Introduction

There is widespread recognition among academics and policy-makers that monetary policy should aim at price stability. A large number of economies have adopted inflation targeting (IT) as the main objective of monetary policy, aiming at maintaining low and predictable inflation rates as a way of preserving the purchasing power of money. Evidence abounds that targeting inflation has been successful in anchoring and stabilising inflation rates and inflation expectations, with no apparent increase in output volatility, from the end of the 1980s to the beginning of this century (Mishkin and Schmidt-Hebbel, 2001, 2007). Moreover, there is mounting evidence that inflation targeting contributes to keeping inflation low and predictable (Fatás, Mihov and Rose, 2006; Angeris and Arestis, 2008; Benati, 2008; Calderón and Schmidt-Hebbel, 2008).

Nonetheless, current monetary policy frameworks have been brought into question by the frequency with which deflation has taken hold of or threatened OECD economies. Japan went through a period of deflation in 2000–6 while the US was perceived to be exposed to a deflation risk in 2001 and again in 2003. A consequence of the current financial and economic crisis is that deflation has once more emerged as a risk, this time in many OECD countries, while it has taken hold of Japan again. While monetary policy frameworks may need to be reassessed with a view to reducing deflation risk, any changes should occur only once current objectives are attained, for fear of undermining confidence in central banks. Against this background an alternative interpretation of price stability has attracted increasing attention from policy-makers in the recent past (Ambler, 2009; Parkin, 2009): it is the notion of price-level stability, where the monetary authority aims at stabilising an aggregate price level around a pre-specified path, instead of its rate of change. A principal advantage of price-level targeting is that it is more consistent with the objective of preserving the long-run purchasing power of money. For instance, following a temporary inflation spike, such as that observed in 2005–7, inflation targeting regimes will leave the purchasing power of money permanently below what would have transpired if inflation targets had been met. A successful price-level-based framework would also avoid a permanent increase in the real value of debt after a period of deflation. Practical experience of price-level targeting is, however, restricted to one historical episode, in Sweden during the 1930s (discussed on p. 312).