National culture and shareholders’ perspective on dividend and cash policy
Abstract
This research studies perspective shareholders have on managements decisions for dividend and cash policies. Other studies already document that national culture and its relation to dividend and cash holdings, however they only investigate the influences of national culture on managers. Under the assumption both have the same cultural background this study examines whether dividend and cash policy come from shareholders’ or
managers’ culturally determined preferences and determines whether an increase of dividend and cash are value enhancing from the shareholders’ perspective. To examine their perspective and its implications for value enhancement, culture is adopted within the valuation regression as proposed by Pinkowitz, Stulz, and
Williamson (2006). In this study culture is defined by Hofstede’s (1980, 2001, 2010) six cultural dimensions and a sample of 8340 firm-years from 25 EU28 countries between 2011 and 2015 is used. Evidence is found that shareholders from countries with a high score on individualism and indulgence see an increase in dividend as value enhancing. Meaning they perceive managements current position on dividend policy as sub-optimal. All other results found are insignificant. This implies that shareholders with high scores these cultural traits do not see an increase in dividend or cash as value enhancing and imply that managers take these cultural traits of investors into account and cater towards their preferences by adjusting cash and dividend policy accordingly.
Name: Kevin Kremer Student number: s2761262 Supervisor: Dr. Silviu Ursu
Introduction
The importance of national culture and its influences on financial and economic phenomena is increasingly being recognized. Various cultural characteristics have shown to influence decision-making of investors and managers. For example, Aggarwal, Kearney and Lucey (2012) and Zheng, Ghoul, and Guedhami (2012) find that culture influences financing structure decisions of companies; Li, Griffin, Yue, and Zhao (2013) report on cultures influence on corporate risk taking; Aggarwal, Kearney, and Lucey (2012) and Shao, Kwok and Zhang (2013) show how culture affects investment decisions. Other literature examines the effects of national culture on managerial decision-making with relation to dividend and cash policies (Chui, Loyd, and Kwok, 2002; Chang and Noorbakhsh 2009; Fidrmuc and Jacob, 2010; Shao, Gwok, and Guedhami, 2010; Chen, Dou, Rhee, Truong, and Veeraraghavan, 2015). However, literature on the investors’ perspective in relation to this subject is limited. Only Orlava, Roa, and Tang (2017) investigate the investors’ culturally determined perspective on cash policy. This study extends on their paper by exploring investors’ views on dividend policy and by extending the amount of cultural traits that are examined for cash policy. One possible reasoning that culture would not matter is that managers cater to shareholders’ preferences. Meaning they already optimise their policies by considering the culturally driven opinion shareholders have by distributing or accumulating cash in accordance with those opinions. However, just as Orlava Roa, and Tang (2017) show, investors and managers from the same cultural background can have various dissimilar preferences in these regards, meaning managers do not always cater and culturally driven agency problems are present. Furthermore, other previous studies do not determine whether current dividend and cash policies are value maximizing. Pinkowitz, Stulz, and
Williamson (2006) provide an adapted methodology of Fama and French (1998) which allows for the
examination of external influences and their implication in the valuation of dividend and liquid holdings. They regard the quality of corporate governance systems as proxies in support of agency theory. However, they do not consider investors’ and managers’ differing perceptions and its possible implications in the principal-agency relationship. Chui, Lloyd, and Kwok (2002) state “Culture does matter because it affects management perception of the cost and risk related to debt finance and agency problems in each country”. Thus, some agency problems might essentially be rooted in preferential differences depending on shareholders’ culturally driven opinion of the value-reducing aspects of managers decision-making. Furthermore, others note that national culture prevalent in the country of origin, at least partially, shapes the relationship between managers and principals (Fidrmuc and Jacob, 2010; Hooghiemstra, Hermes, and Emanuels, 2015).
dimensions. To examine shareholders’ preferences this study tries to answer two main questions. Firstly, are dividends and liquid holdings driven by managerial or shareholders’ cultural background? Though both managers and shareholders are driven by the same national cultural background they may simultaneously face coinciding and conflicting perspectives. In for example high uncertainty avoidance cultures, managers are expected to be more conservative as their personal wealth is directly related to firm performance and survivability. This will cause them to hold more cash and be more reserved in their dividend policy. On the other hand, investors are expected to hold diversified portfolios and as such do not face the same level of dependence on a single firms’ survivability. Rather they would prefer firms to pay-out idle cash through dividend and reinvest those funds in accordance to deal with their personal uncertainty. Secondly, do
shareholders experience the effects of their national culture on cash holdings and dividend as value enhancing? Chang and Noorbakhsh (2009) find that high uncertainty avoidance cultures hold higher cash balances, and Fidrmuc and Jacob (2010) find that high uncertainty avoidance pay less dividends. Question is however if these findings simultaneously translate to higher stock valuations.
The abovementioned research questions are addressed through assessing the role of national culture by
explaining the variation in valuation of dividend and liquid holdings by shareholders. This makes it possible to pinpoint the influences of the cultural characteristics. Furthermore, the focus on the mitigating role of cultural factors on cash and dividend valuation, in contrast to levels, makes it possible to access how shareholders national culture influences their perception on cash and dividend policies. Several other studies investigate the valuation of corporate cash holdings and/or dividends (Pinkowitz, Stulz, and Williamson, 2006; Kyröläinen, Tan, and Karjalainen, 2013; Orlava Roa, and Tang, 2017). Pinkowitz, Stulz, and Williamson (2006) and Kyröläinen, Tan, and Karjalainen, (2013) focus only on firm-level characteristics and corporate governance measures and their implications. Whereas Orlava Roa, and Tang, 2017 examine only three of Hofstede’s cultural dimensions and their relation to cash holdings. This study extends this literature by examine the effects of investors’ perspective driven by their cultural traits on dividend, and by the inclusion of all of Hofstede’s six cultural dimensions for the examination of both dividend and cash policies.
block. Limiting the non-firm level influences on value. Due to the choice of database and methodology the timeframe is from 2011 until 2015.
In this research evidence that shareholders in high individualistic and indulgent societies see an increase in dividends as value enhancing. In other words, they perceive the current dividend policy as sub-optimal, and would prefer to see higher pay-outs. For other dimensions linked to changes in dividend and all six dimensions for changes in cash no significant results are found. There seem to be no differencing preferences between shareholders and agents, implying that managers already cater to shareholders by taking their preferences in to account in their determination of cash and dividend policy.
Firstly, a review of the present academic literature is presented. Secondly, the methodology and hypotheses are presented. Following is the description of the data. Thereafter the results are presented, and finally conclusions are drawn.
Literature review
Within the classical dividend irrelevance theory of Miller and Modigliani (1961) shareholders unanimously support a firm’s dividend policy, because they perceive dividends and capital gains as perfect substitutes for one another. The tax clientele theory supports this statement, however only when firms attract shareholder clienteles facing the same effective marginal tax rates for dividend (Black and Scholes, 1974). However, in the face of asymmetric information, heterogeneous tax rates, and agency costs the views on dividend and capital gains might substantially differ across shareholders. The signalling theory argues that managers (insiders) have superior firm specific information over investors (outsiders). Within this situation of asymmetric information, managers use dividends as an informational deliverance medium conveying useful information about present and future profit and growth expectations (John and Williams, 1985). Bhattacharya (1979. Agency theory in its essence encompasses that the separation of ownership and management lead to principal-agent conflicts, as the managers may pursue their own interest at the expense of the principals. (Jensen and Meckling, 1976; Shleifer and Vishny, 1986). Jensen (1986) states managers exposed to excesses of free cash will rather invest it in negative net present value (NPV) projects than pay it out to shareholders. In other words, shareholders should view an increase in cash holdings as value decreasing as the cash could be wasted on invaluable projects, and they should positively value an increase in dividends.
2000b). Kalcheva and Lins (2007) examine the connection between corporate liquid holdings, firm-level agency costs, and the market value of companies. They find that in situations with agency problems the decision to pay dividend is positively viewed by outside investors. However, contradicting to these findings is the behaviour of firms found by studies like Bates, Kahle, and Stulz (2009). Their research notes a rising trend in corporate cash holdings from 1980 to 2006 for US companies. One of the first studies providing evidence for this trend are Opler, Pinkowitz, Stulz, and Wiliamson (1999). By doing a comprehensive analysis on corporate cash holdings they, among other things, examine the influence of agency costs on cash holdings. Though they do not find evidence for an influence of agency costs on cash holdings, they do find that firms hold cash to protect themselves from harmful cash flow shocks. Others like Bates, Kahle, and Stulz (2009) also find evidence for a precautionary rationale, however also not through an agency-based rationale. Additionally, Duchin (2010) supports the precautionary argument showing that increasing cash flow uncertainty (at least partially) explains increasing corporate cash holdings. Contradictory, others argue in favour of the agency perspective in corporate cash policies (Dittmar, Mahrt-Smit, and Servaes, 2003, Pinkowitz, Stulz, and Williamson, 2006). Dittmar and Mahrt-Smith (2007) find that managers subject to agency problems either invest excess cash inefficiently or that excess cash reduces their incentives to operate efficiently by for example lowering costs or increasing margins. These studies also find that shareholder protection has proven to
influence firms level of corporate cash holdings and shareholders’ valuation of dividend and cash policies. Dittmar, Mahrt-Smit, and Servaes (2003) studied the relationship between shareholder rights and corporate liquid holdings in 45 countries. They find that firms facing low levels of shareholder protection hold less cash than firms who face higher levels of shareholder protection and attribute this behaviour to agency theory. Pinkowitz, Stulz, and Williamson (2006) find a much weaker relationship between dividends and firm value in countries with strong investor protection, and a much weaker relationship between firm value and corporate liquid holdings in countries with weak investor protection.
dividend are already documented within the literature. Chang and Noorbakhsh (2009) find that corporate cash holdings differ substantially across countries with dissimilar cultures. Furthermore, Fidrmuc and Jacob (2010) report similar dissimilarities for dividend. So, one would expect that when culture is already documented to influence managers decision making, it should also influence investors’ preferences and opinion over the current cash and dividend policies. For example, both principal and agent might have a culturally determined dislike for uncertainty. A manager might therefore prefer to hold larger excess cash balances to protect against possible shortages in the future. However, a shareholder might view the idle sitting cash as more uncertain for him then when he could have invested it himself. Thus, he would prefer this cash payed-out to him, giving rise to an agency problem. To examine the exact cultural influences causing these differencing preferences a cultural framework is necessary. Several studies have attempted to distinguish such a framework and its dimensions, but the most influential of the works in cross-cultural psychology comes from Hofstede (1980, 2001). This study is later expanded by Hofstede, Hofstede, and Markov (2010). The theorem identifies primary dimensions of culture attributable to differences between thinking, values, and social behaviour among citizens from over 50 countries. Hofstede's research places these cultural attributes within six primary dimensions known as: Uncertainty Avoidance (UAI), Individualism (IDV), Power Distance (PDI), Masculinity (MAS), Long-Term Orientation (LTO), and Indulgence (IND). Even despite substantiation criticism and the
devolvement of more advanced cultural measurements, Hofstede’s dimensions remain the most prevalent in empirical research on national culture (Beugelsdijk, Kostova, and Roth, 2016). Kirkman, Lowe, and Gibson (2006) investigate various studies using Hofstede’s cultural framework and state that Hofstede’s cultural dimensions, even within their limitations, are clearly relevant for cross-cultural research.
On one hand recent studies already show that the national culture of agents influences the complexion of the agency relationship (Chui, Loyd, and Kwok, 2002; Chang and Noorbakhsh, 2009; Fidrmuc and Jacob, 2010; Hooghiemstra, Hermes, and Emanuels, 2015). On the other hand, the cultural background of shareholders in relation to agency theory has been relatively unexplored, even though shareholders and their relation to firm outcomes have already been broadly examined (Boyd and Solarino, 2016). This might be caused by the weak theoretical link between shareholders’ culture and firm outcomes, or because shareholder composition is often large and diverse making it more difficult to examine how culture influences their combined values and believes in comparison to managers. However, the reasoning that managerial cultural background influences their decision-making argues for similar importance of shareholders’ national culture. There is so far no
theoretical argument why shareholders’ national should not influenced their perception of agency problems in a similar way (Fidrmuc and Jacob, 2010). The only study which examines shareholders’ perspective throughout their valuation of corporate cash holdings with relation to culture is Orlova, Roa, and Kang (2017). They limit their research to three of Hofstede’s cultural dimensions and find that investors’ exhibit different cultural preferences than management culturally driven decision-making with regards to corporate liquid holdings. Their findings imply that agency problems between managers and investors of the same culture exist, as
been linked to a variety of other investor behaviour already. Beugelsdijk and Frijns (2010) find that high uncertainty avoidance explains a lower amount of foreign equity investment. Furthermore, higher levels of uncertainty avoidance cause home bias and under-diversification, however higher levels of long-term
orientation reduce home bias (Anderson, Fedenia, Hirschey, and Skiba, 2011). Individualism has shown links to self-attribution biases, trading activity, momentum effect (Chui, Titman, and Wei, 2010), and more foreign equity investment (Beugelsdijk and Frijns, 2010). Whereas, Anderson, Fedenia, Hirschey, and Skiba (2011) link masculinity to investor overconfidence. Furthermore, Chang and Lin (2015) document the relation between power distance and herding behaviour.
The previous studies mentioned above that all of Hofstede’s dimensions seem to be relevant in explaining parts of their behaviour. Hofstede (1980) states that power distance (PDI) is a measure of the extent to which the less powerful members of a society expect and accept that power is distributed unequally. In high power distance countries society is unequal and investors acquire information in unequal conditions. In such cases investors are more likely to accept implied information from others, like managers (Chang and Lin, 2015). Also, Fidrmuc and Jacob (2010) argue that in high power distance societies lower dividend pay-outs and consequently higher corporate cash holdings are culturally more accepted, as investors are expected accept managements decision-making more easily. In low power distance countries investors have a strong cultural value emphasis on equality in wealth and power, thus investors should expect higher dividends because their need for equality increases agency tensions (Fidrmuc and Jacob, 2010). Thus, this paper hypothesises that in high power distance countries an increase in dividends is expected to be viewed negatively by investors, and an increase in
corporate cash holdings should be viewed positively. Meaning that shareholders should see an increase of dividend as value decreasing, and an increase of cash holdings as value enhancing. This paper therefore hypothesises:
H0: In high PDI countries investors perceive an increase of dividend as value decreasing. H0: In high PDI countries investors perceive an increase of cash as value enhancing.
Orlova, Roa, and Kang (2017) argue that investors in individualistic countries are strongly concerned about their interest, and sensitive to the efficiency of managements handling of excess cash. In individualistic countries managers tend to be overconfident and take excessive risk, thus investors are expected to prefer higher dividend to limit managers spending capabilities on questionable investments. Following their reasoning this paper hypothesises that investors from individualistic societies will see an increase in cash as value enhancing, as this signifies less “wasted” investments in inefficient projects. Similarly, one can argue that increasing dividend also limit managers in sub-optimal investing, thus an increase in dividend should also be seen as value enhancing by investors from individualistic countries.
H0: In high IDV countries investors perceive an increase of dividend as value enhancing. H0: In high IDV countries investors perceive an increase of cash as value enhancing.
Masculinity (MAS) is a measure of the degree of cultural toughness vs. tenderness in a society (Hofstede 1980). It concerns the distinction between gender roles with on one side masculinity, translating to arbitrary, strong, and with a focus on material achievements, and on the other feminity, encompassing tenderness and a focus on the quality of life. In masculine societies people care about materialism and are expected to engage in
opportunistic behaviour to please their greed (Chang and Lin, 2015). Zheng and Ashraf (2014) note something similar and state: “in masculine cultures, agency conflicts are inherently more severe because their members are considered competitive and impatient and are more prone to pursue opportunistic behaviour rather than adhere to others’ decisions and preferences”. Therefore, shareholders are expected to positively value an increase of dividends as this limits the opportunistic behaviour of managers. Consequently, increases in cash would allow for future opportunistic behaviour and shareholders are thus expected to want more control in the allocation of firms’ assets and as such also prefer lower cash holdings. Thus, investors in masculine societies will regard an increase in cash as value diminishing.
H0: In high MAS countries investors perceive an increase of dividend as value enhancing. H0: In high MAS countries investors perceive an increase of cash as value decreasing.
negative relationship between UAI and firms’ dividend levels. On the other hand, they state that investors would prefer an increase of dividend, as they dislike the uncertainty associated with capital gains. Orlova, Roa, and Kang (2017) state that well diversified investors may perceive increases in cash by management as
inefficient and value them negatively. Due to the uncertainty that comes with capital gains investors are expected to value an increase in dividends positively.
H0: In high UAI countries investors perceive an increase of dividend as value enhancing. H0: In high UAI countries investors perceive an increase of cash as value decreasing.
Long-term orientation (LTO) represents values such as patience and preference for long-term results (Hofstede, 2001). According to Chang, and Lin (2015) investors from a high LTO society are expected to focus on future returns, whereas in low LTO environments they are more concerned with current returns.
Zheng and Ashraf (2014) find a negative relation between dividend pay-out and long-term orientation. Furthermore, Chang and Noorbaksh (2009) find that firms located in countries with higher long-term orientation have higher cash balances, as managers are more precautious and prefer a more stable outlook. Zheng and Ashraf (2014) hypothesize that in high LTO countries the greater perseverance, thrift and patience will decrease the severity of agency conflicts. Consequently, investors have a lesser need for dividends to serve as a disciplining mechanism and thus find lower dividend pay-outs culturally more acceptable.
Orlova, Roa, and Kang (2017) argue that shareholders from high LTO countries will put less pressure on managers to produce short-term gains. However, they state that this does not automatically mean that they prefer managers to hold more cash. Instead investors would prefer the distribution of unutilized cash so that they can invest in other firms that suit their LTO preferences. Consequently, this study hypothesises that investors from high LTO societies will value an increase in dividend positively, and they would value an increase in cash holdings negatively.
H0: In high UAI countries investors perceive an increase of dividend as value enhancing. H0: In high UAI countries investors perceive an increase of cash as value decreasing.
Indulgence (IND) is relatively new, and comprehensive research in finance with relation to this dimension is still minimal. Indulgence stands for a society that allows for a relatively free gratification of basic and natural human drives in order to enjoy life and have fun, whereas restraint stands for a society that suppresses
likely to engage in the payment of dividend, though they do not attribute a specific reasoning to their finding. Therefore, by the lack of implications in the previous literature this study is unable to draw expectations for an investors’ perspective on managements’ dividend and cash policy and will investigate if investors from high indulgence countries perceive an increase of dividend and cash as value enhancing or diminishing.
H0: In high IND countries investors perceive an increase of dividend as value enhancing/decreasing. H0: In high IND countries investors perceive an increase of cash as value decreasing/enhancing. Methodology
There are numerous firm-level determinants which have proven to have an impact on the market value of firms. Profitability, growth opportunities, investment opportunities, dividend policy, cash policy, and leverage policy have all be examined for its influences on market value and used for the estimation of it. Other institutional influences like investor protection have also proven to influence investors’ opinion of dividend and cash policies. To properly test the abovementioned hypotheses a thorough method is required which captures the effects of institutional and firm-level determinants on market value, such that solely shareholders’ culturally determined perspective can be examined by looking into their perception of a change in dividend and cash policy.
One such model which fits the aforementioned requirements is presented in Pinkowitz, Stulz, and Williamson (2006). The model was originally developed by Fama and French (1998) to examine the taxation effects of dividend and debt on firm value. It contains a variety of firm-level determinants and the current and future changes in these variables. This method is known as the valuation regression model and their original methodology is presented in equation 1 and looks as follows1:
𝑉𝑖𝑡 = 𝛼 + ß1𝐸𝑖𝑡+ ß2𝑑𝐸𝑖𝑡+ ß3𝑑𝐸𝑖,𝑡+2+ ß4𝑑𝑇𝐴𝑖𝑡+ ß5𝑑𝑇𝐴𝑖,𝑡+2+ ß6𝑅𝐷𝑖𝑡+ ß7𝑑𝑅𝐷𝑖𝑡 + ß8𝑑𝑅𝐷𝑖,𝑡+2+ ß9𝐼𝑖𝑡+ ß10𝑑𝐼𝑖𝑡+ ß11𝑑𝐼𝑖,𝑡+2+ ß12𝐷𝑖𝑡+ ß13𝑑𝐷𝑖𝑡 + ß14𝑑𝐷𝑖,𝑡+2+ 𝛽15𝑑𝑉𝑖,𝑡+2+ 𝜀𝑖𝑡
(1)
Where 𝑉𝑖𝑡is the market value of the firm; 𝐸𝑖𝑡is earnings before extraordinary items plus interest, deferred tax and investment tax credits; 𝑇𝐴𝑖𝑡 is total assets; 𝑅𝐷𝑖𝑡 is research and development (R&D) expense; 𝐼𝑖𝑡 is interest expense; 𝐷𝑖𝑡 and is dividends defined as common dividends paid.
Pinkowitz, Stulz, and Williamson (2006) choose this model as it fits their need for the examination of dividend in countries with low and high shareholder protection, and as it performs well when subjected to a variety of tests. Furthermore, to examine the influences of corporate cash holdings they adjust the model it by splitting of Cash holdings from total assets. To examine the culturally determined perspective of shareholders on both cash
1 Note that all variables are deflated by total assets to control for heteroskedasticity and 𝑉
and dividend this paper therefore uses their adjustment to the valuation regression model. Thus, following this measure the methodology looks as presented in equation 22:
𝑉𝑖𝑡 = 𝛼 + ß1𝐸𝑖𝑡+ ß2𝑑𝐸𝑖𝑡+ ß3𝑑𝐸𝑖,𝑡+2+ ß4𝑑𝑁𝐴𝑖𝑡+ ß5𝑑𝑁𝐴𝑖,𝑡+2+ ß6𝑅𝐷𝑖𝑡 + ß7𝑑𝑅𝐷𝑖𝑡 + ß8𝑑𝑅𝐷𝑖,𝑡+2+ ß9𝐼𝑖𝑡+ ß10𝑑𝐼𝑖𝑡+ ß11𝑑𝐼𝑖,𝑡+2+ ß12𝐷𝑖𝑡+ ß13𝑑𝐷𝑖𝑡 + ß14𝑑𝐷𝑖,𝑡+2+ 𝛽15𝑑𝑉𝑖,𝑡+2+ 𝛽16𝑑𝐿𝑖𝑡+ 𝛽17𝑑𝐿𝑖,𝑡+2+ 𝜀𝑖𝑡
(2)
Where 𝑁𝐴𝑖𝑡 net assets equals total assets minus liquid assets, and 𝐿𝑖𝑡 stands for liquid asset holdings.
Variants of this model has been furthermore used to examine the influences of stock repurchases (Haw, Ho, Hu, and Zang, 2011), and creditor protection (Kyröläinen, Tan, and Karjalainen, 2013) on value.
In equation one and two 𝑋𝑡 is the variable X in year t. 𝑑𝑋𝑡 represents the change in 𝑋𝑡 from t – 2 to time t, 𝑋𝑡− 𝑋𝑡−2. These are included to reflect the change of these variables and represent their growth trend. 𝑑𝑋𝑡+2 is the change 𝑋𝑡 in year t to year t+2, 𝑋𝑡+2− 𝑋𝑡. These are incorporated to absorb changes in expectations. The decision for two years comes from Fama (1990) stating that this is about as far as the market can predict ahead. Both models use the same firm-level proxies as determinants of firm value. Earnings is used as a profitability proxy. The changes in total and net assets proxy for the tangible investment component of cash flows and investment opportunities, whereas R&D is included for the intangible investments part. Furthermore, firms invest when prospects are good, and expected profits are high. Thus, both variables can also pick up
information about profits missed by the other variables (Fama and French, 1998). Finally change in value from time t to t+2 is included to get rid of future changes in other explanatory variables caused by their unexpected components.
This study uses the methodology as proposed by Fama and French (1998) with the separation of liquid holdings as proposed by Pinkowitz, Stulz, and Wiliamson (2006). The variables proposed are largely similar, albeit an addition and some alternative measures. Firstly, in contrast to Fama and French (1998) who measure the effects on the added value of assets by subtracting total assets from the market value this study uses market
Furthermore, instead of interest as a proxy for leverage policy, book leverage equal to long-term debt plus current portion of long-term debt is used as proposed by Barclay, Smith, and Morellec (2006). In contrast to excess cash as measured by Pinkowitz, Stulz, and Williamson (2006) and Kyröläinen, Tan, and Karjalainen, (2013) this study will measure cash holdings as total cash and cash equivalents as presented on the balance sheet to be the determinant for cash policy. Lastly, dividends are similarly established to Fama and French (1998), proxying for dividend policy. This all leads to the following abbreviation of the valuation regression model as presented by equation three:
𝑉𝑖𝑡 = 𝛼 + ß1𝐸𝐵𝐼𝑇𝑖𝑡+ ß2𝑑𝐸𝐵𝐼𝑇𝑖𝑡+ ß3𝑑𝐸𝐵𝐼𝑇𝑖,𝑡+2+ ß4𝑑𝑇𝑄𝑖𝑡 + ß5𝑑𝑇𝑄𝑖,𝑡+2+ ß6𝑑𝑆𝑖𝑡 + ß7𝑑𝑆𝑖,𝑡+2+ ß8𝐿𝐸𝑉𝑖𝑡+ ß9𝑑𝐿𝐸𝑉𝑖𝑡+ ß10𝑑𝐿𝐸𝑉𝑖,𝑡+2+ ß11𝐷𝑖𝑡 + ß12𝑑𝐷𝑖𝑡 + ß13𝑑𝐷𝑖,𝑡+2+ 𝛽14𝑑𝑉𝑖,𝑡+2+ 𝛽15𝑑𝐿𝑖𝑡+ 𝛽16𝑑𝐿𝑖,𝑡+2+ 𝜀𝑖𝑡
(3)
Where 𝑉𝑖𝑡 is the market value equal to market capitalization; 𝐸𝐵𝐼𝑇𝑖𝑡 is earnings before interest and tax; 𝑇𝑄𝑖𝑡 is Tobin’s Q; 𝑆𝑖𝑡 is sales; 𝐿𝐸𝑉𝑖𝑡 is leverage; 𝐷𝑖𝑡 is common dividend paid; 𝐿𝑖𝑡 is cash and cash equivalents. All variables 𝑋𝑡 are deflated by total assets at time t to deal with heteroskedasticity issues (Fama and French, 1998)3, as the results are likely to be dominated by the largest firms. 𝑋𝑡 is the variable X in year t. 𝑑𝑋𝑡 is the amount of change in 𝑋𝑡 from t – 2 at time t, 𝑋𝑡− 𝑋𝑡−2 reflecting the growth trend. 𝑑𝑋𝑡+2 is the change 𝑋𝑡 in year t to year t+2, 𝑋𝑡+2− 𝑋𝑡 which is included to absorb changes in expectations.
The Tobin’s Q as proposed by Peters and Taylor (2017) is defined as follows, they propose a new measure for Tobin’s Q which that also accounts for a measure of intangible capital and conclude this measure as a superior proxy for tangible and intangible investment opportunities. Their version of Tobin’s Q scales firm value by the sum of physical and intangible capital:
Qtotit = Vit
Kitphy+ Kitint (4)
Where Qittot is the proposed proxy for Tobin’s Q; Vit is equal to market capitalization; Kit phy
is physical capital is equal to tangible fixed assets; Kitint intangible capital is equal to purchased intangible fixed assets plus internally created knowledge and organizational capital defined as4:
Kit = (1 − δK)Ki,t−1+ R&Dit+ 0,3 ∗ SG&Ait (5) Where Kit is the in-period intangible capital; Ki,t−1 is the end-of-period stock of internally created intangible capital; δK is the depreciation rate of the internally created intangible capital set at 20%; R&Ditis the research and development cost; SG&Ait are the selling general and administrative expenses; Kit is assumed zero in first year.
To examine shareholders’ culturally determined perspective two interaction terms are needed where the change in cash and the change in dividend interact with one of Hofstede’s cultural dimensions. Therefore, we further
3 Except for Tobin’s Q, which is deflated by the total of physical and intangible capital.
4 This form of Peters and Taylor (2017) determination of intangible capital only holds when an equal depreciation rate is
expand the methodology by including these two terms. Furthermore, rather than having a continuous measure of Hofstede’s cultural dimensions, the sample is split for all of Hofstede’s dimensions using indicator variables that equal one in above-median index scores. With these multiple regressions will be done with each only one of Hofstede’s cultural dimensions to individually examine the effects of all dimensions on the perspective of shareholders. In order to capture the direct effects of the individual cultural traits on firm value the cultural dimension is also individually included within the regression model. Ultimately, a positive sign of an interaction variable will mean that shareholders with a high score in a cultural dimension will perceive an increase in the change of cash or dividend as value enhancing. In other words, they perceive managements position on cash and dividend policy as suboptimal and would prefer higher dividend payments or increased cash holdings.
Additionally, as mentioned in the literature review studies like Pinkowitz, Stulz, and Williamson (2006) find that dividend is valued more by investors in countries with weaker investor protection, and that cash holdings are valued more in countries with stronger investor protection. To control for this, investor protection is added as a control variable. Following Haw, Ho, Hu, and Zhang (2011) investor protection will be proxied by the anti-self-dealing index by Djankov, La Porta, Lopez-de-Silanes, and Shleifer (2008) and included as an above or below median value indicator variable. This leads to the final model as presented in equation three:
𝑉𝑖𝑡 = 𝛼 + ß1𝐸𝐵𝐼𝑇𝑖𝑡+ ß2𝑑𝐸𝐵𝐼𝑇𝑖𝑡+ ß3𝑑𝐸𝐵𝐼𝑇𝑖,𝑡+2+ ß4𝑑𝑇𝑄𝑖𝑡+ ß5𝑑𝑇𝑄𝑖,𝑡+2+ ß6𝑑𝑆𝑖𝑡 + ß7𝑑𝑆𝑖,𝑡+2+ ß8𝐿𝐸𝑉𝑖𝑡+ ß9𝑑𝐿𝐸𝑉𝑖𝑡+ ß10𝑑𝐿𝐸𝑉𝑖,𝑡+2+ ß11𝐷𝑖𝑡+ ß12𝐶𝐷𝑗 ∗ 𝑑𝐷𝑖𝑡 + ß13𝑑𝐷𝑖𝑡+ ß14𝑑𝐷𝑖,𝑡+2+ 𝛽15𝑑𝑉𝑖,𝑡+2+ 𝛽16𝑑𝐿𝑖𝑡+ 𝛽16𝑑𝐿𝑖,𝑡+2 + 𝛽17𝐶𝐷𝑗 ∗ 𝑑𝐿𝑖𝑡 + ß18𝐶𝐷𝑗 + 𝐶𝐷𝑗+ 𝐼𝑃𝑗+ 𝜂 + 𝜔 + 𝜀𝑖𝑡
(3)
Where 𝐶𝐷𝑗 is one of Hofstede’s six cultural dimensions; 𝐼𝑃𝑗 is the measure of investors protection; 𝐶𝐷𝑗∗ 𝑑𝐷𝑖𝑡 and 𝐶𝐷𝑗 ∗ 𝑑𝐿𝑖𝑡 are the cultural interaction terms of interest. Similar to Kyröläinen, Tan, and Karjalainen, (2013), the terms 𝜂 and 𝜔 are included for possible differences in value that prompt from time and omitted country factors for time and omitted country factors. For the inclusion of both time fixed effects, as denoted by 𝜂, and country fixed effects, as denoted by 𝜔, testingwas performed to see if their coefficients where jointly equal to zero. These tests returned significant and thus to control for time and country fixed effects both terms are included. Another important determination when dealing with panel data is to make an appropriate decision about which statistical model is used. To determine the appropriate statistical method for testing the results several tests are performed. Firstly, one which determines the appropriateness of the fixed effects model over OLS. The result of this test was found significant, and thus fixed effects are preferred over pooled OLS. Secondly, a Breusch-Pagan Lagrange multiplier test is done to determine if fixed effects is preferred over pooled OLS. The return of this result is also significant. Lastly, a Hausman test has been done to decide
demeaning process of the variables. Important factors in this study like investor protection and the individually regressed Hofstede’s dimensions will then thus be ignored. Also, for datasets with a large number of
observations (8250 firms) over a short timeframe (5 years) the fixed effects estimation is inconsistent (Baltagi, 2005). The results of the Hausman test however show that there is correlation present between unique errors and the regressors in the model, meaning that the assumption of no correlation between unobserved effects and explanatory variables. This would imply inconsistency in the random effects model due to omitted variables. Therefore, this study opts for the, albeit sub-optimal, pooled Ordinary Least Squares (OLS) regression method. For the results of these tests see appendix B.
Data and descriptive statistics
For the methodology as described in the previous section, firm-level data as well as data on Hofstede’s cultural dimensions for all countries are required. In this section the motivation and properties of the data used will be discussed. Firm-level data is collected using the Orbis World database. The data comes from financial
statements which originate from multiple countries with different accounting conventions. This could cause bias by different specifications of individual items within the financial statements.
For each firm, a variety of financial variables is collected. This includes cash and cash equivalents, dividends paid, total assets, long-term debt, current liabilities, year-end market capitalization, earnings before interest and taxes, research and development expense, interest expenses, selling, and general, and administrative expenses. The choice for time is limited to data availability. Orbis only provides data from 2009-2018, as they only provide the latest 10 years of data. This research limits that timeframe to the years from 2009 until 2017, as data from 2018 contains large amounts of missing observations for a variety of key variables. Furthermore, the methodology requires past changes and future changes of the independent variables, these are the amounts of change over 2 years. At the bottom this implies the past change of variable X from 2009 to 2011 needs to be included in 2011, and at the top the future change from 2015 to 2017 has to be included in 2015. Coincidently, this leaves the dataset with 5 firm-years of observations starting from 2011 until 2015. The firm-level data is gathered using the following data selection process. Firstly, the companies must be listed as value is going to be the independent variable. Secondly, data is limited to firms from EU28 countries. The choice for EU28
countries is partly made for comparability issues like mentioned before, as the EU enforced the mandatory adoption of the IFRS in 2005. Also, countries within the EU operate as an economic and political block and thus are expected to experience similar institutional and economic shocks, limiting non-firm level influences on value such as new trade agreements. Opposite to the similarities mentioned above culture still differs
more things needed to be done to finalize the dataset. Firstly, Orbis only allows to look for market
capitalization from 2017 until 2014 in its search strategy engine. Secondly, the dataset includes firms from Cyprus of which no cultural data is available. After deletion of firms with missing and negative market values for earlier years, and the deletion of firms from Cyprus this leaves a sample of 1650 firms each with 5 firm-year observations translating to a total of 8250 firm-year observations. The sample is comparable in nature to studies like Pinkowitz, Stulz, and Williamson (2006), Kyröläinen, Tan, and Karjalainen (2013), and Orlava, Roa, and Kang (2017), as they also collect cross-country data to examine the valuation of dividends and/or corporate cash holdings.
All variables are deflated by total assets with the exception of Tobin’s Q as mentioned before in the methodology. Furthermore, following Kyröläinen, Tan, and Karjalainen (2013) all the firm-level data is winsorized at 1% in both tails to deal with outliers most likely caused by random mistakes within the dataset. Summary statistics of all variables are given below in table one.
Table 1: Summary statistics dependent and independent variables.
Variable Mean SD Min Max N
Dependent variable
Market Capitalization 1,021 1,216 0,031 7,486 8250
Independent variables
Earnings before Interest and Tax 0,029 0,151 -0,766 0,329 8250
Tobin's Q 2,240 6,273 0,026 45,992 8248
Sales 0,947 0,686 0,001 3,644 8168
Book Leverage Ratio 0,163 0,162 0,000 0,703 8020
Dividend 0,019 0,030 0,000 0,179 8250
Cash and Cash Equivalents 0,135 0,147 0,001 0,757 8250
In the above table mean market capitalization deflated by total assets is slightly higher than one indicating that on average firms included in the sample are worth slightly more than their book value on assets. The only variable which takes on negative values is EBIT. Other variables are all larger than zero, which is in line with the expectations and proves the selection process and winsorization eliminated data which would bias the results. Largest variation can be found in the measure of Tobin’s Q.
Table 2: Summary Statistics per Country Country Num. of Obs. Num. of Firms Market
Capitalization Dividend Cash holdings
Mean SD Mean SD Mean SD
Austria 180 36 0,61 0,50 0,02 0,02 0,10 0,09 Belgium 300 60 0,86 0,89 0,02 0,03 0,11 0,13 Bulgaria 10 2 0,42 0,37 0,02 0,03 0,09 0,06 Czechia 5 1 1,33 0,30 0,07 0,10 0,35 0,10 Germany 1480 296 0,97 1,02 0,02 0,03 0,15 0,15 Denmark 245 49 1,62 2,02 0,02 0,04 0,16 0,20 Estonia 30 6 0,51 0,24 0,01 0,02 0,05 0,03 Spain 275 55 0,80 1,17 0,02 0,03 0,09 0,09 Finland 205 41 1,06 1,07 0,04 0,04 0,13 0,11 France 1175 235 0,73 0,79 0,01 0,02 0,14 0,13 Great- Britain 1790 358 1,44 1,53 0,02 0,03 0,16 0,18 Greece 440 88 0,36 0,52 0,00 0,01 0,10 0,13 Croatia 30 6 0,49 0,30 0,00 0,01 0,06 0,06 Hungary 10 2 0,72 0,40 0,01 0,01 0,11 0,07 Ireland 110 22 1,44 1,46 0,02 0,03 0,20 0,15 Italy 255 51 0,75 0,98 0,02 0,03 0,13 0,10 Luxemburg 55 11 0,48 0,29 0,02 0,02 0,10 0,10 Malta 35 7 1,44 1,58 0,03 0,04 0,09 0,08 Netherlands 260 52 0,99 1,10 0,02 0,02 0,10 0,10 Poland 365 73 0,66 0,73 0,01 0,02 0,09 0,10 Portugal 105 21 0,42 0,38 0,01 0,02 0,07 0,06 Romania 10 2 0,32 0,25 0,01 0,02 0,04 0,04 Sweden 860 172 1,36 1,39 0,03 0,04 0,13 0,15 Slovenia 20 4 0,50 0,43 0,02 0,02 0,03 0,01
Quite some variation can be found across the market-to-book ratio, dividend, and cash holdings. Countries with the highest market-to-book ratios are Denmark, Great-Britain, and Malta with value of 1.62, 1.44, and 1.44 respectively. On average the lowest market-to-book ratio can be found in Bulgaria with a value of 0,42. Dividends seem to be more closely distributed around the mean of the full sample. Though some countries on average stand-out in paying the least amount of dividends to total assets. Firms from Estonia, Greece, Croatia, and Hungary all on average only pay around one percent of total assets out as dividend. Cash holdings of firms also largely vary ranging from one percent of total assets in Slovenia to 35% in Czechia.
A large part of the dataset is dominated by four countries: Germany, France, and Great-Britain. They account for 64% of the total observations, 18%, 14%, 22%, and 10% respectively. Fortunately, there seem to be limited cultural similarities between their Hofstede scores, however to still control for possible bias caused by the domination of these countries within the sample, as robustness a check, the regressions will be done with the exclusion of these four countries.
(UAI), Individuality vs. Collectivism (IND), Long/Short-term Orientation (LTO), Masculinity vs. Femininity (MAS), and Indulgence vs. Restraint (IND). The individual country scores are given in table 3 below.
Table 3: Hofstede’s Cultural Dimensions
Countries PDI UAI IDV MAS LTO IND
Austria 11 70 55 79 60 63 Belgium 65 94 75 54 82 57 Bulgaria 70 85 30 40 69 16 Czechia 57 74 58 57 70 29 Germany 35 65 67 66 83 40 Denmark 18 23 74 16 35 70 Estonia 40 60 60 30 82 16 Spain 57 86 51 42 48 44 Finland 33 59 63 26 38 57 France 68 86 71 43 63 48 Great- Britain 35 35 89 66 51 69 Greece 60 100 35 57 45 50 Croatia 73 80 33 40 58 33 Hungary 46 82 80 88 58 31 Ireland 28 35 70 68 24 65 Italy 50 75 76 70 61 30 Luxemburg 40 70 60 50 64 56 Malta 56 96 59 47 47 66 Netherlands 38 53 80 14 67 68 Poland 68 93 60 64 38 29 Portugal 63 99 27 31 28 33 Romania 90 90 30 42 52 20 Sweden 31 29 71 5 53 78 Slovenia 71 88 27 19 49 48
Results
In this section the results will be presented. To empirically access shareholders’ culturally driven perspective on dividend and cash policy the in equation (3) proposed augmented valuation regression model is tested using the pooled OLS regression. The tests for the choosing the appropriate regression method show that the fixed effects model was preferred over pooled OLS and random effects. However, as reasoned before this paper will use pooled OLS. Also, as mentioned in the methodology time and country level fixed-effects are still included. The results for the testing of the regression method can be found in appendix B.
Next to controlling for heteroskedasticity by deflating the independent variables by total assets, Petersen and Taylor (2009) suggest the use of clustered standard errors in samples with a large number of firms and a short time window. As the collected data consists of 1850 firms with 5 firm-years each these standard errors will be implemented. Furthermore, a large part of the sample is dominated by observations from only four countries, namely Germany, France, Great-Britain, and Sweden. This issue will be addressed in the robustness checks. Other robustness checks will be a regression using the original variables as seen in equation 2, the Fama and Macbeth (1973) regression method, and GLOBE as alternative measure of culture by House, Hanges, Javidan, Dorfman, and Gupta (2004). Table 4 below shows the results for the main equation (3) using pooled OLS. The first column are the results of the interaction effect of the cultural dimensions on the change in dividend. The second column presents the results of the interaction effect on the change in liquid assets.
Table 4: Results of the interaction variables of dividend and liquid holdings with Hofstede’s cultural Dimensions within the valuation regression model.
Change in
Dividend Change in liquid Assets
PDI -2,52 -0,05 (1,66) (,45) IDV 7,21*** 0,15 (2,26) (,61) MAS 2,91 0,15 (1,97) (,61) UAI -1,64 0,18 (1,62) (,44) LTO -2,41 -0,29 (1,9) (,5) IND 3,42* 0,22 (1,85) (,51)
Standard deviations are given between brackets.
within the hypothesis that shareholders should value an increase in dividend as value enhancing. However, this is again insignificant. Opposite to this the signs found for UAI and LTO are opposite to those projected by the hypotheses. In terms of UAI this encompasses that investors would value an increase in dividend lower in societies with high uncertainty avoidance. This would contradict the hypothesis and the implication that they dislike the uncertainty that comes with capital gains. With LTO the sign seems to show that investors value in increase in dividend less. This would contradict the hypothesis and that investors would prefer the distribution of unutilized cash. In this case it could mean that they not only apply reduced pressure on managers to produce short-term goals, but also trust managers with acting in their interest in terms of long-term value creation. However, once more for clarity no significance is found for both UAI and LTO. The insignificance of these four interaction terms could mean that the effects of shareholders’ culturally determined perspective are non-existent. Managers could already cater to shareholders by considering this their views on agency problems and preferences for dividend policy. In other words, shareholders see the current level of dividend as value
maximizing.
In contrast to the findings for PDI, MAS, LTO and UAI significant results are found for the influence of IDV and IND. A positive and significant results at the 1% level is found for the effects of individualism. Consistent with the made hypothesis this means that dividend policy is not optimal in shareholders’ opinion, and that an increase of dividend in individualistic societies is value enhancing. This result is consistent with the reasoning based on Orlova, Roa, and Kang (2017) that investors view management as overconfident and think they take excessive risk. Thus, shareholders prefer higher dividend to limit managers spending capabilities on non-optimal investments. Similarly, a positive and significant result at the 10% level is found for the effects of indulgence. Meaning shareholders in high indulgence countries view an increase of dividend as value enhancing. This result implies the relevance of the IND dimension. Shareholders find the current level of dividend sub-optimal and would prefer an increase in dividend payments. This could imply that even though all the investment information sharing by management, shareholders in high IND countries would rather reinvest these funds themselves as they might satisfy their needs for gratification more this way.
No significant results are found for any of the interaction terms of change in cash holdings and Hofstede’s cultural dimensions. The only two terms which signs are in accordance with expectations as set in the hypotheses are IDV and LTO. With individualism the sign is positive, which could imply that shareholders perceive managements overconfidence and excessive risk taking in investments as value decreasing, and thus see an increase in liquid holdings as value enhancing knowing that it is not expended on invaluable investing opportunities. For long term orientation the sign is negative, which could imply that shareholders in high LTO countries value an increase in corporate cash holdings less. If significant the sign would have proved
For PDI a negative sign is found. Meaning an increase in liquid holdings would be valued less by investors. This would also contradict the reasoning based on Fidrmuc and Jacob (2010) that shareholders need for equality increases the agency tensions. If significant it would seem that shareholders even though increases in corporate cash holdings should be culturally more accepted don’t view that to be value enhancing. The interaction term with MAS has a positive sign. Meaning shareholders would value an increase in liquid holdings positively. This is opposite to the expectations set by Zheng and Ashraf (2004) and might mean that shareholders view an increase in liquid holdings as cash not wasted by managers opportunistic behaviour. With uncertainty avoidance the sign found is positive. Investors here could seem to agree with management that due to uncertainty firms should hold more cash, and value an increase in cash accordingly. Lastly, for indulgence the results could imply that shareholders seem to value an increase in liquid holdings positively. The open information sharing of managers seems to be positively viewed by shareholders and they trust management to make valuable investment opportunities in the future. However, important to note again, the results for these variables are insignificant. In conclusion the insignificance of these terms could mean that the effects of shareholders’ culturally determined perspective with relation to liquid holdings are non-existent. Managers could already cater to shareholders by taking their views on agency problems and preferences in to account when determining cash policy. In other words, shareholders see the current levels of cash holdings as value maximizing.
Robustness
As mentioned before a variety of robustness checks is performed to examine the validity of the results. The tables of these results and the relevant summary statistics are presented in appendix A. Firstly, Hofstede’s cultural dimensions are replaced with cultural dimensions of the GLOBE study of House, Hanges, Javidan, Dorfman, and Gupta (2004) an overview of the cultural scores of GLOBE are given in table A1 and a
comparison of them to Hofstede’s dimensions in table A2. The results for the regression results with globe are given in table A5. Different signs for the change in dividend with the GLOBE are found for the change in dividend interaction variables with PDI, MAS, and LTO. For liquid holdings IDV, MAS, UAI, and LTO do not align with their earlier found signs. However, except for MAS and UAI interacting with the change of dividend these are all not significant. The regression with the GLOBE variables shows a negative relation between these two interaction variables, meaning in both cases investors perceive an increase in dividend as value diminishing.
The results for the robustness check using the Fama and French (1998) variables can be seen in table A6, and the relevant summary statistics in table A3. These results are more in sync with our original results. Correct signs and significance is found for the change in dividend interaction terms with IDV and IND supporting the results. Furthermore, all other terms of dividend in this specification also show significant results. No
Next to robustness checks which use a different set of independent variables, this paper also checks for robustness using the originally proposed statistical regression method which stems from Fama and Macbeth (1973), instead of using OLS. The results for this regression are found in table A7. As for the Fama and French (1998) variables these results show the correct signs and for all the dividend interaction terms and the
significance of all variables support our findings for IND and INV. The signs for liquid holdings are also largely equal, except for MAS. Interestingly, this method returns significance for PDI, IDV, and MAS arguing that investors do perceive changes in liquid holdings as value enhancing or decreasing, and that managers do not cater to investors in relation to these cultural traits.
Lastly, to deal with the dominance of four countries in the sample (Germany, France, Great-Britain, and
Sweden) a regression is done excluding these countries from the sample. An overview of the summary statistics for this subsample can be seen in table A4 and an overview of the results can be seen in table A8. For the subsample significant results are found for the interaction variable of the change in dividend and PDI and IND. Also, a significant result is found for the interaction variable of masculinity with the change in liquid holdings. Overall the evidence found in the robustness checks do not fully support the findings in the results section.
Conclusion
Within this European cross-country study, investors’ culturally determined perceptive on cash and dividend policy is examined. Firstly, for each cultural dimension the different positions of managers and shareholders for dividend and cash policy is determined based on previous literature. Then, using an augmented version of Fama and French (1998) valuation regression model, to determine how they perceive increases or decreases in cash and dividend in countries with high or low scores on different cultural traits. Evidence is found that in societies with high scores in individualism and indulgence an increase in dividends is seen as value enhancing. This means that under influence of the same culture they still have different preferences, investors do not share managers opinion on the choices for dividend policy in these societies and would prefer higher dividend payments. The result for dividend in individualism is in line with the reasoning based on Orlova, Roa, and Kang (2017) stating that investors in individualistic societies will perceive an increase in dividend as value enhancing due to managements’ tendencies for overconfidence and excessive risk taking. The influences of indulgence are far less examined within the literature. This research finds that investors from indulgent countries perceive an increase in dividend as value enhancing. This means that their culturally determined perception does not align with the dividend policy management has set. These findings mean that indulgence is relevance in an investors perception of decisions-making and imply, at least partially, the relevance of
Orlova, Roa, and Kang (2017) use a measure of excess cash. Dittmar and Mahrt-Smith (2007) state that using total liquid holdings might create endogeneity problems because the level cash could be affected by a firms’ investment opportunity set. The choice for total cash holdings could therefore largely contribute to the
insignificant results of the influences of culture on liquid holdings. Secondly, this study focuses only on the EU, while Orlova, Roa, and Kang (2017) focus on firms around the globe. In Hofstede’s research culture is
determined as time invariant and constant. However, in a rapidly changing economic environment these “old” values might no longer be applicable. More so in Europe where countries have converged under the European Union the change in cultural scores might have been more rapid. Also, due to the single market within the EU culture might not be definable by country borders anymore.
The inconsistency of the results with the robustness checks might be caused by a variety of limitations of this study. For example, the assumption that the investor base shares the same cultural background with each other and the firms’ country of origin. Part of the dataset consists of large multinational firms who might not be subject to one prevalent national culture and have an investor composition which is distributed across the globe. Furthermore, cross-border investing has become relatively easy and accessible in Europe due to the free market, meaning investors do not have to share the cultural background as implicated by the country of origin of the firm. Also, the most influential group of investors on firm decision-making are the institutional investors. This group of shareholders might not be equally exposed to national cultural dimensions as others. This argument also goes for managers. There is no guarantee that the managers have the same cultural background as investors or come from the same country the firms are based in. Lastly, this paper assumes that each of Hofstede’s
References
Aggarwal, R., Kearney, C., Lucey, B., 2012. Gravity and culture in foreign portfolio investment. Journal of Banking Finance, 36(2), 525-538.
Anderson, C. W., Fedenia, M., Hirschey, M., Skiba, H., 2011. Cultural influences on home bias and international diversification by institutional investors. Journal of Banking Finance, 35(4), 916-934. Bae, S. C., Chang, K., Kang, E., 2012. Culture, corporate governance, and dividend policy: international
evidence. Journal of Financial Research, 35(2), 289-316.
Baltagi, B. (2008). Econometric analysis of panel data. John Wiley & Sons.
Barclay, M. J., Smith, Jr, C. W., Morellec, E., 2006. On the debt capacity of growth options. The Journal of Business, 79(1), 37-60.
Bates, T. W., Kahle, K. M., Stulz, R. M., 2009. Why do US firms hold so much more cash than they used to? The journal of finance, 64(5), 1985-2021.
Bertrand, M., Mullainathan, S., 2003. Enjoying the quiet life? Corporate governance and managerial preferences. Journal of political Economy, 111(5), 1043-1075.
Beugelsdijk, S., Frijns, B., 2010. A cultural explanation of the foreign bias in international asset allocation. Journal of Banking Finance, 34(9), 2121-2131.
Beugelsdijk, S., Kostova, T., and Roth, K., 2016. An overview of Hofstede-inspired country level culture studies in international business since 2006. Journal of International Business Studies, 48(1), 30-47.
Bhattacharya, S., 1979. Imperfect information, dividend policy, and “the bird in the hand” fallacy. Bell journal of economics, 10(1), 259-270.
Black, F., and Scholes, M., 1974. The effects of dividend yield and dividend policy on common stock prices and returns. Journal of financial economics, 1(1), 1-22.
Boyd, B. K., and Solarino, A. M., 2016. Ownership of corporations: A review, synthesis, and research agenda. Journal of Management, 42(5), 1282-1314.
Breuer, W., Rieger, M. O., Soypak, K. C., 2013. Ambiguity aversion and cash holdings. RWTI Aachen. Chang, K., Noorbakhsh, A., 2009. Does national culture affect international corporate cash holdings?
Journal of Multinational Financial Management, 19(5), 323-342.
Chang, C. H., and Lin, S. J., 2015. The effects of national culture and behavioural pitfalls on investors' decision-making: Herding behaviour in international stock markets. International Review of Economics & Finance, 37, 380-392.
Chen, Y., Dou, P. Y., Rhee, S. G., Truong, C., Veeraraghavan, M., 2015. National culture and corporate cash holdings around the world. Journal of Banking & Finance, 50, 1-18.
Chui, A. C., Titman, S., Wei, K. J., 2010. Individualism and momentum around the world. The Journal of Finance, 65(1), 361-392.
DeAngelo, H., DeAngelo, L., 2006. The irrelevance of the MM dividend irrelevance theorem. Journal of financial economics, 79(2), 293-315.
DeAngelo, H., DeAngelo, L., Stulz, R. M., 2006. Dividend policy and the earned/contributed capital mix: a test of the life-cycle theory. Journal of Financial economics, 81(2), 227-254.
Dittmar, A., Mahrt-Smith, J., Servaes, H., 2003. International corporate governance and corporate cash holdings. Journal of Financial and Quantitative analysis, 38(1), 111-133.
Dittmar, A., Mahrt-Smith, J., 2007. Corporate governance and the value of cash holdings. Journal of financial economics, 83(3), 599-634.
Djankov, S., La Porta, R., Lopez-de-Silanes, F., & Shleifer, A. (2008). The law and economics of self-dealing. Journal of financial economics, 88(3), 430-465.
Duchin, R., 2010. Cash holdings and corporate diversification. The Journal of Finance, 65(3), 955-992. Easterbrook, F. H., 1984. Two agency-cost explanations of dividends. The American economic review,
74(4), 650-659.
Fama, E. F. (1990). Stock returns, expected returns, and real activity. The journal of finance, 45(4), 1089-1108.
Fama, E. F., French, K. R., 1998. Taxes, financing decisions, and firm value. The Journal of Finance, 53(3), 819-843.
Fama, E. F., French, K. R., 2001. Disappearing dividends: changing firm characteristics or lower propensity to pay?. Journal of Financial economics, 60(1), 3-43.
Fama, E. F., MacBeth, J. D., 1973. Risk, return, and equilibrium: Empirical tests. Journal of political economy, 81(3), 607-636.
Farooq, O., Amin, A., 2017. National culture, information environment, and sensitivity of investment to stock prices: Evidence from emerging markets. Research in International Business and Finance, 39, 41-46.
Fidrmuc, J. P., Jacob, M., 2010. Culture, agency costs, and dividends. Journal of Comparative Economics, 38(3), 321-339.
Frésard, L., Salva, C., 2010. The value of excess cash and corporate governance: Evidence from US cross-listings. Journal of financial economics, 98(2), 359-384.
Haw, I. M., Ho, S. S., Hu, B., Zhang, X., 2011. The contribution of stock repurchases to the value of the firm and cash holdings around the world. Journal of Corporate Finance, 17(1), 152-166.
Hofstede, G., 1980. Culture and organizations. International Studies of Management Organization, 10(4), 15-41.
Hofstede, G., 2010. Geert Hofstede. National cultural dimensions.
Hofstede, G., Hofstede, G. J., Minkov, M. 2010. Cultures and organizations: software of the mind (Rev. 3rd ed.). McGraw-Hill. New York.
Hofstede, G., 2011. Dimensionalizing cultures: The Hofstede model in context. Online readings in psychology and culture, 2(1), 8.
Hooghiemstra, R., Hermes, N., Emanuels, J. 2015. National culture and internal control disclosures: A cross‐country analysis. Corporate Governance: An International Review, 23(4), 357-377.
House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., Gupta, V., 2004. Culture, leadership, and organizations: The GLOBE study of 62 societies. Sage publications.
Jensen, M. C., Meckling, W. H., 1976. Theory of the firm: Managerial behaviour, agency costs and ownership structure. Journal of financial economics, 3(4), 305-360.
Jensen, M. C., 1986. Agency costs of free cash flow, corporate finance, and takeovers. The American economic review, 76(2), 323-329.
John, K., Williams, J., 1985. Dividends, dilution, and taxes: A signalling equilibrium. the Journal of Finance, 40(4), 1053-1070.
Kalcheva, I., Lins, K. V., 2007. International evidence on cash holdings and expected managerial agency problems. The Review of Financial Studies, 20(4), 1087-1112.
Khambata, D., Liu, W. W., 2005. Cultural dimensions, risk aversion and corporate dividend policy. Journal of Asia-Pacific Business, 6(4), 31-43.
Kirkman, B. L., Lowe, K. B., Gibson, C. B., 2006. A quarter century of culture's consequences: A review of empirical research incorporating Hofstede's cultural values framework. Journal of International Business Studies, 37(3), 285-320.
Kothari, S. P., Shanken, J., 1992. Stock return variation and expected dividends: A time-series and cross-sectional analysis. Journal of Financial Economics, 31(2), 177-210.
Kyröläinen, P., Tan, I., Karjalainen, P., 2013. How creditor rights affect the value of cash: A cross-country study. Journal of Corporate Finance, 22, 278-298.
La Porta, R., Lopez‐de‐Silanes, F., Shleifer, A., Vishny, R., 2002. Investor protection and corporate valuation. The Journal of Finance, 57(3), 1147-1170.
Lee, C. F., Zumwalt, J. K., 1981. Associations between alternative accounting profitability measures and security returns. Journal of Financial and Quantitative Analysis, 16(1), 71-93.
Li, K., Griffin, D., Yue, H., and Zhao, L., 2013. How does culture influence corporate risk-taking?. Journal of Corporate Finance, 23, 1-22.
Miller, M., Modigliani, F., 1961. Dividend policy, growth, and the valuation of shares.
Newman, K. L., Nollen, S. D., 1996. Culture and congruence: The fit between management practices and national culture. Journal of International Business Studies, 27(4), 753-779.
Once, S., Almagtome, A., 2015. The impact of national cultural values on environmental reporting: a comparative study. Economic and Social Development (Book of Proceedings), 5th Eastern European Economic and Social Development, 251.
Opler, T., Pinkowitz, L., Stulz, R., Williamson, R., 1999. The determinants and implications of corporate cash holdings. Journal of financial economics, 52(1), 3-46.
Orlova, S., Rao, R., and Kang, T., 2017. National culture and the valuation of cash holdings. Journal of Business Finance & Accounting, 44(1-2), 236-270.
Peters, R. H., Taylor, L. A., 2017. Intangible capital and the investment-q relation. Journal of Financial Economics, 123(2), 251-272.
Petersen, M. A., 2009. Estimating standard errors in finance panel data sets: Comparing approaches. The Review of Financial Studies, 22(1), 435-480.
Pinkowitz, L., Stulz, R., Williamson, R., 2006. Does the contribution of corporate cash holdings and dividends to firm value depend on governance? A cross‐country analysis. The Journal of Finance, 61(6), 2725-2751.
Rozeff, M. S., 1982. Growth, beta and agency costs as determinants of dividend payout ratios. Journal of financial Research, 5(3), 249-259.
Schneider, S. C., De Meyer, A., 1991. Interpreting and responding to strategic issues: The impact of national culture. Strategic Management Journal, 12(4), 307-320.
Shao, L., Kwok, C. C., Guedhami, O., 2010. National culture and dividend policy. Journal of International Business Studies, 41(8), 1391-1414.
Shleifer, A., and Vishny, R. W., 1986. Large shareholders and corporate control. Journal of political economy, 94(3, Part 1), 461-488.
Tong, Z., 2010. CEO risk incentives and corporate cash holdings. Journal of Business Finance Accounting, 37(9‐10), 1248-1280.
Woolridge, J. R., 1983. Dividend changes and security prices. The Journal of Finance, 38(5), 1607-1615.
Zheng, X., El Ghoul, S., Guedhami, O. and Kwok, C.C.Y., 2012. National culture and corporate debt maturity. Journal of Banking & Finance 36, 468-488.
Appendix
Appendix A: Robustness data and test results Table A1: Globe values
Countries PDI UAI IDV MAS LTO IND
Austria 5,0 5,2 4,3 3,1 4,5 3,7
Belgium N.A. N.A. N.A. N.A. N.A. N.A.
Bulgaria N.A. N.A. N.A. N.A. N.A. N.A.
Czechia 3,6 4,4 3,6 3,8 3,6 4,2
Germany 5,3 5,2 3,8 3,1 4,3 3,2
Denmark 3,9 5,2 4,8 3,9 4,4 4,4
Estonia N.A. N.A. N.A. N.A. N.A. N.A.
Spain 5,5 4,0 3,9 3,0 3,5 3,3 Finland 4,9 5,0 4,6 3,3 4,2 4,0 France 5,3 4,4 3,9 3,6 3,5 3,4 Great -Britain 5,2 4,7 4,3 3,7 4,3 3,7 Greece 5,4 3,4 3,2 3,5 3,4 3,3
Croatia N.A. N.A. N.A. N.A. N.A. N.A.
Hungary 5,6 3,1 3,5 4,1 3,2 3,3
Ireland 5,2 4,3 4,6 3,2 4,0 5,0
Italy 5,4 3,8 3,7 3,2 3,3 3,6
Luxemburg N.A. N.A. N.A. N.A. N.A. N.A.
Malta N.A. N.A. N.A. N.A. N.A. N.A.
Netherlands 4,1 4,7 4,5 3,5 4,6 3,9
Poland 5,1 3,6 4,5 4,0 3,1 3,6
Portugal 5,4 3,9 3,9 3,7 3,7 3,9
Romania N.A. N.A. N.A. N.A. N.A. N.A.
Sweden 4,9 5,3 5,2 3,8 4,4 4,1
Slovenia 5,3 3,8 4,1 4,0 3,6 3,8
Table A2: Hofstede’s cultural dimensions and its GLOBE equivalents
Hofstede (1980, 2001, 2010) Globe (House, 2004)
Power distance Power distance
Uncertainty avoidance Uncertainty avoidance
Individualism Institutional collectivism
In-group collectivism
Masculinity Gender equalitarianism
Long-term Orientation Future orientation
Indulgence Humane orientation
Assertiveness
Performance orientation
Table A3: Summary statistics independent and depend variables from Fama and French (1998)
Variable Mean SD Min Max N
Dependend variable
Market Capitalization 1,021 1,216 0,031 7,486 8250
Independend variables
Earnings before Tax after interest 0,021 0,143 -0,754 0,280 8134
Net Assets 0,864 0,151 0,008 1,000 8250
R&D 0,022 0,058 0,000 0,383 8249
Interest 0,014 0,014 0,000 0,083 8134
Dividend 0,019 0,030 0,000 0,179 8250
Cash and Cash Equivalents 0,135 0,147 0,001 0,757 8250
Table A4: Summary statistics subsample excluding Germany, Great-Britain, France, and Sweden
Variable Mean SD Min Max N
Dependend variable
Market Capitalization 0,812 1,087 0,031 7,486 2945
Independend variables
Earnings before Interest and Tax 0,042 0,110 -0,766 0,329 2945
Tobin's Q* 1,929 5,903 0,026 45,992 2943
Sales 0,870 0,625 0,001 3,644 2933
Book Leverage Ratio 0,183 0,164 0,000 0,703 2923
Dividend 0,017 0,029 0,000 0,179 2945
Cash and Cash Equivalents 0,109 0,121 0,001 0,757 2945
Table A5: results with GLOBE
Change in
Dividend Change in liquid Assets
PDI 2,81 -0,09 (2,07) (,61) IDV 1,68 -0,14 (1,95) (,56) MAS -3,59* -0,14 (1,88) (,56) UAI -3,38* -0,30 (1,93) (,53) LTO 3,05 0,12 (1,92) (,53) IND 2,98 0,16 (1,84) (,5)
