Antitrust In the Digital Era: A Contextual and Fact-Based Approach

The Network Law Review is pleased to present you with a special issue curated by the Dynamic Competition Initiative (“DCI”). Co-sponsored by UC Berkeley and the EUI, 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. This special issue brings together contributions from speakers and panelists who participated in DCI’s second annual conference in October 2024. This article is authored by Lars Kjølbye, Partner at Latham & Watkins, Brussels.

***

The current European antitrust policy framework was conceived more than two decades ago in a very different economic and political environment. The economy was anchored in the industrial era. Market positions were driven largely by production capacity, efficiency and incremental innovation.[1] And because of a trail of failed government interventions into the economy, there was a belief that in general markets were better at delivering positive outcomes for consumers. Antitrust should intervene only when economic and effects-based analysis confirmed likely harm to competition and consumers. In the industrial era economy that translated into a focus on short-term effects. The European Commission (Commission) issued guidelines setting out an analytical framework for various types of mergers, collaborations and conduct with bright lines to provide legal certainty and economic tests to deliver sound effects-based enforcement.[2] The guidance was relevant across economic sectors because in the industrial era companies tended to use very similar routes to market, collaboration vehicles and commercial practices.

Fast forwarding two decades, the economic and political environment has changed, challenging the industrial era antitrust policy foundations. The economy has entered the digital era characterized by differentiation, complexity and breakthrough innovation. Talent is the key asset and innovation drives success. Production assets have limited importance in many sectors – production can be outsourced – and there are many different business models and routes to market strategies. As a result, the relevance and reliability of our present antitrust analytical framework is questioned. Alongside, the financial crisis gave new life to the belief that governments do better than markets and that intervention is required to steer markets and prevent accumulation of economic power. The antitrust policy response has been to eliminate bright lines, relax tests to make it easier to intervene, and advance new theories of harm.[3]However, we have yet to develop a robust analytical framework fit for the digital era.

1. Competition In The Digital Era

In the digital era, innovation in new and differentiated products and services drives competition. Value is created not by bending metal and refining mechanical functions but through software, data analytics and digitizing products. Software is highly flexible. It is not restricted to the same extent as physical products by the laws of physics and advances in computing technology continuously shifts the barriers for what digitization can deliver. AI is the new frontier which is transforming e.g., the way we work.[4]

The industrial era taught us that potential competition means entering an existing market within a short period of time to compete with someone who is already established in that market.[5] It also taught us to focus on closeness in terms of price and product features. The closer the fit, the greater the risk to competition. In the digital era, markets are fluid and driven by innovation to win customers by solving problems in new ways. Closeness is typically an opportunity loss. Differentiation is key to success. Successful entry requires not looking like the competitors. Zoom succeeded not by being a small version of Cisco – the company Zoom’s founders’ left – but by creating a new simple to use, bandwidth lean video meeting service that made video calling the new business phone. Similarly, it was not an improved PC operating system that made consumers look beyond Windows; it was a new way of doing computing called the smartphone. Innovation competition has created a vast array of digital products and services that integrate with very different business models and monetization strategies. A robust competition policy for the digital era must capture all these elements to support sound enforcement.

2. Fact-based Contextual Analysis

Industrial era homogeneity made antitrust policy development comparatively easy. It was possible to devise economics-based tests to frame the analysis with a sufficient degree of accuracy.[6] While these tests remain useful, they should no longer constitute the core of the analysis. That core should be contextual analysis serving to understand the functioning of a particular space, what drives competition in that space, and where dynamic competition is taking it.

In the EU, the legal foundation for that shift exists. As illustrated by its recent judgment in Google Shopping, the EU Court of Justice has made clear that antitrust analysis must reflect the specific context in which a given conduct or transaction operates.^[7] In that case, the European Commission found that Google had abused its dominant position in online search by demoting competing comparison shopping services in search results and giving prominence to its own comparison shopping service by means of favourable placement and dynamic display features. The European Commission showed that competitors depended on Google’s search engine for a high share of their traffic and that Google’s conduct diverted a significant share of their traffic away from competitors because the first few search results get almost all the clicks. In its judgment upholding the European Commission’s decision, the Court of Justice held that “it cannot be considered that, as a general rule, a dominant undertaking which treats its own products or services more favourably than it treats those of its competitors is engaging in conduct which departs from competition on the merits irrespective of the circumstances of the case.”^[8] A finding of self-preferencing abuse turns on the specific circumstances of the case. Competition on the merits involves attracting customers based on e.g., price, choice, quality or innovation.^[9] Winning on the merits promotes effective competition rather than harming it, even less efficient competitors struggle as a result.^[10] A deep contextual dive helps distinguish anti-competitive foreclosure from competition.

The EU Digital Markets Act illustrates the challenges that arise when the context is ignored. Article 5(2) of the DMA requires designated companies to obtain end user opt-in consent e.g., before combining personal data from a designated service with data from other services. The obligation is inspired by the German competition authority’s antitrust case targeting Facebook’s use of data to personalize ads.[11] Meta is a leading provider of personal social networking services and online non-search advertising. However, Article 5(2) of the DMA applies to any designated service and thus equally to e.g., LinkedIn, which like Facebook is a designated online social networking service. LinkedIn is a small provider of online advertising services, ^[12] and the DMA’s approach may end up limiting LinkedIn’s ability to improve its advertising service which is closely linked to its social networking service.^[13]

Simply requiring contextual facts-based analysis does not suffice. The analysis must be framed so as to capture key elements of dynamic competition. Without such framing, application of the law becomes unmanageable and prone to error. We should aim to develop a set of building blocks that can guide us in framing the analysis. These building blocks should not target first and foremost the status quo;[14] they should capture dynamic competition and its drivers. Capabilities analysis provides a useful source of inspiration in that regard.

3. Capabilities Analysis As Antitrust Building Blocks

The digital era’s high degree of product and business model differentiation reflects individual companies’ capabilities. Each player’s capabilities to engage in dynamic competition reflect its talent base and accumulated skills. An established firm’s internal capabilities frame the innovative paths available to it. Moving in different directions is often challenging because its exiting paths constrain its ability to attract talent – the firm is associated with its existing capabilities. Start-ups are not constrained by exiting capabilities and paths. Experience shows that talented people find start-ups attractive because of their agility and ability to think out of the box. Start-ups are often capable of assembling small teams to drive innovation. Their challenge is often a lack of the capabilities required to transform innovation into a product and bring it to market.

There is increasing focus on potential competition and concerns that large technology companies buy start-ups rather than building the technology themselves.[15] Analysis of capabilities can provide the contextual insights required to test that assumption. In particular, it can help overcome the challenge of identifying potential competition concerns when closeness is losing its predictive value. The relevance of using capabilities as an analytical building block is illustrated by Microsoft’s acquisition of Nuance, a provider of voice recognition and transcription software mainly to physicians. Nuance’s software enabled transcription of patient information directly into electronic health records. The acquisition was cleared by the European Commission in December 2021 and by the UK Competition and Markets Authority in March 2022.[16]

Prior to the transaction, Nuance had entered into a partnership with Microsoft to develop its next generation product called DAX – a cloud-based service that uses sensors around a physician’s office or examination room to passively record doctor-patient conversations and automatically turn them into structured medical notes that can be entered directly into a patient’s medical record.[17] If fully successful, it would be a breakthrough in reducing doctors’ administrative burdens and liberating time to serve patients. The pre-existing partnership led to the conclusion that there was no merger-specific potential competition to consider.[18] However, an analysis of the parties’ respective capabilities would have shown that the merger promoted dynamic competition by significantly increasing the probability that DAX would materialize. The merger brought Microsoft’s and Nuance’s capabilities together and provided a stable environment (compared to the partnership) for the merged entity to work on delivering the next generation product. Nuance’s strength was its deep understanding of the healthcare sector, the specialized vocabulary used, specialized data assets, and integrations with electronic health record systems. Microsoft is a provider of horizontal platforms and systems serving many industries. It had developed general natural language recognition and AI technologies and cloud computing infrastructure. But it lacked Nuance’s specialized capabilities and given the direction of its own capabilities it would have been no easy matter for Microsoft to attract the talent required to create them.

4. Conclusion

A new rigorous framework for antitrust analysis fit for the digital era is needed. The solution is not to simply eliminate elements of the industrial era policy framework that are either consider too limitative or outdated. New thinking is needed to adequately reflect the new competitive dynamics. Because of the heterogeneous and dynamic nature of the competitive environment, the new framework should be anchored in the specific competitive context of a case. Building blocks are required to make such an approach workable and capabilities analysis provides an interesting avenue for developing one of those building blocks.

***

Citation: Lars Kjølbye, Antitrust In the Digital Era: A Contextual and Fact-Based Approach, Network Law Review, Spring 2025.

References:

• [1] As illustration the EU law notion of dominance was anchored in the firm’s volume of production, see e.g., Case 85/76 Hoffmann La-Roche EU:C:1979:36, para. 38.
• [2] See e.g., Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings, OJ [2004] C 31/5; Guidelines on the assessment of non-horizontal mergers under the Council Regulation on the control of concentrations between undertakings, OJ [2008] C 265/6; and Guidelines on the application of Article 81(3) of the Treaty, OJ [2004] C 101/97.
• [3] See e.g., Case M.10615 – Booking Holdings/Etraveli Group, para, 201, where the Commission  departed from its Non-Horizontal Merger Guidelines. It explained that in the face of dynamic and rapidly changing economic reality, those Guidelines did not prevent it, in circumstances that were not envisaged by them, to examine whether the concentration significantly impeded competition by strengthening a dominant position.
• [4] See e.g., E. Brynjolfson, D. Li, and L. Raymond, Generative AI at work, The Quarterly Journal of Economics, available at: https://academic.oup.com/qje/advance-article/doi/10.1093/qje/qjae044/7990658.
• [5] See e.g., Commission Guidelines on the applicability of Article 101 TFEU to horizontal co-operation agreements, OJ [2022] C 259/1, para. 17.
• [6] See e.g., Commission Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings, OJ [2009] C 45/7.
• [7] Case C-48/22 P, Google Shopping, EU:C:2024:726, paras. 166 and 168.
• [8] Id., para. 186.
• [9] Case C-209/10, Post Danmark (I), EU:C:2012:172, para. 22.
• [10] Idem, paras. 21 and 22.
• [11] See https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/2019/07_02_2019_Facebook.html.
• [12] See https://www.statista.com/statistics/290629/digital-ad-revenue-share-of-major-ad-selling-companies-worldwide/#:~:text=In%202023%2C%20Google%27s%20share%20of,with%20an%20expected%20seven%20percent.
• [13] See Commission Designation Decision of 5 September 2023 in Cases DMA.100017 Microsoft – online social networking services; DMA.100023 Microsoft – number-independent interpersonal communications services; and DMA.100026 Microsoft – operating systems, para. 144.
• [14] See e.g., Case C-413/14 P Intel EU:C:2017:632, para. 139, which refers to e.g., market coverage as a relevant factor when assessing capability to foreclose.
• [15] For example, the new UK merger control jurisdictional tests “provide a more comprehensive and effective jurisdictional basis to review mergers involving potential competition or dynamic competition”, see UK Competition and Markets Authority, Mergers: Guidance on the CMA’s jurisdiction and procedure, 2 January 2025, para. 4.82.
• [16] Commission Decision of 21 December 2021, Case M.10290 – Microsoft / Nuance; and UK Competition and Markets Authority Decision of 2 March 2022 in Case ME/6940/21 – Acquisition by Microsoft Corporation of Nuance Communications, Inc.
• [17] Commission Decision, para. 103 with fn. 82.
• [18] UK Competition and Markets Authority Decision, para. 72.