Gatekeepers are not exclusive to digital markets. Gatekeepers exist in many other industries such as banking, telecoms and energy. The new element introduced by digital gatekeepers is their scale and scope: There are digital gatekeepers that have a global scope, have millions of users and are present in several industries. The potential anticompetitive conducts of those digital gatekeepers might have a large impact over consumers. Banks have traditionally been the main gatekeepers and facilitators of the overall financial industry. The banking industry provides an illustration of an industry where regulation has been imposed to reduce the market power of gatekeepers. The analysis of such regulations shows that regulatory remedies necessarily have to be tailored to the relevant competition problems, to the characteristics of the industry and of the business models of the market participants. A too general regulation runs the risk of not being operational and creating uncertainties amongst market players.
By Juan Delgado1
I. INTRODUCTION
Digital gatekeepers can be defined as entities that “have a high degree of control and influence over the relationship between buyers and sellers, or over access by advertisers to potential buyers” which gives these entities “three distinct forms of power: the ability to control access and charge high fees; the ability to manipulate rankings or prominence; and the ability to control reputations.”2 Gatekeepers are not exclusive to digital markets. Gatekeepers exist in many other industries such as banking, telecoms and energy. The new element introduced by digital gatekeepers is their scale and scope: There are digital gatekeepers that have a global scope, have millions of users and are present in several industries. The potential anticompetitive conducts of those digital gatekeepers might have a larger impact over consumers.
During the last years, there has been an increasing debate on whether those gatekeepers should be regulated to prevent market foreclosing. Several competition agencies and governments have commissioned reports analyzing the role and the design of competition law and regulation in the digital era.3 The European Commission recently issued the proposed Digital Markets Act (“DMA”) that aims to establish rules to ensure fair and open digital markets. The DMA sets a list of general obligations to digital gatekeepers and an additional list of further potential obligations susceptible of being imposed on digital gatekeepers.
Digital gatekeepers differ on their nature, the industry they operate and the way they may exercise their gatekeeper role which is linked to their business model.
The regulation applied to gatekeepers in other industries provides an illustration of how regulation can facilitate market entry. The analysis of such industries also shows that regulations are very closely linked to the competition concerns, to the industry where they are implemented, to the competitive dynamics of the industry and to the business model of the gatekeepers. Remedies adopted in those industries are tailored to the competitive concerns detected and to the way markets operate. Thus, general principles and obligations can be a starting point for developing further and more detailed regulations but are no substitute for a detailed evidence-based analysis of the existing competitive concerns.
The banking industry provides an illustration of an industry where regulation has been imposed to reduce the market power of gatekeepers. Banks have traditionally been the main gatekeepers and facilitators of the overall financial industry. Their close customer relationship has traditionally made them the main channel to commercialize financial products. Banks have taken advantage of this fact providing bundles of their own products (checking accounts, investment products, pension plans, insurance, loans and mortgages, etc.). Competition in the banking industry has traditionally mostly developed through competing bundles offered by different banks.
The emergence of fintechs, that commercialize stand-alone products, and the increasing concentration in the banking industry after the economic and financial crisis in 2008 have taken regulators to adopt a number of measures aiming to facilitate the entry of alternative players (in addition to other measures aimed to increase transparency and financial stability which are not the object of this article). This article provides some illustrations of banking regulations that aim to reduce gatekeepers’ market power and promote market entry. The examples do not aim to be comprehensive but to illustrate how existing regulation is being applied and whether lessons can be learnt for the design of regulation in digital markets.
The examples focus on three aspects that are also present in digital markets and addressed by the DMA: the use of data, interoperability and self-preferencing.
II. THE USE OF DATA: OPEN BANKING REGULATIONS
Proprietary data can generate market power and erect barriers to entry. Banks collect information on customers’ financial behavior (revenues, expenses, credit worthiness, investment profile, etc.) which is not available to other competitors. This creates a lock-in effect in customers that reduces competition. Customers’ data provide banks market power and allow them to extract rents from locked-in clients.
Open banking initiatives aim to reduce customer switching costs by allowing customers to instruct their bank to share their account and transaction information with accredited third parties. Open banking regulations are in place in the UK (since 2017), Australia (since 2020), and the European Union (after the adoption of the Payment Services Directive PS2).
The 2018 Open Banking regulations in the UK4 required big banks to permit bank account holders to share information with authorized third parties through common and open standards. Access is made possible through application programming interfaces or APIs, which allow software at one company to have access to information from another. In order to access the data, companies are required to have authorization from the UK regulator as Account Information Service Providers (“AISPs”) and have to prove compliance with regulations including keeping data secure and only using access to provide their stated service.
The data sharing obligations are asymmetric and initially only apply to the nine biggest banks and building societies. The regulation allows access to a limited type of data (account and transaction data), requires the consent of each customers (data is only exchanged for those customers giving consent) and data access is only open to authorized financial service providers (around 135 service providers in 2019).
Open banking regulations reduce switching costs since competing banks can access new clients’ financial history (and can offer loans and other products tailored to the specific needs of each customer), facilitate multihoming (for example, many operators, including incumbents, offer account aggregators which let their customers view their accounts in several banks, making it easier to manage and control their finances) and promote innovation (e.g. it has allowed the development of new products, especially in the payments sector).
The competition concern in this case was created by the collection of data by incumbents that granted them a competitive advantage that could imply a barrier to entry for new competitors and a switching cost for consumers. The regulatory remedy involves an obligation to share such data. The regulatory remedy specifies both the data that has to be shared (and the access format), and which entities can access to the data and for what purposes. The sharing of data should be approved individually by each client.
That is, open banking regulations do not constitute a general mandate to share customers’ data but an obligation to share specific data in a controlled environment under specific conditions. The data-sharing conditions are adapted to the context. In other contexts, data-sharing obligations and conditions necessary to promote competition might be different.5
This basically reflects that data is not a commodity and thus data-sharing regulations cannot be of a generic nature. The nature of data and its role in each industry competitive dynamics matter in the assessment of market power derived from data collection and processing. Certain types of data might be of strategic value for a specific industry but of no value at all to others. For example, purchasing patterns might be of strategic relevance for marketplaces but irrelevant for insurance companies. Data cannot be regulated as a commodity and different types of data in different industries would require a differentiated regulatory approach (if any).
Another important element is the sharing of the data across industries. The open banking regulations do not generally allow the use of banking data outside the banking sector. There are many digital players that are present in several industries. Restricting data-sharing across industries while big digital players can pool their own cross-sector data internally would still imply a competitive advantage for multi-sector companies.
In this respect, the recently announced European Data Strategy of the European Commission proposing, amongst other measures, a “cross-sectoral governance framework for data access and use” and “common European data spaces in strategic sectors and domains of public interest” through pooling European data in key sectors, with EU-wide common and interoperable data spaces, constitutes a vague policy statement of principles that needs to be narrowed down to be operational and which articulation would probably require hundreds of data- and sector-specific regulations.
III. SELF-PREFERENCING: INDEPENDENT FINANCIAL ADVISORS
Gatekeepers may have incentives to treat their own services more favorably than those of their competitors. This is known as self-preferencing.
Banks have traditionally been the main gatekeeper and facilitator of the overall financial industry. Banks have also traditionally offered financial advice to their clients. The financial advice provided by banks may not be fully independent since they may have incentives to promote their own financial products or sponsored third-party products which report them higher commissions.
MiFID II (Markets in Financial Instruments Directive II) is a European Union legislative framework that regulates financial markets strengthening investor protection and improving the functioning of financial markets making them more efficient, resilient and transparent. It entered into force on January 3, 2018. MiFID II introduces a complex set of requirements for firms qualifying as independent financial advisors.
MiFID II makes a distinction between independent and non-independent advisors. Before providing advice, an investment firm must inform a client whether this advice is being provided on an independent or a non-independent basis. Independent advisors must meet a number of requirements. Amongst others, they must include in their product portfolio different types of financial instruments offered by various providers and cannot limit their range of products to financial instruments issued or provided by the investment firm itself.
In addition, the so-called inducements regime, which includes a complex set of rules relating to conflicts of interest and other issues,6 prohibits third-party benefits in relation to portfolio management and independent investment advice. That is, it prohibits independent advisors accepting and retaining benefits received from third parties in relation to the firm’s provision of portfolio management or investment advice services to its underlying clients. This implies independent advisors charging fees to their clients since they can no longer receive fees from financial product providers.
Thus, MiFID II allows financial advisors favoring their own products and third-party sponsored products but, if they do so, they must clearly signal that they are non-independent advisors.
The competition concern in this case was created by the incentives of financial institutions to sell their own products and sponsored products from third parties under the appearance of independence. The remedies consist of a complex set of behavioral incentives and prohibitions. On the one hand, the regulation offers financial institutions incentives to commercialize third-party products by granting them the “independent advisor” denomination. On the other hand, the regulation prohibits third-parties paying for being part of financial institutions portfolios.
MiFID II regulations is complex and often difficult to monitor since they involve verifying not only that the regulations are met on paper but also that they are effectively implemented. For example, the fact that a financial institution offers products from third parties and can thus be considered an “independent advisor” does not guarantee that, in practice, it does not only sell its own products. Also, retail investors are not fully aware and conscious of the importance of independent fee-based advice which reduces the effectiveness of the regulation. The impact of inducement rules has been limited on financial services providers’ decision to offer independent advice or to refrain from it.7
In general, addressing self-preferencing through behavioral remedies requires complex regulatory structures which are often difficult to implement and monitor. It also needs that, as a result of such regulations, firms and users change their behavior, which is not always the case. Such remedies should be carefully designed, tailored to each situation and subject to a detailed impact assessment, especially if such regulations affect the business models of the affected parties.
For example, many digital players are constituted as multisided platforms where the flow of payments between the different sides of the platforms follows the nature and strength of direct and indirect network effects. A MiFID II approach to self-preferencing might imply regulating the flows of payments between the different parties. This may affect the core business model of the platform which is often based on the cross-subsidization between sides. Changing the business model might harm some sides of the platform and might even affect the economic feasibility of the business, reducing ultimately consumer choice.
This is somehow the approach adopted by the European Commission when challenging Google Android’s business model which placed Google’s affiliated services under more favorable conditions than competing services. The solution proposed by Android implied changing its business model and moving towards a license fee model. The welfare consequences of this change of business model are not obvious.
IV. INTEROPERABILITY: PAYMENTS SETTLEMENT
Interoperability refers to the ability of alternative service providers to operate with the gatekeeper’s software or hardware. The lack of interoperability may prevent new entrants to compete with the gatekeeper’s services.
Interoperability has been essential for the introduction of competition in the payment services industry. Payments between accounts located in different payment service providers need a settlement agent. When a customer makes a payment to a business that has an account with a different provider, the customer’s provider owes the business’s provider the value of the payment. This creates a level of risk, so a payment system needs to use an intermediary, known as a settlement agent, for the final settlement of funds between providers.8 In the UK, the Bank of England plays this role backing the payment system and guaranteeing monetary and financial stability.
Until 2018, most non-bank payment service providers (“non-bank PSPs”) accessed the UK payment schemes through a sponsor bank to settle on their behalf, using the sponsor’s settlement account at the Bank of England. Direct settlement with the Bank of England was not available to non-banking institutions. Indirect access arrangements were subject to commercial negotiation. Some Indirect Access providers subscribed a voluntary Code of Conduct for indirect access.9
Most PSPs opted to enter the market through the so-called Bank-as-a-Service (BaaS) platforms10 which allowed them to use an existing bank infrastructure to settle their transactions. The use of BaaS platforms implied that non-bank PSPs were competing with banks to provide payment services while, simultaneously, were accessing the settlement system through the banks they were competing with. The banks had thus access to their competitors’ data and could also limit the operation and innovation of new entrants by limiting their interoperability. Alternatively, PSPs could avoid settling through a sponsor bank by holding a banking license in order to be able to settle directly with the Bank of England. Such alternative required though meeting a number of costly regulations which constituted an expensive market entry ticket.
From 2018, the Bank of England allowed non-bank PSPs to open settlement accounts, subject to appropriate safeguards.11 This means that non-bank PSPs no longer need to use competitors’ BaaS platforms to offer their services, facilitating competition and innovation in payment services. On April 13, 2018, TransferWise, a Financial Conduct Authority-regulated PSP, became the first non-bank PSP joined a UK payment system settling in central bank money. The U.S. Federal Reserve announced in August 2019 that it would develop its instant payment system allowing non-bank PSPs to operate independently of banks.12
The competition concern in this case was the fact that market entry required operating through the infrastructures of competitors. The regulator initially opted for light-handed regulation leaving market players to self-regulate and negotiate agreements privately. The final regulatory solution implied contesting the gatekeeper position of the incumbents by allowing an “outside option”13 which involved settling directly with the Bank of England. This was made possible due to existence of an independent secure and reliable settlement agent that was the central bank. The existence of outside options improves the bargaining position of new entrants making interoperability regulation less necessary.
The existence of outside options plays an important role in eroding the market power of digital gatekeepers. Outside options might not always be available in digital markets and often depend on users’ behavior. For example, if users multihome in several digital platforms, market players might be able to meet through different “routes” and regulating interoperability in each platform might become unnecessary. The context and the users’ behavior thus determine the need (or not) for regulating interoperability. The existence of alternative “gates” to enter the market together with users multihoming reduce the need for regulation.
V. IMPLICATIONS FOR THE REGULATION OF DIGITAL MARKETS
The regulatory process in digital markets requires identifying the specific competition concerns and designing the appropriate remedies to address such concerns. In order to be effective and reduce the distortions created by market interventions, remedies should be objective and proportional and tailored to the competition problems, to the characteristics of the industry and of the business models of the market participants. It is also relevant that remedies do not create competition distortions and that do not make existing business models that benefit consumers not economically feasible.
There are plenty of examples of industries where gatekeepers are subject to some degree of regulation to promote competition and market entry. Despite the scale and scope of some digital players that increase the magnitude of the impact of their actions, some lessons can be learnt from other industries. This article shows different examples of competition concerns and subsequent regulatory responses in the banking sector addressing the use of data, self-preferencing and interoperability. The examples show that solving specific competition problems requires specific analysis of the problems and specific remedies adapted to the conducts, to the industry characteristics and to the business models.
Imposing effective remedies to competition problems requires a deep evidence-based analysis of the competition concerns at hand and of their impact in order to assess whether policy intervention in needed and to select the appropriate policy instrument. Issuing generic obligations might be a starting point but not fully operational, given that competition problems can adopt different forms, differ across different industries and their solution very much depends on the characteristics of the problem, the industry and the business models of the market players.
In this sense, the DMA proposes a list of generic obligations for all gatekeepers and a second list of obligations susceptible of being imposed on digital gatekeepers under certain circumstances. The proposed DMA does not however specify under which circumstances and for which industries and business models each obligation of the list is relevant. Potential competition concerns posed by gatekeepers can be very diverse. Making the DMA more operational would require a more detailed description and classification of the potential competition concerns in each market and for the different market models and establishing the link between the different competition concerns and the potential remedies.
The market investigation tool proposed by the DMA broadens the set of policy tools available for regulators to address competition concerns in digital markets but it should not be perceived as a magic tool. Market investigations might skip the legal hurdles of defining the markets and finding dominant positions, and allow to focus the analysis on the impact of the investigated conducts. However, market investigations require a detailed analysis of the competitive dynamics of the industry and of the impacts of the investigated conducts. Also, the potential proposed remedies would have to be designed on a case-by-case basis depending on the conducts and the market characteristics. Market investigations add a new tool to the toolbox of regulators but there is no guarantee that market investigations will constitute a speedier and more flexible policy instrument.
Finally, the setting-up of Regulatory Sandboxes, that enable a direct testing environment for innovative products and are widely used in financial markets, could be a useful instrument in digital markets to make sure that regulation promotes competition and innovation and that ultimately improves consumers welfare.
1 Global Economics Group.
2 Furman, J., Coyle, D., Fletcher, A., McAuley, D., & Marsden, P. (2019). “Unlocking digital competition,” Report of the Digital Competition Expert Panel. HM Treasury. Stigler Center for the Study of the Economy and the State (2019).
3 See, for example ACCC (2019), Digital platforms inquiry – final report. June 2019. Crémer, J., de Montjoye, Y. A. & Schweitzer, H. (2019) Competition Policy for the Digital Era. Report commissioned by the European Commission, Luxembourg, 2019. Furman, J., Coyle, D., Fletcher, A., McAuley, D., & Marsden, P. (2019). “Unlocking digital competition,” Report of the Digital Competition Expert Panel. HM Treasury. Stigler Center for the Study of the Economy and the State (2019), Committee for the Study of Digital Platforms Market Structure and Antitrust Subcommittee Report, George J. Stigler Center for the Study of the Economy and the State, The University of Chicago Booth School of Business, 2019.
4 See CMA’s Retail Banking Market Investigation Order (2017).
5 In 2019, in the context of the acquisition of the Belgian TV broadcaster De Vijver Media by Telenet, a broadband service provider, the Belgian Competition Authority (BCA) showed some concerns about the potential use of the data on consumer behavior collected by Telenet, to improve the targeted advertising in their own channel. To address this concern, the BCA mandated Telenet to provide other TV broadcasters access to the data concerning their respective channels. The access was limited in scope (broadcasters has access only to data related to their channels) and in the level of detail (access was not provided to raw data but to an upper aggregation level). See Vande Walle, S. (2019), “Sole Control: The Belgian Competition Authority Clears a Vertical Merger in the Audiovisual Sector, Subject to Conditions,” in Concurrences Competition Law Review, Number 3, 2019, pp. 120-122.
6 See “Understanding the MiFID II Inducements Regime,” Latham & Watkins. Oxford Business Law Blog, May 30, 2018.
7 See ESMA – European Securities and Markets Authority (2020). Final Report: ESMA’s Technical Advice to the Commission on the impact of the inducements and costs and charges disclosure requirements under MiFID II, March 31, 2020.
8 See Payments and Settlement webpage at the Bank of England for a description of the functioning of payment services.
9 For example, five UK banks signed a voluntary Code of Conduct for Indirect Access Providers to address concerns around the commercial access arrangements they provide to PSPs requiring Indirect Access Services to UK Payment Systems.
10 BaaS is an end-to-end model that allows digital operator to connect with banks’ systems directly so they can offer their services using the bank’s infrastructure. See “How the banking-as-a-service industry works and BaaS market outlook for 2021” (Business Insider, January 4, 2021) for a brief summary of how BaaS service operate.
11 See Guide on Access to UK Payment Schemes for Non-Bank Payment Service Providers developed by the Bank of England, FCA and Pay.uk (December 2019) for a description of the current settlement options for non-bank PSP.
12 Federal Reserve Press Release: “Federal Reserve announces plan to develop a new round-the-clock real-time payment and settlement service to support faster payments,” August 5, 2019.
13 An outside option is defined to be the best alternative that a player can command if she withdraws unilaterally from the bargaining process.
