Entrepreneurial Activity and the Level of Economic Development

<p>The aim of this thesis is to research the type of &nbsp;interdependence between entrepreneurial activity and the level of economic development measured as GDP per capita. On the one hand, it is expected that countries with advanced institutional sphere, financial sector, infrastructure, democratic laws will demonstrate higher rates of entrepreneurial activity compared to those of less developed nations. However, the data shed light on the opposite relationship between the level of entrepreneurial activity and economy&rsquo;s maturity. In the context of theoretical investigation this thesis represents several approaches used in the literature to determine the concept of entrepreneurship and the role of an entrepreneur in the modern world. Some models of economic development, namely Schumpeter&rsquo;s theory of creative destruction and accumulation, Rostow&rsquo;s theory of five stages of economic growth, and theories of structural transformation were also considered. Some of the sound papers where a U- shaped relationship is tested against an L-shaped one were closely scrutinised. The main factors of entrepreneurial activity were also specified. The work contains information on the sources of endogenous entrepreneurship and its role in the economic progress and in the knowledge spillovers. Econometric analysis of panel data across 119 countries during 2003-2012 was conducted. The data were derived from the databases of the World Bank, the Global Entrepreneurial Monitor, the United Nations. The regression model was run on the percentage of self-employed. The independent variables, reflecting economic, institutional, cultural cross-country differences were included as controlling variables. &nbsp;Pooled regression, fixed-effect and random-effect models were estimated. The tests have supported the fixed-effect model. When running specification tests it became evident that there were no statistically significant differences between a U-shaped and an L-shaped models and that R-squared is not an efficient way to make inferences. &nbsp;Such methods of cluster analysis &nbsp;as hierarchical clustering and k-means were applied. Cluster analysis set out to detect 3 groups of countries in the sample which were compared with the existing classification on factor-, efficiency-, and innovation-driven countries presented in the GEM 2011 Global Report-EXTENDED. The correlation coefficient between the clustered sample and the original GEM&rsquo;s sample is equal to 74.6%.</p>