Measurement Errors in the Consumer Price Index: Perspectives on Numerical Targets for Price Stability in Major Economies

The consumer price index (CPI) is widely used by major central banks in the definition of a numerical target for price stability. In this context, the measurement errors, which became a significant concern following the U.S. Boskin Report in 1996, are still occasionally cited as one of the justifications for setting a positive inflation rate target. This paper presents a review of the current status of discussions and studies on measurement errors in major economies, including Japan, with a particular focus on those published since the Boskin Report. In light of the accumulated research to date, the implementation of solutions to enhance index accuracy in each country, and changes in economic structure, the magnitude of measurement errors that were pointed out in the report appears to have decreased overall, although there are differences across countries and types of errors. Nevertheless, certain areas, particularly those pertaining to service prices, continue to present challenges, and in consideration of the potential impact of changing economic structures, such as e-commerce, measurement errors may fluctuate significantly and remain unresolved in the transition towards a future increase in service consumption, a more digitalized economy, and an aging population. Given these issues, measurement errors could remain a valid basis for setting numerical targets for the CPI.