Parallel Reviews - Parallel Decisions? A look back at the first 12 months of EC & CMA merger reviews

A year in retrospect

Before Brexit formally took effect, resulting in the possibility of parallel merger review by the UK and EU, the UK’s Competition and Markets Authority (CMA) anticipated that a “significant proportion” of its merger reviews would run in parallel with review by the European Commission (EC). In published guidance following Brexit, the CMA then also stated that “where possible and appropriate, [it] will endeavour to coordinate merger reviews relating to the same or related cases” with the EC and other competition authorities.

A year on, we see that approximately 20% of the mergers that the CMA examined in 2021 also faced a parallel review by the EC.  Of those 12 cases (see below), so far, the authorities’ conclusions diverged twice. But, even where the review outcomes were similar, the issues or the remedies accepted differed. In this article, we reflect on some of the features of that apparent divergence.

First things first: the substance

At least in principle, the approaches of the EC and the CMA to merger control are similar. The EC’s “significant impediment to effective competition” (SIEC) test is comparable with the CMA’s “substantial lessening of competition” (SLC) test. They also adopt similar approaches to market definition and consider the same core theories of harm. A key area of difference is, arguably, the threshold for referral to a Phase II investigation, with the CMA applying a lower standard – i.e. a “realistic prospect” of an SLC vs. the EC’s “serious doubts” as to compatibility with the common market.

However, as explained below, evidence from the first year of parallel reviews indicates that similar tests do not always lead to similar outcomes. Indeed, even before Brexit, the CMA was signalling its conceptual divergence with the EC’s approach to remedies in mergers over which the CMA had no jurisdiction, with the CMA’s Andrea Coscelli stating that it would not have “worn” the Google/Fitbit remedy as the CMA is “…quite sceptical about this type of complex, long-running behavioural undertakings that require quite a lot of monitoring and we have rejected similar undertakings in cases over the years in the UK”. In 2021 the CMA also opened an investigation into the divestment of certain assets of Willis Towers Watson to J. Gallagher & Co. – a divestment which formed part of the EC’s approval of the Aon/Willis Towers Watson merger.^[1]

How about procedure?

Even where there is alignment on substance, the authorities’ different review timetables and procedures might mean that streamlining is not easy. The EC has a 25 working day deadline for its Phase I review, extended to 35 if remedies are offered, while the CMA has a longer 40 working day Phase I period, followed by additional days for remedies discussions. The authorities’ “standard” Phase II statutory review periods also differ (24 weeks at the CMA vs c. 18 weeks in the EU) but both are subject to extensions and “stop the clock” decisions.

Particular difficulties may arise if trying to align remedy discussions with the authorities as the CMA would in principle only entertain such discussions at set points in the process which are usually much later than in the EU. This challenge is, however, recognised in the CMA’s updated guidance on jurisdiction and procedure, where it states that the CMA may depart from its standard practice “where there is an appropriate and reasonable justification for doing so, which may include alignment of the CMA’s investigation with the processes of other competition authorities.”

Both authorities are known for their use of lengthy periods of pre-notification which are not subject to statutory limits in either jurisdiction, further complicating the ability to align timetables.

Even if parties start engaging with both authorities simultaneously, there is no guarantee that the formal clock will start at the same time, or with a lag that enables coordination of Phase I reviews. Despite having a shorter Phase I period, the EC was the first to initiate formal reviews in five of the 12 parallel cases examined last year. In certain cases, the CMA started the clock weeks after the EC Phase I review commenced. In IAG/Air Europa, for example, the CMA launched a short-lived Phase I review in November 2021, four months after the EC referred the deal to Phase II.

In appropriate cases, one way for parties to achieve more streamlining in the process could be to ask the CMA to fast-track the case to an in-depth Phase II review – as was done in Cargotec/Konecranes – which resulted in the EC and the CMA simultaneously reviewing the transaction at Phase II.^[2] This strategy is limited though: it assumes a Phase II review is a necessary outcome and relies on the timing for the EC review to be sufficiently crystallised by the time the CMA fast-track application is made.

Same, same, but different

As shown in the table below, in two of the cases examined, at first sight the reviews had diametrically opposed outcomes:

• Meta/Kustomer was cleared unconditionally at Phase I in the UK, but cleared subject to remedies following a Phase II review by the EC. In that case, the regulators pursued broadly similar theories of harm but arrived at different conclusions. The EC’s Phase II review focused in particular on the alleged vertical foreclosure of CRM providers to Meta’s B2C messaging channels (WhatsApp, Messenger or Instagram) with access remedies being required to alleviate the EC’s concerns. The CMA also considered whether customer service and support CRM suppliers that compete with Kustomer could be foreclosed in a similar manner and concluded that Meta would lack the incentive to pursue such a strategy in light of its wider business plans. Both authorities dismissed theories of harm related to data aggregration.
• Veolia/Suez was cleared conditionally at Phase I by the EC, and referred to an in-depth Phase II review by the CMA. The CMA’s review is currently ongoing.

Those two cases do not tell the whole story as we also see cases where the EC and the CMA have conditionally cleared transactions at Phase I but on the basis of different divestment packages. In S&P/IHS Markit, there was significant divergence in terms of the theories of harm pursued. Again, the EC maintained vertical concerns, which were not deemed sufficient to give rise to an SLC by the CMA, and the CMA also pursued a horizontal theory of harm which was not deemed to give rise to an SIEC by the EC. Interestingly, the regulators reached opposite conclusions as regards horizontal effects and closeness of competition between the parties, based on responses to their market investigation, including in cases where it could be expected that the same or similar market participants were contacted. This series of divergences was naturally reflected in the remedies offered, with different parts of the businesses needing to be divested in order to address each regulator’s concerns.

Points of difference also took the form of specific national security concerns (see NVIDIA/Arm), focus on coordinated effects (see SK Hynix/Intel) and approach to vertical relationships (see Alexion/AstraZeneca).

The EC and the CMA in 2022: New Year’s Resolutions?

It seems obvious that in order to mitigate the risk of divergent merger review outcomes, greater inter-agency cooperation is needed. While the CMA has in place a framework for antitrust cooperation with four of its international counterparts (Australia, Canada, New Zealand and the US), there is currently no agreed cooperation protocol with the EC despite both agencies signalling this as a policy priority. It remains to be seen what, in practice, this will add beyond the waivers currently being agreed routinely to facilitate coordination between the two authorities.

In any event, merging parties on relevant transactions should now clearly be prepared for intensive reviews by both agencies, but should not assume that those reviews will generate identical conclusions or outcomes.