Individual characteristics innovation activity of entrepreneurs: Who creates innovations in different countries?

Using data from the Global Entrepreneurship Monitor from 2005 to 2018, we investigate the heterogeneity of the relationship between individual characteristics of entrepreneurs and the probability of their innovation activity in different countries. The dependent variable is a binary variable reflecting the entrepreneur's participation in innovation activities. The independent variables are as follows: past entrepreneurial experience, acquaintance with entrepreneurs, perceived availability of business start-up opportunity, fear of failure, voluntary motivation and higher education. Using a finite mixture logistic regression model, we identify 4 clusters of countries. In the first and fourth clusters, the coefficients in front of the variables of acquaintance with other entrepreneurs and voluntary motivation appeared significant. At the same time, in the first cluster, unlike in the fourth cluster, the coefficients in front of the variables reflecting human capital (higher education and sufficient skills to start a business) are significant. The second cluster consists of countries with the most developed external environment characteristics: the coefficients in front of all variables of interest (except fear of failure) are significant. In the third cluster, on the contrary, only the perceived availability of start-up opportunities shows a positive relationship with innovation activity.