The Inflation Bias Revisited: Theory and Some International Evidence

The Kydland-Prescott, Barro-Gordon inflation bias result relies on the presumption that policymakers aim at achieving a level of employment above potential. Both academics and policymakers have recently questioned this presumption on the ground of realism. We show that even if policymakers are content with the normal level of employment there is an inflation bias if the central bank is uncertain about the future state of the economy, and is more sensitive to policy misses leading to employment below the normal level than to policy misses leading to employment above it. This new view of the inflation bias implies that there should be a positive association between average inflation and the variance of shocks to output. Cross sectional empirical evidence from 21 developed economies supports this implication. The Paper also discusses the consequences for the transparency of monetary policy and for central bank reform.