Abstract
Recent theoretical and empirical research suggests that under certain conditions IMF agreements induce additional inflows of finance from private international capital markets. This article provides new empirical evidence on this catalytic effect using a treatment effects model to correct for selectivity. It concludes that catalysis remains weak or negative overall, with nuances that support recent theory.
Acknowledgements
The authors wish to thank the British Academy for financial assistance and Mejlina Modanu and Jano Bourgeois for research assistance.
Notes
1 Specific data definitions, sources and sample countries are available from the authors.
2 Two observations on the Seychelles were dropped from the sample because of their unusually high capital flows (over 250% of GDP) in 1978 and 1980. These proved to be influential outlier observations, with their inclusion leading to a highly significant negative catalytic effect amounting to over 15% of GDP.
3 A Hausman test confirmed that the fixed effects estimates are preferred to the random effects estimates for this configuration of the equation. Lagged growth had to be dropped as its presence compromised the integrity of the Hausman test.