A Horse Race of Monetary Policy Regimes: An Experimental Investigation

Available as: PDF

Monetary policy has evolved significantly in the aftermath of the global financial crisis and the COVID-19 pandemic. Alternative approaches such as average inflation targeting (AIT), price-level targeting (PLT) and nominal GDP-level targeting (NGDP) have gained attention in academic and policy discussions. These history-dependent regimes can be powerful and effective in stabilizing an economy because the central bank has to make up for past misses. However, their performance depends on forward-looking expectations and people’s understanding of how these regimes work. Despite the recent attention these alternative approaches to monetary policy have received, evidence of the public’s understanding remains limited.

Using a unified experimental framework, we provide a comprehensive assessment of five monetary policy regimes: inflation targeting (IT), dual mandate (DM), AIT, PLT and NGDP. We study how effectively people can understand and form macroeconomic expectations under the different policy frameworks, both throughout periods of economic stability away from the effective lower bound (ELB) and during demand-driven recessions at the ELB. We take the simple New Keynesian model used in the Bank of Canada's own theoretical analysis to the laboratory and run an experimental horse race of the five alternatives.

Our results suggest a distinct ranking of the monetary policy regimes in terms of their ability to achieve macroeconomic stability. Rate-targeting approaches such as IT, DM and AIT significantly outperform level-targeting regimes such as PLT and NGDP in minimizing deviations of inflation, the output gap and nominal interest rates from target. IT, DM, and AIT with a short four-quarter horizon deliver similar performances. AIT with a shorter horizon (4 quarters) performs better than with a longer horizon (10 quarters). Monetary policy regimes that are framed around the inflation rate (e.g., AIT with a 10-quarter horizon) are found to deliver significantly more stable economic outcomes than those that target price levels (PLT).

DOI: https://doi.org/10.34989/swp-2022-33