Seminar on the Statistical Implications of Inflation Targeting

artwork from cover of publication  

Statistical Implications of Inflation Targeting
Getting the Right Numbers and Getting the Numbers Right

Papers presented at an IMF Seminar held in Washington, D.C., February 28–March 1, 2002

Editors: Carol S. Carson, Charles Enoch, Claudia Dziobek

©2002 International Monetary Fund
September 25, 2002

Ordering Information


Foreword, Acknowledgments, and Abbreviations

Part I. Information Requirements for Inflation Targeting

Introduction to Part I

  1. Anchoring Inflation Expectations
    Horst Köhler

  2. The Bank of Canada's Approach to Inflation Targeting
    Malcolm Knight, Robert Fay, and Brian O'Reilly

  3. Statistical Implications of Inflation Targeting in South Africa
    Pali J. Lehohla and Annette Myburgh

  4. Informational Requirements of Inflation Targeting in the Philippines
    Amando M. Tetangco Jr. and Ma. Cyd N. Tuano-Amador

  5. The Experience of Statistics Sweden
    Svante Öberg and Hans Lindblom

  6. Information Requirements for Inflation Targeting: Observations Based on the Australian Experience
    Glenn Stevens

Part II. Defining and Measuring the Target

Introduction to Part II

  1. The U.K. Office for National Statistics and the Inflation Target
    Amanda Rowlatt

  2. Eurostat's HICP and the European Central Bank's
    Definition of Price Stability
    Vítor Gaspar

  3. Design, Measurement, and Communication:
    Chile's Experience with Inflation Targeting
    Pablo García

  4. Price Indices for Inflation Targeting
    Adriaan M. Bloem, Paul A. Armknecht, and Kimberly D. Zieschang

Part III. Deriving Measures of Inflation and Inflation Expectations from the Market Data

Introduction to Part III

  1. The Bank of England's Approach
    Roger Clews

  2. On Market-Based Measures of Inflation Expectations
    Cedric Scholtes

  3. Regime-Oriented Data Coverage Standards:
    Case Study of Inflation Targeting in Israel
    Edward (Akiva) Offenbacher and Meir Sokoler

  4. The Experience of Banco Central do Brasil
    Gustavo Bussinger

Part IV. Implications for the IMF

Introduction to Part IV

  1. The Experience of the Czech National Bank
    Oldrich Dedek

  2. Inflation Targeting: Statistical Issues Suggested by the Experience of Six Emerging Market Countries
    Edgar Ayales, Randall C. Merris, and Alfredo Torrez

  3. Statistical Principles for Inflation-Targeting Regimes and the Role of IMF Data Initiatives
    Armida San José, Graham L. Slack, and Subramanian S. Sriram

  4. Usefulness of Private Inflation Forecasts in Inflation Targeting
    Masahiko Takeda

Part V. Conclusions

Introduction to Part V

  1. Taking the Agenda Forward
    Carol S. Carson, Charles Enoch, and Claudia Dziobek

Seminar Participants and Authors

Chapter Appendices

3.1   Variants of the South African Overall Consumer Price Index and Percentage Changes for the Metropolitan Areas
3.2   Variants of the South African Consumer Price Index and Percentage Changes for the Metropolitan and Other Urban Areas
12.1   The Real Stochastic Discount Factor/Pricing Kernel
12.2   The Inflation Risk Premium
16.1   Exchange Rate Regimes and Anchors of Monetary Policy as of December 31, 2001
17.1   Summary of the Institutional Framework, Operational Issues, and Organizational Implications for Central Banks in Inflation-Targeting Countries
17.2   The IMF's Data Quality Assessment Framework--Generic Framework

Introduction to Part I

IMF Managing Director Horst Köhler (Chapter 1) lays out the relevance of inflation targeting for the IMF's responsibilities to conduct surveillance on global macroeconomic and financial stability and to provide policy advice to member countries. Many countries have moved recently toward flexible exchange rate regimes. Under such regimes inflation targeting provides an anchor for inflation expectations, and the profile and importance of the statistical underpinnings are enhanced. The authors of the remaining chapters in this section concur that information requirements follow directly from the policy requirements of inflation-targeting regimes: a formal commitment to a specific rate of inflation, a designated measure of inflation (for example, the consumer price index), analytical tools to forecast inflation, analytical tools to analyze the monetary transmission mechanism and the central bank's ability to influence inflation, and a communication policy.

Another common theme is that inflation targeting imposes a high degree of discipline on central banks and national statistical offices. Commitment to a target that can be readily observed by the public--for example, the headline inflation figure--makes the central bank's performance much more transparent; any failures to achieve the target will be obvious. A number of statistical implications can be observed. Ongoing communication with the public, including through regularly published inflation reports and analysis, becomes a vehicle for establishing credibility and handling any deviations from the target. To some extent, inflation targeting changes the data requirements for setting monetary policy: there is, for example, more emphasis on forward-looking indicators, including property occupancy rates, labor market information (wage and compensation indices), interest rate yield curves, and long-term bond rates. The extent to which these data are available varies greatly across countries. Countries that have more developed financial markets tend to have a richer set of forward-looking data to rely on, and may therefore have the greatest opportunities in pursuing an inflation-targeting regime.

Malcolm Knight, Robert Fay, and Brian O'Reilly (Chapter 2) reflecting on the Canadian experience, stress the long time lags involved in monetary policy and the great amount of information and study necessary for effective policy. Policy actions are taken with a view toward influencing the price level 18 to 24 months down the road, and effectiveness is closely related to credibility. The uncertainly inherent in statistical models, and hence the need to "diversify" in research and models, leads central banks to process vast amounts of information. The authors provide an overview of the basic statistical elements and the framework for evaluating statistical information as developed by the Bank of Canada.

Pali J. Lehohla and Annette Myburgh (Chapter 3) discuss differences in user needs with respect to price indicators, which led Statistics South Africa to produce, in addition to the overall consumer price index, two core price indices and other variants including different regional coverage.

Amando M. Tetangco, Jr. and Ma. Cyd N. Tuano-Amador (Chapter 4) note that inflation targeting leads central banks to place more emphasis on the medium-term perspective and less on day-to-day developments. Data needs, accordingly, are more focused on forward-looking indicators. In the Philippines and elsewhere, the central bank produces quarterly inflation reports, which help focus research with respect to the reaction function and the transmission mechanism of monetary policy. In the case of deviations from the target, the governor must write an open letter explaining the deviation and how the central bank intends to return to the target path. These information requirements are vehicles for better transparency.

Svante Öberg and Hans Lindblom (Chapter 5) describe the experience in Sweden, where the central bank commissions from Statistics Sweden the publication of two monthly measures of "underlying inflation" which exclude "temporary" effects that enter into the CPI as well as the development of additional, forward-looking indicators.

Glenn Stevens (Chapter 6) notes that, while more frequent data are generally believed to be better, some countries with quarterly data, like Australia, have successful inflation-targeting regimes. More important are the availability of credible CPI series and the analytical capacity to "dissect" and interpret these data, especially to separate temporary from permanent price changes. Independence of the data-producing agencies is key to credibility, as is the central bank's ability to present a comprehensive macroeconomic analysis to the public that includes but is not limited to price developments.

Another theme echoed in many of the chapters is that inflation targeting has meant that central banks or national statistical agencies must develop their ability to research inflation, dissecting data to distinguish between "one-off," or transitory, factors affecting inflation and more permanent changes, and removing volatile items or those independent of central bank action in order to determine appropriate policy responses.