View article

The average impact of the International Monetary Fund (IMF) programmes on the economic growth of the borrowing countries is analysed from a methodological approach not previously applied in this context: the method of matching and difference-in-differences matching. The special feature of this approach is that it controls for ‘selection on observables’ in addition to ‘selection on unobservables’. This second source of selection is what previous studies have tried to model. The data covers the period 1970–1990 and includes all non-major oilproducing developing countries that are members of the IMF. Due to data attrition I run parallel analyses for different data sets that vary with regards to the covariates included in the estimations, and thus the number of observations. Results indicate no positive effects of IMF programmes on per capita GDP growth for three consecutive years after programme participation.

* I am particularly grateful to Karl Ove Moene for his support and advice all along. I am also grateful to John Dagsvik, to Markus Frölich and to colleagues Hege Torp and Pal Schone for useful comments. Thanks also to Adam Przeworski and James R. Vreeland for providing the data used in the analysis. I would also like to thank Mario Damill and Jose M. Fanelli for enlightening discussions during my stay at CEDES (Centro de Estudios de Estado y Sociedad), Buenos Aires in 2000–2001. An earlier version of this paper was presented at LACEA conference, 11–13 October 2002, Madrid, Spain and at the Dep. of Economics, Univ. of Oslo. Financial support from the Norwegian Research Council and the Marshall Foundation is gratefully …