The Phase 2 deal mortality meter: there’s more to death than prohibition

Platypus was launched as a weekly internal UK merger control blog in the Linklaters London competition team. Spotting a gap in the market, it introduced the Phase 2 deal mortality rate, and not a prohibition rate, as our most useful measure for tracking CMA merger control outcomes.

This internal blog is now the nocturnal Platy-day Night Live, while this external version must adapt to the predator hazards of daylight. Regrettably, this makes daytime Platypus more like rice pudding to PNL’s late-night Vindaloo.

But in one key respect it stays true to its origins. The “deal mortality meter” is now in pride of place on the Platypus landing page. It includes (1) prohibitions, but also (2) completed deals that are unwound and (3) mergers abandoned by the parties following adverse encounters with the CMA – i.e. all deals that do not survive the UK merger regime.

Phase 2 deal mortality currently sits around 70% for the period from 1 January 2019: well over twice that of the “early” CMA era from 2014 - 2017. In this post, we explain why we believe this statistic – deal mortality – is a key lens for understanding the trend of substantive outcomes in UK merger control.

Next week we offer some thoughts on what can be made of increased mortality rates. Spoiler alert: Platypus’ Provisional Findings (PFs) are that there is no single cause of death. But, of course, these findings are only provisional.

Focusing on blocked deals only

At various times since the Enterprise Act (the legislation under with the UK merger regime operates) and new EU merger regimes were introduced in 2003 and 2004 respectively, it has been part of European merger control agency narrative to point to the low incidence of prohibitions as a percentage of total outcomes to make a point about interventionism in absolute or relative terms:

• the regime is not unduly lethal – look how few deals we have prohibited; or
• the regime is not more interventionist than the comparator agency – look at the relative rates or total number of deals prohibited.

Fair enough. But of all the single metrics, prohibitions – inherently a small-numbers stat – can be the most misleading of all when consumed in isolation.

First, this stat takes no account of regime deterrence: the degree to which regime intervention in cases that made their way through the regime causes other parties in contemplated M&A not to proceed with higher-risk transactions upon receiving antitrust advice informed by recent intervention outcomes and trends.

The deterrence work of the Office of Fair Trading (OFT) from 2007 (based on surveys) arrived at a 5:1 ratio of deterred transactions to blocked deals, reasoning that this was only a lower bound. Platypus would welcome an updated survey of this type, but in the meantime, the role of competition lawyers and economists in counselling against transactions that should “never have left the boardroom” has been raised by senior CMA officials over the last year. The implication from such sentiment may be: deterrence is not at optimum intensity.

Second, even setting aside deterrence entirely, prohibitions are only a subset of the real-world direct impact of the merger regime on transactions. This fact can be deployed to make the case that greater intervention is to be celebrated or criticised (or neither).

Reasons for a composite metric

To address some of the limitations of relying on blocked deals as a statistic, Platypus instead began tracking deal mortality. The logic of a composite metric of (1) prohibitions of anticipated deals; (2) unwinding of completed deals, and (3) deal abandonments is as follows.

• Prohibitions vs. divestment outcomes. The UK’s merger regime is of course non-suspensory, with the result that the CMA sees a mix of completed and proposed (“anticipated”) deals. Official CMA stats count prohibitions of anticipated deals separately from unwinding of completed deals, which are booked as divestment remedies. The latter is of course technically correct: as a regime process outcome, the compulsory sale of the entirety of the target business is a divestment remedy. However, for most merger regimes and in EU and international merger control parlance, a divestment remedy means conditional clearance, because in most regimes, the deal has not completed and so unwinding the entire thing is not in the frame: if most or all of the deal bothers the agency -- it will block the deal (prohibition). Divestiture does not come into it.
  
  And for the parties and deal-makers, of course, there is a world of difference between the ultimate worst-case scenario – the forced full unwinding of a completed deal, on the one hand -- and a tolerable and potentially even fully planned-for outcome: a deal cleared subject to divestment of a portion of the combined assets (e.g. certain “problem overlap” assets in local/regional markets), on the other. (This will typically leave the industrial logic of the deal intact, especially if it was factored in as a reasonable or likely scenario before signing.)
  
  In fact, a deal unwind surely counts as “worse” for parties than a prohibition: the deal is undone, no less than if a deal is blocked, but with the added costs, disruption and uncertainty of compulsory divestment of the target many months after completion -- potentially (adding injury to insult) at a “forced sale” price considerably lower than the consideration paid on completion. With 20/20 hindsight, the acquirer would never have pursued let alone closed the deal.

• Prohibitions vs. abandonments. Official CMA stats also treat deal abandonments differently from prohibitions. Again, this makes complete sense from a CMA process/decision standpoint. From the parties’ point of view however, they have clearly regarded the deal as fatally wounded by the merger regime no less than if it had been blocked (barring a presumed minority of cases with some other cause for withdrawing the merger plan). Nor are the abandonment numbers trivial. In fact, they dwarf formal prohibition outcomes.

• ‘Early’ or ‘instant Phase 2 mortality’ deal abandonments. When parties abandon an anticipated deal, the CMA cancels the Phase 2 inquiry, as did the Competition Commission (CC) before it. The classic fact pattern is that parties to a smaller transaction abandon their deal on or shortly after referral to Phase 2. In this scenario, the merger regime has (in most cases, absent another cause) resulted in deal mortality, regardless of the odds on the Phase 2 endgame outcome, had the inquiry been concluded. This “not worth it” calculus is generally perceived to be a mixture of assumed risk of an adverse outcome (given the opening of Phase 2, this will never be trivial) and the relatively high cost of UK Phase 2 relative to small deal size, regardless of the view on the odds, and it is neither unusual nor new. There have been 5 such examples since 1.1.2018 (Mole Valley Farmers/Countrywide, Send for Help/SoloProtect, Iconex/Hansol, McGraw-Hill/Cengage, Kingspan/Building Solutions) and 4 such CMA examples in 2014-17 (Pure Gym/The Gym, Joseph Ash/W Corbett, Pearson/Vue, Fenland/Fishers).
• ‘Late stage’ deal abandonments. While no examples feature in calendar 2018, since the beginning of 2019, however, there is an increased incidence of parties abandoning deals some way into Phase 2. In the 2014-17 CMA period, only one Phase 2 anticipated transaction, Clariant/Kilfrost, was abandoned later than (more or less) referral – in that case, after provisional prohibition at Provisional Findings (PFs), and late-stage abandonments were at most a rare occurrence in the CC era. Since 1 January 2019 alone, however, there have been six such examples: 4 cases after adverse PFs (Experian/Clearscore; Thermo Fisher/Roper; Illumina/PacBio; Prosafe/Floatel) and two cases at least 2 months into Phase 2 (TopCashback/Quidco, Taboola/Outbrain), or three times the rate of prohibitions of anticipated deals where the parties persisted in waiting for the PF verdict to be confirmed in the CMA’s Final Report (Sainsbury’s/Asda, Sabre/Farelogix). This trend cannot be disassociated from the relatively new phenomenon of the CMA by which it expressly takes the view in the Notice of Possible Remedies that, at that stage, it appeared to it that prohibition was the only effective remedy (though in some cases this conclusion was implicitly evident to those familiar with CMA remedial principles given the nature and scope of the SLC concerns the CMA had).
• Causation and the counterfactual absent a UK (Phase 2) merger review. Note finally that the stats assume (with good reason, we believe) that the deal abandonment has the CMA process as a material if not dominant cause. If it is clear from public sources that a deal would likely have been dropped even but for the CMA opening Phase 2 or reaching adverse PFs – such as shareholder rejection or some other unfulfilled CP – then we will note this separately, flagging that there is no causal link between UK merger process and deal mortality. There have been no clear examples to date. As set out in the deeper dive, of 7 late-stage deal abandonments in the CMA era, only one deal faced parallel opposition in another jurisdiction (the FTC in Illumina/PacBio) while for the most recent, Taboola/Outbrain (cleared by the DOJ in July 2020) it was the UK and Israeli reviews that were ongoing at time of deal abandonment on 10 September 2020. See further our deeper dive on the issue of the causal link between UK merger regime opposition and late-stage deal abandonments in the CMA era.

The next post will consider the tricky issue of why deal mortality has shot up in the UK.