FRBSF Economic Letter
1997-04 | February 7, 1997
Beginning in the early 1990s, price stability became an increasingly important goal of the monetary authorities in many countries. But some central banks found the traditional approaches–namely, influencing inflation and economic activity by controlling intermediate variables like monetary aggregates or an exchange rate–not very successful. As a result, these central banks faced the serious possibility of losing credibility.
To address this problem, several industrialized countries–New Zealand (1990), Canada (1991), the United Kingdom (1992), and Sweden (1993)–adopted monetary policy regimes that target inflation directly. These regimes are said to be “transparent,” and therefore more credible to the public, because the central bank makes an explicit commitment to conduct monetary policy to meet a specified numerical inflation rate target within a specified time frame. The explicit target provides an anchor for monetary authorities, and it also serves as an anchor for private market expectations.
This Letter discusses the rationale for inflation targeting (IT) and examines how it has been implemented, with special emphasis on the U.K. In addition, it shows that while these countries have had low and stable inflation since adopting IT, indicators of long-term inflation expectations also reveal that there still remain doubts in the market about whether the recent performance will be extended well into the future.
Many central banks historically have targeted an intermediate variable, such as a monetary aggregate or an exchange rate, to achieve their ultimate goals of low inflation and sustainable growth. This procedure assumes both that the intermediate target variable is controllable and that there is a reliable relationship between the intermediate target variable and the ultimate goals.
Unfortunately, this approach has not been uniformly successful; for example, many of the economies of the G-7 countries have had recurring booms and busts and flare-ups of inflation over the past several decades. There are a number of reasons why the approach has not always worked so well. For one thing, at times the central bank cannot pursue both goals–high output and low inflation–simultaneously. Second, this conflict in the goals leads to uncertainty among the public about which goal has top priority; such uncertainty undermines the credibility and hence the efficacy of anti-inflation policy, typically at a time when the public’s confidence in the policy is most crucial for it to work. Finally, the way in which intermediate variables affect output and inflation has varied over time.
Inflation targeting could help alleviate these problems. By explicitly stating that maintaining low inflation is the main goal, monetary authorities reduce uncertainty about the priority of alternative policy goals. Also, the question of the stability of the relationship between intermediate targets and the objective of price stability becomes irrelevant, since price stability is being targeted directly; rather, money aggregates and exchange rates are taken simply as indicators of market conditions.
Before adopting IT, both the U.K. and Sweden had followed fixed exchange rate regimes. For brevity, this section will focus on the U.K. experience only. During most of the 1970s, the U.K. had high inflation and high government budget deficits. The Thatcher administration, which won the general election in 1979, aimed to curb inflation by controlling deficit spending and money growth. Accordingly, the British monetary authorities tried to follow a policy of ensuring smooth and predictable growth in broad money over a medium-term horizon. By 1983, these stabilization efforts brought down inflation in consumer prices to below 5%.
Starting in the second half of the 1980s, however, the monetary policy focus started to shift to exchange rates for several reasons. First, by then, erratic behavior in the velocity of the broad aggregates caused by financial innovations made them unreliable. Second, the pound sterling steadily appreciated against the German mark around the time of the Plaza Accord in 1985 and the Louvre Accord in 1987; under these accords, the major industrialized countries agreed to lower the value of the dollar and to maintain stability in other key exchange rates. Third, partly in reaction to the worsening current accounts situation, they let the exchange rate depreciate, and thereafter sought to stabilize it by shadowing the ERM. This emphasis on the exchange rate culminated in the pound’s entry in the Exchange Rate Mechanism (ERM), which formally prescribed keeping the pound sterling within a narrow band, in October 1990.
During this time, the inflation performance started to deteriorate. A surge in real GDP of 4 to 5% annually between 1985 and 1988 led to accelerating inflation by 1988; by 1990, the inflation rate had reached 11%, but at the same time, the economy was starting to slow. In 1991, real GDP actually contracted by 2%, but the U.K. had little room to ease policy as it had to defend the pegged sterling exchange rate in the face of high German interest rates around the time of Germany’s unification. This situation became more and more untenable, and in September 1992 sterling left the ERM when it came under overwhelming pressure caused by large-scale selling of sterling in the foreign exchange markets. (Sweden left the ERM under similar circumstances.) This withdrawal left no nominal anchor to guide monetary policy and to stabilize expectations. To remedy this situation, the Chancellor of Exchequer announced the adoption of IT.
Implementation of this framework involves first setting a goal in terms of a specific measure of general price inflation. The consumer price index (CPI) is the most common choices as it is perhaps the most familiar measure for the general public. However, a central bank may want to exclude certain components of the index that are inherently problematic for effective inflation targeting. A clear example is housing costs: While the index should go down when monetary policy is tightened, the housing component goes up because it embodies the rise in mortgage interest rates. Thus, many IT countries use the CPI excluding mortgage payments or a housing cost component (e.g., the United Kingdom’s RPIX, retail price index excluding mortgage payments). Similarly, others, such as New Zealand and Canada, exclude commodity prices from their targeted price index on the ground that price changes in commodities like oil are beyond the control of monetary authorities. Countries also differ in their treatment of value added taxes whose changes affect the CPI but are determined by fiscal policy and not monetary policy.
Once a price index is chosen, a target is enumerated in terms of a fixed number and a band. A timetable for achieving the goal also is announced. In the case of the United Kingdom, the goal is to keep the 12-month change in the retail price index at 2 1/2% or lower where the target range is between 1 to 4%. When this goal was first adopted in October 1992, the U.K. government announced its intent to achieve it by spring 1997, when the current parliament’s term ends. Later the current Chancellor extended the goal indefinitely.
This range might appear somewhat broad in light of the low inflation seen in many OECD countries in recent years. However, the British inflation rate has fallen inside this band only about 35% of the time since World War II. For the U.S., the comparable proportion is about 50%. Even in Germany, known for its exemplary inflation performance, the inflation rate has fallen in this band only about 60% of the time. Hence, the British target range is not so permissive when seen from a long-term perspective (Bank of England Quarterly Bulletin 1994). Canada and New Zealand originally adopted more stringent target ranges of 1-3% and 0-2%, respectively. In Canada the current time frame runs up to 1998; in New Zealand it is renewed periodically; in Sweden, the target since 1995 has been 2% with a 1% tolerance band around it.
In addition to the public announcement of goals, most countries that have adopted IT also have granted their central banks greater autonomy and have implemented measures to enhance transparency in monetary policy deliberations. Before adopting IT, central banks in those countries were less autonomous than the central banks of the U.S. or Germany, and policy had been formulated through a consultative process in which the Treasury or Finance Ministry held sway. With the adoption of IT in Canada and New Zealand, a formal arrangement between the Treasury and the central bank clearly delegated operational authority and assigned the responsibility of achieving the goal to the central banks.
No such clear delegation took place in Britain, but measures to increase the transparency of policy have been implemented. A regular monthly consultation between the Chancellor of Exchequer and the Governor of the Bank of England was formally instituted, and minutes of each meeting have been routinely published with about a six-week lag since April 1994. The minutes have shown that recently the Governor disagreed with the Chancellor’s decision on several occasions because of different assessments of incipient price pressures. In addition, at the suggestion of the Chancellor in 1992, the Bank of England started to publish a quarterly report in which the Bank offers an inflation forecast based on the assumption that the current monetary policy stance persists. The Bank uses it as an important vehicle to offer an objective assessment of whether the current policy stance is compatible with achieving the IT goal. For example, the August 1996 issue of the Bank of England’s Inflation Report offered the following assessment: “The Bank’s latest view on inflation two years ahead shows a central projection for RPIX inflation a little above 2 1/2% and rising … the implication of this projection is that a tightening of monetary policy will be necessary at some point to achieve a better-than-even chance of keeping inflation below 2 1/2% in the medium term” (p. 3).
In most countries mentioned in this Letter, inflation in the targeted price indexes has remained within the target range in the past year or so. As a group, the average inflation in IT countries was lower than that for other OECD countries for several years until 1995. However, this better performance is not fully reflected in indicators of long-term inflation expectations, such as long-term interest rates, in some of these countries. Owing to its good inflation track record, Germany’s long-term interest rates contain a relatively small expected inflation and inflation risk premium compared to other countries. Thus, one could gauge how much additional expected inflation and risk premium exist in Britain, for example, by looking at the difference between the yields on 10-year securities in Germany and Britain. As shown in Figure 1, the British yield has been higher than the German counterpart throughout but has narrowed during the 1990s IT period. However, the yield difference between the two has not narrowed to the extent that seems to be justifiable by Britain’s recent favorable inflation performance. Survey measures of inflation expectations in Britain and Sweden also have edged lower at a very slow pace in the same period. Thus it appears that while IT has helped gain credibility for U.K. monetary policy, the markets are not yet fully convinced that policy will stick with its current stringent IT approach in the long run.
Inflation targeting appears to have provided a successful nominal anchor for conducting monetary policy in the countries that have adopted it so far. However, indicators of long-term inflation expectations still seem to reveal lingering doubts about whether current low inflation will be extended into the future. This suggests how difficult it can be to establish the credibility of monetary policy–while institutional changes may enhance credibility, it will take years of solid performance to earn the public’s full belief in the central bank’s commitment to low inflation.
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