The authors begin with the observation that the prices of some goods appear to be more flexible than others. Prices of airline tickets and commodities change daily. The price of haircuts and magazines remain fixed for long periods of time. Whereas Keynesian models generally assume that all prices are sticky and classical models generally assume that all prices are fixed, reality is somewhere in between. The authors set out to see how such differences affect the evolution of macroeconomic aggregates. In these comments I want to first briefly describe their model, highlighting the main assumptions and their consequences. I will then discuss how well their model captures the sectoral consequences of monetary policy, and conclude with some comments on what might be missing from their framework. I begin with the model. To capture the effects of sectoral differences in nominal rigidity, they construct dynamic macroeconomic model with two goods-producing sectors and one representative consumer who owns the capital stock and supplies labor. In one sector goods prices are sticky. In the other they are perfectly flexible. They make five crucial modeling choices. The first concerns the nature of the nominal rigidity. They assume that firms in the sticky-price sector set their nominal prices each period prior to the realization of monetary shocks. This assumption, which only serves to delay adjustment for one period, has two important consequences. First, expected inflation is necessarily high following a monetary expansion, since firms in the sticky-price seetor will be raising their prices in the next period. This puts a lot of upward pressure on short-term interest rates. In fact, in their simulations the nominal interest rate rises in response to a monetary expansion, even though the real interest rate falls. The second consequence of the assumption that prices are preset for only one period is that it eliminates an important source of persistence in the economy's response to monetary shocks. Prices become perfectly flexible the period after the shock. Much research over the past fifteen years has focused how the interaction between the pricing decisions of various agents increases price inertia and propagates the effect of monetary shocks. Alternative assumptions on the nature of the nominal rigidity such as staggered price adjustment (Taylor 1980, Blanchard 1983) or costly price adjustment (Rotemberg 1982, Caplin and Leahy 1991) have been shown to slow the adjustment of prices. The inertia that these alternative assump-
Read full abstract