This article documents time series evidence suggesting the case for a possible structural break in the role of Japan's monetary policy during the 1990s. It uses a simple vector autoregressive framework and offers some suggestive results: While a persistent effect of monetary policy on real output is detected over the full sample of 1975-1998 and the subsample that ends in 1993, such effect disappears with the recent subsample of the 1990s. The stability analysis also provides more specified evidence that there is a break in the reduced form dynamic system in 1995. Some interpretations are offered to intuitively support these findings. (C) 2000 Academic Press. Journal of Economic Literature Classification Numbers: E52, E32.