do international firms with better corporate governance pay out more dividends? Does

country-level governance affect the role of firm-level governance in determining

dividend payouts? We seek to answer these questions using a large sample of companies

from 30 countries with detailed firm-level corporate governance data. Unlike earlier

international dividend studies that primarily employ country-level governance quality

measures, we consider the impact of both firm- and country-level corporate governance

practices. Addressing these research questions sheds light on the mixed empirical

findings on how corporate governance affects the dividend payout policies of

international firms.

We thank the reviewer and the Associate Editor for their insightful comments on our paper. We also

acknowledge the helpful comments and suggestions from John Bai, Imed Chkir, Lamia Chourou, Alfred Davis, and

Ligang Zhong. Our paper has also benefited from feedback from participants at the European Financial

Management Association annual meetings, and from seminar participants at the University of Ottawa and

University of San Diego. Bin Chang, Shantanu Dutta, and Samir Saadi gratefully acknowledge financial support

from the Social Sciences and Humanities Research Council of Canada as well as from the Telfer School of

Management Research Grant. All errors are our own responsibility. Part of this research was conducted while

Samir Saadi was visiting Stern School of Business, New York University.

The Journal of Financial Research Vol. XLI, No. 4 Pages 445–484 Winter 2018

DOI: 10.1111/jfir.12159

445

© 2018 The Southern Finance Association and the Southwestern Finance Association

La Porta et al. (2000) propose two hypotheses regarding the agency theory of

dividends. The “outcome hypothesis”predicts that firms with better governance pay out

more dividends to restrict management’s private benefit of having excess cash. Thus, the

outcome hypothesis suggests a positive relation between governance and dividends. The

“substitution hypothesis”predicts that firms with poor governance and weaker

shareholder rights pay out higher dividends as the substitution of poor governance, to

maintain a good relationship with shareholders and access to external capital in the

future. Thus, the substitution hypothesis suggests a negative relation between

governance and dividends. Although finance theories do not differentiate between

country- and firm-level governance practices, empirical studies are based on either (1) a

cross-country sample using country-level governance variables or (2) a single-country

sample using firm-level governance variables. Cross-country studies generally support a

positive relation between country-level governance and dividends (La Porta et al. 2000),

whereas single-country studies are inconclusive about whether higher firm-level

governance leads to higher dividend payout (Fenn and Liang 2001).

Prior cross-country studies using only country-leve l governance variables

implicitly assume that all firms’corporate policies are equally af fected by country-

level governance. This implies that if there were onl y country-level determinates of

dividend payout policies, all firms in the same country would hav e the same dividend

policy. But this is not the case, as suggested by the single-country literature. The

cross-country literature has not used firm-le vel governance because of the lack of data

availability in a comparable format across countries. Institutional Shareholder Services

(ISS) started constructing comparable firm-level gov ernance data in a cross-country

setup in 2003. We use the ISS governance database and follow the app roach of

Aggarwal et al. (2011) to construct the firm-level ISS gover nance index (ISS41 index).

Our sample shows considerable variation in firm-level governance quality within each

country, suggesting the need to include firm-level gover nance in cross-country studies.

Yet, studies incorporating both country- and firm-leve l governance simultaneously are

scarce.

The seminal study by La Porta et al. (2000) on cross-country dividend policy

focuses on country-level governance in general and does not consider firm-level

governance. Other studies such as Doidge, Karolyi, and Stulz (2007), Aggarwal et al.

(2009), and Aggarwal et al. (2011) find a complementary relation between country- and

firm-level governance, but these studies do not focus on dividend payout policy. As it

appears, the literature does not predict the joint effect of country- and firm-level

governance on dividend payout. Therefore interesting empirical questions arise: Does

firm governance have a role in determining dividend payout after controlling country

governance? If this role exists, does it depend on country-level governance?

Furthermore, earlier studies show that dividend policy is related to firm cash

holdings (e.g., Brockman and Unlu 2009). A firm’s cash position varies with business

and economic cycles. During the financial crisis, firms face more cash crunch. As a result,

the relation between firm-specific corporate governance and dividend payment may

differ between the pre- and during-crisis periods. Better governed firms are likely to pay

more attention to shareholders’interests and a firm’sfinancial health. However, it is not

clear how these firms will behave with respect to their dividend policy during a crisis

446 The Journal of Financial Research

period. This leads to another interesting question: Does the role of firm governance

change in financial crisis?

We take the above questions to the data, using a sample of 4,022 firms across

30 countries from 2003 to 2009. Our results show that after controlling for country-level

variables, firms with better corporate governance pay out higher dividends. This finding

is in line with the outcome hypothesis. Next, we find that this relation is significant only

when the country-level anti-self-dealing index (ASD index) is below the median.

1

In

these countries, the dividend-to-asset ratio is predicted to increase from 0.5% to 2.25% as

the ISS41 index increases from 7 to 29 in our sample. Only when the country-level ASD

index is high do we find that firm-level governance is unrelated to dividend payout in the

full sample. Taken together, our results provide the first systematic evidence that, absent

a strong ASD index, firm governance plays an important role in dividend policy. Our

results are robust to the inclusion of the originality of law, culture, creditors’rights,

alternative measures of dividend payout, individual attributes of firm governance, and

approaches to address endogeneity concerns. Further analysis shows that in countries

with a high ASD index, firm-level governance is positively related to dividend payout

before the financial crisis (2003–2007) but negatively related during the financial crisis

(2008–2009). In countries with a low ASD index, they are always positively related.

We also extend the research to stock repurchases. Cash dividends are “sticky,”

whereas repurchases are less frequent and random. Therefore, we may not see any

systematic relation between corporate governance and repurchase. Yet, as a robustness

test, we examine the relation between firm-specific corporate governance and repurchase

activity. We modify the main model by using the stock repurchase payout ratio as the

dependent variable. We find that ASD and ISS41 are statistically insignificant in

the regression. The result indicates that neither country- nor firm-level governance

affects stock repurchases.

Our article contributes to the literature in several ways. First, we show that firm-

level governance is positively related to dividend payouts in an international setting. The

existing cross-country literature has only used country-level governance and ignores

within-country variation in corporate governance. Our results show that country-level

variables alone do not fully explain the role of governance in dividend payouts. Instead,

in a cross-country empirical setup, a firm-specific governance mechanism plays a

significant role. Second, we show that the role of firm-specific governance in determining

dividend policy is affected by country-level governance. If country-level governance is

weaker (stronger), the quality of firm-level governance becomes more (less) relevant.

Our results indicate that to understand the effect of governance on dividend policy, we

need to consider both country- and firm-level governance structures. Our article

complements the work of Kalcheva and Lins (2007), who study the relation between

investors’valuation of dividend payout and management ownership in an international

setting using only a single year’s (1996) data. Different from them, we focus on the level

of dividend payout, which is a manager’s decision, instead of the value of dividend,

which is an investor’s decision.

1

Following Djankov et al. (2008), we use the ASD index to denote country-level governance quality.

Corporate Governance 447