The recent European Commission decision to prohibit the acquisition of Etraveli by Booking.com marks an important departure from previous practice and guidance for non-horizontal mergers and is of particular relevance to digital mergers involving ecosystem theories of harm. The EC hierarchy has repeatedly said it is willing to entertain novel theories of harm to deal with digital mergers, and that the current non-horizontal merger guidelines reflect past precedent, but do not capture the evolution of the EC’s thinking in recent years. The EC’s main theories of harm in Booking / Etraveli are novel in the sense that it is a conglomerate case in which the concern is not about leveraging dominance into new or nascent markets but about adding functionalities that protect the acquirer´s market position in its core market. It is also a case in which the counterfactual does not reflect prevailing conditions of competition but an assumption that Booking would exit its successful flight Online Travel Agency (OTA) business.

In this blog post, we will discuss the deal, the EC's theories of harm, the points of divergence with the decision of the UK’s Competition and Markets Authority (CMA), and the implications of this decision for future digital mergers.

The deal

The deal involved parties active in two different markets: Booking, which is active in the market for hotel OTAs, and Etraveli, which is active in the market for flight OTAs. Booking’s EEA market share in accommodation is close to 70%. Etraveli’s EEA market share is not mentioned in the EC’s press release (the full EC decision has not yet been published) but its UK market share is between 10-20% according to the CMA’s decision. Booking also offers flight OTA services, but under a partnership agreement with Etraveli. Booking’s rationale for the acquisition was to launch a new seamless service – “The Connected Trip” – to cross-sell between different travel verticals.

The theory of harm

In a case like this, a traditional horizontal theory of harm would focus on the elimination of actual or potential competition between the merging parties in either of the affected markets. For example, in a “killer acquisition” scenario, a key concern would be whether Etraveli is a potential source of future competition that could challenge Booking’s existing dominance in the hotel OTA market. However, this is not what the EC seems to have been concerned about in this case.

Instead, the concern is a conglomerate theory of harm in the sense that the EC focuses on the close link between the hotel OTA and flight OTA markets, and the fact that they serve the same customer groups. A traditional conglomerate theory of harm would normally focus on the risk of leveraging a dominant position in order to obtain an advantage in another market using practices such as tying or bundling. In this case, however, the concern is that adding a seamless service including flights would entrench Booking’s dominance in its core market: accommodation. However, to go as far as to prohibit the acquisition of Etraveli on this basis appears to lower the bar for intervening against digital platform mergers that offer bundling opportunities: a seamless service is a feature which is often viewed by merging parties as a customer benefit resulting from the merger. This is also why critics have labelled the EC’s reasoning as an “efficiency offence” theory of harm that is incompatible with modern, economics-based merger control.

In its 2022 digital mergers workshop, the DG COMP representative mentioned that such theories of harm (not referring to this case, but more broadly) are mentioned in para. 36 of the EC’s horizontal merger guidelines:

“Some proposed mergers would, if allowed to proceed, significantly impede effective competition by leaving the merged firm in a position where it would have the ability and incentive to make the expansion of smaller firms and potential competitors more difficult or otherwise restrict the ability of rival firms to compete. In such a case, competitors may not, either individually or in the aggregate, be able to constrain the merged entity to such a degree that it would not increase prices or take other actions detrimental to competition. For instance, the merged entity may have such a degree of control, or influence over, the supply of inputs or distribution possibilities that expansion or entry by rival firms may be more costly. […]”

The EC press release refers to flight OTAs as an important “customer acquisition channel” for hotel OTAs. Essentially, if Booking’s rivals in hotel OTA services are cut off from a significant part of this channel, it would become (even) more difficult for them to develop a customer base and challenge Booking’s long-standing dominance.

The press release also mentions that the transaction would allow Booking to “expand its travel services ecosystem”. Customer inertia or stickiness means, according to the EC, that a significant share of flight customers would be likely to remain on Booking’s platforms to also shop for other services. This, according to the EC, would create additional barriers for rivals to attract customers and at the same time strengthen Booking’s bargaining position vis-à-vis the hotels. All in all, this could, the EC concludes, result in higher costs for hotels and, possibly, for consumers as well. It is not the first time the EC refers to “ecosystems” alongside traditional relevant markets – see for example the Commission’s decisions in Facebook/Kustomer and Google/Fitbit. But it does, in our view, represent a departure from the existing non-horizontal merger guidelines, a fact the Commission does not deny but, as above, considers the backward-looking nature of EU guidance in contrast to US merger guidance, which represents guidance to US courts (as well as the agencies).

Press reports refer to a potential market share gain of 1-3% in the hotel OTA market as a result of the acquisition of Etraveli, which is notably modest. Without having seen the full decision and foreclosure analysis, it is difficult to say how the EC has distinguished between a market share increase which reflects increased sales due to the customer benefit of a seamless service and one which forecloses rivals and is anti-competitive.

A rare case of EC / CMA divergence on substance: is the CMA still the toughest enforcer on the block?

Another notable feature of this case is that the EC prohibited the deal after the CMA’s Phase I unconditional clearance decision. The theory of harm assessed by the CMA in its 2022 decision was very similar to that of the EC: “the CMA’s investigation focussed on the change in market structure in the supply of accommodation OTA services in the UK brought about by the Transaction and, in particular, on the impact of the potential loss of Etraveli as a customer retention and/or acquisition channel for rival suppliers of accommodation OTA services.” However, unlike the EC, the CMA found that Etraveli was “not a particularly important customer […] acquisition channel” for hotel OTAs. The CMA noted that UK consumers tend to shop around for travel services rather than source all their needs from a sole supplier, that Booking and its rivals in hotel OTAs use several customer acquisition channels (accounting for large shares of revenues) that would be unaffected by the deal, that Etraveli has a modest UK market share in flight OTAs, and that the vast majority of UK consumers (87-89%) purchase flights directly from an airline which means that Booking’s rivals will continue to have access to them.

Without seeing the full text of the EC decision it is difficult to know the extent to which the divergent outcomes were driven by different evidence on market structure and customer behaviour in the EU and the UK. However, it is not immediately obvious this – as opposed to different approaches to the question of foreclosure – is the driver of the difference, and some of the CMA’s findings are quite similar to the EC’s finding in previous flight OTA mergers.

Instead, it would appear that the EC has applied a particularly strict standard to Booking due to its entrenched dominant position in the hotel OTA market (while Booking does not currently meet the thresholds for designation as a “gatekeeper” under the EU DMA and thus was not designated in the first round of designations, it has said it expects to meet the thresholds for designation at the end of this year). DG COMP representatives have stated that they will be very careful when a deal risks raising barriers to entry and expansion in the dominant platform’s core market and there are only limited challengers in the market.

Remedies: a hotel comparison page?

Lastly, Booking offered remedies, but they were rejected by the EC. Booking’s proposal was to display four different hotel options from OTAs, ranked in terms of price, in the flight check-out page. However, the EC had doubts about transparency and monitoring of the proposed behavioural remedy, since Booking’s subsidiary KAYAK would control the algorithms (“working as a black box”). Moreover, appearing in the check-out pages would, according to the EC, not give rivals any visibility in other cross-selling opportunities. Therefore, the EC stated that the feedback received during its market test indicated that the proposed remedies were not sufficiently comprehensive and effective to eliminate the competition concerns it had identified.

Implications

The EC's prohibition of the Booking / Etraveli transaction is a clear indication that the bar for prohibiting non-horizontal mergers has been lowered for EC mergers. In fact, there have been reports of internal friction within the EC hierarchy on the ecosystem theory of harm in which the prohibition was rooted, and the EC has arguably moved the goalposts for non-horizontal mergers in departing from its established guidance. Additionally, the case highlights a potential new point of divergence between EU and UK merger enforcement, where the CMA has often been seen as the more interventionist regulator (see our Platypus blog for posts on earlier trends in EU/UK divergence ratios). Companies with entrenched market positions in digital markets could face increased scrutiny and challenges in pursuing mergers where they have no or limited presence in a neighbouring market. Booking has announced its intention to appeal the EC decision, and the outcome of this appeal could have significant implications for future digital economy mergers involving ecosystem theories of harm.