In this paper we examine price variation in manufacturing industries over the business cycle. This study is motivated by the hypothesis that firms with market power may have incentives to exercise this power differentially over the business cycle. Such differential exercise of market power could exacerbate business cycle fluctuations relative to what would occur in the absence of market power.
Our point of departure is a model of oligopoly investment and pricing that has implications for variations in prices and industry efficiency over the business cycle. This model emphasizes the role of long run production capacity investments that must be made before demand conditions are known. After capacity investments are made, firms learn about the level of product demand and choose prices. The pricing incentives for firms differ depending on the level of demand.
If demand is high, then the short run competitive (market clearing) price is a pure strategy Nash equilibrium. However, if demand is low, then capacity constrained firms have an incentive to deviate from the short run competitive price; the typical result is that firms adopt mixed strategies that yield prices above the short run competitive price and that generate excess production capacity. These results are developed in detail in Reynolds and Wilson (2000). We refer to this as a non-collusive oligopoly model to distinguish it from collusive theories of oligopoly behavior over the business cycle.
This non-collusive model predicts that output prices are procyclical, as do many other theories. The key prediction that we empirically examine is that output price changes will have greater variance during low demand periods than during high demand periods. We search for this differential variance of price changes over the business cycle using a version of time series switching regime filter developed by Hamilton (1989). We estimate this model for each of seventeen manufacturing industries at the two and three-digit Standard Industrial Classification (SIC) level.