Optimal Monetary Policy in a Data-Rich Environment
Joint with: Marc P. Giannoni
This paper considers a framework in which the central bank observes a potentially large set of noisy indicators but is uncertain about the state of the economy. We evaluate the welfare implications of exploiting all available information to assess the state of the economy. We show
that it is possible to characterize in a unified state-space representation the equilibrium evolution of all model variables, whether the central bank sets its instrument following an arbitrary policy rule or commits to optimal policy, and whether the central bank has full information about the state, responds naively to observed indicators, or optimally estimates the state of the economy using available indicators. Using a stylized quantitative model, estimated on US data, we show that filtering out the noise in observable series is crucial to conduct policy appropriately, and argue that under current monetary arrangements, a policy that would systematically exploit all
available information to assess the state of the economy is likely to result in substantial welfare gains.
For more information, contact Frank Schorfheide.