What Characterizes Entrepreneurs and Individuals in Self-Employment?

  Christoph Siemroth  

Table of  Contents  

1.  Introduction  1

2.  What Might Influence the Choice of Becoming Entrepreneur  1

2.1.  Attitude Towards Risk: Less Risk Aversion  1

2.2.  Social Factors  2

2.3.  Liquidity Constraints  4

2.4.  Unattractiveness as Employee  5

2.5.  Income  6

2.6.  Age  7

3.  Summary  8

  List of References  9

Table of  Contents  

Christoph Siemroth

1. Introduction

This paper reviews some empirical studies on entrepreneurship on the level of the individual and identifies factors influencing his occupational choice of becoming self-employed or an employee. 1 The goal is to find representative characteristics of an entrepreneur. Therefore, several traits or prerequisites that might be expected to influence the propensity to become entrepreneur are ana- lyzed, their empirical relevance is evaluated and reasons or explanations for the findings are dis- cussed. Given the formal confinements of this paper, the overview can by no means be exhaustive, but it nonetheless endeavours to reveal the most basic determinants that characterize entrepreneurs in comparison to employees.

2. What Might Influence the Choice of Becoming Entrepreneur?

2.1. Attitude Towards Risk: Less Risk Aversion

Being an entrepreneur, by definition, is a risky occupational choice: Survival of the entrepreneurial endeavour and the future income is uncertain. Most analytic economic papers dealing with entre- preneurship include some uncertainty in their models. For example, in Ilmakunnas/Kanniainen (2001: 199ff.) entrepreneurs face uncertainty about the state of the future market (on which the future entrepreneurial income depends) and Kanniainen/Poutvaara (2007: 5ff.) incorporate risk of project-failure in their model, as do De Meza/Webb (1999: 155ff.). There are also several empirical studies finding earnings of self-employed persons to have a higher variance than the earnings of employees (Ekelund et al. 2005: 651). Given this, we can safely assume that being an entrepreneur is more risky than being an employee. Therefore, ceteris paribus, less risk averse (or more risk seek- ing) individuals should more likely be entrepreneurs (ibid.: 651).

Ekelund et al. (2005) investigated with psychological test data whether or not risk aversion has a significant effect on the occupational choice (self-employed or employee). They used data from questionnaires of a population of Finnish citizens born in the same year (1966) in a geographi- cally defined area, so there is relatively high cultural homogeneity in the sample. The data has been collected in various stages of the subjects’ lives until the age of 31. It consisted of important socio- economic variables such as labour market history and education (ibid.: 652).

The empirical analysis revealed that their risk aversion measure (“harm avoidance”) is in fact significantly lower (p=0.0001) for self-employed individuals than for employees (ibid.: 655). En- dogeneity could be ruled out and so they conclude that the “risk aversion measure has a significant and economically relevant effect on the probability of an individual being self-employed” (ibid.:

1 Being self-employed and being entrepreneur is used interchangeably.

1

Christoph Siemroth

657). Furthermore they argue this effect is even causal, i.e. less risk aversion causes individuals to more likely choose the occupation of the entrepreneur (ibid.: 658).

What would also be interesting to know is whether less risk aversion also means more success as entrepreneur. Among negotiators, for example, the prevalent view is that more risk averse nego- tiators get worse results, because in fear of failed negotiation they accept less than risk seeking negotiators that do not “chicken out” that early. 2 Taking the tenure of self-employment as measure of success (failure or bankruptcy did not occur), Ekelund et al. (2005: 657) find no correlation be- tween the tenure of the self-employed and their degree of risk aversion, suggesting that degree of risk aversion has no significant effect on the success as entrepreneur. 3 Risk aversion, therefore, only seems to play a role in characterizing individuals as entrepreneurs or employees, but not in distin- guishing them as more or less successful entrepreneurs.

Finally, they also found that a significantly (p<0.0001) greater proportion of males are self- employed. This is plausible, since it is consistent with the view that men are mostly less risk averse than women (ibid.: 655). The question remains if this is solely due to the lower risk aversion of males or whether there are some other factors that favour men as entrepreneurs.

In any case, the results are robust, indicating that less risk aversion does not only distinguish the average entrepreneur from the average employee, but it also causes individuals to more likely choose entrepreneurship as occupation. Perhaps as a result of that, men – who generally are less risk averse – are much more likely to become self-employed than women.

2.2. Social Factors

Do social factors such as family or parents influence the propensity to become an entrepreneur? Or does being an entrepreneur have impact on social factors?

A greater proportion of self-employed individuals in Ekelund’s sample is married, but this

difference is not statistically significant.

2 This can be shown analytically with the asymmetric Nash-Bargaining-Solution (it also holds empirically). Suppose there are two negotiators, 1 has a utility function of u 1 =x (and is thus risk neutral) and 2 has the utility function u 2 =(1−x) 0.5 (risk averse), whereas x is the part that 1 receives through bargaining and (1−x) the part that 2 receives. The Nash-Solution is determined by: max [Π = x∙(1−x) 0.5 ] → 0 = Π’ = (1−x) 0.5 −x∙0.5(1−x) −0.5 = (1−x) −0.5 [(1−x)−0.5x] → x = 2/3, thus the risk neutral negotiator has a better result than the risk averse one, all other things being equal.

3 One could still make an argument in favor of the hypothesis that less risk aversion makes better entrepreneurs. The test from Ekelund et al. only checked whether the tenure of self-employed individuals (i.e. not yet failed entrepreneurs) correlated with their risk aversion. It could still be possible that other individuals with higher risk aversion entered the market as entrepreneurs in the past, but were less successful and thus selected out in Darwinian fashion in the market competition. Therefore, the very fact that existing (i.e. surviving) entrepreneurs have significantly lower risk aversion could mean that lower risk aversion characterizes successful entrepreneurs after all. The test Ekelund et al. made does not rule that out, because they only considered the already successful entrepreneurs.

2