Cross-country differences in social and traditional entrepreneurship: The role of national culture

(1)

MSc International Business and Management

Thesis

Cross-country differences in social and traditional

entrepreneurship: The role of national culture

University of Groningen

Faculty of Economics and Business

Department of Global Economics and Management

PO Box 800, 9700 AV Groningen (NL)

January 2018

by Thijmen van der Marel

s2207443

Course: Master’s Thesis IB&M

Course Code: EBM719A20.2017-2018.1

Supervisor: dr. M.J. (Mariko) Klasing

(2)

Abstract

In recent years, social entrepreneurship has gained increasing interest, from both practitioners and academics. The social missions of this type of entrepreneurs, such as contributing to enhanced equitable economic growth, make it an interesting topic of research. While most studies have focused on defining the social entrepreneur, this study will focus on what drives social entrepreneurial activity. More precisely, focusing on the six cultural dimensions developed by Hofstede, this study attempts to find the cultural drivers that promote social and traditional entrepreneurship. Regression analyses reveal that there are several cultural dimensions that affect social entrepreneurial activity. The cultural dimensions ‘long-term orientation’ and ‘indulgence’ affect both traditional and social entrepreneurial activity. The ‘individualism’ dimension solely affects social entrepreneurship. The differences between their effects on social compared to traditional entrepreneurial activity are, however, minimal.

Key words: entrepreneurial activity, social entrepreneurship, Global Entrepreneurship Monitor,

(3)

1. Introduction

The practice of conducting “entrepreneurial activity with an embedded social purpose”, is commonly defined as social entrepreneurship (Austin, Stevenson, & Wei-Skillem, 2006). The phenomena of social entrepreneurship gained increasing interest in the last decade, from both practitioners as well as from the academic field, at a global scale (Alvord, Brown, & Letts, 2004; Leadbeater, 1997; Weerawardena & Mort, 2006). One particular event that amplified attention towards social entrepreneurship is the global financial crisis which erupted in 2008 (Jaén, Fernández-Serrano, Santos, & Liñán, 2017). A multitude of international studies found evidence for greater concentration of income in the upper-income class, generating greater economic inequality (Kotz, 2009). Governments responded to it through focusing on policies supporting commercial entrepreneurship by means of anti-crisis packages (OECD, 2009). Furthermore, increasing fiscal austerity of European governments weakened social protection resulting in a social crisis (Karanikolos, et al., 2013). A combination of these circumstances, which are a direct result of the global crisis, created the context wherein social enterprises started to take over control from the State in their actions to solve problems of inequality. Also, social enterprises took on the role of establishing more equitable economic growth models by means of promoting social change (Jaén, Fernández-Serrano, Santos, & Liñán, 2017). To illustrate the magnitude of this type of business, in 2015, more than half of the United Kingdom’s total nascent entrepreneurial activity (4,0 %) (Kelley, Singer, & Herrington, 2016) can be ascribed to the social side of entrepreneurship (2,3 %) (Bosma, Schott, Terjesen, & kew, 2016).

(5)

5

Finding a relationship between entrepreneurial activity and a country’s specific cultural values is in general a complex process (Jaén & liñán, 2013). It is culture that shapes an individual’s programming patterns of behavior and cognitive schemas such that it is suitable with a nation’s cultural context (Glenn, 2004). The actions of an individual person are therefore partially an outcome of culture’s impact. With this knowledge there is a logical link between culture and entrepreneurship. This link can be expected to be even clearer when the social aspect of entrepreneurship comes into play (Jaén, Fernández-Serrano, Santos, & Liñán, 2017). Remarkable, therefore, is the lack of academic literature searching for the link between culture and social entrepreneurship. The goal of this study is to enhance the literature where social entrepreneurship and culture are combined by means of empirical research.

The following research question is therefore the red line throughout this study:

“Does culture promote social entrepreneurial activity and is this effect different for traditional entrepreneurial activity?”

(6)

6

2. Literature Review

This literature review focuses on social entrepreneurship, a particular type of entrepreneurial activity. Before we can define entrepreneurship in a social context we need to define ‘normal’ entrepreneurship which is a difficult task on itself. Next, I will discuss the theories that focus on social entrepreneurship. The distinction between the two concepts is to this date still relatively blurred. Different ideas about the conceptualization of ‘social’ in entrepreneurship will hence be analyzed in order to make this distinction more clear. Next, the dimensions of culture as defined by Hofstede will briefly be elaborated upon. Making the connection between culture and the different forms of entrepreneurship is the final step of this literature review and summarizes the purpose of this research. Here, I will first discuss the literature linking culture to ‘normal’ entrepreneurship before drawing connections with social entrepreneurship. This will conclude with the development of specific research hypotheses regarding the interplay between national culture and traditional- and social entrepreneurship.

2.1 Entrepreneurship and entrepreneurial activity

Whenever the topic of entrepreneurship is discussed in relation to other factors that might in some way affect or be affected by this phenomenon, the discussion usually begins by exploring the concept of entrepreneurship itself. The large number of articles discussing the definition of entrepreneurship reveals the significant challenges the academic community experiences in giving it a universally accepted definition. The expression on itself is derived, as observed by Cunningham (1996, p. 302) from the French word ‘entreprendre’ meaning ‘to take into one’s own hands’ which again originates from the German word ‘unternehmen’. Both of these words bear the same meaning, ‘to undertake’. The first entrepreneurs came from the French military in the early 16th century and worked as explorers. Later on, in the 17th century, the entrepreneurs included paid employees with the task of building military bridges, fortifications and harbors. So from its origins, the entrepreneur is someone who is contracted to perform a dangerous or risky task. The term has been extended by French economists by including people bearing risk or uncertainty to create innovations (Tan, Williams, & Tan, 2005).

(7)

7

(Baumol, 1993). In the 18th century Robert Turgot and Jean – Baptiste Say defined the entrepreneur as being different from the capitalist. The capitalist is assuming uncertainty or risk-, the entrepreneur obtains and organizes the production factors to create value (Bruyat & Julien, 2001). Their work contains the first of the two basic trends and sees the entrepreneur as a person crafting and developing a new firm or business of any kind. Richard Castillon defined entrepreneurship in the 18th century as well and sees the entrepreneur as a person that is assuming the risk and legitimately appropriates any profits. He saw risk-taking behavior as the essence of entrepreneurship (Herron & Robinson, 1993). Schumpeter’s definition of entrepreneurship is close to the one of Castillon, seeing “entrepreneurs as individuals who exploit market opportunity through technical and/or organizational innovation”. Schumpeter and Castillon’s definition of entrepreneurship forms the second of the two basic trends and says that the entrepreneur is an innovator, thus a person in position of relatively exceptional characteristics enabling him to make changes in the economy, disrupting the market equilibrium (Stevenson & Jarillo, 1990). The personal perspective of Schumpeter and Castillon, who view the entrepreneur as an individual possessing certain characteristics, might explain why not solely economic variables but also other explanatory factors like culture have been argued to explain differences in entrepreneurship rates around the world (Hofstede, et al., 2004). Since this research will focus on the cultural determinants of entrepreneurship, the entrepreneurial concept of Schumpeter and Castillon will be the guiding concept throughout this paper when discussing the determinants of entrepreneurial activity. 2.2 Commercial and social entrepreneurship

(8)

8

Austin, Stevenson and Wei-Skillern (2006) found multiple variables which characterize traditional entrepreneurship, setting it apart from the social counterpart. These variables, acting as theoretical propositions, are meant to distinguish the two forms of entrepreneurship. They found that the commercial entrepreneur steps up in case of market imperfections. Cohen and Winn (2007) back this up by arguing that market imperfections offer significant opportunities for the development of innovative business models and the construction of radical technologies. Another variable where social and traditional entrepreneurship differentiate is the level of resource mobilization. Social entrepreneurs have to tap into a different capital market as a result of a non-distributive restriction on surpluses (Austin, Stevenson, & Wei-Skillem, 2006). This implies that any realized profits should be invested in the company. Investing in social entrepreneurship is hence not attractive for capital investors. The differences in resource mobilization also appear with regard to the available labor market. The two forms of entrepreneurship tap into a different pool of employees as a result of different wages- and worker motivation in the nonprofit and for-profit labor market (Leete, 2000). The social entrepreneurial venture often faces difficulties compensating staff and this is not much of a problem for the traditional market ventures. Employees in non-profit organizations more often value nonpecuniary compensation in return for their efforts (Benz, 2005). Another variable where Austin, Stevenson and Wei-skillern (2006) found social- and commercial entrepreneurship to contrast each other is the way of measuring performance. They argue that traditional entrepreneurs simply rely on market share, product quality, customer satisfaction and financial indicators which are relatively tangible quantifiable measures. The social entrepreneur on the other hand faces greater challenges as performance depends on more varied parties like (non)financial stakeholders. Increasing complexity to manage these relationships makes performance measurement more difficult (Kanter & Summers, Doint well while doing good: Dilemmas of performance measurement in nonprofit organizations and the need for a multiple-constituency approach, 1987).

(9)

9

entrepreneurship and economic growth, 1999; Santos, Roomi, & Liñán, 2016). The causality between entrepreneurship and economic development can go into two directions when distinguishing between social and traditional entrepreneurship. Entrepreneurial activity with relatively high growth potential is a driver of economic development which in turn may increase the level of social entrepreneurial activity (Jaén, Fernández-Serrano, Santos, & Liñán, 2017). This is because economic development results in greater coverage of basic needs among the total population which in turn gives room for starting up a social business aimed at helping others. Furthermore, several personal characteristics are shared among both practitioners. Entrepreneurs are generally characterized as pro-active, creative, and risk-taking individuals who are not afraid to take initiative (Lumpkin & Dess, Claryfying the Enterpreneurial Orientation Construct and Linking It To Performance, 1996). On the other hand, a characteristic that is setting the social entrepreneur apart from the commercial one, concerns the motivation to start a business or making it grow, which in the case of the social entrepreneur is altruism or selflessness (Tan, Williams, & Tan, 2005). The social entrepreneur as an altruistic individual has as her main goal to contribute to solving social problems, thereby making a personal sacrifice to benefit others (Griskevicius, Cantú, & Vugt, 2012; Yunus, 2009). Altruism, according to Adam Smith, is complementary to “individualistic dimension” or “self-interest” and implies reciprocity, a concept often forgotten in economic science (Jaén, Fernández-Serrano, Santos, & Liñán, 2017). Even though the social entrepreneur makes personal sacrifices to run his business, funds are needed to keep the business alive. The social venture therefore needs to make a profit in order to keep the business running in the long run, which is another point of common ground with the commercial entrepreneur. Profits are furthermore needed to make investments and to achieve self-sufficiency in order to be independent from philanthropic or public-aid capital (Jaén, Fernández-Serrano, Santos, & Liñán, 2017).

(10)

10

objective. The results of the survey it reveal that social entrepreneurial activity is spread very unevenly across the participating GEM economies. As an example, only 0,3 percent of the South Korean population is currently trying to make a career as social entrepreneur compared to 10,1 percent in Peru (GEM, 2017). The rate of economic development alone cannot account for these differences across the social entrepreneurship rates, making room for other explanatory variables. At this point culture is a relevant factor as it offers relevant explanations for differences in social, economic, scientific and institutional variables (Jaén & liñán, 2013; Hofstede, et al., 2004). 2.3 National culture and entrepreneurship

(11)

11

individualistic attitudes and psychological traits, favoring entrepreneurial behavior, to originate from a culture sharing pro-entrepreneurial values and patterns (Krueger, 2003; Fernández et al., 2009; Santos et al., 2016). Such a culture will lead to an increased rate of individuals conducting some form of entrepreneurial activity (Mueller & Thomas, 2001). These two mechanism thus imply that a specific national culture will influence the intentions of individuals to conduct some form of entrepreneurship. Furthermore, Jaén and Liñán (2013) in their study on entrepreneurial capital found evidence that cultural values affect entrepreneurial intentions and attitudes of individuals. This means that the level and type of entrepreneurial activity within a country is affected by the arranged cultural values of that country (Ma & Todorovic, 2012).

2.4 Hofstede cultural dimension

(12)

12

Table 1. Definition of Cultural Dimensions Based on Hofstede (2011)

Hofstede is cited over 40,000 times in research conducted primarily in the field of international business and the cultural dimensions are considered to form a leading framework which has a great

Cultural dimension Definition and differentiating Traits

Power Distance – extent to which a society deals

with levels of social power/status (and whether this status is perceived as right or wrong).

High – large degree of tolerance for unequal relationship. Power struggle. Those in power have privileges. Wide salary range from bottom to top.

Low – small degree for tolerance for unequal relationships. Power sharing. Short organization pyramids.

Individualism vs. Collectivism – degree to which

people in a society are integrated into groups.

Individualistic – emphasis on individual accomplishment. “I” mentality. Task prevails over relationship.

Collectivistic – emphasis on group

accomplishment. Dependence on organizations and institutions. “We” mentality.

Uncertainty Avoidance – the extent which

members of a society feel uncomfortable or

comfortable in situations that are different from the usual.

High – ambiguity in life is felt as a continuous threat that must be fought. Anxiety. Need for clarity and structure. Staying in job even if disliked.

Low – ambiguity in life is accepted and each day is taken as it comes. Low stress. Changing jobs no problem.

Masculinity – extent to which a society stresses

achievement or nurture.

High (Masculine) – social gender roles are clearly distinct. Men: assertive and competitive. Women: modest, tender, concerned with quality of life. High degree of materialism. Work prevail over family. Admiration for the strong.

Low (Feminine) – social render roles overlap. Work in order to live. Smaller gender wage gap. Quality of life and people are important.

Long-Term vs. Short-Term Orientation – the

extent to which a society fosters virtues related to either future rewards or past and present.

Long-Term – future oriented, emphasis on perseverance and thrift. Adapt to circumstances. Relationship and market position important. Short-Term – focused on past/present, respect for tradition and preservation of social obligations. Emphasis on quick results. Social spending and consumption.

Indulgence vs. Restraint – national levels of

subjective happiness and life control

(13)

13

impact on today’s management practices (Beugelsdijk, Kostova, & Roth, 2017). Despite broad support, the Hofstede framework does not remain undisputed. A multitude of ethnic groups can often be found within a country’s populations and for this reason Nasif, Al-daeaj, Ebbrahimi and Thibobeaux (1991) criticize the Hofstede model to lack heterogeneity. Also, fragmentation over culture across groups gives reason to believe that using national borders as the determining unit of analysis is not sufficient (DiMaggio, 1997). Additionally, the fact that the underlying data have been generated from a single firm has led to criticism regarding the methodology. Finally, Orr and Hauser (2008) argue that the relevance of the cultural dimensions has diminished and that they are outdated. Yet, one should keep in mind that the Hofstede dimensions do not measure culture in terms of absolute numbers. Instead, the numbers are based on the relative distances between cultures and therefor the points of critique mentioned above do not give any implications to the use of the cultural dimensions for this research. Furthermore, research has shown that the original four classifications of national cultures are still valid today even though they have been published several decades ago (Oudenhoven, 2001). Specifically, Beugelsdijk, Maseland and van Hoorn (2015), in their research on Hofstede’s dimensional scores, found that while a nation’s culture tends to change over time, countries tend to move in similar directions, keeping the relative distance between these cultures relatively stable. Therefore, despite the critique, the hypotheses and conclusions regarding the cultural dimensions will be meaningful. 2.5 Hypothesis Building

Earlier research already confirmed that national culture has an effect on entrepreneurship. Shane (1992) in his research on innovation strategies found that entrepreneurial outcomes at both the national and regional level (based on level of economic development) are affected by culture. Other research found evidence for culture affecting intra-country rates of new-firm formation (Davidsson P. , 1995; Davidsson & Wiklund, 1997). There are however many more aspects that are[ affected by national culture and through this influence entrepreneurial activity. The motives, values, and beliefs that characterize a region’s entrepreneurial activity are another example of the widespread reach of what national culture can affect (Scheinberg & MacMillen, 1988; Mueller & Thomas, 2001; Mitchell, Smith, Seawright, & Morse, 2000). New data generated by the GEM provides an opportunity to reassess the effect of national cultural values on both traditional and social entrepreneurship.

(14)

14

schemas corresponding to the nation’s cultural context. The influence of culture on an individual is therefore represented in his or her actions. In this manner, culture is affecting entrepreneurship. Besides culture, it is likely that other factors influence an individual’s motives to start a career as traditional entrepreneur because of its broad concept. Wankhade (2009) theorized that social entrepreneurship is most prevalent in countries characterized by pro-social behavior. The degree to which a society can be considered as being ‘social’ is in the same way affected by that society’s cultural context (Triandis, 1989). This cultural context can create a pro-social society or a society that is less social. The cultural context of a country is therefore likely to be a strong predictor, - positive or negative-, for social entrepreneurial activity. The first hypothesis is therefore:

H1: National culture has a stronger impact on social entrepreneurship compared to traditional entrepreneurial activity.

Power distance

(15)

15

it is difficult to discover and exploit business opportunities since they lack access to resources, information, and skills (Kirzner, 1997). At the same time, though, individuals positioned low in the hierarchy may see entrepreneurship as their only opportunity to climb the hierarchical ladder, leading to a positive relationship between power distance and entrepreneurial activity. Overall, the majority of studies found empirical evidence for a negative relationship (Hayton, George, & Zahra, 2002; Shane, 1992; Dwyer, Mesak, & Hsu, 2005) between power distance and entrepreneurship rather than a positive relationship (Shane, 1993). Therefor it is hypothesized that:

H2a: Power distance is negatively related to traditional entrepreneurship

The effect of power distance may be different when social entrepreneurial activity is considered. Researchers found that a craving for social change and sustainable social services are drivers of social entrepreneurship (Reis & Clohesy, 2001). The low positioned individuals from a country where high power distance prevails suffer more compared to the low positioned individuals in a low power distance society since they are more limited in their resources (Matsumoto, Yoo, & Nakagawa, 2008). It can therefore be assumed that there is a greater desire for social change among the individuals in a country characterized by high power distance, implying a positive relation. There are however several reasons why a negative relationship between Power Distance and social entrepreneurship can be expected. Countries with a lower score on power distance are less prone to hierarchy. Here, people see each other as equal beings and have a greater tendency to cooperate in order to increase the common good (Jaén, Fernández-Serrano, Santos, & Liñán, 2017). The social entrepreneur is more likely to be accepted in this society and hence social entrepreneurial activity will be higher, implying a negative relationship with power distance. Furthermore, Schwartz (2006) found that people located in countries where high power distance prevails take on a passive role with regard to their acceptation of the country’s economic situation and social order. Since this social inequality is perceived as a natural circumstance, the scope for social entrepreneurship is only limited (Datta & Gailey, 2012). Entrepreneurial activity may hence be expected to be limited in countries where high power distance prevails. The following hypothesis is therefore:

(16)

16

Individualism

The individualism versus collectivism dimension has been the topic of research in many studies concerned with entrepreneurship. Several researcher found empirical evidence for a positive relationship between individualism and new firm creation (Shane, 1992; Mueller & Thomas, 2001). An important reason for this relates to the concept of freedom. Countries scoring high on the individualism dimension value freedom more compared to countries scoring low on this dimensions (Hofstede, 1980). According to Sathe (1988) this freedom is important to entrepreneurs as they have to follow personal convictions. Furthermore, Hofstede (1980) found a positive relationship between individualism and economic development and wealth. Other studies in turn found empirical evidence for enhanced entrepreneurial activity in countries where economic development is on the rise (Gilder, 1988). Other research, however, found evidence for a negative relation between individualism and entrepreneurship, implying that collectivist societies are better at fostering entrepreneurial activity (Hunt & Levie, 2003). Besides the positive and negative relations that are found, there is research arguing that neither of these effects exist. Morris, Avila, and Allen (1993) found proof for entrepreneurship being at its highest level under conditions of balanced individualism-collectivism. The different results concerning this cultural dimension indicate that other factors play a role in the relationship between individualism and entrepreneurship. Hence the following hypothesis:

H3a: Individualism is neither positively nor negatively related to traditional entrepreneurship According to Triandis (1995), individualism is commonly seen to be associated with the pursuit of personal goals. Social entrepreneurs often make self-sacrifices in order to help others as captured in the concept of altruism (Tan, Williams, & Tan, 2005). This is the opposite of an act of individualism where the individual is concerned with improving him or herself. Furthermore, individuals from countries characterized by low individualism expect to be looked after by their family, relatives, and others in return for unquestioned loyalty (Hofstede, 2011). The social entrepreneur finds her motivation in helping others, making this type of entrepreneur more similar to the typical individual found in a collectivistic society. Therefor it is hypothesized that:

(17)

17

Masculinity

The masculinity dimension deals with the allocation of social roles between the sexes. McGrath, MacMillan, and Scheinberg (1992) argued that entrepreneurs take on a highly “masculine” orientation, meaning they tend to be materialistic, thus value money and things, and therefor live to work. Entrepreneurs are, more so than non-entrepreneurs, highly motivated to acquire recognition and monetary compensation. This fits the masculine side of the masculinity-femininity dimension, which is characterized by a need for achievement, instrumentality of wealth and recognition (Hayton, George, & Zahra, 2002). Powell et al.’s (2002) findings back this argument as they found that both women and men perceive a good manager to possess predominantly masculine attributes. According to Lee and Peterson (2000) a culture that is masculine in nature engenders a strong entrepreneurial orientation, which will ultimately result in increased entrepreneurial activity. Furthermore, autonomy, self-confidence, perseverance, decisiveness and high energy levels are all characteristics that simultaneously fit the characteristics of a typical entrepreneur as well as the masculine embodiment of Hofstede’s masculinity-femininity dimension (Chaganti, 1986). The arguments above lead to the following hypothesis:

H4a: Masculinity is positively related to traditional entrepreneurship.

(18)

18

understand and appreciate it, thereby increasing harmony. Hirschi and Fischer (2013) relate this to social entrepreneurs as they found that individuals in a more feminine society seek to contribute to enhanced harmony in their society. However, a masculine culture shares some characteristics suggesting increased entrepreneurial activity. Since social entrepreneurship is a specific form of entrepreneurship it may be considered that a masculine culture also increases social entrepreneurial activity. The similarities though, between a social entrepreneur and a feminine society indicate that the relation between masculinity and social entrepreneurship is different compared to traditional entrepreneurship. The following hypothesis is therefore:

H4b: Masculinity is negatively related to social entrepreneurship.

Uncertainty avoidance

This cultural dimension emphasizes to what extent a culture will adapt to changes and how it copes with uncertainty (Hofstede G. , 1980). Research conducted 30 years ago found empirical evidence for a negative relation between this dimension and national rates of innovation (Shane, 1993). Taking on a neo-classical approach would result in concluding that this decreases entrepreneurial activity. Mueller and Thomas (2000) found evidence for a negative relationship between uncertainty avoidance and risk taking behavior. Risk taking in turn is associated with entrepreneurship. This would suggest a negative relationship between uncertainty avoidance and traditional entrepreneurship (Hayton, George, & Zahra, 2002). McGrath et al. (1992) already confirmed this relationship in their paper comparing entrepreneurs with career professionals. Considering a career move towards entrepreneurship implies coping with uncertainty and complexity in the near future (Gibb, 2002). Swierczek & Ha, 2003 argued that many individual characteristics that are commonly considered to be entrepreneurial, such as high achievement motivations and more risk-taking are more prevalent in low uncertainty avoidance cultures. Wennekers et al. (2007) back this up as they find that uncertainty avoidance changed from being positively related towards being negatively related with new business ownership rates from 1972 to 2007. It is therefore hypothesized that:

(19)

19

Taras, Kirkman and Steel (2010) found evidence for a negative relationship between innovation and uncertainty avoidance at the individual level. Just like traditional entrepreneurship, social entrepreneurship is a process of innovative action (Lepoutre, Justo, Terjesen, & Bosma, 2013). Social entrepreneurship would therefore be less supported in a society that does not tolerate uncertainty. Furthermore, social entrepreneurs in possession of a non-for-profit firm face more uncertainty compared to for-profit firms with regard to their relationship with the government and their ability to secure funding (Weerawardena & Mort, 2006). The negative relationship between uncertainty avoidance and social entrepreneurs can thus be expected to be even stronger compared to the relationship with traditional entrepreneurship. The following hypothesis can thus be derived: H5b: Uncertainty avoidance is negatively related to social entrepreneurship.

Long-term orientation

Hofstede’s fifth dimension, long-term orientation, deals with a society’s inclination to prioritize long-range implications. How links with the past are maintained and how current and future challenges are dealt with differs per society (Lumpkin, Brigham, & Moss, 2010). For an entrepreneur it is of paramount importance to understand the risks and tradeoffs that have to be made which are represented in firm’s strategic control. These risks and tradeoffs are often unpredictable and can be very chaotic (Kanter, 1983). In order to understand the outcomes of such strategic control a long-term orientation is required (Zahra, Hayton, & Salvato, 2004). In this vein, a culture scoring high on the long-term orientation dimension may be expected to have higher entrepreneurial activity compared to cultures scoring low on this dimensions. Zahra, Hayton and Salvato (2004), using financial controls as a proxy for short-term orientation, found proof for a negative relation of this with entrepreneurship. Furthermore, Breton-Miller and Miller (2006) argue that reducing risk by means of good stewardship involves setting long term priorities. Also, they argue that the process of innovation leadership is a long-term goal. Both innovation leadership and reducing risk are evocative of entrepreneurial behavior. It is therefore hypothesized that:

H6a: Long-term orientation is positively related to traditional entrepreneurship.

(20)

20

keeping up relationships with the government and other stakeholders. The uncertainty and risk of traditional entrepreneurial activity is set off against profits. The social entrepreneur does not make such a tradeoff. The higher level of uncertainty and risk is not compensated by profits. Although a long-term orientation is beneficial for understanding risk management, the risk of a social entrepreneur is interpreted different compared to the traditional entrepreneur. This is because the right management of risk does not result in certainty regarding the future. For the traditional entrepreneur, the right management of risk results in earnings which give some certainty regarding the future. Taking this into account, social entrepreneurship may be expected to be less attractive in a culture scoring high on the long-term orientation dimension. A short-term orientation on the other hand might be beneficial for social entrepreneurial activity. The long-term orientation dimension ca be described as the way a society is searching for virtue (Hofstede G. , 2011). The way how social entrepreneurs are passionate about reaching their goals combined with their pragmatic view on the market place (Weerawardena & Mort, 2006) is suggestive for rather radical decision making to get quick results. A society with a culture characterized by short-term orientation is more likely to make radical decisions as they are focused on getting quick results. The following is therefore hypothesized:

H6b: long-term orientation is negatively related to social entrepreneurship.

Indulgence

(21)

21

H7a: Indulgence is positively related with traditional entrepreneurship.

Cultures scoring high on the indulgence dimension tend to take control of life while restraint cultures perceive helplessness (Hofstede G. , 2011). The individuals from a culture scoring low on indulgence are therefore to a lesser extent involved in functional activities. Social entrepreneurship is for a great part integration of social enterprises, thereby encouraging the vulnerable groups of a society to find employment. Individuals from a restraint culture are to a lesser degree expected to conduct social entrepreneurship. The hypothesis is therefore:

H7b: Indulgence is positively related with social entrepreneurship.

Nascent and operational social entrepreneurship

According to Shane and Venkataraman (2000), entrepreneurship should be divided into two processes, the discovering and the exploitation of entrepreneurial opportunities. The same distinction is made for social entrepreneurship. For a given country, it is relevant to understand what affects the capacity of its society to recognize or exploit entrepreneurial opportunities (Davidsson & Honig, 2003) because this can ultimately result in economic growth and increased welfare. I tested for the effect of culture on both nascent and operational entrepreneurship. To get a deeper understanding, I have sought for linear and non-linear relationships.

3. Methodology

This section discusses the research methodology that is used to test the above-derived hypotheses. I will first describe the data sources and sample. What follows is a discussion of the operationalization of the variables. Finally, I will describe the empirical approach used to test the hypotheses.

3.1 Data source

(22)

22

(GEM) and from Hofstede’s (2017) website. The GEM project, Initiated in 1998, assembles harmonized data on an annual basis on entrepreneurial activity measured as the process of new business forming (Sternberg & Wennekers, 2005). The GEM is considered a reliable source of information and multiple key international organizations like the World Bank, World Economic Forum, and the United Nations make use of its data. Data on traditional entrepreneurial activity is retrieved from the 2015 GEM report. This report contains data from 60 economies that completed the Adult Population Survey. Data on social entrepreneurship is retrieved from a special topic report published by GEM (Bosma, Schott, Terjesen, & kew, 2016). The data on social entrepreneurial activity are based on interviews conducted among 167.739 adults in 58 economies in 2015. It is the largest and most up to date comparative study on social entrepreneurship in the world.

The data that make up the Independent variables come from Hofstede’s website1. This database provides scores on 6 cultural dimension at the national level. Besides the independent and dependent variables, some control variables are included. These have been assembled from various standard sources. Among them are The World Bank, The Heritage Foundation, and Barro and Lee (2013). More about the control variables and their sources can be found in Section 3.3 below. 3.2 Sample

The final sample is based on a single requirement, availability of data. This means that for the dependent variables both data on traditional and social entrepreneurship must be available. The GEM report on traditional entrepreneurship includes 62 countries. The special GEM report on social entrepreneurship is less extensive compared to the traditional GEM report as there is information available on only 58 countries. The sample size is further decreased due to a lack of data on Hofstede’s cultural dimension. For a country to be included in the sample, data on all six cultural dimensions must be available. Data on the two most recently added cultural dimensions (long-term orientation and indulgence vs. restraint) is not as extensive as data for the original four dimensions. Hence, the sample had to be trimmed to a total of 46 countries. Availability of data on the control variables did not result in any problems with regard to the sample size. As a result, the total sample size in the analysis is N = 46 countries.

(23)

23

All data on the dependent- and control variables represent the year 2015. This is because the most up to date data on social entrepreneurship is from this year. Data on the independent variables, the cultural dimensions, are based on researched conducted before 2015. However, according to McGrath et al. (1992), the scores on the different dimension of culture constitute long-run values that are slow to change and hence stable over longer periods of time.

3.3 Variables

3.3.1 Dependent variables

Data on the dependent variable used in this study comes from GEM reports. For the purpose of this study, which is comparative in nature, the dependent variable is split in 4 sub-variables.

Total Entrepreneurial Activity (TEA): The TEA rate is a summation of three indicators of entrepreneurial activity measured accordingly to the GEM conceptual framework. The GEM has divided entrepreneurial activity into four business processes and three of them (nascent, new business, established business) encompasses what is labeled in this study as TEA. The fourth business process (discontinuation) is not relevant because the purpose of this study is to find if a certain type of entrepreneurship (traditional vs social) can be triggered by culture and not whether it affects the discontinuation of it. An individual is considered a nascent entrepreneur when he or she is currently setting up a business and is the owner or part-owner of this business, and has not yet been able to initiate business paid wages. The new business entrepreneur owns a business for less than 3.5 years and when this threshold is met, the entrepreneur owns an established business. The values for this variable represent the percentage of the population ages 18-64 conducting either nascent-, new business- and established business entrepreneurial activity.

(24)

24

(2016) is quoted: “any kind of activity, organisation or initiative that has a particularly social, environmental or community objective. This might include providing services or training to socially deprived or disabled persons, activities aimed at reducing pollution or food-waste, organising self-help groups for community action, etc.”

Nascent Social Entrepreneurial Activity (NSEA) and Operational Social Entrepreneurial Activity (OSEA): The values for these variables come from the GEM special topic report (2016) and can be interpreted in the same manner as the former two dependent variables. These variables represent social entrepreneurial activity in the start-up phase and in the operational phase. An individual is considered a nascent social entrepreneur when he or she is currently setting up a business and is the owner or part-owner of this business, and has not yet been able to initiate business paid wages.

3.3.2 Independent variables

Information about a nation’s scores on the different cultural dimension is obtained from the work of Hofstede (2010). Hofstede collected the country scores on the first four dimensions by means of a large IBM study. In this study, conducted in 76 countries around the world, he identified how culture influences specific values in the workplace. The cultural dimension that resulted from this research were power distance (PDI), individualism-collectivism (IDV), masculinity-femininity (MAS), and uncertainty avoidance (UAI). Later on, two other dimensions were added (Hofstede & Minkov, 2012). These cultural dimensions, long-term vs. short-term orientation (LTO) and indulgence vs. restraint (IND) complete the work of Hofstede. All cultural dimensions are incorporated in this research. The scores on each of the cultural dimensions range from a minimum of 1 to a maximum of 100. A score of 1 means that a given cultural orientation is not present in a country (i.e. a score of 1 on IDV indicates that this country does not have individualistic values). A score of 100 means that a given cultural orientation is very much present within a country. For the purpose of the regression analyses presented in Section 4, I will rescale the original scores to fall on the range between 0.1 and 10 rather than 1 to 100.

3.3.3 Control variables

(25)

25

(Wennekers, Thurik, & P.Reynolds, 2005; Hartog & Hoogendoorn, 2011; Tang & Koveos, 2004). The gross domestic product (GDP) per capita retrieved from the World Bank2_{is used as measure} for a country’s income. The GDP is measured in international dollars to adjust for a country’s purchasing power. This conversions makes meaningful comparison possible because it takes into account the relative cost of living and inflation rates of the countries. The second control variable is a country’s unemployment rate because countries with a high unemployment rate are more likely to show increased entrepreneurial activity. The data on unemployment are retrieved from the World Bank3. Higher unemployment gives the individuals within a country a greater incentive to explore self-employment opportunities (Evans & Leighton, 1990). The third control variable is a country’s level of economic freedom. Multiple studies found that economic freedom has an effect on a country’s entrepreneurial activity rate (Nyström, 2008; Stocker, 2005; Bjørnskov & Foss, 2010). The data on a nation’s score on economic freedom were retrieved from the Heritage Foundation4 The data represent a combined score of several other measures, such as security of property rights and judicial effectiveness, regulation of business and labor, regulation of credit and the size of the government. Previous research has found economic freedom to be positively related to entrepreneurship (Nyström, 2008). Finally, because people with higher education have been found to be more likely to participate in social entrepreneurship (Bacq, Hartog, Hoogendoorn, & Lepoutre, 2011), education is also controlled for. The data is obtained from the database created by Barro and Lee (2013)5_{. Among the measures in their database is the average years of total} schooling of the population age 15 and over. This measure contains the years spent in primary, secondary, and tertiary school.

3.3.4 Method of analysis

To find an answer to the main question and to verify the hypotheses, multiple models are tested by means of regression analyses using Ordinary Least Squares (OLS). To obtain the results for hypothesis 1 to 7, the effect of the cultural dimensions on both traditional and social entrepreneurship are tested. In both cases I first test for the effect of each cultural dimensions separately and then test for their joint effects in a multiple linear regression:

2_{https://data.worldbank.org/indicator/NY.GDP.PCAP.PP.CD accessed in October 2015}

3_{https://data.worldbank.org/indicator/SL.UEM.TOTL.ZS accessed in October 2015}

4_{http://www.heritage.org/index/download accessed October 2015}

(26)

26

𝑌 = 𝛼 + 𝛽₁∗ 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝐷𝑖𝑚𝑒𝑛𝑠𝑖𝑜𝑛 + 𝜀

𝑌 = 𝛼 + 𝛽1𝑃𝐷𝐼 + 𝛽2𝐼𝐷𝑉 + 𝛽3𝑀𝐴𝑆 + 𝛽4𝑈𝐴𝐼 + 𝛽5𝐿𝑇𝑂 + 𝛽6𝐼𝑁𝐷 + 𝜀

Later on, the control variables are added to both models. Hypotheses 8 is tested in the same manner as hypotheses 1. Besides the linear regression a non-linear quadratic regression is added in order to get a more thorough understanding of the effect of culture on social entrepreneurship.

𝑌 = 𝛼 + 𝛽₁∗ 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝑑𝑖𝑚𝑒𝑛𝑠𝑖𝑜𝑛 + 𝛽₂∗ 𝐶𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝑑𝑖𝑚𝑒𝑛𝑠𝑖𝑜𝑛2_{+ 𝜀}

Before applying the models, the assumptions for linear regression are checked by conducting several tests. First, a series of scatterplots were created to check for trends in the data between each dependent and independent variable. This visual investigation resulted in the deletion of a single observation because an extreme outlier was revealed. The outlier comes from the country Luxembourg and from the control variable GDP. Because this outlier could have a strong influence on the regression analysis this observation is removed from the final sample. Next, issues of multicollinearity were tested. Although some correlations were high and assume problems of multicollinearity (see table 3), the variance inflation factors (VIF’s) give different results. The VIF scores measure whether there are problems of collinearity among the predictors in an ordinary least- squares regression. All VIF scores for the predictor variables remained below 3,0 which indicates that there is a small, but an acceptable amount of variance among the predictor variables (Zuur, Ieno, & Elphick, 2010). Two control variable exceeded a VIF score of 3,0 but remained below 5 which is still acceptable (Ringle, Wende, & Becker, 2015). Furthermore, I checked whether the data suffers from heteroscedasticity. No major problems of heteroscedasticity were found, therefor there is no need to apply robust standard errors.

(27)

27

4. Results and discussion

In this section I will present the results of the regression analyses. Besides presenting the results I will also discuss the findings, elaborating on any expected or unexpected outcomes. This section is divided in several sub-sections. First of all, the sample of this study is elaborated upon using the descriptive statistics. What follows is a brief description of the correlation matrix. Finally, I will present and discuss the outcomes of the regression analyses.

4.1 descriptive statistics

Table 2 presents an overview of the descriptive statistics for the sample of 45 countries included in this study. Two of the dependent variables that are compared with each other, traditional- and social entrepreneurship, show some differences. TEA ranges from 7,9 to 49,2 percent with a mean of 19,74. The scores for the TSEA range from 1,3 to 16 percent with a mean of 6,71. NSEA ranges from 0,2 to 10,1 percent with a mean of 3,03. OSEA ranges from 0,4 to 8,7 percent with a mean of 3,68. NSEA has a higher maximum score but a lower mean compared to OSEA. A plausible explanation for this is that, although some individuals have the desire to become a social entrepreneur, external factor complicate the transition from nascent entrepreneurship towards being a social entrepreneur in the operational phase.

The scores on the cultural dimensions which form the independent variables range between 0,1 and 10. Two out of the six cultural dimension have relatively high minimum scores combined with a substantially higher mean compared to the other dimensions. The high minimum (2,9) and mean (6,62) of UAI shows that most countries in our sample are relatively risk averse, and thus try to avoid ambiguity in life. The high minimum (2,8) and mean (5,59) for PDI shows that most countries in this study relatively easy accept and to some degree expect unequal distribution of power. The other cultural dimension (IDV, MAS, LTO, and IND) have no abnormal scores. This implies that the cultural scores on these dimensions vary more compared to cultural scores of PDI and UAI which are slightly skewed. Greater variance in the cultural scores is beneficial for the purpose of this study since such variation is required to discover meaningful relationships.

(28)

28

Table 2. Descriptive statistics

Variables Observations Mean St. Deviation Min Max

TEA 45 19,88 9,49 7,9 49,2 TSEA 45 6,56 4,23 1,3 16 NSEA 45 2,94 2,34 0,2 10,1 OSEA 45 3,61 2,43 0,4 8,7 PDI 45 5,99 1,90 2,8 10 IDV 45 4,67 2,34 1,3 9,1 MAS 45 4,86 2,03 0,5 10 UAI 45 6,61 2,12 2,9 10 LTO 45 4,68 2,42 0,7 10 IND 45 4,57 2,06 0,4 9,7 GDP* 45 289,59 167,70 11,32 688,83 Unemployment rate 45 8,25 5,26 0,2 25,1 Economic freedom 45 66,60 8,93 51,4 81,5 Education 45 9,93 2,23 4,2 13,4

* GDP is measured in international dollars

4.1 Correlation analyses

(29)

29

(30)

30

Table 3. Correlation matrix

(31)

31 4.3 Regression results and discussion

In the following I present the outcomes of the regression analysis. The outcomes of the regression analysis are meant to answer the hypotheses and thereby the main research question. What follow is a discussion of the generated outcomes. First, I will discuss the models where traditional entrepreneurship is compared with social entrepreneurship. Afterwards, the focus will be on social entrepreneurial activity and how culture affect both nascent and operational social entrepreneurial activity.

4.3.1 Culture and its effect on TEA and TSEA

(32)

32

In table 5, I present the results of the same analysis but now with the control variables included. When comparing the independent cultural dimension effects on both types of entrepreneurial activity it appears that both traditional and social entrepreneurial activity are affected by the same cultural dimensions. TEA is significantly affected by LTO (p< 0,1) and IND (p< 0,1). TSEA is also affected by LTO (p< 0,01) and IND (p< 0,1). To make any inferences regarding what type of entrepreneurial activity is affected more by culture, a comparison of the betas is useful. For TEA, both cultural variables (LTO, β=-1,132 and IND, β=1,314) show betas larger than 1. The betas for TSEA are smaller (LTO, β=-0,837 and IND, β=0,591). Based on the betas one would conclude that culture has a stronger effect on traditional compared to social entrepreneurship. But the fact that the thoughts and behaviors of an individual are the product of all these cultures combined makes a joint comparison of these cultural dimensions necessary. The results of these joint effects are presented in the last two columns of table 5. Inferences on the relationships in these models can be made since the joint model is significant for both TEA (F=2,760 p< 0,05) and TSEA (F=2,114 p<0,1). As we can see from these models, only TSEA is affected by a cultural dimension, namely LTO (p< 0,05). Measuring the effect of culture on entrepreneurship separately thus results in the same cultures affecting both types of entrepreneurship while in the joint analysis of all cultural dimensions I only find a significant effect of culture on social entrepreneurship. In other words, we can argue that culture affects social entrepreneurial activity more strongly than traditional entrepreneurial activity. Hence, hypothesis 1 is accepted. Certain cultural variables can create a society that is more social than others and this is likely to be represented in the degree of social entrepreneurial activity.

(33)

33

great extent of the variance in TSEA. Some of that variance is captured by the other cultural dimension when I add them to the model. Most of the variance, however, is still captured by LTO and the relationship with TSEA remains significant (p<0,05).

(34)

34

(35)

35

such a case it would hence be difficult to find a linear relationship between culture and entrepreneurial activity. Hofstede’s ‘power distance’ dimension may also suffer from this issue. Graeff and Mehlkop (2003) found that many poor countries in Africa are characterized by highly hierarchical structures. Acceptance of such hierarchical structures corresponds to a high score on Hofstede’s ‘power distance’ dimension. These same countries tend to display also high rates of entrepreneurship. The relationship between power distance and entrepreneurial activity is thus likely to be positive in these countries. This is the opposite of what I hypothesized. The differences between necessity and opportunity-driven entrepreneurship are therefore a possible explanation for the non-significant regression outcomes.

My regression results show a significant negative relationship between traditional entrepreneurial activity and ‘long-term orientation’ even though I expected a positive relation. A possible explanation for this negative relation may be the chaotic and unpredictable nature of entrepreneurship (Kanter, 1983). The risk-averse nature of a country that is relatively long-term oriented has a different effect on the perceived risks involved in entrepreneurship than I expected. I argued that a country scoring high on the long-term orientation dimension foresees these issues and is hence better prepared and more willing to conduct entrepreneurship. Is seems, however, that the risks involved in entrepreneurship are a reason for individuals to withdraw from entrepreneurial activity. Lumpkin et al, (2010) found empirical evidence for a negative relationship between LTO and risk-taking behavior. Although a society with a long term orientation is better in assessing risk, it seems that the risk-averse nature withholds its inhabitant from choosing a career as entrepreneur because of the great risk.

(36)

36

(37)

37

Table 4. Regression of Cultural Dimensions on Traditional & Social Entrepreneurial Activity

Dependent variable TEA TSEA TEA TSEA TEA TSEA TEA TSEA TEA TSEA TEA TSEA TEA TSEA

(38)

38

Table 5. Regresion of Cultural Dimensions on Traditional & Social Entrepreneurial Activity including control variables

Dependent variable TEA TSEA TEA TSEA TEA TSEA TEA TSEA TEA TSEA TEA TSEA TEA TSEA

(39)

39

4.3.2 Nascent and operational social entrepreneurship

I divided social entrepreneurship in nascent and operational entrepreneurship to get a deeper understanding of this concept. Both linear and non-linear quadratic relationships between all cultural dimensions and both forms of social entrepreneurial activity are sought. The results of the linear regression can be found in table 6 while the outcomes of the non-linear regressions are in table 7. Testing for linear relationships between the cultural dimensions and NSEA resulted in two significant relations. Long-term orientation (p<- 0,05) and indulgence (p< 0,10) significantly affect NSEA when measured separately. In the joint model, I only find a significant effect of LTO on NSEA (p< 0,10). I find two significant effects between the cultural variables and OSEA, these are for the individualism dimension (P < 0,10) and the long-term orientation dimension (p< 0,01). When testing for the joint effects of culture on OSEA I only find LTO to significantly affect OSEA (P< 0,01). Searching for non-linear effects of culture on social entrepreneurial activity resulted in one significant effect. Specifically, individualism is found to have a significant non-linear relationship with NSEA (p< 0,10). I will elaborate on this effect after discussing the linear relationship.

From the results we can conclude that the cultural dimensions IDV and LTO both affect nascent and operational social entrepreneurial activity. IND, on the other hand, only effects nascent entrepreneurial activity. Taking into account the linear and non-linear effects, nascent social entrepreneurial activity is affected by more cultural variables compared to the operational entrepreneurial activity.

(40)

40

to become a social entrepreneur, in this case, is developed by means of emotions. These emotions could negatively affect the realization of a social enterprise as they often lead to underestimation of the time and work needed to complete a project (Barona, 1998).

The negative linear effect of LTO on both NSEA and OSEA could be ascribed to the differences in accepting risk. The inhabitants in a short-term oriented culture are more risk-taking (Lumpkin, Brigham, & Moss, 2010). It can therefore be argued that the decision of making a career as social entrepreneur is hence taken more easily compared to inhabitants of a more risk averse, long-term oriented culture.

(41)

41

(42)

42

Table 6. Regression of Cultural Dimension on Nascent & Operational Social Entrepreneurial Activity

Dependent variable NSEA OSEA NSEA OSEA NSEA OSEA NSEA OSEA NSEA OSEA NSEA OSEA NSEA OSEA

Cultural variable PDI -0,030 -0,129 -0,153 0,101 [0,258] [0,243] [0,309] [0,269] IDV 0,011 0,355* -0,211 0,213 [0,199] [0,179] [0,252] [0,220] MAS 0,274 0,223 0,299 0,115 [0,175] [0,167] [0,197] [0,172] UAI -0,039 -0,170 -0,061 -0,115 [0,178] [0,166] [0,183] [0,162] LTO -0,367** -0,480*** -0,314* -0,440*** [0,156] [0,137] [0,185] [0,162] IND 0,332* 0,258 0,119 0,030 [0,184] [0,176] [0,211] [0,184] Obs. (N) 45 45 45 45 45 45 45 45 45 45 45 45 45 45 R-squared 0,000 0,173 0,000 0,244 0,057 0,204 0,000 0,189 0,117 0,368 0,075 0,211 0,079 0,347 F-value 0,982 2,846** 0,980 3,835*** 1,528 3,252** 0,990 3,054** 2,161* 6,121*** 1,714 3,353** 1,376 3,342*** Note: Control variables are inlcuded in the regression analysis

(43)

43

Table 7. Non-linear Regression of Cultural Dimension on Nascent & Traditional Social Entrepreneurial Activity

Dependent variable NSEA OSEA NSEA OSEA NSEA OSEA NSEA OSEA NSEA OSEA NSEA OSEA

Cultural variable PDI -0,431 -2,006 [1,406] [1,340] (PDI)2 0,030 0,141 [0,108] [0,099] IDV -1,534* -0,109 [0,850] [0,796] (IDV)2 0,157* 0,047 [0,084] [0,079] MAS -0,047 -0,194 [0,645] [0,612] (MAS)2 0,034 0,045 [0,066] [0,063] UAI -0,887 -1,694 [1,192] [1,091] (UAI)2 0,066 0,118 [0,091] [0,084] LTO -0,491 0,105 [0,627] [0,542] (LTO)2 0,013 -0,057 [0,059] [0,051] IND 0,663 0,367 [0,733] [0,703] (IND)2 -0,034 -0,011 [0,072] [0,069] Obs. 45 45 45 45 45 45 45 45 45 45 45 45 R-squared 0,000 0,195 0,058 0,231 0,039 0,193 0,000 0,209 0,094 0,372 0,056 0,191 F-value 0,812 2,772** 1,45 3,203** 1,294 2,759** 0,901 2,943** 1,765 5,341*** 1,437 2,729** Note: Control variables are inlcuded in the regression analysis

(44)

44

5. Conclusion

I conclude this thesis by revising the main research question and discussing the implications of the findings of the regression analyses. I will furthermore highlight some of the limitations of this research and share some thoughts regarding future research.

The research question that has been central throughout this study was: “Does culture promote social entrepreneurial activity and is this effect different for traditional entrepreneurial activity?”

The results of the regression analyses reveal that several cultural variables indeed promote the level of social entrepreneurial activity. The directions of the relations between culture and social entrepreneurship however do not differ strongly from that for traditional entrepreneurship. The discovery of an effect of cultural values on social entrepreneurship complements the work of Jaén et al. (2017) who also tried to establish a relation between culture and social entrepreneurship. Social entrepreneurship is promoted by cultures that score high on the ‘individualism’, and the ‘indulgence’ dimension and score low on the ‘long-term orientation dimension’. The cultural dimensions that are related with traditional entrepreneurship, ‘long-term orientation’ and ‘indulgence’, follow the same direction. It thus appears that cultures that are relatively individualistic, more short-term focused and place high value on the instant gratifications of basic needs display the highest level of social entrepreneurial activity. The positive relation of indulgence with social entrepreneurship is in line with the findings of Lepoutre et al. (2013) who found that liberal cultures are more conducive to the creation of social enterprises. I also found a similarity with the recent work of Jean et al. (2017). The U-shaped relationship of individualism with social entrepreneurship also appears in their study. They find social entrepreneurship to be high in countries scoring considerably low on the individualism dimension and in countries scoring very high on this dimension. There are no other similarities between both studies. A possible explanation for this is that Jean et al. (2017) used a different data analyses technique, a different sample, and older data on social entrepreneurial activity. More-over, Jean et al (2017) made a distinction in their sample between developing and developed countries.

(45)

45

Understanding the (cultural) drivers of social entrepreneurial activity is therefor of interest for policy-makers. A better understanding of the effect that culture has on the rate of social entrepreneurial activity could result in more advanced programs that promote these cultural values. Such programs could for instance change certain business policies or even educational systems by aiming towards a more enhanced social entrepreneurial culture. This might ultimately lead to a higher rate of social entrepreneurship.

This research is subject to several limitation that must be considered. The first limitation concerns the application of Hofstede’s cultural dimensions which has received critics from the academic field. The question whether the cultural values are still valid nowadays is the biggest concern. McGrath et al. (1992) however state that culture constitutes a long lasting value which is stable over time and hard to manipulate. It is therefore valid to use the cultural dimension in academic research. Another limitation regarding the use of Hofstede’s model concerns its simplified reflection of the national cultural dimensions (Schepers, 2006). Although culture is found to exist among different social groups like a certain generation or region within a country (Kirkman, Lowe, & Gibson, 2006), Hofstede assumes homogeneity within countries. Regional entrepreneurial activity could be substantially different within a country and regional culture might be a driver of this differences. A theoretical framework where national cultural dimensions are adapted to fit intra-country differences is upon today not available. To counter this limitation, the development of a cultural framework that is able to incorporate intra-country cultures could improve the research regarding the drivers of entrepreneurial activity. For a more precise understanding of the cultural drivers of entrepreneurial activity, further research could conduct individual surveys among entrepreneurs from different regions within a country.

Another limitation concerns the limited sample size that I used to conduct this research. A small sample size decreases the statistical power of the research and may result in a reduction of significant results (S. Button, et al., 2013). Limited data on national social entrepreneurial activity is the cause of this small sample. When data is available for more countries the study results could be improved, it is therefore recommended to include more countries in the next round of the GEM social entrepreneurship survey.

(46)

46

McLean, 2006). This blurred distinction could explain why culture seems to affect traditional entrepreneurial activity and social entrepreneurial activity in a similar fashion. GEM’s broad measure of social entrepreneurial activity allows for a bigger sample size. A more strict classification of social entrepreneurial activity is available but data on this this narrow measure is available for only 31 countries (Bosma, Schott, Terjesen, & kew, 2016). This would make the sample size even smaller and the process of finding significant relations more difficult. An extension of the sample size regarding narrow social entrepreneurial activity is therefore recommended. The tendency for an individual to choose a career as social entrepreneur, following GEM’s narrow measure, could give different insight with regard to the cultural drivers. A great opportunity for further research would therefore be to use a more strict measure of social entrepreneurship.

References

Acs, Z., Desai, S., & Hessels, J. (2008, September 5). Entrepreneurship, economic development and institutions. Small Business Economics, pp. 219-234.

Allen, F. (1993). Strategic management and financial markets. Strategic Management Journal, 11-22. Allison, P. (2012, September 10). Stastistical Horizons. Retrieved from When Can You Safely Ignore

Multicollinearity?: https://statisticalhorizons.com/multicollinearity

Alvord, S., Brown, L., & Letts, C. (2004, September 3). Social Entrepreneuship and Societal

Transformation: An Exploratory Study. The Journal Of Applied Behaverioral Science, pp. 260-282. Austin, J., Stevenson, H., & Wei-Skillem, J. (2006, January). Social and Commercial Entrepreneurship:

Same, Different, or Both? Entrepreneurship Theory and Practice, pp. 1-22.

Autio, E. (2007). Global entrepreneurship monitor: 2007 Global report on high growth entrepreneurship. Wellesley, MA/London: Babson College and London Business School.

Bacq, S., Hartog, C., Hoogendoorn, B., & Lepoutre, J. (2011). Social and commercial entrepreneurship:

Exploring individual and organizational characteristics. EIM Business and Policy Research.

Barona, R. (1998). Cognitive machnisms in entrepreneurship: Why and when entrepreneurs think differently than other people. journal of Business Venturing, 275-294.

Barro, R., & Lee, J. (2013). A New Data Set of Educational Attainment in the World, 1950-2010. Journal of

Cross-country differences in social and traditional entrepreneurship: The role of national culture

MSc International Business and Management

Thesis