IMF programs: Do they work? Can they be made to work better?

A central issue in the debate about a new international financial architecture has been whether programs of policy reform supported by the International Monetary Fund work. The Fund claims that "on balance" they do, because of their positive effect on the balance of payments. Others claim that programs are ineffective, and suggest that they should be discontinued. This paper reviews the econometric evidence dealing with the macroeconomic effects of IMF programs. It goes on to provide additional evidence and judges success against alternative criteria. Although the record is not good, the paper argues that it would be unwise for the Fund to cease lending and to abandon conditionality altogether. IMF programs need to be redesigned and refocused. The paper concludes by identifying a number of principles that should underpin reform. (C) 2001 Elsevier Science Ltd. All rights reserved.