Sticky prices, competition and the Phillips curve

This study analyzes how competition affects price stickiness at the micro level. On the theoretical side, I develop what I call a micro Phillips curve, i.e. a product-specific relation between inflation and economic activity conditional on inflation expectations. I find two opposing effects of competition on the slope of the micro Phillips curve. On the one hand, stronger competition leads to a higher frequency of price revaluations, implying a steeper slope. On the other hand, the stronger competition is, the less firms can transmit changes in economic activity into price changes, implying a flatter slope. Using unique product-level manufacturing panel data, I find that the latter effect clearly dominates and plays an important role in explaining price stickiness. The effect of a marginal increase in economic activity on the likelihood of a price increase is between 63% and 85% lower for products, that face very strong competition, compared to products, that face very weak competition. In line with the theory, prices of products, that face very strong competition, are also less likely to decrease in response to marginal decreases in economic activity. Moreover, it heavily depends on the degree of competition that a product faces whether, and to what extent, the micro Phillips curve is non-linear. The stronger the competition the weaker will be the non-linearity of the micro Phillips curve. My findings imply that effective business cycle policy necessitates good competition policy. Reforms which strengthen the competition in an economy will make stimulus or stabilization policy more effective. Furthermore, the results imply that stimulus or stabilization measures, that target specifically high competition firms or sectors, may be more effective than programs, that follow an indiscriminate all-round principle. -- Price stickiness ; Phillips curve ; real rigidity ; micro data