From Microsoft to the DMA: Through the lens of behavioural economics

At this year’s European Competition Policy Forum in Como, a session discussed the impact that the Microsoft antitrust cases have had on European competition law over the past 25 years. These cases have shaped enforcement; their influence is evident in Article 102 cases and the new Guidelines, [1] as well as the Digital Markets Act (DMA). I joined the panel, chaired by Jorge Padilla, with contributions from Assimakis Komninos (White & Case), Michele Polo (Bocconi University and Compass Lexecon), Marc van der Woude (Erasmus University Rotterdam and former President of the General Court), and Zita Vasas (Compass Lexecon).

This article develops Zita’s contribution to the panel, which examined how behavioural economics can inform the assessment of anticompetitive conduct and design of remedies. It focuses on the two Microsoft “tying” cases: Windows Media Player; and Internet Explorer. [2][3] Their remedies sought to strengthen consumer choice, and they anticipate elements of the DMA’s approach – most directly through the Internet Explorer choice screen, which prompted users to consider alternative browsers. Yet, the evidence that these interventions improved market outcomes is limited, which raises the question whether outcomes improved substantially enough to justify the costs imposed on firms and users. Briefly, this article considers why, and what that implies for the design and timing of future EU competition law interventions.

The views expressed in this article are the views of the authors only and do not necessarily represent the views of Compass Lexecon, its management, its subsidiaries, its affiliates, its employees, or its clients.

In the article, Zita examines six points:

• Why the Microsoft tying cases are best understood through a behavioural economics lens. These cases differ from the interoperability cases, which concerned technical barriers to rival products. Rival media players and browsers could run on Windows, but the design of the choice architecture, i.e., the user interface, encouraged users to adopt and continue using Microsoft’s products rather than alternatives.
• What behavioural insights tell us about the theories of harm. Defaults, salience and familiarity can influence consumers’ choices independently of product quality. But choices can also reflect rational preferences. As the Court of Justice later recognised, once conduct has distorted those choices, its effects can be difficult to disentangle qualitatively from those of product quality.
• How behavioural insights informed remedy design. The Internet Explorer remedy required a choice screen that actively presented rivals, with a design to address framing and ranking effects and avoid choice overload. In contrast, the Windows Media Player remedy required an unbundled version of Windows, but gave little consideration to users’ incentives to choose it.
• How the remedies affected outcomes. Broadly, both remedies failed. The unbundled Windows barely reached users. The browser choice screen did affect outcomes, but not much: it accounted for less than two percentage points in the decline of Internet Explorer’s market share.
• Behavioural insights now underpin EU competition enforcement and regulation. The DMA now builds on many features foreshadowed in the Internet Explorer remedy, including choice screens and safeguards against circumvention. The new Article 102 Guidelines expressly recognise inertia and behavioural bias in tying and self-preferencing.
• Earlier intervention may help improve outcomes, but carries its own risks. The limited impact of the Internet Explorer (and also the Google Android) choice screen may reflect deficiencies in its timing rather than its design. By the time it was introduced, it was too late to address the problem identified. But acting earlier has its own risks and, potentially, they are greater.

---

References