Proponents of industrial policy argue that merger control should consider domestic employment. I propose a model to assess how a product market merger affects rival product entry, consumer welfare, and domestic employment. Firms endogenously decide which products to offer. Domestic jobs depend on production locations and equilibrium quantities in the product market. I estimate the structural parameters of this model for the U.S. home appliance industry. Using the structural model, I examine the impact of Whirlpool's acquisition of Maytag and compare it to the impact of a counterfactual acquisition by a foreign buyer with no prior presence in the U.S. market. Four key findings emerge from the comparison of these two acquisitions: First, rival product entry is mostly independent of the acquirer. Second, a Whirlpool acquisition leads to the removal of more merging party products. Third, it always leads to lower consumer welfare. Fourth, a Whirlpool acquisition leads to a smaller decrease in U.S. employment. I use these results to estimate the job value necessary for domestic employment effects to offset consumer welfare losses.
When is informing consumers about prices most effective at increasing competition? We combine a theoretical model of imperfect information with empirical evidence to show how the effect of informing consumers about prices depends on how well informed they are ex ante. Theoretically, an increase in consumer information decreases prices more the fewer ex ante informed consumers there are. Empirically, we study the introduction of mandatory price disclosure in the German fuel market separately for two fuel types that differ in how informed consumers are ex ante. For each fuel type, we use a difference-in-differences design and the universe of station-level prices in Germany and France. In line with the theory, the decline in prices was stronger for fuel type with ex ante less informed consumers. We also show that the treatment effect declines over time and is intensified by local follow-on information campaigns.
We investigate how the pass-through rate of commodity taxes depends on competition in a setting where consumers have imperfect information about prices. We use a theoretical search model that has two key predictions. First, the larger the number of price sensitive consumers, the higher the pass-through rate. Second, there is a hump-shaped relationship between the average pass-through experienced by consumers and the number of sellers. We test our theoretical predictions by studying pass-through in the context of a tax decrease and increase in the German retail fuel market. We estimate pass-through of these tax changes to diesel and gasoline prices using a unique dataset containing the universe of price changes at fuel stations in Germany and France and a synthetic difference-in-differences strategy. Our empirical results are in line with our theoretical predictions. Finally, we show that our theoretical framework can encompass and reconcile a large number of empirical observations in previous studies.
Data-driven digital products are becoming ubiquitous, traditional markets are rapidly becoming digitised. and a handful of tech companies have come to dominate digital markets. This development has sparked a debate about whether competition authorities are well equipped to handle data-driven industries. Yet, as of today these tech companies are highly innovative, sell many of their products at zero prices for consumers and are among the leading companies for global spending on R&D. This raises the question whether there is really anything amiss that would call for a more active antitrust policy. In this article, we explain the economics of data-driven industries and discuss how policy measures can affect the contestability of these markets. We discuss the need for and implications of changes to merger control, data portability, and mandated data access. None of these measures is a silver bullet. Instead, they could be added to the policy toolbox and used on a case-by-case basis after carefully considering the specificities of the market. Furthermore, the timing of interventions is often crucial. In some cases, it will be sufficient to consider policies in the context of ex post competition enforcement. In other cases, intervention at this stage may come too late and ex ante regulation is preferable.