Over the past twenty years, economists have developed a rich set of tools for studying the design of monetary policy in dynamic and stochastic environments. We know how to build macro models consistent with optimizing behavior by households, firms, and the central bank. Most of these models are highly stylized, more appropriate for gaining basic insights into policy issues than they are for actually guiding the implementation of policy, but these models have also been extended in a wide variety of ways to better incorporate macroeconomic dynamics, to improve their forecasting ability, and to make them more useful in the practice of monetary policy.
Despite these advances, the financial crisis, the Great Recession, and the ongoing fiscal crises serve as painful reminders of how inadequate our models are in understanding the role of financial frictions, the sources of potential shocks, and the manner in which shocks can be amplified and propagated throughout not just the economy in which the shock originates but throughout the international economic system. Fortunately, monetary policy models are rapidly evolving to incorporate many important insights gained from the past two years.
While the theme of this celebration of the 200th anniversary of the Bank of Finland is Monetary policy under resource mobility, most of my focus will be on resource immobility. The reason is that the basic new Keynesian framework that is at the heart of most modern DSGE policy models gives center stage to nominal price and wage rigidities, yet underlying this structure are some very extreme assumptions about the ability of labor and capital to swiftly move among alternative uses. For example, the baseline version of the new Keynesian model has a continuum of firms producing differentiated goods. These firms find it costly to adjust their selling prices, but these same firms can hire and fire workers at zero cost, and both workers and capital can instantly shift from one firm to another.
Baseline models are poor at matching macroeconomic dynamics, and the versions of the models taken to the data generally incorporate features that are often ad hoc but are necessary to capture the types of sluggish adjustments seen in the data. A more careful treatment of the costs that limit rapid movement of labor and capital among alternative uses is likely to be useful in understanding the way monetary policy actions affect the behavior of the economy as well as critical in assessing the costs associate with economic fluctuations.
