On the determinants of central bank independence in open economies

Empirical studies (Romer, 1993, Campillo and Miron, 1997) have suggested that Central Bank Independence does not affect inflation performance in the long run, casting doubts on the idea that commitment is empirically relevant in many sub samples of countries. We test some positive implications of the commitment hypothesis for the design of monetary institutions in open economies, by studying the determinants of central bank independence on a sample of 55 countries, for the period 1980-89. We document fairly consistent empirical evidence in support of the hypothesis that strategic commitment is indeed important to understand cross country variation in the level of central bank degree of dependence. We also address the related question (Romer, 1993) why only highly industrialized countries have relied on such a solution to the inflationary bias of monetary policy whereas others have not. Data suggest that the answer is related to the presence and the size of world-wide common features in the business cycle in each country.