Journal of Evolutionary Economics

, Volume 17, Issue 2, pp 133–160 | Cite as

Uncertainty avoidance and the rate of business ownership across 21 OECD countries, 1976–2004

  • Sander Wennekers
  • Roy Thurik
  • André van Stel
  • Niels Noorderhaven
Open Access
Regular Article

Abstract

Persistent differences in the level of business ownership across countries have attracted the attention of scientific as well as political debate. Cultural as well as economic influences are assumed to play a role. This paper deals with the influence of cultural attitudes towards uncertainty on the rate of business ownership across 21 OECD countries. First, the concepts of uncertainty and risk are elaborated, as well as their relevance for entrepreneurship. An occupational choice model is introduced to underpin our reasoning at the macro-level. Second, regression analysis using pooled macro data for 1976, 1990 and 2004 and controlling for several economic variables, yields evidence that uncertainty avoidance is positively correlated with the prevalence of business ownership. According to our model, a restrictive climate of large organizations in high uncertainty avoidance countries pushes individuals striving for autonomy towards self-employment. Regressions for these 3 years separately show that in 2004, this positive correlation is no longer found, indicating that a compensating pull of entrepreneurship in countries with low uncertainty avoidance may have gained momentum in recent years. Third, an interaction term between uncertainty avoidance and GDP per capita in the pooled panel regressions shows that the historical negative relationship between GDP per capita and the level of business ownership is substantially weaker for countries with lower uncertainty avoidance. This suggests that rising opportunity costs of self-employment play a less important role in this cultural environment, or are being compensated by increasing entrepreneurial opportunities.

Keywords

Business ownership Uncertainty avoidance Cross country study Comparative analysis of economies Cultural economics Entrepreneurship 

JEL Classification

P52 Z1 M13 O11 O57 

1 Introduction

The prevalence of business ownership, expressed as the percentage of owner/managers of incorporated and unincorporated businesses within the labor force, differs strongly between countries. Even within the relatively homogeneous subset of the world’s economically most developed nations (the OECD member countries), the diversity is considerable. In Greece, approximately one in five in the labor force is a (non-agricultural) business owner, whereas in Finland approximately one in 14 operates a business of his own (average rates 1972–2004; see e.g. van Stel 2005; Thurik and Wennekers 2004). A well-known approach explains this disparity by differences in prosperity (Kuznets 1971). The richer the country, the fewer business owners there are. However, the first cracks in this negative relationship appeared in the late 70s (Blau 1987; Acs et al. 1999). In fact, the negative relationship between prosperity and business ownership now seems to be breaking down in several (but not all) of the most prosperous countries, as they have shown a resurgence of business ownership rates in the past decades (see Carree et al. 2002; van Stel and Carree 2004, for an analysis of the relationship between economic development and business ownership, and Carree and Thurik 2003, for a literature survey). Moreover, the dominance of economic variables explaining business ownership rates has been questioned, and other explanatory factors, such as culture, have been brought forward (Hofstede et al. 2004).

Slowly, data material has become available showing that business ownership rates follow some U-shaped path when related to the level of economic development (Thurik and Wennekers 2004). The switch between the downward phase of this U-shape and the upward one has to do with the changing role of entrepreneurial activities. The role of entrepreneurship has changed dramatically, fundamentally shifting between what Audretsch and Thurik 2001) introduced as the model of the managed economy and that of the entrepreneurial economy. In particular, Audretsch and Thurik (2001) argue that the model of the managed economy is the political, social and economic response to an economy dictated by the forces of large-scale production, reflecting the predominance of the production factors of capital and (unskilled) labor as the sources of competitive advantage (see also Audretsch and Thurik 2004). By contrast, the model of the entrepreneurial economy is the political, social and economic response to an economy dictated not just by the dominance of the production factor of knowledge—which can be identified as replacing the more traditional factors as the source of competitive advantage—but also by a very different, but complementary, factor: entrepreneurship capital, or the capacity to engage in and generate entrepreneurial activity. By and large, countries first move from a predominantly rural economy with a high level of business ownership to an industrial one where scale economies dominate, and then again to a service economy where small scale entrepreneurial activities are essential in many industries (see Wennekers et al. 2005, for an example of the U-shape using data material of the Global Entrepreneurship Monitor).

Differences in the business ownership rate between countries seem to be persistent despite the U-shaped path that appears driven by the level of economic development. There is a general intuition that cultural rather than economic variables play a role in explaining these differences, since cultural aspects are relatively time invariant (Noorderhaven et al. 2004). The present paper investigates the role of uncertainty avoidance. Elsewhere the role of variables such as post-materialism and dissatisfaction is studied (Uhlaner and Thurik 2007; Noorderhaven et al. 2004).

At the individual level, the decision to become a business owner can be viewed as the outcome of a process of occupational choice. This approach views agents as utility maximizers making an occupational choice decision—to become employee or business owner—on the grounds of the utility associated with the expected returns from the two activity types.1 Personal characteristics2 as well as cultural, institutional and economic conditions will influence these individual choices. An aggregation of these occupational choices at the level of countries shows the cumulative and interactive influence of the different determinants (Verheul et al. 2002). In the present paper, we will focus on a specific cultural determinant of business ownership, viz., uncertainty avoidance, which to date has received only scant attention.

Our first research question considers the concepts of uncertainty and risk and the relevance of cultural attitudes towards uncertainty for the occupational choice with respect to business ownership. What effects of uncertainty avoidance on the choice for business ownership can be assumed to exist at the individual level, and how does this influence work at the country level? Our second question pertains to the direct influence of uncertainty avoidance on the prevalence of business ownership at the country level. Are differences in business ownership rates at the country level related to differences in uncertainty avoidance? Our third question deals with an indirect role of uncertainty avoidance through an influence on the relationship between GDP per capita and business ownership. Do differences in uncertainty avoidance alter the trade-offs between the opportunity costs and benefits of entrepreneurship in relationship to the level of economic development, and hence indirectly affect business ownership rates?

2 Uncertainty, risk and entrepreneurship

2.1 Some classical and neoclassical views

Since the publication in 1921 of Knight’s dissertation Risk, Uncertainty and Profit, it has become common usage in the social sciences to distinguish between risk and (true) uncertainty (van Praag 1999: 322). Uncertainty is a basic fact of life. We speak of uncertainty when ‘anything might happen.’ Relevant examples in the economic domain are new inventions and changing consumer preferences. Basically, these are unique events. Hence there is no statistical basis for calculating a probability. Risk is a special case of uncertainty. It relates to ‘disagreeable’ events for which past instances may be assembled and analyzed, such as fire-damage or the insolvency of debtors. According to the Oxford Concise Dictionary, tenth edition, 1999, risk is “the possibility that something unpleasant will happen.” Risk is often expressed in a percentage or probability and, accordingly, is to some extent insurable.

According to Knight, the entrepreneur’s main function is bearing the real uncertainty by making judgmental decisions in the face of incalculable and uninsurable business hazards (van Praag 1999: 322–323).3 Knight’s writings present an elaboration and generalization of Cantillon’s views on entrepreneurship that were originally published in 1755 and in which the main entrepreneurial function is arbitrage between supply and demand. “As Cantillon describes it, entrepreneurs buy at a certain price to sell again at an uncertain price, with the difference being their profit or loss” (Hébert and Link 1989: 42). Most (neo)-classical authors, including Say and Marshall, view entrepreneurs as being responsible for risk-bearing (van Praag 1999: 327). Later authors on entrepreneurship, particularly those in the (neo)-Austrian tradition (such as Kirzner), emphasize the entrepreneurial quality of perception of opportunities in the face of uncertainty.

By contrast, Schumpeter (1934) in his well-known Theory of Economic Development (reprinted in Swedberg 2000: 58) emphasizes the innovative function of the entrepreneur, the person who introduces ‘new combinations’ of productive means. Schumpeter’s view “disposes of the conception of the entrepreneur as risk bearer.” In a footnote, Schumpeter continues: “Risk obviously always falls on the owner of the means of production, ...., hence never on the entrepreneur as such.” Finally, Schultz (1975) defines “entrepreneurship as the ability to deal with disequilibria, rather than the ability to deal with uncertainty” (Hébert and Link 1989: 46). For Schultz, the bearing of risk is involved in entrepreneurship but it is “not a unique attribute of entrepreneurs.”

In neoclassical economics, the role of entrepreneurship is limited to the entry that follows profit opportunities (Carree and Thurik 1995). Neoclassical economics suggests that there are a set of possible outcomes and a set of probabilities that each of these outcomes will actually occur (Varian 1992). Then, a distinction is made between risk and uncertainty. The distribution of probabilities says something about the amount of risk. If the probabilities are not known, the term ‘true uncertainty’ is used. In neoclassical economics, usually, the probabilities are assumed to be known. With regard to entrepreneurship and entry, the profit opportunities are supposed to be known and accessible to everybody. Therefore, pure uncertainty is commonly disregarded (Choi 1993; Wubben 1993).

Economists such as Knight and Keynes and economic schools such as the Austrians and the Post-Keynesians have given uncertainty more emphasis (Wubben 1993). They define uncertainty in similar terms, but state that “especially entrepreneurs do not know the full range of outcomes nor their possibilities of occurring” (Lachmann, in Wubben 1993).

2.2 Contemporary views on risk-attitudes of entrepreneurs

The topic of risk (i.e. chance of failure) has remained current in more recent academic literature on entrepreneurship. Kihlstrom and Laffont (1979) emphasize that individuals differ in ‘risk aversion.’ In their model, “more risk averse individuals become workers while the less risk averse become entrepreneurs.” Likewise, Iyigun and Owen (1998) model the occupational choice between ‘inherently risky entrepreneurial ventures’ and relatively ‘safe’ alternatives such as professional activities.

McGrath et al. (1992) compare values, including attitudes towards risk and failure, of entrepreneurs (founder-managers of stand-alone businesses that were at least 2 years old and employed at least one other person) and non-entrepreneurs in eight nations. Entrepreneurs were found to agree more often with statements such as ‘start-up means risk but also excitement.’ whereas non-entrepreneurs agreed more to ‘failure means losing face/respect.’ van Praag (1996) investigates which abilities and attitudes predispose individuals to entrepreneurship. In a sample of 1,763 economically active (Dutch) adults in their early 50s in 1993, more risk averse individuals were found to have a significantly smaller probability of being a business owner or having been one in the past.

Uncertainty is particularly relevant for start-up entrepreneurs because they cannot know the full range of possible outcomes (Bhide 1994). New business founders thus are often unable to calculate their future profits. For example, someone who plans a new outlet of an existing franchise chain might have a fair estimate of its success given the experiences with previous outlets. For founders of new businesses, or more generally for entrepreneurs who introduce an innovation, this does not hold.

2.3 Synthesis of micro-economic views

Uncertainty is a concept that is central to entrepreneurship, as emphasized by eminent economists such as Cantillon, Mangoldt, Knight and Keynes (Hébert and Link 1989; Ekelund and Hébert 1990). Without uncertainty, entrepreneurship would be unnecessary. In the East European socialist planning economies, entrepreneurship was unnecessary and sometimes considered as criminal because a system of complete planning was assumed to result in optimal resource allocation. However, since uncertainty is a fact of economic life, entrepreneurs are needed to arbitrage, to take risks and to innovate (van Praag 1996; Wennekers and Thurik 1999). Entrepreneurs are considered to be the primary agents dealing with uncertainty in the economy. Entrepreneurs are called for in the fast changing economic reality of today’s society (Audretsch and Thurik 2000, 2001). Hébert and Link (1989: 47) attempt to synthesize the many diverging views. Their ‘synthetic’ definition of entrepreneurship incorporates (dealing with) uncertainty, risk, perception of profit opportunities, innovation and change.

Uncertainty is a wider concept, encompassing risks and opportunities as well as distinguishing between degrees of uncertainty. These dimensions are elaborated in Table 1. Across the rows of the table there is a dichotomy distinguishing between possible unpleasant outcomes (‘risks’) and possibilities of business success (‘opportunities’). Next, the columns represent different degrees of uncertainty. Column (1) describes the relatively low uncertainty when the possible outcomes and their probabilities are known. A case in point is selling fire insurance or starting a new outlet of an existing franchise. Column (2) refers to medium-high uncertainty in the sense that there is only a notion of possible outcomes and probabilities, such as may be the case with many new business start-ups. Business founders may not be able to calculate risks and expected profits, but they will often have a perception of the risks, opportunity costs and profit opportunities of their venture. Column (3) describes the ‘true’ uncertainty of future loss or profit, inherent to launching a radical innovation or to investing financial capital in fundamental research.
Table 1

Uncertainty encompassing risk and opportunity

Risks versus opportunities

Degree of uncertainty

(1) Possible outcomes and their probabilities are known

(2) There is a notion of possible outcomes and probabilities

(3) Anything might happen

Possibility of damage, loss or failure

Calculated risks

Perceived risks and opportunity costs

True uncertainty of loss or failure

Opportunity of profit or other business success

Expected profits

Perceived profit opportunities

True uncertainty of profit; serendipity

There is agreement that entrepreneurs (in the sense of business owners) make judgmental decisions in the face of uncertainty, reap the rewards of perceiving and utilizing opportunities and in the process also run the risk of losing their money and their reputation. There is also some consensus that entrepreneurs are less averse to risk, while alternative views hold that entrepreneurs are inherently more optimistic rather than less risk averse or dispose of relevant information reducing uncertainty and risk (Gifford 2003: 37–41).

2.4 Cultural traits with respect to uncertainty

Attitudes, such as risk aversion, pertain to individuals and may show a wide variety within groups. At the ‘ecological level’ of nations, cultural traits related to these individual attitudes may be distinguished. Empirically, these traits may be derived as mean, modal or extreme values of individual observations or through a direct analysis of ‘ecological data’ (pertaining to national practices and achievements). Cultural traits represent a nation’s ‘mental programs’ that are developed in socialization processes in the family in early childhood and reinforced in schools and organizations (Hofstede 2001: xix). Accordingly, cultural traits may differ between societies.

A cultural trait that is strongly associated with individual attitudes towards risk and uncertainty is ‘uncertainty avoidance.’ According to Hofstede (2001: 146), uncertainty avoidance has to do with the extent to which societies tolerate ambiguity. A culture is characterized by high uncertainty avoidance when its members feel threatened by uncertain or unknown situations. People in these cultures “look for structure in their organizations, institutions and relationships, which makes events clearly interpretable and predictable” (Hofstede 2001: 148.) In countries with lower uncertainty avoidance, “not only familiar but also unfamiliar risks are accepted, such as changing jobs and starting activities for which there are no rules.” Low uncertainty avoidance thus implies “willingness to enter into unknown ventures” (Hofstede 2001: 164). Hofstede operationalizes uncertainty avoidance using three survey questions about whether employees feel “company rules should not be broken even when the employee thinks it is in the company’s best interests,” about their personal expected job stability and about how often they feel nervous or tense at work.

2.5 Relevance of uncertainty avoidance for explaining the business ownership rate

2.5.1 Direct effect of uncertainty avoidance

A micro-economic model of occupational choice is introduced to clarify the ways in which uncertainty avoidance may have an impact on the prevalence of business ownership at the country level. In this model, the individual choice between self-employment and wage-employment depends on a personal assessment and utility valuation of the expected material and immaterial rewards of these occupational alternatives, while taking the perceived risks into account (see Wennekers 2006). For simplicity, we operationalize material rewards as the expected personal income generated by self-employment (E(I)SE), compared with the wage one expects to earn in a job (E(I)WE). We reduce the immaterial rewards of self-employment to a gain in autonomy compared with the degree of independence that an individual will experience when working as an employee.

Below, we summarize the model in a schematic manner:
$$ {\text{OC}}^{ * } _{{{\text{SE}},i}} = \alpha _{i} U_{{{\text{MR}},i}} + {\left( {1 - \alpha _{i} } \right)}U_{{{\text{IR}},i}} ;\quad 0 < \alpha _{i} < 1 $$
(1)
$$ U_{{{\text{MR}},i}} = \beta _{i} {\left( {{\left( {1 - \rho _{{{\text{SE}},i}} } \right)}E{\left( I \right)}_{{{\text{SE}},i}} - E{\left( I \right)}_{{{\text{WE}},i}} } \right)};\quad \beta _{i} > 0\,;\quad 0 < \rho _{{{\text{SE}},i}} < 1 $$
(2)
$$ U_{{{\text{IR}},i}} = \gamma _{i} \Delta {\text{AUT}}_{i} ;\quad \gamma _{i} > 0 $$
(3)
Where
OC* SE,i

Latent variable measuring total utility of choice for self-employment (individual i)

U MR,i

Utility of expected change in material rewards due to self-employment

U IR,i

Utility of expected gain in immaterial rewards due to self-employment

E(I) SE,i

Expected income self-employment

E(I) WE,i

Expected income wage-employment

ΔAUT i

Gain in autonomy (self-employment versus wage-employment)

α i

Parameter reflecting the relative weight in utility of material vs immaterial rewards

β i

Parameter transforming expected change in material rewards into utility

γ i

Parameter transforming expected gain in immaterial rewards into utility

ρ SE,i

Discount parameter for perceived risks of self-employment

For empirical application, an observable occupational choice variable OCSE,i might be added, where OCSE,i  = 1 (i is self-employed) when OC*SE,i  > 0 and OCSE,i  = 0 (i is an employee) when OC*SE,i  < 0. Parker (2004: 24–26) elaborates how this micro-economic model might be estimated after transformation into a probit or logit model. This is, however, not necessary for our purpose, i.e. the underpinning of a macro-economic regression model.

We assume that all parameters and variables in the model are idiosyncratic with respect to individuals, i.e. we assume that, for each individual, parameters and variable values are randomly drawn from probability distributions. In addition, we assume that attitudes towards uncertainty and risk play a role in the assessments and utility valuations of the expected material and immaterial rewards. In particular, we assume that the distributions of ρ SE,i , γ i, and E(I) SE,i are systematically influenced by the individual level of uncertainty aversion. That is, ceteris paribus, the distributions of these three parameters and variables are located more to the right or to the left, depending on the individual level of uncertainty aversion. This will be illustrated below. For simplicity let us assume that there are two groups of individuals, a group with a high uncertainty aversion level, H, and a group with a low uncertainty aversion level, L. First, it is assumed that an aversion of uncertainty causes people to perceive fewer profit opportunities and to see more risks in entrepreneurship. This causes a downward bias in their assessments of the expected income of self-employment, i.e. \( \overline{{E{\left( I \right)}}} _{{{\text{SE}},{\text{H}}}} < \overline{{E{\left( I \right)}}} _{{{\text{SE}},{\text{L}}}} \), where the overscore denotes the median value of the distribution. Second, they will also attach a lower utility to a certain expected income when they feel that higher risks are involved, i.e. \( \overline{\rho } _{{{\text{SE}},{\text{H}}}} > \overline{\rho } _{{{\text{SE}},{\text{L}}}} \). Third, it may be assumed that uncertainty averse individuals have a relatively low valuation of autonomy, i.e. \( \overline{\gamma } _{{\text{H}}} < \overline{\gamma } _{{\text{L}}} \).

This model of individual occupational choice presents several bridges to the effects of uncertainty avoidance for the macro-economic business ownership rate. First, a culture of high uncertainty avoidance may imply a higher percentage of uncertainty/risk averse individuals within the population.4 Applying our micro-economic model at the macro level, this implies lower assessments of the expected entrepreneurial income and a higher discount for perceived risks. On the other hand, countries with low uncertainty avoidance will count more individuals with entrepreneurial values who attach a higher utility to the rewards of self-employment. These countries thus have a relatively large supply of potential entrepreneurs (see Shane 1993, for indirect support of this assumption). In terms of our model, this means that there will be more people for whom the utility of the material rewards of self-employment (U MR) is positive (negative) in countries with low (high) uncertainty avoidance.5 This gives rise to the hypothesis that the prevalence of self-employment is diminished by high uncertainty avoidance (UAI+), while it is stimulated by low uncertainty avoidance (UAI−). In our section on Method, two clusters of countries will be defined.

However, there may also be an opposite effect because a culture of high uncertainty avoidance at the country level may be expected to imply a restrictive climate within existing firms and organizations. This would offer a relatively large gain in autonomy (ΔAUT) to individuals choosing self-employment. Even when there are fewer enterprising individuals in such an economy, UAI+ may push many of them towards self-employment. In terms of our model, this means that, on average, the utility of the immaterial rewards of self-employment (U IR) will be higher in countries with high uncertainty avoidance. This leads to the hypothesis that high uncertainty avoidance (UAI+) may stimulate self-employment (see Baum et al. 1993, for an analogous reasoning with respect to the effect of low individualism at the country level).

Summarizing, there are two contradicting hypotheses with respect to the direct influence of uncertainty avoidance on the supply of business owners. On average, an UAI+ culture will result in more individuals with a relatively low value of U MR, but it will also result in more individuals with a relatively high value of U IR. The overall impact of these opposite forces (i.e. the net-effect on the business ownership rate) is a subject for empirical research.

2.5.2 Indirect effect of uncertainty avoidance

Uncertainty avoidance may also have an indirect influence on the rate of business ownership, i.e. the level of uncertainty avoidance in a nation may influence the manner in which other variables determine business ownership. For example, the degree to which increasing per capita income leads to a perception of increasing opportunity costs of entrepreneurship (compared with well-paid, safe jobs) versus a perception of increasing entrepreneurial opportunities (more niches; need for autonomy) may well be dependent on the level of uncertainty avoidance. Likewise, high unemployment levels may be interpreted as a decrease of the opportunity costs associated with business ownership, and hence stimulate entrepreneurship, but also with increased likelihood of failure, and therefore negatively related with business ownership levels, depending on the degree of uncertainty avoidance.

3 Modelling the business ownership rate

The dependent variable in this study is the rate of business ownership in a nation at a certain moment in time. Our major interest is the direct and the indirect contribution of uncertainty avoidance to the variance in business ownership across nations and over time. We position our study within a broad multidisciplinary framework that is based on various strands of the entrepreneurship literature (see Verheul et al. 2002; Wennekers 2006, for a description of this framework). From this framework, we choose control variables for our regression model of the effects of uncertainty avoidance. Table 2 lists economic and demographic determinants of business ownership. Here, we not only focus on the underlying micro-economic studies of occupational choice, but also refer to surveys and empirical macro-economic investigations. The first column also contains an operationalization of the determinants used in the empirical analysis, while the final column indicates the data availability and the sources of the various variables. As we will use data for business ownership (the dependent variable in our study) for the years 1976, 1990 and 2004, and we will use a 4 year lag for the independent variables, we have aimed at collecting data for the years 1972, 1986 and 2000 for the variables in Table 2. However, when data are not available for one of these years, we use data for the closest available year. This is also indicated in the final column of Table 2.
Table 2

Major explanatory variables of the business ownership rate

Economic variables (operationalization)

Relevant literature

Data availability (years); Source

Level of economic development (GDP per capita)

Kuznets 1971; Lucas 1978; Schultz 1990; Yamada 1996

1972, 1986, 2000; OECD National Accounts

Share of services (employment in services divided by total labor force)a

van Stel and Carree 2004

1972, 1986, 2000; OECD National Accounts

Entrepreneurial income relative to the wage rate (labor income share)b

Parker 2004

1972, 1986, 2000; own calculations, based on OECD National Accounts

Unemployment rate

Evans and Leighton 1989; Meager 1992

1972, 1986, 2000; OECD Main Economic Indicators

Social security entitlements (unemployment replacement rate)

Parker and Robson 2004

1972, 1986, 2000; OECD Benefits and Wages

Income disparity (Gini coefficient)c

Ilmakunnas et al. 1999

mid-1980s; 2000; OECD

Cost of capital (long term interest rate)d

Parker 2004

1991; 2000; OECD Economic Outlook 78 database

Assets; collateral (house prices)

Evans and Jovanovic 1989; Evans and Leighton 1989; Parker 2004

Insufficient data on house prices available

Demographic variables

Age composition (number of people aged 25–39 years divided by number of people aged 25–64)

Storey 1994; Blanchflower et al. 2001

1971, 1984, 1991; US Census Bureau, International Data Base

Population density

Audretsch and Keilbach 2004; Bais et al. 1995

1972, 1986, 2000; OECD Labour Force Statistics (population), Grote Winkler Prins encyclopaedia (area)

Educational levels (gross enrollment rates for secondary and tertiary education)

Delmar and Davidsson 2000; Uhlaner and Thurik 2007

1970, 1985, 2000; World Bank EdStats data base

Female labor participation

Delmar and Davidsson 2000; Verheul 2005

1972, 1986, 2000; OECD Labour Force Statistics

aThe services sector is broadly defined here, it contains the sectors Wholesale and retail trade, restaurants and hotels; Transport, storage and communication; Finance, insurance, real estate and business services; and Community, social and personal services.

bThe labor income share has been corrected for the imputed wage income of self-employed individuals. To make the variable better fitting with the (non-agricultural) business ownership rate, the labor income share has been computed excluding the agricultural sector.

cNo data available for 1972. Missing values for Belgium and Spain.

dNo data available for 1972.

3.1 Level of economic development

It has been observed in various studies that the business ownership rate decreases as economies become more developed (Schultz 1990; Yamada 1996; see Carree et al. 2002, for an overview). Economic development is usually measured by per capita income, but it is also reflected in the average wage rate. In the present discussion, we will include both per capita income and the wage rate.

A low level of prosperity usually coincides with a low wage level, implying little pressure to increase efficiency or the average scale of enterprise. Small firms in agriculture, crafts and retail trade are therefore dominant in such an economy. A major route for ambitious wage earners to increase their income, then, is to set up shop and become an entrepreneur.

Subsequently, economic development leads to a rise in wages, which stimulates enterprises to work more efficiently and to reap economies of scale and scope (Chandler 1990). Also, a declining share of agriculture and an increasing share of manufacturing diminish the opportunities for self-employment. At the supply side of the labor market, an additional effect of rising wage levels is an increased attraction of wage-employment, increasing the opportunity cost of self-employment (Lucas 1978). Iyigun and Owen (1998) argue that with economic development the “safe” professional earnings will rise and fewer individuals will be willing to risk becoming business owners.

In recent decades, statistical evidence points at a possible reversal of the negative relationship between real per capita income and self-employment at an advanced level of economic development. With rising per capita income, a differentiation of consumer demand for both goods and services creates new market niches and provides opportunities for business ownership. At the supply side of entrepreneurship, social psychology hypothesizes a hierarchy of human motivations, ranging from material needs to self-realization (Maslow 1970). By providing autonomy, entrepreneurship may become a more attractive occupational choice at higher levels of income.

However, this reversal is not universal, as witnessed by the continued decline of business ownership in some highly developed economies such as France and Japan (Verheul et al. 2002). Two opposing forces may be at play here: while rising wage levels will continue to increase the opportunity costs of self-employment, differentiation of consumer wants will create more opportunities for new enterprises. Occupational choices in countries with low uncertainty avoidance may be influenced more strongly by the latter effect than by the first. In high uncertainty avoidance countries, it may be the other way around. Consequently, at advanced levels of economic development, we conjecture a differential impact of increasing prosperity in low (UAI−) and in high (UAI+) uncertainty avoidance countries. In UAI+ countries, the negative relationship between the level of prosperity and the self-employment rate will be undiminished across economic development. In UAI− countries, the negative relationship between prosperity and the self-employment rate will be weaker or even reverse after a certain turning point.

3.2 Share of services

At the high end of economic development, the share of the services sector usually increases relative to that of manufacturing. On average, self-employment rates in services are considerably higher than in manufacturing (see van Stel and Carree 2004). It requires only relatively modest investments to set up an enterprise in many services. We assume that an increasing share of the services sector will increase the business ownership rate.

So far, empirical research on this compositional effect on the business ownership rate is scant. Wennekers and Folkeringa’s (2002) investigation of long-term trends in the business ownership rate of the Netherlands showed that sector shifts clearly played a part, but within-sector trends turned out to be even more important. For an analogous conclusion about trends in the firm size distribution of six large OECD countries, see Loveman and Sengenberger (1991).

3.3 Relative earnings of self-employment

In a previous section of the present paper we discussed a model of individual occupational choice. This model assumes that relative earnings of self-employment versus wage-employment affect occupational choice. Ceteris paribus, the better the prospects of entrepreneurial income as compared to the wage income of employees, the more people will be attracted to self-employment.6

Parker (2004: 68–70) presents a survey of the empirical evidence for this relationship. Various investigations using a structural probit model, including relative earnings as determinants of individual occupational choice, give mixed results. Two time-series studies at a more aggregate level, also cited by Parker, find a significant contribution of aggregate earnings differentials to explaining trends in the UK self-employment rate.

In our empirical analysis, we use the macro-economic labor income share as a (reverse) proxy for the (expected) entrepreneurial income relative to the wage income. The labor income share is defined as the sum of wages including ‘imputed wage income of self-employed persons,’ expressed as a fraction of total income. This is admittedly a rough proxy.

The relationship between relative earnings and the business ownership rate may be moderated by the degree of uncertainty avoidance. In an occupational choice perspective, the weighing of expected entrepreneurial and wage income against one another also includes an assessment of the risks involved. In UAI− countries, the entrepreneurial risks will be viewed more lightly and accordingly the effect of relative earnings may be stronger than in UAI+ countries.

3.4 Unemployment

(The threat of) unemployment is a factor diminishing the opportunity costs of self-employment, particularly when unemployment benefits are low relative to (minimum or average) wages. However, when structural unemployment is very high, this may indicate bleak business opportunities and discourage business ownership (Hamilton 1989; Meager 1992). Exactly where the negative influence of rising unemployment begins to outbalance the positive effect of decreasing opportunity costs depends on a perception of uncertain future events, and may therefore be related to the level of uncertainty avoidance in a country. Hence we expect the positive effects of unemployment to dominate in UAI− countries, and the negative effects in UAI+ countries.

3.5 Social security entitlements

High social security entitlements for employees contribute to the opportunity costs of entrepreneurship, and may be expected to have a negative influence on the business ownership rate. This has been confirmed in several empirical investigations (Ilmakunnas et al. 1999; Parker and Robson 2004; Hessels et al. 2006) reporting negative effects on self-employment of employers’ social security contributions and/or the unemployment benefit replacement rate.

3.6 Income disparity

Some scholars hypothesize that an equal income distribution may limit the required asset accumulation facilitating enterprise formation, while income disparity may be favorable to entrepreneurship (Ilmakunnas et al. 1999). At the lower end of the income distribution, inequality may act as a push factor to enter self-employment. Additionally, on the demand side of entrepreneurship, income disparity is likely to create a more differentiated demand for goods and services. Empirical research by Ilmakunnas et al. suggests that income inequality positively influences the rate of self-employment, although reversed causality cannot be ruled out.

3.7 Financial variables

Starting and running a business requires financial capital. This capital is needed to purchase or rent the premises, to invest in equipment and/or vehicles, to purchase raw materials, to finance market research and advertising and to advance wages. The need for financial capital differs strongly with the line of business. Financial resources for business start-ups are often derived from self-financing (including savings, gifts, inheritances and lottery wins). Additionally, informal investors, mortgage loans, commercial credit and bank loans and (very rarely) venture capital can also be a source of start-up capital (Bygrave and Hunt 2005).

Capital constraints, often related to lack of assets or collateral, may create serious impediments for business start-ups (Blanchflower and Oswald 1998; Evans and Leighton 1989; van Praag 1996) and for young and growing firms (Chittenden et al. 1996; LeCornu et al. 1996). An influential paper by Evans and Jovanovic (1989) has stimulated research on credit rationing. For a survey of this literature, see Parker (2004). While a clear conclusion on the prevalence of credit rationing seems yet out of reach, there is ample evidence that self-employment rates are positively related to personal wealth (real estate and other assets).

Finally, the direct and indirect (opportunity) costs of financing a business depend on the rate of interest. Higher interest rates may be expected to have a negative effect on business ownership. Parker (2004: 104, 105) weighs the empirical evidence. In particular, several UK and US time-series studies show a significant negative effect of the interest rate on the self-employment rate.

3.8 Demographic characteristics

With respect to gender, in most surveys women are found to be less likely to be involved in either self-employment or early-stage entrepreneurial activity than are men, although the difference varies across nations (Minniti et al. 2005; Verheul 2005). Econometric analysis of a large Swedish dataset of individual business start-ups has shown a remaining ‘pure’ gender effect after correcting for other differences, such as education and previous management experience (Delmar and Davidsson 2000). A higher labor participation rate of women thus in itself means a lower overall business ownership rate in the labor force.

The role of population density at the national level is less obvious. Every local area needs a minimum supply of facilities in retail trade, repair and personal services. Therefore, thinly populated regions will have relatively many small retail outlets, workshops and service providers. Conversely, urban areas will give rise to economies of scale through which small-sized entrepreneurship in particularly retailing comes under pressure (Bais et al. 1995). On the other hand, networks and other supply side factors in urban areas are conducive to new entrepreneurship in many service industries (Audretsch and Keilbach 2004).

Education is somewhat of an anomaly. Research conducted on a Swedish sample at individual level, showing that nascent entrepreneurs attained on average a higher educational level7 than those in a control sample (Delmar and Davidsson 2000), has been reconfirmed in recent investigations across several high-income countries (Acs et al. 2004). On the other hand, macroeconomic research with respect to a static index of entrepreneurship suggests the opposite conclusion. For instance, Uhlaner et al. (2002) found that countries with a higher level of education tend to have a smaller proportion of self-employment. In a recent comparative study across 27 OECD countries, countries with a higher level of enrollment in secondary education show a lower level of early-stage entrepreneurship, while higher enrollment in tertiary education was found to have a positive effect on both early-stage entrepreneurship and total business ownership (Uhlaner and Thurik 2007).

With respect to the age composition of the population, Blanchflower et al. (2001: 686) report that, while ‘older people are more likely to be self-employed, it is younger people who say they would prefer to be self-employed.’ Earlier research also shows that people in the middle age cohorts have the highest prevalence of incumbent business owners (Storey 1994). In many countries, prevalence rates of nascent entrepreneurship are highest in the age group between 25 and 34, while, according to some research, a tendency towards start-ups at a younger age is also apparent.8 Ceteris paribus, the ageing of the population in most developed countries implies a threat to the future development of business ownership.

4 Method and data

4.1 Method

First, we investigate the direct influence of uncertainty avoidance on the business ownership rate by means of a regression analysis of pooled panel data for 21 countries in 1976, 1990 and 2004, given the influence of (4 years lagged) per capita income and some other control variables (also 4 years lagged). The control variables are chosen from Table 2 on the basis of data-availability. We assume that the samples for 1976, 1990 and 2004 are sufficiently independent to warrant pooling them in one regression. Because uncertainty avoidance was measured only once (around 1970), its role in the pooled regression analysis may be interpreted as that of a country-specific time-invariant variable. Next, we use the years 1976, 1990 and 2004 as separate samples to investigate the stability of the direct relationship over time.

Second, we explore the possible indirect influence of uncertainty avoidance on the rate of business ownership. This means that we have added an interaction term between per capita income and uncertainty avoidance to the multiple regression analysis of the pooled sample for 1976, 1990 and 2004. Finally, we repeat this regression, substituting UAI by a dummy variable representing two separate clusters of countries. In our dataset, the following 13 countries form the cluster9 of low uncertainty avoidance: Denmark, Finland, Germany, Ireland, the Netherlands, Sweden, Great Britain, Norway, Switzerland, USA, Canada, Australia and New Zealand. Another eight countries, i.e. Austria, Belgium, France, Italy, Greece, Portugal, Spain and Japan, make up the cluster with high uncertainty avoidance. By comparing these two models we hope to find indications whether the effects of uncertainty avoidance are discrete or continuous (see also Cohen and Cohen 1983).

4.2 Data

Harmonized non-agricultural business ownership rates for 23 OECD-countries are available from EIM’s COMPENDIA data base.10 These data include the owners of incorporated and unincorporated businesses but exclude unpaid family workers. The countries in COMPENDIA include 18 European countries as well as the USA, Japan, Canada, Australia and New Zealand. Data are available for the even years from 1972 onwards.

Additionally, Hofstede (2001) provides data on uncertainty avoidance for 21 of the 23 countries mentioned above.11 Uncertainty avoidance is a key variable in Hofstede’s well-known study12 of cultural dimensions across some 50 different nations and regions. The uncertainty avoidance index (UAI) was computed on the basis of the country mean scores for three different survey questions already mentioned in a previous section of the present paper. Because the surveys on which the index was based were held between 1967 and 1973, the stability of the index is a crucial aspect for our study into the rate of business ownership in the years 1976, 1990 and 2004. Hofstede (2001: 34) claims that national cultures are extremely stable over time. He argues that ‘ ... this stability can be explained from the reinforcement of culture patterns by the institutions that themselves are products of the dominant cultural value systems.’ In the long run, ‘cultures shift, but they shift in formation, so that the differences between them remain intact’ (Hofstede 2001: 255). Chapters 2 and 4 of Hofstede’s book present abundant statistical information about the stability and reliability of the uncertainty avoidance index. Our best assessment is that this index can be used for explaining national rates of entrepreneurship during several decades following the measurement of the index.

An alternative would have been to use the uncertainty avoidance data reported by the GLOBE project (House et al. 2004). We refrain from doing so for two reasons. First, Hofstede’s uncertainty avoidance index is well understood, and has been used in many previous studies. Hofstede (2001) also reports extensively on correlates of his uncertainty avoidance index with measures from over one hundred other studies. Comparable validation of the GLOBE uncertainty avoidance scales is not available. Second, there are some conceptual difficulties with the GLOBE uncertainty avoidance scales. GLOBE constructed two scales, the actual use of uncertainty avoidance mechanism in the respondent’s society (“practices”), and the desired use of uncertainty avoidance mechanisms (“values”). These two scales are negatively correlated. The GLOBE practices scale is also negatively correlated to Hofstede’s UAI scale, the GLOBE values scale positively (Sully de Luque and Javidan 2004).13 This makes the GLOBE scales difficult to interpret. The authors note that most countries with high uncertainty avoidance practices are technologically developed nations (Sully de Luque and Javidan 2004: 621). This makes the index less relevant for the current study, as we are comparing levels of business ownership across developed countries only. Both GLOBE uncertainty avoidance scales are strongly correlated with economic prosperity, the “practices” scale positively, and the “values” scale negatively (Sully de Luque and Javidan 2004: 631). Hofstede’s UAI, in contrast, is only weakly correlated to economic prosperity (Hofstede 2001: 201). Consequently, Hofstede’s index measures cultural characteristics of countries that are relatively independent of wealth, and thus this index forms a good complement to the economic indicators we also use in this study.

For the operationalization and sources of the control variables, we refer to Table 2. Besides the controls included in Table 2, we also include year dummies in our analysis. Recent decades have witnessed a worldwide diffusion of new information and communication technologies, as well as a widespread tendency towards deregulation of markets. Both phenomena have created opportunities for small scale business and new entrepreneurship. Audretsch and Thurik (2000, 2001) label this as a regime switch from ‘a managed to an entrepreneurial economy.’ We try to catch these developments using year dummies as controls in our analysis.

Table 3 presents the correlation matrix of the pooled sample for 1976, 1990 and 2004. The highest (positive) correlations among the control variables include those between per capita income, on the one hand, and the female labor share and tertiary education, on the other. Uncertainty avoidance and per capita income show a moderate degree of (negative) correlation in our sample.
Table 3

Correlation matrix

 

1

2

3

4

5

6

7

8

9

10

11

12

13

14

1. Business ownership rate

1

             

2. Uncertainty avoidance

0.49***

1

            

3. GDP per Capita

−0.24*

−0.38***

1

           

4. Labor income share

−0.42***

−0.42***

0.27**

1

          

5. Unemployment

0.31**

0.09

−0.01

−0.10

1

         

6. Female labor share

−0.21

−0.22

0.67***

0.19

0.001

1

        

7. Population density

−0.03

0.28**

0.04

0.22*

−0.01

−0.14

1

       

8. Share services

0.21*

−0.26**

0.54***

0.10

−0.02

0.18

−0.07

1

      

9. Replacement rate unempl.

−0.26**

−0.18

0.32**

0.05

0.14

0.30**

0.27**

−0.035

1

     

10. Gini index

0.71***

0.42***

−0.21

−0.35**

0.32*

−0.32*

−0.06

0.33**

−0.54***

1

    

11. Long term interest rate

0.08

−0.14

−0.56***

−0.08

0.38**

−0.36**

−0.24

−0.25

−0.04

−0.26

1

   

12. Share age group 25–39 in adult pop. (25–64 year)

0.05

−0.23*

0.42***

−0.15

0.18

0.32**

−0.15

0.49***

0.29**

−0.11

0.27*

1

  

13. Secondary education

−0.06

−0.23*

0.59***

0.13

0.18

0.64***

0.10

0.26**

0.48***

−0.21

−0.48***

0.27**

1

 

14. Tertiary education

0.06

−0.16

0.73***

−0.01

0.27**

0.66***

−0.13

0.44***

0.30**

0.11

−0.60***

0.38***

0.70***

1

N

63

63

63

63

63

63

63

63

63

37

40

63

62

62

Mean

0.11

62.9

14,762

0.77

0.056

0.39

1.20

0.42

0.25

0.29

0.074

0.43

0.96

0.36

Standard deviation

0.035

24.7

4,770

0.06

0.038

0.059

1.12

0.12

0.14

0.044

0.025

0.033

0.23

0.20

Correlations are based on a pooled sample of 1976, 1990 and 2004 (63 observations maximum).

*p < 0.10; **p < 0.05; ***p < 0.01

5 Results

5.1 Direct influence of uncertainty avoidance

Table 4 presents the regressions on the pooled sample for 21 countries in 1976, 1990 and 2004. First, we regress business ownership on uncertainty avoidance, GDP per capita and the year dummy variables. This is the ‘base model’, shown in the first column of the table. The significantly positive coefficient for uncertainty avoidance is support for Baum’s hypothesis, stating that dissatisfaction with a climate of high uncertainty avoidance in large organizations may push enterprising individuals towards self-employment. GDP per capita and the year dummies are also significant and have the expected sign. Next, we introduce the other control variables one by one. In all but one of these regressions, the coefficient for uncertainty avoidance is significantly positive. The only exception is the regression including the Gini index, which is based on 37 observations only. With respect to the significant control variables, the only counterintuitive result is the positive sign for the long term interest rate. Subsequently, as shown in the second to last column of Table 4, we regress business ownership on uncertainty avoidance while including all control variables that are significant in the previous regressions.14 Finally, the last column shows the variables that are significant in a ‘complete model.’ These are uncertainty avoidance (+), per capita income (−), the share of services (+), the unemployment replacement rate (−) and the dummy variables for 1990 and 2004 (+).
Table 4

The direct influence of uncertainty avoidance on business ownership, pooled sample 1976, 1990 and 2004 (21 countries)

 

Base model

 

 

 

 

 

 

 

 

 

 

 

 

Complete model

Constant

0.11 (7.0)

0.18 (3.8)

0.10 (4.6)

0.18 (5.0)

0.11 (7.0)

0.076 (4.5)

0.13 (7.7)

0.001 (0.0)

0.043 (0.9)

0.026 (0.6)

0.14 (4.4)

0.11 (7.0)

0.098 (1.8)

0.094 (5.0)

Uncertainty Avoidance

0.041 (3.1)

0.036 (2.9)

0.043 (3.0)

0.036 (2.4)

0.048 (3.5)

0.045 (3.9)

0.031 (2.4)

0.009 (0.4)

0.041 (2.2)

0.044 (3.6)

0.036 (2.7)

0.043 (3.3)

0.035 (2.4)

0.038 (3.3)

GDP per Capita

−0.38 (3.9)

−0.32 (2.8)

−0.32 (2.6)

−0.29 (2.6)

−0.35 (3.8)

−0.59 (7.2)

−0.39 (4.8)

−0.29 (1.8)

−0.26 (2.0)

−0.43 (4.3)

−0.37 (3.9)

−0.40 (4.1)

−0.50 (4.6)

−0.57 (7.0)

Labor income share

 

−0.089 (1.3)

            

Unemployment

  

0.13 (1.1)

           

Female labor share

   

−0.21 (2.6)

        

−0.13 (1.5)

 

Population density

    

−0.004 (1.4)

         

Share services

     

0.15 (5.4)

      

0.10 (3.1)

0.14 (4.6)

Replacement rate unempl.

      

−0.082 (3.6)

     

−0.063 (2.8)

−0.052 (2.2)

Gini index

       

0.51 (4.8)

      

Long term interest rate

        

0.78 (2.2)

     

Share age group 25–39 in adult pop. (25–64 year)

         

0.22 (2.0)

  

0.12 (1.0)

 

Secondary education

          

−0.025 (0.8)

   

Tertiary education

           

0.031 (1.1)

  

Year dummy 1990

0.023 (2.8)

0.022 (2.4)

0.015 (1.3)

0.033 (3.3)

0.023 (2.7)

0.023 (3.3)

0.032 (3.9)

  

0.018 (1.9)

0.026 (2.4)

0.019 (2.1)

0.033 (3.2)

0.029 (3.8)

Year dummy 2004

0.046 (4.3)

0.041 (3.4)

0.037 (2.9)

0.060 (4.5)

0.044 (4.2)

0.048 (6.1)

0.059 (5.9)

0.009 (0.9)

0.051 (3.3)

0.045 (4.1)

0.055 (3.3)

0.034 (2.5)

0.065 (4.9)

0.055 (7.0)

N

63

63

63

63

63

63

63

37

40

63

62

62

63

63

R 2

0.368

0.385

0.381

0.429

0.383

0.567

0.459

0.592

0.361

0.399

0.377

0.373

0.627

0.601

Dependent variable: number of non-agricultural business owners per labor force.

Absolute heteroskedasticity consistent t-values are between brackets.

Table 5 presents the regressions in three separate sample years 1976, 1990 and 2004. For each year, the left-hand column presents a regression including the control variables that were listed in the second to last column of Table 4, while the right-hand column reports significant control variables only. The main finding for the sample of 1976 is a significantly positive influence of uncertainty avoidance on the rate of business ownership. In 1990, the coefficient of uncertainty avoidance is again positive, but no longer fully significant. In 2004, no influence of UAI is found. All regressions confirm the well-known negative influence of GDP per capita.
Table 5

The direct influence of uncertainty avoidance on business ownership, separate samples 1976, 1990 and 2004 (21 countries)

 

1976

1976

1990

1990

2004

2004

Constant

0.093 (1.4)

0.060** (2.2)

0.16* (1.9)

0.14*** (3.4)

0.32* (1.8)

0.33*(2.0)

Uncertainty avoidance

0.054* (1.9)

0.063*** (3.1)

0.029 (1.3)

0.032 (1.4)

−0.007 (0.4)

−0.0095 (0.6)

GDP per capita

−0.39 (1.3)

−0.52** (2.7)

−0.63*** (3.0)

−0.58*** (3.1)

−0.62*** (4.4)

−0.59*** (4.0)

Female labor share

−0.15 (1.3)

 

0.041 (0.3)

 

−0.59 (1.5)

−0.67* (2.0)

Share services

0.11 (1.5)

0.15** (2.2)

0.16** (2.2)

0.13** (2.7)

0.047 (0.7)

 

Replacement rate unemployment

−0.030 (0.8)

 

−0.093 (1.7)

−0.10** (2.3)

−0.056 (1.5)

−0.072** (2.3)

Share age group 25–39 in adult pop. (25–64 year)

0.066 (0.6)

 

−0.091 (0.5)

 

0.40 (1.7)

0.50*** (3.2)

N

21

21

21

21

21

21

R 2

0.614

0.558

0.749

0.747

0.686

0.672

Dependent variable: number of non-agricultural business owners per labor force.

Absolute heteroskedasticity consistent t-values are between brackets.

*p < 0.10; **p < 0.05; ***p < 0.01.

The main outcome of Table 5 is that the positive effect of uncertainty avoidance fades away over time. Our interpretation is that the advent of the entrepreneurial economy in recent years, as discussed in our Introduction, has created new pull factors mobilizing the relatively abundant supply of potential entrepreneurial capital in countries with low uncertainty avoidance. So Baum’s push hypothesis for high uncertainty avoidance and Shane’s pull hypothesis for low uncertainty avoidance may now be equally valid, effectively countervailing one another in the regression for 2004. Another explanation could be that the measurement of uncertainty avoidance (which was carried out around 1970) has lost some of its validity 30 years onwards, but the arguments discussed in the Data section offer no specific support for this interpretation. The coefficients for GDP per capita and to a lesser extent for the share of services and the replacement rate are relatively stable over time. The two other control variables, i.e. the female labor share and the share of the age group 25–39, are only significant for 2004. These results are consistent with the findings in the last two columns of Table 4.

We conclude that there is evidence for a push effect of high uncertainty avoidance on the rate of business ownership. However, in recent years, a pull towards entrepreneurship in a climate of low uncertainty avoidance may have gained dominance vis-à-vis this longstanding historical push effect of high uncertainty avoidance. We have also found consistent confirmation of the well-known observation of a negative bearing of per capita income on business ownership. Finally, most results support a positive influence of the share of services and a negative effect of the replacement rate.

5.2 Indirect influence of uncertainty avoidance

Next, we explore the possible indirect influence of uncertainty avoidance on the rate of business ownership, by adding an interaction term between per capita income and uncertainty avoidance to the pooled panel regressions. The two left-hand columns of Table 6 compare the results of the base model including this interaction term in addition to uncertainty avoidance, GDP per capita and the year dummy variables with the original base model as presented in Table 4. The main outcome is a significant (at 10% level) intermediate effect of uncertainty avoidance on the influence of GDP per capita.15 What do these results mean in a quantitative sense? As an illustration, the results imply that for the country with the highest UAI-rate in the sample (Greece), an increase in real per capita income with $ 1.000 would imply a decrease of the business ownership rate with 0.61 percentage points, while for the country with the lowest uncertainty avoidance rate (Denmark), this increase in income would mean a decline in business ownership with 0.14 percentage points only. These differences show that the indirect effect exists indeed.
Table 6

The indirect influence of uncertainty avoidance on business ownership, pooled samples 1976, 1990 and 2004 (21 countries); base model using UAI and per capita income only

 

UAI continuous effect

UAI discrete effect

Constant

0.11*** (7.0)

0.064** (2.6)

0.14*** (8.2)

0.12*** (7.9)

Uncertainty avoidance (UAI)

0.041*** (3.1)

0.11*** (2.9)

  

Dummy UAI strong

  

0.016* (1.8)

0.060** (2.5)

GDP per capita (YCAP)

−0.38*** (3.9)

−0.017 (0.1)

−0.42*** (3.5)

−0.28*** (2.8)

UAI * YCAP

 

−0.53* (1.8)

  

Dummy UAI strong * YCAP

   

−0.31* (1.8)

Year dummy 1990

0.023*** (2.8)

0.022*** (2.7)

0.025*** (2.8)

0.024*** (2.8)

Year dummy 2004

0.046*** (4.3)

0.043*** (4.1)

0.050*** (4.2)

0.047*** (4.3)

N

63

63

63

63

R 2

0.368

0.400

0.347

0.383

Loglikelihood

137.3

139.0

136.3

138.1

Dependent variable: number of non-agricultural business owners per labor force.

Absolute heteroskedasticity consistent t-values are between brackets.

*p < 0.10; **p < 0.05; ***p < 0.01.

The right-hand columns of Table 6 present regressions in which uncertainty avoidance has been substituted by a dummy variable representing a high and a low uncertainty avoidance cluster of countries, as explained in the section on Method and Data. The results are similar to those including the continuous scale for UAI, as discussed in the previous paragraph. Again, the model including both a direct and an indirect effect of uncertainty avoidance outperforms the model including a direct effect only (at 10% level). An increase in real per capita income with $ 1.000 implies a decrease of the business ownership rate with 0.59 percentage points in the UAI+ countries and a decline with 0.28 percentage points in the UAI− countries.

Next, we test the robustness of the indirect effect by adding the share of services and the replacement rate as control variables. Table 7 reports the results. As can be seen from columns 2 and 4, the indirect effect then becomes somewhat smaller and is no longer fully significant.
Table 7

The indirect influence of uncertainty avoidance on business ownership, pooled samples 1976, 1990 and 2004 (21 countries), complete model

 

UAI continuous effect

UAI discrete effect

Constant

0.094*** (5.0)

0.058** (2.0)

0.11*** (6.8)

0.093*** (5.3)

Uncertainty avoidance (UAI)

0.038*** (3.3)

0.089** (2.6)

  

Dummy UAI strong

  

0.022*** (2.9)

0.048** (2.2)

GDP per capita (YCAP)

−0.57*** (7.0)

−0.31** (1.8)

−0.59*** (6.8)

−0.50*** (5.5)

UAI * YCAP

 

−0.38 (1.5)

  

Dummy UAI strong * YCAP

   

−0.18 (1.1)

Share services

0.14*** (4.6)

0.14*** (4.6)

0.15*** (5.5)

0.15*** (5.3)

Replacement rate unemployment

−0.052** (2.2)

−0.047** (1.9)

−0.052** (2.5)

−0.046** (2.0)

Year dummy 1990

0.029*** (3.8)

0.027*** (3.6)

0.028*** (3.9)

0.027*** (3.6)

Year dummy 2004

0.055*** (7.0)

0.052*** (6.1)

0.055*** (7.2)

0.053*** (6.5)

N

63

63

63

63

R 2

0.601

0.617

0.617

0.629

Loglikelihood

151.8

153.1

153.1

154.1

Dependent variable: number of non-agricultural business owners per labor force.

Absolute heteroskedasticity consistent t-values are between brackets.

*p < 0.10; **p < 0.05; ***p < 0.01.

By and large, there are serious indications of a differential effect of per capita income on entrepreneurship across the rate of uncertainty avoidance, but the robustness of these results is limited. A final observation on the basis of Tables 6 and 7 is that the statistical fit of a ‘discrete effect’ of uncertainty avoidance is not significantly better than that of a ‘continuous effect.’16

6 Conclusions

The prevalence of entrepreneurship, expressed as the percentage of business owners in the labor force, differs strongly between countries. The causes of this disparity do not only have an economic basis, but also stem from cultural differences between countries (Hofstede et al. 2004; Noorderhaven et al. 1999). The persistence of the country differences throughout the economic cycles points at cultural determinants, which are relatively constant per country.

Using a pooled dataset of a large number of OECD countries in 1976, 1990 and 2004, we have found a positive direct influence of uncertainty avoidance on business ownership rates, indicating that, in those years, a climate of high uncertainty avoidance in existing firms and organizations may push enterprising individuals towards self-employment (Baum’s hypothesis, as discussed before). These findings also show that a personal trait (risk aversion) and its cultural counterpart (uncertainty avoidance) may have a diverging impact on entrepreneurship. Repeating these regressions in three separate sample years confirms these results for 1976 and 1990. However, for the year 2004, the main outcome is that uncertainty avoidance no longer has any direct influence on business ownership. Our interpretation is that the advent of the entrepreneurial economy in recent years has created pull factors mobilizing the relatively abundant supply of potential entrepreneurial capital in countries with low uncertainty avoidance. In recent years, a pull towards entrepreneurship in a climate of low uncertainty avoidance has gained dominance vis-à-vis a longstanding historical push effect of high uncertainty avoidance.

We also found evidence for a negative indirect influence of uncertainty avoidance through a moderating effect on the influence of per capita income on business ownership. In low uncertainty avoidance countries, the negative influence of per capita income on the rate of business ownership is clearly smaller than in high uncertainty avoidance countries. In a group of eight high-uncertainty avoidance countries, a relatively strong negative relationship between GDP per capita and the level of business ownership suggests that rising opportunity costs of entrepreneurship are the dominant perception in this cultural environment. On the other hand, in a group of 13 low-uncertainty avoidance countries, the relatively weak negative relationship between business ownership and per capita income suggests that rising opportunities are a countervailing force in an environment of low uncertainty avoidance.

A closer look at the underlying development of the business ownership rate in all 21 countries between 1972 and 2004 reveals the following. In the group of low-uncertainty avoidance countries, eight out of 13 nations show either a clear U-shape (Finland, Germany, the Netherlands and New Zealand) or a vaguely U-shape trend (Australia, Great Britain, Sweden and USA), three show a continuously upward trend in entrepreneurship (Canada, Ireland and Switzerland), one shows a stabilization in the last 20 years (Denmark), while only one (Norway) shows a decreasing trend.17 In the group of high-uncertainty avoidance countries, two out of eight countries (France and Japan)18 show a strongly decreasing trend, while six show an increase or a U-shape, sometimes followed by stabilization. While the large number of countries with rising business ownership rates across both groups bear witness to a worldwide trend toward more entrepreneurship related to ICT and deregulation, the differential indirect effects of uncertainty avoidance also suggest that, in modern service economies, high uncertainty avoidance may indirectly have a negative impact on the development of business ownership and may hamper the exploitation of new economic opportunities.

Our study has some limitations that should be borne in mind when interpreting the results. First, the modest explanatory power of most of our regressions suggests that other cultural and psycho-sociological variables may also play a role.19 Second, our paper only studies the effect of uncertainty avoidance on the level of entrepreneurship. It would be relevant to repeat the study for the dynamics of entrepreneurship, although a lack of time series of harmonized business start-up data across countries may hamper the latter at least in the near future. Finally, business ownership rates are available for a far smaller number of countries than uncertainty avoidance data. This inhibits fuller testing of the direct and the indirect effect of uncertainty avoidance.

Nonetheless, the present results may already have some relevance for policymakers trying to promote entrepreneurship. While we would not advocate social engineering, the results do suggest that countries should investigate to what extent their educational system and relevant labor market, social and fiscal legislation foster a low or a high degree of uncertainty avoidance within the population.

Footnotes

  1. 1.

    This approach is rooted in the work of Knight (1921) and starts from the functions of the provision of entrepreneurial ability and the bearing of risks. The second function underlines the importance of risk attitudes in the occupational choice process. See, for instance, Kihlstrom and Laffont (1979), Parker (1997), where the degree of risk aversion and the differences in risk of becoming a business owner vis-à-vis an employee are given the central role in the determination of the occupational choice. See also Freytag and Thurik (2007).

  2. 2.

    See Blanchflower and Meyer (1994), Blanchflower and Oswald (1998), Douglas and Shepherd (2002), Evans and Leighton (1989, 1990), Grilo and Irigoyen (2006), Grilo and Thurik (2005a,b), Lin et al. (2000) for empirical work.

  3. 3.

    There is no generally accepted definition of entrepreneurship. See Wennekers and Thurik (1999) for an overview. In the present paper, we adopt a pragmatic approach by equating entrepreneurship, business ownership and self-employment, and an entrepreneur will simply be understood to be the owner/manager of either an unincorporated or an incorporated business. See also Thurik and Wennekers (2004), Davidsson (2004).

  4. 4.

    In terms of our illustration above, the group of individuals with a high uncertainty aversion level, H, is larger than in a culture with low uncertainty avoidance.

  5. 5.

    Note again that OC*SE,i in Eq. 1 has to be positive in order for an individual to choose for self-employment, as the utility variables in the model are defined relative to the situation of wage–employment.

  6. 6.

    In a micro-economic model of occupational choice (de Wit 1993), equality of entrepreneurial income and wages determine the equilibrium number of self-employed. In this model, an exogenous wage increase lowers the (equilibrium) number of self-employed while an exogenous upward shift of profits raises the equilibrium.

  7. 7.

    In addition, nascent entrepreneurs were found to have more management experience.

  8. 8.

    Delmar and Davidsson (2000), EIM/EZ (2000), van Gelderen (1999:21) and various annual Executive Reports published by the Global Entrepreneurship Monitor.

  9. 9.

    The clustering was carried out with the K-means algorithm. See Noorderhaven et al. (1999).

  10. 10.

    COMParative ENtrepreneurship Data for International Analysis. See van Stel (2005). In the current paper, data from COMPENDIA version 2004.2 are used.

  11. 11.

    No data on Hofstede’s indices are available for Iceland, whereas for Luxembourg there are estimates that we have used for clustering only. See Noorderhaven et al. (1999).

  12. 12.

    This study was first published in 1980, but the second edition, published in 2001, gives more information on stability and cross-validation of the data.

  13. 13.

    For the 19 countries in our dataset for which we have both Hofstede and GLOBE data on uncertainty avoidance, the correlations are: Hofstede UAI x GLOBE practices: −0.643; Hofstede UAI x GLOBE values: 0.607; GLOBE practices x GLOBE values: −0.869. All these correlations are significant at the 1% level.

  14. 14.

    Excluding the Gini coefficient and the long term interest rate, due to the smaller available number of observations of these variables.

  15. 15.

    This appears both from the significance level of the interaction term (p value is 0.078) and from a loglikelihood test comparing the models in the first two columns of Table 6. The LR test statistic is 3.4 while the 10% critical value is 2.71.

  16. 16.

    Note that, in Table 6, the R 2 of the continuous effect model is (slightly) higher compared to the discrete effect model while in Table 7, this is the other way around.

  17. 17.

    However, in 2004, Norway had a significant rise in the business ownership rate compared to 2002, possibly indicating a stabilization or even reversal of the downward trend. By contrast, while Canada and Switzerland show an increasing trend over the period 1972–1998, the business ownership rates of these countries are decreasing since 1998.

  18. 18.

    In addition, Luxemburg, that estimates show to be a high uncertainty avoidance country, also has a declining trend.

  19. 19.

    This includes individualism (Shane 1993), post-materialism (Uhlaner and Thurik 2007) and dissatisfaction (Noorderhaven et al. 2004).

Notes

Acknowledgement

An early version of the present paper was read at the Workshop on Entrepreneurship and Culture, held at the Max Planck Institute of Economics in Jena on 7 February 2005. The authors would like to thank Geert Hofstede, Lorraine Uhlaner and two anonymous referees for helpful comments. The paper has been written in the framework of the research program SCALES which is carried out by EIM and is financed by the Dutch Ministry of Economic Affairs.

References

  1. Acs ZJ, Carlson B, Karlsson Ch (1999) The linkages among entrepreneurship, SMEs and the macroeconomy. In: Acs ZJ, Carlson B, Karlsson Ch (eds) Entrepreneurship, small and medium-sized enterprises and the macroeconomy. Cambridge University Press, Cambridge, UK, pp 3–42Google Scholar
  2. Acs Z, Arenius P, Hay M, Minniti M (2004) Global entrepreneurship monitor: 2004 executive report. Babson College and London Business SchoolGoogle Scholar
  3. Audretsch DB, Keilbach M (2004) Entrepreneurship capital and economic performance. Reg Stud 38(8):949–959CrossRefGoogle Scholar
  4. Audretsch DB, Thurik AR (2000) Capitalism and democracy in the 21st century: from the managed to the entrepreneurial economy. J Evol Econ 10(1):17–34CrossRefGoogle Scholar
  5. Audretsch DB, Thurik AR (2001) What is new about the new economy: sources of growth in the managed and entrepreneurial economies. Ind Corp Change 10(1):267–315CrossRefGoogle Scholar
  6. Audretsch DB, Thurik AR (2004) A model of the entrepreneurial economy. Int J Entrep Educ 2(2):143–166Google Scholar
  7. Bais J, van der Hoeven WHM, Verhoeven WHJ (1995), Determinanten van zelfstandig ondernemerschap: een internationale vergelijking, Den Haag: OSA-werkdocumentGoogle Scholar
  8. Baum JR, Olian JD, Erez M, Schnell ER, Smith KG, Sims HP, Scully JS, Smith KA (1993) Nationality and work role interactions: a cultural contrast of Israeli and U.S. entrepreneurs’ versus managers’ needs. J Bus Venturing 8:499–512CrossRefGoogle Scholar
  9. Bhide A (1994) How entrepreneurs craft strategies that work. Harvard Bus Rev 72(2):150–161, March–AprilGoogle Scholar
  10. Blanchflower DG, Meyer BD (1994) A longitudinal analysis of the young self-employed in Australia and the United States. Small Bus Econ 6:1–19CrossRefGoogle Scholar
  11. Blanchflower DG, Oswald AJ (1998) What makes an entrepreneur? J Labor Econ 16(1):26–60CrossRefGoogle Scholar
  12. Blanchflower DG, Oswald A, Stutzer A (2001) Latent entrepreneurship across nations. Eur Econ Rev 45:680–691CrossRefGoogle Scholar
  13. Blau D (1987) A time series analysis of self-employment. J Polit Econ 95:445–467CrossRefGoogle Scholar
  14. Bygrave WD, Hunt SA (2005) Global entrepreneurship monitor 2004 financing report, Babson College and London Business SchoolGoogle Scholar
  15. Carree M, van Stel A, Thurik R, Wennekers S (2002) Economic development and business ownership: an analysis using data of 23 OECD countries in the period 1976–1996. Small Bus Econ 19(3):271–290CrossRefGoogle Scholar
  16. Carree M, Thurik R (1995) Profitability and number of firms: their dynamic interaction in Dutch retailing. In: A. van Witteloostuijn (ed) Studies in industrial organization: market evolution: competition and cooperation. Kluwer, Boston, MA, pp. 257–266Google Scholar
  17. Carree M, Thurik R (2003) The impact of entrepreneurship on economic growth. In: Acs ZJ, Audretsch DB (eds) Handbook of entrepreneurship research. Kluwer, Boston, MA, pp. 437–471Google Scholar
  18. Chandler AD Jr (1990) Scale and scope: the dynamics of industrial capitalism. Harvard University, CambridgeGoogle Scholar
  19. Chittenden F, Hall G, Hutchinson P (1996) Small firm growth, access to capital markets and financial structure: review of issues and an empirical investigation. Small Bus Econ 8(1):59–67CrossRefGoogle Scholar
  20. Choi YB (1993) Paradigms and conventions: uncertainty, decision making and entrepreneurship. University of Michigan Press, Ann Arbor, MIGoogle Scholar
  21. Cohen J, Cohen P (1983) Applied multiple regression/correlation analysis for the behavioral sciences. Erlbaum, Hillsdale, NJGoogle Scholar
  22. Davidsson P (2004) Researching Entrepreneurship, International Studies in Entrepreneurship. Springer, Berlin Heidelberg New YorkGoogle Scholar
  23. de Wit G (1993) Models of self-employment in a competitive market. J Econ Surv 7:367–397CrossRefGoogle Scholar
  24. Delmar F, Davidsson P (2000) Where do they come from? Prevalence and characteristics of nascent entrepreneurs. Entrep Reg Dev 12:1–23CrossRefGoogle Scholar
  25. Douglas EJ, Shepherd DA (2002) Self-employment as a career choice: attitudes, entrepreneurial intentions, and utility maximization. Entrep Theory Pract 26(3):81–90Google Scholar
  26. EIM/EZ (2000) Entrepreneurship in the Netherlands; opportunities and threats to nascent entrepreneurship. EIM, ZoetermeerGoogle Scholar
  27. Ekelund RB, Hébert RF (1990) A history of economic theory and method. McGraw-Hill, New YorkGoogle Scholar
  28. Evans DS, Jovanovic B (1989) An estimated model of entrepreneurial choice under liquidity constraints. J Polit Econ 97(4):808–827CrossRefGoogle Scholar
  29. Evans DS, Leighton LS (1989) Some empirical aspects of entrepreneurship. Am Econ Rev 79(3):519–535Google Scholar
  30. Evans DS, Leighton LS (1990) Small business formation by unemployed and employed workers. Small Bus Econ 2:319–330CrossRefGoogle Scholar
  31. Freytag A, Thurik AR (2007) Entrepreneurship and its determinants. J Evol Econ (this issue)Google Scholar
  32. Gifford S (2003) Risk and uncertainty. In: Acs ZJ, Audretsch DB (eds) Handbook of entrepreneurship research, Boston, MA: Kluwer, pp 37–53Google Scholar
  33. Grilo I, Irigoyen J (2006) Entrepreneurship in the EU: to wish and not to be. Small Bus Econ 26(4):305–318CrossRefGoogle Scholar
  34. Grilo I, Thurik AR (2005a) Determinants of entrepreneurial engagement levels in Europe and the US, Max Planck Institute Discussion Paper on Entrepreneurship, Growth and Public Policy 25-2005, Jena, Germany: Max Planck Institute of EconomicsGoogle Scholar
  35. Grilo I, Thurik AR (2005b) Latent and actual entrepreneurship in Europe and the US: some recent developments. Int Entrep Manag J 1(4):441–459CrossRefGoogle Scholar
  36. Hamilton RT (1989) Unemployment and business formation rates: reconciling time series and cross-sections. Environ Plann A 11:249–255CrossRefGoogle Scholar
  37. Hébert RF, Link AN (1989) In search of the meaning of entrepreneurship. Small Bus Econ 1:39–49CrossRefGoogle Scholar
  38. Hessels SJA, van Stel AJ, Brouwer P, Wennekers ARM (2006) Social security arrangements and early-stage entrepreneurial activity; an empirical analysis. EIM Scales Paper N200518. EIM, Zoetermeer, NetherlandsGoogle Scholar
  39. Hofstede G (2001) Culture’s consequences; comparing values, behaviors, institutions and organizations across nations, 2nd edition. Sage, Thousand OaksGoogle Scholar
  40. Hofstede G, Noorderhaven NG, Thurik AR, Uhlaner LM, Wennekers ARM, Wildeman RE (2004) Culture’s role in entrepreneurship: self-employment out of dissatisfaction. In: Ulijn J, Brown T (eds) Innovation, entrepreneurship and culture: the interaction between technology, progress and economic growth. Cheltenham, UK and Brookfield. Edward Elgar, USA, pp 162–203Google Scholar
  41. House RJ, Hanges PJ, Javidan M, Dorfman PW, Gupta V (eds) (2004). Culture, leadership and organizations: the GLOBE study of 62 societies. Sage, Thousand OaksGoogle Scholar
  42. Ilmakunnas P, Kanniainen V, Lammi U (1999) Entrepreneurship, economic risks, and risk-insurance in the welfare state. Discussion paper No. 453, Department of Economics, University of HelsinkiGoogle Scholar
  43. Iyigun MF, Owen AL (1998) Risk, entrepreneurship, and human-capital accumulation. AEA Papers and Proceedings 88(2):454–457Google Scholar
  44. Kihlstrom R, Laffont JJ (1979) A general equilibrium entrepreneurial theory of firm formation based on risk aversion. J Polit Econ 87:719–748CrossRefGoogle Scholar
  45. Knight FH (1921) Risk, uncertainty and profit. Houghton Mifflin, New YorkGoogle Scholar
  46. Kuznets S (1971) Economic growth of nations, total output and production structure. Harvard University Press, Cambridge, MAGoogle Scholar
  47. LeCornu M, McMahon R, Forsaith D, Stanger A (1996) The small enterprise financial objective function. J Small Bus Manag 34(3):1–14Google Scholar
  48. Lin Z, Picot G, Compton J (2000) The entry and exit dynamics of self-employment in Canada. Small Bus Econ 15:105–125CrossRefGoogle Scholar
  49. Loveman G, Sengenberger W (1991) The re-emergence of small-scale production; an international comparison. Small Bus Econ 3:1–37CrossRefGoogle Scholar
  50. Lucas RE Jr (1978) On the size distribution of business firms. Bell J Econ 9:508–523CrossRefGoogle Scholar
  51. Maslow AH (1970) Motivation and personality. Harper and Row, New YorkGoogle Scholar
  52. McGrath RG, MacMillan IC, Scheinberg S (1992) Elitists, risk-takers, and rugged individualists? An exploratory analysis of cultural differences between entrepreneurs and non-entrepreneurs. J Bus Venturing 7:115–135CrossRefGoogle Scholar
  53. Meager N (1992) Does unemployment lead to self-employment? Small Bus Econ 4:87–103CrossRefGoogle Scholar
  54. Minniti M, Arenius P, Langowitz N (2005) Global entrepreneurship monitor 2004 report on women and entrepreneurship, Center for Women’s Leadership at Babson CollegeGoogle Scholar
  55. Noorderhaven N, Wennekers S, Hofstede G, Thurik R, Wildeman R (1999) Self-employment out of dissatisfaction; an international study. Discussion paper TI 99–089/3. Tinbergen Institute, Erasmus University RotterdamGoogle Scholar
  56. Noorderhaven N, Thurik R, Wennekers S, van Stel A (2004) The role of dissatisfaction and per capita income in explaining self-employment across 15 European countries. Entrep Theory Pract 28(5):447–466CrossRefGoogle Scholar
  57. Parker SC (1997) The effects of risk on self-employment. Small Bus Econ 9:515–522CrossRefGoogle Scholar
  58. Parker SC (2004) The economics of self-employment and entrepreneurship. Cambridge University Press, Cambridge, UKGoogle Scholar
  59. Parker SC, Robson MT (2004) Explaining international variations in self-employment: evidence from a panel of OECD countries. South Econ J 71:287–301Google Scholar
  60. Schultz TP (1990) Women’s changing participation in the labor force: a world perspective. Econ Dev Cult Change 38:457–488CrossRefGoogle Scholar
  61. Schultz TW (1975) The value of the ability to deal with disequilibria. J Econ Lit 13:827–846Google Scholar
  62. Schumpeter J (1934) The theory of economic development. Harvard University Press, Cambridge, MA (a translation of the second edition from 1926 of a work that originally appeared in 1911)Google Scholar
  63. Shane SA (1993) Cultural influences on national rates of innovation. J Bus Venturing 8:59–73CrossRefGoogle Scholar
  64. Storey DJ (1994) Understanding the small business sector. Routledge, New YorkGoogle Scholar
  65. Sully de Luque M, Javidan M (2004) Uncertainty avoidance. In: House RJ, Hanges PJ, Javidan M, Dorfman PW, Gupta V (eds) Culture, leadership and organizations: the GLOBE study of 62 societies. Sage, Thousand Oaks, pp 602–653Google Scholar
  66. Swedberg R (2000) Entrepreneurship; the social science view. Oxford University Press, OxfordGoogle Scholar
  67. Thurik AR, Wennekers S (2004) Entrepreneurship, small business and economic growth. J Small Bus Enterp Dev 11(1):140–149CrossRefGoogle Scholar
  68. Uhlaner LM, Thurik AR, Hutjes J (2002) Post-materialism: a cultural factor influencing entrepreneurial activity across nations, Erasmus Research Institute for Management Report ERS-2002-62-STR. Erasmus University, RotterdamGoogle Scholar
  69. Uhlaner LM, Thurik AR (2007) Post-materialism influencing total entrepreneurial activity across nations. J Evol Econ (this issue)Google Scholar
  70. van Gelderen MW (1999) Ontluikend ondernemerschap, een studie naar mensen die bezig zijn met het opzetten van een bedrijf (nascent entrepreneurs). Strategische Verkenning. EIM, ZoetermeerGoogle Scholar
  71. van Praag M (1996) Determinants of successful entrepreneurship. Amsterdam: Thesis PublishersGoogle Scholar
  72. van Praag CM (1999) Some classic views on entrepreneurship. De Economist 147:311–335CrossRefGoogle Scholar
  73. van Stel A (2005) COMPENDIA: harmonizing business ownership data across countries and over time. Int Entrep Manag J 1(1):105–123CrossRefGoogle Scholar
  74. van Stel A, Carree M (2004) Business ownership and sectoral growth; an empirical analysis of 21 OECD countries. Int Small Bus J 22(4):389–419CrossRefGoogle Scholar
  75. Varian HR (1992) Microeconomic analysis. Norton, New YorkGoogle Scholar
  76. Verheul I (2005) Is there a (fe)male approach? Understanding gender differences in entrepreneurship, ERIM PhD. Series Research in Management. Erasmus Research Institute of Management, Erasmus University RotterdamGoogle Scholar
  77. Verheul I, Wennekers S, Audretsch D, Thurik R (2002) An eclectic theory of entrepreneurship: policies, institutions and culture. In: Audretsch DB, Thurik AR, Verheul I, Wennekers ARM (eds) Entrepreneurship: determinants and policy in a European–US comparison. Kluwer, Boston, MA, pp 11–81CrossRefGoogle Scholar
  78. Wennekers S (2006) Entrepreneurship at country level: economic and non-economic determinants. Rotterdam: ERIMGoogle Scholar
  79. Wennekers S, Folkeringa M (2002) The development of the business ownership rate in the Netherlands 1899–1998; a decomposition into sector shift and within sector trends. Paper presented at BKERC 2002 Conference. Boulder, ColoradoGoogle Scholar
  80. Wennekers S, Thurik R (1999) Linking entrepreneurship and economic growth. Small Bus Econ 13:27–55CrossRefGoogle Scholar
  81. Wennekers S, van Stel A, Thurik R, Reynolds P (2005) Nascent entrepreneurship and the level of economic development. Small Bus Econ 24(3):293–309CrossRefGoogle Scholar
  82. Wubben EM (1993) Markets, uncertainty and decision making: a history of the introduction of uncertainty into economics. Thesis Publishers, AmsterdamGoogle Scholar
  83. Yamada G (1996) Urban informal employment and self-employment in developing countries: theory and evidence. Econ Dev Cult Change 44:289–314CrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2007

Authors and Affiliations

  • Sander Wennekers
    • 1
  • Roy Thurik
    • 1
    • 2
    • 3
  • André van Stel
    • 1
    • 2
    • 4
  • Niels Noorderhaven
    • 5
  1. 1.EIM Business and Policy ResearchZoetermeerThe Netherlands
  2. 2.Erasmus University RotterdamRotterdamThe Netherlands
  3. 3.Max Planck Institute of EconomicsJenaGermany
  4. 4.Cranfield School of ManagementCranfield UniversityCranfieldUK
  5. 5.Tilburg UniversityTilburgThe Netherlands

Personalised recommendations