Essays in industrial organization

Abstract

This thesis is a collection of three chapters that investigate burgeoning empirical issues in industrial organization. In the first chapter, I study platform fee policy with a specific focus on two-sided online marketplaces. The main contributions of the paper are threefold. First, I study a setting with coordinated price experimentation along the three different fee dimensions that are common to such marketplaces. Second, I describe the empirical impact of incomplete fee salience on equilibrium outcomes. Finally, I quantify the network externalities that must be present in order for observed fees to constitute an equilibrium. In the paper, I begin by developing a tractable model of the platform's problem that generates testable predictions and yields equilibrium conditions in terms of estimable quantities. Then, using estimates from experimental data obtained from a large online marketplace, I quantify the salience and network effects.

To conclude, I consider the counterfactual level and composition of equilibrium platform fees under when these effects are muted or absent. In the second chapter, using data from the same source as in chapter one, I study small sellers competing on the supply side of online marketplaces. As these platforms grow and markets become increasingly disintermediated, an important concern is whether small sellers, who may have limited experience or attention, can individually compete effectively with larger, often professional sellers operating on the same marketplaces. To answer this question, I develop and estimate a structural model that incorporates essential features of the empirical setting, including large and rapidly changing choice sets and buyer heterogeneity. Using the estimated model, I compute optimal pricing policies under various informational and computational restrictions.

I find that small sellers adhering to a simple strategy can obtain nearly optimal expected revenue and that this strategy's information requirements are easily satisfied in the online setting. Additionally, I present suggestive evidence that sellers learn to approximate such a strategy through repeated market interactions. In the third and final chapter, I investigate the industrial impacts of firm control rights, which confer discretion over firm policy and are usually shared between debt and equity holders. Control rights operate along a continuum and are difficult to measure. As a proxy, I consider the discontinuous shift in control from equity holders to creditors due to loan covenant violations, a common form of technical default. This paper contributes to the growing covenants literature in two ways. First, I consider the impact of and response to covenant violations at the industry level, inclusive of firms never in technical default.

Second, I empirically document the effects of violations on contemporary product markets. I find that control rights transfers to creditors make firms tough in product markets, consistent with the predictions of a stylized model, and that markups decline at the industry level, though the declines are sharpest for firms directly affected.