chevron-down Created with Sketch Beta.

ARTICLE

Lessons on Airline Partnerships

Yajing Jiang and Chieh-Hsuan Hu

Lessons on Airline Partnerships
Michael H via Getty Images

Recently, there has been a revived interest in the airline industry among antitrust practitioners, economists. and the antitrust authorities. Of the hotly debated topics, airline partnerships repeatedly appear in the theoretical and empirical literature, focusing on their effects on market outcomes. As is well known to the antitrust community, in the past 12 months, there have been at least two notable district court decisions in this industry, one regarding an alliance between American Airlines and JetBlue, the other regarding a proposed acquisition of Spirit Airlines made by JetBlue. In this article, we briefly explain the intuitions of canonical theoretical models, then survey the empirical studies in the past decade or so, discussing their estimated price and non-price effects of airline code sharing with and without the antitrust immunity (ATI) of joint price-setting, revenue- or cost-sharing joint ventures (JV), and mergers. After conducting an extensive literature review, we provide our final thoughts for practitioners: as Oscar Wilde puts it, “the truth is rarely pure and never simple,” a close examination of facts backed by data and documents is recommended before drawing conclusions recklessly from these research findings.

1. Economic models: code sharing, joint ventures, and foreclosure

Without distinguishing the various partnership forms and treating them all as effective mergers, economic theory suggests that the integration of horizontal competitors could lead to increased market power but could also lead to cost-savings, leaving the net outcome undetermined. Building upon this basic theory, various airline partnership types can be modelled differently and generate different testable implications for empirical researchers.

First, code-sharing occurs when carriers not operating a route contract with other airlines to market and sell their seats Code-sharing carriers often share their facilities and marketing efforts, leading to fixed-cost savings. Additionally, for the hub-and-spoke networks operated by major airlines, researchers have considered a phenomenon called “economies of traffic density,” in that the marginal cost of serving an additional passenger on a given segment diminishes with traffic. This combined with code-sharing, could result in reduced variable costs of operation due to increased load factors in the spoke routes of the itineraries; on the flip side, code-sharing could lead to reduced traffic for hub-to-hub routes, which elevates the marginal costs via the economies of traffic density. Since there are more routes between hubs and spokes than between hubs, people believe that on net, economies of traffic density could lead to cost synergies. Notably, economies of traffic density cannot be assumed to hold true everywhere: when an airport experiences severe traffic congestion, the marginal cost per passenger could increase in traffic.

Some code-sharing agreements come with the so-called “antitrust immunity” (ATI), which allows partners to jointly determine airfares of the interline trips., From an economic standpoint, code-sharing with ATI resembles a vertical relationship in that a carrier operating any trip segment acts as the upstream supplier, and the ticketing carrier operating a segment of the trip serves as the downstream distributor. This complementary relationship results in an additional benefit to interline flight passengers: the elimination of double marginalization (EDM), where partners’ joint pricing decisions drive down prices to the actual marginal cost of the interline routes. That said, literature also finds that such an EDM may not be fully realized if a partner also offers an online product that competes with the interline route, which softens the partner’s incentive to reduce prices to the marginal cost level, as it could recapture some profits if keeping prices high.

One policy-relevant question is whether ATI per se creates any marginal benefit to consumers beyond code-sharing. While the theoretical models remain to be further developed, early work suggests that when airlines compete in price setting and when consumers have differentiated tastes over the number of stops in an itinerary, compared to code-sharing agreements without ATI, having ATI does not further reduce airfares of interline routes because price competition enables partners to altogether remove rents from double marginalization, regardless of whether the partner carriers can jointly set prices.

A deeper form of partnership is a joint venture (JV), which allows partners to obtain a certain level of revenue-sharing (via, for example, network planning) and/or cost-sharing (via, for example, jointly purchasing supplies) beyond the scope of cooperation discussed above. There has not been much theory work around the effects of JVs, with one paper finding that, under economies of traffic density, JVs could benefit consumers from the aligned incentives between partners in both the interline and interhub markets, thereby realizing more efficiencies and avoiding the increased marginal costs due to decreased traffic in hub-to-hub routes. This result also highlights the importance of understanding the degree of size gaps between interline and interhub routes, as well as the cost structure measured by the degree of economies of traffic density.

Finally, airline partnerships sometimes raise foreclosure concerns, as partner carriers may need more incentives to accept flights to their hub cities from outside carriers, which could lead to the entire exclusion of the outside carriers and elevated airfares. Theorists are also exploring the possibility that partnerships fueled with ATI may dampen partner airlines’ incentives to enter new markets with non-stop flights. Separately, although there have been concerns around collusion among airline partners, empirical work shows mixed findings and further theoretical work is yet to be developed.

2. Recent empirical literature: net price and non-price effects

We now survey empirical research about airline partnerships. Table 1 outlines key findings from 11 recent empirical papers, focusing on alliances encompassing code-sharing with and without ATI, JVs, and mergers, revealing the price and non-price effects. Three out of 11 papers find that airline partnerships tend to result in lower airfares due to EDM. Nevertheless, another four papers also find increased airfares, particularly in gateway-to-gateway (GTG) flights, nonstop flights, online flights, and in markets where the alliances possess more market power., For the non-price effects, the literature finds that the elevated prices for nonstop flights may prompt consumers to opt for connecting flights, resulting in increased travel time. Alternatively, alliance formation enables carriers to optimize their flight schedules, potentially leading to higher flight frequencies, increased seat capacity, or expanded hub networks. Quantifying whether consumers benefit from these price/non-price effects can be challenging. To tackle this, some economists opt to link data to theory models by adopting the so-called “structural approach”, which enables them to compare against the “but-for” world without such partnerships to gauge the impacts on consumer welfare. Among the structural paper we surveyed, one paper finds reduced consumer surplus due to the upstream margins with non-ATI code-sharing alliance, while another paper finds no impact on consumer surplus. One paper suggests increasing consumer surplus with airline mergers due to the expansion of network and increasing number of direct flights.

Using a structural approach, economists make explicit assumptions about consumer preferences. By solving for the equilibrium that reflects both consumers’ and airlines’ optimal decision-making, economists estimate a set of parameters that “mimic” the data as closely as possible. This enables them to hypothetically “shut down” airline partnerships to “simulate” market outcomes that ultimately imply changes in consumer welfare. In contrast, reduced-form models refrain from explicit assumption-making or equilibrium-solving and instead rely on regressions to pin down the causal impact. For instance, a reduced-form model may find a negative effect on prices but cannot inform whether this is due to cost-savings or expanded networks. Structural models further allow researchers to estimate marginal costs and markups that are typically not publicly available.

That said, structural models can rarely escape from substantial estimation costs, and researchers may need help defending certain key assumptions. Reduced-form models can instead unpack causal relationships more parsimoniously and robustly.

Regardless of the chosen approach, understanding the assumptions, empirical setups, and data is crucial when interpreting results. For instance, as the table shows, while Brueckner & Singer (2019) find that GTG airfares could be elevated by JVs, Bilotkach & Hüschelrath (2019) find JVs increase capacity on GTG routes but reduce load factors. As another example, Bontemps et al. (2023) find that the American Airlines/US Airways merger modestly improves consumer welfare due to airlines’ endogenous network expansion decisions, which Das (2019) does not consider.

Table 1 Summary of recent empirical literature on airline partnerships and mergers

Paper

Partnership type

Key findings

Net effects

“Reduced-form” approach

 

Ito & Lee (2007)

Code-sharing without ATI

  • Virtual code-sharing itineraries are 5–6% cheaper than online products and can attract price-sensitive consumers.

Price (+).

Brueckner & Singer (2019)

Code-sharing with ATI and JV

  • Airfares by JVs are 7.2% lower than interline fares.
  • ATI or JV leads to higher prices on GTG fares due to elevated market power.

Opposite prices effects in different markets.

Bilotkach & Hüschelrath (2019)

JVs of SkyTeam in 2008, Star alliance in 2009 and Oneworld in 2010.

  • A 3-5% seat capacity increase between partner airlines' hub airports on JV routes, but a 3-5% decrease on routes between competing JV hub airports.
  • Load factors on JV routes decrease by 0.5-5% compared to ATI-only routes due to the elevated market power and increased airfares.

Opposite capacity effects in different markets.

Das (2019)

Merger between American Airlines and US Airways in 2013

  • Price reduction post-merger, more so for markets with larger populations.
  • Departure/arrival delays increase by 15 minutes.

Price (-).

Quality (-)

Yimga (2022)

Code-sharing without ATI in 2002-2012

  • Higher path quality via more direct flights offered.

Quality (+).

Tan & Zhang (2022)

JV of five major carriers from the Oneworld alliance in 2010

  • Airfares of online flights in behind-to-gateway markets increase by 3.5-4.5% due to dampened competition.

Price (+).

Fageda et al. (2020)

ATI and Non-ATI code-sharing in 2010-2017

  • Non-ATI alliances increase flight frequencies by 17-24%, while only a modest increase for ATI alliances.
  • Higher fares for non-ATI flights but lower fares for ATI flights.

Flight frequency (+). Price (+) with non-ATI. Price (-) on ATI.

“Structural” approach

 

Armantier & Richard (2008)

Code-sharing without ATI of Continental/Northwest in 1999

  • The impact of code-sharing on average consumer surplus is not significant.
  • Consumers benefit from lower prices for connecting flights at the costs of increasing stopovers and travel time.

No impact on consumer surplus. Price (-). Quality (-).

Gayle (2013)

Code-sharing without ATI in 2006

  • “Upstream” margin on code-sharing products remains since higher prices help counter competition “downstream” for the carrier's own products.

Upstream margin (+).

Gayle & Brown (2014)

Code-sharing without ATI of Delta/Continental/Northwest in 2003

  • Demand rises, especially in markets with higher joint share, as loyalty program members are more incentivized by better perks.

No collusive pricing behavior by alliances.

Bontemps et al. (2023)

Merger between American Airlines and US Airways in 2013

  • Consumer surplus increases by 0.5%-0.8% due to expanded network and more direct flights offered.

Consumer surplus (+).

 

3. Take-aways for practitioners

A well-developed strand of economic theory literature explains the basic mechanisms behind airline partnerships. However, further work is needed for highly policy-relevant topics, such as the marginal impact of various partnership elements (e.g., revenue- and cost-sharing) on consumers and how those elements distinguish the partnership from a full-fledged merger in the relevant market. Recent empirical research presents cutting-edge tools to evaluate the causal impact of partnerships on market outcomes. In 3 out of 11 papers, price increases are observed in code-sharing without ATI or JV, while another 3 papers reveal price decreases in code-sharing without ATI/ATI and merger scenarios. That is, some research finds that the effect of the elimination of double-marginalization stemming from code-sharing partnerships can be substantial enough to offset upward pricing incentives. Others find elevated ticket prices for markets where the parties possess enhanced market power, such as the nonstop flights. As a result, consumers may be worse off by having to switch to cheaper connecting flights. Given some of the different assumptions and empirical settings in the existing research, it is not surprising to see mixed findings. Accordingly, practitioners should be aware of and understand the existing research, but they should exercise caution and consider facts backed by documents and data when citing numerical estimates from the literature.

 

References

Armantier, O., & Richard, O. (2008). Domestic Airline Alliances and Consumer Welfare. The RAND Journal of Economics, 39(3), 875–904.

Bilotkach, V. (2005). Price Competition between International Airline Alliances. Journal of Transport Economics and Policy, 39(2), 167–189.

Bilotkach, V. (2019). Airline Partnerships, Antitrust Immunity, and Joint Ventures: What We Know and What I Think We Would Like to Know. Review of Industrial Organization, 54(1), 37–60. https://doi.org/10.1007/s11151-018-9636-x

Bilotkach, V., & Hüschelrath, K. (2011). ANTITRUST IMMUNITY FOR AIRLINE ALLIANCES. Journal of Competition Law & Economics, 7(2), 335–380. https://doi.org/10.1093/joclec/nhq029

Bilotkach, V., & Hüschelrath, K. (2019). Balancing competition and cooperation: Evidence from transatlantic airline markets. Transportation Research Part A: Policy and Practice, 120, 1–16. https://doi.org/10.1016/j.tra.2018.12.008

Bontemps, C., Gualdani, C., & Remmy, K. (2023). Price Competition and Endogenous Product Choice in Networks: Evidence from the US Airline Industry.

Brueckner, J. K. (2001). The economics of international code-sharing: An analysis of airline alliances. International Journal of Industrial Organization, 19(10), 1475–1498. https://doi.org/10.1016/S0167-7187(00)00068-0

Brueckner, J. K. (2003). International Airfares in the Age of Alliances: The Effects of Code-sharing and Antitrust Immunity. The Review of Economics and Statistics, 85(1), 105–118.

Brueckner, J. K., & Singer, E. (2019). Pricing by international airline alliances: A retrospective study. Economics of Transportation, 20, 100139. https://doi.org/10.1016/j.ecotra.2019.100139

Chen, Y., & Gayle, P. G. (2007). Vertical contracting between airlines: An equilibrium analysis of code-share alliances. International Journal of Industrial Organization, 25(5), 1046–1060. https://doi.org/10.1016/j.ijindorg.2006.12.002

Das, S. (2019). Effect of Merger on Market Price and Product Quality: American and US Airways. Review of Industrial Organization, 55(3), 339–374.

Fageda, X., Flores-Fillol, R., & Lin, M. H. (2020). Vertical differentiation and airline alliances: The effect of antitrust immunity. Regional Science and Urban Economics, 81, 103517. https://doi.org/10.1016/j.regsciurbeco.2020.103517

Fageda, X., Flores-Fillol, R., & Theilen, B. (2019). Hybrid cooperation agreements in networks: The case of the airline industry. International Journal of Industrial Organization, 62, 194–227. https://doi.org/10.1016/j.ijindorg.2017.10.008

Gayle, P. G. (2008). An Empirical Analysis of the Competitive Effects of the Delta/Continental/Northwest Code‐Share Alliance. The Journal of Law and Economics, 51(4), 743–766. https://doi.org/10.1086/595865

Gayle, P. G. (2013). On the Efficiency of Code-share Contracts between Airlines: Is Double Marginalization Eliminated? American Economic Journal: Microeconomics, 5(4), 244–273. https://doi.org/10.1257/mic.5.4.244

Gayle, P. G., & Brown, D. (2014). Airline strategic alliances in overlapping markets: Should policymakers be concerned? Economics of Transportation, 3(4), 243–256. https://doi.org/10.1016/j.ecotra.2015.02.002

Ito, H., & Lee, D. (2007). Domestic Code Sharing, Alliances, and Airfares in the U.S. Airline Industry. The Journal of Law and Economics, 50(2), 355–380. https://doi.org/10.1086/511318

Tan, G., & Zhang, Y. (2022). Competitive Effects of Joint Ventures in the International Airline Industry. Transportation Research Record, 2676(2), 59–68. https://doi.org/10.1177/03611981211036342

Whinston, M. D. (2007). Chapter 36 Antitrust Policy toward Horizontal Mergers. In Handbook of Industrial Organization (Vol. 3, pp. 2369–2440). Elsevier. https://doi.org/10.1016/S1573-448X(06)03036-6

Yimga, J. (2022). Code-sharing agreements and path quality in the US airline industry. Transport Policy, 116, 369–385. https://doi.org/10.1016/j.tranpol.2021.12.021

    Authors