Essays in Industrial Organization

Abstract

This thesis develops methods aimed at studying transport markets, and applies them to the analysis of specific industries. The first Chapter studies platform competition in the app-based transportation industry. It presents a model of competing platforms in transport equilibrium, characterizing analytically the profit-maximizing allocations and prices, and estimate it using high-frequency data on the operations of the two main platforms in New York City. The estimated model is then used to study the impact of a platform merger on traffic volumes and customer welfare, and to study common policies aimed at reducing traffic, such as congestion pricing and entry restrictions. The second Chapter explores efficiency and optimal policy in decentralized transport markets. It shows that search frictions can distort the transport network and the dynamic allocation of carriers over space, identifying the main sources of externalities and deriving explicit conditions for efficiency. Using data from dry bulk shipping, it finds sizeable social loss and spatial misallocation of carriers, and shows that optimal policy can restore efficiency by favoring locations that are central in the trade network and might be preferable to centralization. The third Chapter develops methods - which are applied in the first two Chapters - for estimating undiscounted Markov decision problems: a class of models with widespread applications to transportation and logistics.