Quantitative growth economists often have to deal with model uncertainty (Barro et al. (2003)) and the issue of open-endedness of theories (Brock and Durlauf (2001)). Bayesian Model Averaging (BMA) is the best statistical tool to evaluate the variables to include in a growth regression.

This work aims to investigate the robustness of the determinants of growth in Europe from 2002 to 2019. Our dataset is composed of 70 explanatory variables for 19 European countries. We compare different BMA estimates by combining 2 model priors with 5 coefficient priors and we find that no variable is robust to all our specifications.

Our results support neoclassical growth theories, as the initial level of GDP per capita and savings are robust determinants of growth. Other robust determinants include the share of manufacturing in GDP, demography, public accounts, wage and labor contract regulation, and fixed capital accumulation.