Interest on reserves and monetary policy transmission

Abstract

Using monthly U.S. data, I estimate a structural vector autoregression with short- and long-run restrictions to compare the effects of monetary base shocks before and after the introduction of interest on reserves (IOR) in October 2008. The subsamples cover January 1967–October 2008 and November 2008–October 2018. Before IOR, a one-standard-deviation shock raises output by approximately 0.31% at its peak and the price level permanently by approximately 0.29%. The 68% error band for output is above zero for 57 of 60 months. After IOR, both responses are muted and statistically imprecise. Monetary policy shocks account for 86.5% of monetary base variation on impact before IOR and 51.1% afterward, with money demand shocks accounting for the difference. These findings are consistent with weaker aggregate transmission of monetary base expansions to the real economy.