Credibility and Time-Consistency in Monetary Policy

Abstract

There is a strong presumption that policies work better when they are more credible. Impressive support for this is not difficult to find and one case in particular has received considerable attention of late. Consider a government embarking on a programme of disinflation. Such a programme might involve either a short period of severe monetary restraint (the ‘cold turkey’ approach) or else a less radical policy aimed at reducing monetary growth more slowly over time (the ‘gradualist’ plan). The relative merits of these schemes rest ultimately upon the speed of market adjustment. Regardless of which one is adopted, the success of either in bringing down inflation at little or no cost to the real economy depends crucially upon the programme being credible. If, for whatever reason, the public is sceptical about the policy, the economy will experience the familiar output-inflation trade-off, being plunged into a recession by high inflationary expectations. Only if the public has confidence in the programme will the danger of sliding down the Phillips curve be somewhat mitigated. Moreover, to the extent that inflationary expectations partly influence the actual inflationary process, a believable disinflation has the added bonus of contributing to the disinflation itself.