Reflections on the European Commission’s draft Merger Guidelines: Where should we focus?

After the initial noise, where should attention focus as the European Commission consults on its draft Merger Guidelines? [1] Drawing on their experience before the Commission, Jorge, Neil, Martina, Cecilia, and Roman share their reflections on the ambition in the draft Guidelines – and, critically, how to ensure that it works in practice.

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.

What are your first impressions of the draft Guidelines?

Guidelines are just that: guidance. But they signal the institutional priorities and process that shape outcomes. So, the Commission’s draft matters a lot. It is ambitious, and it indicates a genuine desire to improve the merger control process. Whether that leads to an evolution or revolution in practice remains to be seen – but there are three encouraging elements.

Introducing a formal “Theory of Benefit”: taking scrutiny of the business case seriously 

In Europe, not just at the Commission, merger control has been underpinned by an assumption that all transactions are either bad for competition, or neutral. Never beneficial. That was the implicit starting point, woven into the merger assessment process itself.[2]

A regulator should be sceptical; that’s its responsibility. But, in practice, we had a bad equilibrium. Parties knew that their commercial rationale for a transaction would struggle to carry weight, so they often did not invest enough time in presenting it rigorously. In some cases, parties chose not to emphasise synergies at all, concerned that doing so might appear defensive or invite greater suspicion. Of course, that made the Commission’s scepticism self-fulfilling.

Contrast that with other regimes. In the U.S., and more recently in the CMA’s assessment of Vodafone/3, the business case was central: to understanding the commercial rationale, and to the analysis. That does not imply credulity. The detailed scrutiny applied in that case was impressive. But it does ensure that both the positive and negative effects of a transaction are properly tested and weighed, rather than filtered asymmetrically.

The draft Guidelines signal a genuine attempt to recalibrate. The current Guidelines already have the words needed to accommodate efficiencies, at least in principle. The issue has been cultural and operational: efficiencies sit at the margins of people’s minds, and the process. In that light, placing a “Theory of Benefit” alongside the “Theory of Harm” offers more than a branding exercise. The symmetry in terminology presents the opportunity for a meaningful shift in mindsets – from treating efficiencies as a defensive add-on, to recognising them as an integral part of the competitive assessment.

A forward-looking assessment: taking dynamic effects seriously

Currently, European merger control often defaults to a static analytic framework; both in the evidence it relies upon and, more importantly, in the counterfactual against which transactions are assessed.

The Commission has, at times, sought to engage with a forward-looking assessment. But even in its much-discussed headline cases – Dow/DuPont and Bayer/Monsanto – the analytical framework was essentially static in its application. The focus on R&D overlaps captured observable competition within existing projects; it said much less about the underlying forces that drove innovation, or how those forces might evolve over time.

Against that background, the draft Guidelines signal a real desire at the Commission to take dynamic effects seriously. It doesn’t pretend to have all the answers. But it’s asking the right questions, which is a crucial step forward.

First, their emphasis on companies’ competitive and complementary capabilities is notable. Not only because it appears repeatedly; its prominence is telling. This reflects a more fundamental point: competition is driven by companies’ capabilities. Overlaps are just an effect. Static proxies will still have a role, but they offer only a limited and restrictive way to analyse dynamic effects. You see people suggest applying the approach in digital sectors, but it doesn’t work in markets with less structured innovation pipelines. It doesn’t capture the relevant factors.

Second, the dedicated section on the counterfactual is important. It explicitly recognises that “pre-merger conditions may not be reflective of normal or foreseeable future market conditions – and therefore may not constitute the relevant counterfactual”.[3] That is an obvious point, but not one that always bears weight in current practice. There is an understandable tendency – for both authorities and parties – to ‘pragmatically’ rely on what is observable and measurable, even when it addresses the wrong question. Here, the draft Guidelines specify the right questions.

The “innovation shield” will attract attention, but it’s a welcome recalibration following Illumina/Grail. Concerns about potential “killer acquisitions” expanded well beyond the factual and economic basis that initially motivated them. The original academic study identified a real but narrow risk.[4] Its application in merger control, at times, extended beyond that domain, particularly in digital and technology sectors where complementarities are central and the evidentiary basis is less clear.

Broadly, the draft Guidelines might disappoint anyone that hoped for an instruction manual on assessing dynamic effects. But that’s not a realistic expectation. They do, however, provide a structure and starting point for developing a more disciplined and serious assessment of dynamic effects. It is also a challenge. Parties will need to engage earlier, and with greater rigour, to demonstrate how their markets are likely to evolve, not simply where they stand today. 

Introducing taxonomies, structuring the issues without constraining them

There’s a clear signal that the Commission is seeking to avoid “cheap talk”, in general and especially on efficiencies.

The effort to introduce taxonomies that impose structure on the issues and evidence is admirable. The draft Guidelines could have stopped at high-level headings, but the granularity of its taxonomies – such as the 7 direct synergies and 8 dynamic efficiencies – is welcome.[5] People like us have been calling for that, and many of the initial reactions focused on it.

The more important point, however, is how the draft frames these taxonomies. They are not presented as exhaustive checklists or prescriptive templates. The draft Guidelines make clear that these categories are illustrative and non-exhaustive – designed to help parties structure and evidence their arguments, not to foreclose lines of reasoning that fall outside a predefined list.[6]

That distinction matters. Structure and rules can help an assessment: it’s more predictable for the notifying parties, helping them prepare; and it formalises the “quick wins”, focussing analysis where it is needed. But they can be too restrictive. Ultimately, it’s the outcomes that matter, and simplistic checklists break down in the real world. The Commission seems to have made a genuine attempt to get the balance between clear principles and a clear process right. Contrast that with the formalism in the DMA and the draft Article 102 guidelines.[7] It has not adopted the same approach here, and that must be commended.

Whether that balance holds in practice will depend less on the text than on how it is applied. A taxonomy can easily harden into a checklist if treated mechanistically. The flexibility that the draft Guidelines provide could also become meaningless if its application is completely unconstrained. Establishing that balance in practice will, frankly, require good faith and on-going discipline from all stakeholders: the Commission, parties, and practitioners.

Do you have any concerns?

Guidelines are just guidelines. What matters is how the Commission – and others – apply them in practice. There are three areas that temper optimism. 

Maintaining focus: a broad set of objectives can lead to no objectives

The breadth of ambition in the draft Guidelines is one of their most compelling features, and a potential source of difficulty in practice.

The Commission recognises a wide set of issues and competitive parameters. Some are front and centre: scale, innovation, and resilience. But other non-price parameters of competition have increased prominence: diversity of choice, capacity, investment, privacy, and sustainability among others. That’s good. Before publication, a lot of outsiders were concerned that the draft Guidelines may either be led by the political context that preceded them, or deaf to it. The draft navigates that well, demonstrating what the Commission means when it says it is mindful of issues that are “politically relevant, not politically motivated”.

The challenge lies in operationalising them.

The first issue is that, for several of these parameters, it is not yet clear how they will be assessed in a merger control context. Developing workable methodologies will require sustained engagement between the Commission and practitioners. In some areas, that process is already advancing. Resilience, for example, is being explored in some of our current cases before the Commission. In others, the learning curve is steeper. Sustainability is a good illustration: while authorities such as the Dutch ACM have done pioneering work, it is a leap to translate that into merger control, as envisaged in the draft Guidelines.

The second, and more fundamental, issue is how competing objectives will be reconciled. Expanding the set of relevant considerations inevitably raises the question of trade-offs. Once many things can matter, there’s a risk that nothing matters. The Commission may acknowledge that scale, and resilience, and sustainability are important, but will any of them ever be important enough to change a decision? That’s the litmus test. Efficiencies are recognised in principle in the current guidelines, but in practice the hurdles to getting them accepted have often seemed impossibly high – as this new draft implicitly accepts.

From experience, this is not a trivial challenge. Assessing trade-offs requires a degree of specificity in the harms and benefits being balanced, and a willingness to accept uncertainty when doing so. As the draft Guidelines discuss, it also involves comparing effects that are not readily commensurable: price and non-price dimensions, short-term and long-term impacts, effects across different customer groups.[8] Expanding the range of relevant factors only increases that complexity.

There is a risk, therefore, that a wide focus leads to a lack of focus. If many considerations are relevant but they are assessed vaguely, “in the round”, then it becomes an impenetrable fog that the Guidelines can’t guide us through.

Disciplined discretion: an essential but risky feature in any guidelines

Guidelines leave a regulator with significant room for interpretation, and reinterpretation. The U.S. 2023 Merger Guidelines illustrate the point. Initially seen as a statement of intent under a more interventionist FTC, they have since been retained under new leadership, with a markedly different tone and set of priorities. The text remains; its application differs. The margin for manoeuvre is substantial.

The freedom for regulatory discretion is greater in Europe. Unlike in the U.S., where courts are embedded in the process, European merger control operates under a paler shadow of judicial oversight. Appeals are rare, and many transactions are abandoned even before a formal prohibition decision is reached. In practice, this gives the Commission considerable autonomy – not only in the conclusions it reaches, but in how it frames the issues, selects the evidence, and weighs competing arguments.

Discretion is inevitable and even desirable to an extent. No set of guidelines can anticipate every market context or theory of harm, nor should it attempt to. There is always a balance to be struck between clarity of rules and flexibility of principles.

The concern arises where that discretion becomes unconstrained. It is notable that most of the references in the draft Guidelines to a “margin of discretion” relate to evidentiary assessment. Flexibility in handling evidence is important. But if it translates into vague or shifting standards – where the relevance or weight of evidence becomes difficult to predict – the process risks losing coherence. At that point, the guidance ceases to guide.

The issue, then, is not discretion per se, but discipline: whether the framework meaningfully constrains how discretion will be deployed in practice. 

An expanded scope for adequate remedies, the increased importance of planning early

The draft Guidelines focus on assessment. But that inevitably shifts pressure onto what follows: how outcomes will be secured after the merger. Remedies have always been prominent in the Commission’s decisions; the ratio of transactions cleared with remedies compared with fatalities, that are prohibited or abandoned, is relatively high (Figure 1). However, the draft Guidelines will have implications for their role, and what constitutes an adequate remedy.

The CMA’s decision in Vodafone/3 is instructive. It marked a willingness to engage seriously with efficiency claims. But it also highlights the conditions under which that is feasible. The case involved a relatively well-understood business rationale, in a sector with a strong and engaged regulator, capable of both informing the assessment and supporting an effective post-merger monitoring framework.

Those conditions will not always be present.

In many sectors where dynamic effects and efficiencies are most relevant, the institutional environment isn’t comparable. That raises a practical challenge: it will not be sufficient to demonstrate that efficiencies are plausible. Parties will need to show, early and credibly, that the merged entity can “stick the landing” – and that delivery can be observed and, where necessary, enforced.

This has practical implications for remedies. Obviously, non-structural remedies will play a prominent role, as we saw in the UK. But the assessment of dynamic effects might also affect the shape of adequate structural remedies, compared with what would have been required after a more static assessment. A “wait and see” approach – developing remedies reactively, once concerns crystallise – will be less viable where the assessment itself is more forward-looking and contingent on delivery.

Figure 1: Significant merger outcomes, 2004-2025

^{Notes: (1) “1 – Fatalities” includes all cases under Art 8.3 and Art 8.4 decisions and cases withdrawn during phase 2; (2) “2 - P1/P2 clearances with remedies” includes all Art 6.1(b) and Art 8.2 decisions; (3) “3 - Unconditional P2 clearance” includes Art 8.1 decisions; (4) “1 – Fatalities” excludes any cases where the parties withdrew due to antitrust reasons during phase 1; (5) The EC does not state what date is used for grouping mergers; however, we understand it is the outcome date.}

^{Source: https://competition-policy.ec..... Accessed 05/05/2026.}

Where should we now focus, so that the Guidelines work in practice?

The starting point is that we need serious, sustained, and constructive engagement from all stakeholders. Not only during the consultation period, but well beyond it. 

“All the talents”: drawing on both experience and new ideas

Making the final Guidelines work in practice will depend on how the Commission, practitioners, and economists collectively finalise, interpret, and apply them across different sectors and circumstances. That requires the full depth and breadth of experience, drawing on what worked before. And also a willingness to develop and test new approaches, tailoring the framework to suit real cases.

The consultation period is an important first step. It provides an opportunity to identify areas of consensus and – just as importantly – where further work is required. Some issues are already clear candidates. The section on failing firms, for example, seems to be a missed opportunity to address Europe’s lack of competitiveness; particularly, if you contrast it with discussion of the important competitive force that mavericks provide.[9] Similarly, the treatment of sustainability and privacy will need more concrete analytical frameworks.

But once the initial flurry dies down, the real test will come after the consultation closes. Clarity, expertise, and persistence will determine whether the ambition of the Guidelines translates into a workable framework. This is a long-term commitment.

That said, several themes stand out.

Strengthening the evidentiary framework, and how that evidence is interpreted

The draft Guidelines discuss potential evidence sources, particularly in relation to dynamic efficiencies. That is a positive development and should be applauded. However, we need to identify ways to better scrutinise the Commission’s common sources of evidence, in general and particularly with respect to efficiencies.

Firstly, the role of market testing needs to be refined. Responses from competitors and customers play a key role in the Commission’s current approach. In many cases, it is given disproportionate weight, lacking a disciplined assessment that distinguishes between circumstances where such evidence is compelling, and when it’s misleading. The deference given to market testing could be especially problematic for efficiencies, as their confidentiality means they cannot be tested that way. It risks baking in an evidentiary imbalance that isn’t at all necessary.

Secondly, the role of internal documents needs to improve. Over recent years, these have become increasingly important for the Commission (Figure 2). That is understandable but it can be problematic. In some cases, the exercise became little more than identifying “concerning” language in emails and board papers, rather than understanding the underlying commercial and economic context.

The issue is not the source of the evidence itself, but the framework through which it is interpreted. The draft Guidelines describe types of internal documents that may help establish efficiencies. These include those presented to the board or investment committee when developing the business case, such as pre-merger studies, historical innovation data, and independent expert analysis; especially where prepared “in tempore non suspecto”.[10] Clearly, these can provide valuable insight into commercial thinking. But they are not, in themselves, an assessment of competitive effects. Numbers don’t speak for themselves. What they reveal about incentives, capabilities, or long-term market dynamics requires economic interpretation. That, generally, is what the process has lacked when we criticise the use of internal documents.

So, the priority is not to identify evidence – whether external testimony or internal documents – that provides a shortcut in the assessment, displacing rigorous analysis. Rather, it’s to identify the evidence that best informs and prompts that rigorous analysis. The business case for the transaction is a good example.

Figure 2: Evidence referred to in EC merger decisions, significant outcomes only

^{Notes: (1) The analysis is based on data scraped by the Compass Lexecon Data Science team. The analysis includes all documents published by the EC. Within each document, we then conducted a string search for terms that could be grouped under the above headings (2) The last data included in the table was 20/01/2026. (3) The process does not capture whether the reference made by the EC was positive or negative (4) The chart includes all mergers with a significant outcome, that is phase 1 remedy cases under 6(1)b and phase 2 cases. (5) Mergers are grouped by the opening date of the merger assessment by the EC. In some cases, the EC does not record this date, so a proxy was used based on the first recorded document date. (6) The chart includes only cases closed prior to 20/01/2026. (7) The chart only includes closed decisions. (8) The analysis considers only decisions published in English.}

^{Source: Compass Lexecon analysis of published EC documents. Data extracted on 20/01/2026.}

Ensuring rigorous analysis is at the heart of decisions, particularly in key transactions

There is a broader point here. The Commission was, historically, at the forefront of applying economic analysis in competition law. In recent years, there has been a degree of retrenchment – both in the use of quantitative tools and, more fundamentally, in the application of underlying economic principles. To an extent, it is understandable. The reaction to cases such as UPS/TNT or Facebook/Instagram – both by the General Court, and court of popular opinion – casts a long shadow over subsequent regulatory practice.

But there is a risk that caution becomes aversion. Take market definition as an example. There became a view that it got in the way: that it was often unnecessary at best, and misleading at worst. So, in practice, it’s not simply that we see the Hypothetical Monopolist Test or SSNIP test applied infrequently. It’s that the principles that underpin them aren’t applied either.[11] The recent U.S. experience in FTC v Meta illustrates why that matters. Intuition might suggest that Facebook and Instagram dominate their market. Many people took that view. Yet a rigorous analysis of market definition led to a different conclusion. The point is not the outcome, but the process: where markets are evolving, intuition and precedent can be a poor guide.[12]

That applies with even greater force in forward-looking assessments. One can’t apply the same empirical analysis. But as the Guidelines move towards a more dynamic framework, the need for disciplined and principled economic reasoning increases, not decreases.

Assessing trade-offs between incommensurable factors

At face value, one might think static theories of harm will no longer be the central preoccupation. And superficially, the draft Guidelines suggest that. However, the test will come when a clear static harm, such as a minor price increase, is weighed against benefits that are not directly comparable. At that point, if the Commission reverts to safer ground, the process will essentially remain where it is now.

The draft Guidelines acknowledge the challenge explicitly: how should price effects be weighed against non-price parameters? How should short-term harms be balanced against longer-term benefits? How should likelihood and magnitude interact?

The draft’s discussion of symmetric and asymmetric benefits and harms is a good starting point.[13] A precise quantification, such as the NPV calculation it mentions, will not always be possible. But we can fruitfully scrutinise the underlying logic and inputs of such a calculation, considering relative magnitudes, likelihoods, risk, and time horizons. That still provides a framework for disciplined analysis.

Similarly, there is scope to make greater use of tools that help translate different dimensions into a common analytical language. In practice, approaches such as quality-adjusted pricing, discrete choice analysis, and merger simulation can play an important role in bridging the gap between price and non-price effects. These techniques are already used in some sectors; their application could be broader. 

Developing a workable framework for assessing likelihood, magnitude and uncertainty of effects

The draft Guidelines do not introduce a formal “balance of harms” test. They still require the Commission to conclude whether a merger is more likely than not to result in a Significant Impediment to Effective Competition. However, it includes many of the important elements of one: when assessing benefits and harms, the Commission says it will consider the parameters affected, the timeframe and likelihood that each effect will materialise, and their respective magnitude. In popular terms, that allows one to consider high-impact-low-likelihood “Black Swan” events, that a strict “balance of probabilities” assessment would overlook.

That is the right approach. You will not find many economists who disagree with the principle. The debate is whether it can be adequately implemented in practice. And perhaps that is why the Commission has not introduced a full balancing test. Certainly, that was the basis on which the CMA rejected the Furman report’s recommendation that it should adopt a balance of harms approach.

Frankly, economists should focus on getting out of their own way on this issue. As a profession, we need to help establish how to ensure the approach works in practice. Sceptics have a point. But too many of them conflate being precise with being rigorous, and being pragmatic with being unprincipled. It asserts a false dichotomy between unrealistic precision and unstructured judgement. That is wrong and unhelpful.

In reality, decision-making already operates in this space. Courts and authorities routinely form views on what is “likely”, what is “material”, and how confident they are about those assessments. They don’t analyse that probability to three decimal places, but they engage with the principles.

The role of economics is to refine that engagement. Not by replacing the use of judgement with falsely precise numbers. Rather, it should structure the use of judgement, so that the thought process underpinning it is rigorous. That means identifying what constitutes a plausible benchmark, assessing the robustness of different scenarios, and clarifying where uncertainty lies – and how much weight it should carry.

The draft’s emphasis on structured, but non-exhaustive, frameworks is consistent with that approach. The task now is to apply it in a way that avoids both extremes: the opacity of vague balancing, and the false certainty of over-quantification.

---

References