The Welfare Effects of Incomplete Vertical Integration and Relaxed Price Competition in Common-distributor Channels: an Empirical Analysis of the U.s. Carbonated Soft Drink Industry

Abstract : This paper studies the welfare consequences of vertical integration, focusing on the following two aspects, which have been relatively neglected in the literature, but are considered practically important: (i) vertical integration may facilitate inter-firm coordinated pricing, and (ii) it may not be as successful as expected or even harmful to the integrated entity, due to intra-firm causes such as organizational failure. For this purpose, I use and examine data of two vertical mergers (PepsiCo and Coca-Cola) from the U.S. carbonated soft drink industry to propose an empirical framework to consider these two issues as well as well-known elimination of double marginalization and foreclosure. I first conduct a difference-in-differences (DID) analysis of the price effects of vertical integration, and the estimation results suggest these two causes. I then estimate a structural model of upstream and downstream wholesale bargaining and downward price competition in consideration of common-distributor channels. My analysis suggests that only PepsiCo’s supply chain bene…ted after the merger wave: the other two chains (Coca-Cola and Dr Pepper) lost their pro…ts by more than 15%, and consumer welfare also lowered by 3%.

---

Vertical Integration in Motion Picture Industry

Abstract : This paper examines the impacts of vertical integration on pricing, screening, run length and demand in motion picture industry using movie-theater-day level data from China. Our results indicate that integrated theaters charge higher prices, allocate more screens to and run longer lengths for their own movies than movies distributed by distributors owning another theater chain, which are the evidence of vertical foreclosure. Although the higher prices reduce demand, integrated theaters promote the demand of their own movies by allocating more screens to them and other measures. Finally, we find that the foreclosure is more prevalent when the own movies of integrated theaters have lower quality and faces competition from movies in the same genre, the integrated theaters have higher quality, face a stronger competition from other theaters and is owned by the central government.

---

Vertical Integration, Supplier Behavior, and Quality Upgrading Among Exporters

Abstract : We study the relationship between exporters' organizational structure and output quality. If only input quantity is easily observable, theory predicts that vertical integration may be necessary to incentivize suppliers to take input quality-increasing actions. Using direct measures, from the Peruvian fishmeal manufacturing industry, of (i) suppliers' behavior, (ii) suppliers' ownership, (iii) all within- and across-firm supply transactions, (iv) manufacturers' output by quality grade, and (v) all export transactions, we show the following. After integrating with the plant being supplied (before and after integration) and losing access to alternative pay-per-kilo buyers, suppliers take more quality-increasing and less quantity-increasing actions (and vice versa for de-integrations). Integration consequently causally increases output quality, as we show using exogenous variation in firms' access to suppliers they currently own relative to independent suppliers. Finally, we demonstrate that meeting the demand for high quality output is one of firms' motivations for integrating suppliers, while meeting low quality demand is not. We do so both by interacting (instrumented) changes in the relative price of high quality grades with firms' prior location on the quality ladder, and by interacting changes in relative demand from destinations that import high quality grades with firms' prior export destinations.

---

In-house and Arm's Length: Productivity Heterogeneity and Variation in Organizational Form.

Abstract : This paper studies variation in firm boundaries in the US hotel industry. For most hotel brands, the properties with the lowest and highest room occupancy rates are managed at arm's length by franchisees---a form of outsourcing. Properties with intermediate levels of occupancy tend to be managed by company employees, meaning that hotel management is vertically integrated. We propose a model of organizational form choice that can explain these patterns as the result of trade offs between relative fixed costs, control right assignment, and the option to include state-contingent performance incentives in contracts. Hotel brands prefer to manage properties at intermediate levels of productivity because retaining control increases the brand's relationship-specific investment. Low occupancy properties are outsourced because of lower fixed costs. High occupancy properties are outsourced under rent sharing agreements with the franchisee that complement the franchisee's incentives to invest in the relationship arising from control rights. We show that the complementarity between these incentive mechanisms outweighs the brand's concerns about hold up for high occupancy properties in all but the highest quality brands and in all locations other than where performance incentives are costly due to high local taxes.

---