Beginning with a review of the adoption of inflation targeting in a broad group of countries, Paulin focuses on changes in the design of inflation-targeting frameworks in light of fifteen years of accumulated experience. Included in the discussion are the use of numerical targets and ranges, the policy horizon, supporting institutional policy structures, and communication, including the publication of forecasts. A recurring theme is how much flexibility an inflation-targeting regime allows. The article concludes that the changes made to the frameworks have been relatively modest since their adoption, but in concert with the improved credibility that has resulted from central banks meeting their inflation-control targets, they have allowed an increasingly nuanced response to economic shocks.
Since the Bank of Canada adopted inflation targeting in 1991, it has focused on a measure of core inflation as a shorter-term guide for monetary policy. When the targets were renewed in 2001, the Bank adopted CPIX as its measure of core inflation because of the advantages it offered. Leflèche and Armour review the experience with CPIX and whether the criteria used to select it in 2001 still favour the measure today. They describe the various measures of core inflation monitored by the Bank and evaluate them on the basis of the volatility of the components, the volatility of the core measures themselves, absence of bias relative to total CPI, predictive power, and certain practical criteria, including timeliness and credibility. They conclude that CPIX still satisfies all the empirical and practical criteria.
The conduct of monetary policy within an inflation-targeting framework requires the establishment of an inflation-target horizon, which is the average time it takes inflation to return to the target. Policy-makers have an interest in communicating this horizon, since it is likely to help anchor inflation expectations. This article focuses on the determination of the appropriate policy horizon by reporting on two recent Bank of Canada studies. The evidence suggests that the current target horizon of six to eight quarters remains appropriate. It is important to note that the duration of the optimal inflation-target horizon varies widely, depending on the combination of shocks to the economy. In rare cases when the financial accelerator is triggered by a persistent shock, such as an asset-price bubble, it may be appropriate to take a longer view of the inflation-target horizon.