26/6/2014: Explaining the Gender Gap in Entrepreneurship

A new paper by Caliendo, Marco and Fossen, Frank M. and Kritikos, Alexander and Wetter, Miriam, titled "The Gender Gap in Entrepreneurship: Not Just a Matter of Personality" (May 23, 2014: CESifo Working Paper Series No. 4803 http://ssrn.com/abstract=2457841) tackles a very important and highly sensitive issue of gender gap in entrepreneurship.

The authors ask "Why do entrepreneurship rates differ so markedly by gender?"

The paper uses data from a large, representative German household panel, covering 2000-2009 period, to "investigate to what extent personality traits, human capital, and the employment history influence the start-up decision and can explain the gender gap in entrepreneurship."

"In contrast to previous research the main advantage of our data set is that it contains not only information on the socio-demographic background of the respondents, but also on a broad set of personality constructs that elicit the Big Five traits and several specific personality characteristics."

Note: Big Five Factor Model of personality (McCrae and Costa, 2008) "describes the personality by the factors openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism (or, reversely, emotional stability)."

Per authors, "To the best of our knowledge information on the Big Five approach has not been used to assess the gender gap in entrepreneurship. We are the first to simultaneously analyse the effects of the Big Five factors, risk aversion, locus of control, and the ability to trust others ..., as well as of a variety of variables controlling for human capital, employment status, and other socio-demographic factors on the gender specific decision to enter self-employment."

The findings are very far-reaching and substantially in dispute with commonly held views:

1. "Applying a decomposition analysis, we observe that the higher risk aversion among women explains a large share of the entrepreneurial gender gap."
2. "We also find an education effect contributing to the gender difference." More specifically: "On average, working aged women in Germany are still less educated than men and are, therefore, less inclined to start a business."
3. "Thirdly, the current employment state has a strong effect into the opposite direction: If the share of women in wage employment were as high as the male share, holding everything else constant, their entry rate into self-employment would be much smaller."
4. "…we show that personality traits help explain the gender gap in nuanced ways. While specific characteristics, in particular risk attitudes, are able to explain a substantial amount of the gender gap, the overall influence of the Big Five personality constructs point to the opposite direction. This means that if women were endowed with the same scores in the Big Five as men, the gap would be even larger."

Overall: "the explained gap is therefore negative meaning that if women exhibited in all observable variables the same parameter values as men, the entry rate of women would be even smaller than actually observed."