Developing an underlying inflation gauge for China

BIS Working Papers No 465
September 2014

The headline consumer price inflation (CPI) is often considered too noisy, narrowly defined, and/or slowly available for policymaking. On the other hand, traditional core inflation measures may reduce volatility but do not address other issues and may even exclude important information. This paper develops a new underlying inflation gauge (UIG) for China which differentiates between trend and noise, is available daily and uses a broad set of variables that potentially influence inflation. Its construction follows the works at other major central banks, adopts the methodology of a dynamic factor model that extracts the lower frequency components as developed by Forni et al. (2000) and draws on the experience of the People's Bank of China in modelling inflation. The paper is the first application of this type of dynamic factor model for inflation to any large emerging market economy. Our UIG for China is less noisy but still closely tracks the headline CPI. It does not suffer from the excess volatility reduction that plagues traditional core inflation measures and instead provides additional information. Finally, when forecasting the headline CPI, our UIG for China outperforms traditional core measures over different samples.

JEL classification: C13, C33, C43, E31, E37, G15

Keywords: Inflation, Dynamic Factor Models, Core Inflation, Monetary Policy, Forecasting, China