Introduction

On May 24, 2024, the United Kingdom’s Competition and Markets Authority (CMA) issued a call for consultation regarding their draft guidance for the digital markets competition regime (draft guidance) following the royal assent of the new Digital Markets, Competition and Consumers Act (DMCC). The DMCC grants “new responsibilities to the CMA to promote competition in digital markets through the new, forward-looking digital markets competition regime.”[1] These new powers include a special designation of firms in one or more digital activity as holding “Strategic Market Status” (SMS), enforcing company specific conduct requirements (CRs), as well as the implementation of pro-competitive interventions (PCIs). The draft guidance provides needed clarity as to how the CMA will enforce the DMCC, including the criteria for applying SMS status to firms, the standards for imposing CRs, the demonstration of procompetitive benefits, the use of PCIs, and how the CMA will administer remedies.

The Information Technology and Innovation Foundation (ITIF) appreciates the opportunity to comment on the draft guidance, specifically to address concerns about how to minimize the DMCC’s negative effects on digital competition, innovation, and consumer welfare. ITIF is a nonprofit, nonpartisan research and educational institute that has been recognized repeatedly as the world’s leading think tank for science and technology policy. ITIF’s comment proceeds in five parts.

First, the CMA should avoid overdesignating firms as SMS, which would unnecessarily impose costs on digital businesses, and instead properly account for the dynamic and often drastic nature of competition in digital markets. Second, the CMA must prioritize tethering CRs to likely harm to competition rather than mere harm to competitors or exploitative abuses. Third, the countervailing benefits exemption (CBE) in the draft guidance should be applied using a disproportionality test to mitigate the problems associated with conduct that often involves tradeoffs between static harms and dynamic benefits. Fourth, under the PCI framework, the use of structural relief by the CMA should be clearly stated as an extraordinary remedy to be deployed only in rare circumstances and when behavioral remedies have failed. Fifth, ITIF recommends that the CMA revise its draft guidance to better allow competition and innovation to flourish within the new DMCC regime. A brief conclusion highlights the DMCC’s potential geopolitical implications and the commensurate need for appropriate enforcement discretion by the CMA.

SMS Designations and Leapfrog Competition

SMS designation is the first point of entry by a firm into the new digital markets competition regime and the threshold issue for determining whether the DMCC’s requirements will apply. To be designated as an SMS, the CMA will conduct an SMS investigation to assess, as elaborated upon in the draft guidance, whether four conditions are met: digital activity, jurisdiction and turnover, substantial and entrenched market power, and position of strategic significance. With respect to substantial and entrenched market power, the draft guidance explains that a forward-looking assessment of at least five years must be undertaken, paying attention to foreseeable activity if the CMA does not intervene, and developments that may affect the undertaking’s conduct in carrying out the digital activity.

The approach taken by the draft guidance to substantial and entrenched market power risks being insufficiently sensitive to the unique nature of competition in digital markets. Since the groundbreaking work of Joseph Schumpeter in the 1940s, it has long been understood that competition is not merely an equilibrium where price equals marginal cost, as neoclassical economics holds. Rather, as Schumpeter explained, innovation or dynamic competition occurs through “gales of creative destruction” whereby firms compete for the market by creating a new product, only to be challenged by additional “leapfrog competition” that supplants the formerly dominant firm with a still newer product that not just dazzles consumers but allows for the firm to recoup the costs of its innovation.[2]

As such, that a firm may have substantial market power over a five-year period does not mean that there is any substantial and entrenched market power. Rather, competition may already be occurring in another market as part of a dynamic process of leapfrog competition, or still impending in the form of a drastic Schumpeterian wave. For example, that Microsoft may have long enjoyed a high share in computer operating systems is not evidence that its digital competition has failed: Google and Apple both leapfrogged Microsoft with their respective search and mobile operating platforms that have brought tremendous benefits to consumers. And now, a drastic Schumpeterian gale is underway in the form of artificial intelligence, which is already challenging established digital giants like Google. At bottom, rather than narrowly focus on a particular market, SMS designations must consider market power in a broader context that accounts for the “leapfrog” and sometimes drastic nature of dynamic competition that defines digital markets.

Defining Legal Standards For Conduct Requirements

CRs, or restrictions on the business practices of digital firms, are the centerpiece of DMCC enforcement. As the guidance paper explains, these CRs will give effect to the DMCC’s objectives to promote fair dealing, open choices, trust, and transparency. Moreover, the guidance paper explains that “the CMA may only impose a CR which is of a permitted type,” which includes discrimination, self-preferencing, leveraging, tying, refusal to interoperate, anti-steering, and other behaviour.[3] In addition to explaining that a CR must be consistent with the stipulated objectives of the DMCC and fall within the permitted types, the draft guidance notes that the CMA will also “consider the proportionality of any CRs that it proposes to impose,” which involves consideration of the “potential effects” of the CR on the SMS firm, third parties, consumers, as well as the market more broadly.[4]

In so doing, the draft guidance fails to clearly articulate that CRs will only be applied to address behaviour that results in likely harm to competition, as opposed to, for example, potential (or likely) harm to competitors. For this reason, the DMCC risks being enforced in a way that brings about the harms that have already resulted from the European Union’s DMA, which expressly disclaims any inquiry into anticompetitive effects in lieu of per se rules of illegality untethered to consumer harm. For example, the DMA’s bans on self-preferencing with respect to Google’s design of its Maps product have not only harmed consumers through an interface that requires consumers to make more clicks to get the answers they seek, but also small businesses like hotels and restaurants who have seen their traffic go to Google’s competitors like TripAdvisor and Booking.com—a classic case of the unintended consequences of regulation resulting the picking of winners and losers rather than serving the public interest.  

As such, the guidance paper should make clear that CRs should only be imposed to address behaviour which is likely to harm competition, as distinct both from conduct that merely harms competitors, as well as exploitative conduct that may harm consumers but is unaccompanied by any harm to the competitive process, like unfair or excessive pricing. Indeed, doing so is crucial to avoiding two common problems with ex ante regulatory regimes like the DMCC. First, as indicated above, by allowing conduct that merely harms competitors to be proscribed, the CMA exacerbates concerns about regulatory capture—a concern which is particularly serious with regimes like the DMCC, which specify CRs on a company specific rather general marketwide basis. Furthermore, in merely condemning behaviour that results in “unfair” market outcomes as distinct from any harm to competition, the DMCC risks making manifest its own limited knowledge about optimal market outcomes, a problem which is also specifically acute in fast moving digital markets.

Unduly Restricting the Countervailing Benefits Exemption

As distinct from the burden of proof to show anticompetitive harm, the draft guidance also implicates the crucially important question concerning the standard that will be applied to demonstrate that conduct results in procompetitive benefits, and the CBE. As stipulated in the DMCC, the CBE only applies if a number of conditions are met, which include not just that “the conduct to which the investigation relates gives rise to benefits to users or potential users of the digital activity in respect of which the CR in question applies” but that the procompetitive benefits “outweigh any actual or likely detrimental impact on competition resulting from a breach of the CR.”[5]

The language in the draft guidance suggests that this CBE balancing will come at the expense of sound enforcement. For example, as many of the permitted types of CRs will very often entail concrete procompetitive justifications, attempting to weigh, as the guidance paper elsewhere suggests, the “effects of the CR(s) expected to be felt in the short term [] against others expected to arise in the future” will result in the CMA having to deal with the long recognized difficulties associated with calculating these types of static and dynamic tradeoffs.[6] Accordingly, U.S. courts rarely apply simple balancing tests to address the permitted types of behaviour, whether refusals to deal,[7] predatory innovation,[8] or even technological tying as in Microsoft,[9] which did not engage in any balancing analysis.[10]

As such, to the extent the CMA must insist on implementing some form of a balancing test for purposes of applying the CBE, it should at the very least utilize what has been termed the “disproportionality test,” whereby the anticompetitive effects of the conduct at issue must substantially outweigh any procompetitive benefits. In this way, the CMA can reduce the administrative costs associated with attempting to apply a strict balancing test to weigh short run static harms against long-run dynamic benefits. Moreover, a disproportionality test will also help the CMA economize on the error costs associated with false positives in the form of reduced innovation that harms UK markets and consumers by condemning behaviour that has dynamic benefits which far outweigh any short-run harms.

In addition to the demonstration of verifiable benefits balanced against competitive harms, procompetitive justifications are often subjected to a causation requirement—in other words, that the conduct is in some way responsible for the procompetitive benefits. This can include inquiries into whether the conduct is likely to produce the procompetitive benefits, as well as in some cases the extent to which the procompetitive benefits could have been achieved without the conduct—with the latter showing being expressly not required under the Microsoft standard for evaluating unilateral exclusionary conduct.[11] Unfortunately, not only does the draft guidance sanction the indispensability test of section 9(1)(b) of the Competition Act 1998 (which goes against the final Parliamentary position to reject indispensability) but appears to apply an additional necessary causation requirement in the form of a de facto least restrictive alternative test in its proportionality analysis, the combination of which is likely to chill procompetitive conduct by unduly limiting the ability for SMS firms to offer defenses. 

Structural Relief Should be Strongly Discouraged

The draft guidance also provides details on the DMCC’s regulatory regime concerning the remedies associated with PCIs, which seek to rectify an adverse effect on competition (AEC) found after a market investigation. Specifically, in addition to the already broad powers extended to the CMA, the draft guidance provides more clarity on “structural remedies” such as divestitures, which it explains can be used to “address competition problems by changing structural aspects of a sector and/or the lack of rivalry resulting from those aspects.”[12] The draft guidance also elaborates as to how structural and behavioural remedies can be used in tandem, and in particular how the “CMA may rely on a mix of behavioural and structural measures” in the course of a PCI.[13]

Unfortunately, the draft guidance fails to distinguish between the fundamentally different natures of behavioural and structural remedies, especially when remedying the unilateral exclusionary conduct encompassed by the permitted types of CRs. As has long been U.S. and international best practice, in contrast to behavioural remedies, which are used to remedy exclusionary conduct, structural relief, or the break-up of a company’s business, is typically only appropriate for remedying mergers which result in collusive behaviour that cannot be addressed through conduct provisions (i.e., the legality of tacit collusion). Only in extraordinarily rare cases are structural remedies applied to remedy exclusionary conduct—and only those which involve some type of leveraging between multiple markets, such as tying—and only where behavioural remedy is for some reason not sufficient to address the anticompetitive harms, a set of conditions which was notably absent even in cases like U.S. v. Microsoft case.[14]

Rather than treat behavioural and structural remedies as merely two different tools in the CMA’s remedial toolbox, the draft guidance should provide clear principles articulating the narrow conditions under which structural relief should be used in the context of a PCI. For example, and firstly, the draft guidance should confirm that behavioural remedies are the presumptive form of relief for PCIs, as well as that structural remedies will only be applied both under extraordinarily rare circumstances and if a behavioural remedy cannot reasonably remedy the competitive harm the PCI is meant to address. Second, the draft guidance should also, in the context of its proportionality assessment, confirm that structural relief is presumptively disproportionate given the costs it imposes on the target SMS, and should only be invoked in cases where the harm from the competitive behaviour is vastly disproportionate to both the benefits and the costs to the SMS of divestiture.

Recommendations

In view of the above, ITIF offers the following recommendations to the CMA as it considers amending its draft guidance:

1. Avoid overdesignating firms as having Strategic Market Status: The CMA should avoid overdesignating digital firms as SMS and creating regulatory barriers to entry that benefit existing incumbents, but instead take into account the leapfrog nature of Schumpeterian competition that defines many digital markets.

2. Tether CRs to likely harm to competition and consumers: To mitigate issues associated with regulatory capture and incomplete information, the CMA should only condemn behaviour that results in likely harms to competition, as opposed to simply harms to competitors or even exploitative harms to consumers untethered to any harm to the competitive process.

3. Avoid overly restrictive application of the countervailing benefits exemption: If the CMA wishes to promote a workable digital regime, it must eschew a strict balancing analysis that risks creating administrative difficulties and false positives when evaluating the merits of business conduct by SMS firms that may have both static harms and dynamic benefits, and instead implement a more practical disproportionality test.

4. Structural relief should be used only in the most extraordinary circumstances:Given the substantial costs on businesses associated with structural relief, the CMA must make clear that structural relief is a presumptively disproportionate remedy that will only be applied in extraordinarily rare circumstances, and where behavioural relief proves wholly inadequate to address the CMA’s competitive concerns.

Conclusion

The UK’s DMCC regime comes at a time when flourishing digital markets are key for social and economic prosperity, not least given the ongoing improvements in artificial intelligence. It also comes at a time when the UK continues to define its future outside the European Union and amidst a rising China. In this challenging world, strong techno-economic cooperation between the United States and its closest allies like the UK is all the more critical, but can be undermined by the abuse of digital regulation that ultimately ends up having the primary effect of targeting and placing undue burdens largely on American firms—a reality that is already becoming apparent in Europe with the DMA. As the U.S. and UK look forward to the possibility of even deeper trade relations, the CMA must ensure that the application of the DMCC benefits UK innovation and consumers rather than kneecap America’s tech giants, who continue to contribute so much to its digital markets and broader economy.

Thank you for your consideration.

Endnotes