Competition and Markets Authority v Apple Inc & Ors

The central issue arising upon this appeal concerns the circumstances under the Enterprise Act 2002 (“EA 2002”) in which the Competition and Markets Authority (“CMA”) has jurisdiction to make a market investigation reference (“MIR”) having, in the past, concluded that it would not conduct such an investigation. To understand the appeal, it is necessary to have an appreciation of the various powers the CMA is entitled to exercise in order to conduct an investigation. There are two principal regimes under which the CMA has jurisdiction to make a MIR.

The first (“the standalone power”) is broad in nature. It is subject to certain conditions precedent the most important of which is that that CMA has reasonable grounds for suspecting that any feature or combination features of a market for goods or services prevents, restricts or distorts competition in the United Kingdom or a part thereof. This power is found in section 131(1) EA 2002.

The second (“the MSN power”) arises where the CMA wishes to investigate a matter but considers that it does not, at that point in time, possess sufficient information. Under the EA 2002 it is empowered to exercise its power to collect information by publishing a market study notice (“MSN”). In such circumstances it is required, under sections 131A and 131B, to adhere to a procedure the ultimate effect of which is to require the CMA within fixed time periods to take a decision upon whether it intends to investigate or not, and why.

In the present case, the CMA adopted this second course of action. In accordance with the statutory procedure and timescale, it adopted a decision not to conduct a MIR. This was upon the basis that, whilst the CMA entertained concerns about the conduct of Apple and Google in the markets that it was investigating, it confidently anticipated that the Government would, shortly, introduce new legislation in Parliament which would, in the view of the CMA, confer upon it a different and more sophisticated and suitable regime of regulation of the tech-industry. The CMA indicated that if there were delays to the introduction of this legislation it reserved the right to revisit its concerns and, if appropriate, make a MIR.

In this regard the UK Government was intending to introduce legislation in order to march broadly in line with a seminal piece of legislation adopted at the EU level, namely the Digital Markets Act (“DMA”): Regulation (EU) 2022/1925 of the European Parliament and of the Council (14^th September 2022) on contestable and fair markets in the digital sector (OJL265/1). Under the DMA the European Commission is empowered to adopt decisions identifying certain operators in the digital sector as “gatekeepers”. These are large digital platforms providing so called core platform services, such as online search engines, app stores, messenger services, etc. Gatekeepers, so designated will be required to comply with a series of obligations and prohibitions listed in the DMA. The legislation amounts to a paradigm shift in the approach to regulation conferring an ex ante, before the event, jurisdiction to prevent harm arising at all; in contrast with the largely ex post, after the event, jurisdictions which presently exist whereby regulators must await the emergence of harm before acting. As of the date of the appeal hearing, the DMA was on the statute book but was in the transitional period prior to its coming into full force and effect. In anticipation of domestic legislation, the CMA established a Digital Markets Unit (“DMU”) to begin work preparing for the new domestic regime but it only becomes operational upon the coming into force of the new law. Notwithstanding a clear statement of intent upon the part of the Government, the anticipated UK equivalent legislation did not emerge. In fact the proposed law (the Digital Markets, Competition and Consumers Bill) was included in the King’s Speech on 7^th November 2023 and is not therefore yet in force though might be enacted during the present 4^th session of Parliament.

In the circumstances, the CMA decided not to proceed with a reference. It set this out in a decision dated 14^th December 2021 which was 6 months after the issuance of the MSN. However, upon the expiry of the 12 month period from the date of the MSN, when it had become clear that the new legislation was not imminent, the CMA made a proposal to exercise the broader standalone power under section 131(1) EA and commence a new procedure leading to a possible MIR.

This culminated in a decision of 22^nd November 2022 to initiate a MIR. This is the decision under challenge in these proceedings (“the Decision”). Apple, the respondent, contends that the regulatory regime set out in sections 131A and 131B of the EA 2002 permits the CMA to take one, and only one, decision on whether to conduct a MIR in relation to a particular matter unless the CMA issues a further MSN. Having decided not to initiate such an investigation, and absent the issue of a further MSN, the CMA had no enduring jurisdiction or power to take another. Apple therefore challenged the Decision as ultra vires the powers of the CMA. The CAT agreed.

If this is a correct interpretation of the EA 2002 it could have serious consequences. It would mean that the CMA had no jurisdiction, even some years later, to investigate concerns into the behaviour of an undertaking such as Apple or Google, even if such concerns were objectively justified. The CMA would be unable to impose a remedy designed to protect consumers in the public interest. Appreciating this, the CAT, having endorsed a strict and preclusive interpretation of the EA 2002, sought to mitigate the rigour of its construction by postulating, though without making definitive rulings, that there could be exceptions to the prohibition upon subsequent and fresh regulatory intervention. Nonetheless, the CAT still indicated, though again without so finding, that the sorts of justifications advanced by the CMA for conducting a fresh investigation in this case were contrary to general principles of public law. The net effect, in practice, of the judgment of the CAT is that there is very little scope, if any, for the CMA to initiate any further investigation into the conduct of Apple or Google, howsoever deep or justifiable the CMA’s concerns about their conduct.

The issue arising upon this appeal is narrow. It concerns the interpretation of EA 2002 and the circumstances in which the CMA might, lawfully, conduct a fresh market investigation having already come to a prior conclusion that it would not investigate. It is common ground between the parties that the central issue arising is one of vires i.e. the jurisdiction of the CMA to act. The respondent, Apple, was clear that this was the only issue it wished to address. In response to a question from the Court, Mr Otty KC confirmed that he was not seeking to either address or justify the analysis of public law set out in the CAT judgment. For its part, the CMA also accepts that the central issue is one of vires. It challenges the CAT’s analysis of the EA 2002 in its totality. However, it also challenges the CAT’s public law analysis, not least because it is concerned that there could be a lingering and adverse effect upon policy if the Court of Appeal did not correct the reasoning in the judgment.

The central question on the appeal concerns the interpretation of the EA 2002 and the jurisdiction or vires of the CMA to conduct a fresh investigation subsequent to an earlier decision not to conduct an investigation. It will though be necessary also to address, in some measure, the public law analysis of the CAT.

The CMA has a statutory function under section 5 EA 2002: “... of obtaining, compiling and keeping under review information about matters relating to the carrying out of its functions”. Section 5 provides:

“Acquisition of information etc.

(1)

The CMA has the function of obtaining, compiling and keeping under review information about matters relating to the carrying out of its functions.

(2)

That function is to be carried out with a view to (among other things) ensuring that the CMA has sufficient information to take informed decisions and to carry out its other functions effectively.

(3)

In carrying out that function the CMA may carry out, commission or support (financially or otherwise) research."

The CMA has additional powers pursuant to section 174 EA 2002, which it may exercise for a “permitted purpose” which includes requiring a person to attend to give evidence, produce documents, and/or provide estimates, forecasts returns or other information. Enforcement powers are provided for in section 174A. According to Market Guidance published by the CMA (Footnote: ¹), a MSN may lead to a variety of outcomes, including: (i) no action; (ii) actions improving the quality and accessibility of information to consumers; (iii) encouragement to businesses in the market to self- regulate; (iv) making recommendations to the Government to change regulations or public policy; (v) taking competition or consumer enforcement action; and (vi), making a MIR or accepting one or more undertakings in lieu of such a reference. A MSN is not therefore necessarily a precursor to a MIR, though it might be.

Section 131 confers a power (cf “may”) to make a MIR where three express conditions (one positive and two negative) are met. These are: (i) that the CMA has “reasonable grounds for suspecting” a competition concern (the positive condition - section 131(1)); and (ii), that neither of the two negative conditions in section 131(4) exist. There is a further condition which flows from the existence of the power in section 131(1). This arises out of the normal common law constraints imposed upon the exercise of any statutory power, such as the duty to act rationally and in accordance with the proper purpose for which the power was conferred by Parliament in the first place.

Section 131 EA 2002 provides as follows:

“(1)

The CMA may, subject to subsection (4), make a reference to its chair for the constitution of a group under Schedule 4 to the Enterprise and Regulatory Reform Act 2013 if the CMA has reasonable grounds for suspecting that any feature, or combination of features, of a market in the United Kingdom for goods or services prevents, restricts or distorts competition in connection with the supply or acquisition of any goods or services in the United Kingdom or a part of the United Kingdom.

(2)

For the purposes of this Part any reference to a feature of a market in the United Kingdom for goods or services shall be construed as a reference to –

(a)

the structure of the market concerned or any aspect of that structure;

(b)

any conduct (whether or not in the market concerned) of one or more than one person who supplies or acquires goods or services in the market concerned; or

(c)

any conduct relating to the market concerned of customers of any person who supplies or acquires goods or services.

(2A) In a case where the feature or each of the features concerned falls within subsection (2)(b) or (c), a reference under subsection (1) may be made in relation to more than one market in the United Kingdom for goods or services.

(3)

In subsection (2) “conduct” includes any failure to act (whether or not intentional) and any other unintentional conduct.

(4)

No reference shall be made under this section if –

(a)

the making of the reference is prevented by section 156(A1) or (1); or

(b)

a reference has been made under section 132 or 140A(6) in relation to the same matter but has not been finally determined.

(5)

References in this Part to a market investigation reference being finally determined shall be construed in accordance with section 183(3) to (6).

(6)

In this Part –

“cross-market reference” means a reference under this section which falls within subsection (2A) or a reference under section 132 which falls within subsection (3A) of that section (and see section 140A);

“market in the United Kingdom” includes –

(a)

so far as it operates in the United Kingdom or a part of the United Kingdom, any market which operates there and in another country or territory or in a part of another country or territory; and

(b)

any market which operates only in a part of the United Kingdom;

“market investigation reference” means a reference under this section or section 132 or 140A(6);

“ordinary reference” means a reference under this section or section 132 which is not a cross-market reference (and see section 140A);

and references to a market for goods or services include references to a market for goods and services.”

In the present case the CMA exercised the section 5 powers and published a MSN. This was in accordance with the powers in sections 130A, 131A and 131B EA 2002.

Section 130A provides:

“Duty to publish market study notice

(1)

Where the CMA is proposing to carry out its functions under section 5 in relation to a matter for the purposes mentioned in subsection (2), the CMA must publish a notice under this section (referred to in this Part as a “market study notice”).

(2)

The purposes are—

(a)

to consider the extent to which a matter in relation to the acquisition or supply of goods or services of one or more than one description in the United Kingdom has or may have effects adverse to the interests of consumers; and

(b)

to assess the extent to which steps can and should be taken to remedy, mitigate or prevent any such adverse effects.

(3)

A market study notice shall, in particular, specify— (a) the matter in relation to which the CMA is proposing to carry out its functions under section 5; (b) the period during which representations may be made to the CMA in relation to the matter; and (c) the dates by which the CMA is required to comply with the requirements imposed on it by sections 131A and 131B.”

Section 131A(1) starts by describing a set of circumstances, namely where the CMA has published a MSN and proposes to make a reference “in relation to the matter specified in the notice”. Where these circumstances exist there is a consequential duty: (i) to publish notice of the proposal concerned; and (ii), to consult relevant persons about the proposal. It provides:

“131A Decisions about references under section 131: consultation

(1)

This section applies to a case where the CMA has published a market study notice and—

(a)

the CMA is proposing to make a reference under section 131 in relation to the matter specified in the notice; or

(b)

a representation has been made to the CMA within the period specified in the notice under section 130A(3)(b) to the effect that such a reference should be made but the CMA is proposing not to make such a reference.

(2)

The CMA shall—

(a)

publish notice of the proposal concerned; and

(b)

consult the relevant persons about the proposal, in such manner as it considers practicable, before deciding whether to make a reference.

(3)

The CMA may, for the purposes of subsection (1), ignore any representation which it considers to be frivolous or vexatious.

(4)

For the purposes of subsection (2), a person is a “relevant person” if the CMA considers that its decision whether to make a reference is likely to have a substantial impact on the person's interests.

(5)

In consulting a person for the purposes of this section, the CMA shall, so far as practicable, give its reasons for the proposal.

(6)

In considering what is practicable for the purposes of this section, the CMA shall, in particular, have regard to—

(a)

the restrictions imposed by the time-table for making the decision (see section 131B); and

(b)

any need to keep what is proposed, or the reasons for it, confidential.”

Section 131B concerns time limits for the publication of the notice and commencement of the consultation provided for in section 131A. It also sets out (sub-sections (2) and (3)) mandatory time periods within which the final decision of the CMA must be taken. A decision of the CMA not to make a MIR must be taken within 6 months and a decision to make a MIR must be taken within 12 months, both time periods starting from the date on which the CMA publishes the MSN:

“Section 131B “Market studies and the making of decisions to refer: time-limits

(1)

Where the CMA has published a market study notice in a case to which section 131A applies, the CMA shall, within the period of 6 months beginning with the date on which it publishes the notice—

(a)

publish the notice under section 131A(2)(a); and

(b)

begin the process of consultation under section 131A(2)(b) (but the CMA need not complete the process within that period).

(2)

Subsection (3) applies where—

(a)

the CMA has published a market study notice;

(b)

no representation has been made to the CMA within the period specified in the notice under section 130A(3)(b) to the effect that a reference under section 131 should be made in relation to the matter specified in the notice; and

(c)

the CMA has decided not to make such a reference.

(3)

The CMA shall, within the period of 6 months beginning with the date on which it publishes the market study notice, publish notice of the decision not to make a reference.

(4)

Where the CMA has published a market study notice it shall, within the period of 12 months beginning with the date on which it publishes the notice, prepare and publish a report (referred to in this Part as a “market study report”) which sets out—

(a)

the findings of the CMA in relation to the matter specified in the notice;

and

(b)

the action (if any) which the CMA proposes to take in relation to the matter.

(5)

In a case to which section 131A applies, the market study report shall, in particular, contain—

(a)

the decision of the CMA to make a reference under section 131 in relation to the matter specified in the market study notice, the decision to accept an undertaking under section 154 instead of making such a reference or (as the case may be) the decision otherwise not to make such a reference;

(b)

the CMA's reasons for the decision; and

(c)

such information as the CMA considers appropriate for facilitating a proper understanding of its reasons for the decision.

(6)

Where a market study report contains a decision of the CMA to make a reference under section 131 in relation to a matter, the CMA shall, at the same time as it publishes the report, make the reference.

(7)

This section is subject to section 140A (duty of Secretary of State to refer in public interest intervention cases).”

On 15 June 2021 the CMA published a MSN entitled “Mobile Ecosystems”. The CMA indicated that it proposed to exercise its functions under section 5 EA 2002 in relation to the supply of mobile ecosystems in the United Kingdom and to consider the extent to which any matter in relation to the supply of such services had or could have effects adverse to the interests of consumers and to assess the extent to which steps could and should be taken to remedy, mitigate or prevent such adverse effects. The expression “mobile ecosystems” referred to the supply of smartphones and tablets, and associated software such as operating systems, app stores, browsers and applications. The expression “browsers” referred to all web browsers and associated browser engines accessible via mobile or desktop devices. The CMA elaborated in a document annexed to the MSN entitled “Statement of Scope”.

The CMA invited any person wishing to make representations to do so in writing no later than 26 July 2021. The CMA confirmed that representations could include whether the CMA should make a MIR pursuant to section 131 EA 2002. The MSN went on to describe the procedure that it was required to comply with, including as to time limits.

The consultation was conducted pursuant to the rules set out in section 131A EA 2002. Where the CMA has published a MSN and where no representations are made to the CMA within the specified period to the effect that a MIR should be made, and where the CMA decides not make such a reference then it is required to publish notice of the decision not to make a reference, within a period of six months beginning with the date upon which it published the MSN. In the present case, the CMA received no representations to the effect that a reference under section 131 should be made. The CMA concluded however that it should await the coming into legislative effect of new regulatory powers, and, accordingly, on 14 December 2021, within the six month statutory period, it issued a decision entitled “Mobile Ecosystems: Notice of Decision not to make a Market Reference under Section 131 of the Enterprise Act 2002”. Paragraphs [1] – [4] of the decision summarised the procedure that had been adopted. The reasons behind the decision are not set out. The CMA explained that it had published a further document, the “Interim Report”, in which a full analysis of its competition concerns and the reasons for the decision were set out. In paragraph [8] the CMA stated:

“The decision not to make a market investigation reference should not in any way be interpreted as the CMA finding no concerns in the sector, only that it considers that any potential concerns would not be best addressed through a market investigation at this time. The CMA has published today alongside this notice its Interim Report, in which it has set out its preliminary views on potential measures that may be required to address certain concerns it has identified. It has invited submissions on those views.”

For obvious reasons, since the CMA decided not to make a MIR, there was no challenge to the decision by Apple or Google.

As contemplated in paragraph [8] of the decision on that same day, 14 December 2021, the CMA published an Interim Report entitled “Mobile ecosystems: Market study interim report” (“the Interim Report”). This set out the CMA’s analysis and invited submissions. This detailed study comprises nearly 450 pages of analysis, excluding appendices. It explained that the CMA had structured analytical work under four so-called “themes”: (1) competition in the supply of mobile devices and operating systems; (2) competition in the distribution of mobile apps; (3) competition in the supply of mobile browsers and browser engines; and (4), the role of Apple and Google in competition between app developers.

Initial findings were set out in relation to each theme. It is not relevant to this appeal to go into the detail. A flavour can be obtained from the Executive Summary. In paragraph [5] the CMA observed that the operating system on a mobile device determined and controlled a range of features that were important to users, ranging from the appearance of the user interface, through to the speed, technical performance and security of the device. They could also determine what kinds of software (applications) could run on top. The CMA observed: “As suppliers of the two key mobile operating systems in the UK, Apple and Google are able to make a number of key decisions that can have significant implications for the products and services that are accessed online”.

In paragraph [6] the CMA observed that Apple and Google also controlled the key gateways through which users accessed content on mobile devices and through which content providers accessed potential customers. For example, Apple’s App Store was the only permitted app store on iOS devices and Google operated the Play Store which was used for the discovery and download of over 90% of all native apps on Android devices. The CMA stated: “Apple and Google are in a position to determine which apps are allowed in their store, how apps are ranked and discovered, and also often charge significant levels of commission (up to 30%) on app developers’ revenues from in-app transactions, by requiring these transactions to be made through their own in-app payment systems. At the same time, Apple and Google also offer their own “first-party” apps to users”.

In paragraph [9] the CMA recognised that the control of Apple and Google over their respective ecosystems could generate positive outcomes for consumers. For example, products could work seamlessly together and were easy and convenient for users which were traits valued by consumers. It was also recognised that Apple, Google and others had acted in an innovative manner improving the features themselves, their functionality and performance, making devices quicker, more powerful and effective. Further, the stewardship of Apple and Google of their ecosystems had engendered consumer confidence and trust which had facilitated market entry for small start-ups and otherwise unknown brands. Revenues generated had been reinvested in other valued services.

In paragraph [10] the CMA observed that, notwithstanding these acknowledged benefits, the level of control exerted by Apple and Google could create barriers to entry for other ecosystems: “Further, because Apple and Google control the way that browsers perform on their devices; and also set the terms for access to their app stores for native apps, they are able to limit competition from third parties in various ways within their ecosystems”. At paragraph [11] the CMA summarised its concerns about adverse consumer effects:

“Weak competition within and between Apple’s and Google’s mobile ecosystems can affect consumers in the following ways:

• Innovation: barriers to competition (particularly from third parties) risk holding back innovation in digital markets. For example, certain types of service may not be available to users (such as cloud gaming services on iOS devices), or certain developments in technology may be held up where Apple or Google do not have a clear incentive to promote these (such as web apps on iOS devices). Further, third parties investing in innovative products such as apps, services or connected devices which could complement the existing ecosystems may be discouraged from doing so, for example due to a fear of their data being used in order to further the development of Apple’s and Google’s own apps. Consumers may also lose out indirectly where, for example, the way that app stores are designed (including the ranking of apps) or terms imposed on app developers by Apple and Google, such as high rates of commission, have an impact on which apps succeed.

• The user experience: although overall satisfaction with smartphones is high there may be some ways in which users are not making informed and effective choices within mobile ecosystems. For example, the preinstallation of certain apps or setting certain apps as the ‘default’ can have significant impacts on user behaviour and give an advantage to Apple’s and Google’s own apps. The design of app stores and in particular the way in which search results are ranked can have a significant impact on which apps succeed.

• Privacy, security, and safety online: through design choice or other policies, Apple and Google are often in the position of acting in a quasi-regulatory capacity in relation to users’ security, privacy, and online safety. In many cases they opt to make decisions on behalf of consumers. However, it is not always clear if these numerous choices – ranging from restrictions on browser functionality to policies that affect targeted advertising – are in all cases made fully in the interests of consumers. For example, in many cases it seems decisions made on the grounds of protecting users’ security and privacy would also serve to give an advantage to first-party apps, or otherwise limit consumer choice.

• Prices: both Apple and Google are consistently making substantial profits with high margins, meaning that their prices go well beyond recovering the costs of providing these goods and services. In particular, Apple’s device sales, as well as for app distribution and search advertising revenue for both firms, are all highly profitable. We can infer from this that the prices charged for Apple’s devices, Google’s search advertising fees and each firms’ app store commissions, are likely to be above a competitive rate in each case. These high prices will in most cases ultimately be borne, directly or indirectly, by consumers.”

In paragraphs [88] – [97] the CMA considered the implications of statements made by Government that it would introduce legislation broadly equivalent to that contained within the EU DMA. The conferral of an ex ante jurisdiction could be an effective means of implementing interventions that the CMA was considering. The codes being contemplated would enable the CMA to impose upon the undertaking concerned obligations such as a requirement: to trade on fair and reasonable contractual terms; not to apply unduly discriminatory terms, conditions or policies to certain customers; not unreasonably to restrict how customers could use platform services; not to influence competitive processes or outcomes in a manner that unduly provided self-preference to a platform’s own service or to a service from which the platform derived a commercial benefit over rival services, including through use of preferential access to data; not to bundle or tie the provision of products or services in markets where the platform had market power with other services in the way which exerted an adverse effect on users; not unreasonably to restrict interoperability with third party technologies where this could exert an adverse effect upon users; to hold own apps/services accountable to the same privacy standards as imposed upon third parties and, not unreasonably to restrict APIs or hardware in a manner which exerted an adverse effect upon users.

The CMA noted that under the proposed legislation it would acquire powers to impose pro-competitive interventions (“PCIs”) to open up competition to SMS platforms. PCIs would be targeted at the sources of market power and would work in parallel with codes of conduct:

“91.

In summary, our initial view is that if the Government implements the framework broadly as currently envisaged, the framework for codes of conduct and PCIs envisaged in its consultation could be effective in addressing the types of concerns associated with exploitation of market power in the markets within the scope of this study, in addition to reducing market power for particular activities over time.”

In paragraphs [92] – [97] the CMA set out its reasoning for its decision not to make a MIR. In paragraph [93] the CMA concluded that, based upon its initial findings, there were reasonable grounds for suspecting that features of the market could be restricting or distorting competition in the UK. In paragraph [94] the CMA recorded that it had a discretion whether or not to make a MIR and that one of the factors to be taken into account was whether a MIR was the most appropriate mechanism for assessing the issues arising and delivering the required outcomes. In paragraph [95] it recorded that a further matter it was required to take into account was whether it had received any requests for a MIR to be made in response to a statement of scope or in the course of subsequent engagement with stakeholders. The CMA stated that it had received no such requests. Paragraphs [96] and [97] state as follows:

“96.

Our current assessment is that the DMU – through a combination of the anticipated enforceable codes of conduct and pro-competitive interventions – will in principle be best placed to tackle the competition concerns identified by this market study to date. In particular, this is because the interconnected nature of the activities carried out by Apple and Google within their ecosystems is likely to necessitate a package of interventions aimed at assessing potential harms to competition from a number of different angles, which in some cases potentially requires iterative design, testing, and trialling.

97.

On this basis, the CMA has decided not to make a market investigation reference. Notwithstanding this, the CMA will continue to keep under review the potential use of all its available tools during and following the second half of the market study, taking into account any relevant market or legislative developments that may arise. This includes the possibility of making a market investigation reference at a later point in time or taking enforcement action under our competition or consumer powers.” (original emphasis)

On 10 June 2022, the CMA launched a new consultation on a proposed MIR. This is a relatively short document of 2 pages. It repeats, concisely, the reasons which led to its decision not to refer. In paragraph [1.8] the CMA records that since it published the Interim Report it had received submissions from parties with a broad range of interests, including from browser vendors urging the CMA “…to take action now”. The CMA had also received responses from individual web developers and small businesses detailing concerns regarding mobile browsers and mobile browser engines which included specific examples of burdens and restrictions they confronted which they attributed to the conduct of Apple and Google. They explained how this affected the quality of their current and future goods and services offered to mobile users. The CMA referred to further concerns from a wider set of cloud gaming providers about the adverse impact of restrictions imposed by Apple upon their businesses. The CMA stated that they had, independently, undertaken additional analysis which gave it “increased confidence that interventions to remove certain restrictions – in particular those relating to mobile browsers and cloud gaming services – could be implemented without compromising safety, security, or privacy over people’s data”.

The CMA also stated this:

“The Government has now published its response to the consultation on the new regime which confirmed that it intends to bring in legislation for the DMU, but we now understand this will not be in the current Parliamentary session (i.e. within the next year).”

In view of the lack of imminence of new legislative powers, the CMA concluded that now was the right time to take targeted action by way of a MIR.

In paragraph [2.0] the CMA explained its jurisdiction or vires to act. It recorded that it had jurisdiction to make a MIR when it had reasonable grounds for suspecting that a feature or a combination of features of a market or markets in the United Kingdom prevented restricted or distorted competition and a MIR appeared to be an appropriate response. In paragraph [2.3] the CMA stated that when the reference test was met, the CMA then acquired a discretion to make a MIR. The CMA identified four criteria relevant to the exercise of that discretion. These were as follows:

“(a)

The scale of the suspected problem, in terms of its adverse effect on competition, is such that a reference would be an appropriate response.

(b)

There is a reasonable chance that appropriate remedies would be available.

(c)

It would not be more appropriate to address the concerns through undertakings in lieu of a reference (UILs).

(d)

It would not be more appropriate to address the competition problems through alternative powers available to the CMA or through the powers of sectoral regulators.”

The CMA also took into consideration that a MIR led to significant costs to the CMA itself and the public purse but also to the parties involved.

On 22 November 2022, the CMA published the Decision. The document is concise covering 21 pages. It explains the procedural background. In paragraphs [2.4] – [2.9] the CMA explained that it had received 43 written responses to the consultation; 37 were supportive of a reference either as proposed by the CMA or with a broader scope. Apple and Google did not support any further investigation. In Section [3] the CMA summarised its competition concerns in order to establish that it had jurisdiction to make a MIR and then it set out criteria as to the appropriateness of making a MIR. It considered: the scale of the suspected problem; the availability of appropriate remedies through a MIR; the availability of undertakings in lieu of a reference; the availability of alternative powers available to the CMA or to sectorial regulators; and, whether traditional competition law prohibitions and consumer law powers were more appropriate. Having considered these matters the CMA concluded that it was “…best placed to take forward action in this area”, exercising powers under section 131.

Apple challenged the Decision by way of appeal.

As set out above, the CAT allowed the appeal. In paragraphs [43] – [45] it explained why the Decision was ultra vires. The essential reasoning was that whilst the Decision was purportedly taken under section 131 it was, in reality, no more than a continuation of the earlier procedure under sections 131A and 131B and as such it did not comply with the time limits in section 131B, which had expired. Having taken a negative decision the CMA could not start over again this time using the standalone power in section 131. The CAT’s detailed reasoning can be summarised as follows.

Section 131A applied because: the CMA had published a MSN in June 2021; the CMA was proposing to make a MIR under section 131(1) which was “clearly” the effect of the MSN; the proposal to make a MIR was “in relation to the matter specified in the notice”; and accordingly, section 131A(1) applied and the CMA was obliged to comply with the time limits laid down in section 131B. But it had not done so; the Decision unlawfully post-dated expiry of the section 131B time periods. In paragraph [44(3)] the CAT stated:

“The CMA has failed to comply with both of these time limits. It was not suggested by the CMA that if these time limits apply, they could in some way be extended or waived. The CMA accepted that there were hard-edged jurisdictional limits, and we agree.”

In short, the procedures initiated by the Decision purportedly under section 131(1) “were too late”. The CMA “had no jurisdiction”. The decision lacked the statutory prerequisites for a valid decision and was ultra vires the CMA’s powers.

In paragraphs [47] – [51] the CAT set out a “sense check”. It rejected the submission of the CMA that its analysis “…could make no sense and would, irresponsibly, thwart the public functions of the CMA…”.

The words “in relation to” in section 131A(1)(a) were “key words” (paragraph [51]). Where a proposed MIR was “in relation to the matter specified” in a MSN, then the procedure and time limits in sections 131A and 131B had to be adhered to. This analysis made good sense:

“50.

The constraints that we have described thus make good sense in terms of efficient use of the CMA’s resources; and in terms of the extent to which a matter under investigation (whether by way of a market study or by way of a market investigation) is considered only once and not time-after-time. Participants in a market – having been the subject of a market study and, if the CMA is so advised, of a market investigation – are entitled to be left alone.”

The CAT recognised, however, that applied rigidly this could lead to “most undesirable” results. In paragraph [54] the CAT stated:

“In circumstances where the CMA obviously considers – for reasons set out in the Market Study Interim Report and in the Market Study Final Report – a market investigation reference to be desirable this is a conclusion that raises an important question of competition enforcement. We consider it to be most undesirable for the CMA to be constrained without good reason from making a market investigation reference. Yet – because of the Earlier Decision – that would appear to be precisely what this Judgment achieves.”

The use of the phrase “in relation to”, as the lynchpin for determining when the CMA could commence a new investigation, gave rise to the problem that if, say, the CMA decided not to conduct an investigation in year one but, five years later, wished to conduct a new investigation, it was precluded from so doing if the new investigation remained “in relation to” the same matter. In paragraph [51] the CAT observed that the expression “in relation to the matter specified” constituted “wide words that would derive their significance, if not their meaning, from the factual context”. Given the breadth of this key phrase it necessarily followed that the block placed on future investigation could be substantial. The CAT then went on as follows:

“Clearly, where there the subject matter of the reference is a matter entirely different from an anterior market study notice, no constraint will arise at all. But we would go further than this: the effluxion of time, in and of itself, may cause the “matter” of a market study notice (either in terms of its subject matter or the conditions irrelevant to the CMA’s decision making) to be sufficiently different from the “matter” of a later proposed reference such that the proposed notice and subsequent reference are not “in relation to the matter specified” in a market study notice. Things change over time, and it is the CMA’s continuing duty to keep itself informed under section 5 of the Enterprise Act 2002. We would not want it to be said that simply because the CMA could not point to something that had materially changed, it could not rely on the effluxion of time to justify revisiting a market whether by way of a market study or self-standing market investigation reference.”

As applied to the present case two consequences flowed (Judgment paragraph [53]). First, the CMA could only make a new standalone reference under section 131 where it was not “in relation” to the matter specified in the June 2021 MSN, but the new MIR was clearly in relation to the old matters. Secondly, the CMA could make a new MIR under sections 131A and 131B where a fresh MSN was made but were the CMA to attempt this on the present facts it “…may well be challengeable on other public law grounds”.

Given this conundrum the CAT was anxious to “articulate the way out of this situation” (Judgment paragraph [55]). In paragraph [55(1)] the CAT suggested, albeit without expressing any definitive view, that mistake of fact, misrepresentation, and change of circumstance, could amount to circumstances which, if they existed, would mean that a new investigation was not “in relation to the matter specified” in the earlier proceedings. The CAT found, however, such circumstances did not arise on the facts of the present case. The CAT rejected the argument of the CMA that its concerns about the conduct of Apple or Google, or the fact that new consultation responses had come in, were relevant. Such matters did not entitle the CMA “to re-visit matters” (Judgment paragraph [57]). Further, the attempt by the CMA to reserve its position in the event that new statutory powers became available to it, could amount to an error in law and/or the taking into account of an immaterial consideration (Judgment paragraphs [59] – [60]).

In a nutshell, the CMA had driven itself up a blind alleyway. The CMA argues that the CAT erred in law.

The starting point is to consider the power the CMA purported to exercise in the Decision. This was the standalone power in section 131 EA 2002. As to this all the positive and negative conditions (see paragraph [12] above) are satisfied. First, the CMA must have “reasonable grounds for suspecting” that there is a competition concern. The details of the CMA’s concern are set out in the Decision and Interim Report and examples are given at paragraphs [22] – [29] above. There is no suggestion from the respondent, Apple, that this positive condition is not satisfied. Secondly, neither of the negative conditions in section 131(4) exists which would otherwise preclude a MIR. Again, this is common ground. Thirdly, there are no public law grounds raised which, even if the positive and negative conditions are satisfied, would nonetheless prevent the CMA from exercising the standalone power in section 131(1). There is no challenge, on public law grounds, to the exercise of discretion on the part of the CMA in this regard. For example, it is not said that Apple had any legitimate expectation that, having taken a decision not to make a MIR, the CMA would never in the future do so.

Prima facie, the CMA has vires to issue the Decision under section 131. The question that then arises is whether there is anything in sections 131A or 131B which, by inference or implication, curbs the otherwise lawful exercise of the power by the CMA under section 131(1). I will deal with the arguments raised.

First, there is the role played by the expression “…in relation to the matter specified in the notice…” in section 131A(1)(a). Apple argues that the CAT was correct to conclude that the expression curbed the exercise of the section 131(1) standalone power. It is said that these words amount to a broad limitation upon the ability of the CMA in any case where it has in the past published a MSN, from taking a decision outside of the statutory timetable in section 131B. I disagree. These words are found in section 131A which is under the heading “Decisions about references under section 131: consultation”. Here, Parliament ascribed to them a strictly limited task. They describe the nexus (“in relation to”) between a proposed reference and the matters specified in the notice and in practical terms they describe an evaluative exercise to be performed by the CMA which, by its nature, should be cursory and obvious. As the CAT correctly observed in its Judgment, the words do not describe a demanding threshold or hurdle to surpass. Where the CMA performs that exercise and concludes that the nexus exists, the effect is only that the duties in section 131A(2) are triggered. These involve (i) publishing notice of the proposal concerned and (ii) consultation of relevant persons about the proposal. The remainder of section 131A concerns the modalities of the consultation exercise. It is no part of the function of these words to create some additional, wide-ranging, restriction upon the exercise of the different standalone power in section 131(1). The phrase plays a part during the early stages of the sections 131A and 131B procedure. Its stated function comes to an end upon publication of the notice and commencement of the consultation as required by section 131A(2). Nothing in the statutory language accords those limited words any wider and more enduring significance and effect. With respect to the CAT it has taken this phrase out of its proper context.

Secondly, if Parliament had intended to use that phrase as an additional restriction or fetter upon the exercise of the standalone power in section 131(1) it would surely have said so. The logical, and in my view inevitable, place in which such a limitation would have been inserted would be in section 131(4). This provision was designed by Parliament to identify those cases whereby because of the “related” nature of the exercise of other procedural powers concerning the same subject matter, the CMA was precluded from exercising the standalone power. Section 131(4) (set out at paragraph [13] above) identifies two such cases. The first is where the Secretary of State has issued a MIR under section 156(A1) or (1). This will happen when the Secretary of State disagrees with the failure of the CMA to make a MIR so initiates an independent MIR. The second is where the CMA is in the process of negotiating statutory undertakings to compromise an existing MIR. There is no third category in section 131(4) where the CMA has, at an earlier point in time, issued a decision under section 131B following a section 5 EA 2002 inquiry and where the CMA now wishes to make a MIR in a “related” matter. Section 131(4) is the place where Parliament, having addressed its mind to those situations where, because of a related investigation, the CMA was to be prohibited from exercising the standalone power, would surely have placed this variant of that same scenario if it had intended there to be such a constraint. But it did not do so. Parliament might, one supposes, additionally or alternatively, have added a new subsection to section 131B, probably to be inserted after subsection (4), providing that the CMA was prohibited from issuing a new MIR in circumstances where the twelve month period set out in section 131B(4) had expired. If Parliament had intended the expiry of the sections 131A and 131B procedures to create a permanent bar to future action, it would have said so. But, again, it did not do so.

Thirdly, there is the material said to be admissible to show Parliamentary intent. In paragraphs [49] and [50] of its Judgment the CAT identified a number of policy reasons which, it concluded, made “perfect sense” and supported its analysis of the statutory language as creating an additional constraint. The CAT recorded that a MIR involved a great deal of work not only for the CMA but also for the undertaking under investigation to be performed in a “very short space of time”. Restrictions upon the CMA making a fresh MIR under section 131 were justified in terms of the efficient use of the CMA’s resources and in terms of the extent to which a matter under investigation would be considered “only once and not time-after-time”. Participants, having been made the subject of a MSN procedure, were entitled “to be left alone”. Mr Otty KC, for Apple, sought to buttress these policy considerations by reference to material admissible as a guide to construction of the EA 2002. He drew our attention to the Explanatory Notes which highlighted the importance of legal certainty and which acknowledged the intrusive nature of a MIR for the undertakings concerned. The difficulty with the analysis of both the CAT and the respondent is that they have described only part of the overall picture. The Explanatory Notes also refer to the importance of the statutory purpose of promoting competition and protecting consumers. The CMA also points out that pursuant to section 25(3) Enterprise and Regulatory Reform Act 2013, it has an overarching statutory duty to “…seek to promote competition…for the benefit of consumers”. When applying a purposive construction to the EA 2002, whilst protection of investigated undertakings from undue investigatory burdens is a relevant consideration, the principal purpose of the Act is to promote competition and protect consumers. Where, as here, there is no challenge to the conclusion of the CMA that it had reasonable grounds for suspecting the existence of a competition problem, then some level of burdensome regulatory intervention is an inevitability for the undertakings concerned. In my judgement, the CAT lost sight of this consideration. There is no overarching principle that an undertaking is entitled to be investigated once, and once only.

Finally, the CAT considered that public law played a role in constraining the exercise of power by the CMA. Public law does play a role through the statutory conferral of a discretion in section 131(1) via the “may”. The scope for public law intervention might be limited by the fact that Parliament identified positive and negative conditions which had to be satisfied before the discretion could be exercised. These might, otherwise, have been issues which could have sounded in a public law challenge. Nonetheless, it is accepted that there is a residual constraining role for public law to play. Apple has not complained that any of the conditions are breached nor suggested that there is any public law basis justifying complaint. The short point is that the existence of the statutory power in section 131 carries with it the important safeguard that its exercise is subject to public law control, such as rationality and the proper purpose principle. There is no need to introduce some additional fetter arising from the language of section 131A to ensure that the standalone power in section 131 is used properly.

In conclusion the Decision was intra vires to the CMA and hence lawful. I would allow the appeal of the CMA on this point.

I next address the public law points made by the CAT. The CMA challenges the analysis root and branch. Apple did not defend the CAT’s analysis. The CAT’s analysis and its desire to describe possible “get outs” from the straitjacket it had, otherwise, imposed, focus upon perceived wriggle room in the expression, “in relation to”. That analysis, however, is irrelevant to the public law constraints that might arise to limit the exercise of power under section 131(1).

As already observed where a statutory power exists, its exercise will always be subject to constraints imposed by public law. This will include the duty to exercise the power in accordance with its proper purpose; Padfield v Minister of Agriculture, Fisheries and Food [1968] AC 997. Mr Williams KC, for the CMA, expressly conceded that the power in 131(1) EA 2002 was subject to such constraints. In written submissions the CMA set out a number of examples where it said that, at least arguably, it would be exercising the power in an unlawful way, notwithstanding that otherwise it was entitled to exercise the standalone power. By way of illustration, Mr Williams acknowledged that had the CMA sought to exercise the power under section 131(1) purely and simply to circumvent, artificially, the time limits in section 131A and 131B, that it would be open to challenge on public law grounds. There is, I should add, no suggestion that this is the case on the facts.

There are two particular matters the CMA is concerned about which the CAT addresses under section 131A(2) but which might resonate under section 131(1). These are: (i) the relevance of the existence of possible alternative remedies; and (ii), the emergence of new evidence from third parties. I take each in turn.

As to the relevance of possible new legislation, this did not play a part in the reasoning of the Decision under challenge, although it is clearly part of the historical baggage leading up to that decision. The Decision, as drafted, is based squarely upon the existence of reasonable grounds for suspecting a competition concern.

Nonetheless, Mr Williams KC contended, because the matter was of broader public importance and might have future ramifications for the CMA, that it was within the statutory purpose behind the EA 2002 generally and section 131(1) specifically for the CMA to take account of such matters. He relied upon the judgment of the Divisional Court in R v Secretary of State for the Environment ex parte Birmingham City Council [1987] RVR 53. That case requires careful consideration. There, Birmingham City Council (“BCC”) sought an order of mandamus directing the Secretary of State to perform his statutory duty under section 60 Local Government Planning and Land Act 1980 by making a Rate Support Grant Report for the financial year 1987/88 and laying it before the House of Commons for its approval. It was common ground that the Secretary of State was under a statutory duty to make and lay a Rate Support Grant Report. The dispute concerned the point in time at which the report had to be laid. BCC contended that the Secretary of State was under an implied statutory duty to lay the report in sufficient time to enable local authorities to take it into account in preparing budgets. In the alternative, it was said that the Secretary of State had a power or discretion as to the point in time at which the report would be laid but had erred in delaying the laying of the report because he had taken account of irrelevant considerations or mis-directed himself in law. The complication arose because there were two views as to what was included in the term “rate fund”. When the system was first introduced in 1981/2 the consensus was that a narrow view was correct. However subsequently, in 1986, the Secretary of State was advised that a wide view was correct. Later that year the Secretary of State came to Parliament to explain the mistake and announce an intention on the part of the Government to introduce urgent clarificatory legislation. As of the date of the litigation, the legislation had not received Royal Assent. In these circumstances the Secretary of State deferred the laying of the Report. This was the basis upon which it was argued that the Secretary of State was bound to adhere to the law as it presently stood, and not by reference to a regime anticipated for the future. The Divisional Court considered that the submission was “contrary to the plainest common sense”. The Secretary of State found himself in a “extremely awkward position”. When the error was discovered, he had two alternatives, one was to start the consultation process anew, the other was to seek new legislation. The Court considered that “the advantage of the latter course must have been manifest”:

“In the first place it would enable the existing practice to continue, as desired by all, including, it would seem (though there is no evidence on this point) Birmingham City Council. Secondly, it would validate what had been done in the past. Thirdly, it would save a great deal of expense and confusion. Lastly, there must at least have been a chance that the Bill would become law before a report on the wide basis could be made. To say in those circumstances that the Secretary of State had to disregard the second alternative would be absurd. It would bring the law into needless disrepute.”

The Court drew a distinction between the power of the Secretary of State and the position of the courts. The latter must enforce the law “as it stands”. The courts had no choice. But the Secretary of State had a choice. He had to decide what to do and to disqualify him from considering the possibility of seeking a change in the law was “absurd”.

How does one apply this to the present case? The EA 2002 was not enacted with the prospect of ex ante, prophylactic, future regulation in mind. It creates a regime largely based upon, after the event, intervention. Further, the CMA, unlike the Secretary of State, is not in a position to introduce legislation into Parliament. Nonetheless, the CMA has a broad and overarching duty to promote competition and protect consumers. When deciding whether to act in a case where, prima facie, it has jurisdiction, it also looks to see whether it, as opposed to some other regulator exercising different power, is best placed to act. Further, upon the evidence it had strong and reasonable grounds for believing that the law would, imminently, change, not least so that regulators in the UK were not at a disadvantage with EU counterparts.

We did not hear full argument upon the point and it is not squarely before the Court on the appeal. Nonetheless, the CAT has suggested, though without deciding, that the taking into account of such a consideration is likely to be challengeable upon public law grounds. In my view, it is arguable that the CMA would not be without the statutory purpose under section 131 if it took such a matter into account. I would go no further than this. Ultimately, this must be an argument for another day, if it were ever to arise again, which might be unlikely.

Finally, as to the new evidence from affected persons, it is not said that the CMA erred in concluding that it had a sufficient evidential base for its concerns and that the power in section 131 did not prima facie arise. Nor is it said that the fact that during the earlier MSN process no consultee came forward to propose a MIR is relevant, given other evidence the CMA already held. So, the incremental effect of new consultees coming forward proposing a MIR is nothing to the point. Mr. Williams KC, for the CMA, speculated that the reason that there were no responses to the first consultation (see paragraph [29] above) was because consultees knew that the position of the CMA was that it was biding its time pending new legislation, so that there was no practical purpose to be served by submitting evidence at that juncture. He might be right. There is however no proof of this. Standing back, at a high level of abstraction, the fact that following a consultation a significant volume of adverse material is submitted by consultees must be a matter that is relevant to the exercise of the power.

In conclusion, I would allow the appeal. I would set aside the order of the CAT.

I have had more difficulty with this appeal than my brethren. The reason for my hesitation is as follows. Mr Williams KC accepted at one point during the course of his submissions in reply that, read literally, section 131A(1)(a) EA 2002 applied to the facts of this case: the CMA had published an MSN and it was proposing to make a MIR in relation to the matter specified in the MSN. Furthermore, he disavowed any contention that the matter in relation to which the CMA was proposing to make a MIR was different to the matter which had been the subject of the MSN. It is common ground that, if section 131A applies, then the time limits specified in section 131B apply. It is also common ground that, if the time limits specified in section 131B apply, then the Decision was out of time. This is, in a nutshell, why the CAT held that the Decision was ultra vires.

The CMA’s case is that section 131A(1)(a) should not be interpreted literally, but purposively. In short, the CMA says that section 131A(1)(a) cannot have been intended by Parliament to apply where the CMA is exercising the standalone power in section 131(1). In the end, I have been persuaded that, for the reasons given by Green LJ, the CMA is correct, and therefore the appeal should be allowed.

I agree that the appeal should be allowed for the reasons given by Green LJ.