Did IMCO Forget That Consumer Protection Is in Its DNA?
The IMCO draft opinion on the Digital Networks Act (DNA), reported by Euractiv and circulated after a Politico leak, takes the Commission’s already-imperfect proposal and systematically worsens it on net neutrality across three dimensions simultaneously. The central move — making the conciliation mechanism mandatory — does what the Commission explicitly declined to do: it builds a structural pathway for network fees directly into EU telecoms law.
In parallel, a joint statement signed by almost 40 organisations — led by BEUC, Internet Society and epicenter.works— calls on EU co-legislators to remove Articles 191 to 193 of the DNA entirely and to preserve the Open Internet Regulation as a standalone legal instrument. As we reported in our Hot item, the breadth of that coalition matters: it spans consumer groups, civil liberties organisations, infrastructure operators, media associations, libraries, and gaming developers. IMCO’s mandate covers consumer protection alongside the internal market. The organisations raising the alarm on 9 June are, in large part, the constituency that mandate exists to serve.
Voluntary to Mandatory: The Core Move
The Commission’s original DNA proposal built a “voluntary conciliation facility” into Articles 191 to 193. In a January blog post, we explained why that label was misleading: a mechanism that permits one party to summon another before a regulator is not voluntary in any meaningful sense, regardless of what the recitals say. The threat of initiation is itself the instrument of leverage.
The IMCO rapporteur, Stéphanie Yon-Courtin, at least has the honesty to remove the label and not even try to hide what the intention behind this superfluous mechanism. Across at least ten amendments, the word “voluntary” is deleted and replaced with “mandatory.” The conciliatory meeting becomes mandatory (Amendment 4). The facility becomes a “facility for mandatory conciliation” (Amendment 64). BEREC’s reporting obligation, the review timeline, and the institutional architecture all follow the same substitution. Amendment 71 adds enforcement teeth: where a conciliatory meeting produces no agreement, the national regulatory authority “shall have the power to impose binding measures to resolve the dispute between the parties.”
That is not conciliation. That is arbitration without the procedural safeguards that arbitration normally requires. And arbitration over IP interconnection arrangements, supervised by telecoms regulators who answer to governments with distinct views on how value should flow in the digital economy, is a managed market in everything but name.
The Commission rejected mandatory contributions and network fees after its own 2024 White Paper process concluded they were legally risky, economically distortive, and incompatible with competition law. IMCO is now asking Parliament to write the rules the Commission refused to write. The difference is structural rather than cosmetic: mandatory conciliation with binding resolution powers does not create a “fair share” payment directly, but it creates the regulatory infrastructure through which one becomes almost inevitable over time.
Experts, including BEREC, have noted consistently that no market failure has been identified that would justify such intervention. The current IP interconnection landscape operates largely without contracts, on dynamically negotiated terms, at standards understood across the industry. The European internet works precisely because that arrangement is not subjected to regulatory adjudication. Introducing mandatory conciliation with binding outcomes would produce the opposite of stability: rigid, contested arrangements, an incentive for dominant telcos to manufacture disputes, and routing decisions ultimately made by telecoms regulators rather than engineers.
A New Obligation on the Wrong Side of the Pipe
The mandatory conciliation shift has attracted most of the external commentary. A second set of amendments deserves equal scrutiny because the analytical problem it creates is more fundamental.
Amendment 36 inserts a new obligation into Article 93, the open internet access article, directed not at internet access service providers but at “all providers of content, applications, and services that affect traffic management.” The proposed text states that such providers “must not interfere with, degrade, restrict, or discriminate against open access to the internet.”
Net neutrality, as it has existed in EU law since the Open Internet Regulation came into force in 2016, is a constraint on network operators. The principle is structurally asymmetric: ISPs sit between users and the rest of the internet and therefore hold the technical capacity to discriminate between traffic. Content providers do not. A streaming platform cannot, by definition, degrade the network between its servers and a user’s screen; only the network operator carrying that traffic can.
Extending the non-discrimination obligation to content providers is not a strengthening of net neutrality. It is a reclassification of who bears the burden. The practical effect would be to bring platforms, cloud providers, CDN operators, public broadcasters, universities, and any other entity that “affects traffic management” into scope of telecoms regulatory obligations, without any framework for what that supervision looks like, who enforces it, or what conduct it is actually designed to prevent.
BEREC has spent a decade developing guidance on what discriminatory traffic management by ISPs means in practice. No equivalent analytical framework exists for content providers, because the problem it would address does not exist. Inserting this obligation into Article 93 does not fill a gap. It creates a new category of regulated entity without a methodology, and without the market failure that would justify the intervention.
Amendment 38 adds further complexity to the traffic management exception in Article 93(3), permitting differentiation “among specific categories of traffic subject to different technical requirements as regards the level of quality of service.” The OIR’s traffic management exception already allows operators to manage traffic for objectively different technical quality requirements. The IMCO language appears to go further by permitting differentiation by traffic category rather than by technical characteristic, which shifts the boundary in the direction that commercial differentiation arguments have long sought to exploit.
The Institutional Risk Is Cumulative
Each of these three elements would be concerning individually. Together, they reconfigure the legislative architecture of the open internet at EU level in a way that is difficult to reverse once embedded in a regulation.
The joint statement published on 9 June identifies a fourth dimension that the IMCO amendments do nothing to address: the DNA as a whole removes 18 of the 19 recitals of the Open Internet Regulation, stripping out the interpretative guidance that has been central to both BEREC guidelines and CJEU case law since the OIR came into force. By absorbing the OIR’s substantive provisions into a broader regulation while discarding nearly all of their interpretative context, the DNA places net neutrality alongside regulatory objectives, such as network performance, resilience, and ecosystem cooperation, that can be used to condition or override it. The open letter signatories recommend that the OIR remain a separate legal instrument. The IMCO rapporteur’s amendments push in the opposite direction.
Mandatory conciliation with binding enforcement powers creates a forum in which network fees can emerge through case-by-case decisions rather than through explicit legislative choice. Amendment 36 creates a new class of regulated actor with obligations but no corresponding rights or procedural framework. Amendment 38 loosens the traffic management exception at the margin. None of these changes strikes directly at the principles in Article 93. They work on the surrounding architecture instead, which is precisely how the DNA’s risks were described in the earlier posts in this series: the principles are not the problem; the governance around them is.
The three-year review clause, which IMCO shortens from 36 to 24 months, provides the next leverage point. A mandatory conciliation mechanism that has produced contested decisions and bilateral arrangements under regulatory pressure creates exactly the paper trail that can be used in 2028 to argue for even harder obligations. This is how soft-law mechanisms harden across legislative cycles.
The Internet Society identified this dynamic clearly in response to the original DNA proposal. The IMCO amendments accelerate the timeline.
The Available Course Correction
Parliament’s IMCO committee is not the final word on this legislation, and the ITRE committee, which leads on the DNA, will produce its own report. What the rapporteur’s draft does carry is a question it cannot easily sidestep: BEUC, the organisation whose mandate most directly mirrors IMCO’s own, signed the 9 June joint statement calling for Articles 191 to 193 to be removed entirely. A committee rapporteur tasked with consumer protection who proceeds in the opposite direction of the European Consumer Organisation has some explaining to do.
The minimum required correction is straightforward. Amendments 4, 56, 64, 65, 66, 67, 71, 72, and 73 should be rejected and the voluntary framing of the original Commission text removed, on the basis that no market failure has been established that justifies mandatory arbitration over IP interconnection. Amendment 36 should be struck from the text entirely: net neutrality obligations belong on network operators, not on the services those operators carry. Amendment 38 should revert to the Commission text to avoid widening the traffic management exception beyond what a decade of BEREC guidance has established as the workable limit.
The Commission’s DNA was already a legislative vehicle that required careful handling on open internet questions, as this series has argued since January. The IMCO rapporteur’s draft does not handle it carefully. The Parliament as a whole should.
Written by Caroline De Cock, LL.M., Head of Research
