Regulatory Choices in the Digital Fairness Act: Banning Practices Without Rethinking Unfairness?

Share

A blog post by Monika Namysłowska [1] and Kasper Drążewski [2]

In October 2024, the Commission published the results of its Fitness Check of EU consumer law on digital fairness. [3] It found that EU consumer law, including the Unfair Commercial Practices Directive (UCPD), was only partly effective in protecting consumers online, pointing to dark patterns, addictive design, personalisation that targets vulnerabilities, hard-to- cancel subscriptions and influencer marketing. [4]

Two years later, on 11 November 2026, the Commission proposed the Digital Fairness Act (DFA), a regulation that would repeal and replace the UCPD. [5] Its new rules address, above all, the practices named in 2024, although the digital environment has not stood still since. If the legislative procedure takes another two years and the regulation applies 12 months after its entry into force (Article 40), those rules would take effect around the end of the decade. The new targeted bans would then arrive more than five years after the diagnosis, while the general clause, the safety net for whatever no specific rule catches, stays as it was in 2005.

This post argues that the DFA, despite its name, is a longer blacklist with a faster update mechanism, attached to an unfairness test from 2005.

The Instrument: Why Is an Act for All Practices Called Digital?

The DFA is a regulation ‘on promoting fair business-to-consumer commercial practices in the offline and digital single market’. It applies to business-to-consumer commercial practices whether offline or online (Article 2(1)), and its Chapter II carries over Articles 5 to 9 UCPD, that is, the general clause and the rules on misleading and aggressive practices. Most of the new substantive rules are in Chapter III, ‘Digital fairness’ (Articles 10 to 29), although the DFA does not define a digital (unfair) commercial practice.
The explanatory memorandum justifies a regulation with the European Council’s preference for regulations, less legal uncertainty and the ‘legal precondition’ for the Commission’s own enforcement powers. [6] It does not discuss a separate instrument for digital practices, nor does it explain the name.

‘Digital Fairness Act’ would suit a separate act on digital commercial practices, which one of us proposed in 2024. [7] For an act covering all practices, we would change one letter of the acronym and call it the Consumer Fairness Act (CFA).

Only one of the DFA’s five chapters is about digital fairness, and the name hides that the act replaces the UCPD for all practices.

The Digital Chapter: A Second Blacklist?

Chapter III consists largely of specific duties and prohibitions. Breaching any of them would be unfair in itself (Article 5(4)(c)), without passing through the test of Article 5(2). Most of the chapter works like a second blacklist beside Annex I, and much of it regulates design. It tells traders which features consumers must be able to switch off (Article 17), how a contract can be terminated online (Article 23) and what a withdrawal button must say (Article 24). Annex I would grow from 47 to 62 points. Five new bans join the misleading practices in Part I, and a new Part III lists ten dark patterns. [8]

Under the UCPD, Annex I can be modified only by revising the Directive (Article 5(5) UCPD). [9] The DFA would let the Commission add dark patterns to Part III (Article 10(3)) and features to Article 17 (Article 17(2)) by delegated acts.

A blacklist of unfair digital commercial practices that the Commission can update by delegated acts is what one of us called for in 2024, [10] and per se rules spare authorities and courts the burden of proving distortion in technologically complex cases. [11] But the delegation is too narrow. The Commission should be able to update the whole blacklist and not just Part III. [12] The bigger problem is that a list is always devised after the practices it bans, and no list can cover every harmful practice. [13]

Specific rules and delegated acts make the lists longer and quicker to amend, but they say nothing about a practice that is on none of them.

The General Clause: Can the 2005 Test Handle New Digital Practices?

A practice that is neither listed nor misleading or aggressive is left to Article 5(2). It is unfair if it is contrary to professional diligence and likely to materially distort the economic behaviour of the average consumer. That provision and the definition of professional diligence (Article 3(9)) repeat the UCPD almost word for word. [14]

This test is ill-suited to digital practices. It has rarely been applied, and its reference to honest market practice gives little guidance where market standards are weak or absent. [15] The Fitness Check found a widening gap between the average consumer, the benchmark of the test, and the way consumers actually behave online. [16] Consumer organisations asked for the benchmark to be redefined. [17] Academics proposed a new standard of conduct for traders who design digital choice environments, [18] or a reading of the existing standard in that light. [19]

The response is modest. The explanatory memorandum announces updated definitions of the average and vulnerable consumer and a ‘general fairness-by-design obligation’. [20] Article 3 defines neither consumer, and no article contains the obligation. The DFA does offer recitals 24 and 25, which acknowledge cognitive biases, following Compass Banca, [21] as well as information overload and limited digital literacy. Article 5(3) would also cover practices that distort the behaviour ‘primarily’, rather than ‘only’, of particularly vulnerable consumers. [22]

The closest Chapter III comes to a general clause is Article 10(1), which prohibits online interfaces likely to deceive or manipulate ‘consumers’ or otherwise to materially distort their economic behaviour. Unlike a blacklisted practice, such an interface must be assessed case by case, yet the provision does not specify against whom. The average consumer appears in Article 5(2) and in recital 23 but not here. Article 10(1) is a general clause for interfaces only, and one without a benchmark. Is the measure the average consumer, the average member of a targeted group or the individual user? A regulation chosen partly to reduce legal uncertainty leaves the answer to national courts.

The Commission added rules around the general clause and left the clause itself untouched, while the one new open rule it wrote has no benchmark.

Proposed Wording: What Would a General Clause for Digital Practices Look Like?

Chapter III needs a general clause of its own that says what it measures. One of us proposed such a clause in 2024, built on a new standard of digital professional diligence, [23] and others have since called for that standard. [24] Placed in the DFA, it would read: [25]

Article 3, new point: ‘Digital professional diligence means not exploiting digital asymmetry and/or digital vulnerability by a trader towards consumers.’
Chapter III, new Article: ‘A digital commercial practice shall be unfair if a) it is contrary to the requirements of digital professional diligence and/or the law, and b) it materially distorts or is likely to materially distort a consumer’s autonomous decision-making.’

Such a clause would catch what the lists miss. [26] The average consumer, which we consider unfit as a benchmark for digital practices, [27] is gone. The test asks whether the trader exploited digital asymmetry or digital vulnerability, a state common to all consumers, [28] and whether the practice distorted a consumer’s autonomous decision-making. Through the reference to ‘the law’, a breach of other digital legislation could also make a practice unfair. [29]

The lists of the DFA need a general clause behind them, and for digital practices Article 5(2), taken over from 2005, is the wrong one.

Conclusion: Which Choices Are Still Open?

The DFA would ban more practices and let the Commission update the bans faster, and we welcome both. The European Parliament and the Council can still change the name and open the whole blacklist to delegated acts. The harder task is a general clause for digital practices with a standard and a measure of its own. Without one, a practice the lists do not cover would fall to Article 5(2) as written in 2005 or, for interfaces, to Article 10(1), which does not say whose behaviour counts.

For now, the answer to the question in the title is yes: the DFA would ban practices without rethinking unfairness itself.

[1] Prof. dr hab. Monika Namysłowska, Head of the Department of European Economic Law, Faculty of Law and Administration, University of Lodz, Poland.
[2] Dr Kasper Drążewski, Centre for IT & IP Law (CiTiP), KU Leuven, Belgium.
[3] Commission, ‘Fitness Check of EU Consumer Law on Digital Fairness’ (Commission Staff Working Document) SWD(2024) 230 final <https://commission.europa.eu/document/707d7404-78e5-4aef-acfa-
82b4cf639f55_en> accessed 12 November 2026.
[4] Commission, ‘Executive Summary of the Fitness Check of EU Consumer Law on Digital Fairness’ (Commission Staff Working Document) SWD(2024) 231 final, 2–3.
[5] Commission, ‘Proposal for a Regulation of the European Parliament and of the Council on promoting fair business-to-consumer commercial practices in the offline and digital single market, and repealing Directive 2005/29/EC (Digital Fairness Act)’ COM(2026) 1100 final (DFA proposal). Articles, recitals and annexes cited without further indication are those of this proposal.

[6] DFA proposal (n 5) explanatory memorandum 2, 5–6; recital 5.
[7] Monika Namysłowska, ‘Future-Proofing the Unfairness Test’ in Natali Helberger and others, Digital Fairness for Consumers (BEUC 2024) 163, 230.
[8] The new points are 11, 12, 14, 33 and 34 in Part I and 53–62 in Part III: see the correlation table in Annex II.
[9] Directive 2005/29/EC of the European Parliament and of the Council of 11 May 2005 concerning unfair business-to-consumer commercial practices in the internal market [2005] OJ L149/22 (UCPD).
[10] Namysłowska (n 7) 224, 227.
[11] Fitness Check (n 3) 38.
[12] To the same effect, Anne-Lise Sibony and others, ‘Assessing the EU Digital Fairness Act and Its Promise of Consumer Protection’ (Tech Policy Press, 11 November 2026) <https://www.techpolicy.press/assessing-the-eu-digital-fairness-act-and-its-promise-of-consumer-protection/> accessed 11 November 2026.
[13] Namysłowska (n 7) 205.

[14] cf UCPD (n 9) arts 2(h) and 5(2).
[15] Namysłowska (n 7) 168, 170, 175–76; see also Monika Namysłowska, ‘The Silent Death of EU Consumer Law and Its Resilient Revival: Reinventing Consumer Protection Against Unfair Digital Commercial Practices’ (2025) 48 Journal of Consumer Policy 317 (‘The Silent Death’).
[16] Fitness Check (n 3) 46, 48.
[17] BEUC, ‘Towards the Digital Fairness Act’ (Position paper BEUC-X-2025-110, 2 December 2025) 3, 7<https://www.beuc.eu/sites/default/files/publications/BEUC-X-2025-
110_Towards_the_Digital_Fairness_Act.pdf> accessed 10 November 2026.
[18] Namysłowska (n 7) 189–90.
[19] Christoph Busch, Amelia Fletcher and Michèle Ledger, Towards an EU Consumer Law Fit for the Digital Age (CERRE 2026) 1 <https://cerre.eu/publications/towards-an-eu-consumer-law-fit-for-the-digital-age/> accessed 10 November 2026.
[20] DFA proposal (n 5) explanatory memorandum 9–10.
[21] Case C-646/22 Compass Banca SpA v Autorità Garante della Concorrenza e del Mercato EU:C:2024:957, operative part, point 1.
[22] cf UCPD (n 9) art 5(3).

[23] Namysłowska (n 7) 217–19; Helberger and others (n 7) 270–71 (proposed arts 2 and 5a UCPD).
[24] John Albert, Marijn Sax and Natali Helberger, ‘Digital Fairness Act: Why We Need an Ambitious DFA to Protect Digital Consumers from Manipulative and Addictive Design Practices’ (DSA Observatory, 19 May 2026) <https://dsa-observatory.eu/2026/05/19/digital-fairness-act-why-we-need-an-ambitious-dfa-to-protect-digital-consumers-from-manipulative-and-addictive-design-practices/> accessed 10 November 2026.
[25] Namysłowska, ‘The Silent Death’ (n 15) 331. The 2024 version added that the distortion ‘causes or is likely to cause harm’ and came with definitions of digital commercial practice, digital asymmetry and digital vulnerability: Helberger and others (n 7) 270–71.
[26] Our earlier digital fairness advocacy work discussed the clause in the context of proposed amendments to the existing Directive. In the context of the existing DFA proposal, the formulation should also carry the weight of including the relevant consumer benchmarks.
[27] Namysłowska, ‘The Silent Death’ (n 15) 322.
[28] Helberger and others (n 7) 271 (‘a universal state of susceptibility’).
[29] Namysłowska (n 7) 221–22.