The last time I wrote a short take about inflation targeting was in April 2001. It was mostly an international phenomenon in places like New Zealand and Australia. But it was also practiced in some form in the UK and Canada as well as that 'upstart', the European Monetary Union. It is also a feature of Scandinavian monetary policy as well as for many Latin American countries. Now it is increasingly talked about as a strategy for the U.S. Federal Reserve and even the Bank of Japan.

The catalyst for the shift in monetary policy began in the 1990s, a period of considerable reform and innovation in central banks across the world. Many new central banks were established and many existing ones were given expanded responsibilities along with greater independence from their governments. Independence was often in exchange for a clear commitment to meet specific targets for inflation. Early inflation targeting usually included government involvement in setting an appropriate target. For example Canada, which adopted inflation targeting in 1991, used a joint announcement by the Ministry of Finance and the governor of the Bank of Canada. This announcement, and subsequent joint announcements, established a target range for price stability but left the details and the responsibility for policy in the hands of the Bank.

The first country to adopt inflation targeting was New Zealand. And it did so with strong support and fairly specific instructions from its legislature. The Reserve Bank of New Zealand Act of 1989 established the basic framework. This act gave the central bank the objective "of achieving and maintaining stability in the general level of prices," with "regard for the efficiency and soundness of the financial system." The act provided that the government and Reserve Bank jointly determine the specific inflation target and other policy objectives through Policy Target Agreements. The first of these target agreements defined price stability as a range of 0 percent to 2 percent in New Zealand's Consumers Price Index (CPI), set a goal of achieving price stability within two years, and gave conditions that could justify going outside the price stability band. Subsequent agreements have modified the target CPI, widened the tolerance band to zero to 3 percent, and introduced other objectives such as the stability of output growth. These modifications have generally made New Zealand's inflation-targeting regime much more flexible.

The UK adopted formal inflation targets in 1992 after the country left the Exchange Rate Mechanism (a monetary union prior to the formulation of the European Monetary Union). This marked a decisive break with the past. And in 1997, the Bank of England was granted operational independence and an institutional framework was put in place that entrenched and enhanced the credibility of inflation targeting. In the UK as in many other inflation targeting countries, a track record of success has been built up for more than a decade and has reinforced the credibility of these targets. As a result, people and firms have increasingly come to expect inflation to stay close to the official target - a belief which itself helps to keep it there. Setting the target is the responsibility of the Chancellor of the Exchequer.

The government was less explicitly involved for the next group of countries adopting inflation targeting. In Sweden, which adopted inflation targeting in 1993, the government had previously announced that controlling inflation was an overriding goal for the Riksbank. In Australia, the Reserve Bank governor announced its inflation targeting regime in a speech, using broad, previously delegated authority. In Norway, the Norges Bank operates under a governmental mandate declaring that the long term objective of monetary policy is to maintain the domestic and international value of its currency. The ECB operates under authority mandated by the Treaty Establishing the European Community and with the primary goal of price stability established by the treaty.

What is inflation targeting?
The primary goal for most major central banks is to contain inflation. Inflation targeting generally identifies price stability as the primary objective in monetary policy. An explicit numerical target for inflation is set including a time period over which any deviation should be corrected. While the theme is common, there are various ways in which the target can be defined. But they boil down to either a range or a specific numerical target or a combination of both. Some are dogmatic in pursuing the target while others are more flexible and include other considerations besides inflation into their monetary policy objective. Other goals such as sustained growth and employment growth are secondary in many cases. However, two major central banks have yet to establish an inflation target as a specific goal - The Bank of Japan and the U.S. Federal Reserve. While both may have unofficial goals for price increases, there is no formal target.

The target is defined using a specific price measure. The most common among the practitioners is the year-on-year percent change in the consumer price index. While some use the total CPI, others use a core measure, usually excluding food and energy. Another criterion is the span of time during which inflation is expected to return to the established target. There are vast differences here, anywhere to one year to over the span of the business cycle. Some variations do provide escape clauses related to the pace of return to price stability.

U.S. monetary policy focuses on growth and employment while containing inflation. The Federal Reserve has been under pressure to consider using an inflation target and has studied the topic intensively. Federal Reserve Act calls on the Fed to maintain growth of credit and the money supply "commensurate with the economy's long-run potential to increase production, so as to promote effectively the goals of maximum employment, stable prices, and moderate long-term interest rates." The Fed's mandate states directly that monetary policy should aim at having the U.S. economy operate at full capacity. And politically, the Fed would have a hard time convincing Congress that they have not abandoned the dual mandate. The Fed is said to prefer the personal consumption expenditure (PCE) price index rather than the consumer price index as an inflation guide. The PCE is preferred because it covers a greater range of consumer expenditures. But others prefer the CPI. The CPI is not subject to revisions while the PCE is - it is part of the U.S. GDP calculations and would be an unstable target. In addition, about 75 percent of the data used to calculate the PCE is derived directly from the CPI.

Unlike other central banks, the Bank of Japan had another issue to deal with - deflation. Some urged the BoJ to establish an inflation target as a means of ridding the economy of deflation. But the Bank took an unorthodox approach - quantitative easing - which flooded the market with cash and a zero interest rate policy. In March 2006, the Bank declared that the deflation fight had been won and began to withdraw the surplus funds from the marketplace. And in July 2006, the Bank increased their policy rate by 25 basis points to 0.25 percent.

Types of inflation targeting
The three more prominent monetary policies that rely on inflation targeting are those of the European Central Bank and the Banks of England and Canada. The ECB is mandated by treaty to target inflation, while the Bank of England has its inflation target set by the Chancellor of the Exchequer in his budget message. The Bank of Canada's target range is decided by the Bank and the government. Both the ECB and Bank of England use a specific numerical inflation target - 2 percent - but with a difference. While the ECB sets a ceiling above which inflation should not climb, the Bank of England tends towards its target but allows a range of plus or minus 1 percent around the target. Should inflation exceed/drop more than the 1 percent, they are required to write a letter to the Chancellor explaining why this happened. The Bank of Canada uses a range of 1 to 3 percent and tends toward the 2 percent mid-point.

What is the best monetary policy for growth?
Over the years, central banks have tried various frameworks in attempting to find a way to sustain growth. First, central banks tried fixing exchange rates to gold; most later tried fixing their exchange rates to those of other countries. Some central banks tried to target credit or the growth of monetary aggregates, while many relied solely on their own judgment. Suffice it to say that all of these frameworks have had their problems.

Over time, many think that the best way for monetary policy to promote sustainable economic growth is to anchor expectations about the future purchasing power of money. Bitter experience of many countries is that when monetary policy chases short-term goals, mistakes are made, uncertainty is increased, and fluctuations in economic activity are aggravated. Focusing on domestic price stability is the best contribution monetary policy can make to economic stabilization and sustainable long-term growth.

Measures of inflation for targeting purposes
Banks tend to use different measures of inflation for policy purposes. And all are calculated somewhat differently. ECB uses the harmonized index of consumer prices (HICP), the inflation measure for the European Union. The Bank of England recently changed its inflation measure from the retail price index excluding mortgage interest payments to a consumer price index modeled after the HICP. The Bank of Canada uses a consumer price index as well.

In a 2005 speech, David Dodge, governor of the Bank of Canada, explained the choice of the CPI. He said that "the key reasons were that it is widely understood and is the measure of inflation most familiar to Canadians. Choosing a well-known indicator as a target makes it easier to explain our actions and to be accountable to Canadians. However, movements in the prices of particularly volatile components of the CPI can cause the index to fluctuate sharply. So we use a measure of core inflation as an operational guide. This measure strips out the most volatile components and the effect of changes in indirect taxes on the rest of the index, giving us a better understanding of the trend of inflation." The Bank of Canada clearly states that it is using a symmetric target and is not fixated by a specific numerical target. They worry just as much about inflation falling below target as they do about it rising above target. The symmetry is an answer to the charge that central banks target inflation at the expense of growth.

Bottom line
Over their short history, inflation-targeting regimes have evolved to give central banks greater flexibility in conducting monetary policy. Language suggesting the Banks are entirely ignoring output stabilization is viewed as an extreme position and not a desirable option. It has been argued that there has been a convergence toward "flexible inflation targeting." This means that inflation targeting systems, in practice, take into account deviations in both output and inflation from their respective goals. Such an evolution has brought many inflation targeting regimes closer in practice to a dual mandate, not unlike the United States.

Experience with inflation targeting has been universally positive. No country that has adopted it has turned back yet. Inflation targeting has already weathered several financial sector shocks, commodity price shocks as well as large currency movements. Importantly, different types of disturbances require different responses - and inflation targeting has shown it can accommodate them.

This short take article is adapted from "International Economic Indicators and Central Banks" which was written by the author and will be published by John Wiley and Sons in the first quarter of 2007.

Anne D. Picker, Chief Economist, Econoday