Anthony W. Swisher is a Partner, and Jody Boudreault is a Senior Associate, in the Washington, DC office of Baker Botts L.L.P. Mr. Swisher serves as the WLF Legal Pulse blog’s Featured Expert Contributor on Antitrust & Competition Policy—U.S. Department of Justice.

On May 12, 2020, the Department of Justice moved the U.S. Court of Appeals for the Third Circuit to vacate the U.S. District Court for the District of Delaware’s U.S. v. Sabre Corp. opinion.  In an opinion released on April 7, the court rejected DOJ’s challenge to the proposed merger of Sabre and Farelogix.  Despite winning the DOJ litigation, the parties abandoned their merger before DOJ filed any appeal brief because the U.K. competition authority challenged their deal.  Although the transaction was abandoned, DOJ’s motion to vacate reflects concern that the court’s reliance on the 2018 Ohio v. American Express Co. Supreme Court decision “could affect antitrust enforcement beyond the instant case.”

Specifically, DOJ argued that “this ruling—if not vacated—could have an outsized effect on cases involving competition in the digital economy, where it is not uncommon for multi-sided platforms to face competition from one-sided rivals.”  Perhaps DOJ means that the Sabre holding (in the Third Circuit where the holding is persuasive) might make it much more difficult for the agencies to challenge certain mergers.

Sabre was a litigated challenge to a leading firm’s acquisition of a small rival—the Sabre acquisition of Farelogix in an alleged “booking services” market.  Notably, the court’s analysis of Amex, product market, and economic models to determine competitive effects may heighten the agencies’ burden of proof and inform defendants’ strategy in future merger challenges.

The district court’s reliance on Amex may increase the agencies’ burden of proof in future cases, making it difficult for an agency to prevail in litigation

The district court seems to have increased the agency burden of proof with three holdings in Sabre based on Amex.  First, the district court found that under Amex, two-sided transaction platforms only compete against other two-sided platforms.  Slip Op. at 69-72.  Because Farelogix is “indisputably” “not a two-sided platform” and the Second Circuit had held that Sabre is a two-sided transaction platform, the court held the merging parties do not compete.  Id. at 69.

Further complicating merger enforcement, the decision seems to prohibit the agencies from parsing separate products out of a platform’s offerings.  The district court wrote, “Amex establishes that two-sided transaction platforms such as the Sabre GDS supply ‘only one product’: the transactions that link both sides of the market.”  Id. at 70.

Finally, the court relied on Amex to hold that DOJ must prove harm to competition on both sides of a two-sided platform in these types of cases.  “Even if all of this were not the case,” the court declared, when a two-sided platform buys a one-sided competitor, the government must show harm to competition on both sides of the two-sided market to make the two-sided “platform market, overall, less competitive.”  Id. at 72-73.  In Sabre, the court found that “DOJ did not even try to meet this burden.”  Id. at 73.

Amex decidedly tipped the scales in favor of the defense for the Sabre court.  If the Sabre decision is not vacated, might it do so for other defendants in future mergers?

Relevant product-market definition based on an IT functionality may be successfully portrayed as arbitrary

The district court was persuaded by the disconnect between DOJ’s narrow product-market definitions versus evidence about the parties’ products.  This may make it more difficult for the agencies to define a narrow market around differentiated IT services.  Ultimately, the size of the product market affects the agencies’ burden of proof for the “competitive harm” element of a Clayton Act merger case.

DOJ had to contend with one especially difficult fact: Sabre does not sell any standalone “booking services” products.  Despite this, DOJ argued that the markets were for booking services—transmitting, receiving, and processing flight offers and bookings between airlines and different types of travel agencies.  They alleged that both parties sell products that include booking-service functions.

Rejecting those arguments, the district court held that the parties’ products are in separate product markets.  Sabre does not participate in the same market as Farelogix because Sabre does not sell booking services as a separate product that has separate demand.  Slip Op. at 76 n.17 & 77 n.18.  Rather, booking services is a functionality among the services Sabre “actually sells.”  Id. at 46.  The court noted more than once that DOJ could not determine a value or price for either of the parties’ “booking services.”  E.g., id. at 76.  The court found that in selectively and arbitrarily dissecting Sabre’s service bundle, DOJ’s proposed market “fails to ‘correspond to the commercial realities of the industry.’”  Id. at 77 (quoting Amex, 138 S. Ct. 2274, 2285 (2018)).

Economic models may be successfully attacked as unrealistic

The district court was persuaded by attacks on DOJ’s hypothetical monopolist test and HHI analysis.  The agencies often rely on these economic models to prove anticompetitive effects.  While this line of the court’s analysis is fact-specific, it may also have implications for defendants in other markets.

DOJ had broken down the product market into two distribution channels.  It argued that sales through online travel agencies are a different product from sales through traditional physical travel agencies because they target different consumers.  Furthermore, DOJ argued, neither of these two products includes airlines’ direct ticket sales to consumers on

The district court found the proposed online booking product market too narrow based on a flaw in DOJ’s hypothetical monopolist test.  The hypothetical monopolist test determines what products are in the relevant market.  It asked whether a monopolist who controlled the parties’ booking services could successfully impose a 5% price increase on airlines or would the airlines walk away?  Defendants argued that the binary choice model—either pay a price increase or walk away—is not what happens in the real world.  The district court agreed: Airlines could also “try to steer traffic . . . to” and replace “much of” the revenues derived through online travel agencies.  Slip Op. at 49-50.  The court found that “[f]or an airline, the ‘closest alternative’ to distribution through an [online travel agency] is distribution through its own website.” Id.  As a result, the court included in the online booking market.

Including in the online booking market effectively doubled the size of the market and decreased the parties’ market shares.  The smaller market shares had litigation consequences.  The court said DOJ’s “HHI cannot be relied on because it ignores industry realities.”  Id. at 83.  The DOJ routinely uses an index of HHIs (an index of market concentration equal to the sum of the squares of each firm’s market share) and HHI deltas (the difference between pre- and post-merger HHIs) to show a presumption of anticompetitive harm.  With smaller market shares for the parties in online bookings, both the HHIs and the HHI delta were also smaller.  They were so small in the online booking services market that it left DOJ without a presumption of anticompetitive harm.  This ultimately resulted in increasing DOJ’s evidentiary burden as it could no longer use HHI arithmetic to demonstrate anticompetitive effects in online bookings.

In the traditional bookings market, the court also disregarded DOJ’s HHI analysis.  The government’s mistake there was in attributing sales to Farelogix “even though [another firm] controls the sale and the commercial relationship.”  Id. at 84.

As a result of the failure of its economic models, the district court held that DOJ failed to show any anticompetitive effects.


While the Sabre opinion does not have precedential binding effect, DOJ’s motion to vacate suggests that they fear it will influence other courts.  If that were to happen, this small merger case may produce big wins for future defendants, making it more difficult for the agencies to challenge certain mergers.  Whether DOJ is successful in having this unfavorable opinion vacated remains to be seen.  Even if they are successful, however, the decision provides a roadmap of arguments available to future litigants.