Why Article 102 TFEU is about equally efficient rivals: legal certainty, causality and competition on the merits
As I mentioned last week, some of the most basic tenets of the post-modernisation consensus are being questioned (as much as the post-modernisation consensus itself). The idea that the vast majority of practices should only be prohibited following a case-by-case analysis of their likely effects is one of them.
More to the point, it has also become relatively frequent to challenge the idea that Article 102 TFEU is concerned with competitors that are as efficient as the dominant firm (at least so as a matter of principle).
This principle tends to be associated with the modernisation of the EU competition law system and the ‘more economics-based approach’. However, it has long been part of the case law. An explicit reference to equally efficient rivals can be traced back to the seminal AKZO ruling of 1991, where the Court explained that below-cost pricing is potentially exclusionary insofar as it is capable of foreclosing firms that are as efficient as the dominant player. In subsequent cases, the principle has been confirmed by the Court, perhaps more emphatically in Post Danmark I and Intel.
Two conclusions can be drawn from the case law. First, the idea that Article 102 TFEU is only concerned about the exclusion of equally efficient rivals is broader than the ‘as efficient competitor test’. The latter (‘AEC test’) is merely a manifestation of the broader principle. Accordingly, whether or not the AEC test is applied, Article 102 TFEU will still be concerned with the exclusion of equally efficient rivals. Similarly, the inquiry (is the practice likely to exclude equally efficient rivals?) is the same, irrespective of the instrument used.
Second, the principle is relevant across all parameters of competition. Because the AEC test and the broader principle tend to be conflated, there is a tendency to think that the latter applies only to price-based competition. However, the case law is unequivocal in this regard. In Post Danmark I, the Court made it clear that principle made it clear that Article 102 TFEU is not concerned with the exclusion of less efficient rivals in terms of, inter alia, ‘price, choice, quality or innovation’.
As we undergo times of change, it makes sense to look back and explain the logic of the case law. The hearing in Qualcomm (which took place last week) and the (UK) Royal Mail ruling have brought the discussion to the fore again. The analysis seems useful for two reasons. First, it is important to distinguish between the (narrow) AEC test and the broader principle, as the conflation of the two is relatively frequent. Second, it is an aspect of the case law that wonderfully exemplifies the extent to which law and economics go hand in hand.
The essence of the case law can be summarised as follows:
Competition on the merits: Article 102 TFEU seeks to ensure that firms remain willing and able to compete on the merits. The exclusion of firms that are less efficient is the very manifestation of the process that EU competition law is intended to preserve: it is the logical and expected outcome of a system based on undistorted competition. Protecting inefficient firms would alter the competitive process as much as a subsidy intended to keep a firm artificially afloat.
EU competition law protects a process, it does not engineer market structures: In the same vein, the point of Article 102 TFEU (and EU competition law) is not to design markets in accordance with a preconceived vision. Similarly, it is not for a competition authority to decide how many players should compete on the market (and for how long or with which assets). The protection of the competitive process as enshrined in the Treaty is far more modest in its ambitions: instead of determining outcomes and engineering market structures, Article 102 TFEU is there to ensure that rivals that have the ability to do so can thrive in spite of the presence of a dominant firm.
Causality: It is clear from the case law (think in particular of Post Danmark II), that any actual or potential effects must be attributable to the behaviour of the dominant firm. In other words, the Court makes it necessary for an authority or claimant to establish a causal link between the latter and the former. Where a firm is less efficient than the dominant player, any actual or potential effects cannot be attributed to the dominant firm, but to the fact that it is less attractive in terms of quality, price or any other parameter of competition. In other words, the causal chain would break in such a scenario.
Legal certainty: The case law suggests a final rationale. A dominant firm should be in a position to anticipate when it is in breach of Article 102 TFEU. For instance, a dominant firm knows its costs. Accordingly, it is aware of when it is pricing below cost (and thus where an equally efficient rival would be selling at a loss). Similarly, it can evaluate whether a rebate scheme is capable and/or likely to exclude a competitor that is at least as efficient as itself. On the other hand, a dominant player cannot be expected to be aware of the cost structure of a less efficient rival and, by the same token, it would not be able to tell in advance whether or not it is in breach of the law.
While some aspects of the case law are yet to be addressed, the principle has consistently been confirmed over the years. As we rethink EU competition law, the fundamental question we should be asking is whether there are compelling reasons to depart from it, and interpret Article 102 TFEU along different lines. I very much look forward to your comments on this point.
Chillin'Competition 
Dear Pablo,
Thanks a lot for this post (as always it is instructive and fun to read). I completely agree with you that EU competition law protects a process and not an outcome. However, at the conceptual level, I am wondering why competition on the merits can be conceptualized only as rivalry between equally efficient rivals. Intel seems to push forward this understanding but I am not sure that all case law corresponds to such a clear-cut articulation. Leaving the case law aside for a moment, I am wondering whether you conceptualize competition on the merits by using any criterion other than efficiency. Here your caution against over-inclusivity and in favour of “meaningfully constrain administrative action” provides two very helpful criteria. Do you think that only efficiency can meet these standards?
On the causality point, and as of clarification: Post Danmark II indeed says that the effects should be attributable to the dominant company (para 47) since conduct (e.g. rebate schemes) implemented by the dominant undertaking should be capable of having an exclusionary effect on the market, yet some case law does not require a causal connection between dominance and abusive behaviour. For example, in Continental Can, the Court dismissed undertaking’s argument that there was a break in the causal chain because market power was not used to effect the merger (paras 19-21, 27) and in Hoffmann-La Roche considered that “the interpretation suggested by the applicant that an abuse implies that the use of the economic power bestowed by a dominant position is the means whereby the abuse has been brought about cannot be accepted”. Hence according to Hoffman La Roche causation might be treated differently according to the nature of the abuse in question (see also Opinion of AG). In addition, dominance, abuse and effects might be in defferent markets (British Gypsum v Commission, Tetra Pak II para 27). So would you say that these cases are in the same line or there is a change in Court’s thinking regarding causality?
Stavros Makris
10 May 2021 at 6:44 pm
Thanks, Stavros!
On the first point: I guess I am more interested in the positive analysis of the EU legal order. From that perspective (which includes State aid and free movement), it looks like the exclusion of firms that do not deliver in terms of, inter alia, price quality or innovation is seen as the natural manifestation of the competitive process. Competition, by its very nature, results in the exclusion of such undertakings.
On the second point: I fail to see any contradiction. What the two judgments you mention clarify is that there need not be a link between dominance and abuse. Post Danmark II, on the other hand, emphasises that there must be a link between the practice and any actual or potential effects. They address two different points. The idea that the abuse and the effects may be manifested in different markets is not new (it is the scenario in any leveraging case) and is not at odds with the point about attributability.
Pablo Ibanez Colomo
10 May 2021 at 11:24 pm
Thanks a lot for your response!
I agree with you that a positive analysis of EU competition law suggests that firms that do not deliver on certain parameters is a manifestation of the competitive process. However, this formulation is not necessarily attached to the concept of efficiency. It can also be attached to/derive from the distinction between ‘performance competition’ (Leistungswettbewerb) and ‘impediment competition’ (Behinderungswettbewerb). This can have a practical implication on the scope of Art. 102. And this is the reason why (if you ask me) the AECT is only a ‘useful tool’ among others for finding certain abuses and there is no legal obligation for always using (Post Danmark II paras 55-62). Efficiency is one proxy for distinguishing between performance and impediment competition.
Re the second point: I agree with you that there is no contradiction. However, the conception of attributability is quite thin on this basis. For instance, a strong conception would demand a causal link between market power, behaviour, and effects. When the Court breaks these links and asks only for a link between effects and conduct it might hard to trace a clear-cut (neoclassical) economic story consisting of the basic elements of exercise of market power, reduction of output and consumer harm.
Stavros Makris
11 May 2021 at 10:03 am
Thank you for (yet) another thought-provoking post. It seems that the previous comment already alludes to it, so here is a question on Post Danmark II. The post postulates that Article 102 TFEU is still concerned with the exclusion of equally efficient rivals. However, is that really so, assuming that PD II (in particular, para 59 and 60) is still valid and not changed by Intel?
WW
10 May 2021 at 10:15 pm
Thanks for the great question. I have addressed that point in my writings (see in particular here: https://ssrn.com/abstract=3599407)
As I explain there, one can think of exceptions to the principle as described above. You rightly mention Post Danmark II, which provides a wonderful example: if the dominant firm benefits from a competitive advantage deriving from exclusive rights in a partially liberalised industry, one cannot properly speak of competition on the merits.
So, to sum up: there may be exceptions to the principle. I explain in the paper that it would be for a claimant or authority to show why, in the specific circumstances of a particular case it is justified to depart from that principle
Pablo Ibanez Colomo
10 May 2021 at 11:29 pm
Thank you for the reply. Indeed, this interpretation makes sense. And thank you for the link, will read with great interest.
WW
11 May 2021 at 8:40 am
I agree with your comment about distinguishing between what Art 102 is trying to achieve and the tools used to achieve that (e.g. the AECT). But I don’t really agree with the statement that competition law is “only concerned with the exclusion of as efficient rivals” when stated as an overriding principle of law (I understand the economic and legal arguments that it should be limited in that way but I do not agree that it is as things stand). I think it must be right that Art 102 is concerned with such exclusion but that’s not the be all and end all (e.g. all the cases refer to a “hampering” of the competition process as well as “elimination” of competition and the European Commission’s guidance recognises the role in the competitive process of less efficient rivals).
The Royal Mail case you note there is a prime example and I hope this isn’t thought of as simply an (ex) state monopoly case from a (now) obscure jurisdiction. It’s also not a case of a return to “form based” exclusions. The Court of Appeal does a good job, I thought, of contextualising the Post Danmark I and II and Intel cases within the wider scope of competition law and the objectives of Art 102.
I see the exclusive focus on the supposed efficiency of entrants as risking looking at the question the wrong way around. In an enforcement an authority is looking at what the dominant firm has done – have they unlawfully distorted the competitive process in a way that harms competition, etc. What you get back is that it doesn’t matter what the dominant firm has done because the rival is less efficient on some metric (maybe they are just a smaller entrant to the market). To combine a real world example with the absurd, who cares if I dug up all the railway tracks if the rival didn’t have a hope anyway? Alternatively, can a dominant firm just keep snuffing out smaller rivals before they become “as efficient”?
I see the law as saying that all firms – big, small, good, bad – have a right to compete on the merits of their own products and services. The dominant firm has a special responsibility not to act in a way that distorts that competitive process. They can compete hard and no one has a right to stay in the market if their product isn’t good enough. In lots of cases, the right way to look at a problem will be to ask whether an as efficient rival has been excluded from the market and then you can debate what tools to use. In other cases that won’t be the right question. Case-by-case assessment of the factual and economic situation is necessary.
Restricting Art 102 “only” to those cases in which an as efficient rival is excluded makes it a rather narrow tool and would allow too much anti-competitive behaviour go unchecked. All my opinion, of course. The debate will go on!
ND
11 May 2021 at 4:57 pm
ND’s comment reflects my understanding as well. Competition law doesn’t allow a monopolist to claim efficiency while excluding, hampering or eliminating other competitors. Or are you, Pablo, saying that it is necessary to show (and prove) that all non efficient competition will be eliminated?
There has to be a showing that the conduct in question affects the competitive process; but that depends on the nature, structure and operation of the markets concerned.( I posted the relevant sections from Telia Sonera separately).
The protection of the process and the promotion of competition as a process that delivers innovation and consumer benefit is the objective; not (short term) consumer welfare or (short term) efficiency.
In blunt terms some markets are easer to dominate than others. Where, as in energy water, transport, telecoms, and tech markets the industries benefit from huge scale and scope economies and high fixed costs and network externalities, then entry will be a challenge, and no entrant can hope to be “as efficient” as the incumbent. To think that represented the law would cast its remit very very narrowly and would effectively limit its use to a small number of markets and make it almost entirely useless in digital and technology markets. While that might be an interpretation favored by defendants its isn’t the law.
Note also that the time dimension is potentially being overlooked. What is the timeframe? Should there be a timeframe over which consumer welfare and entry and meeting “historic” demand through established solutions is to be measured?
Is it correct to even think in these terms anymore? Isn’t end user or consumer demand for solutions through any alternative and competing means the real driver of markets?
Aren’t many markets now being digitized such that the switching of platform capability to focus on a specific demand can be achieved using sunk-cost and supply side resources at the flick of a switch – meaning that measurement of efficient supply on historic cost and alternative comparisons is limiting the frame of review to a backward looking basis – and one largely irrelevant to the nature of the markets in question?
Competition from alternative suppliers, big or small, from a fringe of others or from alternatives is, in reality, more important in meeting demand and driving innovation isn’t it? I think this is recognized in the caselaw too.
tim cowen
11 May 2021 at 6:35 pm
Thanks so much, Tim. Really wonderful to get your views here.
You are right to write that certain markets are easier to dominate than others. And you are right to refer to network industries such as telecoms, transport and tech as an example of such markets.
What is interesting is that the principle that Article 102 TFEU is concerned with equally efficient rivals was developed in cases relating to those very markets.
You mention TeliaSonera, which is a wonderful example in the telecoms industry (featuring both high fixed costs and network effects). In the TeliaSonera judgment alone, I have counted no fewer than 21 references to an equally efficient competitor/as efficient competitor. It is central to the assessment of the lawfulness of the conduct.
Post Danmark I is another wonderful example, in which the Court emphasised the principle. Again, it is a network industry featuring very high fixed costs. In addition, the industry was not entirely liberalised at the time of the reference (thereby giving a major competitive advantage to the incumbent).
Lithuanian Railways, to which ND referred, yet another example.
This background suggests that even in industries that feature some segments with natural monopoly features the lawfulness of conduct is to be assessed by reference to an equally efficient rival. What my post attempted is to explain the logic underpinning this consistent choice made by the Court in these cases.
Again, really grateful!
Pablo Ibanez Colomo
11 May 2021 at 11:15 pm
Thanks both for the thoughtful comments. I really appreciate it.
ND: the main comment I have is that you seem to assume that the relative efficiency of rivals is exogenous to the question of whether a dominant firm has ‘unlawfully distorted the competitive process’.
Your view does not appear to capture the essence of the case law as I understand it. Far from being an irrelevant consideration, the consistent interpretation of Article 102 TFEU by the Court suggests that the relative efficiency of rivals is an essential factor when drawing the line between lawful and unlawful conduct.
For instance, it is clear, following Post Danmark I, that pricing aggressively is not abusive where an equally efficient rival would not be forced to sell below cost. In fact, we have an example of a dominant firm removing the railway tracks, as you know (Lithuanian Railways). Even in that case, an as efficient competitor was the benchmark to assess the lawfulness and impact of the conduct.
So when looking at ‘what the dominant firm has done’ the relative efficiency of rivals is central when crafting legal tests. In this sense, you conclude that ‘much anticompetitive behaviour’ would ‘go unchecked’ if Article 102 TFEU were only concerned with equally efficient rivals. However, what amounts to ‘anticompetitive behaviour’ to begin with is determined by reference to equally efficient rivals.
Pablo Ibanez Colomo
11 May 2021 at 11:06 pm
hi – I leave the discussion of case law to you, Tim and ND, but if you are right, and the case law requires us to follow the equally efficient competitor principle then I would have to say that far from exemplifying the extent to which law and economics go hand in hand, it manages to do quite the opposite. The principle does feature in the prioritization guidance (and to be clear any exclusion that does exclude an equally efficient competitor is certainly problem, though not necessarily one that should be prioritized), however as an economic matter it is perfectly possible for less efficient rivals to be excluded through anti-competitive conduct, and such exclusion distorts competition just as excluding equally efficient rivals does. It is true that causation is always difficult but there’s no reason to think it cannot be established in such cases. Therefore this unfortunately seems be a case where competition law departs from economics. Maybe for reasons of legal certainty that is necessary, or maybe it is just because the answers that economics provides have implications that are inconveniently interventionist for some. I don’t know. In any case this is another area where those that consider that excessive reliance on economics has led to underenforcement should think again and focus more on the importance accorded to legal certainty (which in this cases requires us to accept that intentional and avoidable mistakes will be made, and hence justice denied for some (less efficient firms will be foreclosed and consumers will suffer harm) in order to simplify decision-making for others – a debate for another day perhaps).
chris pike
12 May 2021 at 1:12 pm
The key points from the judgement in TeliaSonera are given below. Note that the starting point is what evidence has to be taken into account in an abuse by margin squeeze case.
As a practical matter the court says we should start with the costs of the incumbent (para 41 & 46). If the prices are below incumbent costs and intended to drive out the entrant the enquiry ends. Para 40. It addresses evidence of intent from the “strategy” of the incumbent (para 41). Consistency with AKZO suggests that the costs to be taken are Average Total Costs (probably on a historic cost basis).
The reason is given in para 43: if the costs are below own cost and the strategy of driving out the entrant in the downstream business will likely be exclusionary.
Importantly, it does not call up or require an investigation into the efficiency, actually or potentially, of the entrant, either at the time of the enquiry or in future. It accepts that the entrant has “smaller financial resources” and is “perhaps” as efficient as the incumbent. But no enquiry has to take place if the imcumbent is seeking to drive out a rival and pricing below own costs.
The principle is that the dominant incumbent can’t price below its own costs with intent to exclude rivals.
The “efficiency” in question is incumbent efficiency meaning efficiency vs entrant in the downstream market. As stated in 43 if the incumbent is making a loss in the downstream market entrants that are equally or less efficient would not survive; the issue is the obligation on the incumbent and its duty toward others not to abuse its dominance ( 44).
Importantly the case can proceed from the perspective of likely effects and there is no need to wait until rivals have been excluded.
Para 45 deals with special cases. For example what to do if the entrant is in practice more efficient or if the incumbent hasn’t got an attributable cost – or its written off – or there are other special reasons to look further in an individual case, only then should the new entrants costs be investigated. Where the defendant puts forward evidence that the new entrant is for example, more efficient than it is, then that may be a special case where further investigation could take place.
Like Intel or Pfizer/Flynn, where the incumbent puts forward evidence in its defence, such as the fact that its pricing is below own costs but above new entrant costs and hence not exclusionary, that can’t simply be ignored and needs to be investigated.
However, unless there is evidence put forward by the incumbent it is not necessary to do a detailed investigation of ATC/AVC or what is or is not variable or marginal cost or joint costs or what time period should be taken into account- nor does it discuss the concept of cost and what the relevant accounting policies or which policies are accepted that allow costs to be depreciated.
So there is no investigation of actual or relative efficiency required absent special circumstances. The court is not defining an efficient market outcome objective. It is, at best, defining an outcome which means that entry has to be better than or equal to the (historic) downstream incumbent, which is likely to be based on evidence available most of the time but a lot depends on evidence.
Also the court was clearly assuming that the telco has to keep costs and accounts under telecoms regulation: in most other major dominant businesses there may be no relevant cost accounting available. (There is an argument that part of the duty of a dominant company is to keep such accounting in order to be sure of compliance).
A highly relevant factor when considering efficiency is that the telco is regulated because it will, absent regulation, have such scale and scope and externality economies that it will inevitably foreclose competition (and this did happen for over 100 years in most countries). To be as efficient as the incumbent telco in all activities, a competitor would have to enter at the same scale. This is highly unlikely to happen in practice overnight.
So, we know that the CJEU statements about “as efficient competitors” are not absolute statements about being as efficient as incumbents in all activities.
CJEU:
“The prices to be taken into account
38 The Stockholms tingsrätt seeks to ascertain, first, whether, for that purpose, account should be taken not only of the retail prices applied by the dominant undertaking for services to end users, but also those applied by competitors for those services.
39 It must be recalled, in that regard, that the Court has already made clear that Article 102 TFEU prohibits a dominant undertaking from, inter alia, adopting pricing practices which have an exclusionary effect on its equally efficient actual or potential competitors (see, to that effect, Deutsche Telekom v Commission, paragraph 177 and case-law cited).
40 Where an undertaking introduces a pricing policy intended to drive from the market competitors who are perhaps as efficient as that dominant undertaking but who, because of their smaller financial resources, are incapable of withstanding the competition waged against them, that undertaking is, accordingly, abusing its dominant position (see, to that effect, Deutsche Telekom v Commission, paragraph 199).
41 In order to assess the lawfulness of the pricing policy applied by a dominant undertaking, reference should be made, as a general rule, to pricing criteria based on the costs incurred by the dominant undertaking itself and on its strategy (see, to that effect, Case C‑62/86 AKZO v Commission [1991] ECR I‑3359, paragraph 74, and France Télécom v Commission, paragraph 108).
42 In particular, as regards a pricing practice which causes margin squeeze, the use of such analytical criteria can establish whether that undertaking would have been sufficiently efficient to offer its retail services to end users otherwise than at a loss if it had first been obliged to pay its own wholesale prices for the intermediary services (see, to that effect, Deutsche Telekom v Commission, paragraph 201).
43 If that undertaking would have been unable to offer its retail services otherwise than at a loss, that would mean that competitors who might be excluded by the application of the pricing practice in question could not be considered to be less efficient than the dominant undertaking and, consequently, that the risk of their exclusion was due to distorted competition. Such competition would not be based solely on the respective merits of the undertakings concerned.
44 Furthermore, the validity of such an approach is reinforced by the fact that it conforms to the general principle of legal certainty, since taking into account the costs and prices of the dominant undertaking enables that undertaking to assess the lawfulness of its own conduct, which is consistent with its special responsibility under Article 102 TFEU, as stated in paragraph 24 of this judgment. While a dominant undertaking knows its own costs and prices, it does not as a general rule know those of its competitors (Deutsche Telekom v Commission, paragraph 202).
45 That said, it cannot be ruled out that the costs and prices of competitors may be relevant to the examination of the pricing practice at issue in the main proceedings. That might in particular be the case where the cost structure of the dominant undertaking is not precisely identifiable for objective reasons, or where the service supplied to competitors consists in the mere use of an infrastructure the production cost of which has already been written off, so that access to such an infrastructure no longer represents a cost for the dominant undertaking which is economically comparable to the cost which its competitors have to incur to have access to it, or again where the particular market conditions of competition dictate it, by reason, for example, of the fact that the level of the dominant undertaking’s costs is specifically attributable to the competitively advantageous situation in which its dominant position places it.
46 It must therefore be concluded that, when assessing whether a pricing practice which causes a margin squeeze is abusive, account should as a general rule be taken primarily of the prices and costs of the undertaking concerned on the retail services market. Only where it is not possible, in particular circumstances, to refer to those prices and costs should those of its competitors on the same market be examined.”
tim cowen
12 May 2021 at 2:54 pm
Hi Chris and Tim
I very much hope your comments will get the attention they deserve (and hopefully provide the basis for further discussions beyond this blog).
Chris: the points you raise deserve a full seminar. I would point out the following here: any legal system will lead to Type I and Type II errors. It is inevitable if it is to provide clarity and certainty. The challenge is to design the system in a way that errors are minimised.
Competition law is not, and has never been, a tool to achieve optimal outcomes in every case. You may see this as a departure from economics (in the sense that the legal system tolerates errors), but I do not. It is a legal system informed by economics and relying upon tools that assist courts and authorities in minimising errors.
We could of course conceive a system in which achieving optimal outcomes in individual cases is prioritised over legal certainty and the predictability of administrative action.
In such a system, which would give authorities broad discretion, to intervene in every case and stakeholders would not be able to anticipate intervention and/or self-assess their conduct, would amount to policy-making without law.
Some are advocating for such a system, which represents a marked departure from the current one. Whether it is preferable (including from an economic perspective) is, I agree, a debate for another day (or days, given the ramifications of the issue).
Thanks so much again!
Tim: I have the impression that your analysis conflates two questions. I understand your point as suggesting that the principle whereby Article 102 TFEU is concerned with equally efficient rivals implies that an authority would need to show, in a given case, that rivals are as efficient as the dominant firm. This is not the point I was discussing, and it is not something that follows from the above.
The paragraphs you very helpfully copied in your response exemplify the way in which the principle was applied.
If it turns out that the dominant firm would not force equally efficient rivals to sell ‘otherwise than at a loss’ (paras 42-43), then the practice is not abusive. In such circumstances, an equally efficient firm would have the ability and incentive to compete and, in accordance with the principle, the practice would fall outside the scope of Article 102 TFEU.
I cannot think of a better illustration of the issues raised in the post. Thanks so much again for the active involvement!
Pablo Ibanez Colomo
15 May 2021 at 12:48 pm
Pablo, I worry that we may be missing each other. Put simply you keep making the point that the law is concerned with equally efficient “firms” or equally efficient “rivals”. The thing that you appear to be saying is that the new entrant has to be as efficient as the incumbent firm – at the level of the firm.
That is not what I am saying. The law is about exclusion and not being excluded from a downstream market.
The difference is between a firm and a market.
For example BT Nordics, which as Telenordia was one of the original complainants in the Telia Sonera case, was only competing in a limited way in a narrow downstream market. It was buying inputs from TeliaSonera at wholesale. Telia Sonera was the incumbent with a huge network covering all of Sweden. Telia Sonera was competing downstream and upstream – as a local loop operator and in many other locations worldwide. It was making a profit overall so no issue of predation.
The case was about whether the incumbent could make a profit on the wholesale business and price below cost on the retail business. This became known as a margin squeeze case but was simply a case of exclusion in a downstream market. Where the rival is only in the retail business abs there was no margin in retail the actions of the dominant firm excluded it from that org do the market. The vertically integrated incumbent is almost certainly more efficient than the entrant at a firm level. Indeed many economists argue that vertical integration is a more efficient market structure. The issue that was raised was why protect a new entrant that is only seeking to provide retail services. So why should the law prop up an inefficient rival that chooses only to compete in retail?
The decision focuses on the downstream market. The new entrant was considered excluded from the downstream market if the incumbent priced and intended to price below the costs of its own downstream retail business. That was likely to prevent entry in the retail business by any firm seeking to enter – at that level – not leaving enough margin at retail for entry to take place at the retail level.
The court said that exclusion at the retail level was enough. There was no investigation of vertical integration or the fact that a more efficient business structure would have been vertical integration whereby any number of firms might have been able to compete both retail and wholesale level.
So it’s not correct to say that the law is about equally efficient rivals. The truth is that efficiency is investigated, if at all, if there is insufficient evidence of intent to exclude or there is some other defence mounted by the incumbent.
What this leads to is a promotion of the competitive fringe by narrow market. The same is exemplified by the Microsoft case isn’t it?
If there is a market over which a vertically integrated firm may exercise a dominant position: and exclude rivals, then that firm will be stopped in relation to that market. This allows a competitive fringe to develop. In Microsoft the integration of windows media player and the Os was stopped. MS was also prevented from withdrawing interoperability from rivals.
As a matter if economics this is an example of a preference for innovation in narrow markets and the opportunity for innovation in narrow markets that use dominant platforms over vertical firm efficiency. It allows investment in retail from which potential competition at wholesale may then develop.
In terms of the point you were discussing this is very much a type of common carrier intervention.
The fact that the new entrant might be a part of a bigger global group wasn’t investigated either as far as I can recall. It could of course have been, had the incumbent made that case. From memory we sold up and exited the Nordics before that could take place – but there were a lot of other complainants and I believe they pursued the case and some recovered a lot of damages.
Overall, the economic efficiency and benefit of being vertically integrated is undone through this application of the law. The law can’t be said to be about equality of efficiency with an incumbent; no firm level enquiry takes place. I hope that is now clear.
If at all the only enquiry that could take place on the evidence is if the incumbent puts forward evidence of new entrant efficiency – put another way – if the incumbent can show a plausible case that it’s exclusion strategy will not have an effect on the market because of the resources of the entrant, then the law will not intervene to protect the entrant. As I said this is not a principle of efficiency it’s a matter of the legal duty on the incumbent and controlling behaviour to support rivals and competition as a process: more beneficial to innovation than to efficiency though.
I hope this is clear and you don’t think I have conflated anything! As a different point – when you talk about type I and II errors I think you will find that is based on Frank Easterbrook’s work and his assumption was that markets are self righting in the long run. We know that to be a position that only holds in very limited circumstances. It is certainly the wrong way round for high fixed cost low variable cost high network externality businesses that have broad scope and big scale economies. Lack of intervention in such markets will inevitably lead to monopoly – as it did for 100 years in Telecoms.
So the mistake is in failing to intervene in monopolistic market and in intervening too much in non monopolistic markets. Unfortunately that is a situation that persists to this day as the prioritisation principles and enforcement actions often fail to focus on the markets where enforcement is most needed.
Which is why big tech platforms have developed on the back of the telecoms regime. In effect those with the economic power to compete with the new big tech players (telcos) have been prevented from doing so. As a consequence the big tech platforms have built overlay networks and benefitted from the same economies that drive the entire digital “stack”. It’s a major example of regulatory arbitrage. It is not terribly clever industrial strategy – but has developed piecemeal and now would be a good time to put it right. Thankfully we are now seeing a G7 conference due in September that, hopefully,will knit together the governments and policy people with competition people, and we have a chance to rectify the mistakes of the past.
Tim Cowen
15 May 2021 at 3:48 pm
Thanks for the very interesting post Pablo! How does the case law on tying and bundling fit into this picture?
Justin Lindeboom
14 May 2021 at 10:31 am
Thanks, Justin! The issue discussed in the post is central in tying and bundling cases. If you take a look at cases like Microsoft or Android you will see how much the Commission emphasises that rivals’ products are just as good, if not better, as those of the dominant firm. Insofar as this is the case, that case law and administrative practice seems very much in line with the above discussion.
Pablo Ibanez Colomo
15 May 2021 at 12:33 pm