Determinants of Economic Growth:
A Bayesian Panel Data Approach

Abstract

Model uncertainty hampers consensus on the key determinants of economic growth. Some recent cross-country cross-sectional analyses have employed Bayesian Model Averaging to address the issue of model uncertainty. This paper extends that approach to panel data models with country-specific fixed effects. The empirical results show that the most robust growth determinants are the price of investment goods, distance to major world cities, and political rights. This suggests that growth-promoting policy strategies should aim to reduce taxes and distortions that raise the prices of investment goods; improve access to international markets; and promote democracy-enhancing institutional reforms. Moreover, the empirical results are robust to different prior assumptions on expected model size.

JEL Classification: C11, C23, O4.
Keywords: Growth Determinants, Model Uncertainty, Bayesian Model Averaging, Dynamic Panel Estimation.

World Bank Policy Research Working Paper No. 4830. (Revise and Resubmit The Review of Economics and Statistics)