Interest on Reserves and Monetary Policy Transmission

Using monthly U.S. data from 1967M1 to 2018M10, I estimate a structural vector autoregression with short-and long-run restrictions to test whether monetary base expansions had weaker effects on output and prices after the Federal Reserve began paying interest on reserves (IOR) in October 2008. I compare the macroeconomic effects of monetary base shocks across pre-IOR (1967M1-2008M10) and post-IOR (2008M11-2018M10) subsamples. In the pre-IOR period, a one-standard-deviation monetary base shock raises output by approximately 0.31 percent at its peak and produces a permanent increase in the price level of approximately 0.29 percent, with the output response significantly above zero for roughly 55 months.

In the post-IOR period, the same shock produces muted and statistically imprecise responses for both variables. Forecast error variance decompositions reveal a complementary shift: monetary policy shocks account for 86 percent of monetary base variation on impact in the pre-IOR period but only 44 percent in the post-IOR period, with money demand shocks absorbing the difference. These results are consistent with a weakening of the aggregate transmission of monetary base expansions to the real economy.