Finding the Goldilocks Data Collection Frequency for the Consumer Price Index

Abstract

This work addresses a well-defined question: the lack of a theoretical framework to assist National Statistical Institutes (NSIs) in determining the adequacy of their data collection for the Consumer Price Index – whether it is too sparse, too frequent, or just right. We propose a framework designed to achieve a balance between reducing uncertainty in price measurement and minimizing the expenses associated with data acquisition, processing, and storage. This cost-benefit analysis is particularly relevant with the emergence of big data and alternative data sources, alongside regulatory requirements for NSIs to archive their data over extended periods. An illustrative application is provided through an examination of electricity and gas utility prices in the Italian unregulated market during year 2023, items that were notably affected by the energy price crisis stemming from Russia’s invasion of Ukraine.