FRBSF Economic Letter
1996-25 | September 9, 1996
Two recent news stories offered examples of dramatically contrasting relationships between a government and the authority charged with monetary policy. In Russia, President Boris Yeltsin pressured the Central Bank of Russia into providing $1 billion for new government spending, even though officials of the central bank protested that Yeltsin’s demands were a threat to the bank’s independence.
In New Zealand, the inflation rate, as measured by the Reserve Bank of New Zealand’s estimate of underlying inflation, was 2.1 percent for the year ending March 1996. When this statistic was released, the Minister of Finance immediately issued a press statement demanding an explanation for this breach of the Reserve Bank’s 0-2 percent inflation target. Both the Chairman of the Non-Executive Directors and the Governor of the Reserve Bank responded with public letters in which they expressed their concern over the breach and explained that policies were in place to bring inflation back within the target range.
It is hard to imagine a sharper contrast: In one case, a politically dependent central bank; in the other, a central bank with an externally mandated goal to which the elected officials can hold it accountable. And both stories offer a contrast to the situation in the U.S., where the Fed has the independence to resist heavy-handed political pressures but lacks a clear mandate for which it can be held accountable.
The desire for greater central bank independence and for some means of assessing central bank performance helps to explain why many countries are establishing explicit inflation objectives for monetary policy. The popularity of inflation targeting is based on the belief that explicit targets lend credibility to low-inflation policies and that explicit targets provide a transparent means of measuring the central bank’s performance.
This type of goal dependence–in which the explicit goal of monetary policy is defined by the legislative process–combined with the freedom to conduct policy to achieve the stated goal–often characterized as instrument independence–does provide a natural measure of the success or failure of policy. But what actually happens if the goal is not achieved? This question is particularly appropriate when, as is the case in some countries, the central bank itself has adopted an inflation targeting objective or where the target inflation rate is based on a measure produced internally by the central bank.
A review of New Zealand’s recent experiences with inflation targeting, focusing on whether the policy reforms there have succeeded in establishing a framework with clear goals and a mechanism for accountability, offers some lessons for other countries moving towards the adoption of targets for inflation.
Many nations have revised their central banking structures over the past five years, but New Zealand’s Reserve Bank Act of 1989 continues to remain a focus of interest for several reasons. First, New Zealand was in the vanguard of the recent wave of central bank reforms, and as such provided an early example for other nations looking to insulate their central bank from direct electoral influences. Second, the New Zealand approach is the most explicit attempt to establish a policy structure that includes means to ensure accountability. Third, and perhaps most importantly, New Zealand has enjoyed remarkable success in achieving and maintaining relatively low rates of inflation, rates that certainly represent a contrast from its earlier history. From 1973-1985, for example, New Zealand’s inflation rate averaged over 12 percent, compared to roughly 7 percent in the U.S. And while average inflation in the U.S. averaged only 3.8 percent from 1986 to 1992, it was 6.9 percent in New Zealand. However, between 1992 and 1994, the inflation rate in New Zealand, at 1.35 percent, was less than half that experienced in the U.S.
The basic outlines of the New Zealand reforms are well known (see, for example, Walsh 1995). Price stability is defined by the legislative act as the sole objective of monetary policy, and the actual implementation of policy is guided by the Policy Targets Agreement (the PTA) between the Minister of Finance and the Governor of the Reserve Bank. A mechanism for holding the Reserve Bank Governor accountable was written into the Act by (1) making the Governor directly responsible for the conduct of monetary policy, and (2) allowing for the Governor’s dismissal if the goals established in the PTA are not met.
The PTA’s have defined the target to be 0-2 percent inflation as measured by the Consumer Price Index (CPI). Exceptions that would allow the range to be exceeded include various supply side shocks (increases in indirect taxes or government charges, livestock diseases, terms of trade shocks, etc.). In such cases, the Reserve Bank Act allows the government and the Reserve Bank to negotiate a new PTA.
This structure makes responsibility clear: The Governor, and not a committee such as the FOMC, is responsible for policy decisions, and success or failure to fulfill the goals of the PTA are transparent–the inflation rate as measured by the CPI is frequently available and widely publicized.
How has this structure worked in practice? Spiegel (1995) argued that the Reserve Bank of New Zealand appeared to be operating under a “two-tiered” system: As long as CPI inflation remained below 2 percent, the Reserve Bank enjoyed full independence; whenever inflation rose above 2 percent, however, some degree of independence was lost because the Reserve Bank was forced to justify its policies.
During the last two years, however, this is not quite how the system has worked. As CPI inflation rose above the 0-2 percent range in late 1994, the Reserve Bank focused more and more on its own measure of inflation, the so called “underlying rate.” And the CPI inflation rate started being referred to in Reserve Bank publications as the “headline” inflation rate.
At the time, there were reasonable grounds for skepticism over these developments. After all, one of the attractive features of the New Zealand policy framework was its carefully designed mechanism for accountability. Letting the Reserve Bank define its own inflation rate, especially one that looked like it would peak at just under the 2% upper limit allowed under the PTA, would appear to have threatened the whole notion of accountability.
In fact, however, the Reserve Bank has maintained its credibility. The two-tiered system Spiegel identified is now based on the Reserve Bank’s own measure of inflation. In response to the recent very small deviation of the underlying rate above 2 percent, the Bank Directors stated that they “regret the 0 to 2 percent underlying inflation target range has been breached…” (Sir Peter Elworthy to Rt. Hon. W.F. Birch, Minister of Finance, 19 April 1996) and they publicly reaffirmed their commitment to the 0-2 percent range. The Directors also affirmed their support for the job being done by the Governor, stating that the breach should not “call into question the Governor’s performance or his continued employment.”
Notice that the statement says that the inflation target range applies to “underlying inflation”; that is, the Reserve Bank is to be judged by whether a statistic that it constructs remains within the target range. Interestingly, underlying inflation for the year ending July rose further to 2.3 percent, while headline inflation fell to 2.0 percent.
A policy regime with a clear goal, defined in terms of an easily observed measure, facilitates policy accountability. But the use of actual inflation, whether measured by the central bank or not, raises a variety of implementation problems.
One problem arises because the CPI may not be the appropriate measure of inflation. In New Zealand, for example, the inclusion of interest costs in the CPI means that contractionary monetary policy–policy that boosts interest rates in the short term–increases inflation as measured by the CPI. This provides a misleading indicator of policy and is a major reason the Reserve Bank shifted focus away from the CPI. There are also many factors in addition to policy actions that affect the inflation rate, so the central bank’s ability to control inflation is imperfect, and the central bank’s performance evaluation should not be based on the effects of unforeseen events.
But perhaps most importantly, a breach of the target range reveals something about monetary policy a year or more earlier; it says little about whether current policy is appropriate. Thus, while using the actual variable in terms of which the goal of policy has been defined as a measure of performance has much to commend it, it also has problems: It can be misleading or it can simply provide an assessment of policy that comes too late.
Lars Svensson (1996) has suggested that inflation forecasts may solve some of the difficulties that arise in using actual inflation in an inflation targeting system. Inflation forecasts have all the desirable properties of a good intermediate target. For example, such forecasts are highly correlated with the variable of ultimate interest, i.e., the actual rate of inflation. Furthermore, they are more easily controllable than the actual rate of inflation. In fact, just as was the case with the Reserve Bank Act of 1989, perhaps New Zealand is again out in front. According to a description of policy by Mayes and Riches (1996) in the Reserve Bank’s Bulletin, “The current operational framework employed by the Reserve Bank is based directly on forecasts of inflation” (p. 7).
If the central bank’s forecasting model is made public, inflation forecasts could provide a transparent measure of how well the central bank is doing. Of course, central banks might have an incentive not to reveal their true forecasts, but in this case, market expectations or private forecasts of inflation could play a role in assessing the job being done by the central bank.
During the 1990s, New Zealand has achieved one of the most enviable inflation records, and several lessons can be drawn from this experience. First, the establishment of explicit goals contributes to the transparency of monetary policy. But these advantages are diminished if performance is based on internally generated statistics. The successful period from 1991 until 1994 in which “headline” inflation was brought down to and kept within the 0-2 percent range was undoubtedly critical in establishing the policy credibility needed for markets to accept the Reserve Bank’s “underlying” inflation measure. So despite the formal reforms in New Zealand, the Reserve Bank still had to earn credibility by actually delivering low inflation.
Second, a strict inflation targeting regime is probably not feasible because it ignores the effects of supply side disturbances to the economy. That is why the Reserve Bank has been successful in shifting the debate away from “headline” inflation and onto its measure of underlying inflation. This implicitly allows it to adjust its inflation target in light of supply side disturbances while still maintaining the transparent framework provided by inflation targeting.
Third, inflation targeting, in practice, is likely to mean inflation forecast targeting. So either market-based expectations or the central bank’s own internal forecasts, made public, should be used to assess the conduct of policy. Perhaps the introduction of inflation-indexed bonds in the U.S., which would allow a market-based forecast of inflation to be derived from the interest rates on indexed and nonindexed bonds, will serve a useful role in providing an assessment of U.S. monetary policy.
Carl E. Walsh
Professor of Economics
University of California, Santa Cruz
Federal Reserve Bank of San Francisco
Mayes, David G., and Brendon Riches. 1996. “The Effectiveness of Monetary Policy in New Zealand.” Reserve Bank of New Zealand Bulletin 56 (1) (March) pp. 5-20.
Spiegel, Mark. 1995. “Rules vs. Discretion in New Zealand Monetary Policy.” FRBSF Weekly Letter 95-09 (March 3).
Svensson, Lars E.O. 1996. “Inflation Forecast Targeting: Implementing and Monitoring Inflation Targets.”Institute for International Economic Studies, Stockholm (June).
Walsh, Carl E. 1995. “Is New Zealand’s Reserve Bank Act of 1989 an Optimal Central Bank Contract?” Journal of Money, Credit and Banking 27(4) pt. 1 (November) pp. 1179-1191.
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.
Please send editorial comments and requests for reprint permission to
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120