The Revolving Door: Duration and Recidivism in IMF Programs
利用1974-2003年季度数据，通过风险分析和误差修正估计，发现参与IMF项目次数越多，新项目持续时间越短且间隔越短，支持“旋转门”假说。
Quarterly evidence on participation in IMF programs for the period 1974-2003 is examined using the techniques of hazard analysis and error-correction estimation. Three hypotheses are proposed and tested. An increase in cumulative prior participation in IMF programs is found to cause a reduction on average in the length of a new spell of participation. The length of time between participation spells is reduced significantly with an increase in prior participation in IMF programs. Reserve adequacy is shown to be significantly and positively affected by participation in IMF programs. The results support a "revolving door" explanation of participation in IMF programs. Copyright by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.