Multinational Enterprises and their impact on Entrepreneurial activity through Socioeconomic development: A cross-country analysis on the impact of big players

Multinational Enterprises and their impact on

Entrepreneurial activity through Socioeconomic

development: A cross-country analysis on the impact of

big players

Eirik Solvang Pedersen Student Number: 2783797

Supervisor: Dr. Miriam Wilhelm Co-supervisor: Dr. Rian Drogendijk

University of Groningen Faculty of Economics and Business

2

Abstract

This thesis elaborates the mediating effect that Socioeconomic development has on the relationship between the presence of Multinational Enterprises (MNEs) in host countries and Early-stage Entrepreneurial Activity. Early-stage Entrepreneurial Activity is divided into three sub measurements consisting of Necessity-driven early-stage entrepreneurial activity (NEA), Opportunity-driven early-stage entrepreneurial activity (OEA) and Total Early-Stage Entrepreneurial Activity (TEA). The study is a cross-cultural study with sample data on early-stage entrepreneurial activity collected from 31 country cases through the Global Entrepreneurship Monitor’s databases (GEM). A multiple year delta approach will be applied to measure change from the one year (2005) to another (2013). This research tests for both the direct relationship between the various types of entrepreneurial activity as well as the indirect relationship, mediated by socioeconomic development. Results show no significant relationships between any of the variables with the exception of the relationship between socioeconomic development and NEA.

The findings of the research show little support for the hypothesis that was developed. The empirical results do however find support for a negative relation between socioeconomic development and NEA and some close to significant relations that are likely to be relevant.

3

Table of content

2.1.1 Introduction to the field of Entrepreneurship ... 9

2.1.2 Types of entrepreneurial activity ... 10

2.1.3 Stages of entrepreneurial activity ... 11

2.1.4 How entrepreneurial behavior is stimulated. ... 11

2.2. How socioeconomic variables affect entrepreneurial activity. ... 12

2.3 Multinational Enterprises and their position in foreign economies ... 15

2.3.1 MNEs influence on local economy ... 16

2.3.2 The impact of MNEs on entrepreneurship ... 17

2.3.3 How multinationals affect Socioeconomic variables in developed and less developed economies. ... 18

3. Hypothesis development. ... 20

3.1 MNE Presence and the effect on Necessity-Driven and Opportunity-Driven early-stage Entrepreneurial Activity ... 20

3.2 MNE Presence and the direct effect on Total Entrepreneurial Activity ... 21

3.3 MNE Presence and the mediated effect of socioeconomic development on Necessity-Driven – and Opportunity-Driven Entrepreneurial early-stage Activity and the Total early-stage Entrepreneurial Activity ... 21

4. Methods ... 25

4.1. Sample ... 25

4.1.1 Database and sample description ... 25

4.2 Operationalization of the variables ... 28

4.2.1 Entrepreneurial activity ... 28

4.2.2 Necessity-Driven early-stage Entrepreneurial Activity (NEA) ... 28

4.2.3 Opportunity-Driven early-stage Entrepreneurial Activity (OEA) ... 29

4.2.4 Multinational presence ... 29

4.2.5 Socioeconimic development ... 30

4

4.2.7 Control Variables: Region ... 31

4.2.8 Control Variables: OECD Membership ... 31

4.2.9 Analysis. ... 32 5. Assumptions ... 32 5.1 Linearity ... 32 5.2 Sample Size ... 33 5.3 Outliers ... 33 5.4 Correlations ... 34 5.5 Multicolleniarity ... 34 6. Empirical Results ... 35

6.1 Necessity Driven Entrepreneurial Activity ... 35

6.2 Opportunity-driven early-stage Entrepreneurial Activity... 36

6.3 Total Entrepreneurial Activity ... 37

6.4 Sensitivity analysis. ... 38

7. Discussion ... 39

7.1 MNE presence and the lacking effect on socioeconomic development. ... 39

7.2 MNE presence and entrepreneurial activity ... 41

7.3 Socioeconomic development and entrepreneurial activity ... 43

7.4 Limitations ... 44

7.5 Suggestions for further research ... 46

7.6 Management implications ... 48

Chapter 8: Conclusion ... 48

References ... 50

Appendix A: Country list ... 57

Appendix B: ... 59

Appendix C: Boxplots illustrations ... 62

Appendix D: Multi-collinearity analysis summaries ... 65

Appendix E: Sensitivity regression analysis Variant 1 ... 69

Appendix F: Sensitivity regression analysis Variant 2 ... 83

5 Acknowledgements

The process of writing this thesis has been a tough and demanding process that I would not have been able to do without the help of a number of persons.

I would like to express my thankfulness to my supervisor Dr. Miriam Wilhelm for her helpful comments, and support through this period. My sincere thanks also go to my co-supervisor Dr. Rian Drogendijk who provided me helpful guidance and assistance.

6 1. Introduction

Explorers like Christopher Columbus and Vasco da Gama crossed the great seas and connected America and Asia with Europe at the end of the 15th century. These Voyages of Discovery introduced the world to new key & future indispensable routes, bridging Europe to the “new” continents, inaugurating meanwhile a fresh era of intercontinental trade. This era was dominated by a selection of key players, the entitled ‘Chartered Trading Companies’ (Jones, 2000). The first traders of this period were the Portuguese and later the Hudson’s Bay Company (HBC), Dutch East India Company (VOC) and the East India Company (EIC), which all were founded in the beginning of or late 17th century (Jones, 2000). Nevertheless, such trading companies did not just act as transporters. In this manner, they established large networks of bases with trading posts, offices and warehouses along the coast on the way to and on the Asian and American continent (Rijksmuseum, 2015). Between the 16th and 18th century these companies dominated the world’s trade through their monopolistic governmental contracts (Jones, 2000). Through the geographical presence and the intercontinental trade between Europe and the rest of the world, the HBC, VOC and EIC developed into the first modern ‘multinational’ enterprises (Jones, 2000).

During the 18th century, the invention of the steam powered engine (Hills, 1993), a “General-Purpose Technology”, turned to be main power source of the 19th century. It took the world into the first step of industrialization (Rosenberg and Trajtenberg, 2004). The steam engine had the fundamental characteristic that made it independent from geographical dependencies (Rosenberg and Trajtenberg, 2004). These influential individuals were entrepreneurs who initiated great change in how to do business.

7

Fluctuations in demand in supply variables represent, correspondingly, one of the essential sources of opportunities. These can be created by MNEs. MNEs entering new regions create demands in the market that previously did not existed or their planned activities contribute to exceed the previous market demand. As of this, the establishment of MNEs in new markets creates new opportunities that could be taken advantage of by entrepreneurs with an opportunity driven motivation. When MNEs establish, they are likely to be initiating positive changes within the country (Washington and Chapman, 2014), which could lead to an improved socioeconomic situation. Teaching and training of local employees is often a necessity for MNEs to be able to operate in the host country. This training of the local labor is important for host country as it participate to improved socioeconomic development.

This ambition of this study is to elucidate in a coherent and unequivocal fashion how entrepreneurial activty is influenced by MNEs through their impact on the socioeconomic development in a country. There are multiple possibilities for measuring socioeconomic developments in a country. The focal point of our study is supported by three central pylons. Capital and entrepreneurship is argued to be closely related to each other considering the fact that the entrepreneur is dependent on demand in order of having opportunities to exploit (Alvarez and Barney, 2014). The need for capital in terms of venture capitalists and FDI is therefore essential in this study. Likewise, the second mediator that is capable of shaping the development and advancement of countries is illustrated through the labor in the country. As Washington and Chapman discuss, the presence of MNEs leads to increased Labor demand in the host country. The final mediator that will be studied is the influence of Human Capital. Blomström and Kokko (1998) argue how the presence of MNEs lead to increased Human Capital through the learning and copying of processes or gain profits from their operations.

8

Debackere (2003) who explain how FDI could increase the crowding out effect of domestic entrepreneurs short term, but on longer term have a less crucial effect and even be reversing the crowding out effect. The relation between MNEs and Socioeconomic variables have been researched by scholars like LIU et al. (2014) who find evidence for increased Human Capital among entrepreneurs in China, gained through experience at MNEs, being a cause for increased performance in comparison to entrepreneurs without experience from MNEs. Balsvik (2011) provides similar evidence claiming that workers with experience from MNEs contribute to a higher productivity in domestic firms than do workers without MNE experience.

This study provides a cross-country study on the impact of the MNEs, measured in three socioeconomic indicators, on local entrepreneurial activity. This leads to the following research question:

How can socioeconomic development explain the effect of multinational presence on entrepreneurship?

The outline of the paper is the following: While the first part provided a short introduction to the topic of the thesis, the second part contains the literature review, including Entrepreneurship-, Internationalization and the Globalization theory. The third part encloses the development of the Hypothesis and the fourth encompass the Methodology strategy. In part five the assumptions of the thesis are presented. Further on, part sixth contains the empirical results and the seventh consist of the discussion. The finale part will present the conclusion which has been drawn from this research.

2. Literature review

This chapter will provide a foundation about the existing literature related to the areas of research

that this study is built upon. Entrepreneurship is argued to be one of the most important drivers for economic growth

9

Establishments (Market expansions) of MNEs may serve as a vital influencer on the economic development through its impact on socioeconomic variables such as demand after labor, flow of capital and the level of human capital.

2.1. Entrepreneurship

2.1.1 Introduction to the field of Entrepreneurship

The first academic to recognize entrepreneurship as a indispensable key function for economic growth was Cantillon (1775) who described the economic role of the entrepreneur as risk-bearing in the process of buying at certain prices and selling at uncertain prices (Stevenson and Jarillo, 1990). Schumpeter had adopted a more modern view on entrepreneurship in his expression of entrepreneurship as the engine in economic development (1934). Schumpeter’s, and Baumol’s (1968) view on entrepreneurship from 1934 is close to today’s researchers understanding of entrepreneurship as a phenomenon by emphasizing the importance of entrepreneurship for economic growth and spreading of innovation (Ahmad and Saymour, 2008). In this spirit, we can distinguish three variables as crucial aspects of a aforesaid economic growth sphere, namely; the entrepreneur, entrepreneurial activity and entrepreneurship. Ahmad and Saymour (2008) have defined these terms well aligned with the research perspective of this study. By definition, the entrepreneur is “a person (business owner) who seek to generate value, through the creation or

expansion of economic activity, by identifying and exploiting new products, processes and markets” (Ahmad and Saymour, 2008, p. 14). The notion of entrepreneurial activity as “the enterprising human action in pursuit of the generation of value, through the creation or expansion of economic activity, by identifying and exploiting new products, processes or markets” (Ahmad and Saymour, 2008, p. 14). Ahmad and Saymour’s (2008) understanding of

10

and the use of resources and institutions’ (1999 p.47). The most essential in Wennekers and Thurik’s definition in regards to completing the definition of Ahmad and Saymour, is that they also include internal entrepreneurial activity and entrepreneurship occurring within teams.

In the following section the types of entrepreneurial activity will be described.

2.1.2 Types of entrepreneurial activity

A distinguishment is made in the literature between two frames where entrepreneurship occurs, internal and external entrepreneurship. Entrepreneurship within existing organizations, expressed as intrapreneurship, is key for the growth of SMEs as much as for large enterprises (Carrier, 1994; Anotoncic and Hisrich, 2001; Anotoncic and Hisrich, 2003). Amongst other, intrapreneurship could lead to new business ventures (Burgelman, 1983; Stopford and Baden-Fuller, 1994) as well as to new internal “developments of products, services, technologies, administrative techniques, strategies, and competitive postures” (Anotoncic and Hisrich, 2003 p. 498). Entrepreneurship taking place outside existing organizations, business venture entrepreneurship involve starting new business based upon opportunity or necessity driven motivation, also known as the Push-Pull model (Oxfeldt 1943; Johnson and Darnell, 1976). Necessity-driven motivation (Push) is an instinct driven motivation for establishing new businesses, as it is initiated due to lacking possibilities to obtain resources (Singer et al., 2015). Individuals starting their own business, based on necessity driven motivation, tend to have “low education, run smaller firms, expect their firms to grow less, but are likely to stay in the market” according to Poschke (2013 p.1). By the same token, necessity-driven Necessity driven entrepreneurial activity is more noteworthy in efficiency driven economies. Without first and so to some degree in innovations-driven economies, as well, but its considerable and tremendous relevance is most obviously and noticeably contemplated in factor-driven economies (Raynolds et al., 2002; Grilo and Thurik, 2006; Hessels et al., 2008; Amorós and Bosma, 2013).

11

advantage of discovered business opportunities, due to greater potential and not as a reaction of lack of other opportunities (Singer et al., 2015). Entrepreneurs, who base their motivation on discovered opportunities, expect their ventures to grow and to create more vacancies (Acs et al., 2008). These entrepreneurs are people that have started a business with the intention of increasing their earnings to a higher level than what they are offered as compensation as an employee and/or to achieve a higher independency (Singer et al., 2015).

The next section will scrutinize the different stages of entrepreneurial activity.

2.1.3 Stages of entrepreneurial activity

Researchers such as Singer et al. (2015) divide entrepreneurial activity into four phases: Nascent, new business, established business, discontinuation of business, by relating it to the ‘organizational life-cycle approach’ (Singer et al., 2015). Nascent entrepreneurial activity represents the entrepreneurs who actively are “involved with setting up a business they will own or co-own; this business has not paid salaries, wages, or any other payments to the owners for more than three months” (Singer et al., 2015, p.24). Owners-Managers of a New Business are entrepreneurs currently “owning and managing a running business that has paid salaries, wages or any other payments to the owners for more than three months, but no more than 42 months” (Singer et al, 2015, p. 24). Established business entrepreneurs are people owning and managing a business for longer than 42 months while business discontinuation entrepreneurs are people who have “discontinued a business within the last 12 months”.

The following section will elaborate upon how the entrepreneur gets stimulates to perform entrepreneurial activity.

2.1.4 How entrepreneurial behavior is stimulated.

Much research has focused on examining what drives entrepreneurial behavior. Verheul et al. (2001) has categorized the stimulating drivers into two subgroups consisting of demand (influencers on market demand) and supply (factors that increase entrepreneurial behavior).

12

characteristics” to be of special importance in this relation. Hisrich (1990) on the other hand find personal characters such as having leader skills and impact on their culture, to be highly relevant for entrepreneurs. Researchers such as Sexton and Bowman (1986) have recognized the following psychological characteristic traits among entrepreneurs; “1) Tolerant of ambiguous situations, 2) prefer autonomy (autonomy may be described as self-reliance, dominance, and independence), 3) resist conformity, 4) be interpersonally aloof yet socially adroit, 5) enjoy risk-taking, 6) adapt readily to change, and 7) have a low need for support” (pp.129-130). Steward et al. (1999) finds partly the same results, and although they distinguish between two types of business owners in their study and find propensity of risk-taking to generalizable for both and differentiating the entrepreneur from the manager. De Wit and Van Winden (1989) look into family relations and have found support for a relation between entrepreneurs and parents who themselves have been involved in entrepreneurial activity. Another perspective of drivers is related to age with Bosma et al. (2012) arguing how young people are more likely to be having a more fresh way of thinking, be born in the digital world and to have less responsibilities, while older people gain advantages through more experience, contacts and possibly capital. The function of gender and entrepreneurship is discussed by Sexton and Bowman-Upton (1990) who argue that Female entrepreneurs are less likely, compared to their male counterparts, to involve themselves into taking high risk and that they have a lover energy level which is something argued to be very much needed in business formation. Marlow and Patton (2005) continues on this with their results providing evidence for female entrepreneurs having to deal with more barriers due to their gender, when seeking finance. Jamali (2009) express the gender issue more thoroughly through his Lebanese research where he finds support for how this culture strongly influence women’s low participation rate in the workforce.

The next section will more thoroughly elaborate upon how socioeconomic variables are of large importance for entrepreneurial activity.

2.2. How socioeconomic variables affect entrepreneurial activity.

13

14

15

In regards to developing countries, researchers such as Cooper and Dunkelberg (1987) have been looking into education’s influence on entrepreneurship and have found significantly higher levels of education among entrepreneurs than non-entrepreneurs in US. Thompson (1986) found similar results in Canada, an average of 13 years of education among Canadian entrepreneurs in his studies. Bhola et al. (2006) analyzes this more thoroughly and finds support for a distinguishing difference: they find positive relation between higher level of education and the opportunity drive, as well as between low levels of education and self-employment due to necessity. When studying the past employee compensation of new entrepreneurs in Canada, Amit et al. (1995) found that entrepreneurs, who left their jobs to precede a career as entrepreneurs, had a significantly lower employee income prior to the entrepreneurial career than those who did not. Research upon economic growth have been done by Wong et al. (2005) who have found relations between entrepreneurship and economic growth and their results support a significant relation between High Potential Total Entrepreneurial Activity (HTEA) and Economic growth. Over and above, the access to finance is crucial for entrepreneurship and Verheul et al. (2001) describe how governments in developed countries provide support for local entrepreneurs through direct payment as well as discounted advisory.

Apparently socioeconomic development influence rates in the entrepreneurial activity. One of the factors that could contribute to increased socioeconomic development is the presence of multinational enterprises.

2.3 Multinational Enterprises and their position in foreign economies

16

how Multinational Enterprises influence the local economies where they make foreign direct investments.

2.3.1 MNEs influence on local economy

The MNEs operating today, similar to the ‘Chartered Trading Companies’, often hold a powerful position in the economy through their market dominance and a large global network of bases and partners. Gachino (2010) argues how the globalization has been a key factor leading to the increasing levels of FDI in the global economy. This is the reason why firms can offshore tasks due to the productivity or quality level as of the fact that handling tasks independently no longer justifies the extra costs of producing internally (Baldwin, 2006) and thereafter specialize on the high value adding tasks of the value chain of their core business. The share of FDI inflows in world gross domestic product (WGDP) has increased from 1% to 3% from 1990 to 2005 (Gachino, 2010). This indicates illustrate how the focus of venture capitalists is changing and how the globalization and foreign economies are increasing its interest among venture capitalists. During the last decades, the share of FDI measured in total foreign equity flow has been shifting from being invested in developed countries to dominantly being invested in developing countries (Goldstein and Razin, 2006). This shift take place due to lower production costs which on the other hand increases the profits and returns of the investments (Goldstein and Razin, 2006). The research of Yeaple (2006) provide supplying information about this as his results support a FDI-offshoring of capital intensive industries to capital scarce economies, while he explain R&D intense business to be more likely to be offshored to economies with richly skilled labor. Noeth and Sengupta (2012) on the other hand argue that globalization was what made rich countries capable of becoming important contributors to the outgoing FDI. Goldstein and Razin (2006) results confirm that multinationals operate as venture capitalists and important players in the FDI investment market. Previous studies have illustrated the relation to institutions and how globalization has participated to opening up emerging markets in which politicians are willing to make changes and to modernize their economy to fit the international market (Kunnanatt, 2013). However, Agosin and Machado (2005) argue that many developing countries don’t have the necessary resources to implement efficient screening policies to control FDI and that the adaption of the policies

17

The access to new technology is important for most economies and Blomström and Kokko (1998) describe that the key reasons for countries attempt to attract foreign investment is the eager after gaining insides on new technology related to products, processes and distribution in

addition to marketing and management skills. Other researchers have looked into the linkages between multinationals and local firms and

claimed that it is of importance for economies. Lall (1980) mention the following MNE decisions to be important in these relations; input sourcing locally or as import, make or buy, and the formation of relations between MNE and local supplier. A very important factor is the productivity spillovers that take place as MNEs get established new markets. Blomström and Kokko (1998) argue that productivity spillovers are of high impact on local economy, independently of the ownership structure of the new affiliate. In general one may say that FDI leads to development in an economy firstly due to the resources invested, secondly since it provide a platform to get first-hand experience which make economies better suited to innovate and thirdly it create valuable spillover effects (Barry et al., 2003; Noeth and Sengupta, 2012). The next section of will elaborate more upon the impact that MNEs has on entrepreneurship in a national economy.

2.3.2 The impact of MNEs on entrepreneurship

18

Belgian economy, which led to a significant positive effect on the level of domestic entrepreneurial activity. The increased level of IFDI did lead to increased competition and indeed, at first represent a threat that scared away domestic entrepreneurial activity, but on the long run the effects of increased multinational presence in the economy led to favorable

influences on the local entrepreneurial activity. MNEs deciding to make FDIs in new markets by establishing subsidiaries will most often need to

find strategic partners in the new market to gain resources (Blomström and Kokko, 1998). This situation creates opportunities for entrepreneurs and local ventures. The results of Blomström (1983) and Spencer (2008) supports that key expertise and knowledge (e.g. technological superiority and strong managerial capabilities) from developed country MNEs can be imitated by domestic firms and lead to benefits in local market. A country study from two developed countries, Ireland and UK supports this spillover function as it indicate that multinationals possibly holds a position as a source of learning and education for potential entrepreneurs (McKeon et al. 2004). The knowledge that MNEs import is essential not only for the MNE, but for the development within existing and of new firms in the domestic market (Zhang et al., 2010). Another knowledge transfer takes place through horizontal spillovers. Spencer (2008) and Zhang et al. (2010) mention that horizontal spillovers take place when domestic firms absorb knowledge from MNC and thereby improve their performance. Norman and Ramirey (1993) show this by the example of IKEA that seeks suppliers for long-term relationship and provides them with access to the global market. Moreover, they assist locals with technical assistance, leased equipment and advice on bringing production up to world quality (Norman and Ramírez, 1993). As comes forward here, multinational presence in new local environments can have benefits for local entrepreneurial activity. Their presence increases opportunities for locals to start an entrepreneurial venture. On the other hand, their presence leads to more jobs, which decreases the amount of necessity driven entrepreneurship. The next section will explain the effect that

multinational enterprises have on socioeconomic variables.

2.3.3 How multinationals affect Socioeconomic variables in developed and less developed economies.

19

host countries. Introduction of new industries or products, which link host country closer to the rest of the world are some of these key effects that MNEs have on new markets where they establish (Lipsey, 2004). Washington and Chapman also support this as they mention that MNEs participate to development by offering learning, demonstration, as well as networking and linkages which all transfer potential value to the local environment. When MNEs establish in already existing industries however, Lipsey (2004) explain that wage spillovers appears, possibly due to the increased competition after skilled workers, which again will lead to increased average wage levels locally in host economy. Another way that MNEs could influence local economies is through their sources of power; capital, skills and job opportunities which again could influence institutional framework conditions, such as access to finance, government policies, support programs, educational activities and social and cultural norms. How and to what degree multinationals influence socioeconomic variables is dependent by the level of development in host country and therefore some aspects that may be influenced by this will be presented next. Alguacil et al. (2008) finds support for positive, although modest effects on developed EU countries due to MNE establishment in fields such as technological upgrading, training of labor, acquisit of skills and introduction of new management technics. Washington and Chapman (2014) state that multinational enterprises train and educate local employees, which leads to developing the labor pool and increasing the skills in the host country. Such value adding activities are important for developing countries through increased productivity and increased possibilities for opportunity driven entrepreneurship. Hoekman et al. (2005) argue that R&D intensive FDI leads to higher total factor productivity in developing countries and that technology intensive FDI leads to positive influence on locals due to spillovers on know-how about

international quality levels and standards. Fazekas (2005) argue that MNEs are likely to increase the labor participation in countries with

20

Washington and Chapman (2014). Noorbakhsh and Youssef however argue that countries that mainly attract FDI due to low-cost, low-skilled labor or natural resources meet difficulties with attractive high-value adding industries and therefore experience slower economic growth.

The next chapter will present the Hypothesis on the variables, the mediators and the conceptual model of the thesis.

3. Hypothesis development.

3.1 MNE Presence and the effect on Necessity-Driven and Opportunity-Driven early-stage Entrepreneurial Activity

According to Spencer (2008), the presence of MNEs in local economies might be a threat to local business ventures. Due to increased levels of competition and the increase in vacancies created by the presence of the large international firms, individuals that initially were forced to go into entrepreneurial activity to make a living are now able to find a job at a large multinational venture. Moreover, multinationals create a competitive situation in which local firms are able to compete for a relational position with the MNE as well as the possibility for locals to start up ventures that serve the necessities of the MNE. This way, the individuals that are involved in entrepreneurial activity are less likely to do this out of necessity, but rather of a choice of living as an entrepreneur. It is therefore expected that the level of necessity driven entrepreneurship is decreased by the presence of MNEs. This leads to the following hypotheses:

Hypotheses 1A: Multinational Presence leads to decreased levels of necessity-driven entrepreneurial activity.

21

finding strategic partners is important for the success of multinational enterprises in new markets. Local business ventures can profit from this need by aligning their goals with multinational needs, as well as expanding and increasing their business activities to be a successful partner. It is evident here that not only new ventures can be founded, but there is also the opportunity for expanding current business ventures. It is therefore likely that MNE presence increases the level of opportunity-driven entrepreneurial activity. This leads to hypotheses 1B:

Hypotheses 1B: Multinational Presence leads to increased levels of opportunity-driven entrepreneurial activity.

3.2 MNE Presence and the direct effect on Total Entrepreneurial Activity

As becomes evident above, the effect of MNE presence is noticeable in the level of entrepreneurial activity. Not only does it influence the types of entrepreneurial activity, but also the total amount. By increasing the level of available jobs (Norman and Ramírez, 1993) MNEs decrease the amount of individuals participating in entrepreneurial activity. Moreover, increased competition due to the strong bargaining power and increased levels of competition (Spencer, 2008) allows for less entrepreneurial ventures to survive. It is therefore expected that:

Hypotheses 2A: MNE presence negatively influences the total amount of entrepreneurial activity.

3.3 MNE Presence and the mediated effect of socioeconomic development on Necessity-Driven – and Opportunity-Necessity-Driven Entrepreneurial stage Activity and the Total early-stage Entrepreneurial Activity

22

that increased wage levels occur due to multinational presence, resulting in higher levels of wealth and making individuals less likely to be forced into necessity driven entrepreneurial activity. Moreover, the presence of MNEs can improve the institutional situation in the host country, thereby increasing the availability of finance and other factors that make acceptable living standards more likely to occur for individuals. It is therefore likely that the presence of MNEs does not influence necessity driven entrepreneurial activity directly, but through its effect on socioeconomic development. It is therefore expected that the relationship between MNE presence and necessity driven entrepreneurship is mediated by socioeconomic development.

Hypotheses 3A: The relationship between MNE presence and necessity-driven entrepreneurial activity is mediated by socioeconomic development. MNE presence is positively related to socioeconomic development and socioeconomic development is negatively related to necessity-driven entrepreneurial activity.

23

country as well thereby indirectly stimulating entrepreneurial behavior. This leads to hypotheses 3B:

Hypotheses 3B: The relationship between MNE presence and opportunity-driven early-stage entrepreneurial activity is mediated by socioeconomic development. MNE presence is positively related to socioeconomic development and socioeconomic development is positively related to opportunity-driven early-stage entrepreneurial activity.

While MNE presence is likely to decrease the amount of necessity-driven entrepreneurship and increase the amount of opportunity-driven entrepreneurship, the total level of entrepreneurship is also influenced by the effect that MNEs have on the socioeconomic situation. The institutional environment is influenced by their presence. A larger presence of MNEs will result in more corporate taxes. Djankov et al. (2008) argues that rises in corporate taxes decrease the level of entrepreneurial activity. This could be explained by the rise in corporate taxes decreasing firms profit and thereby increasing barriers linked to starting up new business venturing. Loveman and Sengen Berger (1991) argue that technological developments transferred through MNEs can lead to opportunities for existing firms rather than new business ventures, thereby increasing possibilities for growth for existing firms. This way, possibilities open up for individuals to find a job at an expanding company rather than to engage in business venturing themselves. Moreover, since MNEs usually still staff their foreign subsidiaries with expatriate topmanagers, locals realize that a need for skills is highly important in order to make a career which promotes the development of strong skills rather than individual business venturing. It is therefore likely that the effect of multinational presence on the total level of entrepreneurial activity is influenced by socioeconomic development. This leads to the final hypotheses:

Hypotheses 3C: The relationship between MNE presence and total entrepreneurial activity is mediated by socioeconomic development. MNE presence is positively related to socioeconomic development and socioeconomic development is negatively related to total entrepreneurial activity.

24

The abovementioned relationships are visualized in the following conceptual models.

Figure 1: Conceptual model: Multinational presence and Necessity driven entrepreneurship

25

Figure 3: Conceptual Model: Multinational Presence and Total early-stage Entrepreneurial Activity

4. Methods

4.1. Sample

The data for this research will be collected from secondary databases, meaning no first hand data will be acquired for this project. To collect first hand data on this scale as well as the cross-country nature, will make it time consuming and extremely expensive. Therefore, the data will be collected from external database sources well acknowledged both within the field of research as well as among corporates and institutions. The disadvantage about this is that the data will not be collected for this particular project.

4.1.1 Database and sample description

In this research, five sources of data have been selected based on their relevance to the study, as

well as their well acknowledged quality. The data about the dependent variables, which consists of Necessity-driven early-stage

Entrepreneurial Activity (NEA), Opportunity-driven early-stage Entrepreneurial activity (OEA) and Total early-stage Entrepreneurial Activity (TEA), is gathered from the GEM database. GEM

26

was established in 1999 with data from 10 countries (Reynolds et al., 1999). Since then, they have produced annual publications where they include data on entrepreneurial activity on a country level, collected the previous year (Singh et al., 2015). GEMs last publication covered 73 economies which together represented 72,4% of the world’s population and 90% of the world’s GDP (Singh et al., 2015). This paper finds its samples in the GEM database which differs in its size and in regards to its participating countries based on what yearly report that comes in interest.

27

social and environmental situation (OECD, 2015). The last control variable is the region. By scanning the samples, one could establish that the study covers countries from the following five continents; Africa, Asia, Europe, North- and South America. Europe will function as a reference continent and not an IV. Europe is represented with most countries (17) and Africa has the least participating countries (1) (CIA, 2015).

28 4.2 Operationalization of the variables

4.2.1 Entrepreneurial activity

The Total early-stage Entrepreneurial Activity (TEA) in a country will be measured through the data collected by Global Entrepreneurship Monitor (GEM). The TEA data is collected through telephone interviews, with approximately 2000 members of the adult population (18-64 years) in each country. Both Nascent entrepreneurship, which are people that are actively involved in setting up a business, and New business owners, which are owners managers of a new business which exists less than 42 months are a part of the TEA rate (Singh et al., 2015). The TEA data distinguishes between Necessity-driven and Opportunity-driven Entrepreneurship, which are percentages of the TEA representing respectively the amount of Necessity-driven and

Opportunity-driven Entrepreneurs.

It is important to be aware of the fact that calculating the two aforementioned drives of motivation not will give a 100% sum of TEA since there also are individuals who not consider

themselves as part of either of the groups. Year 2013 is selected as this is the second most recent dataset available and since the number of

countries that participate both in 2005 and 2013 is much higher than of those in the comparison of 2005 and 2014(the most recent dataset). Therefore, the country selection has its ground in availability of data in the GEM database.

4.2.2 Necessity-Driven early-stage Entrepreneurial Activity (NEA)

29

motivation represent of TEA. The data is collected from 2005 and 2013 and the change between these two years is measured.

4.2.3 Opportunity-Driven early-stage Entrepreneurial Activity (OEA)

With opportunity-driven early-stage entrepreneurial activity, it is meant that the participation in business venturing caused by the nature of voluntarily participation. This kind of entrepreneurial activity is based in individuals who involve themselves into business venturing due to a recognized opportunity in the market. These entrepreneurs take advantages of possible business opportunities based on an intention of increasing their earnings or to achieve a higher level of independency. Asc et al., (2008) mention that entrepreneurs with this kind of drive are expecting their business to provide more jobs in comparison to the necessity-driven entrepreneur. This data is collected on the same way as NEA, where the sum of TEA activity is used to calculating the percentage that NEA, OEA and other motivation represent. The information of OEA is based on the change in data from 2005 and 2013.

4.2.4 Multinational presence

30 4.2.5 Socioeconimic development

Since no universal measure of socioeconomic development exists, there are many variables used in different types of research. According to Petrakis and Kostis (2013) the basic indicators are Capital, Labor and Human Capital. To collect data about these variables, different sources of data must be used. When increasing the amount of data-sources, you automatically decrease the reliability. To receive more reliable results, a reduction of sources is therefore recommended. Human Development Index encompasses almost all of the abovementioned variables, in one variable. The index was established by United Nations Development Programme, with the intention of measuring the development of a country with focus on its inhabitants and their capabilities instead of the economic growth alone. Therefore HDI is a measurement to measure people’s development on a country level. It uses the following four components to measure this; Life expectancy in Birth (years), Mean years of schooling (years), Expected years of schooling (years) and Gross national income (GNI) per capita (“year” PPP $). HDI is a well-used and acknowledged variable for measuring the socioeconomic development of a country (Basabe et al., 2002; Jacobsen and Koopmann, 2005; McLaren, 2007), but it does not directly measure labor participation. Through GNI per capital, does this partly get measured, but not as well as through an own variable and could cause biases. The use of HDI data will increases the reliability of the data and will therefore be used to measure the socio-economic development. The data is collected from 2005 and 2013 and these two reference years will be used to measure the change between them.

The level of entrepreneurial activity in an economy will probably be influence by other factors in addition to multinational presence through socioeconomic development. This will be tested by applying three control variables to control the relative impact of the independent variables.

4.2.6 Control variables: Democracy level

31

socioeconomic situation in a country. Instability is negative as it creates a shift in the country’s focus from working, saving and investing to worrying about conflicts and extreme political changes (Hussain, 2014). Its impact could have direct effects on the market through changes in laws and regulations, but also have more long-term effect such as declining trust in the government. Factors that are of influence when discussing political stability in relation to economic growth are according to Hussain (2014) “the rule of law, strong institutions rather than powerful individuals, an efficient bureaucracy, low corruption and an investment enabling business climate”. On the other hand does political stability attract venture capitalists interest, but it could also lead to decline in growth (Hussain, 2014). To measure the level of democracy, the database of Polity IV Project will be used (Systemic Peace, 2015). This index measure the democratization on a scale from 1 – 10, where 10 represent the highest level of democratization (Systemic Peace, 2015). The level of democracy will be compared with the other variables and in regards to differences between the two selected years, 2005 and 2013.

4.2.7 Control Variables: Region

The next control variable is regions. Continents are probably one of the most common understandings of a clustering based on regions. When clustering countries into continents one could finds trends of more and less developed economies which again could be used as a variable to find region specific characteristics. Four dummy variables have been created for the regions represented in the study(Africa, Asia, North and South of America) and Europe is selected as a reference region due to the high frequent of European countries in the sample. Oceania is not represented in the sample and Africa is represented with one country only (South Africa).

4.2.8 Control Variables: OECD Membership

32

membership in OECD it becomes transparent which countries that are most likely to be particularly dedicated into spreading the principles of the free market economy. The data is collected from the 2013 country list (OECD, 2015)

4.2.9 Analysis.

The analysis will take use of regression analysis to test the relations mentioned in the Hypothesis development section. It will be taken use of a multiple regression analysis to measure the aforementioned relationships. The analysis will consist of multiple parts. First the direct relations between MNE presence and NEA, OEA and TEA respectively will be tested. After that, the mediating effect of socioeconomic development will be examined for each of the three independent variables. Additionally, the control variables will be tested for.

5. Assumptions

This thesis will test the hypothesis mentioned in section three by using a linear regression analysis. In order to perform this analysis a number of assumptions need to be checked. According to Osborne and Waters (2002) it is necessary to fulfill these assumptions, otherwise overestimation and underestimation of the significance of the results can occur. The assumptions that need testing according to Osborne and Waters (2002) are linearity, sample size, outliers and correlations. These will be discussed in the following sections.

5.1 Linearity

33

Opportunity-driven Entrepreneurial Activity). By controlling the scatterplots, all of the plots can be assumed to be linear. Scatterplots can be found in Appendix B.

5.2 Sample Size

The size of the sample is important to assure reliable results. Researchers such as Norusis (2008) argue that a number between 10 and 20 individual cases should be added per independent variable. Since this thesis uses eight independent variables, the two main IVs Socio-economic development and Multinational presence and the control variables, it is recommended to use between 80 and 160 cases. It is important to involve enough cases to make sure that the regression line has stability and to be sure that the results will be the same if the test is to be reproduced.

5.3 Outliers

34

there are differences among the dataset. These outliers are not recommended to remove as it would be shrinking the natural variation in the sample.

5.4 Correlations

In table 1, the correlations between the independent and the dependent variables are visually illustrated. Relations between independent variables are not fortunate as they negatively influence the possible causes of the relations between dependent and independent variables. If independent variables relate too much (0,8 or higher) with each other, Pallant and Manual (2010) claim that this may cause uncertainty about which IV actually influencing the dependent variable. When interpreting the table, a relationship between Necessity-driven Entrepreneurial Activity and Socio-Economic Development is found with r= -,403 at a 0,01 significance level. There is also a significant relationship at 0,05 level between Multinational presence and North American region with a correlation coefficient of r=,609.

Table 1: Correlation Matrix

5.5 Multicolleniarity

35

is a warning indication which should be taken into consideration (Pallant, 2010). The results from the regression analysis illustrate that all of the VIFs have low levels, below 5, and that no Tolerance issues are current. Therefore no multicollinearity is measured among the IVs used in this thesis.

6. Empirical Results

This chapter will describe the results of the regression analysis. After standardizing the variables a regression analysis with a mediator is conducted for each of the three dependent variables. In the first step, the direct effect of MNE presence on each of the dependent variables will be tested. In the second step, the effect of MNE presence on socioeconomic development will be tested and in the third step, both MNE presence and socioeconomic development will be regressed on the dependent variable in order to check for a mediating effect. In this third step, the control variables will be added to the model as well.

6.1 Necessity Driven Entrepreneurial Activity

36

lower necessity-driven entrepreneurial activity. None of the control variables show any significant effects.

Table 2: Results of MNE presence & NEA

Table 3: Results of MNE & HDI

Table 4: Full model results

6.2 Opportunity-driven early-stage Entrepreneurial Activity

37

Table 5: MNEs influence on OE

Table 6: Full model results

6.3 Total Entrepreneurial Activity

38

Table 7: MNE influence on TEA

Table 8: Full model results

The empirical results illustrate very few findings among the relations that have been tested. Some relations where found, one that was significant and one close to be significant and therefor also

worth mentioning. First it was found a relation in the full model of NEA, where socioeconomic development

revealed to be of a negative significant relation (β=-0.406, p<.05). Second, although it not indicated a significant relation, OECD membership and TEA had a high effect size, which applies that it still will be of interest to look closer into.

6.4 Sensitivity analysis.

39

used to test the DVs with each of the control variables, separately. This additional analysis was performed to control for potential relations that not was found in the multiple regression analysis with a larger amount of independent variables. The results of regression analysis variant 1, showed two significant relation between TEA and OECD (β=997, p<.01) and OEA and OECD

(β=654, p<.01)

Variant 2 provided no support for significant relationships.

In the next section, the empirical results will be taken under the loupe and discussed more thoroughly.

7. Discussion

The previous chapter enlightened the study with the empirical results. The next chapter will discuss these results and discuss possible approaches for the practical application. This chapter will also reveal the limitations of the study, as well as recommendations for further research.

7.1 MNE presence and the lacking effect on socioeconomic development.

The empirical results imply that MNE presence does not have influence on socioeconomic development. Therefore, the mediating effect that was predicted in hypothesis 3a and 3b could not be defended through these results. This section will elaborate more upon adjustments of, and possible explanations for the empirical results.

40

more developed countries. It is therefore legitimate to propose that an increased diversity among the sample could have increased the impact of MNEs on socioeconomic development.

The measurements of the variables, such as HDI stands as a second fundamental variable in this socioeconomic development equation. HDI does not directly measure labor, which possibly could have been causing relations between MNE presence and socioeconomic development. HDI, as previously mentioned, does not measure the labor participation in a country directly. It measures the GNI per capita, which is likely to increase when unemployment is low, but that also could be low as a result of low average wages in an economy. There is an ongoing discussion about MNEs effect on the unemployment in their host countries. Some researchers, with interest in low-wage countries mention that MNEs are likely to increase the labor participation (Fazekas, 2005) through their participation to output and export growth (Blanchard and Jimeno, 1999; Aktar and Ozturk, 2009), while others argue that MNEs in developed countries increases unemployment (Bailey and Driffield, 2007). Therefore, as labor is excluded from the HDI data, this effect is also not measured. Though if labor would have been included, the financial crisis which occurred in between the two reference years could have caused a misrepresenting effect of MNE presence on the period and thereby created biases. As many of the countries included in the sample have suffered from increased unemployment in the years after the financial crisis, this could be leading to misinterpreted results in the analysis. The effects on socioeconomic development of MNE presence could be understood as a negative while it actually is other causes that are responsible for the registered effects. Despite the fact that labor is not measured directly, the effects of the financial crisis possibly still to have effects on the HDI in regards to the GNI per capital.

41

high level. Again, the composition of the sample could possibly have caused biases and a with a larger and more diversified sample the relations could have visible.

In addition to a selection of measurements and sample composition, is the possibility to measure knowledge transfers a potential for lacking results. MNEs offer learning and demonstration to the local environment (Washington and Chapman, 2004). Although this might be concordant with facts, this does not influence the amount of years that individuals in a country take part in the formal education systems. Therefore, although MNEs possibly contribute to increase the human capital by adding skills and knowledge to locals, it will not be measured through the use of HDI. As these sorts of transfers are arduous and problematic when it comes to weighing, especially in in quantitative studies it is not feasible that companies will keep register in their records of such transactions.

In the next section we will discuss the empirical results received after testing the relation between MNE presence and entrepreneurial activity.

7.2 MNE presence and entrepreneurial activity

The parade of the results bring into view that there are no direct effects of MNE presence on the level of entrepreneurial activity. There are many explanations that possibly could be providing answers to better understand these lacking direct effects.

42

a demand is likely to require a certain level of skills and resources, it would have been more likely that opportunity-driven, and not the necessity-driven entrepreneur would take advantage of such an opportunity. Since the sample is dominated by more developed countries, might this be a cause to the missing direct relation between MNE presence and Opportunity-driven early stage Entrepreneurial Activity (OEA).

Not only is the necessity for new establishments, but the lacking need for low-skilled labor is also likely to influence the relationship. No significant change in relation between Multinational Enterprise (MNE) presence and Necessity-driven early-stage Entrepreneurial Activity (NEA) due to the lacking need for low-skilled labor. Jenkins (2006) elaborate upon this when mentioning that due to higher productivity among the affiliates, the demand after new labor is limited. Jenkins (2006) discusses this particular related to the labor-intensive industries where low-skilled labor is needed. Bailey and Driffield (2007) argue upon the same, as MNE presence leads to increased competition after high-skilled workers, but a nonexistent or negative effect on unemployment rates.

These low-skilled labors are likely to be individuals who will stay in NEAs if no new low-skilled jobs appear and thereby not creating changes in the rates.

Moreover, the missing relation between MNE presence and OEA might exist due to the low demand after local goods in host country. Jenkins (2006) argues that MNEs handle their own need of inputs directly through direct imports. Thereby, the MNE does not create value-creating linkages to local firms. The low creation of demand as of such circumstances, gives opportunities driven entrepreneurs less scope for taking advantage of new opportunities (Jenkins, 2006).

43

opportunities than NEAs. Since the sample is dominated by richer countries, this is likely to be less dominating.

The next section will be discussing the empirical results related to socioeconomic developments

influence on entrepreneurial activity.

7.3 Socioeconomic development and entrepreneurial activity

The empirical results added findings to the relationship between socioeconomic development and entrepreneurial activity. The analysis found a weak, but negative and significant relation between socioeconomic development and NEA. The significant results imply that when the socioeconomic conditions in a country increase, it is likely that the NEA rate in the same country will decrease. Since the increase of socioeconomic variables interpret that the living standard of the people in the country increase, it is reasonable to argue that less people depend upon gaining their earning through forced entrepreneurial activity. As Tambunan’s (2009) argue, the connection between measurements such as years of education is likely to be influencing this relation. A possible cause of the relation between socioeconomic development and NEA, such as increase in years of education, does however not automatically imply that a change in the level of OEA in a country will take place. This is also supported by the omitted direct link between socioeconomic development and OEA from the empirical results. If we look at the OEA definition of Singer et al. (2015), it defines the opportunity-driven entrepreneur to be driven by factors such as increasing their earnings and/or to achieve higher levels of independencies. These individuals are likely to already have reached a relative high/acceptable level of wealth (in terms of standard of living) and will therefore be less dependent by changes in the socioeconomic levels in the country. This would imply that socioeconomic development influence poor people stronger and that it only has a scarce impact on people who have reached a certain level of living standard, in respect to entrepreneurial activity that is. This could be an explanation to why no relation has been detected in the relation between socioeconomic development and OEA.

44

This lacking change in the total numbers of entrepreneurial activity could therefore be a result of the countries represented in the study.

The socioeconomic development could have an influence on entrepreneurship when considering other relations which need to be taken into consideration such as the costs related to starting a business. One could assume that starting a NEA business is related with fewer costs than it is to start an OEA. To start a NEA could as an example mean investing in a shoe cream and a brush and thereby be capable of polishing people’s shoes for a living, while an OEA maybe involving investing in machines or heavy equipment

The next section contains substantial remarks and understanding related to the limitations of the study.

7.4 Limitations

There are multiple limitations that could be mentioned related to this study. First, the selection of data is a very important source of limitations. All data applied in this study is secondary data which means that the data was collected for other purposes that the intentions of this study. The GEM data is collected by volunteers from each country and it is likely to assume that some of the data is not. When using secondary data it is expected that the data not perfectly match the needs of the study where it is applied. For example, although HDI is an aggregated set of data that increased the reliability and contained most of the socioeconomic variables that was intended to be tested in the analysis, it does have some weaknesses. Labor is not directly tested in this index. It is partly measured through the GNI per capital which will decrease by lower and increase by higher labor participation, it will although also be dependent on the average wage level in the economy. Labor participation is important to measure as this potentially is a strong factor that could influence the level of NEA in an economy. Low access to jobs is likely to increase NEA for low-skilled workers (Verheul et al., 2001).

45

Arora, 2014). Even though the questionnaires are standardized, the work of the researchers may differ in its execution and this may cause biases in the data. Collecting primary data would have been a solution to this, but considering the time and expenditure of obtaining primary data it would not have been possible to achieve this within the frame of this study.

Second, the modest size of the sample stands as another limitation of this paper. Consisting of only 31 country cases, the size is by far more cramped than the 80-160 cases recommended by Norusis (2008). The scarce size of the sample leads to less reliable results. The diversity among regions suggests a more narrow and restricted picture, as only one country from Africa was represented, and none from Oceania. Europe on the other hand is heavily overrepresented, which is another issue related to the sample size. With a larger sample it would be likely that the countries participated would have been better distributed and that all continents would have increased their participation rates.

Third, the same issue is relevant for the diversity among countries with various degrees of economic development. With a larger sample it is likely that the levels of economic development would have been better represented among the cases. These limitations exist as a result of the lacking means among countries to participate in multiple GEM studies. The number of cases could have been increased by adjusting the research design of the study by not measuring the delta effect between two years but instead look at differences among countries within the same year. If the most resent GEM database would have been used for this intention, it would have more than doubled the sample size from 31 to 73 cases included. This would, however, not have led to the same results where the change from one year to another is measured instead of just a current status of a number of countries.

Forth, there are also limitations connected to the literature and data available on Africa. This is unfortunate since Africa is an interesting continent with a large population and resources which have the potential to be of interest for MNEs. The influence of modernization and development is probably highly present, but the limited access to information makes the continent less accessible.

46

markets in these new, as well as existing member states. The aftermath of the financial crisis is another example which hit the economic markets, but that is difficult to detect in the data. Many countries still struggle after the struck that their economies took during this crisis. By adding more years in the study, this could have been prevented since it then would have been possible to comprehend the changes before and after the crisis occurred.

The next section will discuss the suggestions that would be recommended for future studies within similar fields of research.

7.5 Suggestions for further research

The limitations presented in the previous paragraph leaves room for multiple improvements that can be implemented into future research of the field.

First, it would be recommended to increase the sample size for future studies. More data available will increase the reliability of the study and generalizability of the empirical results in addition to possibly make some specific connections. Increased data from certain parts of the world, such as on Africa, underdeveloped countries in Asia as well as former Soviet countries from Eastern Europe would be of importance to get a more diversified sample. Since the GEM database releases reports on annually basis and the tendency is leading towards a yearly rise in amount of cases, the importance of this limitation will decrease as more reports are being published.

Second, it seems possible that the level of development possibly holds a key position in the relationship between the tested DVs and IVs. It is therefore recommended for future researchers to shift the spotlight towards such factors that lead to higher development.

47

understanding of the institutional environment and its influence on local economies is assumed to be advantageous.

Fourth, to take the financial crisis into consideration is another recommendation. This have created worse situations in many countries and taken time to remove the aftermath of. This could be avoided by increasing years tested.

Fifth, the control variables used in the research could also have been extended. For example, the level of economic development could have been controlled for. This could have been important as this influence many variables due to different institutions, political priorities and so on. It could also have been tested for membership in trade alliances such as EU or NAFTA. It has already been registered a nearly significant relation between OECD and TEA, so it is likely that other organizations may also have influence.

Another recommendation would be to include more reference years between 2005 and 2013. By adding more years, it will possible to receive a more realistic picture of the changes that have taken place from year to year. Since this study have used only two years, causalities may have caused biases. When adding more years, the results would offer more descriptive results that could be linked with other known factors that have caused impact on economies or trends. It would for example be possible to better understand the change from 2005 until brake out of the financial crisis and the recovering after. Thereby, when adding more reference years, the future researchers could possible better discover irregularities and changes in the economies in a general basis.

48 7.6 Management implications

Although this study provided only scarce significant statistical evidence related to the relationships tested in analysis, there are still key notes to be made from this research. We have seen that socioeconomic development leads to less people involved in necessity driven entrepreneurship, but no significant relation between MNE presence and NEA, OEA or TEA. This relation does affect MNEs in host countries. It is likely that MNEs will be exposed for increasing numbers of ex-entrepreneurs who will need to convert from acting as a self-employed to functioning as an employee. By implementing structures with specific procedures to handle these kinds of issues, it is likely that the transformation process will be more productive both for MNE and for the employee. This does however imply that socioeconomic development has influence on its environments and it is able to activate it to become more or less suitable for FDI depending of the needs of MNEs. his is important for corporates when developing strategies for affiliate establishment or other kinds of FDI. The awareness is also applicable when concerning corporate social responsibility (CSR). Firms contributing to less developed countries may have positively repercussion on the socioeconomic development. The discussions about firm’s responsibility in societies where they are active have been alive for a longer time. Some researchers claim firm’s CSR to be of pure economic intensions (Friedman, 1970), while others that ignoring CSR is expensive (Estes, 1996) and to crucial to excluded from corporate strategy (Caroll and Hoy, 1984). It could also be of interest to be aware of the impact of forums such as OECD when selecting host countries for FDI. This is due to their potential power on fields that possibly could promote economic entrepreneurial activity.

Chapter 8: Conclusion

49

generalizable findings. Scholars have also discussed upon effect of socioeconomic development on entrepreneurship previously, but the adoption of the three variables in a triangle research with a cross-cultural perspective seems to be representing a gap in the literature. Thereby, the following research question was developed.

How can socioeconomic development explain the effect of multinational presence on entrepreneurship?

This thesis did find some support in the empirical analysis, but more importantly it provided interpretations which have contributed with viable answers where analysis showed lacking support and in addition it has added a value to the understanding to this triangular relationship. The thesis did not find any significant triangular relation between Multinational presence, socioeconomic development and entrepreneurship, not with NEA or OEA, and also not with the TEA. It is argued how this lacking relation may have been caused by the selection of the variables for the analysis. These variables, especially HDI which measures socioeconomic development, seem not evaluate the variables that MNEs have the most influence on. Labor participation and knowledge transfers from MNE to individuals or local firms are two of these. This would represent an imaginable explanation to the lacking support for the relationship in the empirical results.

The direct links between MNE and the different entrepreneurial activities did likewise not serve any supporting relations. By adding more countries to the sample with a larger diversity, this could possibly have led to different results, although the empirical results do not offer any significant relations confirming this.