TY - JOUR
T1 - Revisiting the Determinants of Entrepreneurship
T2 - A Bayesian Approach
AU - Arin, K. Peren
AU - Huang, Victor Zengyu
AU - Minniti, Maria
AU - Nandialath, Anup Menon
AU - Reich, Otto F.M.
N1 - Publisher Copyright:
© The Author(s) 2014.
PY - 2015/2/27
Y1 - 2015/2/27
N2 - Entrepreneurship has long been seen as an important instrument in stimulating and generating economic growth. The amount of research trying to identify key factors that drive entrepreneurship is considerable; yet, little consensus has been achieved. We argue that this lack of consensus could be on account of model uncertainty as empirical studies often tend to be selective on what variables are included in the final model. Drawing on recent literature, we demonstrate the benefits of Bayesian model averaging (BMA) in reducing the impact of model uncertainty on empirical research in entrepreneurship. Additionally, BMA provides measures of variable importance and can be seen as a complementary approach to dominance/relative importance analysis. We show that when model uncertainty is corrected for, gross domestic product per capita, unemployment, the marginal tax rate, and the volatility of inflation are the only macro variables significantly and universally associated with aggregate entrepreneurship. Furthermore, the emphasis on inflation and taxation suggests that governments have the power to influence the quantity and distribution of entrepreneurial activity by setting incentives that are not entrepreneurship specific but overlap significantly with general and fundamental principles of economic stability.
AB - Entrepreneurship has long been seen as an important instrument in stimulating and generating economic growth. The amount of research trying to identify key factors that drive entrepreneurship is considerable; yet, little consensus has been achieved. We argue that this lack of consensus could be on account of model uncertainty as empirical studies often tend to be selective on what variables are included in the final model. Drawing on recent literature, we demonstrate the benefits of Bayesian model averaging (BMA) in reducing the impact of model uncertainty on empirical research in entrepreneurship. Additionally, BMA provides measures of variable importance and can be seen as a complementary approach to dominance/relative importance analysis. We show that when model uncertainty is corrected for, gross domestic product per capita, unemployment, the marginal tax rate, and the volatility of inflation are the only macro variables significantly and universally associated with aggregate entrepreneurship. Furthermore, the emphasis on inflation and taxation suggests that governments have the power to influence the quantity and distribution of entrepreneurial activity by setting incentives that are not entrepreneurship specific but overlap significantly with general and fundamental principles of economic stability.
KW - macro policy and entrepreneurial activity
KW - model averaging
KW - model uncertainty
KW - variable selection
UR - http://www.scopus.com/inward/record.url?scp=84922021563&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84922021563&partnerID=8YFLogxK
U2 - 10.1177/0149206314558488
DO - 10.1177/0149206314558488
M3 - Article
AN - SCOPUS:84922021563
SN - 0149-2063
VL - 41
SP - 607
EP - 631
JO - Journal of Management
JF - Journal of Management
IS - 2
ER -