Disentangling the Price Effects of Joint Ventures in the International Airline Industry: Policy Lessons for South Korea

Over the last three decades, global airlines have joined hands in cooperative efforts that have transformed the industry. Code sharing opened the door to the formation of broader alliances; these groups eventually sought and obtained antitrust immunity (ATI) to lawfully coordinate pricing and schedules on international routes. Today, the apotheosis of industrial collaboration is represented by an especially integrated form of cooperation known as the profit‑sharing, metal‑neutral joint venture (JV). Typically, national competition authorities review the legality of these schemes on a case-by-case basis. JVs allow partners to coordinate fares, schedules, and capacity across routes in a manner that, in some markets, approximates the competitive consequences of a merger, while also promising greater efficiencies in terms of network alignment. Theoretically, the net pro- or anti-competitive effect of any given JV is ambiguous. On the one hand, coordination can reduce costs and resolve double marginalization on connecting itineraries: when two carriers price sequential segments without coordination, each imposes a markup on the through itinerary; a JV internalizes this externality and can lower end‑to‑end fares. On the other hand, joint control over schedules and capacity on overlapping routes can soften competition and allow for higher prices, particularly on gateway‑to‑gateway corridors with concentrated supply. The empirical literature reflects this ambiguity, with studies reporting both fare decreases on connecting flows and fare increases on nonstop overlaps. A key reason for the mixed evidence is methodological: most studies take the JV as a single binary treatment, implicitly averaging over distinct mechanisms that work in opposite directions. This paper opens the black box by separating shared marketing from shared operations at the itinerary level and estimating their separate effects on fares. Using a difference‑in‑differences (DiD) design on transatlantic two‑segment connecting itineraries observed before (2004 to 2006) and after (2014 to 2016) the wave of JV approvals that occurred from 2008 to 2010, we find that each additional JV‑marketing segment is associated with a fare reduction from roughly 8.4 to 10.6 percent, whereas each additional JV‑operating segment is associated with a fare increase from 4.5 to 8.1 percent. In a typical case, where one segment is both marketed and operated within a JV, the implied net effect is a decrease of three to four percent. These mechanism‑specific results explain why average JV effects vary across contexts and provide a more robust foundation for antitrust scrutiny.