According to Research Fellow Danny Quah and Shaun P Vahey in Discussion Paper No. 1153, arguments over the right measure of inflation are interminable. In one interpretation the reason for this is theoretical: for many coherent models of economic equilibrium, the rate of aggregate inflation is irrelevant. Under such views any definition of the rate of inflation is correct. Another interpretation, however, is that the reason for the never-ending debate is extremely practical. Policy-makers in many countries have assigned themselves the task of keeping aggregate inflation within precisely specified bands. As a result, financial markets monitor and react to every little tick in inflation, anticipating (rightly or wrongly) discrete policy changes. Interest rates, exchange rates, and equity prices respond to statistical reports that in truth need be no more than best guess estimates of an underlying noise-ridden reality. Hence, the precise figure for aggregate inflation matters a great deal.

The authors argue that measured retail price index (RPI) inflation is conceptually mismatched with core inflation: the difference is more than just `measurement error'. A technique is proposed for measuring core inflation based on an explicit long-run economic hypothesis. Core inflation is defined as that component of measured inflation that has no (medium- to) long-run impact on real output, a notion that is consistent with the vertical long-run Phillips curve interpretation of co-movements in inflation and output. Lastly, a measure of core inflation is constructed by placing dynamic restrictions on a vector autoregression (VAR) system. Applied to the UK, it suggests that the real economy adjusts quickly to movements in core inflation, and that disturbances to core inflation have a persistent impact on a standard measure of inflation, the RPI.

Measuring Core Inflation
Danny T Quah and Shaun P Vahey

Discussion Paper No. 1153, March 1995 (IM)