| Chapter 1 sets up the framework for this research in the context of the U.S. cable industry. Chapter 2 investigates the effects of vertical integration in this industry. Theory shows that vertical integration can promote efficiency by eliminating successive monopoly markups, internalizing service, and mitigating contractual problems between firms. It can also facilitate the strategic practice of market foreclosure, by which an integrated firm denies a rival access to their markets for the purpose of gaining market power. This study empirically estimates the relative size of these two effects towards understanding the welfare implications of vertical integration in the cable industry Unlike previous studies, I focus on a single vertical merger and estimate its average treatment effects on the several market outcomes. The findings suggest that there was an efficiency gain from the merger. The merged systems were found to carry affiliated networks more frequently than the non-merged systems, there was a larger price decrease in the merged markets, and the new affiliated networks increased their subscribers more in the merged markets than in the non-merged markets. On the other hand, there was weak evidence of foreclosure. Chapter 3 looks into the industry practice of bundling multiple cable networks. Recent studies show that a probability distribution of consumer valuation for a collection of goods has proportionately more mass near the mean; hence, the seller is able to extract more of a consumer's surplus than would be possible without bundling. Using an empirical framework based on this bundling theory for a multi-product monopolist, I estimate consumers' preference distributions for bundles of cable networks in the U.S. cable industry. The obtained distributions are then used to measure the extent of surplus extraction by a monopolist. I found the percentage of surplus extraction ranged from 13 to 46%. Contrast with the bundling theory predictions of such figures as Bakos and Briynjolfsson (1999), the surplus extraction does not increase with the bundle size. This is due to decreasing valuation, as suggested by Geng et al. (2005). The valuation for a network decreases about 87.9% when the bundle size increases from 1 to 30. |
