Essays on economics of automobile industry

This dissertation studies different topics in the field of Industrial Organization. Each of the three chapters in this thesis constitute a separate research papers. All the three chapters provide empirical analysis of automobile industry, primarily focusing on the European countries.These three chapters are not only connected  through Empirical Industrial Organization literature, but also through the motive to contribute to evaluation of antitrust policies  as well as the economic welfare aspect.

In the first chapter, I investigate the effects of an automobile merger on the equilibrium dealership network and consumer welfare. I estimate a structural model by using dataset from Belgian automobile industry, where firms make endogenous decisions on prices as well as their dealership networks. I use these estimates to simulate a counterfactual world focusing on the recent acquisition of Opel brand by the PSA group. The counterfactual simulations show that car prices increase after the merger. However, the number of dealers selling cars of the merged firm and its close competitors decreases. The results suggest that, while an increase in prices following a merger unambiguously harms consumers, the investment incentives after a merger, leading to rearrangements in the dealership network may have heterogeneous implications for consumer welfare. In particular, the dealership network shrinks significantly in relatively rural regions leading to a fall in consumer welfare in those local markets. The firms however add dealers in relatively urban and more populated regions which improve consumer welfare in those geographic locations. These results have important implications for antitrust policies.

The second chapter is an extension of first chapter in terms of methodology related to estimation of fixed cost of dealers.In this chapter, I estimate a structural model of the automobile industry by considering nine large European countries. I explicitly model  the manufacturers' decision of number of dealers to uncover fixed costs.The existing literature on the estimation of fixed costs (e.g. Pakes et al. (2015), Fan and Yang (2020)) has modeled the entry into a market as a strategic game while treating the decision about whether to enter as a discrete choice variable. In contrast, in this analysis, I treat the firm's decision on the number of dealers in a given market as a continuous variable. The assumption of the number of dealers as a continuous choice variable provides a significant advantage in lowering the computational burden while estimating the entry game. The results show that the average of the estimated fixed costs of adding a dealer under this assumption matches closely with the fixed cost of entry when the number of dealers is treated as a discrete variable.

The third chapter is related to the previous chapters in terms of industry and methodologies. This chapter documents the patterns of product characteristics and  markups over a period of thirty years (1970-99) in European automobile industry.I use detailed data on car prices, quantities, and other car characteristics in five European countries to estimate demand systems with flexible consumer preferences.The evidence confirms the patterns in Knittel (2011)  that horsepower, size, and fuel efficiency have improved significantly over this period of time. I estimate the markup of each car model, under the assumption that the car manufacturers maximize their variable profits. I find that the markup increases by approximately 33% across all the countries. The marginal cost of production has also increased due to the increasing quality of cars and technological growth in the automobile industry, leading to higher car prices over time.