Abstract
The Harmonized Index of Consumer Prices (HICP) is at the core of the monetary policy strategy of the European Central Bank (ECB). It is the basis for the quantitative definition of price stability that is the ECB's principal objective. For operational purposes, in October 1998 the Governing Council of the ECB originally announced that its definition of price stability would be an annual increase in the HICP of 'below 2 percent'. In May 2003, this was changed to 'close to 2 percent'. But is 2% the right number? Our analysis suggests that the answer is no, and that a modest upward redefinition of HICP inflation consistent with price stability is warranted. We evaluate the ECB's quest for price stability during the first years of monetary union from a measurement perspective. That is, we start by considering what the HICP is designed to measure and how accurate it is in terms of its stated objective, and then ask whether there is any sense in which HICP inflation of 2% can be said to be too low. We conclude that the conceptual underpinnings of the HICP remain sufficiently vague so that it is difficult to compare with other indexes or come to any hard conclusions about its accuracy. However, it is possible that the HICP is susceptible to the biases that are known to affect other measures of inflation at the consumer or household level, and if forced to quantify potential bias, a point estimate of 1% strikes us as reasonable. Bias of this magnitude, in conjunction with the inherent noisiness of the headline number and the well-known aversion of central bankers to deflation, lead us to conclude that a target of 2% is in fact too low.