Borrowed Reserves, Fed Funds Rate Targets and the Term Structure

Abstract

When examined for the period 1985-1992 as a whole, the impact of changes in the targeted Fed funds rate on U.S. treasury bill rates has been weaker than during previous periods. The period, however, should be viewed as three separate regimes. First, I show significant differences between the Greenspan and Volcker eras arising from management style. Volcker emitted ambiguous signals to the market, which induced heterogenous expectations. Greenspan's market leadership forged more homogeneous expectations allowing treasury rate responses prior to the target change to be attributed to market anticipations. Further, I find strong annual seasonality in borrowed reserves targets, corresponding to an agricultural cycle. It suggests the Fed no longer defends a target for borrowed reserves and instead allows them to follow the cycle. Subsequently, the Fed funds rate adheres more closely to a target rate and Treasury rates react similarly to an earlier direct Fed funds target regime. Thus, I conclude that as of late 1989, the Fed again directly targets the Fed funds rate. Target rate changes induce starkly different reactions by treasury rates of various maturities over the three regimes. Applied to the term structure, I find that the average monthly interest rates implied by the expectations theory is consistent with the pattern of actual treasury rate reactions. I compare implied monthly rate reactions after a policy shock between direct Fed funds rate regimes of the latter Greenspan period and the 1975-1979 Fed. I find that after a policy easing, the market expects the monthly rate to remain at a lower level throughout a horzion of at least 20 years. This is consistent with expectations of a Fed that is able and willing to contain inflation: although easing, the public has confidence it will not inflate over the forseeable future.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 9 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.