Debates amongst competition scholars and practitioners have been buzzing with references to sustainability recently. Several books and articles were written about the topic in the last two years and several competition agencies have written position papers about it. Most of the debate about the interaction between competition law and sustainability concerns agreements between competitors that (purportedly) aim to promote sustainability, e.g. by phasing out unsustainable products or business processes.

There are various aspects and subtopics which are debated in this respect, but one of the most significant ones is the requirement in Article 101(3) TFEU that a sustainability agreement that restricts competition can only be lawful if it “allow[s] consumers a fair share of the resulting benefit” of the agreement. This requirement has been interpreted rather strictly by the Commission in the past decades. In its Guidelines on the interpretation of Article 101(3) TFEU, the Commission states that this requirement implies that the agreement “at least compensate” the “direct or indirect users” of the products or services covered by the agreement “for any actual or likely negative impact” caused to them by the restriction of competition caused by the agreement (paras 84-85).

But what if it is really the wishes of the consumers themselves which are at the heart of the sustainability concern? What if consumers themselves want products that pollute the environment, that harm fauna and flora, or that can only be produced by exploiting workers? If companies supplying those goods (or services) would decide to stop doing so (because of sustainability concerns), consumers may just move to competitors who are willing to fulfil their desire for these (unsustainable) products. In the absence of legislation, making such products more sustainable would require a coordinated stance of the entire industry to simply no longer make unsustainable products available to consumers and to force them to opt for sustainable alternatives instead.

In its draft revised Guidelines on horizontal cooperation agreements which it published on 1 March 2022, the Commission recognizes that the egoistic choices of consumers may harm society and that agreements between competitors may be necessary to overcome this. Indeed, “[a]s the sustainability impact from individual consumption accrues not necessarily to the consuming individual but to a larger group, a collective action, such as a cooperation agreement, may be needed to internalise negative externalities and bring about sustainability benefits to a larger group of the society” (para 601).

However, the Commission insists that, if such “collective benefits” are taken into account under Article 101(3) TFEU, the users of the products in question should be “part of the beneficiaries” (para 603). In para 604 of the draft Guidelines, the Commission provides two examples to clarify this: one example of a sustainability agreement that also benefits the users of the product in question and may therefore benefit from the exemption of Article 101(3) TFEU, and one example where this is not the case.

  • The first example concerns a sustainability agreement between fuel producers to make petrol less polluting. The Commission explains: “To the extent that a substantial overlap of consumers (the drivers in this example) and the beneficiaries (citizens) can be established, the sustainability benefits from cleaner air are in principle relevant for the assessment and can be taken into account if they are significant enough to compensate consumers in the relevant market for the harm suffered.”
  • The second example concerns a sustainability agreement between cotton growers to only supply cotton that is produced in a way that reduces chemicals and water use on the land where it is cultivated. The Commission explains: “Such environmental benefits could in principle be taken into account as collective benefits. However, there is likely no substantial overlap between the consumers of the clothing and the beneficiaries of these environmental benefits that occur only in the area where the cotton is grown.”

These examples show how fundamentally iniquitous the requirement of full compensation for the users is. Indeed, the Commission states that, if an agreement between competitors benefits (presumably rich European) citizens by providing them with clean air, it is acceptable. On the other hand, such an agreement is not acceptable if it merely benefits (presumably poor Indian) cotton pickers by not exposing them to hazardous chemicals. To make the contrast even starker: an agreement between competitors to improve production processes and thereby reduce air pollution could benefit from Article 101(3) TFEU if the factories in question are located in the EU (since European citizens, including the users of the product in question, would benefit from it), but it could not benefit from the exception if the factories are located outside of the EU (since European citizens would not benefit from it). This difference in treatment is all the more problematic since the need for sustainability cooperation will be more acute outside of the European Union, where government regulation of polluting emissions (or of hazardous chemicals, working conditions, etc.) is generally much less stringent.

Obviously, European competition law cannot solve all the problems of this world, but if these problems are caused by European consumers, why must they be the ones that are compensated for the cost of any solutions to such problems? If it is European consumers that demand cotton t-shirts for € 5 and this means that cotton pickers in India are exploited, why should European consumers be compensated if the industry agrees to an improvement in working conditions for these cotton pickers? If European competition law takes into account the reduction of air pollution as an efficiency, shouldn’t clean air outside Europe count just as much as clean air in Europe itself?

In the Commission’s defence, the requirement that an agreement “allow[s] consumers a fair share of the resulting benefit” is contained in Article 101(3) TFEU itself and is therefore not subject to the Commission’s discretion. On the other hand, it is not uncommon in European law that certain provisions of the treaties are interpreted in a manner that is inconsistent with their literal wording (although this obviously requires sanctioning by the European Court of Justice). But, more fundamentally, if (an interpretation of) Article 101(3) TFEU gives rise to such a glaring injustice, there is simply something wrong with that rule.

Indeed, I would contend that the exclusive focus of European competition law on consumer benefits (understood in a narrow sense as fulfilling consumer desires) is what makes that body of law hard to reconcile with several legitimate public policy concerns which are caused by unbridled consumerism (some but not all of which can be subsumed under the term “sustainability”). Obviously, the consumer welfare standard is a standard that is easier to apply (and easier to apply consistently across the members of the European Competition Network) than a broader standard of justice or fairness, but if the consumer welfare standard leads to iniquitous outcomes, it needs to be set aside (at least in this case).

In conclusion: if cleaner air for European citizens counts as a legitimate excuse for an otherwise anti-competitive agreement then a less hazardous working environment for cotton pickers should count in the same way. Alternatively, neither of these concerns should be a legitimate excuse for such an agreement, and such matters should be left to legislators (while possibly even more serious concerns, such as fighting climate change – which affects every citizen of the globe –, could still act as a legitimate excuse under Article 101(3) TFEU) – although European legislation may be less effective than business cooperation in enforcing more sustainability outside of the EU.


To make sure you do not miss out on regular updates from the Kluwer Competition Law Blog, please subscribe here.

Kluwer Competition Law

The 2022 Future Ready Lawyer survey showed that 79% of lawyers are coping with increased volume & complexity of information. Kluwer Competition Law enables you to make more informed decisions, more quickly from every preferred location. Are you, as a competition lawyer, ready for the future?

Learn how Kluwer Competition Law can support you.

Kluwer Competition Law
This page as PDF


  1. Agree with al
    l the sentiments here but would add one comment. Jan suggests one needs to interpret 101(3) in a manner “ inconsistent with their literal wording”. His essential reason is that a literal interpretation would give rise to a “glaring injustice”. To me that means it would be unfair. But if it is unfair that means the consumers have got more than a “ fair share”. The law says they should get a “ fair share” so a literal interpretation works. There is no need for an interpretation “inconsistent “ with the literal wording.

    1. Thank you for your comment. It is indeed possible to argue that, in some circumstances, a ‘fair share’ means ‘no share’. Of course, in that case, the second condition of Art. 101(3) TFEU risks losing much of its purpose. That condition is traditionally interpreted as preventing undertakings from arguing that an agreement should be exempted because it creates efficiencies even though these efficiencies only benefit the undertakings themselves (e.g. lower costs which are not passed on to consumers). If the second condition is reduced to the question of ‘fairness’, it will become difficult to apply since ‘fairness’ is a notoriously vague notion (one could say that most if not all of the law is about fairness). Why is it ever ‘fair’ that consumers get some of the benefit that undertakings achieve when they themselves create efficiencies? So, interpreting ‘fair share’ as possibly including ‘no share’ is indeed not inconsistent with the literal wording, but it creates other problems.

  2. Interesting observations! Thanks for pointing this issue out. I agree with Simon that a “fair share” might be interpreted more broadly. Many consumers actually want that products are more sustainable – so the collective action problem is among European consumers (rather than just between Europeans and people somewhere else). Each individual’s consumption only has very little impact so it is rational for everyone to not alter their consumption but hope that everyone else does. Solving this free riding problem should therefore benefit European consumers by catering the market better to their actual preferences. Of course this is difficult to reconcile with revealed preferences approach, so likely counterargument is paternalism and maybe I am too confident in preferences for sustainability. But at least where we have enough evidence that consumers would actually prefer more sustainable products this would be another avenue.

  3. Interesting blog, thank you! I do not think there is a need to ignore the literal wording of “fair share”.

    • Perhaps starting with a bit of a philosophical comment: Why is it assumed that “the resulting benefit” is material or monetary”? It can also be immaterial: We all benefit from justice in society. We are not just individuals; humans are defined by relationships to others and to society. And short term local damage can lead to long-term world-wide trains on society.

    • Perhaps more convincing: Receiving a “fair share” means first undoing an unfairness caused by one’s own action (negative externality, or damage caused to others) before claiming a share of a positive externality. If a person damages a public transport bus, it is fair that he must first repair it before being able to claim a right to ride it. (I welcome a better analogy)

  4. Super blog post and comments.
    My two cents: the balancing under Art 101(3) should be between the harm to competition and the first requirement of art 101(3) i.e. overall benefits of the agreement, not between the harm to competition and the possible impact on consumers. The overall benefits have no territorial limit. Fair share of the ‘resulting’ benefit means that some of the overall gains, but not all, have to go to the consumer. So the cotton example: achieving a sustainable development goal goes to the first condition. Fair share could be demonstrated by indicating that there is some consumer demand for sustainably produced goods but I disagree with the commission’s reading that fair share should be greater than or equal to the harm to competition.

    What we need is a test for fairness. I have a paper coming out on this and the suggestion is that we draw on behavioural economics insights on what is fair that reflect community standards.

    On zero benefits, observe that in Asnef-Equifax the ECJ raised the point (but did not discuss it fully) that a consumer who does not get a loan from a bank because the bank has, thanks to an information exchange agreement, worked out that they are a credit risk, is possibly a beneficiary of this agreement because they don’t get into a risk of bankruptcy.

Leave a Reply

Your email address will not be published. Required fields are marked *