Full length article
National culture and dividend policy: International evidence from banking

https://doi.org/10.1016/j.jbef.2014.07.002Get rights and content

Abstract

We examine the relations between three dimensions of national culture and dividend policies of banks using a sample of banks from 51 countries over the period 1998–2007. In our main analysis, we employ three dimensions of Hofstede et al. (2010) and find that banks in high uncertainty avoidance, high long-term orientation and low masculinity countries pay lower amount of dividends and, are less likely to pay dividends. To confirm our main results, we also employ comparable three dimensions of national culture of House et al. (2004) and find that banks in high uncertainty avoidance, high future orientation and low assertiveness countries pay lower amount of dividends and, are less likely to pay dividends, findings confirming our above results. In sum, we find significant influence of the three dimensions of national culture on bank dividend policies.

Introduction

We employ three dimensions of national culture from two culture measuring studies of Hofstede et al. (2010) and House et al. (2004), and an international sample of banks to study the effects of national culture on bank dividend policies. Cross-country differences in banks’ propensity to pay dividends and level of dividend payments are likely to be affected by differences in corporate governance and legal institutions, as well by softer dimensions such as national culture that may influence excessive earnings retention or payment as dividends.

The extant literature has studied the relation between national culture and dividend policies, but has excluded banking firms (Khambata and Liu, 2005, Shao et al., 2009, Fidrmuc and Jacob, 2010, Bae et al., 2012). A large amount of literature has documented that financial systems promote economic growth by mobilizing savings, reducing information asymmetries, providing risk-sharing and facilitating exchange (Levine, 1997, Levine, 2005, Hassan et al., 2011). Due to such an important role performed by this highly leveraged industry for national economies and quite different nature of banking firms as compared to industrial firms, it is important to examine the relation of cultural forces with bank dividend policies. Further, numerous studies have examined the bank dividend policies separately from the dividend policies of industrial firms (Casey and Dickens, 2000, Dickens et al., 2002, Theis and Dutta, 2009).

In addition, more cash holdings can encourage bank managers to take more risk by extending poor credit quality loans. In a global survey on factors that created the conditions for the banking crisis conducted in May 2008 by Economist Intelligence Unit and PricewaterhouseCoopers, 58% of survey participants put the blame on ‘ineffective regulatory oversight’, 31% on ‘monetary policy’, and an impressive 73% on ‘culture and excessive risk-taking’ (Kanagaretnam et al., 2011). Cross-country differences became more apparent in the recent financial crisis, which had a significantly larger adverse effect on banks in certain countries (e.g., the UK and the USA) than in others (e.g., Australia and Canada). Given these findings, an examination of the influence of national culture on bank dividend policies is clearly warranted.

Culture is generally defined as a set of norms, beliefs, expected behaviors and shared values that serves as guiding principles in people’s lives (Schwartz, 1994, Hofstede, 2001). By guiding human behavior, cultural values reflect what a society/group considers to be legitimate or illegitimate, good or bad, acceptable or unacceptable, or ethical or unethical (Hofstede, 2001). On the other hand, extant literature argues the important role of dividend policies in reducing agency costs and minimizing information asymmetries (Rozeff, 1982, Easterbrook, 1984, La Porta et al., 2000, Denis and Osobov, 2008, Eije and Megginson, 2008, Brockman and Unlu, 2009). And, in a recent study, John et al. (2010) argue that agency costs are more severe in banks due to their higher leverage. Therefore, we base our theoretical framework on Fidrmuc and Jacob (2010) who offer an agency theory based explanation of dividends for national cultural dimensions. This approach considers the preferences and behaviors of economic agents inherent in the cultural values for determining dividend policies of firms. So, in this paper we argue that bank dividend policies are a special case of social norms reflecting the legitimacy or acceptability of certain dividend payout strategies in a society. Across countries, social norms governing dividend payout policies may vary because differing cultural value emphases breed different behaviors, aspirations, beliefs, and preferences, and therefore alter the severity and nature of agency conflicts.

An alternative view is that because banks operate in highly regulated environments, and are strictly monitored by central banks for capital levels, therefore, country-level national cultural factors may not be as important in influencing bank dividend policies. In this regard, banks may face different levels of regulatory pressure both at bank- and country-levels. On bank-level, low capitalized banks can be under severe regulatory pressure for increasing capital by not paying dividends than the banks which either meet regulatory capital requirements or operate with the capital levels significantly above regulatory minimum requirements. Similarly, on country-level, banks in some countries may face more stringent capital requirements than the banks in other countries which have less stringent capital requirements. If we find significant cultural effects on bank dividend policies after controlling for bank- and country-level regulatory pressure, then we can argue that cultural forces are even more important for other less-regulated or un-regulated industries.

Bae et al. (2012) hypothesize and find that national culture influences dividend policies of nonfinancial firms. More specifically, they document that three dimensions of national culture–uncertainty avoidance, masculinity and long-term orientation–have negative relations with dividend payout amounts. Given their findings, we examine the relations between same three dimensions2 of national culture, and bank dividend policies. We focus on the pre-financial crisis period (e.g., the period 1998–2007) in our analysis.

We derive three hypotheses based on three dimensions of national culture. First, we posit that bank insiders and outside shareholders in high uncertainty avoidance societies put heavy emphasis on the certainty that dividend payment expectations are met each period. Also, uncertainty-avert minority shareholders and bank insiders may prefer higher retained profits because they are cash resources to hedge against unforeseen financial distress. Therefore, we expect lower dividend payouts in high uncertainty avoidance societies. Second, we conjecture that in masculine cultures, agency conflicts are inherently more severe because their members are considered competitive and impatient, and are more prone to pursue opportunistic behavior rather than adhere to others’ decisions and preferences. Therefore, outside shareholders are expected to demand higher level of dividends to discipline the opportunistic behavior of bank insiders. Finally, we hypothesize that in more long-term orientation countries, which tend to show greater perseverance, thrift and patience, the severity of agency conflicts is inherently lower. Consequently, investors have a lower preference for dividends as a disciplining mechanism and find lower dividend payouts culturally more acceptable.

We employ common dividends paid to total assets ratio and a dummy variable, equal to one for dividend paying banks and zero otherwise, to test the relations between three dimensions of national culture, and dividend payout amounts and the propensity to pay dividends, respectively.

We use an international sample of banks from Bankscope database representing 51 countries over the period 1998–2007 to test our predictions about cultural dimensions and bank dividend policies. We begin by examining the impact of cultural dimensions on dividend payout amounts while controlling for bank- and country-level regulatory pressure, bank size, profitability, asset growth, minority shareholder rights, creditor rights and level of financial market development. At bank-level, we use equity to total assets ratio of each bank to control for regulatory pressure because well capitalized (weakly capitalized) banks may face lower regulatory pressure (higher regulatory pressure) while deciding about dividend payments. At country-level, bank regulators in some countries may impose more stringent capital requirements for banks and then force banks to meet capital requirements before paying any dividends. Therefore, we use regulatory capital index of Barth et al. (2013) to account for cross-country heterogeneity in capital requirements and regulatory pressure.

We estimate tobit panel regressions to examine the cultural effects on dividend payout amounts. We find negative and significant relations between uncertainty avoidance and long-term orientation cultural dimensions and dividend payout amounts, and a positive and significant relation between masculinity dimension and dividend payout amounts.

Next, we estimate logit panel regressions to examine the cultural effects on the probability to pay dividends. Similar to dividend amounts results, we find that uncertainty avoidance and long-term orientation have negative and significant relations with the probability of paying dividends, whereas, masculinity has significantly positive relation with the probability of paying dividends.

To further confirm our results, we use three cultural dimensions–uncertainty avoidance, assertiveness and future orientation–from a more recent cultural framework of House et al. (2004). These three cultural dimensions measure almost same cultural values and, therefore, are comparable with three dimensions of Hofstede. We estimate all our models with these cultural dimensions and find similar results; uncertainty avoidance and future orientation show significantly negative and assertiveness shows significantly positive associations with both, dividend payments and the probability to pay dividends, variables.

Our paper is innovative in several aspects: First, we provide new evidence on relations between national cultural dimensions and dividend policies from banking after controlling for regulatory pressure, an area ignored by previous research. Second, in contrast to previous research in this area, we follow Brockman and Unlu (2009) and estimate tobit panel regressions to examine the cultural effects on dividend payout amounts. Tobit model is considered more appropriate when dependent variable has many same values in one tail. Third, we, first time, use cultural dimensions of House et al. (2004) in dividend policies research.

Our study contributes to the literature in several ways: First, to best of our knowledge, our study is the first to study the bank dividend policies in an international context. We confirm bank-level factors such as size, profitability, growth and capital; and country-level factors such as corporate governance, legal institutions, financial market development and national culture as significant determinants of bank dividend policies in an international setting. Second, we extend law and finance literature by confirming the importance of minority shareholder rights and creditor rights for bank dividend policies. Third, we contribute to national culture and finance literature (see Aggarwal and Goodell (2014, for a review)), in general, and to national culture and banking literature (Kanagaretnam et al., 2011, Zheng et al., 2013), in particular, by identifying the national culture’s significance for dividend policies even in a regulated industry such as banking. Fourth, we contribute to literature which argues that financial structures of higher uncertainty-avoidance cultures are more bank-based (Kwok and Tadesse, 2006, Aggarwal and Goodell, 2009). We examine micro-level dividend policy channel that banks in higher uncertainty-avoidance countries pay lower dividends, and one possible reason for these lower dividend payments may be to keep more funds for extending more bank loans. Finally, as Fidrmuc and Jacob (2010) relate three dimensions of national culture (uncertainty avoidance, individualism and power distance) to dividend policies based on agency problems explanation, we extend this explanation to two more cultural dimensions of masculinity and long-term orientation.

The rest of the paper is organized as follows. In Section  2, we review literature and provide testable hypotheses. Section  3 introduces data. Section  4 presents empirical results. And, the final section concludes the study.

Section snippets

Literature review

Since the publication of the Miller and Modigliani (1961) dividend irrelevance propositions, an extensive research has been done to resolve the puzzle that “why do firms pay dividends”. Despite this expansive literature,3

Data

We download annual financial accounting information for bank holding companies, and commercial, cooperative and savings banks from Bankscope database. Cultural variables are obtained from two culture measuring studies of Hofstede et al. (2010) and House et al. (2004). Regulatory capital index is taken from Barth et al. (2013). We obtain shareholder rights and creditor rights information from Djankov et al. (2008) and Djankov et al. (2007), respectively—two studies that update the La Porta

Summary statistics

Table 1 reports summary statistics for the main variables. The mean value for our dividends paid to total assets ratio, Dividends, is 0.11%. The mean of payer dummy variable is 38%, suggesting that the sample has a majority of dividend non-paying banks. Other bank-level variables also show considerable variation. In Table 2, we report the distribution of banks across countries. Germany has highest bank observations (14,948) and New Zealand has lowest (22). We note that like most of the

Conclusion

We address the primary research question that how various dimensions of national culture influence the dividend policies of banks. Given the important role of national culture for the cross-country differences in managerial behavior and the important role of banking for national as well as global economies, absence of prior evidence on the implications of national culture for bank dividend policies is surprising.

Our empirical analyses are based on an international sample of 41,343 bank-year

References (66)

  • K.M. Casey et al.

    The effects of tax and regulatory changes on commercial bank dividend policy

    Q. Rev. Econ. Finance

    (2000)
  • H. DeAngelo et al.

    The irrelevance of the MM dividend irrelevance theorem

    J. Financ. Econ.

    (2006)
  • H. DeAngelo et al.

    Dividend policy and the earned/contributed capital mix: a test of the life-cycle theory

    J. Financ. Econ.

    (2006)
  • D.J. Denis et al.

    Why do firms pay dividends? International evidence on the determinants of dividend policy

    J. Financ. Econ.

    (2008)
  • S. Djankov et al.

    The law and economics of self-dealing

    J. Financ. Econ.

    (2008)
  • S. Djankov et al.

    Private credit in 129 countries

    J. Financ. Econ.

    (2007)
  • E.F. Fama et al.

    Disappearing dividends: changing firm characteristics or lower propensity to pay?

    J. Financ. Econ.

    (2001)
  • J.P. Fidrmuc et al.

    Culture, agency costs, and dividends

    J. Comparat. Econ.

    (2010)
  • M.K. Hassan et al.

    Financial development and economic growth: New evidence from panel data

    Q. Rev. Econ. Finance

    (2011)
  • J.F. Houston et al.

    Creditor rights, information sharing, and bank risk taking

    J. Financ. Econ.

    (2010)
  • K. Imran et al.

    Banks dividend policy: evidence from Pakistan

    Econ. Modelling

    (2013)
  • K. John et al.

    Outside monitoring and CEO compensation in the banking industry

    J. Corp. Finance

    (2010)
  • R. Levine

    Finance and growth: theory and evidence

    Handb. Econ. Growth

    (2005)
  • A.N. Licht et al.

    Culture, law, and corporate governance

    Int. Rev. Law Econ.

    (2005)
  • M.H. Miller et al.

    Dividends and taxes

    J. Financ. Econ.

    (1978)
  • R.R. Pettit

    Taxes, transactions costs and the clientele effect of dividends

    J. Financ. Econ.

    (1977)
  • R.M. Stulz et al.

    Culture, openness, and finance

    J. Financ. Econ.

    (2003)
  • F. Allen et al.

    A theory of dividends based on tax clienteles

    J. Finance

    (2000)
  • S.C. Bae et al.

    Culture, corporate governance, and dividend policy: international evidence

    J. Financ. Res.

    (2012)
  • K.H. Bae et al.

    Creditor rights, enforcement, and bank loans

    J. Finance

    (2009)
  • H.K. Baker et al.

    Factors influencing dividend policy decisions of Nasdaq firms

    Financ. Rev.

    (2001)
  • M. Baker et al.

    A catering theory of dividends

    J. Finance

    (2004)
  • J. Barth et al.

    Bank regulation and supervision in 180 countries from 1999 to 2011

    J. Financ. Econ. Policy

    (2013)
  • Cited by (53)

    • Culture and payout policy: International evidence

      2023, Journal of Multinational Financial Management
    • Foreign bank lending: The role of home country culture during prosperous and crisis periods

      2022, Journal of Multinational Financial Management
      Citation Excerpt :

      Recently, Berger et al. (2021) extended this research focus by analyzing the relationship between culture and banks’ failures, whereas Haq et al. (2018) investigated banks’ leverage. Meanwhile, fewer studies focus on other aspects of cultural effects regarding, for example, banks’ performance (Halkos and Tzeremes, 2011; Boubakri et al., 2017) or banks’ propensity to pay dividends (Zheng and Ashraf, 2014). In our opinion, two gaps in the banking literature are particularly striking.

    • Journal of Behavioral and Experimental Finance: A bibliometric overview

      2022, Journal of Behavioral and Experimental Finance
    View all citing articles on Scopus

    This paper is funded by the project of National Natural Science Foundation of China (NSFC) (Grant # 71173077). We thank the editor, Brian Lucey, and the anonymous referees for their very constructive comments.

    1

    Address: Room #530, School of Management, Huazhong University of Science and Technology, 1037 Luoyu Road, Hongshan District, Wuhan 430074, Hubei, China. Tel.: +8613971135486.

    View full text