One Step Forward, Two Steps Back: Why the Draft Merger Guidelines Matter for Europe’s Economy and Its Democracies

Two civil society organizations assess the European Commission’s draft Merger Guidelines.

Merger control exists to keep market power in check. The Commission's draft Guidelines equip it to do that, then hand merging parties several new arguments for clearance, among them a favorable framing of scale, a theory of benefit and an innovation shield.

The evidence the Commission itself commissioned does not make the case for any of them. One step forward, two steps back, at a moment when Europe needs merger control to reverse the merger concentration trend rather than ease it.

First Published: CPI Antitrust Chronicle · October 2026 | By Hilary Jennings & Claire Lavin¹

I. INTRODUCTION

The European Commission is overhauling its framework for the assessment of mergers, which is the most significant change to European merger control since the 2004 Merger Regulation was introduced. This follows two decades during which the economic landscape has shifted significantly.

Concentration has risen significantly in many European markets, and a small number of digital firms now control services on which businesses, citizens and public bodies depend. Markups have increased at a time of acute pressure on household budgets, while dependence on a few suppliers of critical inputs has exposed the fragility of supply chains. Much of the news media has seen its business model hollowed out by platforms that sit between publishers and their audiences.

Our two organizations welcomed the review for these reasons. The Horizontal and Non-Horizontal Merger Guidelines of 2004 and 2008 were written before ecosystems, data-driven network effects and serial acquisitions of nascent rivals became central to competition in large parts of the economy and at a time where our economies were not in the hands of a few powerful corporations. An update gives the Commission the opportunity to use merger control to stop this concentration trend.

The Merger Regulation leaves little doubt about the purpose of the exercise. Article 2(3) directs the Commission to block concentrations that would significantly impede effective competition. In other words, it aims to keep market power in check, ensure markets remain open and dynamic, and foster innovation. In turn, this should help the digital world and media sphere to remain open and diverse and therefore preserve our democracies.

Article 2(1)(b) of the EU Merger Regulation allows technical and economic progress into the assessment only where it is to consumers’ advantage and does not form an obstacle to competition.

We would add that the stakes run wider than the price and output effects on which analysis has traditionally been centered. Concentrated control of communications platforms and media shapes public debate; concentrated buyer power depresses wages and the incomes of suppliers; concentrated supply of strategic inputs leaves the economy exposed and weak when shocks come.

Judged against that purpose, the draft published on 30 April 2026 is a mixed document. It takes one step forward, with its treatment of theories of harm, market power and dynamic competition that would equip the Commission far better than the current guidelines.

But alongside these improvements, the draft introduces a “theory of benefit,” an explicitly favorable stance towards mergers that create scale, and an innovation shield for certain acquisitions of innovative firms. These are new policy choices, and the ex-post evidence does not support them.

Taken together they risk cancelling out much of what the draft gains elsewhere, two steps back against one step forward. The two economic studies the Commission commissioned for this review, both published in the past month, simply support that concern.²

II. THE RECORD AND THE EVIDENCE

The political case for relaxing merger control rests on the claim that the Commission has been too interventionist and that this has held back European growth. The record does not bear this out.

Between 2015 and 2024 the Commission reviewed 2,833 notified mergers and prohibited nine, and around 5 percent were cleared with remedies. Natalia Fabra, Massimo Motta & Martin Peitz have observed that with intervention this rare, it is difficult to argue that merger policy has been an obstacle to EU productivity or growth.³

Enforcement by national authorities has arguably been weaker still. DG Competition’s own research documented rising concentration and markups.⁴

Anyone arguing that the Commission should recalibrate in favor of mergers needs to show that it has been blocking or deterring beneficial transactions in significant numbers. We have seen no such evidence.

The Commission has also tested the claim against its own caseload. In September DG Competition published an ex-post evaluation by Joel Stiebale & Florian Szücs of 3,825 mergers the Commission reviewed and cleared between 1990 and 2024, all of them cases at the enforcement margin.⁵

In the five years after clearance, patenting fell by about 18 percent at the merging firms and by about the same at the rivals named in the Commission’s own decisions, while markups rose by around 4 percent and 3 percent respectively.

The findings on rivals are the hardest to dispute. A merger that cuts duplicated research gives a firm outside the deal no reason to innovate less or to widen its margins. Here rivals did both, which points to weaker competition.

The study also found substantial losses in the sectors that European industrial policy treats as strategic. Rivals’ innovation fell by around two-thirds in advanced semiconductors and by around half in clean technologies. The merging firms’ own innovation fell by about a third in pharmaceuticals.

The authors conclude that their results provide “little support for the view that EU merger enforcement has systematically been too stringent,” and little support for a general presumption that mergers in strategically important sectors raise innovation. If anything, they suggest, enforcement has been too lenient.

Patents are an incomplete measure of innovation and the estimates are averages. Because outcomes are measured at firm level, however, the effects in the markets actually affected are probably larger than those reported.

Oxera’s study for the Commission, drawing on more than 450 publications, finds that the market power a merger creates in future markets is “not a suitable basis for an ‘innovation defense.” That study judges mergers only by their effect on consumers, which is the narrowest test available. Even on that test, these Commission advisers give no support for the additional latitude the draft goes on to offer.

The gaps in the current framework lie elsewhere. The Commission needed tools capable of capturing market power and harms that sit awkwardly within traditional methodologies. These include the leverage that comes with control of a digital ecosystem, the advantages conferred by data, markets prone to tipping, the elimination of nascent competitors, and harm that accumulates through a series of acquisitions that each look unremarkable.

It also needed a framework that looks beyond a narrow consumer welfare test to effects on workers, suppliers, resilience and media plurality.

The allocation of the burden of proof should reflect the fact that merging parties hold almost all the relevant information about their own plans, costs and pipelines. Parties therefore have the ability (and likely incentive) to present evidence in a way that supports their case.

III. WHERE THE DRAFT MOVES FORWARD

On theories of harm, the draft represents a substantial advance. It requires the Commission to articulate and substantiate a theory of harm in each case, and it recognizes harms the current text handles poorly, namely the loss of investment and innovation competition, the elimination of potential competitors, entrenchment of a dominant position, ecosystem and portfolio effects, and the effects of minority and common shareholdings.

Several of these reflect the Commission’s recent practice, including Booking/eTraveli.⁶ Codifying them gives case teams a footing they lacked under the 2004 and 2008 texts.

One of the draft’s strongest features is how much wider it goes on what counts as harm. Its definition of a theory of harm names innovation, privacy, sustainability and resilience among the parameters on which a merger can damage competition, and buyer power over workers is treated as a harm in its own right.

It also recognizes that merger control helps maintain the balance of power on which democratic societies depend, and that limiting concentration in media markets protects diversity and plurality of information. Reasoning of that kind has not appeared in EU merger guidance before. This is part of the draft we would most want to see in the final version.

However, recognition is not the same as a workable test. Resilience, sustainability and plurality can already be advanced by merging parties as reasons to clear a deal, and they need the same development as grounds for finding harm.

The draft is also better on how that power is measured. It accepts that market shares can understate power in dynamic markets. Where concentration is high, it comes close to a structural presumption: the parties must bring substantial evidence that harm is unlikely.

Paragraph 35 states that the more market power the merged entity holds, the less likely it is that efficiencies can outweigh the harm. A merger approaching monopoly will almost never be saved by them. We regard this principle as essential and would resist any weakening of it in the final text.

The introduction of dynamic competition is also particularly positive. It allows the Commission to assess how competition is likely to develop over a realistic horizon, as well as the market’s current state. In technology and life sciences, the rivalry that matters most is often competition to invest, innovate and enter.

A dynamic framework also strengthens the analysis of non-horizontal mergers, where concerns have too often been set aside as speculative. The economic study prepared for the Commission by Oxera sets out, with support from the literature, how a vertical or conglomerate merger can weaken rivals’ incentives to invest.⁷

The merged firm can hold up rival investment, raise rivals’ costs through refusal to deal or degraded interoperability, or deter entry that would have to occur at several levels at once. It can also leverage a secondary market to protect profits in its core market or acquire an input that a smaller rival needs to catch up. The study notes that some of these mechanisms carry no offsetting efficiency at all.

These updated principles and methodologies will only make merger control effective if they are operationalized. In CK Telecoms, the Court of Justice confirmed that the Commission must establish, on a sufficiently cogent and consistent body of evidence, that it is more likely than not that the concentration would significantly impede effective competition.⁸ Theories of harm that the Guidelines describe but do not operationalize will struggle to meet that standard.

The final text should explain, for each new theory of harm, what evidence the Commission will look for and how it will be weighed. That means covering overlaps in pipelines and research capabilities, the innovative capacity of rivals, internal documents on R&D priorities, the acquirer’s acquisition history and the indicators of ecosystem power.

IV. WHERE THE DRAFT MOVES BACK

Set against the evidence from the Commission’s studies and report, the draft gives us cause for concern. The first step back, and most concerning, is scale. Scale arguments are not new (and the Commission has rejected them before, in Siemens/Alstrom among other cases).⁹

What is new is that the draft treats scale as a positive in its own right, set out before the harms of concentration and independent of the conditions that efficiency claims have to meet, The draft states that the Commission “regards positively mergers that increase procompetitive scale,” and sets out the circumstances in which such mergers may contribute to competitiveness. It says very little about when they harm competition. Merging parties will read that imbalance as an invitation to present consolidation as industrial policy.

Scale does not have to come through acquisition. Firms can expand organically, form joint ventures, license technology, share infrastructure, enter distribution agreements or raise outside capital.

The Draghi Report itself identifies fragmented capital markets as a structural cause of Europe’s difficulty in scaling up firms. The European Commission President’s own State of the Union address in September 2026 also located the problem in fragmentation: “we still fragment our capital, networks and purchasing power.”¹⁰

Those barriers are for the single market and the savings and investments union to remove. Loosening the treatment of consolidation in merger guidelines does nothing to remove them.

The evidence on strategic sectors cited above sits badly with the premise that consolidation is the route to innovation at scale. Telecoms, the sector most often cited in support of scale, tells the same story: in-market consolidation has tended to raise prices without delivering the promised gains in investment or quality.¹¹

Scale achieved by removing a rival is different from scale achieved by growth on the merits, and competitiveness is served by contestable markets in which many firms can scale.

No evidence supports the use of benefits and scale as Trojan horses to serve corporate interests to the detriment of consumers, workers, innovation and resilience. The final text should remove the priority given to scale.

The second concern is efficiencies. The draft asks merging parties to put forward a “theory of benefit.” Paragraph 36 states that a preliminary finding of harm is not a condition for efficiency claims to be submitted and considered.

For more than twenty years the Commission has first established whether a merger is likely to harm competition, and only then asked whether verified efficiencies offset that harm.

The draft moves towards a balancing exercise in which claimed benefits can shape the analysis from the outset. It also gives the parties’ narrative a formal status equivalent to the Commission’s own case. Greater emphasis would be placed on efficiencies, leaving more room for businesses to see efficiency claims succeed even when cases are problematic.

Efficiency claims are prospective assertions, prepared by the merging parties and advisers whose objective is clearance. As forward-looking claims, they are inherently prone to uncertainty. The Commission has limited means to test them, and nobody checks whether they materialize once the deal has closed.

The resources required to build an elaborate theory of benefit are greatest precisely where market power is greatest. The Merger Regulation itself speaks of taking into account efficiencies that are substantiated and likely, put forward by the undertakings concerned.

Ex-post evidence gives little reason to extend more credit to such claims.

In a submission to the merger Guidelines consultation, Annika Stöhr and Tommaso Valletti reviewed 41 studies and found that the evidence weakens at each stage from post-merger profitability to merger-specific cost savings to benefits passed on to consumers. They identify no systematic evidence of all three in mergers scrutinized by competition authorities.¹² The rival results set out above support the same conclusion.

These concerns apply with most force to dynamic efficiencies, where the promised benefits lie years ahead and are least verifiable. The Oxera study accepts that dynamic analysis requires longer time horizons. It insists, however, that claimed benefits be held to the same standards of evidence, probability and discounting as the harm. It treats investment plans prepared in contemplation of a merger review as having low probative value.

It also warns that closing overlapping research programs reduces innovative effort and should not be counted as a synergy. The draft’s extensive guidance on long-term efficiencies contains none of these safeguards.

We would retain the existing cumulative conditions of verifiability, merger specificity and pass-on to consumers, with the burden on the parties and the bar rising with market power in line with paragraph 35 of the draft. Oxera considers that framework adequate for dynamic effects. A separate theory of benefit adds nothing to it except another route to clearance for problematic transactions.

The third is the innovation shield. Paragraph 192 of the draft creates a safe harbor under which the Commission will not, in principle, find a significant impediment to effective competition in the acquisition of an innovative firm or start-up. The conditions turn largely on market definition and the number of remaining competitors. This is an innovation defense recast as a safe harbor.

A safe harbor looks at the market as it stands, and that is the difficulty. The harm from buying a young innovator lies in the loss of competition in the future, which a test built on today’s market shares and today’s count of competitors would not pick up.

Some of the acquisitions that most warrant scrutiny would fall inside the shield. The Stiebale & Szücs study found that innovation at an acquired firm roughly halves after the deal, which is consistent with the literature on killer acquisitions.¹³

The carve-out is too narrow as well. It is closed to the largest firm in a market and to designated gatekeepers, leaving out firms whose power lies across an ecosystem rather than in any single market.

As it is, Big Tech or Big Pharma could exploit this mechanism and keep on acquiring small businesses with a view to killing competition.

Because each deal is judged on its own, a firm can buy its way across a sector without any single transaction raising a concern, and since the Court of Justice’s ruling in Illumina many of these deals are not reviewed at all.¹⁴

We would remove the shield. The draft’s own tests for potential and nascent competition ask the right questions, and the Commission should add a screen for cumulative acquisitions so that a run of deals by the same buyer is judged on what it adds up to.

V. GETTING THE FINAL GUIDELINES RIGHT

The Commission deserves credit for the way it has run this review. Two rounds of public consultation, seven topic papers and a series of stakeholder workshops and conferences have opened the process to a far wider range of voices than earlier revisions, and civil society organizations, our own among them, have welcomed the opportunity to take part.

What matters now is how those contributions are weighed. Business associations, merging parties and the law and economics firms that advise them account for a large share of the responses, and they have far greater resources to engage than civil society, trade unions or smaller firms.

The final text should follow the weight of the evidence. The number of submissions from those with a commercial interest in a more permissive regime should count for little.

Nor should the Guidelines become a political football, and the pressure to make them one is considerable. The European Commission President has called for an environment that will produce ““Europe’s next champions.”¹⁵

The new Director-General for Competition, Anthony Whelan, told the Financial Times that “the opening to the importance of pro-competitive scale can enable the creation of champions, including European champions,” and that the Commission was “consciously making some explicit and hopefully clear signals” that it would seriously consider such arguments. He did not know, he said, how many transactions were “being dusted off from a filing cabinet.”¹⁶

That is the effect we fear the scale language will have: deals once judged unlikely to clear, revived because the rules seem to have moved.

Mr. Whelan also described the “seriousness and neutrality” with which the Commission applies its rules as “a very precious asset.” We agree, and that neutrality is difficult to reconcile with guidance intended to produce champions.

The capitals do not speak with one voice either: Finland, Ireland, the Czech Republic, Estonia, Romania and Latvia have warned that “size in itself should not be the primary objective.”¹⁷

Merger guidelines create a legitimate expectation that the Commission will follow them, and decisions taken under them must be defended before the Union courts. The political mood of a single year is a poor foundation for rules that may stand for another twenty.

Europe is at a crossroads. If the final Guidelines keep wording that helps dominant companies argue their way past clear competition concerns, the Commission would be siding with the interests of powerful corporations and a few governments, against consumers, smaller competitors and workers.

The Commission’s own studies say otherwise. Concentrated corporate power is already eroding democratic life across Europe, and dominant firms have both the means and the incentive to resist regulation and contest enforcement.

Merger control is a powerful ex-ante tool. If used effectively, it can prevent the rise and entrenchment of monopolies, rather than leaving competition law and regulation to deal with them afterwards.

For the final text, we would ask the Commission to keep and operationalize its improvements on theories of harm, market power, dynamic competition and the broader parameters of competition, including media plurality, supply chain resilience and labor markets.

It should strip the theory of benefit of its separate standing and hold efficiencies to the existing cumulative conditions. The priority given to scale should go, and so should the innovation shield. A screen for cumulative acquisitions should be added in place of the latter.

In certain sectors where companies hold excessive market power, such as tech and pharma, the Commission should view mergers holistically instead of reviewing them in silos. It should also press for call-in powers in each Member State, so that harmful transactions below the thresholds do not escape scrutiny.

Guidance alone will not be enough. The Commission must make full use of its enforcement powers in mergers. That means reviewing all potential harms on sound evidence, preferring structural remedies to behavioral ones, and prohibiting deals where remedies cannot address the harm. It should be stricter with firms that already hold substantial market power.

It should also commit to a standing program of ex-post evaluation, so that the next revision of these Guidelines rests on evidence from the outset. Tougher merger enforcement is now more important than ever to preserve our democracies and economies.

Notes

1 Respectively: Senior Fellow, Balanced Economy Project / Research Fellow, Open Markets Institute Europe.

2 Prof. Dr. Joel Stiebale & Prof. Dr. Florian Szücs, Study on the impact of mergers reviewed by the European Commission on innovation and markups An empirical investigation (September 2026); Oxera et al., Economic study on the dynamic effects of mergers, report for the European Commission (August 2026).

3 Natalia Fabra, Massimo Motta & Martin Peitz, How to update the EU merger guidelines, CEPR (21 August 2025), https://cepr.org/voxeu/columns/how-update-eu-merger-guidelines.

4 European Commission, Protecting competition in a changing world (2024).

5 Stiebale & Szücs, Study on the impact of mergers reviewed by the European Commission on innovation and markups, report for the European Commission (September 2026). Innovation is measured by patents weighted by the citations they receive.

6 Case M.10615 Bookings Holdings / eTraveli Group, 25 September 2026. The decision was upheld by European courts (General Court, T-1139/23, Booking Holdings v Commission, 9 September 2026).

7 Oxera et al., Economic study on the dynamic effects of mergers, report for the European Commission (August 2026).

8 Case C-376/20 P, Commission v CK Telecoms UK Investments, 13 July 2023.

9 Case M.8677 Siemens/Alstom, 6 February 2019.

10 Ursula von der Leyen, State of the Union Address, Strasbourg, 16 September 2026.

11 Christos Genakos, Tommaso Valletti & Frank Verboven, “Evaluating market consolidation in mobile communications,” Economic Policy 33 (2018); Lear et al. (2024), as reported in Oxera et al., Volume 1, paragraph 7.21.

12 Annika Stöhr & Tommaso Valletti, The Empirical Record on Merger Efficiencies: Evidence, Limitations, and Implications for the Assessment Framework, Submission to the European Commission public consultation on the Draft Merger Guidelines, 26 June 2026.

13 Colleen Cunningham, Florian Ederer & Song Ma, “Killer Acquisitions,” Journal of Political Economy 129 (2021).

14 Joined Cases C-611/22 P and C-625/22 P, Illumina v Commission, 3 September 2024.

15 European Commission, Press Release, Commission opens consultation on draft of new Merger Guidelines, Brussels, 30 April 2026.

16 B. Moens, “EU opens door to bigger corporate mergers,” Financial Times, 8 September 2026.

17 Non-paper from the Czech Republic, Estonia, Finland, Ireland, Latvia, Romania and Slovenia, Effective Competition Policy is a Cornerstone of well-functioning, resilient and competitive Single Market, meeting of the Competitiveness Council, 26 February 2026.