Amazon/Deliveroo: Dynamic Counterfactual Analysis and the Failing-Firm Defence

Amazon logo

The economic and financial impact of the Covid-19 pandemic foreshadows an increase in the number of deals where the so-called “failing-firm defence” (“FFD”) might come under discussion, as recently demonstrated by the provisional clearance of Amazon’s investment in Deliveroo by the Competition and Markets Authority (“CMA”). This blogpost addresses the interplay between the FFD and dynamic counterfactual analysis, in light of the CMA’s decision and the overall approach employed by the European Commission (“Commission”) at the EU level.

I. The FFD and Dynamic Counterfactual Analysis: a refresher

Competition agencies may decide to clear an otherwise problematic merger if one of the merging parties is a failing firm, the basic requirement being that the deterioration of the competitive structure that follows the merger cannot be said to be caused by the merger. Agencies’ approaches to the FFD are broadly similar and parties have a high bar to meet, as shown by the very limited number of cases in which the FFD has been successfully argued. The CMA applies a three-limb test as follows:

Limb 1: the assessment of whether the firm would have exited the market (through failure or otherwise) absent the transaction (“limb 1”). Consideration is given both to whether the firm is unable to meet its financial obligations in the near future and to whether it is unable to restructure itself successfully;

Limb 2: the assessment of whether there would have been an alternative purchaser for the firm or its assets than the envisaged merging partner (“limb 2”). If there are alternative purchasers, the CMA is expected to consider what impact an acquisition by one of these would have on competition compared to the transaction as part of the significant lessening of competition analysis (see here, para. 4.3.11); and

Limb 3: the assessment of what the impact on competition would be if the failing firm exited the market and how this would compare to the impact of the acquisition (“limb 3”). The wording in its Merger Guidelines (see para. 4.38) is somewhat misleading (see here, para. 20). Depending on the nature of the markets at issue, the CMA will not only consider what might happen to the sales of the merging party but will also consider the impact that the deal is likely to have on “competition more broadly“. This may thus require a more dynamic and forward-looking counterfactual analysis about future competition with and without the effects of the deal.

The EU Commission’s Guidelines set forth similar criteria, although, historically, some have criticized it for employing a narrower approach and establishing an excessively high standard of proof (see here). The Commission’s approach to the FFD, first developed in Kali und Salz (see para. 19 and ff.) was further refined in BASF/Eurodiol/Pantochim (see paras. 142 and 143), where it specifically considered that “the application of the concept of the rescue merger requires that the deterioration of the competitive structure through the merger is at least no worse than in the absence of the merger”. Thereafter, the FFD was successfully argued in Nynas/Shell, a case that concerned the acquisition of a division of Shell, its Harburg refinery, and in Aegean/Olyimpic II, an acquisition which had been forbidden two years before.

At the outset, it is also worth recalling that parties may wish to consider dynamic counterfactual analysis within the assessment method even prior to the FFD stage. The FFD is – by definition – a solution for a problematic deal. Having regard to the broader competitive environment and the financial impact of Covid-19 on the merging parties, they may wish to ponder whether the deal does lead to a significant impediment to effective competition in the first place, notably by the forward-looking financial projections of the parties and indeed their weaker competitive position in the absence of the merger (see, for instance, T-Mobile/Tele2, para. 565). The CMA has been indeed progressively relying upon dynamic counterfactuals, specifically within the context of digital markets. As summarised in Paypal/Izettle: “(…) [it] is a fast-moving and dynamic market (…) we have taken account of a range of forward-looking evidence (…) that allows us to form expectations about future competition, both with and without the effects of the Merger. This has required us to investigate thoroughly evidence available from internal documents, of the parties”.

II. Amazon/Deliveroo, dynamic counterfactual analysis and the FFD

In April 2020, the CMA provisionally cleared Amazon’s investment in Deliveroo, as a result of the application of the FFD. Previously, the enforcer had announced the Covid-19 pandemic has not brought about any relaxation of the standards by which mergers are assessed or the CMA’s investigational standards and that its approach to the FFD would remain unchanged.

In December 2019, the CMA took the view that Amazon’s acquisition of a minority stake in Deliveroo could reduce the likelihood of Amazon re-entering the online restaurant delivery market and thus initiated a Phase II investigation. The CMA thus considered a dynamic counterfactual: in the absence of the merger, there would be enhanced competition in the market by the future re-entry of Amazon.

The enforcer found that the online restaurant delivery market is highly concentrated with just three large suppliers operating in the UK: Deliveroo, Just Eat and Uber Eats. Amazon operated a competing online food platform in the UK until November 2018 (para. 7). The CMA went on to hold that the deal gave rise to a realistic prospect of a significant lessening of competition as a result of horizontal unilateral effects in the supply of online convenience groceries in the UK. According to the CMA’s provisional findings, Amazon’s investment in Deliveroo was strategic, offering rapid food delivery is important to the company, and so it may have looked to invest in an alternative business absent the merger (paras. 8 and 9; paras. 163 to 179).

Yet, in the aftermath of the Covid-19 pandemic and its resulting impact on Deliveroo’s revenues, the CMA has provisionally concluded that, with respect to the FFD and having regard to the Covid-19 pandemic (para. 4.24 and ff.):

Limb 1: Deliveroo would, most likely, exit the market without receiving funding from Amazon through the acquisition. The company has a pressing need for additional funding and Covid-19 lockdown caused significant declines in Deliveroo revenues (on the supply side, the company has experienced a significant reduction in the number of restaurants operating during the crisis, and the restaurants that have closed or reduced operations have included major brands; on the demand side, Deliveroo stated it had experienced a significant decline in customer orders);

Limb 2: Deliveroo has no alternative options available to it to preserve its business. The three main options explored by the company for raising additional funds were additional funding from existing shareholders; external funding from potential new shareholders; or some form of debt funding. The CMA considered the particular structure of Deliveroo’s business and its need for regular external funding to continue to grow its business and compete. It held that while securing additional funding from other sources may have been possible before the coronavirus outbreak, the pandemic has severely limited the availability of finance for early-stage businesses. Furthermore, alternative sources of funding (e.g. from new shareholders) would not likely be available within the required timeframe. The CMA thus considered that this requirement was met due to a “combination of exceptional circumstances”;

Limb 3: as the case did not involve a full acquisition of Deliveroo by Amazon, the CMA applied the general principle underlying this test, thus assessing whether the effect of the transaction is “substantially less competitive than Deliveroo’s exit”. In its assessment, the CMA concluded that Deliveroo’s exit would be more detrimental to customers and competition than the transaction, as the market is highly concentrated. Deliveroo has had a substantial and growing market share in recent years and it has been a major driver of competition for new customers, as well as an important competitive constraint on the largest existing supplier, Just Eat. Furthermore, exit by Deliveroo would result in significantly weaker competition (higher prices and/or less expansion or innovation from the other two leading players) over an extended period of time, even if Amazon ultimately re-entered the market successfully.

III. Key Takeaways for a Post-Covid-19 FFD

Limb 1: it is worth recalling that, in the 2011 Olympic/Aegean Airlines decision, despite the global impact of the financial crisis, most notably in Greece, the Commission considered that the parties had met “some financial difficulties” over the recent months but ultimately, for a number of reasons (see paras. 1992 to 2070), it considered that the first limb of the FFD test was not met. Two years later, in 2013, in what has been called an “embarrassing reversal” (see here, page 10), the parties re-notified the transaction and the Commission cleared the deal based on the FFD (see here and here). In Amazon/Deliveroo, despite the overall economic uncertainty as to the impact of Covid-19, the CMA held that the current crisis was having a significant impact on Deliveroo’s business. Having regard to the cashflow forecasts and financial information provided by Deliveroo, and the advice received by its directors, the CMA was persuaded that, without an assurance that it will receive additional funds in the immediate future, Deliveroo’s directors would be obliged to declare it insolvent. The CMA thus concluded that Deliveroo would have been likely to exit the market as a result of the Covid-19 crisis without additional funds. Inevitably, while a case-by-case analysis and strong evidence-based submissions will be required and certain sectors will be more affected than others, competition agencies should be ready to face the Covid-19 economic impact as leading to severe long-lasting effects.

Limb 2: in Amazon/Deliveroo, although some potential alternative investors may have been available prior to the coronavirus outbreak, that was no longer likely to be the case having regard to the pandemic.  While it is true that parties may lack time to provide convincing evidence suggesting that they have sought to find less anti-competitive purchasers, Amazon/Deliveroo suggests that this requirement can actually be met. This is particularly likely to occur should the target be especially affected by Covid-19 and depend upon regular external funding to continue its business. Indeed, these circumstances interpreted under the balance of probabilities standard (“the most likely counterfactual”, at para. 4.65; see also here) should offer some encouragement to parties wishing to rely on the FFD.

Limb 3: the FFD is successful where the parties persuade competition agencies that, absent the merger, the same restriction of competition in the market would arise, as the CMA held for instance in East Coast Buses/First Scotland (para. 73) or the former Competition Commission held in British Salt/New Cheshire Salt Works (para. 5.25). However, in Amazon/Deliveroo the CMA held that, absent the merger, competition would be significantly weakened. The rationale appears to be somewhat similar to the Commission’s reasoning in BASF/Eurodiol/Pantochim (para. 151). Indeed, following other cases such as Bottomline Technologies/Experian Limited or Paypal/Izettle, in Amazon/Deliveroo the counterfactual was one of a dynamic nature and the case draws the attention for its impact on the FFD (see here). That is to say, beyond a formalistic analysis of causality (see here), the case confirms that parties may wish to submit evidence-based dynamic counterfactual analysis as to the overall competitiveness of the market: as mentioned, a key aspect of the CMA’s assessment in Amazon/Deliveroo was the fact that Deliveroo was a major driver of market competition and, absent the merger, competition would be weakened due to the company’s exit, even if Amazon ultimately re-entered the market successfully.

It is therefore clear that events which occur during the review of a deal, including the business impact of Covid-19, but which are not a result of the merger, can be incorporated into the counterfactual. Overall, the CMA appears ready to consider new evidence regarding financial difficulties caused by Covid-19 as “a matter of urgency”, react quickly and reassess its findings (see here). Indeed, while it remains a difficult argument to run, Amazon/Deliveroo reinforces the parties’ incentives to build a strong FFD, provided they can support their propositions, thereby seeking clearance of otherwise problematic deals.



Virgilio Pereira

Virgilio Pereira is a PhD Candidate at King’s College London. He was a Blue Book Trainee at the Directorate General for Competition of the European Commission.

Zusammenhängende Posts

18. Mrz 2024
von Daniel Mandrescu
competition law, abuse of dominance, apple app store, the digital markets act

The Apple App Store – A New Kind of Hallmark Case

After almost three years since the Commission sent Apple its statement of objections, which was significantly trimmed down, the Commission reached a finding of abuse for which it imposed a whopping fine of 1.8 billion euros. Alongside this case, Apple was also involved in an almost identical case running parallel in the Netherlands, with similar findings. Meanwhile, during these procedures, […]
07. Nov 2023
Features von Daniel Mandrescu
app store, apple, abuse of dominance, platforms, ACM, art. 102 TFEU.

The ACM vs. Apple AppStore – A Second Chance To Get It Right

The Dutch case concerning the Apple App Store appears to make a (welcome) comeback. The case that started in 2019 came to a rather disappointing end in the summer of 2022 when the Dutch competition authority issued a public statement that gave the impression that it was satisfied with Apple’s adjustments to the App Store front in the Netherlands. This […]
26. Okt 2023
von Daniel Mandrescu
airport travel

Booking / eTraveli: assessing envelopment strategies and mixing up market power thresholds

About a month ago the European Commission announced that it was prohibiting the acquisition of eTraveli by Booking Holdings ( The prohibition, which is a rare occurrence in itself, did not attract much attention beyond comments on the ‘ecosystem’ theory of harm which it may have introduced. But this case offers more than that. First, it shows that current practice […]
12. Sep 2023
Features von Daniel Mandrescu
Microsoft teams antitrust claim, abuse of dominance, European commission

Microsoft III – Paving The Way To A Tying Trilogy?

This summer the European commission (finally) announced it will start a formal investigation against Microsoft following Slack’s complaint concerning the (abusive) tying or bundling or Teams to the Microsoft and Office 365 suites. Not long after, Microsoft came out with an official statement concerning the changes in its pricing and distribution strategy  of Teams it will introduce in order to […]
31. Aug 2023
von Parsa Tonkaboni
The ECJ Judgment in CK Telecoms – Setting the Record Straight? - 0122 Blog post

The ECJ Judgment in CK Telecoms – Setting the Record Straight?

Introduction On 13 July 2023, the European Court of Justice (‘ECJ’) delivered its highly anticipated ruling in CK Telecoms UK Investments v European Commission (‘CK Telecoms’). The Grand Chamber judgment is significant at the most fundamental level. It clarifies some of the core legal concepts and principles at the very heart of EU merger control. The five crucial issues the […]
08. Mrz 2023
Features von Friso Bostoen
Requiem for an objection: the Commission drops half of its App Store case - zhiyue 7DOU5NlNIcE unsplash

Requiem for an objection: the Commission drops half of its App Store case

On 28 February 2023, the European Commission (EC) sent Apple a new Statement of Objections (SO) ‘clarifying its concerns over App Store rules for music streaming providers’. Rather than a clarification, or an expansion of the previous SO, the new SO dropped one of the two objections—an unusual move, especially at this stage of the proceedings. When a startup shuts […]
18. Jan 2023
Features von Daniel Mandrescu
competition law, abuse of dominance, refusal to supply, Lithuanian railways, bronner, essential facility, art. 102 TFEU

Case C-42/21P Lithuanian Railways – another clarification on the Bronner case law and the non-exhaustive character of art. 102 TFEU

The recent case of Lithuanian Railways provides yet another clarification on the scope of application of the Bronner case law. The Judgement of the CJEU reconfirms exceptional character of the Bronner case law and the type of situations it is intended to apply to. By doing so the CJEU potentially helps prevent future disputes of a similar  nature in the […]
03. Jan 2023
Features von Daniel Mandrescu
facebook, competition law, abuse of dominance, art. 102 TFEU, multisided platforms, dominant position, tying and bundling, unfair trading conditions, competition economics, european commission,

On-platform Tying or Another Case of Leveraging- A Discussion on Facebook Marketplace

Just before 2022 ended the Commission sent a statement of objections to Meta regarding the potential abusive behaviour of Facebook. According to the statement of objections, Facebook may be engaging in (i) abusive tying practices with regard to Facebook Marketplace as users (i.e. consumers) that log into Facebook and are automatically also offered access to the Facebook Marketplace, without the […]
07. Dez 2022
Features von Daniel Mandrescu
market definition notice, relevant market, market power, market analysis, notice update, digital platforms, multisided markets, multisided platforms, online platforms, SSNIP test, SSNDQ test, Google android, Google shopping, merger control, abuse of dominance

The draft notice on market definition and multisided (digital) platforms – avoiding rather than resolving some of the main challenges

Approximately a month ago the Commission published its draft notice on the definition of the relevant market. The new notice is supposed to replace the old one that dates back to 1997 and thereby bring the entire process up to date with today’s new challenges, particularly in the context of digital markets. A first read of this long awaited document […]
27. Okt 2022
Features von Daniel Mandrescu
tv broadcasting; competition law; art. 102 TFEU; antitrust; merger control

Opinion of AG Kokott in Case-449/21 (Towercast): filling gaps in EU merger control and creating new routes for dealing with killer acquisitions through the DMA 

Earlier this month AG Kokott delivered an opinion that quickly caught the attention of the (EU) competition law community. It covered a matter which has long been left unaddressed after the introduction of EU (and national) merger control rules, namely the possibility to apply art. 102 TFEU to concentrations.  According to AG Kokott, this possibility, which has been thought to […]

Melden Sie sich für unseren Newsletter an, um regelmäßig über unsere kommenden Konferenzen, Lexxion Trainings, Vor-Ort-Workshops und die neuesten Veröffentlichungen von Lexxion informiert zu werden.

Verpassen Sie keine Neuigkeiten und abonnieren Sie unseren kostenlosen Newsletter. Jetzt anmelden!