A horse race of monetary policy regimes : an experimental investigation

We provide a comprehensive assessment of five monetary policy regimes-inflation targeting (IT), dual mandate (DM), average inflation targeting under 4-period (AIT-4) and 10-period (AIT-10) horizons, price level targeting (PLT), and nominal GDP level targeting (NGDP)-in a unified experimental framework. We study how participants can understand different regimes and form expectations during periods of economic stability and during a demand-driven recession that temporarily brings the economy to the ELB. Our results suggest a distinct ranking of policy regimes in terms of their ability to achieve macroeconomic stability. DM and IT are the most stabilizing regimes, followed by AIT and then level-targeting regimes PLT and NGDP. AIT with a shorter horizon performs better than AIT with a longer horizon. Monetary policy regimes that are framed around the inflation rate (e.g., AIT-10) are found to deliver significantly more stable economic outcomes than those that target price levels (PLT). Participants have greater difficulty understanding regimes that are more history-dependent and forecasting in the rationally expected direction. They instead rely on trend-chasing heuristics to form their expectations. Trend-chasing forecasting is more destabilizing for regimes with more history dependence. Participants also "need to see it to believe it." PLT and NGDP initially have mixed success at achieving their targets, and these regimes do not gain credibility before the economy enters into the ELB, where credibility is needed more than ever.