Robust Determinants of Public Sector Size in IRAN: Bayesian Averaging of Classical Estimates Approach

This paper examines the robust determinants of public sector size (government) in IRAN in the years 1979-2012 and the uncertainty condition of the model. For this purpose is used the Bayesian Averaging of Classical Estimates (BACE) approach due to favorable characteristics for the assumption of model uncertainty. With estimate 16,000 regression and Bayesian averaging from the coefficients, effective variables were identified. The results show that robust determinants of the public sector size were as follows: lag the government size, population, growth rate of oil revenues, the ratio of population less than 15 and over 65 years to total population, bureaucracy and economic globalization.