Let’s call the whole thing off: Outcome-critical UK intervention in US-centric M&A

This … is not a “killer acquisition” … [it] does not fit the “archetype” … [and is] far from… Facebook/Instagram

– Sabre, Response to CMA Issues Statement in Sabre/Farelogix


You say tom-ay-tuh, I say tom-ah-to, you eat pot-ay-tuh, I eat pot-ah-to …

You say no-overlap … I say … Killer Acquisition …

And the music stops. Texas-based Sabre Corp. had beaten the DOJ in federal court in the agency’s challenge to its $360m deal to buy Miami-based Farelogix Inc. But on April 9th, only two days later, the acquisition was blocked by the UK’s CMA. Well before this point, Sabre had felt obliged to protest to the CMA head-on that its deal was “not a killer acquisition”.

But the CMA treated this argument as close to surreal. Sure, the famous painting by René Magritte may be captioned «ceci n'est pas une pipe»… but to the agency it looked damn near certain to be a pipe. On substance, the DOJ and CMA both called “fire” having smelled the “smoke” in certain provocative Sabre text messages and Farelogix statements. Both characterised the deal as Sabre eliminating a “disruptive force” that would reduce airlines’ bargaining power and harm innovation.

The transatlantic similarities end there, however. Farelogix supplies technology to (primarily US) airlines that enables them to create and transmit end-user sales offers to travel agents either directly and bypassing the traditional GDS platform (where Sabre is US #1 and global #2) or using GDS as a passive distribution channel and therefore more cheaply. But as far as UK deal nexus goes, Farelogix had no direct British airline customers. Or any other UK customer, or revenue, or market share, it seems.

How to get around this jurisdictional stumbling block? In essence, the CMA asserted jurisdiction via Farelogix’s supply to American Airlines, which included an interline component with its oneworld alliance partner, British Airways, such that Farelogix “derived value” from and therefore engaged in “supply” to BA and in turn “in the UK” (though it had never received any revenue from BA). As soon as there is any increment, however small, the CMA’s “25% share of supply in the UK” test is met.

On May 1st, Sabre terminated the deal, despite the home court win. Sabre will challenge the CMA’s assertion of jurisdiction as unlawful as a matter of principle in the Competition Appeal Tribunal. Each case (and error of law claim) turns on its own facts and circumstances, but the track record of merger judicial review in the UK is sobering for aggrieved parties, and challenges to jurisdiction have thus far ultimately gone the regulator’s way (South Yorks Transport in the House of Lords, Eurotunnel in the Supreme Court).

While unique in some respects, the Sabre/Farelogix prohibition is not a one-off or outlier but forms part of an emerging pattern with respect to UK merger enforcement against US-centric and other foreign-to-foreign deals in dynamic, innovation-heavy markets.

  • Last June, Thermo Fisher abandoned its $925m acquisition of California’s Gatan, active in the global electron microscope systems market, despite no EU filing and no Second Request. The reason given was CMA opposition (it had provisionally blocked the deal).
  • In January, Illumina dropped its $1.2bn bid for Silicon Valley’s Pacific Biosciences, active in the global DNA sequencing sector, despite no EU filing and no US trial process. It is true that the FTC was ready to litigate but the CMA provisional prohibition – where the CMA plays prosecutor and judge – cannot have helped; as U.S v. Sabre and the FTC’s loss in the Steris/Synergy Health matter show, the US agencies often lose such cases in court).
  • In contrast, in tandem with the FTC, the CMA cleared Roche’s $4.3bn acquisition of Spark, a potential entrant via gene therapy (GT) into haemophilia A treatment, where Roche had a blockbuster non-GT drug. Like in Sabre, Spark had no UK sales, customers, or users. In essence, the CMA asserted jurisdiction based on (at least some of) Spark’s R&D personnel in the UK, who worked (at least partially) on the matters in question. The CMA also issued a worldwide injunction on post-closing integration for the duration of its review.

In each of these cases, one can reasonably assume that the UK accounted for a small fraction (and probably less than 10% or even 5%) of the parties’ sales in the relevant market, if the target had any UK sales at all (cf. Sabre, Roche cases).

On the other hand, many of these deals were probably viewed by the CMA as “potential killers”. The UK debate, via the Furman Report and the CMA, has appropriated the term “killer acquisition” from the influential empirical study of pharma transactions, which defines this to mean a deal in which the target (or its R&D) was shut down to preserve market power. The UK is not alone (see earlier blog post) and the term has been extrapolated, enthusiastically and without hesitation, to Big Tech and more broadly to any incumbent vs. challenger setting in digital, e-commerce, healthcare or other innovation markets. To round out a top five that would have fitted the CMA’s bill in addition to the Sabre, Roche and Illumina cases: Experian/Clearscore (free credit scores, deal provisionally blocked and abandoned in 2019) and PayPal/iZettle (mobile offline and online payments, Phase 2 clearance, 2019).

In 2020, the CMA has strong incentives to assert jurisdiction over any deal that matches the rough-and-ready incumbent/challenger sketch description on the UK’s Most Wanted: Killer Acquisitions poster -- ever since the remorse began in earnest (in 2018 onwards) following on the poster-child of “naive” UK merger decisions, 2012’s Facebook/Instagram. The theory of harm focus, however, is not just new candidates based on the CMA’s “post-mortem” take on Facebook/Instagram – that the target may, but for the merger, have become a challenger to the acquirer’s dominance in its core (platform) market. The CMA has also considered the inverse: the potential for the acquirer to innovate into, or expand or enter (alone or via other investment/M&A) into the target’s complementary area of strength: see e.g. Sabre and PayPal cases. Consistent with the evocative power of the terminology in the enforcement narrative, these may become known in influential circles as “reverse killer acquisitions”.

Moreover, for the CMA’s potential competition radar, the target need not be small or nascent but can itself be a highly capable incumbent in other markets. In the view of three most influential expert reports in UK circles on digital platforms (Furman, Cremer et al, Stigler) this fact pattern can reinforce the dominance of a “platform ecosystem”. Influential commentators like Shapiro have pointed to the likes of Big Tech M&A targets such as DoubleClick, YouTube and LinkedIn; Facebook’s acquisition of WhatsApp (EU Phase 1 clearance) is often mentioned in the same breath as Instagram (UK Phase 1 clearance). CMA cases where such a theory no doubt formed part of the drive to investigate include Google/Looker Data Sciences (after the DOJ had cleared it) and Salesforce.com/Tableau (both 2019). Both these cases, as well as examples such as Sabre, Roche, PayPal and Facebook/Instagram itself, were own-initiative cases “called in” by the CMA’s Mergers Intelligence unit.

Finally, the broad palette of potential competition concerns can feature two incumbents in different (product or geographic) markets: cases include the CMA’s hypothesis of Amazon as a (re)entrant into UK restaurant home delivery (it exited in 2018; see Amazon/Deliveroo, 2020) which it also applied to Takeaway.com (it exited in 2016; see Takeaway.com/Just Eat, 2020).

The issue for US and other players in the tech innovation sectors: the tension between the enforcement policy interest of the CMA (worried about loss of potential competition in global markets which would impact the UK as elsewhere) and traditional understandings of legal and economic nexus in multi-jurisdictional deals (exemplified in UK decisions from a now-forgotten era, such as Draeger/Air Shields, where a behavioural remedy was imposed given the UK was a small part of the market and the production assets were extraterritorial).

As noted in an earlier post, the CMA has arguably the world’s most elastic jurisdictional thresholds to the point where it is fair to wonder whether it wouldn’t simply be better for legal certainty not to have them at all. This seems paradoxical... until one considers the current system in action, which drives eye-popping CMA origami on the share of supply test and creates a false sense of security – both as to jurisdictional “safe harbours” and a nominally “voluntary” regime – for those in the US and elsewhere unfamiliar with its nuances. The UK regime requires no mandatory filing but is simply not voluntary (except for sellers), and no deal involving incumbents in tech, healthcare and other markets should rule out engaging with the CMA, whether on grounds that the target is too small, that there is no overlap, that there are no UK sales, customers or users or other superficially compelling grounds that – in the hard reality of UK practice – may mismanage expectations.

Just as Astaire and Rogers make dancing on roller skates look deceptively easy, the same could be said of the UK on the merger control map. But as the above deal casualties show, the regime is a slippery surface; the CMA has stopped the music and caused more than one dancing pair to call the whole thing off. To avoid breaking M&A bones requires learning how to dance on roller skates in potentially extreme British weather conditions. This means mandatory risk assessment, good skating and accent coaches and careful preparation – immediately, even before the UK exits the EUMR one-stop-shop from the end of 2020.