Risk attitudes of nascent entrepreneurs–new evidence from an experimentally validated survey

Abstract

The influence of risk aversion on the decision to become self-employed is a much discussed topic in the entrepreneurial literature. Conventional wisdom asserts that being an entrepreneur means making risky decisions; hence more risk-averse individuals are less likely to become entrepreneurs. In contrast to previous research, we are able to examine empirically whether the decision of starting a business is influenced by objectively measurable risk attitudes at the time when this decision is made. Our results show that in general, individuals with lower risk aversion are more likely to become self-employed. Sensitivity analysis reveals, however, that this is true only for people coming out of regular employment, whereas for individuals coming out of unemployment or inactivity, risk attitudes do not seem to play a role in the decision process.

Appendices

Appendix A: Arrow–Pratt coefficient of relative risk aversion

From the responses to the hypothetical investment question, under certain assumptions, we can calculate proxies for individual Arrow–Pratt coefficients of RRA. Utility is a function of wealth. Individuals may choose to invest an amount inv between zero and x = 100,000 Euros (the hypothetical windfall gain). There are equal probabilities α of earning a profit of inv and losing half of it. Thus, the agent maximizes his/her expected utility subject to the budget constraint:

$$ \begin{array}{ll} {\hbox{max}}&\left(\alpha U(x+inv)+\alpha U\left(x-\frac{inv}{2}\right)\right)={\hbox{max}}(f(inv))\\ {\hbox{s.t.}}&0\leq inv \leq 100,000\end{array} $$

(2)

The problem is solved by finding the null of the first derivative:

$$ \begin{array}{lll} f^{\prime}(inv)=0&\mathop{\Longrightarrow}\limits^{\alpha > 0} & U^{\prime}(x+inv)=\frac{1}{2}U^{\prime}\left(x-\frac{inv}{2}\right)\\ &\mathop{\Longrightarrow}\limits^{Taylor-approx.} & U^{\prime}(x)+inv U^{\prime\prime}(x)\approx \frac{1}{2}U^{\prime}(x)-\frac{inv}{4}U^{\prime\prime}(x)\\&\mathop{\Longrightarrow}\limits^{inv\neq 0\wedge U^{\prime}(x)\neq 0} &\rho _{ARA}=\frac{-U^{\prime\prime}(x)}{U^{\prime}(x)}\approx\frac{2}{5inv}\end{array} $$

(3)

Individuals cannot indicate risk-neutral or risk-loving attitudes by construction of the hypothetical investment question, which implies the assumption $U''(x) < 0\forall x.$ As α > 0, it follows that

$$ f^{\prime\prime}(inv)= \alpha U''(x+inv)+\frac{1}{4}\alpha U^{\prime\prime}(x-\frac{inv}{2}) < 0\forall inv\\ $$

(4)

Thus, f(inv) reaches its global maximum at the null. ρ_ARA is the Arrow–Pratt measure of absolute risk aversion. We approximate the individual’s total wealth endowment w with the hypothetical 100,000 Euros to calculate the coefficient of RRA ρ_RRA:

$$ \rho_{\rm RRA} = \rho _{\rm ARA} \cdot w \approx \frac{2}{5inv}\cdot 100,000 $$

(5)

As inv is nonnegative, ρ_RRA is always positive (risk averse agents). If an individual chooses to invest nothing (inv = 0), we arbitrarily set his/her ρ_RRA to 4 (double the coefficient of an individual choosing the smallest investment possible, i.e., 20,000 Euros). The results remain basically unchanged, albeit somewhat less significant, if ρ_RRA is set to 8 or 16.

Appendix B: Definition of variables used in the estimations

Table B.1 Detailed description of the risk measures used

Table B.2 Detailed description of the variables used