Are IMF Loans More Effective without Policy Advice? A Switching VAR Assessment

Since the current global financial crisis started in 2008, the international Monetary Fund (IMF) has been gaining power again. Before this crisis, the IMF appeared to be sliding towards terminal irrelevance. For several years, emerging economies had been making fewer and fewer demands on its rescue services. The economic emergency room with nearly no patients a year ago is now back in place. But questions are swirling about just how effective the new-look IMF will be fighting crisis. Where once it demanded a dramatic policy change following its advice, the IMF is now taking softer stance and attaching few restrictions to its massive loans. Doesn’t the fund’s unwillingness to press for drastic policy shifts imply that borrowers miss opportunities to improve competitiveness? Would IMF loans really be more effective without policy advice? In this paper we develop a new approach to evaluate the effectiveness of IMF lending programs on the program country’s economy. Instead of including a program dummy in a static regression model in a conventional fashion, we set up a vector autoregression (VAR) model with a switching policy reaction function and estimate the system together with the program participation equation. In this way, we can estimate the loan effect and the policy advice effect separately. Surprisingly, the effectiveness of IMF programs appears to come largely from the policy shifts rather than from the loan itself.