Abstract. This paper analyses the role of institutions in enhancing the economic efficiencies across countries in a two stage Double Bootstrap DEA framework based on nonparametric frontier analysis as proposed by Simar and Wilson (2007). In the first stage, cross country workers’ efficiency is estimated using a bootstrapped DEA approach over the period of 1990-2000 for 78 countries. In the second stage, the impact of institutions on these efficiency estimates is analyzed in a truncated bootstrapped regression. Twenty-nine institutional indicators from the same period have been utilized to extract three orthogonal factors based on principal component analysis. These factors namely institutional and policy rents, political rents and risk reducing technologies, along with their aggregated index are used as institutional variables. Findings suggest that inefficiencies in the economy are less where institutions are stronger. This study also shows that institutions that curb corruption, bureaucratic inefficiencies, lax regulations and unfriendly business policies, tend to have a larger effect on workers efficiency than other two indices that curb political rents and those that reduce transactional risks. Furthermore, when they are aggregated, the combined impact is more than the individual impact.

Keywords: institutions; DEA; efficiency; bootstrap.