Sustainable Competition law; Competition Law Kills Coal Closure Plan, Or Does It?

This post concerns a  bit of a Dutch thing, namely the ‘position’ of the Dutch National Competition Authority ACM on an agreement by electricity producers active on the Dutch market, but it is interesting more generally for those who are interested in the relation between (EU) competition law and other issues like sustainability. The trigger for this position by the ACM is a plan in the national Energieakkoord  which is an agreement between organisations representing employers, employees, environmental NGO’s, companies and other social actors that aims to benefit the transition to a more sustainable energy policy and sustainable economic development in the Netherlands. Part of this Akkoord is the deal between four electricity producers to close down five older coal fired power plants (all constructed in the 1980s) in a coordinated manner. This get-together of four competitors to reduce production capacity has obvious competition law implications, so the Netherlands Competition Authority (ACM) was consulted on the compatibility of this plan with Article 101 TFEU and the Netherlands equivalent, Article 6 of the Competition Act. As the title suggests, the ACM considered the plan incompatible with competition law in a very preliminary and barely reasoned finding.

Interestingly, the ACM has just consulted the public on its position paper concerning competition law and sustainability and this case may well be the first real test case for the position contained in that paper. Firstly, though, some background to this matter. The relation between sustainability or public interest objectives and competition law has been controversial for quite some time. This is reflected in the various Commission guidelines on state aid for environmental matters (now in their sixth edition since 1974) and the numerous cases that have dealt with competition law and public interests. It has also been the focus of some excellent legal research (most recently by Chris Townley and Suzanne Kingston) and heated debates on the fundamental question whether there should be a role for such considerations in (EU) competition law in the first place. This question, however, has been answered in the affirmative, with Commission decisions like the one in CECED and Court judgments like those in Wouters, OTOC and Sydhavnens. The question today is how public interests should play a role in competition law and the cartel ban (Article 101 TFEU and the equivalent Article 6 of the Netherlands competition act) in particular.

In this regard it is interesting to see that the ACM has published a position paper on this matter in the first place, whereas the Commission seems keen on erasing its CECED precedent and the special attention for environmental agreements from the books. It may be recalled that the CECED agreement between the European importers and manufacturers of washing machines no longer to market energy inefficient washing machines was exempted under Article 101(3) in part on the basis of the environmental benefits that follow from the increased energy efficiency. Whereas the CECED decision was used as an example of an environmental agreement that passes competition scrutiny in the 2001 Guidelines on Horizontal Agreements (OJ 2001 C 3/2) in a special chapter devoted to environmental agreements, this category of agreement no longer has a special chapter devoted to it in the 2011 Guidelines on the application of Article 101 TFEU to horizontal agreements, whereas the CECED case is presented as an example of the standardisation agreement (OJ 2011 C 11/1, para. 329. Within the ECN, therefore, the guidance by the Commission is limited and with the position paper and this case the ACM appears to want to play the leading role.

Let’s have a look at the findings then. The essential summary according to the ACM, set out in the press release, is that the benefits are not outweighed by the costs to consumers. Energy prices will rise, according to the ACM, whereas the benefits are too limited to compensate the disadvantages arising from the higher prices, according to the president of the executive board of the ACM. This means that the ACM is seriously barking up the consumer welfare tree, looking only at narrowly defined costs and benefits, in a way that I least don’t remember from the Commission.

These findings are set out in slightly more detail in the analysis that is also available online from the ACM website.

Here we first find the description of the Energieakkoord and the way it was concluded. This involves over forty organisations from different industry sectors and representing different interests, including the consumer interest. The ACM then moves on to check the compatibility with the first paragraph of Article 101 TFEU. Its findings start with the observation that this closure plan involves approximately 10% of the generation capacity installed in the Netherlands. Moreover, coal fired generation capacity is situated towards the bottom end of the merit order curve. This means that the marginal costs of a unit of production of this type of plant is relatively low (in fact only hydro and nuclear have lower marginal costs), so this type of production capacity will first be used to meet demand for electricity. Only if demand exceeds the generation lower merit order capacity will higher order, more expensive, capacity such as gas fired plants be used. Phasing out this coal fired capacity will thus mean that higher order capacity will need to be used at an earlier stage, involving higher costs and thus leading to higher electricity prices. According to the ACM these are ‘clues that the upward pressure on prices that can be expected can be of real significance’ [translation by author]. Now this seems a bit speculative to me, but this is of course only a preliminary assessment.

Moreover, this assessment starts from the premise that removing 10% of the installed capacity in the Netherlands will have this effect. On a wider European market, the effects may be negligible. Indeed, what is to keep a Dutch supply company from simply buying coal fired electricity from, say, a German coal fired plant (like the one in Bund Naturschutz, C-115/09) and then import into the Netherlands to sell it to meet demand in the Netherlands? Poor interconnection leading to limited cross-border transmission capacity may be what is keeping this from turning into a reality. According to the ACM, there is more interconnection, ‘but this is not complete’ [translation by author]. In the 2009 decision on the takeover of Dutch utility Essent by German utility RWE, the Commission distinguished between peak and off peak hours, with markets being national at peak hours whereas there is a north west European market at off peak times (para. 30 et seq.) We find none of this deeper insight in this preliminary assessment. On a similar note other arguments put forward by the energy companies are set aside in a way that is gnomic as well as succinct. The result is that we don’t even know whether this is seen as a hard core restraint, for which no market analysis is required, or rather given a – albeit short – effects-based analysis. This matters, because for restrictions in the object, or hard core, category, no such effects analysis is required, greatly reducing the possibilities for redeeming such restrictions.

The ACM then moves on to the third paragraph. Here, the environmental effects are weighed in. As a general statement, the ACM considers that a ‘more sustainable production can be welfare enhancing’ [translation by author]. So, what are the environmental benefits? In a nutshell, closing down coal fired power plants means less emissions of CO2, SO2, NOx and fine particulate matter.

As far as CO2 emissions reductions are concerned, the ACM observes the so-called waterbed effect. This refers to the situation where the EU emissions trading scheme (ETS) for greenhouse gasses functions like a waterbed. The extra emissions reduction in the Netherlands reduce demand for greenhouse gas allowances on the European market, thus lowering the price and allowing for an increase of emissions elsewhere in the EU. In the opinion of the ACM, this means that the associated advantages will spread over the entire ETS-area (the entire EU) resulting in only a very small share for Dutch electricity users. The result is that the benefits in terms of CO2 emission reduction are considered irrelevant. The other emissions, however, are taken into account as they generate effects in the Netherlands. Because of the National Emissions Ceilings Directive (NEC Directive), NOx and SO2 emissions reductions in the Netherlands allow the Netherlands to emit more from other sources or to take less other measures to reduce emissions. This translates into avoided costs amounting to € 9,40 / kilo NOx and  € 5,40 / kilo SO2.

For fine particulate matter there is no such ceiling, but the emissions of this have an effect on the health of citizens in the Netherlands. This is estimated to factor in at € 44,30 / kilo. To my mind this is where we enter the realm of the purely speculative. There is, to the best of my knowledge, no safe limit for fine particulate matter exposure and estimates are that in the Netherlands alone every year between 2300 and 3500 people die prematurely as a result of short term exposure to fine particulate matter. Mortality as a result to long term exposure is expected to be much higher, as the Netherlands Planbureau voor de Leefomgeving (Environmental Planning Agency) has stated in a report. I personally fail to see how any price tag can be attached to what is essentially the increased chance of dying prematurely, but the plot thickens. In the same report the Planbureau finds that the Netherlands is a net exporter of such fine particulate matter. This is a nice way of saying that fine particulate matter produced in the Netherlands is likely to cause premature death in Germany and Denmark (because of the southwesterly winds), and these cost effects are not taken into account. So much for a waterbed effect… All in all, the benefits to Dutch electricity consumers are calculated to be in the order of € 180 million over the duration of the agreement.

The costs as a result of the capacity reduction, however, are equally significant and weigh in at € 450 million over the duration of the agreement. The conclusion is that, in the view of the ACM, there is no advantage and a fair share for consumers, meaning that the last two conditions of Article 101(3) need no further scrutiny. The ACM’s preliminary finding is therefore that the coal closure plan qualifies as a prohibited restriction of competition. In this regard it is interesting to see that VEMW, an ardent proponent of the interests of large energy consumers, is one of the parties to the Energieakkoord. Apparently, they considered these costs acceptable.

What, then, is the fate of sustainable competition law? For one thing, focusing on costs and benefits for a limited number of consumers alone certainly sees at odds with the essence of sustainability. Climate change and environmental deterioration more generally affect many people beyond the inhabitants of this little kingdom by the sea. Dogmatic competition law would also look radically different if the focus were on consumer cost effects alone. For one, the analysis of recoupment as a requirement for predatory pricing would need to be reconsidered on this side of the Atlantic.

Finally, sustainable competition law requires legal certainty. The utilities involved in the coal closure plan have sought to achieve this by asking the ACM for its opinion. What they got is really little more than a vaguely worded preliminary assessment that is open to significant criticism. In this criticism we read that the Akkoord still stands and will not be terminated because of the ACM’s position. What could happen is that the plan is put into effect, the ACM finds a prohibited agreement and issues a decision that would then be challenged, leading ultimately to a preliminary ruling. This would create the much needed legal certainty, unless the ACM follows the Commission’s example and decides to opt for a commitment or settlement procedure that would allow it to stick to its preliminary assessment without conducting a fully blown competition law and economics analysis. Then again, this blog entry is only a preliminary assessment, so don’t take my word for it.


  1. Christopher Townley

    I think that this is a really fascinating note. Thank you.
    It is a really interesting time for competition law in the Netherlands with, as you point out, the draft ACM position paper on sustainability. The Position Paper is an excellent attempt to bring the application of EU competition law by the National Competition Authorities back in line with the EU Courts’ judgments.
    Yet, here is a wonderful example of an opinion that, at least as presented in your note, seems to row back from this promise.
    You, rightly, focus on the relevance of environmental considerations here, and you do so powerfully. Another issue of particular interest to me that you allude to is that of whether one should analyse the costs and benefits of an agreement in each market that it affects seperately; or, can one aggregate these costs and benefits across different relevant markets?
    The answer to this question has major implications for the impact of public policy (particularly environmental) considerations on the outcomes of competition assessments. Often, consumer welfare harm in cases such as the one you point to is felt in different markets (and territories) from the environmental benefits. To say that one cannot aggregate across markets means that, where agreements generate environmental benefits that outweigh their competition harm, they cannot be brought to bear in the Article 101 balance.
    The relevant Commission notice says: “The assessment under Article 81(3) [now Article 101(3)] of benefits flowing from restrictive agreements is in principle made within the confines of each relevant market to which the agreement relates.” (European Commission Guidelines on the application of art.81(3) of the Treaty [2004] OJ C101/8, para 43. Similarly, Guidelines on the assessment of horizontal merger under the Council Regulation on the control of concentrations between undertakings [2004] OJ C31/3, para 79 has a rule against aggregating across markets; and see the discussion in OFT, Article 101(3)—A Discussion of Narrow versus Broad Definition of Benefits: discussion note for an OFT breakfast roundtable (London, Office of Fair Trading, May 12, 2010), available at paras.2.4, 4.1 and A10).
    In an ECLR paper from 2011 ( ) I argue that the EU courts do aggregate across markets. The paper discusses the advantages of aggregating across markets; and the disadvantages. It concludes that, in part because of the environmental impact, the arguments in favour of aggregating across markets outweigh those against. I recommend that EU competition authorities that are not already aggregating across markets should start soon.
    This sort of change may have had an impact in cases just like the one that you address.

  2. Laurens Ankersmit

    Thanks for your comment Chris. I have my reservations to your point that the ACM position paper aligns NCA practice with CJEU case law. In particular, the CJEU has never accepted neoclassical welfare analysis within 101. Also, Wouters is ignored on the argument that that case is insufficiently clear.

    To me it seems the ACM position paper brings the NCA practice more in line with the Commission practice, not too much with the Court’s case-law.

  3. Martin Holterman

    Does the Dutch competition authority even have the legal authority to clear a cartel that is of net detriment to Dutch social welfare on the grounds that it is has a net benefit when viewed at a European or global level? I.e. are they even allowed to weigh in the interests of people outside the Netherlands?

  4. Hans Vedder

    Dear Chris, Laurens and Martin,

    Many thanks for your thought-provoking comments and in relation to Chris’ comments I see a general tendency on the part of competition authorities to ignore their own past practice or certain elements of the Court’s jurisprudence if this in the interest of their perception of procudural economy and indeed aggregating across markets is probably the case in point. Now, the question whether or not the ACM can do this is down to what is the essential objective of the Competition Act. This is probably to safeguard effective competition in the Netherlands (applying Metro II to the Netherlands), but at the same time the Netherlands Competition Act wants to be no more or less stringent than the EU competition rules and the latter are increasingly (insofar as the Commission is concerned) informed by a consumer welfare standard, and the question then becomes to what extent this means that only consumer welfare effects in the Netherlands can (must) be part of the equation and yet another question is whether this means that price effects equate to consumer welfare effects. So what is the legal authority or legal basis for all of this? I’m not sure whether this ‘traditional’ way of looking at competition authorities as merely/simply applying legal norms that contain clear standards really works at all in a competition context. But I guess that gets an ever bigger issue involved and one that is also addressed during an ACLE conference on 12 December.


  5. Edwin Woerdman

    Dear Hans, Chris, Laurens and Martin,

    As a law & economics scholar, I mostly agree with the line of reasoning of the competition authority ACM in the Netherlands.

    In principle, the early closure of 5 coal-fired power plants raises power prices and must be regarded as anti-competitive. This could be exempted in case of environmental benefits that outweigh these costs, but the ACM calculated (largely correctly) that this is not the case. CO2 emissions in the EU will not go down when the Netherlands closes these old plants. This is primarily due to the waterbed-effect of the EU ETS, where the (auctioned) emission allowances that would remain after closure will be sold to an emitter somewhere else in the EU. As a result, there would be no CO2 benefit, as the ACM correctly says. SO2, NOx and particulates reduction is also a benefit, but too small to offset the anti-competitive effect. The ACM thus arrives at the correct conclusion that the closure of these plants should not be allowed.

    On the one hand, you may have some doubts about the calculation of the benefits of reducing SO2, NOx and particulates, about the exclusion of cross-border pollution and about the exclusion of other parts of the broader Energy Accord (Energieakkoord) concluded by civil society. On the other hand, the calculation of those benefits is pretty state-of-the-art (but can be improved as recent studies suggest, see for instance Miriam Nijland’s dissertation), while stretching the legal reasoning to all other aspects of the Energy Accord could basically legitimize every cartel. In the margin, however, I do admit that there is an imbalance in the reasoning of the ACM by (a) taking into account the cross-border waterbed effect of local CO2 reduction (which evaporates under the EU ETS), while (b) not taking into account the cross-border pollution effects of local SO2, NOx and particulates reduction.

    What worries me most in the legal debate, however, are the limited views that lawyers have of what law & economics and welfare economics is all about. Consumer welfare is not a zombie version of welfare economics; it is the very essence of welfare economics. Welfare economics is grounded in consumer welfare. And law & economics, for that matter, is all about weighing costs and benefits, for instance when judging the potential anti-competitive effect of companies agreeing to close 5 old coal-fired power plants in the Netherlands. If the ACM assesses this potential closure by analyzing consumer welfare effects (costs) and environmental effects (benefits), it is doing precisely what a law & economics scholar would prescribe to do in order to come to an efficient decision in the context of Article 101 TFEU and Article 6 of the Dutch Competition Act.

    It may not feel good that the Netherlands thus keeps these old coal-fired power plants open for a few more years. On a personal, more emotional level, I would love to see those plants being shut down today, just like you (as I would guess). But as a scientist, thinking about it from a law & economics point of view, the assessment changes. What remains is the burning question: why did power companies, in the Energy Accord, not also decide to voluntarily cancel (‘burn’) the emission allowances after the closure of those plants? Then we would also have a CO2 benefit of the closure, which could significantly alter the analysis.

  6. Hans Vedder

    Hi Edwin,
    to start with the last question: I’m guessing that they did not cancel the allowances because that would cost them even more (cf. the opportunity cost reasoning in Joined Cases C-566/11 et seq., para. 34). As to your concern, I’m not sure whether I (or lawyers in general) see consumer welfare as a zombie version of welfare economics. What I do believe is that the ACM uses a balancing mechanism (that may be in accordance with welfare economics) that simply doesn’t conform to the legal framework for that mechanism. As Laurens has also pointed out: there is no requirement in Article 101 that costs to consumers have to be outweighed by the benefits. We see this where Henk Don (chair of the ACM) defends this decision in the Financieel Dagblad using, inter alia, the following reasoning:

    “Als er voldoende voordelen aan de afspraak zijn verbonden, dan kan de wettelijke uitzonderingsbepaling op dat verbod van toepassing zijn. Daarvoor gelden strenge eisen, met name dat de concurrentiebeperking noodzakelijk is om de beoogde voordelen te halen en dat de afnemers dankzij de voordelen gecompenseerd worden voor de nadelen die zij ondervinden van de concurrentiebeperking.”

    This roughly translates into: if there are sufficient advantages to an agreement the legal exception to the prohibition can be applicable. This is subject to strict requirements, in particular that the restriction has to be necessary to achieve the benefits and that consumers, though the benefits, are compensated for the disadvantages that they experience from the restriction. This I believe refers to Article 101(3) TFEU or the identical Article 6(3) of the Competition Act and this provision only requires an advantage and, most importantly, a fair share for consumers in this benefit. Even when it makes limited sense from a welfare economics perspective, the legal framework simply doesn’t allow for such an approach.

  7. Martin Holterman

    Hi Hans,

    Thanks for your reply. Here’s a question I was emailing with Edwin about:

    Do you think the costs of a loss of competition to consumers can be offset by benefits enjoyed by them in another capacity?

    Obviously, under the Commission’s guidelines you can’t offset a higher price paid for the product by Dutch consumers due to a lessening of competition against a benefit enjoyed by Germans. But does the offsetting benefit have to be enjoyed by Dutch consumers in their capacity of consumer, i.e. in the same product market? What if instead part of the benefit comes in the form of lower income taxes? Or not having their house flooded quite so often? Those benefits are also “in a different market”, if they can be said to be in a market at all. Not including such benefits seems bizarrely narrow, but on a strict reading of the guidelines I would say they are out.

  8. Pingback: Ferries & Competition | Martin Holterman
  9. Pingback: Competition Law, Sustainability and Geography | Martin Holterman
  10. Pingback: Business must collaborate – without it the world is brutal and terrifying | Enjeux énergies et environnement
  11. Pingback: Business must collaborate - without it the world is brutal and terrifying | Expressivus
  12. Pingback: EU Emissions Trading Scheme and Europe’s Climate Policy: A Flagship Floundering – Oxford Martin Programme on Integrating Renewable Energy

Leave a Reply