Decisive Influenc-er? The Proposed UK Backstop to Block the Next ‘Facebook/Instagram’

Part 2: The Case for the Prosecutor-ial

The UK Government’s proposal to confer on the CMA the substantive power to block a GAFA deal using a lower Phase 1 standard is controversial: for the first time in any major regime, it allows (global) prohibition of a merger that is by definition, unlikely to give rise to competitive harm (albeit on the premise the harm would be very large, if realised).

Our Part 1 blog post considered increasing the international legitimacy of the UK proposal by market testing it with peer regimes with typically the most nexus to such deals, notably the US and the EU. This Part 2 considers another potential route to making the proposal more legitimate: by sliding the accountability and robustness scale to the maximum the UK legal system has to offer.

A stacked deck

By international standards, the CMA’s Phase 2 decision-making deck is already heavily stacked in its favour. By virtue of the CMA’s administrative model, and the high deference granted to the regulator in UK judicial review, the CMA has radically more discretion (a.k.a. power) when compared to its counterparts in the US and the EU. Allowing the CMA to intervene where there is only a realistic prospect of harm to competition means a greatly increased scope for the CMA to make judgments that (inherently, given the test) can only be reasonably speculative, perhaps highly so.

As to this Phase 1 test, Platypus would suggest it is a fairly settled view among the UK antitrust bar that it is not really possible to challenge a Phase 1 referral decision to Phase 2 in the UK courts. This is not because, as an interim decision, it is non-justiciable, as an EU ‘serious doubts’ decision to open Phase 2 is under EU law. In the UK, no one has ever spent money attempting such a challenge in 18 years, in part because it is thought better to argue the case in Phase 2 (given the prize of winning in court is a remittal to have the CMA do a do-over) but also because CMA discretion is so wide  because the Phase 1 test is so low. Conversely, Phase 1 clearance decisions have been appealed several times (IBA Health, Unichem, Celesio, AC Nielsen) partly because… the Phase 1 test is so low (as per the Court of Appeal in IBA: it stretches down to “more than fanciful”).

The legitimacy index: with greater power, should come greater accountability

If the proposed special merger regime (applying a Phase 1 test at Phase 2) is an idea whose time has come, perhaps it is time to rethink the process by which these decisions are stress-tested.

That would not be the light-touch judicial review that currently applies to UK merger control or even the merits appeal system that applies to cartel and abuse of dominance cases (under the Competition Act 1998 equivalent to Articles 101/102 TFEU). Rather, it would be a system familiar to and endorsed by the United States: the prosecutorial model.

Calls for reform in favour of a prosecutorial model have consistently been resisted in the UK for ordinary merger control. But what is being proposed for SMS is extraordinary. And while the CMA may not adopt this label, and indeed may resist it, in substance, it has shown itself to be a prosecutor par excellence for cases where it decides there is a problem.

Platypus’ proposal is therefore this: if the new test is to be introduced, then harness this prosecutorial strength with another strength of the UK regime, a specialist competition court a.k.a. the Competition Appeal Tribunal (CAT). Under this model, the CMA would prosecute an SMS merger in the CAT. The CAT would decide whether the evidence supported the Phase 1 threshold, e.g., “realistic prospect”. (Platypus believes that this is the wrong formulation for a lower test, but that is for another post). In doing so, the CMA and the merging parties would not only have recourse to documentary evidence and expert economic evidence, but also live witness evidence to assist the CAT in passing judgment.

The crucible of cross-examination: what did you mean when you wrote that?

A prime theory of harm in incumbent/challenger cases is potential competition and in turn the respective strategic intents of the target and the acquirer but for the transaction (i.e., would Instagram have become a social media competitor to Facebook? According to Instagram's senior executives? According to Facebook’s? What were Facebook’s motivations for the deal?)

The prime starting point of evidence for strategic intent is email and other internal documents: the Government in its Consultation on Facebook/Instagram cites a Mark Zuckerberg email quote (and emails are prominent in FTC v. Facebook). The fact that the ex post Lear Report commissioned by the CMA said it could not conclude whether the Instagram deal was net anti-competitive in retrospect is often glossed over in the prominent citation of this case as a justification for radical reform. However, in fairness to Lear, we should note that the CMA probably did not review internal emails in 2012, and, for Lear, its review (published in June 2019) predated by about a year those Facebook emails that have entered the public domain (as a result of the US House Judiciary subcommittee inquiry in mid-2020 and which feature in the FTC’s subsequent case).

Against this backdrop, the ability to have examination and cross-examination of witnesses in court, alongside internal documents, would be an invaluable addition and surely assist the CAT in reaching the right outcome in the context of the CMA’s revised lower test. This would allow evidence (especially inferences from internal documents) to be tested for reliability “in the crucible of cross-examination” as established in English common law. The robustness, reliability and independence of the UK judiciary goes in large part to explain why the forum of choice for foreign litigants is so often English law (see also: New York law).

British courts are not easily captured

Are there good reasons to think the CAT could not cope with such a task? At present, the CAT does review CMA merger decisions for error of law, irrationality and other judicial review grounds but does not reach its own view on the merits of whether the legal test has been met.

However, SMS merger theories of harm, whether taking out a challenger, or foreclosure, are analogous to an abuse of dominance (on which, at present, the English courts can and do hear private causes of action (Streetmap v Google and Arriva v Luton Airport in the High Court; Socrates v Law Society and Achilles v Network Rail in the CAT). The CAT also hears appeals on the merits to CMA decisions on abuse of dominance (see Pfizer v CMA). Platypus sees nothing inherent about such cases that render a specialist competition court ill-equipped to the proposed task, which can include economists on the bench, and addresses some of the issues that arise under a general court system where judges lack familiarity with the nuances of competition law.

Some may object that a prosecutorial model is a pretty heavy-handed and burdensome (expensive, time-consuming) system. We nonetheless propose it because blocking global mergers that are not likely to harm competition is itself pretty heavy-handed and intrusive. The stakes are very high, and the Government’s proposal does not envisage a large volume of cases.

Pro-enforcement voices may also argue that the judges wouldn’t “get it” or be too lenient or seduced by business executive testimony. However, the pool of judges on the CAT bench could include not only the present pool of highly-regarded judges but also the calibre of excellent individuals that serve on CMA Panels. In the case of the latter, the difference would be not the qualified individuals but the institutional and procedural context of their decision-making, arrived at via the adversarial framework of common law litigation. This would arguably deal with user dissatisfaction with the Phase 2 process, where there is no “day in court” and where the case team has radically more (weekly) access to the Panel at critical decision-making points than do the merging parties. In short, the objection in other regimes that “judges don’t get it” is not, fairly viewed, a deal-stopper in the UK.

A further answer to this is that, if the prosecution’s evidence really can establish a “realistic prospect”, then it should be capable of persuading an independent tribunal that this low threshold is met. If it cannot, perhaps the prospect is not realistic, in which case the policy concern (avoiding under-enforcement even if it requires some over-enforcement) does not apply in the case in question.

By separating the role of prosecution and judge, the CMA can maintain a coherent policy position, win or lose, without outside perceptions (either in the UK or beyond) that the CMA or individuals within it reached judgments that were politically influenced in political cases. The unintended consequence of reform could be the (actual or perceived) politicisation of the CMA in high-profile cases so as to damage its brand.

Conclusion: Max Britannica?

The Platypus brand took inspiration from the fact that the UK merger regime is unique and, if it did not exist, it would probably not have been invented now. As much as we are fond of its strengths, as a global blueprint for other regimes, one might fairly say: “I wouldn't start from here”. And yet the proposed SMS merger reforms keep all the idiosyncratic process and architecture in place – but add more discretionary power to block what will mostly be foreign-to-foreign mergers on global markets.

A proposed UK prosecutorial model, as a proportionate quality control mechanism for such extensive new powers, would substantially aid in achieving cross-border legitimacy and de-risk politicisation. It may be a compromise that neither satisfies the CMA (who may understandably wish to retain the power to decide) nor certain SMS firms (who would understandably resist any lowering of the test at all).

But it is precisely this element of compromise that the Government should consider if it really does wish to proceed with a new regime. It may wish to embrace fully the pro-reform view but should recognise that building the SMS merger regime into CMA architecture is a marked increase in power wielded by one already very powerful authority, and with global ramifications. A combined CMA/CAT solution to this issue would utilise the UK regime in a way that, relative to the counterfactual of reform that is currently on the table, would increase legitimacy and decrease the risk of politicisation.

If the UK Government really does opt for all-British on its wish list (see Part 1), then on this issue, a combination of the CMA and the UK judiciary is really the best all-British that Britain has to offer: Max Britannica.