Inflation Targeting and Target Instability

摘要

Monetary policy is modeled as being governed by a known rule, except for a time-211u001evarying target rate of inflation. The variable target is taken as representing 211u001eeither discretionary deviations from the rule, or as the outcome of a 211u001epolicymaking committee that is unable to arrive at a consensus. Stochastic 211u001esimulations of FRB/US, the Board of Governors large, rational-expectations model 211u001eof the U.S. economy, are used to examine the benefits of reducing the variability 211u001ein the target rate of inflation. We find that putting credible boundaries on 211u001etarget variability introduces an important non-linearity in expectations. The 211u001eeffect of this is to improve policy performance by focussing agents expectations 211u001eon policy objectives. But improvements are limited; it does not generally pay to 211u001ereduce target variability to zero. More important, this non-linearity in 211u001eexpectations allows for policy to be conducted, at the margin, with greater 211u001eattention to output stabilization than would other-wise be the case. The results 211u001eprovide insights as to why inflation targeting countries use bands and why the 211u001ebands they use are narrower than studies suggest they should be. A side benefit 211u001eof the paper is the demonstration of a numerical technique that approximates to 211u001earbitrary precision a non-linear process with a linear method, thereby greatly 211u001espeeding and making more robust the computation of simulation results.