Chronicle of the Liquidity Effect Foretold

Abstract

Conventional accounts of the immediate mechanism whereby the Federal Reserve raises short-term interest rates require the Fed to reduce the supply of reserves in order to raise interest rates. This paper shows in popular monetary models that although the Fed can have an immediate effect on interest rates purely by announcing a change in the target without changing reserves, these models imply that by the end of the maintenance period, one should see some combination of a drop in nonborrowed reserves, drop in excess reserves, or increase in borrowed reserves in order to ratify the Fed's intended interest rate hike. Using aggregate reserves of depository institutions data, this paper looks at what happens to various reserve measures during maintenance periods in which there was a change in the Fed funds target. An institutionally motivated time-varying parameter model is built to predict each monetary aggregate. Using three different measures of monetary policy, two of an unanticipated change in the funds target and the change in the target itself, a negative correlation between the target change and what happens to three of the five corresponding reserve aggregates is found. The sign and magnitude of the finding for the target is consistent with the standard results from the literature regarding the liquidity effect. Nevertheless, these results are not replicated when using the proxies for unanticipated changes.