Estimates of the Sticky-Information Phillips Curve for the United States, Canada, and the United Kingdom

Mankiw and Reis (2001a) have proposed a "sticky-information"-based Phillips curve (SIPC) to address some of the concerns with the "sticky-price"-based new Keynesian Phillips curve. In this paper, we present a methodology for the empirical implementation of the SIPC for closed and open economies. We estimate its key structural parameter—the average duration of information stickiness—for the United States, Canada, and the United Kingdom. The benchmark results (with forecasting horizons of firms of seven to eight quarters) indicate that: (i) the average frequency of information updates is four quarters in the United States, between four and five quarters in Canada, and over seven quarters in the United Kingdom, and (ii) the open-economy estimates for Canada and the United Kingdom are similar to the closed-economy ones. Mankiw and Reis (2001a) assume information stickiness of four quarters to generate inflation dynamics similar to that observed in the U.S. data. We interpret our estimates in terms of the slope of the Phillips curve. Our structural estimates can be used to calibrate dynamic general-equilibrium models for monetary policy analysis that use informational inertia to match the stylized facts on inflation dynamics.