Search results

AbstractRising consumer prices may reflect shifts by consumers to new higher‐priced products, mostly for durable and semi‐durable goods. I apply Bils' (2009) methodology to newly available Canadian consumer price data for non‐shelter goods and services to estimate how price increases can be divided between quality growth and price inflation. I find that less than one third of observed price increases during model changeovers should be attributed to quality growth. This implies overall price inflation close to inflation measured by the official index. I conclude that, according to Bils' methodology, the quality bias is not an important source of potential mismeasurement of CPI inflation in Canada.

Models with imperfect information that generate persistent monetary nonneutrality predominantly rely on assumptions leading to substantial heterogeneity of information across price-setters. This paper develops a quantitative general equilibrium model in which the degree of heterogeneity of information is determined endogenously. In the model, firms use two technologies to acquire information: costly updating to full information and costless learning from publicly observed market signals. Price changes of firms that update information infrequently are synchronized with market signals. This leads to an externality whereby less frequent updating increases the information conveyed by prices and quantities. When the model is calibrated to moments from a panel of BLS commodity sectors, it is found that the private value of costly updating to full information is close to zero, market signals are informative, and the real effects of monetary shocks are small.

Recent New Keynesian models of macroeconomy view nominal cost rigidities, rather than nominal price rigidities, as the key feature that accounts for the observed persistence in output and inflation. Kryvtsov and Midrigan (2010a,b) reassess these conclusions by combining a theory based on nominal rigidities and storable goods with direct evidence on inventories for the U.S. This paper applies Kryvtsov and Midrigan's model to the case of Canada. The model predicts that if costs of production are sticky and markups do not vary much in response to, say, expansionary monetary policy, firms react by excessively accumulating inventories in anticipation of future cost increases. In contrast, in the Canadian data inventories are fairly constant over the cycle and in response to changes in monetary policy. Similarly to Kryvtsov and Midrigan, we show that markups must decline sufficiently in times of a monetary expansion in order to reduce firms' incentive to hold inventories and thus bring the model's inventory predictions in line with the data. The model consistent with salient features of the dynamics of inventories in the Canadian data implies that countercyclical markups account for a sizable (50-80%) fraction of the response of real variables to monetary shocks.

Kryvtsov and Midrigan (2008) study the behavior of inventories in an economy with menu costs, fixed ordering costs and the possibility of stock-outs. This paper extends their analysis to a richer setting that is capable of more closely accounting for the dynamics of the US business cycle. We find that the original conclusion survives in this setting: namely, the model requires an elasticity of real marginal cost to output approximately equal to the inverse intertemporal elasticity of substitution in consumption in order to account for the countercyclicality of the aggregate inventory-to-sales ratio in the data.

Real rigidities that limit the responsiveness of real marginal cost to output are a key ingredient of sticky price models necessary to account for the dynamics of output and inflation. We argue here, in the spirit of Bils and Kahn (2000), that the behavior of marginal cost over the cycle is directly related to that of inventories, data on which is readily available.We study a menu cost economy in which firms hold inventories in order to avoid stockouts and to economize on fixed ordering costs. We find that, for low rates of depreciation similar to those in the data, inventories are highly sensitive to changes in the cost of holding and acquiring them over the cycle. This implies that the model requires an elasticity of real marginal cost to output approximately equal to the inverse of the elasticity of intertemporal substitution in order to account for the countercyclical inventory-tosales ratio in the data. Stronger real rigidities lower the cost of acquiring and holding inventories during booms and counterfactually predict a procyclical inventory-to-sales ratio.

This paper asks: What is the effect of government policy on output and inequality in an environment with education and labor-supply decisions? The answer is given in a general equilibrium model, consistent with the post 1960s facts on male wage inequality and labor supply in the U.S. In the model, education and labor-supply decisions depend on progressive income taxation, the education system, the social security system, and technology-driven wage differentials. Government policies affect output and inequality through two channels. First, a policy change leads to an asymmetric adjustment of working hours and savings of schooled and unschooled individuals. Second, there is a redistribution of the workforce between schooled and unschooled workers. Using a battery of proposed government policies, we demonstrate that skill redistribution dampens the response of wage inequality to a policy change and amplifies the response of output by an additional 1 to 2 percent.

We assess the implications of price indexation for estimated frequency of price adjustment in sticky price models of business cycles. These models predominantly assume that non-reoptimized prices are indexed to lagged or average inflation. The assumption of price indexation adds tractability although it is not likely reflective of the price practices of firms at the micro level. Under indexation firms have less incentive to adjust their prices, which implies downward bias in the estimated frequency of price changes. To evaluate the bias, we generate data with Calvo-type models without indexation. The artificial data are then used to estimate the frequency of price changes with indexation. Considering different assumptions about the degree of price rigidity and the level of trend inflation in the data-generating model, we find that the estimated indexation bias can be substantial, ranging up to 12 quarters in some cases.