It is important but difficult to distinguish between desirable and undesirable effects of unemployment insurance (UI) that are observationally equivalent when designing optimal UI schemes. For example, a UI-induced rise in the wage rate caused by workers taking more time to match their skills with job vacancies is desirable. However, another view of the same observation is that UI causes permanently higher involuntary unemployment by raising the reservation wage. This paper avoids this problem by regarding the trade-off between the UI replacement rates and unemployment as an intermediate relationship that matters only as far as it impacts economic growth. An empirical analysis of UI replacement rates, unemployment rates, and growth rates using annual panel data finds UI replacement rates are associated with higher unemployment. However, no significant relationship is found between UI-related unemployment and the real growth rate of gross domestic product. An earlier version of this paper was presented at the Fiftieth International Atlantic Economic Conference, October 15-18, 2000, Charleston, South Carolina. Financial support from the Scottish Economic Society and the University of Stirling is gratefully acknowledged. The Organization for Economic Cooperation and Development generously provided access to their database on benefit entitlements and gross replacement rates. The authors are grateful to an anonymous referee for constructive comments.