MASTER THESIS
NATIONAL CULTURE AND SHORT-TERM
THINKING
YANNICK DÉJEAN
10000755
SUPERVISOR:
DR. ILIR HAXHI
SECOND READER: PROF. SUZANA RODRIGUES
UNIVERSITEIT VAN AMSTERDAM
FACULTY OF ECONOMICS AND BUSINESS
MSC. BUSINESS STUDIES
INTERNATIONAL MANAGEMENT
ABSTRACT
Short-term thinking has been seen as a major threat to the health of business industries and systems since the 1980’s; however, a relatively limited researched topic. The recent financial crisis has emphasized the role of short-term thinking in the financial sector. While there might be many factors that instigate short-term behaviour, this study focuses on the cultural influences and its drivers. Building on national cultural dimensions, in the current study we explore the effect of national culture on short-termism. We argue that the more individualistic a society, the more short-term its behaviour. On the other hand, when there is more uncertainty avoidance and long-term orientation in a society, there is less short-term thinking. In this study, short-term behaviour is measured by dividend yield and R&D expenses, where a higher dividend yield and lower R&D expenses indicate short-term behaviour.
For a sample of 4132 companies from 15 different countries, we find that the uncertainty avoidance dimension affects negatively the short-term behaviour. This study found evidence that the individualism dimension leads to less short-term behaviour as well, though only when it comes to R&D expenses. Hofstede’s long-term orientation has a positive effect on short-term behaviour however, meaning that the higher the score on this dimension, the more short-term thinking there is in the business system.
This study found empirical evidence that short-term thinking is culture-driven. Even though its effects are limited, some cultures and countries are more susceptible to short-term behaviour than others, simply because of certain traits specific to their culture. On the other hand, criticism on the Anglo-American way of doing business, arguing that it is detrimental as it leans too much on ‘quarterly capitalism’, has been proven wrong. Secondary implications add to the criticism on Hofstede’s cultural dimensions and Hofstede’s long-term orientation dimension in specific, as in this research Hofstede’s long-term orientation dimension actually is a driver of short-term behaviour.
1.! INTRODUCTION+...+4! 2.! LITERATURE+REVIEW+...+7! 2.1! CORPORATE+GOVERNANCE+...+8! 2.2! SHORT9TERMISM,+THE+ROLE+OF+CAPITAL+MARKETS+&+HOSTILE+TAKEOVERS+...+11! 2.3! CULTURE+...+13! 3.! HYPOTHESES+DEVELOPMENT+...+14! 4.! METHOD+...+19! 4.1! DATA+COLLECTION+...+19! 4.2! SAMPLE+DESCRIPTION+...+21! 4.3! DEPENDENT+VARIABLES+...+22! 4.3.1! DIVIDEND+YIELD+...+22! 4.3.2! RESEARCH+&+DEVELOPMENT+EXPENSES+...+23! 4.4! INDEPENDENT+VARIABLES+...+24! 4.5! CONTROL+VARIABLES+...+27! 4.6! DATA+ANALYSIS+...+28! 5.! DATA+ANALYSIS+&+RESULTS+...+29! 5.1! DESCRIPTIVE+STATISTICS+...+29! 5.2! CORRELATIONS+...+30! 5.3! MULTICOLLINEARITY+...+33! 5.4! MODEL+FOR+DIVIDEND+YIELD+...+34! 5.5! MODEL+FOR+R&D+EXPENSES+...+37! 6.! DISCUSSION+...+41! 6.1! FINDINGS+...+41! 6.2! IMPLICATIONS+...+44! 6.3! LIMITATIONS+&+RECOMMENDATIONS+...+46! 7.! CONCLUSION+...+47! 8.! BIBLIOGRAPHY+...+49!
1. INTRODUCTION
Eighty-six per cent. That is the amount of respondents in a McKinsey research among board members and ‘C-suite’ executives that believed that corporate performance would improve when adopting a longer time-horizon. However, another 63% of the same group also felt that, in the last five years (since the start of the 2008 financial crisis), they were pressured more and more by financial markets to come up with short-term financial gains (Barton & Wiseman, 2014). In a previous article, the consultancy McKinsey (via its Managing Director, Dominic Barton) explained that in their eyes “having a long-term perspective is the competitive advantage of many Asian economies and businesses today” (Barton, 2011).
McKinsey attributes the inability of managers to take on a more long-term view, something that managers themselves also believe to be better for their firm’s performance, to the pressures of the financial markets. These financial markets cause firms to take the perspective of ‘quarterly capitalism’, an unsustainable practice that would hurt performance in the long-term (Barton & Wiseman, 2014). One of the main recommendations of McKinsey is that large asset owners would be more active in controlling the firms that they own, and shift away from a short-term focus to a long-term perspective.
However, before the economic crises of 2008 and later, there were some that advocated for a large role for financial markets in corporate governance. In 2002, Cuervo advised continental European countries to stop with creating codes of good governance, and rather rely on the market to provide discipline for their firms (Cuervo, 2002). This plea to move from the traditional continental Germanic model to a more Anglo-Saxon model means that, in Cuervo’s opinion, European countries should let the stock exchange be the ones to control the managers of its firms, much like in countries such as the United Kingdom and the United States of America. This, he argues, is a more effective way to maximize firm value, as codes of good governance are too soft to be a real way to discipline firms.
So on one hand there is a stream of criticism from the business world itself that believes that the pressures from the capital markets is growing so big, that it is pushing firms and their managers into making short-term decisions (Barton & Wiseman, 2014; Dallas, 2011; Porter, 1992). The aforementioned McKinsey report even urges large asset- and shareholders to take a more active stance – a characteristic of the network-based models of corporate governance. On the other hand, there are scholars who argue that controlling firms through the market is more efficient and who urge countries to open up more for the financial markets (Cuervo, 2002; Franks & Mayer, 1997; Meulbroek, Mitchell, Mulherin, Netter, & Poulsen,
The short-term behaviour that is researched in this study is about managerial decisions. It relates to the willingness of firms and their managers to forego expenditures that in the long-term would benefit the firm, but in the short-term lead to negative cash flows. While the firm knows the investment would be beneficial in the long-term, management purposely refrains from making the investment because they prefer to take short-term gains (Stein, 1989). For example, a manager decides not to buy a new high-tech machine that in the long-term would be a good investment, as the purchase would mean a lower profit, and could therefore lead to a lower stock price of the firm in the short-term. This short-termism is also known as managerial myopia (Stein, 1988).
In the field of corporate governance, there are two traditional models. The Anglo-Saxon, market-based model, believes that executives need to be controlled by the market (Moerland, 1995a). The traditional alternative to the Anglo-Saxon model is the German model. This model can generally be identified more as a network-based model of corporate governance (Moerland, 1995a). It is argued that most countries in continental Europe have based their corporate governance model on the German one. Based on the findings of Barton (2014) and others, one could make the simple assumption to say that firms in countries that have an Anglo-American culture, and model of corporate governance, where the market plays a more prominent role, are always focused more on the short-term than firm in countries with a German model.
This direct relation has never been studied. While there has been research done on how managers feel pressures to short-term thinking in the firms for which they work (Demirag, 1998), or on the influence of stock markets and threats of hostile takeovers (Stein, 1988), there has been no cross-country research on how short-termism is related to different national cultures. Previous research on short-termism has been focused on specific parts of a financial system, for example on hostile takeovers (Stein, 1988), the method of raising capital (Bhojraj & Libby, 2005; Stein, 1989) and the role of different types of ownership of a firm (Edmans, 2009). However, there has been no research on how national culture might have an influence on this process.
There are previous studies that research the influence of short-term behaviour in cultures. Geert Hofstede found the ‘Long-Term Orientation’ dimension for his explanation for culture, for example (Hofstede & Minkov, 2010). A similar dimension exists in the GLOBE study, which is a similar cross-national study about culture, where it is named Future Orientation (R. J. House, Hanges, Javidan, Dorfman, & Gupta, 2004). These two studies argue that there are differences in long-term orientation between cultures, and that they affect
societies and businesses as well. Other studies found that national culture has an influence (to a certain degree) on corporate culture (Hofstede, 1985; Kirkman, Lowe, & Gibson, 2006). National cultural traits influence how business systems are set up and how they function. The studies found a direct effect of national culture and business culture as well. In the context of this study this is important, as this study tries to find a direct relationship between short-term thinking on a national level and the business level. However, these studies are all based on surveys and not on empirical data, which shows the need for a study that finds hard evidence for such relations.
The aforementioned critiques of Dominic Barton and Michael Porter claim that the short-term mind-set of the USA is culturally bound. They argue that because of the national culture of the USA, where there is a lot of individualism, risk taking and focus on the present, there is more short-term thinking in American firms (Barton & Wiseman, 2014; Porter, 1992). However, much like the long-term orientation dimension of Hofstede, the propositions of Porter are not based on empirical data. Barton and Wiseman’s research was based on a survey. As short-term thinking in firms nowadays is an important and topical subject, there is a need for an objective, empirical study whether these claims actually hold in the eyes of quantitative data.
Following the line of reasoning of Barton, Porter and Hofstede, this study argues that certain traits that are specific to a country’s national culture might be instigators of short-term behaviour within companies. For example, in a country where there is more focus on the long-term wellbeing of a society as a whole, the firms in these countries will also be more focused on the long-term. The other way around, when a country is more individualistic and more focussed on the term, the firms based in these countries will also show more short-term behaviour.
This research will therefore set out to investigate the following research question: how does culture affect the level of short-term behaviour in countries?
As there has not been a lot of previous literature on the relation between national culture and short-termism, this paper is a theoretical contribution that researches the links between these factors. It sets out to find empirical evidence to investigate claims that short-term thinking in companies is influenced by national culture. Previous research found that national culture does have a significant influence on the behaviour of firms. In this study, the focus will be on specific dimensions of national culture, and how specific individual traits of national culture influence short-term behaviour in firms. Furthermore, this study looks at what
traits have different types and magnitudes of influence on short-term behaviour in companies. This study tries to identify these relationships in an empirical manner. Next, it will either help to prove or disprove earlier, often contradicting studies on short-term behaviour within firms. These include Porter (1992) and Barton (2011), who say that cultural factors, specific to the USA, have created a business system that is based on short-term thinking. But it also includes Cuervo (2002), who argues that the American situation is good, as the researcher believes that a free market situation is imperative for a good business system, and that American culture actually encourages this preferred situation.
To find the answer to the research question, this study will research how Hofstede’s dimensions are related to several factors that are indicators of short-term behaviour. In particular, Hofstede’s Individualism, Uncertainty Avoidance and Long-Term Orientation dimensions will be used. R&D expenses and dividend yield are the indicators of short-term thinking in firms. By putting these variables into a regression, the relation between Hofstede’s dimensions and short-term behaviour can be determined.
This study is a contribution to academic literature as it is the first to identify specific cultural drivers of short-term behaviour in companies and to research how big their influence is. It looks at claims (often not supported with empirical data) that certain cultures, and in specific the American culture, inhibit more short-term behaviour than others. Lastly, the study will investigate whether Hofstede’s controversial long-term orientation is actually measuring short-term behaviour, or that the claims that it is not explaining what it pretends to are supported by empirical data.
This thesis is structured as follows: this introduction is followed by a literature review that critically reviews literature on key concepts that are used in this research. Next, a method section specifies how the research question of this paper is going to be answered and what data is used for this purpose. After the results are presented, the discussion part will go deeper on the implications and limitations of this study.
2. LITERATURE REVIEW
The literature review will critically summarise previous research that is relevant for this particular study. Firstly, the review briefly reviews the field of corporate governance and the different models that exist. Next, the review will focus on the capital market aspect of the different corporate governance models, followed by research on capital market myopia and managerial myopia. Lastly, national culture will be explained and how it can be measured using Hofstede’s dimensions.
2.1 CORPORATE GOVERNANCE
As the concept of corporate governance has many dimensions, there are several different definitions of what corporate governance exactly is (Aguilera & Jackson, 2010). Schleifer & Vishny (1997), for example, say that corporate governance mostly has to do with the ways in which investors or other suppliers of capital can control their investments. One of the major aspects in this is the management of the principal-agent problem, or agency theory. Simply said, agency theory relates to the relationship between the owners of a corporation (i.e. the shareholders) and those that lead the firm on a daily basis (i.e. (top) management of a firm). The owners of the firm want the management of a company to work in their best interest, which is the maximisation of the share price, or firm value. A higher share price leads of course to more value for the owners (Eisenhardt, 1989).
However, solely focussing on the owners and managers in a firm is a too narrow view. There are many other stakeholders that have an interest in the running of a firm. The employees of a firm have certain interests for example, (local) governments as well. Customers might be one of the most important stakeholders. These various stakeholders need to be managed properly (Donaldson & Preston, 1995).
But there is also a legal side to corporate governance. There are certain rules and regulations to which firms and organisations need to comply and these rules give certain duties and rights to stakeholders in a firm!(La!Porta,!Lopez-de-Silanes,!Shleifer,!&!Vishny,! 1997). These authors argue that a country’s corporate governance model is largely influenced by the degree of investor protection and the quality of enforcement of the regulations.
This legal side immediately brings in a role for institutions in corporate governance. Institutions can be defined as “the rules of the game in a society, or more formally, (…) the humanly devised constraints that define human interaction” (North, 1990). These institutions can shape a model of corporate governance, and are therefore very influential. Furthermore, as these institutions do not necessarily change quickly, firms need to adapt to the institutional environment (Williamson, 2000).
To compare corporate governance models with each other, these different aspects of corporate governance need to be aggregated. For their paper on comparative corporate governance, Aguilera & Jackson (2010) broadly define comparative corporate governance as “the study of relationships between parties with a stake in the firm and how their influence on strategic corporate decision making is shaped by institutions in different countries.” (p. 491).
Traditionally, several different models of corporate governance models have been identified. At first, scholars identified two major models. There is the Anglo-Saxon model1 and the continental model2 (Moerland, 1995a; Shleifer & Vishny, 1997). The Anglo-Saxon model is primarily characterised by a high amount of publicly traded firms. The capital markets are well developed and they play a big role in corporate governance models, as the main way for firms to raise money is through equity (Moerland, 1995b). Firms may have many owners, as a lot of individuals hold stock. This also means that there are many parties that only hold a small amount of stock, and there are few, if any, block holders that can exert a lot of influence on management with their voting power (Shleifer & Vishny, 1986). The management of a company is actually controlled by the market in which all of these small shareholders play a role (Denis & McConnell, 2003; Moerland, 1995b). The controlling feature of the capital markets will be explained in detail in the next part of this literature review. Another characteristic of the Anglo-Saxon model is that management typically has no close relationships with any stakeholder, whether they are investors or employees. The main goal of management is that shareholder value is maximised and top management only represent the shareholders of the firm (Aguilera & Jackson, 2010). The CEO of a firm is the most important person in the company, dictating the strategy and having full control over the firm, together with the responsibility (Denis & McConnell, 2003). Normally there is one board that governs the whole organisation, where management and supervision (over management) are combined in one group (Moerland, 1995a). Lastly, as mentioned before, in this model there is little room for close relationships between management and any stakeholders, including the employees of a firm. Employees are not represented in the board, nor do they have any other influence over management in comparison to other models of corporate governance. This complies with the general belief that this system is the most economically efficient, and would therefore bring the most value for the rest of society (Cuervo, 2002).
In the continental model, the relative amount of small and medium enterprises is higher than in Anglo-Saxon countries. The main source of capital of these firms is taking on
1 Also known as the market-oriented, outsider, shareholder (-centered) model.
2 Also known as the network-oriented, insider, stakeholder (-centred), German, Rhineland
debt, as the firms borrow the capital they need from banks. These organisations fund their investments with these debts and with retained earnings (Corbett & Jenkinson, 1997). In this model, firms are less likely to be publicly listed compared to Anglo-Saxon firms, and there are more large blockholders that hold large blocks of stock. These blockholders often include the same banks that have lent money to the firm, because they can control their investments. These large shareholders can exert a lot of influence, as they hold a lot of voting power. It is also common that firms hold large blocks of stock of each other (Moerland, 1995a). This shows that in this model many firms are connected with each other, and that business is done along the lines of these relationships. In general, personal relationships are more important in the continental model. Managers try to keep a good relationship with all stakeholders of the firm, and these stakeholders are sometimes even legally protected and granted a say in the running of the firm (Aguilera & Jackson, 2003; Moerland, 1995b). This can be seen in the composition of the boards of firms for example. Opposite to the Anglo-Saxon firms, in the continental model of corporate governance firms mostly have a dual-board system.3 In this system, there is an executive board, which runs the firms on a daily basis, and an independent supervisory board, which supervises the executive board (Shleifer & Vishny, 1997). The large shareholders often hold a seat in the supervisory board, and banks and employees are also represented. In Germany it is specified by law how many seats in the supervisory board are allocated to employees.
There are however more than these two simplified models. Each country has a specific model of corporate governance that deviates from these rather standardised models (Aguilera & Jackson, 2010). The Netherlands for example, can be seen as a hybrid model with features of both of these models (Moerland, 1995a). The Japanese model has many of the same characteristics of the continental model, however there are of course also differences.
While this paragraph has taken a quick glance of corporate governance models of general, the next paragraph will focus on a specific part of these models. The role of capital markets will be explained in more detail, and also how the markets correct firms and their managers.
3 German law requires that German firms have a dual board system. In a country such as the
2.2 SHORT-TERMISM, THE ROLE OF CAPITAL MARKETS & HOSTILE TAKEOVERS
Research on short-termism and its consequences, managerial myopia, came to prominence with a paper by Stein (1988). In this paper, it was proposed that there are internal and external incentives, the threat of a takeover in particular, and pressures on managers to come up with short-term financial gains. In order to do this, managers often (have to) forgo long-term investment that will pay out more in the future. This consequence is called ‘managerial myopia’.
The capital markets have the power to control executives of a firm, and punish them with (the threat of) a hostile take-over when the firm is underperforming (Jensen & Ruback, 1983). When a company performs well, the stock price will go up. This is because investors believe that present earnings are a good measure for future earnings, and when a company presents results that are above expectations, more investors will try to buy a share of the firm, which, in turn, drives up the share price (Holmstrom & Kaplan, 2001).
However, when a firm is underperforming, the price of a stock will go down. Investors voice their discontent not by influencing management personally, but by selling their shares. This also creates the possibility of a hostile takeover. When a specific party would like to radically change the course of a firm, it can decide to offer the dispersed shareholders to buy their stock, without consulting the firm itself (Jensen, 1988). If they succeed to gain a majority of the shares of a firm, the new owners can remove the board and install one that would better satisfy their demands (Martin & McConnell, 1991).
After the publication of Stein’s 1988 article, researchers argued that its theory was not completely convincing. Two studies found that firms that installed anti-takeover measures eventually spend less on R&D and other long-term investments (Mahoney, Sundaramurthy, & Mahoney, 1997; Meulbroek et al., 1990). However, this research focused on anti-takeover measures, and not whether the whole system of hostile takeovers would be good for business. More recent research actually found support for Stein’s hostile takeover theory, as they find that takeover protection actually leads to less short-term behaviour in firms (Zhao, Chen, Zhang, & Davis, 2012). However, once again, this paper did not thoroughly explain whether the whole system of hostile takeover pressure is actually provoking short-termism in firms.
But there is also a different way of how financial markets can evoke short-term behaviour in firms. Michael Porter raised some problems with the American capital investment system. He argues that inefficiencies and short-term thinking in American firms can mainly be explained by the fact that the American business system has to rely too much
on fluid capital, and that the focus of firms is too much on return on investments. This is caused because American firms have to rely so much on the capital markets to raise money (Porter, 1992).
In the Anglo-Saxon model in general, the capital markets are very important, as they are the main providers of capital to firms (Corbett & Jenkinson, 1997). Rather than borrowing money from banks, firms raise money by publicly offering shares of their firms on stock exchanges. Even after they have initially sold their shares on the market, and cashed the proceeds of these transactions, the stock price remains important. Many corporate takeovers are not done with cash, but are paid in stock. A higher stock price means that corporations could do more takeovers or do higher bids to win from competition for example (Corbett & Jenkinson, 1997).
Yet, as the capital markets in Anglo-Saxon are so well developed, there are many actors in the stock market. Investors buy and sell stock in a much faster pace than in the German model of corporate governance. But because of this, investors flock from company to company to get the highest return, rather than supporting the firm they have invested in and making sure that value is created in the long-term (Porter, 1992).
The above-mentioned pressures by the financial markets on firms, via the threat of hostile takeovers, are a driver of short-termism and managerial myopia according to further research. Bhojraj & Libby for example, found evidence that managers preferred projects that would lead to short-term gains rather than projects that would lead to maximised total cash flow. So managers chose short-term profits over making investments that would lead to higher total profits in the future. As one of the major pressures for this behaviour, the authors identified capital markets as an important source (Bhojraj & Libby, 2005).
Another study that seems to underwrite the differences of perceived pressure of financial markets on managers between Anglo-Saxon model countries and German model countries was a survey among American and Swedish managers. This survey found that American managers felt much more pressure from financial markets to come up with short-term gains than their Swedish counterparts. As the Swedish model of corporate governance is much more like the German model, this proves that this pressure from financial markets is more prominent in Anglo-Saxon models than in German ones (Segelod, 2000).
However, there is also evidence that the capital markets do not necessarily increase short-term thinking. A study found that the pressure from the capital market does not lead to short-term behaviour and managerial myopia (Samuel, 2000). This research also found
evidence that investments from institutional shareholders leads to lower R&D spending in firms, which is one of the factors that explains short-termism.
So, all in all, there are two main reasons how financial markets can lead to short-termism in firms, via the threat of hostile takeovers and because of the short timespan that many investors have. Multiple studies have also shown that managers actually feel pressured by capital markets to come up with short-term profits. The next section will describe how culture might be an influence on short-term thinking.
2.3 CULTURE
This section will explain how culture might affect short-termism. For this research, Hofstede’s cultural dimensions are used to explain differences in national culture. Hofstede’s famous work on culture led to the identification of five different dimensions, including Power Distance (PDI), Individualism (IDV), Uncertainty Avoidance (UAI), Masculinity (MAS) and Long-Term Orientation (LTO) (Hofstede, 1984).
Hofstede was the first one to come with a quantitative description of national culture. He created a massive study on cross-cultural difference, for which he provided surveys to employees at IBM, a multinational company focused on computers and business software. As IBM in the 1980’s was a company with many foreign subsidiaries and with employees from all over the world, Hofstede was able to discern the first four dimensions. Furthermore, he was able to score each country in his survey on these dimensions. In later years Hofstede added a fifth dimension (LTO) and was able to provide scores on all the dimensions for many other countries (Hofstede, 1984; Hofstede & Bond, 1988). PDI is about the (un)equal distribution of power between the members of a culture. MAS is about whether a society is masculine, with more competitiveness and assertiveness, or feminine, where the focus is more on working together (Hofstede, 1984).
IDV relates to the extent to which a society is a collective, or where the individual is more important. In countries with a high score for IDV, the focus is more on the individual. Personal achievements are very important, and people make decisions that would benefit themselves, rather than a larger group. In collectivist societies, loyalty to the group is much more important (Hofstede, 1984).
UAI is the degree of which certain cultures like, and deal with, risk. It shows whether societies feel comfortable with anxiety, uncertainty, and/or change. Societies with low UAI
are more open to change and are more tolerant. On the other hand, societies with high UAI adhere to strict rules and do not like instability (Hofstede, 1984).
The last dimension, LTO, was added later on. This dimension is about to what extent a society is focused on the long-term, and to what extent it has built its society to reach that goal. It started with an article in which Confucian Dynamism was discovered as a new dimension in culture. The foundation of this new dimension lay in the teachings of the old Chinese scholar Confucius. Using surveys, a ranking was drawn across countries for this dimension (Hofstede & Bond, 1988). Soon after however, this dimension was renamed Long-Term Orientation, as this dimension was used more and more with countries that had little to do with Confucius (Hofstede, 1991).
However, Hofstede’s dimensions are controversial and many critiques have been written about his research. A major point of criticism is that Hofstede does not properly look at cultural differences within a single country, as the dimensions are calculated on a national level (Baskerville, 2003). Furthermore, some argue that the dimensions are defined too broadly, and that therefore their significance decreases (Kwok & Tadesse, 2006). Lastly, Hofstede uses only one firm, IBM, in his research (Javidan, House, Dorfman, Hanges, & De Luque, 2006). On the other hand, Hofstede’s dimensions have been widely used in other studies and are a simple way of defining culture, and therefore they are used in this particular research.
An overview of these scores of the countries that are in this research can be found in part 4 of this paper.
The following part of this paper will elaborate on the hypotheses that are at the foundation of this research. These hypotheses have been based on the literature described above.
3. HYPOTHESES DEVELOPMENT
In this part, the hypotheses that are tested to find an answer to the research question are explained. There are three hypotheses, all of which will be explained below. Figure 1 gives a short and clear overview of the relation of the hypotheses.
Figure 1
As mentioned before, there has not been a lot of work on how national culture influences short-term behaviour within firms. Those studies that were done on this topic involved surveys and mostly showed that there actually is short-term behaviour at all in certain firms and that there are differences between countries (Demirag, 1998; Segelod, 2000). While these studies suggested that national culture is a potential cause of these differences, the studies cannot back up these propositions with empirical evidence.
The goal of this study is to find a clear relationship between national culture and short-term behaviour, and to come up with empirical evidence for this. Previous research has shown that national culture does have a significant influence on the operations of an individual company (Hofstede, 1985). The way in which a firm operates is significantly affected by the culture of the country in which it operates. Previous research regarding short-term thinking has claimed that specific parts of a country’s culture do have an effect on firms. These suspicions have not been backed up by empirical evidence though. Furthermore, it is even more interesting to identify what specific traits in particular have an influence on short-term behaviour. By using Hofstede’s national cultural dimensions, this study Therefore, in line with suggestions of Demirag and Segelod, this study attempts to find a relationship between national culture, identified by Hofstede’s cultural dimensions, and short-term behaviour, which is measured by R&D investments and dividend pay-outs. While all the variables used in this study are backed by theory, it must be kept in mind that there has not been a similar empirical study on this topic before and that the variables to measure short-term thinking might be controversial.
LTO IDV UAI Short-term behaviour (+) H1 (-) H2 (-) H3
This study will take Hofstede’s Individualism, Uncertainty Avoidance and Long-Term orientation dimensions as focal point. The two other dimensions, Power Distance and Masculinity are not used in this research, as they are about other factors of a nation’s society than what is needed for this study. As there are no similar studies on this topic it is hard to benchmark these variables against other work. However, as Hofstede’s dimensions have been used in many other studies regarding national culture, they are logical variables in the context of this study. As a matter of fact, previous studies have shown that Hofstede’s dimensions are still the most widely used measures when it comes to cross-national culture research (Kirkman et al., 2006).
First, in individualistic cultures people tend to make decisions that would benefit themselves, and not necessarily the rest of the community (Hofstede, 1984). In the context of short-term thinking this could mean that investors might demand higher dividends from firms, as that would benefit them more (Porter, 1992). It could also mean that within a firm, managers might strive to get short-term gains, as that would benefit their bonus compensation. Greed is also partially related to the IDV dimension (Green, Deschamps, & Paez, 2005). Porter sees an overload of greed as one of the causes of ‘quarterly capitalism’ (Porter, 1992). The greed of shareholders and investors for quick returns on one hand, and the greed by managers to maximise their compensation on the other hand, leads to a focus on short-term profit taking.
Especially as the criticism is directed mainly on the United States, the individualism dimension is important. With a score of 92 for the USA, this country has the highest score for IDV in the world. On the other hand, it can be derived from the literature that countries with a lower score on individualism, such as Germany, are not necessarily focused on personal gains (Hofstede, 1984). In these societies there is much more attention for others, and decisions are not based on individuals and their profits, but on what is best for a society as a whole.
With this in mind, it seems that there should be a relationship between the IDV dimension and short-term behaviour. Firms from countries with higher IDV score will take decisions without caring so much about their consequences for society as a whole. As individualism leads to greed, companies will also be greedier. This would mean that firms prefer quick, large profits over long-term investments that will only pay out in a couple of years. And this is exactly what short-term behaviour is about. Therefore, hypothesis 1 is:
The second hypothesis is based on Hofstede’s uncertainty avoidance dimension. This dimension is about whether a society is risk-averse. Firms and investors in countries with high uncertainty avoidance will want to hedge the risks of their investments in firms, and demand a long-term growth strategy (Hofstede, 1984). A higher Uncertainty Avoidance score means that people dislike not knowing what is in the future.
This would mean that companies from low UAI countries are more engaged in activities that are either more riskier, or of which the consequences on the long-term are unknown. These companies would therefore be more at ease to focus on short-term profits, without really regarding the long-term ramifications of their actions. And this in turn is exactly the criticism of several authors on the American short sightedness (Barton, 2011; Barton & Wiseman, 2014; Porter, 1992).
In countries with high UAI scores there is more need and demand for planning for the long-term. As investors, managers, employees and others are focused more on making sure that the firm still exists in a couple of years, the amount of risk taking is lower and business strategies will be centred around the long-term. A higher score on the UAI dimension would therefore mean less short-term behaviour (Hofstede, 1984).
On the other hand, it must also be noted that uncertainty avoidance can lead to a ‘fear’ of innovation. People from countries with a high score for UAI often like it how things are at present. Change means uncertainty, as the outcome of change is not always known. High uncertainty avoidance could therefore also mean that innovation, and the change that comes with it, is less than in countries with lower uncertainty avoidance (Shane, 1993). This could mean that in countries with a high score for UAI there is more short-term behaviour. However, as there has not been a previous research using this dimension in the context of uncertainty avoidance, and as the first reading regarding this dimension fits the literature more, hypothesis 2 is:
H2: In countries with a higher score for UAI, there is less short-term behaviour.
The last hypothesis is about the long-term orientation (LTO) dimension. This dimension is about how much societies are focusing on the long-term. Firms from countries with more LTO might have plans for a longer period than firms from countries with less LTO. While the validity of this dimension is disputed (Fang, 2003), this dimension seems to be the most relevant to this research.
Hofstede added the LTO dimension years after he came up with the other four dimensions of his national culture framework. This dimension was found after Hofstede researched Asian culture extensively. LTO is related to Confucianism, a philosophy that is very prominent in China, and to an extent in its surroundings. LTO has to do with multiple aspects. Hofstede says that cultures were there is a lot of perseverance, pride, respect for predecessors and focus on tradition, can be seen as high LTO cultures. As mentioned before, China is supposed to be the main country when it comes to LTO. Hofstede believes that in China, the whole mind-set of the culture is centred around focusing on the future, and on future generations to come (Hofstede & Bond, 1988).
In the context of this study, it is argued that in countries with higher scores for LTO, firms are more focused on long-term profits. This means that the companies in these countries have a strategy of directing their resources in such a way that in a couple of years they pay out, sometimes at the expense of short-term profits. As LTO is about is the future, it is argued that firms from high LTO countries are also focused on the future.
It must be noted that the LTO dimension is rather controversial. The main critique about this dimension is that its definition is too vague, and that its name is misleading (Fang, 2003). There have not been a lot of studies taking this dimension central, and none have truly tested the LTO dimension in the context of long-term and short-term thinking. On the other hand, theoretically, this dimension should be the best measure of short-termism in the context of national culture. If it really explains what Hofstede claims, the LTO dimension is very relevant to this study. Therefore hypothesis 3 is:
H3: In countries with a higher score for LTO, there is less short-term behaviour.
In these hypotheses, the short-term is measured by looking at dividend yield and R&D expenses. In the next part, the method of this research is described.
4. METHOD
This part explains the methods of this research. Firstly, the data collection procedure is explained, followed with a description of the sample. Next, the dependent, independent and control variables are detailed.
4.1 DATA COLLECTION
As data was needed about R&D expenses and dividend yield, it was decided to use information of publicly listed firms. For listed firms, this data should be available and reasonably accessible. The data for this research was collected using the COMPUSTAT database. By using either the ticker symbol of a firm (for American and Canadian firms) or the International Security Identification Number (European firms) the necessary data could be extrapolated from the COMPUSTAT database. At first information on European firms was obtained through the AMADEUS database, operated by Bureau van Dijk. However, after crosschecks with the COMPUSTAT database, the latter was used for all firms. While the AMADEUS database is tailored for European firms, the COMPUSTAT database had more complete information about the requested factors for this research. Furthermore, by using a single database there was less need to manually transfer of transform data, leading to less possible mistakes in the database itself.
Lists of listed firms were obtained on the website of stock exchanges, and used as input in COMPUSTAT to obtain data. Therefore, all firms that are used in this research are publicly quoted in 2014. Table 1 provides an overview of the stock exchanges that were used in this research. While most stock exchanges only list firms, the London Stock Exchange and the New York Stock Exchange (NYSE) in particular also have listed trade funds. As funds are not companies that produce goods or services, it was decided to filter them out. This was done checking the name of the fund, and the description of the security that Standard & Poor’s, the owner of the COMPUSTAT database, had provided.
Table 1
Stock exchange Location
Wiener Börse Vienna, Austria
Euronext Brussels Brussels, Belgium
SIX Swiss Exchange Zürich, Switzerland
Deutsche Börse Frankfurt, Germany
OMX Copenhagen Copenhagen, Denmark
OMX Helsinki Helsinki, Finland
Euronext Paris Paris, France
London Stock Exchange London, Great Britain
Borsa Italiana Milan, Italy
Irish Stock Exchange Dublin, Ireland
Euronext Amsterdam Amsterdam, The Netherlands
Euronext Lisbon Lisbon, Portugal
OMX Stockholm Stockholm, Sweden
New York Stock Exchange New York, USA
NASDAQ New York, USA
Furthermore, there are instances when companies are listed on multiple stock exchanges. Royal Dutch Shell NV is listed in Amsterdam, while Royal Dutch Shell plc. is listed in London for example. These double listed firms were also filtered out, by cross checking names and ISIN codes. As the American stock exchanges work with ticker symbols, and not so much with ISIN codes, it was decided to clear all non-American and non-Canadian firms. Firms registered with an ISIN belonging to Isle of Man, Jersey or Guernsey are seen as British firms. In the end, all companies are from the countries listed in table 1, plus Canada. Firms from other countries are left out.
The year 2012 was used as the benchmark for looking at the variables used in this research. This is because 2012 was the sole year for which information for all firms was complete. As data in 2012 was used for firms that are quoted in 2014, firms that exist today but were not traded in 2012 are excluded from analysis. All currencies have been converted to US dollar, using the exchange rates of 31 December 2012, as the vast majority of end dates of
4.2 SAMPLE DESCRIPTION
The sample consists of in total 4132 firms, from 15 different countries. Table 2 shows the distribution of the countries of origin. These countries were chosen as they represent multiple types of corporate governance models, and because of availability of data on firms from these countries. While the United States accounts for over 35% of the sample, the US portion of the sample is very varied (in terms of size and of different industries), and there are enough firms from other countries to compensate for the effect of the size of America’s part of the sample. Table 3 shows how the firms are divided per industry. The industries have been identified in COMPUSTAT by their Standard Industrial Classification (SIC) code. While there are different levels of specificity in this system, it was decided to stick with the top domain, leading to 11 different categories. This was done, as the industry level is not the focal point of this paper.
Table 2
Frequency Percent Cumulative Percent
Austria 41 1,0 1,0 Belgium 67 1,6 2,6 Switzerland 142 3,4 6,0 Germany 333 8,1 14,1 Denmark 91 2,2 16,3 Finland 100 2,4 18,7 France 370 9,0 27,7 Great Britain 935 22,6 50,3 Ireland 38 ,9 51,2 Italy 159 3,8 55,0 Netherlands 65 1,6 56,6 Portugal 28 ,7 57,3 Sweden 188 4,5 61,8 Canada 112 2,7 64,5 United States 1463 35,4 100,0 Total 4132 100,0
Table 3
Frequency Percent Cumulative Percent Agriculture, Forestry & Fishing 19 ,5 ,5
Mining 247 6,0 6,5
Construction 70 1,7 8,2
Manufacturing 1710 41,4 49,6
Transportation,Communications,
Electric, Gas & Sanitary service 351 8,5 58,1
Wholesale Trade 132 3,2 61,3
Retail Trade 232 5,6 66,9
Finance, Insurance & Real Estate 589 14,2 81,1
Services 767 18,6 99,7
Other 14 ,3 100,0
Total 4132 100,0
4.3 DEPENDENT VARIABLES
The variables that are used in this research to measure short-term thinking are dividend yield and R&D expenses. Both variables have to do with how companies use their financial assets for the future. Firms can freely decide how high they set their dividend yield, and how much they spend on R&D, so they are both independent measures of short-term behaviour.
4.3.1 DIVIDEND YIELD
Dividends are direct payments of firms to its shareholders. These payments are paid out of the profits of a firm. A company can decide to return all, or a part, of its profits to its shareholders by paying out dividends. However, paying out dividends also means that there is less money available to make investments. As dividends are direct payments that reduce the availability of potential future investments, it is a payment based on short-term profit taking. The
shareholders of a firm get a payment now, while forgoing future growth of the stock price through investments of the company. Therefore, it is a good measure of short-term behaviour.
Previous studies have shown that firms that pay fewer dividends spend more on R&D budgets and other investments (Bah & Dumontier, 2001). The researchers found a direct relation between higher dividends and lower R&D expenses, affirming the suspicions that dividend pay outs are an example of short-term profit taking by investors. This, together with the large availability of data on dividends, makes this a good variable for this study.
However, taking total dividends is not a perfect measure. Large firms may pay out more dividends in total, however they might be lower relative to the size of a company. Therefore a ratio is used to create a variable that makes it possible to make a better comparison between large and small firms. Dividend yield is used as a measure of how much dividend a firm pays to its shareholders. It is a ratio that calculates how much dividend is paid out relative to the share price.
A high dividend yield means that firms return a relatively large sum of the profits to its shareholders. On the other hand, a low dividend yield means that a company does not pay out a lot of its profits directly to its investors, but uses the money for other purposes.
It must be noted that companies do not necessarily invest all of the profits that they do not pay out as dividends. Firms can also decide to pay back loans, or hoard the money. Therefore, this variable only partly explains short-term behaviour, and other variables are needed. The next variable to explain short-term thinking is R&D expenses.
4.3.2 RESEARCH & DEVELOPMENT EXPENSES
Research and Development expenses are also an important factor. Where dividends are a measure of short-term behaviour, R&D expenses are indicators of a long-term view. Research & Development investments are often investments of considerable amounts of money, without a clear outcome. The projects in which firms put their money can be hugely profitable to the company, or they can fail miserably and come up with nothing. As research costs time, these investments often take months or years to return with good value. However, in competitive markets these risky investments are necessary for any firm. Because while there is a risk that they do not necessarily pay out as predicted, R&D investments are important as they prepare the firm for the future.
Innovation in competitive markets in very important, as it creates competitive advantages for the firm, or destructs those of its competitors. It also protects a firm from new,
more innovative entrants and even creates new markets or niches for its own. While innovation is not limited to R&D expenses alone, they are very important for the process. Previous research has shown that in the long-term, R&D investments create above normal returns (Eberhart, Maxwell, & Siddique, 2004). Firms that relatively spend more than their competitors were found to be able to be more profitable than their peers, and to be profitable over a longer period of time. The Eberhart et al. study underlines the importance of R&D investments in the innovation process, and therefore also their importance in the long-term orientation of a company.
As larger firms will spend more on R&D in absolute terms than smaller companies, firm size will be controlled for. The same holds for industry. A high-tech industry company will spend more on R&D than a construction company for example. A more detailed description of the control variables will follow soon. As in this sample there was a large variance between the amounts of R&D expenses (ranging from USD 0 to more than USD 11 billion), it was decided to transform the data by taking the root square of the R&D budgets. Otherwise many of the largest firms had to be treated as an outlier. By taking the root square of this variable, the sample could be kept at a large number without damaging the integrity of the statistics.
4.4 INDEPENDENT VARIABLES
This paper focuses on how national culture influences short-term behaviour in firms. Therefore, this paper will make use of Hofstede’s cultural dimension scores (Hofstede, 1984). Geert Hofstede was the first to come up with a quantitative description of national culture where he was able to identify five different cultural dimensions. These dimensions are Power Distance (PDI), Masculinity (MAS), Individualism (IDV), Uncertainty Avoidance (UAI) and Long-Term Orientation (LTO).
Hofstede found these dimensions after an extensive study using employees of IBM. In the 1980’s, IBM had many foreign subsidiaries and a workforce existing of many different nationalities. By asking the very diverse population of IBM to fill in surveys about their culture, Hofstede succeeded in creating a worldwide study on cultural differences. He then proceeded to calculate scores for separate countries on each dimension, leading to the first global dataset on national culture (Hofstede, 1984).
For this study, the IDV, UAI and LTO dimensions are particularly important, as these are used as independent variables. The country scores for these dimensions have been calculated
for each country, and Table 4 shows the scores for the countries that are used in this paper. The rest of this part will explain the dimensions that are used in this study, the critique on Hofstede’s research and the alternative to Hofstede’s research, the GLOBE project.
The individualism dimension will be used, as this dimension shows to what degree a society is focused on the individual. The higher the score for this dimension, the more a society will go for own gains, and not for the common good. As long-term investments often benefit the collective more than some individuals, societies with high scores for IDV will probably have more short-term behaviour as well. Investors and shareholders in high IDV countries will want to maximise their own profits with less regard for other stakeholders. Managers would like to maximise their pay. As the compensation of many managers is based on performance-based pay, they would like to maximise a firm’s profits a soon as possible as well.
On the other hand, the uncertainty avoidance dimension has to do with the amount of risk a society is willing to take, and to what extent they think about this risk. Societies with high scores for UAI will therefore plan ahead more, and feel the need of long-term thinking more than societies with low scores. Therefore, in these societies there will be more focus on the long-term, by planning ahead and by making sure companies have enough assets to remain profitable in the future.
The last independent variable is the long-term orientation dimension. As mentioned before, this dimension is based on Confucianism. Hofstede argues that Confucianism is a philosophy that has everything to do with the past and the future. Tradition, pride and perseverance are quite important in this dimension, as they should make a society focus on the long-term wellbeing of their families and societies (Hofstede & Bond, 1988; Hofstede & Minkov, 2010).
There has been some critique on this dimension. Some say it is too vague and that it does not measure what it claims to do, namely long-term orientation. Especially as this dimension is based on a certain philosophy that is dominant in certain parts of Asia, this dimension is not independent enough to compare different cultures and nations from all over the globe (Fang, 2003).
This is not the only critique on Hofstede’s dimensions. From the start, Hofstede’s cultural dimensions have not been without controversy. For one, Hofstede treats all countries and societies as homogenous. In his scores, local populations and intra-national differences are not calculated. Furthermore, the fact that Hofstede only used information gathered at one single multinational company, IBM, is debateable. It could very well be that his data is
influenced by IBM’s own organizational culture for example. Lastly the critique is that Hofstede’s dimensions are often too broadly defined, and that some of the dimensions have overlapping characteristics (Javidan et al., 2006).
An alternative to Hofstede’s cultural dimensions could be GLOBE project (R. House, Javidan, Hanges, & Dorfman, 2002), which is a more recent study of global cross-cultural differences. However, it was decided to use Hofstede’s cultural dimensions for this study because of the following reasons. This focal point of this study is on a country level. As Hofstede’s work, despite its shortcomings, is also based on the national level, it is very useful to use his cultural dimensions model. As a matter of fact, Hofstede’s model is still the most widely used model when it comes to cross-cultural research. Most of the previous studies on culture have used his theories (Haxhi & Van Ees, 2010). Therefore, it seems best to use Hofstede’s dimensions for this study as well.
Table 4
Country IDV UAI LTO
Austria 55 70 31 Belgium 75 94 38 Canada 80 48 23 Denmark 74 23 46 Finland 63 59 41 France 71 86 39 Germany 67 65 66 Great Britain 89 35 25 Ireland 70 35 43 Italy 76 75 34 Netherlands 80 53 44 Portugal 27 104 30 Sweden 71 29 33 Switzerland 68 58 40 United States 91 46 29
4.5 CONTROL VARIABLES
As there are a lot of different types of firms in this research, it is important to control for external factors. The dataset used for this research contains small and large firms, from all different types of industries. As in one industry there are different standards for dividend yield and R&D expenses than in the other, there is a need to control for this. The control variables for this research are the industry in which a firm operates, and the total revenue of a firm.
The industry in which a firm operates also influences the amount of R&D expenses and a firm’s dividend yield. High-tech companies will probably have more R&D investments than an agricultural firm, for example. In the high-tech sector there is much more innovation needed for a company to survive and to excel. The agricultural sector is much more commodity-like, where there is little to no innovation. Another example is the financial sector. While there is a lot of money involved in this sector, financial service firms have very few R&D expenses, as their investments are less in tangible materials and more in intangible ideas from their staff. So there is a need to control for the industry in which the firm operates. The companies in the dataset have been identified by their Standard Industry Code, and are distributed among 10 different industries. Table 3 (on page 20) provides the different industry sectors. By creating dummy variables, where a 1 means that the firm is in the concerned industry, and a 0 means it is not, all the firms have been identified.
The second control variable is total revenue. Total revenue is used to control for the size of a firm. It would be logical that a large firm with more revenue is involved with higher absolute dividend and R&D expenses than a firm with less revenue. By controlling for firm size, it is possible to look at the results in a relative context, where small and large firms can be compared. There are multiple ways to control for firm size. Besides total revenue, one could use the number of employees, or the amount of units sold for example. However, it was decided to use total revenue as main measure. This is because the number of employees can fluctuate (sometimes greatly) in a fiscal year, making it difficult to pick a benchmark number. A measure such as amount of units sold is also not ideal, as there is a lot of difference between products. A multinational company such as the Dutch ASML, which sells high-tech computer chip fabrication machines, has a revenue of several billion euros, while it only sells around 100 of these machines per year, for example. Therefore total revenue is the best variable to control for size in this study.
4.6 DATA ANALYSIS
Regression analysis is the main tool that is used to test the hypotheses of this research. Using the statistical software package SPSS, a multiple regression analysis was done. Two models have been constructed, one taking dividend yield as dependent variable, and one taking R&D expenses as dependent variable. The models are hierarchical, and variables go into the model using the (forced) entry mode in SPSS. Table 5 explains how the variables are entered in these hierarchical models, for both the regression for Dividend Yield and R&D expenses.
Table 5
Model Control variable Hofstede dimensions
Industry Total
revenue UAI IDV LTO
Model 1 X X
Model 2 X X X
Model 3 X X X
Model 4 X X X
Model 5 X X X X X
For both regressions, at first the control variables will be entered. As explained in the previous part, there are 11 different industries to which all the firms belong. Total revenue is used as a measure to control for the size of a firm. After the control variables have been entered, Hofstede’s UAI, IDV and LTO dimensions will be put into the regression. Again, the same procedure is done for both dividend yield and R&D expenses. They are the dependent variable for their respective regression models. As the industries have been entered as a dummy variable, SPSS has automatically used Manufacturing as the constant variable, and will therefore not be shown in the regression output.
This concludes the methods section of this thesis. In the next part the results of the data analysis of the study will be presented.
5. DATA ANALYSIS & RESULTS
This section will present the results. Firstly, the descriptive statistics will be presented, followed by the collinearity statistics. This is followed by the correlations of the variables, and the results of the regressions that were done for this study.
5.1 DESCRIPTIVE STATISTICS
Table 6 presents the descriptive statistics for the variables that are used in this study. The descriptive statistics show that there is a very big fluctuation in total revenue (total revenue is in 100.000 USD). The mean of total revenue is around 3.7 billion USD, with a standard deviation of more than 16 billion USD. This shows that there is quite a lot of diversity between companies. There are some companies that only have a couple of hundred thousand dollars of revenue that are in this database, and there are some multibillion companies as well. The results for Hofstede’s dimensions show that the sample leans toward a moderate score on uncertainty avoidance, a high score on individualism, and a low score on long-term orientation. This is because of the relative bias of firms from Anglo-Saxon countries (Canada, UK and USA). Most of the firms in the database are located in these countries. Interestingly enough, there are differences between the standard deviations of the three different dimensions. Even with the Anglo-Saxon bias, there is a rather large deviation of uncertainty avoidance between the countries in this database. This means that there are some countries with very high, and very low scores on this dimension. For individualism there is also a significant standard deviation. For long-term orientation the standard deviation is rather low though. It shows that Western countries score rather homogenous on this dimension. As the LTO dimension was at first based on Confucianism and on East-Asian countries, it might be that there is not that much difference between the LTO scores for the countries in this dataset.
For dividend yield, the average percentage is 2,99% for all firms. This means that on average, for every dollar worth of share price, a shareholders receives 2,99 cents. The standard deviation is 7,6 percentage points. While there are some exceptions, most companies have a relative low dividend yield. There are many that do not pay out any dividend at all, and there are some that have dividend yields of up to 90%. As it is impossible to have a dividend yield of more than 100%, and because of errors in the database where firms actually had a dividend yield of more than 100%, the firms were filtered, resulting in an N of 4053 for this variable.
With regards to the R&D expenses, it must be noted that in the regression the root square of R&D expenses was used. For the purpose of complete information, the descriptive statistics for the original R&D expenses are also in table 5. For the original number, the mean was 73 million USD, with a standard deviation of more than 516 million USD. This shows that there is a very large difference between the total amounts of R&D paid. Therefore it was decided to use the square root of the R&D expenses, to make the statistics more robust. This leads to a mean score of 3031.2, and a standard deviation of 7904.7.
Table 6 Descriptive Statistics Mean Std. Deviation N Total Revenue 3731,8094551 16634,95737908 4101 Hofstede UAI 52,82 15,876 4132 Hofstede IDV 78,91 11,183 4132 Hofstede LTO 33,59 5,962 4132 Dividend Yield 2,986092 7,587210 4053 R&D expenses in USD 7297820,2 516092921 4132 R&D expenses in USD (root sq.) 3031,2347 7904,7164 4132 5.2 CORRELATIONS
When looking at the correlations between the factors, there are some interesting preliminary findings (table 7). Firstly, regarding Hofstede’s dimensions, it must be noted that there is a high positive correlation between the uncertainty avoidance and the long-term orientation dimensions (r = .511, p < .01). This means that in this sample, countries that do not like uncertainty focus more on the long-term. Furthermore, the individualism dimension is highly negatively correlated with the other dimensions. So, more individualistic countries are oriented on the short-term, and feel more at ease with handling uncertainty.
When it comes to Dividend Yield, it seems that there is a small, positive relationship (r = .044, p < .01) with the LTO dimension. It seems that firms from countries with more long-term orientation generally issue more dividend to their shareholders. However, when taking a look at the UAI dimension, there is a different story. Here there is a small negative
level, which is fairly close. Anyhow, this shows that firms from countries that are more risk averse pay out fewer dividends. The relation between the IDV dimension and dividend yield is not significant.
When taking a look at R&D expenses, for both the UAI and LTO there is a positive, significant relationship (r = .059; r = .048 resp., p < .01 for both). This means that firms from countries that are more risk averse, and firms that come from countries that are more long-term oriented, spend more on R&D investments. For the IDV dimension there is a significant negative relation (r = -.041, p < .01), showing that firms from countries with more individualistic traits spend less on R&D.
Lastly, it is interesting to see that having a higher revenue means that R&D expenses are higher – this is predictable. However, having more revenue also means a high dividend yield (r = .259, p < .01). This means that the largest firms have a higher dividend yield than smaller firms.
32 Table 7 Correlations Country identifier Industry identifier Total Revenue Hofstede UAI Hofstede IDV Hofstede LTO Dividend Yield R&D in USD (root sq.) Country identifier Pearson Correlation 1 Sig. (2-tailed) Industry identifier Pearson Correlation ,109** 1 Sig. (2-tailed) ,000 Total Revenue Pearson Correlation ,015 -,007 1 Sig. (2-tailed) ,173 ,561 Hofstede UAI Pearson Correlation -,371** -,001 -,013 1 Sig. (2-tailed) ,000 ,942 ,311 Hofstede IDV Pearson Correlation ,708** ,073** -,028* -,588** 1 Sig. (2-tailed) ,000 ,000 ,023 ,000 Hofstede LTO Pearson Correlation -,403** -,032 ,029* ,511** -,632** 1 Sig. (2-tailed) ,000 ,109 ,017 ,000 ,000 Dividend Yield Pearson Correlation -,021 -,029 ,259** -,029 -,015 ,044** 1 Sig. (2-tailed) ,086 ,128 ,000 ,061 ,219 ,000 R&D in USD (root sq.) Pearson Correlation -,009 -,041** ,275** ,059** -,041** ,048** -,007 1 Sig. (2-tailed) ,461 ,001 ,000 ,000 ,003 ,000 ,619 **. Correlation is significant at the 0.01 level (2-tailed).
