Essays on the demand for caffeinated beverages

Abstract

This dissertation consists of two chapters focusing on structurally estimating the demand for caffeinated beverages. Accurate demand estimation is essential because analysis of market competition, such as the competitive impacts of mergers, crucially depends on both the type and strength of the demand relationship between products of rival firms. Furthermore, such competition analysis signals policymakers whether to approve a merger between those attempting firms. There exists a history of cooperation between leading soda and coffee firms. Whether a merger between these firms has negative impacts on society heavily relies on the extent of consumer demand substitutability between coffee and soda products. The first chapter specifies a structural demand model framework that incorporates both types of caffeinated beverages, and estimates this model using sales of soda and coffee products in a sample of US markets. Based on the demand parameter estimates, we simulate the hypothetical merger effects on prices and welfare. The counterfactual experiments reveal that mergers between leading coffee and soda firms increase firms’ variable profit, but decrease consumer surplus assuming no merger-induced efficiency gains. Importantly, without a certain magnitude of merger-induced cost efficiency gain, which we document in the findings, the gains of firms are not sufficient to compensate for the welfare losses of consumers. Therefore, the results suggest that policymakers exercise caution in deciding whether to approve mergers between these caffeinated beverage firms. The second chapter illustrates that, compared to static discrete choice demand models, a dynamic discrete choice demand model can better capture "complementary" type consumer choice behavior among pairs of differentiated products. Measuring the competitive impacts of mergers crucially depend on both the type and strength of the relationship between products of rival firms, where sufficiently strong complementarity between products of the merging firms can result in lower price-cost markups post-merger, an unattainable outcome when relevant products are substitutes. Accordingly, hypothetical merger simulations between leading caffeinated beverage firms selling several complementary products predict lower price-cost markups on many products post-merger.