Mark-Ups and Common Ownership: An IO Perspective
Friday, Jan. 3, 2020 2:30 PM - 4:30 PM (PDT)
- Chair: Julie Holland Mortimer, Boston College
Structure, Conduct, and Contact: Competition in Closely-Related Markets
AbstractWe perform a simultaneous empirical study of competitive conduct in 40 product categories of the Israeli food sector. In each category we estimate a differentiated product demand model, and compute several empirical indicators of the category's ability to support cooperative price equilibria. Specifically, we test the hypothesis that the data were generated by a cooperative pricing equilibrium, and compute threshold discount factors: those that would support a hypothetical collusive regime in a standard repeated-game framework. We then investigate several pervasive questions in the Industrial Organization literature. In particular, we examine the extent to which cooperative pricing regimes are facilitated by both within-category concentration and by cross-category interactions (multimarket contact). We further examine the relationship between a category's concentration and its aggregate demand elasticity. By combining these analyses we provide a multi-faceted picture of competition and concentration in closely related markets.
The Evolution of Market Power in the United States Auto Industry
AbstractWe estimate markups and welfare in the U.S. car and light truck market from 1970-2015. During this time the set of available products changed dramatically. Some examples of these changes include the rise of import competition, the introduction of new body-styles, the change in automobile characteristics (such as size and fuel efficiency), and the introduction of new features (such as anti-lock brakes). We document these trends in the set of available products and relate them to trends in market power. We establish how market power is related to innovation in product characteristics and the speed at which competitors compete away these rents by introducing similar products.
Testing the Production Approach to Markup Estimation
AbstractUnder the underlying assumptions of the production approach to markup estimation, one should recover the same markup from any flexible input. I test this approach by comparing estimates of markups derived using labor and materials for four national manufacturing censuses and store-level data from a nationwide US retailer. Across all five datasets, I find that markups estimated using labor are negatively correlated with those estimated using materials, exhibit much greater dispersion, and have different trends over time. I also find different correlations with the degree of competition faced by the store for the retailer. I examine several mechanisms that may explain these findings, and find evidence that they are consistent with plant level heterogeneity in production technology. In order to successfully measure markups, the production approach will have to model such heterogeneity.
- L1 - Market Structure, Firm Strategy, and Market Performance