Carmelo Cennamo: “The EU Digital Markets Act: It Is Not About Markets But Ecosystem Failures!”

The Network Law Review is pleased to present you with a Dynamic Competition Initiative (“DCI”) symposium. Co-sponsored by UC Berkeley, EUI, and Vrije Universiteit Amsterdam’s ALTI, the DCI seeks to develop and advance innovation-based dynamic competition theories, tools, and policy processes adapted to the nature and pace of innovation in the 21st century. The symposium features guest speakers and panelists from DCI’s first annual conference held in April 2023. This contribution is signed by Carmelo Cennamo, Professor with special responsibilities of Strategy and Entrepreneurship at Copenhagen Business School.


1. Introduction

The Digital Markets Act (DMA) represents a new institutional framework redefining the principles of what constitutes anticompetitive behaviour and abuse of advantageous position in the digital economy, and perhaps more importantly, the domain within which such conduct is assessed. It imposes a list of obligations on firms operating large digital platforms, the so-called “gatekeepers,”1“Gatekeepers” in the DMA are very large digital platforms with a “systemic role in the internal market” due to their bottleneck position, and are identified according to the criteria set out in Art. 3 DMA of having a “significant impact on the internal market,” providing “a core platform service” and enjoying “an entrenched and durable position.” to promote fairness and contestability in digital markets. As the DMA enters its implementation phase, there are still questions about what its specific objectives are, and how these obligations will be implemented. There is still some level of ambiguity about what the DMA is really trying to achieve. This, along with the interpretation of the DMA’s littera, will largely depend on the underlying conception of digital platforms and the role they play for and in the digital economy. The idea advanced in this short article is that the way we see platforms, as pure market infrastructures or as ecosystem organizations (providing the organizing principles for joint value production and consumption), greatly change the assumptions about what constitutes harmful conduct and the remedies to such conduct.

The DMA does not define what is the “relevant digital market” one must consider. The DMA identifies a set of core platform services (CPS) representing main gateways to related digital markets; yet, it defines quantitative revenue and user base thresholds to designate the firm holding the “gatekeeper” position, which thus is subjected to the obligations.2The DMA does not formulate explicitly an overarching principle for what constitutes a firm’s anticompetitive behaviour and abuse of dominant position; instead, it anchors a detailed list of obligations (“dos” and “don’ts”) to the two overarching objectives, “fairness” and “contestability.” The obligations to “desist” (the list of don’ts) refer to conduct that is presumed either as “unfair” towards platform’s business users or end users, or blockading “contestability” of the gatekeeper position. The obligations to “act” (the list of dos) refer to conduct that the gatekeeper must adopt to promote fairness towards business/end users.

In fact, the DMA is not about traditional anticompetitive conduct, nor about (digital) markets; it is largely about preserving the “health” of digital ecosystems! Essentially, it is about limiting potentially prevaricating (and abusive) practices by the firm controlling the platform core services over the entire digital ecosystem gravitating around the platform CPS; shortly, guaranteeing the well-functioning of the ecosystem. The DMA does not use that language, while it should! It would make the regulatory framework clearer and signal that this is a new institutional framework departing from traditional antitrust law.

  • It replaces the concept of relevant market, for instance, with Core Platform Services, around which an ecosystem emerges.
  • It replaces the concept of dominant position with the gatekeeper designation.
  • It replaces ex-post assessment of economic efficiencies against the abusive conduct with ex-ante obligations for the preservation of the ecosystem health!

By defining the new organizing principles (the “rules of the game”) and the structures of economic relations wherein (joint) value production and consumption take place, ecosystems are the new relevant locus of innovation and competition in the digital age. Understanding the objectives and obligations of the DMA from an ecosystem rather than market perspective would make the DMA objectives and obligations coherent, clearer, and provide anchoring criteria for the implementation of such obligations. This is not just semantics. As better articulated below, whether one considers digital platforms just markets (“gateways”) or ecosystems (inter-organizational “value architectures”) has important implications for the theories of harm (i.e., how we see the two objectives of fairness and contestability) and for the application of the obligations (i.e., how we shall pursue such objectives).

2. Digital Platforms: Markets or Organizations?

A common view has it that all digital platforms, independently from how they function, are market intermediaries – gateway to customers –, facilitating transactions between business providers and end users interacting through the platform. As gatekeepers, platforms control transactions, extract value by monetizing the matching between the two sides, and cumulate data, which they use to strengthen their market position. This means that the business model adopted by a very large platform is neither relevant to determine whether a platform is to be considered a gatekeeper, nor to establish which of the new obligations are to be implemented by it. Accordingly, contestability would refer to the ability of any business provider to overcome entry barriers to “the market,” where the market is the platform’s specific CPS. Fairness instead would attain more to the relationship between the platform provider and its business users, and more specifically to the asymmetric distribution of the value realized and exchanged on and through the platform.

This is a reductive view of the digital economy, considered as an additional market channel for existing activities (and goods), one which does not account for how value is produced in the first place. It takes a static view at competition, consider the dynamism of dynamic competition as just “future” entry to the (same!) “market” offering similar (!) competing product offerings. As maintained elsewhere,3See Cennamo et al. (2022); Jacobides et al. (2022). this is not how dynamic competition unfolds in digital domains orchestrated by platforms. Platforms are more than just markets; they are a new organizing structure for collaboration towards the joint production and consumption of goods and services. Digital platforms do not just facilitate existing transactions between business users and end-users, they also enable new interactions that would not occur in the absence of the platform. These interactions are linked to the production of novel kinds of data which further contribute to the innovativeness of platform ecosystems.4Alaimo, C. and Kallinikos, J. ‘Computing the everyday: Social media as data platforms’ (2017) The Information Society 33(4): 175-191; Alaimo, C., Kallinikos, J., and Valderrama, E. ‘Platforms as service ecosystems: Lessons from social media’ (2020) Journal of Information Technology 35(1): 25-48. It is the capacity to generate these novel interactions that unlock innovation in ecosystems, not just of products or services but about new ways to attain to a customer need (such as new ways to search for or rate a product/service – e.g., à la; new ways to fund a venture business idea – e.g., à la Kickstarter; new ways to provide financial services to the unbanked ones – e.g., à la Kiva). It is innovation in the forms of organizing economic activity that represents the most powerful form of dynamic ecosystem competition, with one ecosystem organized differently offering alternative solutions to existing ecosystems. Viewing digital platforms (and their CPS) as ecosystems highlights their organizational role as architects of new “value architectures” providing the organizing principles for joint value production and the system within which complementarities can be attained in production and consumption.

Ecosystems are an integrated solution for heterogenous customers in contexts where such solution is lacking due to firms’ failure to cooperate and/or co-specialize their assets/activities for joint value production.5See Jacobides et al. (2022). Ecosystems offer inter-organizational blueprints to structure relationships among complementors and define the roles and rules of collaboration. They serve as governance arrangements to internalize the externalities of these cooperation interdependencies. This facilitates customer-facing integrated solutions, driven by autonomous complementor contributions, set within the boundaries of ecosystem governance. As advanced in Jacobides, Cennamo and Gawer (2022), “ecosystems therefore minimize the costs of coordination and cooperation and allow the comparative advantages in innovation to be leveraged by firms that specialize in each module, while also guaranteeing system-level integration of those modules into a coherent set of value options for the customer.” In other words, ecosystems allow for complementarities to emerge and expand by integrating assets and capabilities of the central, platform firm with those of complementor firms.

If we see platforms for their broader, organizational role they play in the digital economy, the focus on fairness and contestability in the DMA should be directed not just to access to the customer base of a given platform but to the underlying “core assets” and interorganizational arrangements that enable firms to contribute to the joint value proposition (i.e., build complementarities) leveraging their specialized capabilities. Focus should then be on preserving ecosystem health and limiting “ecosystem failures,”6In their recent paper, Jacobides M., Cennamo C., and Gawer A., ‘Complementarities and externalities in platforms and ecosystems: From value creation to inherent failures’ (2022) Working paper: London Business School, lay out a framework to understand the kind of market failures that digital platforms help solve as new organizational models, and the inherent, post hoc ‘ecosystem failures’ that may emerge as a result of these new structures. They distinguish between functional failures—problems with the inherent ability of platforms and ecosystems to deliver value to the final customer; and distributional failures—issues associated with ecosystem participants’ inability to capture value proportional to their joint contribution. that is, ecosystems not functioning properly in their capacity as inter-organizational structures7Kretschmer T., Leiponen A., Schilling M., Vasudeva G. ‘Platform ecosystems as meta-organizations: Implications for platform strategies’ (2022) Strategic Management Journal 43: 405-424. to stimulate innovations that need to come together to create specific value propositions for consumers.8see e.g., Cennamo C., and Santaló J. ‘Generativity tension and value creation in platform ecosystems’ (2019) Organization Science 30(3): 617-641. Table 1 summarizes these different views and the related emphasis’ elements in each.

Platforms as gateway to market Platforms as organization of joint value
Main role: coordination of market participants Main role: organization of joint value production
Main Benefits: Enhance efficiency of market transactions for existing intermediaries Main Benefits: Enable innovation in terms of new activities and interactions
Variety drives value: more complementors/products is better Complementarity drives value: stronger complementarities in production or consumption is better (joint value proposition)
Competitive focus is on (multi)product market competition Competitive focus is on value system competition
Dynamic competition: future entry to the market of providers of similar (hopefully) better solutions Dynamic competition: ecosystem-level innovation offering alternative configurations and solutions to heterogeneous customer needs
Antitrust focus: Need to open platforms and data to all participants Antitrust focus: Need to preserve ecosystem health given the differentiation of approaches across ecosystems

Table 1. The different views on platforms: markets vs. ecosystems 

3. The Ecosystem View: Implications For the DMA

What do we gain by viewing platforms as organizations and considering the entire ecosystem as the relevant unit of analysis? I see important implications for the DMA, the most immediate of which are briefly discussed below.

  1. Gatekeeper designation: the DMA focuses on size thresholds of the firm operating specific CPS. Yet, this does not help identify what is the core element, i.e., the “bottleneck” that grants the platform provider the gatekeeper power over end users and thus, bargaining power vis-à-vis business providers. It is just assumed that that power supposedly comes from the network effects around the platform, themselves a function of platform size. An ecosystem view would shift the focus on ecosystem dynamics and consider the gatekeeping position as function of the “architectural” control the firm has, i.e., the control over the core asset which constitutes the “kernel” of the ecosystem anchoring both end users and business providers across multiple markets.
  2. Application of the obligations: Identifying the relevant ecosystem is critical to understand the domain of application of the obligations in a specific case, including which components should be made open and interoperable:
    1. Identifying what is the core component without which the ecosystem cannot work properly is critical to understand the relevant ecosystem, which often transcends multiple CPS, as well as the externalities it helps to internalize and fix. Such analysis is needed to develop a contextualized understanding of the potential ecosystem failures and thus separate out the core, needed component of the ecosystem (i.e., the “kernel”) from the other, add-on expanded components that might be centralized by the gatekeeper for value capture purposes or anti-competitive (exclusionary) purposes, without any added value to the ecosystem functioning.9In these regards, Cennamo et al. (2022) provide a framework with real examples contextualized to specific provisions of the DMA.
      • implication for contestability: one should open up those add-on, centralized components rather than every CPS.
      • implication for fairness: at the level of single CPS, some exclusionary conduct may not be “unfair” if closeness of the individual CPS implies better overall user experience or differentiation of the whole ecosystem.
  1. DMA objectives: Addressing ecosystem failures is a more substantive way to foster fairness and contestability. Promoting meaningful contestability requires a focus on real opportunities for growth for players who have a chance to exert some significant control in an ecosystem. That implies assessing and addressing “ecosystem failures” rather than assuming that by just “opening up” each CPS will fix the underlying market failures. Focus should be on dynamic competition between ecosystems as affected by the core assets and capabilities, of the central ecosystem orchestrator and the complementors. This implies assessing at the ecosystem level how (the configuration of) those assets and capabilities affect the health of the ecosystem and/or blockade the development of alternative ecosystem configurations by undermining the ability of other firms to come up with such alternative, differentiated ecosystem solutions.
      • we must run proper counterfactuals by considering alternative ecosystem configurations to understand where the bottlenecks to the dynamic competition process may rest.
      • we must consider the broader value architecture and customer journey within and across ecosystems to identify the sources, object, and domain of dynamic competition: competition not on standalone products in a given market but on integrated solutions (systemic innovation) of multiple products across markets; competition on assets and capabilities for building these value architectures. This allows us to understand which part of the platform technology or complementary services is an important, central aspect of the customer journey, missing which, the customer is less likely to interact, transact, or use the platform.

4. Moving Forward

This brief discussion makes it clear how the way we see platforms can greatly change the focus of application of the DMA. In its “reductionist” implementation, focusing on platforms as markets, the DMA may end up protecting specific players by imposing one-fits-all organizational structural forms to platform CPS that might not necessarily be the one unlocking most value opportunities in the digital economy. The risk, in fact, is of reducing plurality of approaches (i.e., different ways of organizing CPS) and by this, reducing, instead of enhancing dynamic competition between ecosystems.

We shall instead look more broadly at limitations to the innovation and competitive process, at the context where these limitations can take place, and use the DMA to remove those constraints. An “holistic” implementation of the DMA focusing on platforms as ecosystems will explicitly recognize the interdependencies between firms/activities in and across ecosystems affecting the competitive process, consider ecosystem-level innovation as the source of dynamic competition, and have as an overarching objective the well-functioning of the whole ecosystem.

In closing, I see some key challenges to moving forward with such implementation of the DMA.

  1. The obligations seem to be too rigid to have any practical, analytical efficacy. Some contextualization is needed to foster dynamic competition, accounting for the reality of a plurality of digital platform business models and ways different ecosystems work.
  2. When considering contestability, there is no distinction between the different levels of competition: within a given ecosystem between players in the same role (i.e., competitors) or complementary role (i.e., complementors to a given joint service), and between rival ecosystems, and the different forms of competition – competing on the ground of ecosystem scale/efficiency/cost vs. ecosystem differentiation (different way of producing, delivering and consuming things – differentiated joint value proposition and consumption experience). This can be problematic for the same implementation of the DMA provisions because it may prefigure cases in which multiple objectives are at play – fairness and contestability – and at trade-off. In such cases, shall we still apply strictly these provisions? If not, how can we accommodate for this while still making the law applicable ex-ante?
  3. In a world of ecosystems redesigning the rules of competition and the roles firms (should) play in the collective enterprise that is the ecosystem, where do we put the boundaries to such collectives and when shall managerial discretion (to design these systems) be constrained? When the interest of one collective (ecosystem) is possibly at conflict with interests of other collectives (be they rival ecosystems or the general public), how do we sort this and prioritize interests?

Carmelo Cennamo

Citation: Carmelo Cennamo, “The EU Digital Markets Act: It Is Not About Markets But Ecosystem Failures!”, Network Law Review, Summer 2023.


Related Posts