Stop Interrupting: An Empirical Analysis of the Implementation of IMF Programs

Summary For many years the analysis of IMF conditionality overlooked the extent to which it was implemented. However, more recently, increasing attention has been paid to implementation. Theoretical contributions have focused on the importance of special interest groups, but empirical evidence has failed to provide compelling support for the theory. Indeed, empirical studies have reported mixed results that sometimes seem to be conflicting. This paper identifies a range of economic, political and institutional factors that may, in principle, influence implementation. Focusing in particular on the irreversible interruption of IMF programs, it tests an econometric model designed to capture these influences 'over 1992-2004' exploiting improved sources of data. The results suggest that significant determinants of interruption are trade openness, the existence of veto players and the amount of resources committed by the Fund. The paper interprets the results, tests their robustness, briefly examines cases that initially appear to be inconsistent with the overall findings and discusses the implications of the results for policy.