Master Thesis Small Business and Entrepreneurship MSc BA

(1)

Master Thesis

Small Business and Entrepreneurship MSc BA

Self-employment and Subjective Well-Being:

The Moderating Role of Perceived Institutional Quality

2021

Rijksuniversiteit Groningen

Faculty of Economics and Business

Kimmy Coleen Ignacio

S4213793

Supervised by Dr. Michael Wyrwich

Co-assessed by Dr. Erzsi Meerstra-de Haan

(2)

Abstract

This study provides a comparison between self-employed and dependent employees in terms of their subjective well-being. In particular, this research tests whether self-employed experience a greater degree of subjective well-being than those who engage in paid employment (Hypothesis 1). Additionally, light is shed on the role of government quality (as formal institution) and institutional trust (as informal institution) in explaining individual well-being. Drawing on existing literature, it is to expect that government quality and institutional trust positively influence subjective well-being (Hypothesis 2a, Hypothesis 3a). In this regard, it is also investigated whether self-employed benefit more from institutional quality, determined by government quality and institutional trust, than dependent employees. Two further hypotheses derived from the theoretical discussion are tested in the analysis: The positive effect of government quality is stronger for self-employed than for the dependent employees (Hypothesis 2b) and the positive effect of institutional trust is stronger for self-employed than for the dependent employees (Hypothesis 3b). In order to test these relationships, Ordinary Least Squares (OLS) regression is performed using data from rounds 1 to 8 (2002-2016) of the European Social Survey (ESS), covering 32 countries. The total number of observations is 234,941.

The results obtained support the expectation that the well-being of self-employed is significantly higher than the well-being of dependent employees. Also, it is found that there are significant direct positive effects of government quality and institutional trust on perceived well-being, underscoring the importance of improving the overall institutional framework for an individual’s well-being. In contrast to the assumptions, these effects are not more pronounced for the group of self-employed. Implications and future research suggestions on subjective well-being are also discussed within the paper.

Keywords: Subjective Well-being, Self-employment, Government Quality, Institutional Trust,

(3)

1. Introduction

It is generally acknowledged that entrepreneurship can be a potential source of employment generation, economic growth and technology development (Schumpeter, 1934; Fritsch, 2013). Indeed, the review by Shepherd, Wennberg, Suddaby, and Wiklund (2019) found that over the past two decades the main dependent variable used in entrepreneurship studies has been firm performance. However, it is important to note that over time more and more academic researchers have considered to pay greater attention to other dependent variables such as well-being (Shepherd et al., 2019). Subjective well-being and happiness of entrepreneurs play a crucial role in generating greater economic growth (Wincent, Örtqvist, & Drnovsek, 2008). It is therefore important to study the aspects that increase the well-being of entrepreneurs. While many studies confirm the positive impact of entrepreneurship on overall economic welfare, there is lack of empirical evidence on the interaction between institutional quality and entrepreneurship in determining the well-being of the self-employed.

Based on previous literature, entrepreneurship has a positive effect on well-being (Croson & Minniti, 2012). This is confirmed by a recent study of Stephan (2018) who points out that entrepreneurship is related to personal development, growth and subjective well-being. In comparison to paid employees, self-employed people experience higher levels of job and life satisfaction despite the fact that self-employment is often associated with lower incomes and longer working hours (Frey, Benz, & Stutzer, 2004; Benz & Frey, 2008a, 2008b). In fact, it has been found that greater independence (Benz & Frey, 2008a), autonomy (Van Gelderen & Jansen, 2006) as well as self-determination (Benz & Frey, 2008a) are determining factors for individuals of choosing self-employment. In this regard, Benz & Frey (2008a) use the concept of procedural utility, which refers to the value gained from the work process itself, to explain the high job satisfaction of the self-employed.

Alongside the increasing number of studies highlighting the importance of entrepreneurs’ well-being, organisations like the European Commission (2005) emphasised in the European Commission Mental Health Green Paper that the improvement of mental health determines the social well-being and economic prosperity of the European population. Furthermore, the United Nations recognises that “the pursuit of happiness is a fundamental goal” and appeals to Member States to develop “additional measures that better capture the importance of the pursuit of happiness and well-being in development with a view to guiding their public policies” (UN General Assembly, 2012, para. 1). In this context, institutional mechanisms can play a moderating role in the relationship between self-employment and subjective well-being. Nonetheless, the effect of institutions on well-being may vary among the self-employed and paid employees (Fritsch, Sorgner, & Wyrwich, 2019a).

(5)

(North, 1994, p. 361). Formal and informal institutions, which together form the institutional quality of a country, can thus facilitate or restrict business conduct of entrepreneurs and, in turn, affect their subjective well-being. According to Fritsch, Sorgner, and Wyrwich (2019b) and based on Elert, Henrekson, and Stenkula (2017) the former refers to the rule of law, economic freedom, and protection of property rights whereas the latter is characterised by the presence of an entrepreneurial culture (Freytag & Thurik, 2007; Fritsch & Wyrwich, 2017), social legitimacy (Kibler, Kautonen, & Fink, 2014), and institutional trust (Welter & Smallbone, 2006).

Since entrepreneurs’ independent decision-making is associated with overall well-being (Benz & Frey, 2008a), institutions limiting their opportunities can have a negative impact on their subjective well-being. For example, there is evidence that corruption may negatively affect the level of well-being among the self-employed and paid employees (Avnimelech, Zelekha, & Sharabi, 2014). On the other hand, a high level of transparency in regulatory processes and better protection of property rights can lead to better organisation of business conduct and thus may positively affect subjective well-being (Elert et al., 2017). Whilst it is acknowledged that institutional factors can facilitate entrepreneurship, there have been very few studies devoted to the influence of institutional quality on well-being among the self-employed and paid employees (Fritsch et al., 2019b).

In conclusion, it is important to know whether certain institutional mechanisms, specifically formal and informal institutions, can lead to an increase in subjective well-being of the self-employed. Hence, I am interested in investigating whether institutional quality is actually conducive to the degree of subjective well-being. Taken together, the above-mentioned literature shows that entrepreneurship and institutions are relevant determinants that influence well-being. However, there is still a need to empirically test the moderating role of institutional quality on perceived well-being. My research question is therefore as follows: Does entrepreneurship relate to higher subjective well-being among

the self-employed (in comparison to dependent employees) and to what extent does institutional quality moderate this relationship?

(6)

(7)

2. Literature Review

2.1. Self-employment and Subjective Well-being

In the past few years, scientific research on entrepreneurship has placed personal well-being in the foreground. Entrepreneurship research tends to accentuate results at the company-level, such as performance and growth. Most people, however, become entrepreneurs for a variety of different reasons (Carter, Gartner, Shaver, & Gatewood, 2003).According to Stephan (2018) entrepreneurship is related to growth, development as well as individual well-being. Hence, the question of how entrepreneurship is linked with satisfaction and well-being is of paramount importance.

Literature provides evidence that entrepreneurship has a positive effect on well-being (Croson & Minniti, 2012). Although self-employed people (relative to paid employees) often report lower incomes and longer working hours, they experience greater degrees of job and life satisfaction by establishing and running a new company (Frey et al., 2004; Benz & Frey, 2008a, 2008b). Self-employment is associated with higher levels of self-reinforcement and self-determination which in turns leads to enhancement of well-being, measured as work and life satisfaction (e.g., Frey et al., 2004; Benz & Frey, 2008a). In contrast to other professions, entrepreneurs benefit from a different level of flexibility and freedom that allows them to pursue more meaningful work, leverage their strengths and competencies, and take self-organised initiatives (Wood, Bylund, & Bradley, 2016). Additionally, a person may pursue entrepreneurship to gain independence, effectively realise ideas and gather positive experiences (Licht, Goldschmidt, & Schwartz, 2007).

(8)

confirms that self-employed individuals experience higher job satisfaction than paid employees (e.g., Blanchflower, 2000; Binder & Coad, 2013).

A majority of the studies analysing the relationship between entrepreneurship and well-being have either chosen to focus on overall well-being measured as life satisfaction and positive affect or well-being referred to as job-related satisfaction (e.g., Benz & Frey, 2008a; Block & Koellinger, 2009). Fritsch et al. (2019a) even combined the two measures of life and work satisfaction to determine an individual’s subjective well-being. Moreover, psychologists often define subjective well-being as the general inner state of mental well-being. It includes the hedonic dimension which is the achievement of pleasure and the avoidance of unpleasant sensations, and on the other hand, the eudaimonic dimension which refers to positive functioning, fulfilment and self-realisation (Deci, Koestner, & Ryan, 2001; Keyes, 2006). This research uses life satisfaction as an indicator of subjective well-being which also follows on from Veenhoven’s definition (2012) that life satisfaction can be described as “the degree to which an individual judges the overall quality of his/her own life-as-a-whole favorably” (p. 5). Based on these insights, it can be hypothesised that:

H1: Self-employed experience a greater degree of subjective well-being than those who engage in

paid employment.

2.2. Institutions, Self-employment and Subjective Well-being

Institutional conditions may substantially influence the decision of people to move towards self-employment and therefore change the number of self-employed individuals (e.g., Baumol, 1990). This is confirmed by a recent study of Nyström (2008) who states that “institutions can obviously be expected to be influential in a person’s decision to become an entrepreneur” (p. 270). Additionally, Chowdhury, Audretsch, and Belitski (2018) argue that an entrepreneur’s decision-making process and cognition are influenced by formal and informal institutions. Regulatory limitations can force people to adjust their decisions and thus reconsider otherwise interesting business opportunities (Wood et al., 2016).

In general, institutions can be described as “the rules of the game” in a society that shapes human interaction (North, 1994, p. 361). Based on the study of Chowdhury et al. (2018) institutional factors can determine the quantity of entrepreneurship, which is line with Baumol’s (1990) notion, as well as the quality of entrepreneurship which is associated with innovation, development and productivity. Well-developed institutions may increase the number of opportunity-driven entrepreneurs which can result in greater well-being of the self-employed (Binder & Coad, 2016; Block & Koellinger, 2009).

(9)

self-employed” (p. 2).Also, the authors found that entrepreneurs operating in “liberal” market economies (e.g., the UK and Ireland) and “coordinated” market regimes (e.g., Germany and Austria) as opposed to many Mediterranean and Eastern European countries report on average higher job and life satisfaction. Since entrepreneurs’ subjective well-being is related to independent decision-making (Benz & Frey, 2008a), institutions limiting their choices can have a negative effect on their general well-being. Other previous studies showed that costs resulting from corruption (Anokhin & Schulze, 2009) and tax issues (Bruce & Mohsin, 2006) determine business performance and possibly also certain elements of well-being.

In view of the above, it can be argued that institutional quality influences the relationship between self-employment and well-being. Institutional quality can be explained by the presence of regulations, certain sanction proceedings, cultural legitimacy and the consequent level of an individual’s economic freedom (Sims, Gong, & Ruppel, 2012). Generally, the literature draws a distinction between formal and informal institutions. According to Fritsch et al. (2019b) and based on Elert et al. (2017) the former refers to the rule of law, economic freedom, and protection of property rights whereas the latter is characterised by the presence of an entrepreneurial culture (Freytag & Thurik, 2007; Fritsch & Wyrwich, 2017), social legitimacy of entrepreneurship (Kibler et al., 2014), and institutional trust (Welter & Smallbone, 2006). This study focuses on government quality as formal institution and institutional trust as informal institution. Together they form the two aspects of institutional quality which will be discussed in more detail in the following sections.

2.3. Government Quality on Subjective Well-being

At present, there is no universal definition of government quality because existing descriptions are either enormously broad or focus merely on corruption (Rothstein & Teorell, 2008). The definition of quality of government was inspired by the World Bank’s definition of governance which refers to “the traditions and institutions by which authority in a country is exercised” (Kaufmann, Kraay, & Zoido-Lobatón, 1999, p. 1). Kaufmann and associates at the World Bank (1999) suggested six dimensions of governance which were frequently used as governance measures in previous empirical research (e.g., Alam, Uddin, & Yazdifar, 2019). These include government effectiveness and regulatory quality which were also used as indicators for government quality in the study of Elert et al. (2017). The former is defined as “perceptions of the quality of public and civil services and the degree of their independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies” whereas the latter denotes “perceptions of the ability of the government to formulate and implement sound policies and regulations that permit and promote private-sector development” (World Governance Indicator (WGI), World Bank1_).

(10)

Government quality can affect the subjective well-being of citizens because governments have the means, including formal institutions, resources, and public policies that can influence an individual’s life satisfaction (Kim & Kim, 2012). More specifically, a good government and the rule of law can enhance subjective well-being by ensuring an acceptable level of social equality, justice, healthcare, safety, enforcement of property rights and the possibility to participate in elections (Bjørnskov, Dreher, & Fischer, 2010; Ott, 2010). Apart from these conditions, Nikolova (2016) points out that “good institutions imply a strong social contract guided by norms of civic engagement, which is also conducive to greater well-being” (p. 133). By contrast, weak institutions associated with corruption and inefficiency can cause high psychological costs due to dealing with a dysfunctional system, ultimately reducing subjective well-being (Welsch, 2008). Moreover, Whiteley, Clarke, Sanders, and Stewart (2010) analysed policy outcomes and life satisfaction and found that government quality can significantly influence well-being of individuals through economic and other policy outcomes. It can therefore be expected that government quality (as formal institution) positively influences subjective well-being of individuals.

Well-developed institutions with high quality of government facilitate market entry while corruption, bureaucracies and a large unofficial market are likely to discourage entrepreneurs to start a new company (Djankov, Porta, Lopez-De-Silanes, & Shleifer, 2002). In this respect, Fritsch et al. (2019b) state that “inefficient regulations, bureaucracies, and infrastructures” may result in business formation delays and desperation of entrepreneurs, which in turn negatively affects their subjective well-being (p. 4). In the same vein, the greater the costs of enforcing contracts and corporate taxes, the less enjoyable it is to engage in business activities, leading to lower well-being (Fritsch et al., 2019b). Furthermore, Goel and Nelson (2014) claim that laws and regulations can effectively reduce corruption in a country. The extent of corruption may negatively influence the level of well-being among the self-employed and paid employees (Avnimelech et al., 2014).

The quality of government determines the effectiveness of the system of property rights protection (Elert et al., 2017), which in turn influences the well-being of entrepreneurs. By investigating the link between government quality and R&D investment, Alam et al. (2019) found that higher government quality results in more R&D investment by companies. The availability of effective intellectual property protection can reduce uncertainty and increase personal safety of entrepreneurs, hence increasing the life satisfaction of entrepreneurs.

(11)

Furthermore, high government quality can decrease the negative effects of government intervention on companies and therefore increase the life satisfaction of entrepreneurs. While government intervention has a macroeconomic control function, it may also hinder the growth of businesses. Lu and Pan (2016) examined the relationship between government quality and subjective well-being of entrepreneurs. The authors argue that entrepreneurs may feel more frustrated if they believe that administrative and financial costs resulting from government intervention are increasing, thus reducing the well-being of entrepreneurs.

In the light of the foregoing, regulations for the protection of rights (Elert et al., 2017), favourable institutional environment (Estrin et al., 2013) and government intervention (Lu & Pan, 2016) can contribute to establishing entrepreneurial-friendly regulations. Institutions facilitating entrepreneurship may play an important role in decreasing the administrative burden on business operations. Consequently, entrepreneurs can run their business more effectively and with less frustration. In fact, there is evidence that reduced regulatory burden to promote entrepreneurship leads to an increased level of well-being among the self-employed (Fritsch et al., 2019a). In other words, institutional quality that makes it easier to engage in business activities should feedback positively on entrepreneurs’ life satisfaction.

Subjective well-being of dependent employees can also be affected by government quality, specifically by the presence or absence of corruption, institutional environment, and state intervention (Avnimelech et al., 2014; Lu & Pan, 2016). Nevertheless, it can be anticipated that the effects of these institutional factors have a less direct impact on the group of paid employees than on the group of self-employed. This is likely to be the case for institutions promoting business entry, institutions facilitating business operations as well as a system for property rights protection. Taking the previously discussed literature into account, one may assume that government quality can have a positive impact on an individual’s well-being. Further, it is to expect that that the relationship between government quality and paid employees’ well-being is less strong as compared with the relationship between government quality and self-employed well-being. In line with this institutional thinking, I propose the following hypotheses:

H2a: Government quality positively influences subjective well-being of individuals.

H2b: The positive effect of government quality is stronger for self-employed than for the dependent

employees.

2.4. Institutional Trust on Subjective Well-being

(12)

and its informal rules” (p. 224). In other words, institutional trust denotes trust in the institutional framework which involves formal organisations, enforcement of sanctions, as well as informal codes of behaviour and norms (Welter & Smallbone, 2006). The associated mechanisms are fairness, impartiality and neutrality (Su et al., 2020) which is in line with the notion of Rothstein (2005) who describes generalised trust as the unbiased and efficient performance of public organisations, especially institutions such as the judiciary that upholds order and good government. Furthermore, institutional trust is created when people feel that institutions act in their interest (Hardin, 1999). In this regard, it has been proven that trust “depends on people’s perceptions of the extent to which institutions live up to such normative expectations as incorruptibility and honesty” (Grönlund & Setälä, 2012, p. 523).

The research of Hudson (2006) examines institutional trust and well-being and reports that institutional trust is related to some aspects of well-being. Based on the results of this study, trust in particular institutions positively influences individual well-being. In particular, the law, the national government, the United Nations (UN), the European Central Bank (ECB), and the European Union (EU) were included in the analysis. It was concluded that institutional trust depends on the level of performance of these institutions which is consistent with Rothstein’s (2005) definition of institutional trust. In sum, Hudson (2006) confirms the findings of Frey and Stutzer (2002) that subjective well-being is not merely determined by endogenous factors but also by institutional performance. These results are similar to those reported by Su et al. (2020) who show that trust has a direct impact on subjective well-being. Accordingly, it is reasonable to assume that institutional trust (as informal institution) positively influences subjective well-being of individuals.

Informal institutions have an impact on the functioning of formal institutions but can also be significant in themselves to promote entrepreneurship (Elert et al., 2017). The authors discuss that standards, customs and, especially trust that contribute to the ease of doing business need to be reinforced. It is acknowledged that unstable institutions impede democracy and the workings of the free market economy. Based on the available literature, high-trust environments are related to business growth, competition and productive entrepreneurship (Fukuyama, 1995). An explanation for this relationship is that informal institutions such as institutional trust can replace the function of formal institutions to decrease transactions costs (Arrow, 1972). On the other hand, in an environment with high levels of distrust, people need to invest time in order to identify risks and in some cases take them (Hardin, 2001). This is in line with the conclusion of Arrow (1972) who argues that individuals incur higher transactions costs in low-trust environment because of a greater need to continuously monitor economic transactions. Since self-employed seek to innovate and achieve business growth (Zhao, 2005), it can be assumed that self-employed compared to dependent employees make more use of the benefits their high-trust society provides, leading to higher subjective well-being.

(13)

other words, institutional trust can promote the ease of conducting the business, which is helpful to improve the well-being of entrepreneurs (Fritsch et al., 2019b). In comparison to dependent employees, self-employed individuals attach greater importance to working in an independent and effective manner (Benz & Frey, 2008a; Van Gelderen & Jansen, 2006), therefore they may derive greater satisfaction from their job in a high-trust environment. Secondly, higher levels of trust may support in creating a steady business environment for investments and other financial operations (Knack & Keefer, 1997). In addition, it has been suggested that economic actors work more efficiently and profitably in an environment with mutual confidence (Fafchamps, 2006). A stable environment is not solely important for economic growth, but also provides more certainty and personal security. Due to the fact that self-employment is related to financial uncertainty, it is possible that the positive effect of institutional trust in enabling a more predictable environment is stronger for the self-employed than for organisationally employed individuals.

In the light of the aforementioned considerations, there are good reasons that trust in institutions can have a positive impact on an individual’s well-being. These observations suggest that, while institutional trust can enhance the subjective well-being of both self-employed and dependent employees, the effect of institutional trust is expected to be more pronounced for the former than for the latter. As such, it is hypothesised that:

H3a: Institutional trust positively influences subjective well-being of individuals.

H3b: The positive effect of institutional trust is stronger for self-employed than for the dependent

employees.

2.5. Conceptual Research Model

The following conceptual model represents the proposed relationships in this study. All relationships are controlled for gender, age, level of education, industry and survey year.

Figure 1: Conceptual Research Model

H

3a

+

H

2a

+

H

3b

+

H

2b

+

Self-employment

Subjective Well-

_being

(14)

3. Methodology

3.1. Data and Sample

The empirical analysis of this paper is based on data from the European Social Survey (ESS) which is an academically driven cross-national survey. The dataset provides data on socio-political attitudes and public perceptions within Europe. The data can be accessed online without restrictions by simple and quick registration. The ESS started in 2001 and the survey takes place every two years in over 30 countries to date. The data is collected on an individual level and the respondents are aged 15+ regardless of their nationality, citizenship or legal status in the participating countries. One half of the questionnaire used in each face-to-face interview is asked in each round. The core topics are, for example, trust in institutions, moral and social values, subjective well-being and security. The second half includes rotating topics such as attitudes towards welfare, economic morality and family, work and well-being.

In this research ESS data collected from the rounds 1 to 8 are used, covering the period between 2002 and 2016. The total number of observations is 234,941. The selected sample consists of individuals who are either self-employed (26,995) or dependent employees (207,946). This study follows the definition of OECD (2020) that being self-employed means to work for themselves as opposed to for an employer. Hence, participants working for their own family business are discarded from the sample because it remains unclear whether they are mainly responsible for the company or are considered as a regular employee in the firm. The data includes people between the age of 18 and 65, with an average age of 43 years, of which 48.33% are males (113,551) and 51.67% females (121,390). Moreover, the sample provides details on level of education, life satisfaction (as a proxy for subjective well-being), as well as country-level factors, such as government quality and institutional trust.

3.2. Measures

Subjective well-being. Drawing on the emotional model theory of subjective well-being by Diener

(1984), people’s own evaluation of their own lives determines the level of their well-being (Winkelmann, 2009). According to Fritsch et al. (2019b) “Life satisfaction is intended to represent a broad, reflective appraisal a person makes of his or her life” (p. 4). Based on previous studies, life satisfaction is the most widely used and best validated indicator to represent well-being (Pavot & Diener, 2008). In this research, the concept of life satisfaction is therefore used to measure the subjective well-being of the participants. The corresponding ESS item is as follows: “All things considered, how

satisfied are you with your life as a whole nowadays?” (European Social Survey, 2016). It is measured

(15)

Self-employment. As for employment status, respondents of the European Social Survey will

be distinguished between being self-employed and being a dependent employee. The occupation status of the individuals is reflected in the ESS item “Employment relation” with the corresponding values 1 = employee, 2 = self-employed, 3 = working for own family business. The question asked is: “In your

main job are/were you an employee, self-employed, or working for your own family’s business?”

(European Social Survey, 2016). For the empirical analysis, participants who indicated to work for own family business (3) were excluded. The reason for this lies in the ambiguity with respect to their entrepreneurial or non-entrepreneurial position within the company. The ESS item “Employment relation” was transformed into a binary variable called “Self-employment” and coded as 0 = employee and 1 = self-employed. In accordance with the approach of previous studies, self-employed individuals are considered to represent entrepreneurs (Nyström, 2008).

Government quality. This research follows the WGI, World Bank definition of government

effectiveness and regulatory quality which were also used to measure government quality in the study of Elert et al. (2017). The former is defined as “perceptions of the quality of public and civil services and the degree of their independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies” whereas the latter denotes “perceptions of the ability of the government to formulate and implement sound policies and regulations that permit and promote private-sector development” (World Governance Indicator (WGI), World Bank). Thus, government quality is measured with the item: “Now thinking about the

[country] government, how satisfied are you with the way it is doing its job?” (European Social Survey,

2016). It is measured on a Likert scale from 0 (Extremely dissatisfied) to 10 (Extremely satisfied).

Institutional trust. Based on available literature, generalised trust is described as the unbiased

and efficient performance of public organisations, especially institutions such as the judiciary that upholds order and good government (Rothstein, 2005). Su et al. (2020) define institutional trust, which is in line with the study of Rothstein and Stolle (2008), as “judgement of the functioning of the overall political, legal or economic framework and its informal rules” (p. 224). In this research, institutional trust is therefore measured with the item: “How much do you personally trust the legal system?” (European Social Survey, 2016). It is measured on a Likert scale from 0 (No trust at all) to 10 (Complete trust).

Control variables. In this study, all relationships are controlled for gender, age, level of

(16)

Classification of Education (ISCED) which makes it possible to compare national education systems across countries. Based on these standards, the lowest level of education is classified as ISCED 0-1 (less than lower secondary education) and the highest level of education as ISCED 5-6 (tertiary education completed). Furthermore, industry sector was used as a control variable. Due to the fact that different NACE (Statistical Classification of Economic Activities in the European Community) classifications were used across the survey years, more specifically NACE rev.1 for round 1, NACE rev.1.1 for rounds 2-4, NACE rev.2 for rounds 5-8, a new variable is created to reduce the industries to more general categories. Summarising the wide variety of industries into groups resulted in the creation of 10 dummy variables to account for the various industry sectors. In that respect, correspondence tables and the table for “high-level SNA/ISIC aggregation A*10/11” provided by Eurostat are used for the categorisation. A detailed overview of the 10 industries included in this study can be seen in the Appendix (Table A1). Moreover, survey year dummies are included in all models to capture year-specific effects.

3.3. Method of Analysis

(17)

4. Analysis and Results

4.1. Descriptive Statistics

Table 1 provides a summary statistic which describes the basic features of the data used in this study. The average subjective well-being across ESS countries is quite high with a mean of 6.89, which is measured on a Likert Scale from 0 (Extremely dissatisfied) to 10 (Extremely satisfied). The independent variable self-employment is transformed into a binary variable with the values 0 = employees and 1 = employed. A mean of 0.115 indicates that 11.5% of the total sample used in this research are self-employed, and thus 88.5% are dependent employees. Based on Table 1, the average government quality throughout the ESS area is rather low with a rating of 4.13 points out of 10 (Extremely satisfied). With respect to the level of institutional trust, respondents rate their institutional trust with an average score of 5.11 out of 10 (Complete trust).

As far as the control variables are concerned, age shows that the average age of participants in the total sample is 43 years. Individuals between the age of 18 and 65 are included in the final dataset, of which 48.33% are males and 51.67% females. Gender as a control variable is coded as 0 = male and 1 = female. Another control variable is the level of education which shows a mean of 3.44, suggesting that overall, respondents completed the upper secondary education. Lastly, industry was also used to control for all relationships. This variable includes 10 sectors such as agriculture, manufacturing, construction, wholesale and retail trade which are based on the general categorisation provided by Eurostat (Appendix, Table A1).

Variable Mean Standard

Deviation Minimum Maximum Subjective well-being 6.891 2.259 0 10 Self-employment 0.115 0.319 0 1 Government quality 4.131 2.445 0 10 Institutional trust 5.114 2.683 0 10 Age 43.179 12.886 18 65 Gender 0.517 0.5 0 1 Level of education 3.443 1.258 1 5 Industry 5.531 3.012 1 10

Table 1: Descriptive Statistics. Number of Observations = 234,941

4.2. Comparison between Countries in Terms of Their Average Subjective Well-being, Average Government Quality and Average Institutional Trust

(18)

32 participating countries: Austria, Belgium, Bulgaria, Croatia, Cyprus, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Lithuania, Luxembourg, Netherlands, Norway, Poland, Portugal, the Russian Federation, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, Ukraine and the United Kingdom. In total, the sample consists of 234,941 participants.

It emerges from Table 2 that there are notable differences between countries in terms of average subjective well-being of their people. While Denmark appears to be the happiest country with an average level of subjective well-being of M = 8.40, Ukraine seems to be the least happy country with

M = 4.50. At the same time, it can be observed that people in Luxembourg rate the country’s government

quality with an overall score of M = 6.18, representing the highest rating across the ESS area, in comparison to people in Ukraine who evaluate their government quality with merely M = 2.41. Finally, the data suggests that Denmark has the highest points pertaining to institutional trust with M = 7.43 whereas Ukraine shows once again the lowest points with M = 2.24.

(19)

Slovakia 5579 2.37 6.23 4.03 3.72 Slovenia 6427 2.74 _6.82 _3.50 _3.69 Spain 9836 4.19 7.18 3.50 4.19 Sweden 9708 4.13 7.82 5.14 6.23 Switzerland 9376 3.99 _8.00 _5.89 _6.39 Turkey 1549 0.66 5.63 4.66 6.07 Ukraine 5508 2.34 4.50 2.41 2.24 United Kingdom 11435 4.87 _7.01 _4.05 _5.25 Total 234941 100.00 _6.89 _4.13 _5.11

Table 2: Number of Respondents per Country, Average Subjective Well-being, Average Government

Quality, Average Institutional Trust

4.3. OLS Regression on Subjective Well-being

To begin with, it should be mentioned that the results of models 1-3 are analysed in this section whereas models 4-6 are explained in section 4.5. (Robustness checks). Before analysing the results, it is necessary to assess whether the regression models in Table 3 fit the data. By examining the R-squared (R2_{), it is possible to determine what percentage of the variability in the dependent variable is explained}

by the model. Hence, the adjusted R-squared of 0.164 in Table 3 indicates that approximately 16.4% of the variance in the dependent variable (subjective well-being) can be predicted from the independent variables (self-employment, government quality, institutional trust) and the control variables (gender, age, level of education, industry) within the model. Additionally, the F-value and its associated p-value F (22, 234918) = 1711.57, p < 0.001 show that the group of independent variables have a statistically significant relationship with the dependent variable, thus confirming a good overall model fit.

Table 3 provides the results of the OLS regression with the above-mentioned dependent, independent and control variables. While model 1 shows the effect of self-employment on subjective well-being, model 2 looks at the direct effects of government quality and institutional trust on subjective well-being. Further, model 3 displays the interaction effects between self-employment and government quality as well as between self-employment and institutional trust on subjective well-being.

Based on the results obtained from the regression analysis, hypothesis 1, which expects that self-employed are happier than dependent employees, can be confirmed. Model 1-3 clearly indicate the significant positive impact of being self-employed on an individual’s subjective well-being at the p < 0.001 level. The strength of this relationship is reflected in model 1 with ß = 0.325, model 2 with ß = 0.330 and model 3 with ß = 0.622. Self-employment results in higher levels of well-being compared to regular employment which is consistent with the findings of prior studies. Thus, hypothesis 1 is confirmed.

(20)

becomes clear from the output that there is a significant positive impact of government quality on the level of subjective well-being (β = 0.207, p < 0.001). An increase of one point on the scale of government quality results in an increase of 0.207 points on the scale of subjective well-being. Therefore, hypothesis 2a is supported. Similarly, the same model provides evidence that institutional trust has a direct significant positive effect on the level of subjective well-being (β = 0.172, p < 0.001). An increase of one point on the scale of institutional trust leads to an increase of 0.172 points on the scale of subjective well-being. Hence, hypothesis 3a is confirmed.

The third model presents the interaction effect between self-employment and government quality on the level of subjective well-being. For the corresponding hypothesis 2b, stating that the positive effect of government quality is stronger for self-employed than for the dependent employees, the results do not support this expectation based on the literature. The analysis shows that there is a slightly negative interaction effect between self-employment and government quality on subjective well-being (β = - 0.001) and with no significant difference among the self-employed and dependent employees. Consequently, hypothesis 2b is rejected.

At the same time, model 3 indicates the interaction effect between self-employment and institutional trust on the level of subjective well-being. By multiplying the variable institutional trust with self-employment, it becomes apparent that there is a significant difference between the levels of subjective well-being among the self-employed and dependent employees. In terms of effect size, the positive effect of institutional trust on self-employed well-being is ß = 0.122 (0.179 – 0.057). The interaction effect with ß = - 0.057, p < 0.001 means that self-employed are less positively affected by institutional trust than dependent employees. This finding contradicts hypothesis 3b which assumes that the positive effect of institutional trust is stronger for self-employed than for the dependent employees. Based on the results, dependent employees benefit significantly more from institutional trust than the group of self-employed. Therefore, hypothesis 3b is rejected.

(21)

VARIABLES Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Self-employment (0=Employee, 1=Self-employed) 0.325*** 0.330*** 0.622*** 0.200*** 0.214*** 0.565***

(0.015) (0.014) (0.038) (0.014) (0.013) (0.038)

Government quality 0.207*** 0.207*** 0.210*** 0.210***

(0.002) (0.002) (0.002) (0.002)

Institutional trust 0.172*** 0.179*** 0.188*** 0.196***

(0.002) (0.002) (0.002) (0.002)

Self-employment x Government quality -0.001 -0.005

(0.006) (0.006)

Self-employment x Institutional trust -0.057*** -0.066***

(0.006) (0.006)

Age -0.013*** -0.011*** -0.011***

(0.000) (0.000) (0.000)

Gender (0=Male, 1=Female) -0.116*** -0.023* -0.022*

(0.010) (0.009) (0.009)

Level of education 0.138*** 0.070*** 0.070***

(0.004) (0.004) (0.004) Industry = 2, Manufacturing, mining and quarrying and other industry 0.294*** 0.314*** 0.312***

(0.029) (0.026) (0.026)

Industry = 3, Construction 0.250*** 0.270*** 0.267***

(0.032) (0.029) (0.029) Industry = 4, Wholesale and retail trade, transportation and storage, accommodation

and food service activities

0.355*** 0.352*** 0.347*** (0.028) (0.026) (0.026)

Industry = 5, Information and communication 0.670*** 0.533*** 0.530***

(0.038) (0.035) (0.035)

Industry = 6, Financial and insurance activities 0.940*** 0.752*** 0.745***

(0.035) (0.032) (0.032)

Industry = 7, Real estate activities 0.566*** 0.448*** 0.447***

(0.059) (0.054) (0.054) Industry = 8, Professional, scientific, technical, administration and support service

activities

(22)

Industry = 9, Public administration, defence, education, human health and social

work activities 0.802*** 0.653*** 0.646***

(0.028) (0.026) (0.026)

Industry = 10, Other services 0.501*** 0.434*** 0.430***

(0.032) (0.030) (0.030) Year = 2004 -0.128*** -0.070*** -0.070*** (0.019) (0.018) (0.018) Year = 2006 -0.302*** -0.223*** -0.222*** (0.020) (0.018) (0.018) Year = 2008 -0.464*** -0.264*** -0.262*** (0.019) (0.018) (0.018) Year = 2010 -0.455*** -0.190*** -0.189*** (0.019) (0.018) (0.018) Year = 2012 -0.342*** -0.131*** -0.130*** (0.019) (0.018) (0.018) Year = 2014 -0.053** 0.034 0.035 (0.019) (0.018) (0.018) Year = 2016 0.012 0.044* 0.044* (0.018) (0.018) (0.018) Constant 6.725*** 5.043*** 5.007*** 6.868*** 5.037*** 4.995*** (0.036) (0.035) (0.035) (0.005) (0.013) (0.013) Observations 234,941 234,941 234,941 234,941 234,941 234,941 Adjusted R-squared 0.035 0.164 0.164 0.001 0.149 0.149

Robust standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

(23)

In order to get the best possible OLS estimates and thus draw reliable conclusions, the regression analysis was conducted with robust standard errors that allow for the presence of heteroscedasticity. After performing the Breusch-Pagan test in Stata, a test for heteroscedasticity of errors in regression, one may assume that there are signs for non-normal distribution of the error terms (residuals). Since homoscedasticity is a classical OLS assumption, robust standard errors are used to obtain unbiased standard errors of OLS coefficients. Also, for statistical accuracy it is necessary to mention that multicollinearity is not present. The correlation matrix (Table A2) in the Appendix shows that there are no correlations above 0.7 between the independent variables. Lastly, the Variation Inflation Factor (VIF), which measures the severity of multicollinearity in an OLS regression analysis, for all main variables is under 4.

4.4. Effect of Self-employment on Predicted Subjective Well-being at Different Levels of Government Quality and Institutional Trust

Due to the fact that the OLS regression does not show the effect of self-employment on subjective well-being at different levels of institutional quality, determined by government quality and institutional trust, two separate plots (Figure 2 and Figure 3) are displayed to show these relationships. As regards Figure 2, the outcomes resemble the findings of Table 3 pertaining to self-employed being happier than dependent employees. Furthermore, the effect of being self-employed and being a dependent employee on predicted subjective well-being at different points of the distribution of government quality is relatively constant between the two groups. This can be seen by the gap between the line for the self-employed (red line) and the line for the dependent employees (blue line) as well as by how the two lines run fairly parallel to one another. However, it is notable that there is no difference in the level of predicted well-being when self-employed and dependent employees are extremely satisfied with the government quality (= 10 on the government quality scale). This is reflected in the overlap of the red line (self-employed) with the blue line (dependent employees).

(24)

Figure 2: Effect of Self-employment on Predicted Subjective Well-being at Different Levels of

Government Quality

Figure 3: Effect of Self-employment on Predicted Subjective Well-being at Different Levels of

(25)

4.5. Robustness Checks

In order to test whether the results of the OLS regression in models 1-3 (Table 3) are robust, the same analysis was conducted again in models 4-6 (Table 3) but without any control variables. It can be concluded that there are indeed no significant changes once sociodemographic variables, industry and year effects are excluded.

The OLS regression in Table 3 examined the interaction effect between self-employment and government quality as well as between self-employment and institutional trust on subjective well-being. However, one cannot conclude whether these interaction effects differ between countries. It is for this reason that an additional analysis for a subset of countries is performed to determine the robustness of the results among all ESS countries. In order to draw conclusions for all countries, a group of disparate countries with varying levels of subjective well-being are selected. The countries included in the second robustness check are Denmark, the Netherlands, Spain, Ukraine and Bulgaria.

(26)

Denmark The Netherlands Spain

VARIABLES I II III I II III I II III

Self-employment 0.140* 0.128* 0.696** 0.115* 0.125* 0.395 0.219*** 0.218*** 0.144 (0=Employee, 1=Self-employed) (0.059) (0.059) (0.261) (0.052) (0.051) (0.227) (0.057) (0.056) (0.131) Government quality 0.107*** 0.111*** 0.160*** 0.157*** 0.091*** 0.078*** (0.008) (0.009) (0.010) (0.011) (0.010) (0.011) Institutional trust 0.111*** 0.115*** 0.087*** 0.095*** 0.046*** 0.053*** (0.011) (0.012) (0.010) (0.010) (0.009) (0.010)

Self-employment x Government quality -0.041 0.025 0.072**

(0.024) (0.032) (0.024)

Self-employment x Institutional trust -0.045 -0.067* -0.043

(0.033) (0.030) (0.024)

Control variables1 _Included _Included _Included _Included _Included _Included _Included _Included _Included

Constant 8.145*** 6.797*** 6.773*** 7.428*** 6.450*** 6.399*** 6.836*** 6.294*** 6.293*** (0.140) (0.166) (0.168) (0.134) (0.140) (0.143) (0.136) (0.142) (0.143) Observations 7,657 7,657 7,657 10,649 10,649 10,649 9,836 9,836 9,836 Adjusted R-squared 0.010 0.068 0.068 0.023 0.088 0.089 0.028 0.047 0.048

1_{Age, Gender, Level of education, Industry, Year}

(27)

Ukraine Bulgaria

VARIABLES I II III I II III

Self-employment 0.275 0.343* 0.288 0.627*** 0.645*** 0.581** (0=Employee, 1=Self-employed) (0.149) (0.140) (0.222) (0.133) (0.124) (0.220) Government quality 0.310*** 0.310*** 0.317*** 0.313*** (0.017) (0.017) (0.018) (0.019) Institutional trust 0.108*** 0.106*** 0.104*** 0.107*** (0.016) (0.016) (0.016) (0.017)

Self-employment x Government quality 0.000 0.060

(0.072) (0.062)

Self-employment x Institutional trust 0.026 -0.045

(0.067) (0.067)

Control variables1 _Included _Included _Included _Included _Included _Included

Constant 4.772*** 3.052*** 3.055*** 4.072*** 3.093*** 3.091*** (0.231) (0.230) (0.230) (0.235) (0.223) (0.223) Observations 5,508 5,508 5,508 4,881 4,881 4,881 Adjusted R-squared 0.051 0.153 0.153 0.117 0.224 0.224

1_{Age, Gender, Level of education, Industry, Year}

(28)

5. Discussion and Conclusion

5.1. Discussion and Interpretation of Results

The empirical analysis of this research indicates the positive association between self-employment and subjective well-being. The results provide evidence that self-employed are happier than dependent employees, confirming hypothesis 1. A likely reason for this relationship is the higher levels of self-reinforcement and self-determination which is related to self-employment (e.g., Frey et al., 2004; Benz & Frey, 2008a). In comparison to dependent employees, entrepreneurs experience greater flexibility and freedom that allows them to pursue more meaningful work and make use of their strengths and competencies (Wood et al., 2016). It is the key factors such as independence, autonomy and self-determination (Benz & Frey, 2008a; Van Gelderen & Jansen, 2006) attributed to the group of entrepreneurs that may explain the significant difference in well-being among the self-employed and dependent employees. Drawing on the concept of procedural utility, individuals not only derive value from end results but also from actively engaging in the work process itself (Frey et al., 2004). Therefore, the achievement of personal goals through self-employment can arouse a sense of self-effectiveness, which in turn increases their job and life satisfaction.

As regards the first component of institutional quality in this study, it can be concluded that government quality has a significant direct positive effect on individual well-being. This result seems to be robust throughout the ESS area. Therefore, hypothesis 2a is supported. This positive relationship between government quality and subjective well-being is in line with the expectations based on available literature. Government quality can positively influence life satisfaction through economic growth, good governance and social policy (Lapinski, Riemann, Shapiro, Stevens, & Jacobs, 1998; Ott, 2010). In other words, the government plays a significant role because it has “the capability to mobilize resources, implement policies, and establish institutions, all of which can influence quality of life” (Kim & Kim, 2012, p. 895).

In contrast to what I anticipated based on the literature review, although not significant, the results of the OLS regression show that dependent employees slightly benefit more from government quality compared to the self-employed. Therefore, hypothesis 2b is rejected. One possible explanation for this is that dependent employees may benefit more from the design of the tax system, thus increasing their subjective well-being (Elert et al., 2017). Another explanation would be that paid employees are able to make full use of the system’s benefits pertaining to, for example, disability, sickness and parental leave (Elert et al., 2017). It is also likely that institutional environment, influenced by the quality of government, stimulates competition which ultimately drive large firms to improve employee working conditions (Fritsch et al., 2019a).

(29)

fact that government quality comprises several aspects that can influence an individual’s happiness. As mentioned previously, at present “quality of government” does not have a universal definition (Rothstein & Teorell, 2008). The term can include factors related to, for instance, corruption, property rights protection, institutional environment and government intervention.

Interestingly, it seems that there is no difference in the level of well-being when self-employed and dependent employees are extremely satisfied with the government quality (= 10 on the government quality scale). An explanation for this could be the impact of welfare state policies on subjective income insecurity (Anderson & Pontusson, 2007) which is associated with individual well-being. According to Annink, Gorgievski, and Den Dulk (2016) and based on Hemerijck (2013) institutions are “important providers of social security and may act as stabilizers for the consequences of economic insecurity” (p. 648). Hence, self-employed may feel more secure by knowing that they can rely upon unemployment benefits in case their earnings should prove insufficient. In fact, a recent study provided evidence for the positive effect of unemployment allowances on subjective well-being due to reduced income insecurity (Annink et al., 2016). Similarly, financial assistance in the form of unemployment pay and pensions can lead to higher subjective well-being of employees. Taken together, the same level of predicted well-being among the self-employed and dependent employees in case of perfect government quality may be attributed to an effective system of social support.

As can be taken from Table 4, the robustness check shows that the relationships between the variables work differently in the selected countries. For example, the effect size of the significant positive effect of government quality on subjective well-being is much larger in Ukraine and Bulgaria than in Denmark, the Netherlands and Spain. From there, it appears that good government is more important for countries with poorer governance and lower incomes (Helliwell & Huang, 2008). Baltatescu (2009)concluded that “in less richer countries from Eastern Europe the performance of the government has more influence on individual well-being and, in turn, on individual happiness than in more economically developed countries” (p. 20). This can be traced back to the political culture and increased welfare dependency in post-communist countries, which Arts, Hermkens, and van Wijk (1995) justify with people’s experiences with an equalitarian regime. Another noteworthy finding from the robustness check is that in Spain the interaction effect between self-employment and government quality proved to be significant whereas in the other countries (Denmark, the Netherlands, Ukraine, Bulgaria) this effect is insignificant. One explanation for this might be the differences in institutional conditions for entrepreneurs between countries. In this regard, Fritsch et al. (2019a) found that entrepreneurs’ well-being differs across institutional regimes. Therefore, it can be expected that government quality has varying effects on subjective well-being among the self-employed in different institutional regimes.

(30)

consistent with earlier studies. For example, Su et al. (2020) shows that trust, which serves as an informal institution, positively influences entrepreneurs’ subjective well-being. In a similar vein, trust in the legal system, government and police has been found to increase life satisfaction (Hudson, 2006). Trust in the government rests on its good performance in different areas, for instance welfare provision, and on the belief of being treated fairly by the authorities (Hough, Jackson, Bradford, Myhill, & Quinton, 2010).

Contrary to the assumptions of hypothesis 3b, stating that self-employed draw more satisfaction from institutional trust, the results of this study suggest that it is the dependent employee that profits significantly more from institutional trust. Thus, hypothesis 3b is rejected. One possible explanation for this unexpected but significant finding is the generalised trust on the welfare system, which is often designed for the benefit of employees (Henrekson, 2005). Consequently, dependent employees face higher opportunity costs of leaving their secure job to become self-employed. In fact, labour market regulation can increase the benefits of employment relative to self-employment (Van Stel, Storey, & Thurik, 2007). Trust in the legal system regarding severance pay and wide-ranging employment protection regulation can help reduce worries related to the risk of being laid off, ultimately increasing subjective well-being of employees.

Another explanation is that institutional trust may be related to trust in society’s ability to provide security and predictability in life (Koivusilta, 2018). The need for predictability is particularly ascribed to regular employment. Individuals choose regular employment over self-employment for stability and a relatively more secure income. Liu, Gao, and Huang (2019) state that trust in the government “implies citizens’ acknowledgment of its good will and of its capacity for governance— which will, in turn, enhance their sense of belonging to communities, a feeling of safety in the social environment, and a perception of happiness and satisfaction with life” (p.975-976). Accordingly, dependent employees may significantly derive more satisfaction from institutional trust compared to self-employed.

Furthermore, it is possible to establish a link between institutional trust and employee engagement, which increases employee job satisfaction. Earlier studies have raised awareness for the positive association between institutional trust and compliance with laws, regulations and civic duty (Letki, 2006). Under the assumption that institutions promote Corporate Social Responsibility (CSR) and organisations formally acknowledge and engage in CSR initiatives, employee engagement, highly driven by CSR (Gross & Holland, 2011), can develop an emotional attachment towards the place of work or community which ultimately increases employee job satisfaction. Indeed, Galaskiewicz (1991) proposes that normative or cultural institutions can induce corporate actors to act in a socially responsible manner.

(31)

Shleifer, and Vishny (1996), generalised trust reduces the need of opportunistic behaviour within companies, therefore, complete contracts are less necessary. This allows employees to have more autonomy at work which, in turn, can lead to more creativity and more entrepreneurial spirit (Elert et al., 2019). Generalised trust can influence employees to take a more entrepreneurial attitude, which has a positive impact on their subjective well-being. From the point of view of the entrepreneur, the self-employed profits from institutional trust by decreasing the possibility of opportunistic behaviour, thereby reducing transaction costs and uncertainty which should positively feedback into their perceived well-being. In sum, in case of complete institutional trust, both groups may experience the same level of well-being as they can fully engage in entrepreneurial activities, leading to greater self-determination and happiness.

5.2. Implications

The review by Shepherd et al. (2019) found that over the past two decades the main dependent variable used in entrepreneurship studies has been firm performance. However, an increasing number of scholars have drawn attention to other dependent variables such as well-being (Shepherd et al., 2019). Hence, this study reacts to the request of Wiklund, Nikolaev, Shir, Foo, and Bradley (2019) to investigate the subjective well-being of entrepreneurs as an important dependent variable. In particular, it extends entrepreneurship literature by examining the role of government quality and institutional trust, as indicators of institutional quality, in improving the well-being of the self-employed. This study aims to answer the question whether entrepreneurship relates to higher subjective well-being among the self-employed (in comparison to dependent employees) and to what extent institutional quality moderates this relationship. Moreover, it aims to contribute to the current state of knowledge by analysing a large-scale dataset which includes respondents from 32 countries with varying socio-political attitudes and public perceptions.

(32)

whereas necessity entrepreneurship does not affect economic development (Acs, 2006). Moreover, current research on self-employment and freelancing shows that not every entrepreneur and freelancer intend to hire employees (Bögenhold & Klinglmair, 2016). Thus, policy makers should carefully consider the creation and implementation of new regulations which ultimately benefit the society as a whole. On this note, it is important to stress that “policymakers must not lose sight of the long-term goal of liberalization to promote entrepreneurship, innovation, and growth” (Elert et al., 2017).

5.3. Limitations

Although this study is based on a large-scale dataset covering many European countries and several industries, it is also subject to limitations that might limit the applicability of findings. First, the constructs used in this research were measured with single items which may not sufficiently assess the underlying themes. Secondly, the subjective well-being of an individual depends on various aspects, including individual-level and country-level factors. Consequently, the main variables included in this research (self-employment, government quality, institutional trust) only partly explains a person’s subjective well-being. Thirdly, due to the fact that this study did not distinguish between different categories of self-employed, for example, opportunity-driven entrepreneurs vs. necessity-driven entrepreneurs or self-employed with employees vs. self-employed without employees, the outcome of this study might be different if heterogeneity of the self-employed was considered. Lastly, this study did not include GDP and income inequality in the analysis. However, these macro factors could influence self-employment and subjective well-being (Fritsch et al., 2019a) and therefore also alter the results.

5.4. Further Research Suggestions

(33)

scholars could explore the heterogeneity of the self-employed by examining the well-being of, for example, entrepreneurs with and without employees or opportunity-driven and necessity-driven entrepreneurs, and how the institutional framework moderates these relationships. Finally, the emphasis of this research lies in the macro factors that affect perceived well-being, nonetheless micro factors are also of paramount importance. Thus, future studies should assess the role of micro factors such as social support and psychological capital (e.g., optimism) in explaining the well-being among the self-employed.

6. Acknowledgements

(34)

7. References

Acs, Z. (2006). How Is Entrepreneurship Good for Economic Growth? Innovations: Technology,

Governance, Globalization, 1(1), 97–107.

Alam, A., Uddin, M., & Yazdifar, H. (2019). Institutional determinants of R&D investment: Evidence from emerging markets. Technological Forecasting and Social Change, 138, 34–44.

Anderson, C. J., & Pontusson, J. (2007). Workers, worries and welfare states: Social protection and job insecurity in 15 OECD countries. European Journal of Political Research, 46(2), 211-235.

Annink, A., Gorgievski, M., & Den Dulk, L. (2016). Financial hardship and well-being: A cross-national comparison among the European self-employed. European Journal of Work and Organizational

Psychology, 25(5), 645-657.

Anokhin, S., & Schulze, W. S. (2009). Entrepreneurship, innovation, and corruption. Journal of Business

Venturing, 24(5), 465-476.

Arrow, K. J. (1972). Gifts and exchanges. Philosophy and Public Affairs, 1(4), 343–362.

Arts, W. A., Hermkens, P., & van Wijk, P. (1995). Justice Evaluation of Income Distribution in East and West. In J. R. Kluegel, D. S. Mason, & B. Wegener (Eds.), Social Justice and Political Change:

Public Opinion in Capitalist and Post-communist States (pp. 131-150). New York: Aldine de Gruyter.

Avnimelech, G., Zelekha, Y., & Sharabi, E. (2014). The effect of corruption on entrepreneurship in developed vs non-developed countries. International Journal of Entrepreneurial Behavior &

Research, 20(3), 237-262.

Baltatescu, S. (2009). Differential Effects of Interpersonal and Political Trust on Happiness and Life Satisfaction. SSRN Electronic Journal, 1-24.

Baumol, W. J. (1990). Entrepreneurship: Productive, Unproductive, and Destructive. Journal of Political

Economy, 98(5, Part 1), 893-921.

Benz, M. S., & Frey, B. S. (2008a). Being Independent is a Great Thing: Subjective Evaluations of Self-Employment and Hierarchy. Economica, 75(298), 362-383.

Benz, M. S., & Frey, B. S. (2008b). The value of doing what you like: Evidence from the self-employed in 23 countries. Journal of Economic Behavior & Organization, 68(3-4), 445-455.

Binder, M., & Coad, A. (2013). Life satisfaction and self-employment: a matching approach. Small

Business Economics, 40(4), 1009–1033.

Binder, M., & Coad, A. (2016). How Satisfied are the Self-Employed? A Life Domain View. Journal of

Happiness Studies, 17(4), 1409–1433.

Bjørnskov, C., Dreher, A., & Fischer, J. A. (2010). Formal institutions and subjective well-being: Revisiting the cross-country evidence. European Journal of Political Economy, 26(4), 419–430. Blanchflower, D. G. (2000). Self-employment in OECD countries. Labour Economics, 7(5), 471–505. Block, J., & Koellinger, P. (2009). I Can’t Get No Satisfaction-Necessity Entrepreneurship and