Academic journal article Journal of Money, Credit & Banking

The Optimal Design of Interest Rate Target Changes

Academic journal article Journal of Money, Credit & Banking

The Optimal Design of Interest Rate Target Changes

Article excerpt

MOST CENTRAL BANKS currently use interest rate targets as their operating objective in the implementation of monetary policy. Central banks in Australia, Canada, Japan, and the U.S. all target an overnight interbank interest rate, while in most other countries short-term interest rates (tender rates) are targeted. This paper asks: "What is the optimal form for such interest rate targeting, given the objectives facing central banks?" The question is motivated by the practices of the Federal Reserve, which has long targeted the federal funds rate, either directly or indirectly. Goodfriend (1991) suggests the Fed targets the fed funds rate to achieve its ultimate policy objectives, but in doing so it is careful not to "whipsaw the market" and waits till sufficient information has been accumulated before changing the target rate. In particular, he notes that adjustments to the target rate are made at irregular intervals in relatively small steps. Some target changes occur in relatively rapid succession, but when this occurs the changes are in the same direction; target changes are not soon reversed. He also suggests target changes are essentially unpredictable at forecast horizons longer than a month or two. Rudebusch (1995) has confirmed these features empirically for two periods of explicit funds rate targeting: from September 1974 to September 1979 and from March 1984 to September 1992. He shows that target changes are conducted in small, standardized steps with an approximate equality in the size of increases and decreases and that in the first few weeks after a target change the Fed is fairly likely to change the target again in the same direction (but very unlikely to reverse its previous change), while after five weeks without a change there is only a small (but equal) likelihood of an increase or decrease in the target rate.

These practices are not unique to the Fed. The Bank for International Settlements (1998, p. 68) examines 12 industrial countries that target interest rates and finds that central banks generally move interest rates several times in the same direction before reversing policy and that the interval between policy adjustments is typically considerably longer when the direction is changed. The Bank for International Settlements' data end in March 1998, but starting dates vary across countries. Using consistent starting dates and longer sample periods for five of these countries, Goodhart (1997) reports similar findings. Given the range of sample periods and countries considered, these findings span a variety of institutional and macroeconomic environments, suggesting a common explanation may be appropriate. Yet, despite the prevalence of interest rate targeting, no model exists that explains this set of puzzling facts. (1) This paper provides such a model.

The model we develop is based on the following assumptions. At each point in time the central bank has a preferred level of some short-term interest rate that is determined by its ultimate objectives (such as inflation and output). (2) We start by taking the simplest case and assume this preferred rate evolves according to driftless Brownian motion. This assumption shows that we do not need to assume mean-reversion in the preferred rate in order to generate positive autocorrelation in target rate changes. (3) The central bank is assumed to suffer flow costs that are quadratic in the difference between the actual interest rate and this preferred level. With just this assumption, the central bank's optimal policy would be to make the target rate exactly track its preferred rate, thus adjusting its target rate by infinitesimal amounts at every point in time. However, we do not observe this in practice. Instead, we observe infrequent discrete changes that suggest the central bank faces some costs to changing its interest rate target.

We think of these adjustment costs as arising from the central bank's additional role of maintaining orderly financial markets. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.