Booking/eTraveli: General Court Upholds Ecosystem Theory of Harm

On September 9, in a landmark judgment, the General Court upheld the Commission’s 2023 decision prohibiting Booking’s acquisition of eTraveli.

The ruling has far-reaching implications for acquisitions by dominant firms in markets characterized by network effects and allegedly high entry barriers. The Court held that the Commission may rely on novel theories of harm to challenge mergers that consolidate or perpetuate pre-merger competitive conditions, reinforce network effects, and allow an already-dominant company to create an “ecosystem” that rivals cannot replicate, even if that merger involves only a de minimis improvement in the dominant firm’s pre-existing offering or market position.

Background

In September 2023, the Commission blocked the acquisition by Booking—a leading hotel online travel agency (OTA)—of eTraveli—a flight OTA—based on a novel “reverse leveraging” theory of harm.

The Commission found that Booking’s acquisition of eTraveli would allow it to cross-sell hotel rooms to flight customers, making it harder for competitors to match Booking’s one-stop-shop offering and “entrenching” Booking’s existing position in hotel OTAs. The Commission treated such cross-selling as a merger-specific foreclosure mechanism. It concluded that the improvement in Booking’s market position would not be achievable to the same extent through Booking’s pre-existing commercial relationship with eTraveli (which already powered Booking’s flight feature) or through organic growth. By anchoring its prohibition on an “ecosystem entrenchment theory of harm” based on the cross-selling

of complementary products, the Commission departed from the established framework for assessing non-horizontal mergers, which does not rely on cross-selling or any other efficiency-enhancing effect but instead requires anticompetitive effects through post-transaction foreclosure that harms rivals.

Why Is the Judgment Significant?

In upholding the Commission’s prohibition decision, the Court expanded the circumstances in which conglomerate mergers that involve no horizontal overlaps or vertical relationships may be challenged, effectively lowering the intervention threshold under the “significant impediment to effective competition” (SIEC) test. 

• No need to strictly follow the Non-Horizontal Merger Guidelines. The Court confirmed that the Commission is not strictly tied to the theories of harm expressly listed in the Non-Horizontal Merger Guidelines (NHMG). The guidelines merely lay down a “general framework” rather than an exhaustive list of applicable theories of harm. The Court held that the Commission has discretion to develop novel theories of harm—such as reverse leveraging or ecosystem entrenchment.
• No reduction in competitive pressure required where competition is already low. The Court held that an SIEC does not require demonstrating a post-transaction reduction in competitive pressure in markets with “an already low level of competition existing.” In other words, an SIEC can arise, according to the tenth chamber of the General Court, where a merger “consolidate[s] and perpetuate[s]” an already weak competitive structure, making a dominant undertaking’s position “less contestable”—for example by raising entry barriers—even where the market share increment resulting from the merger is negligible and, importantly, even when the mechanism to achieve such minimal market share improvement is not anchored on anticompetitive bundling or tying but on the mere ability and incentive to offer a one-stop-shop and cross-sell. 
  
  Indeed, despite finding several flaws in the Commission’s calculations and concluding that the increment to Booking’s hotels share from the transaction might only be “a few tenths of a per cent,” the Court ruled that in markets characterized by strong network effects, some customer loyalty and inertia, and a wide gap between the dominant firm and its rivals, even a minimal increase—in this case through a negligible growth in market share through one of the few channels for acquiring hotel customers where Booking did not already have a notable presence—could strengthen network effects, chill rival expansion, and lead to an SIEC. 
  
  In essence, although the Court acknowledged that there is no “automatic link” between the dominance and SIEC tests, it relied on Booking’s pre-existing dominance and the factors underpinning it to conclude that the transaction would strengthen that position and give rise to a SIEC. In doing so, the Court did not require the Commission to show that, absent the transaction, an actual or potential competitive constraint would otherwise have grown or become more effective—and that the transaction would prevent such outcome.
• Pro-competitive behavior as a means of foreclosure. The judgment characterizes the merged firm’s enhanced capacity for cross-selling complementary travel services as a foreclosure mechanism. According to the Court, this kind of pro-competitive behavior may impair the ability and incentive of rivals to compete, even where the merged entity does not have the ability and incentive to tie or bundle, and absent any indication that such behavior would prevent customers from switching or multi-homing, or incite them to remain with Booking despite more competitive offers to any significant extent. By treating the fact that the transaction could create a one-stop-shop offering as exclusionary (rather than as competition on the merits), the ruling effectively creates an ‘efficiency offense’.
• Reliance on qualitative indicators alone is sufficient, even where quantitative evidence is clearly available and has been used to establish that none of the alleged qualitative factors could have any significant impact, let alone strengthen the dominant firm’s position. Specifically, while the Parties had been able to quantify inventory allocations, multi-homing practices, user traffic conversion rates, and loyalty/inertia effects and show that they would have a de minimis effect on market shares, the Court nevertheless relied on the same underlying qualitative factors to establish a reinforcement of existing network effects and an entrenchment of an already dominant position. 
  • Despite finding that the Commission did not establish that hotels would transfer any additional hotel room inventory to Booking post-transaction or terminate relationships with rival OTAs, and despite acknowledging that the Commission did not provide an approximate number of additional hotel customers (other than through flawed market share increment calculations), the Court nevertheless inferred that network effects would strengthen.
  • In essence, the Court found that, despite not being the main operator on the flight OTA market, eTraveli had “unique attributes” that would allow Booking to create a travel ecosystem. This was not based on a “static assessment” but “rather a prospective analysis”, focusing on Booking’s ability to accelerate flight OTA growth through eTraveli.
  • The Court also found that the transaction would increase customer loyalty and inertia based on evidence of pre-existing loyalty among Booking’s hotel customers and on Booking’s internal documents.
  • In other terms, to establish the SIEC, the Court “relied” on the very same qualitative factors that the parties had quantified and used as evidence to demonstrate that the transaction would only have a de minimis effect on rivals.
• No impact of the precise counterfactual. The Court ruled that, on the facts of the case, the precise counterfactual scenario had limited impact as the market share increment was marginal under both Booking’s and the Commission’s counterfactual models, and the precise choice of baseline had no bearing on the finding of an SIEC.
• Efficiency claims must be raised early and remain subject to a high and asymmetrical evidentiary burden of proof. The Court rejected Booking’s efficiency claims with regard to hotel customers and hotels as, according to the court, they were not raised during the administrative proceedings. It also rejected Booking’s claimed benefits for flight customers because it had not shown (including based on contemporaneous internal documents) that these would be passed on to customers and, in any event, they would not offset the harm to hotels and hotel customers given the limited overlap between end-users of hotel and flight OTAs. 

Implications?

The judgment validates the Commission’s “entrenchment” theory of harm as codified in its draft Merger Guidelines and gives the Commission wide discretion to block acquisitions by dominant firms in digital markets characterized by network effects, even where the merging companies do not compete and any strengthening in the dominant firm’s position is marginal. 

By endorsing “reverse leveraging” and “ecosystem entrenchment,” the Court confirmed that the Commission can prohibit transactions that give rise to a negligible market share increment and involve no direct loss of competition, or price increases, based on qualitative evidence alone. As such, any acquisition that marginally consolidates a firm’s pre-existing dominance can be framed as “perpetuating” a weak market structure. Importantly, the Court considered that the ability and incentive of the merging parties to cross-sell and offer a more convenient product could not be treated as competition on the merits. Instead, it was reframed as a foreclosure mechanism that could strengthen a dominant position and lead to an SIEC. 

The ruling also establishes a stark evidentiary asymmetry: the Commission may establish harm through qualitative evidence alone (even where there is quantitative evidence to the contrary), while merging parties face an exceptionally high hurdle to defend transactions on efficiency grounds.  Specifically, they must adduce concrete, pre-existing internal evidence demonstrating that any claimed efficiencies are merger-specific, verifiable, and directly benefit largely the same consumers as are impacted by the merger.   

This judgment calls for particularly rigorous evidence of efficiencies to be shown in contemporaneous documents and developed from the outset in markets involving network effects, customer loyalty/inertia and complementary products. Any claimed benefits must be quantified, and merging parties must identify which customers will receive them and how they will be passed on. Even then, because the Court determined that the Commission need not quantify the alleged strengthening of Booking’s pre-existing dominance, it is difficult to see how Booking (or a dominant firm in an analogous situation) could demonstrate that the alleged SIEC would be outweighed by efficiencies, however reliably they were calculated. 

Finally, by relying on the efficiencies claimed by Booking to explain its rationale for the transaction as a justification for challenging the merger, the Commission reincarnated an ‘efficiencies offense’ that many believed had been abandoned. In so doing, merging companies in similar situations will need to think carefully before proactively advancing efficiencies in case they are used as evidence of entrenchment.