Growth, Institutions and Productivity: An empirical analysis using the Bayesian approach

In this paper we explore how the environment offered by institutions influences long-run growth. In order for the estimation results to be trustworthy we control the reliability of the estimates in several ways. Firstly, we include the lagged level of output per worker in the model to control the effect of conditional convergence. Secondly, we use the production function theory to form an environment of other inputs which may affect the parameter value of institutions and handle the issue of endogeneity using convenient instruments for institutions and other inputs. Thirdly, we use institutional indicator which is built using 18 indicators which all reflect the ability of institutions to create an environment in which the citizens can manage their risks they encounter during their life time. Finally, we study the sensitivity of estimation and control the effect of outliers and bad quality of data using a subsample of 22 industrial countries in addition to the total sample of 86 non-oil countries. Our cross-country analysis - based on Bayesian inference - confirms that the production environment offered by institutions has a significant role on economic growth, but it does not seem as dramatical as some may have expected.