Forget about consumer welfare: it’s the law vs discretion divide that will mark the future of competition law (my presentation at the IEE in Brussels)
Considerable energy has been devoted to attack (and defend) the consumer welfare standard on both sides of the Atlantic.
There is a real (perhaps growing) divide in the competition law community. Now that we are at a crossroads, this divide matters. It will mark the field for the years to come.
The above said, I think defending the consumer welfare standard is a waste of energy. For those who challenge it, consumer welfare is just a proxy war. Their end destination is not to embrace an alternative standard, but to change the shape and understanding of competition law.
(EU) competition law has progressively become a legalist discipline. Competition authorities are subject to constraints coming from the law and coming from mainstream economics. In the same vein, effective judicial review is known to be an indispensable ingredient in the system.
Under a legalist approach, consistency and predictability are paramount (for could we say that competition policy is implemented through law if enforcement were not consistent and predictable?). It is accepted that these values may occasionally lead to under-enforcement.
Those who attack consumer welfare, as much as those who say to endorse fairness or a wider range of values are, in essence, pleading to move away from legalism to embrace a discretionalist approach to competition law.
A discretionalist approach does not feel bound by consensus positions in economics, and is characteristically sceptical of them. Disregarding mainstream economics is not seen as a big deal. In fact, it may even be desirable – the consensus may be flawed for a variety of reasons.
What matters, under a discretionalist approach, is to reach the outcome that is deemed optimal in any given case. This is something that I explained already in the blog a while ago. Avoiding under-enforcement, accordingly, matters more than the consistency and/or predictability of the system. And avoiding under-enforcement may occasionally involve introducing a plurality of values or considerations.
These are the ideas I presented on Friday of last week in Brussels. My slides can be found here. I try to explain why people tend to develop a fascination with the goals of competition law, and why this debate is, by and large, irrelevant (or overly superficial).
I then move on to explain why law vs discretion is the real divide in the competition law community. As I already explained here, economic analysis can be legalism’s best friend, in the same way that formalism is often its worst foe.
I look forward to your comments!
Chillin'Competition 
This is a very important post. Thanks Pablo for sharing your insightful thoughts
Mario Solimani
14 September 2018 at 6:51 am
Interesting post, Pablo. I am quite critical of the consumer welfare standard interpreted as (allocative) efficiency, but I’m not sure that discretionism is the inevitable alternative: other standards may leave just as little (or as much) discretion as the consumer welfare standard, depending on how they are interpreted.
Indeed, as you point out, the consumer welfare standard is itself ambiguous: it can be interpreted in a discretionary or a legalistic way. Originally, the consumer welfare standard was meant to be all about effects analysis (as opposed to the formalism of earlier years) but, since this was too difficult, “proxies” were invented (Easterbrook already talked about this in 1984, even before the consumer welfare standard was firmly established). And so we now have a new formalism which is basically concerned with pricing being below or above (average variable) cost. This formalism suffers from the same ills as the old formalism namely that, as you say, reality is messier than formalism allows.
There will always be a tension between legalism and discretionism in regulatory fields, but when standards are not or no longer useful or succesful in achieving the objectives of a law (as they are understood today, and in the case of the allocative efficiency standard, I would argue, also as they were understood when antitrust legislation was first enacted), new standards will be invented.
Jan Blockx
14 September 2018 at 9:25 am
Dear Pablo, that is a very interesting matrix.
However, just as Jan, I disagree that being critical of the consumer welfare standard means being in favour of a more discretion based approach (you can put me in what you call the “formal legalists” box in your matrix). I believe that economics should play an essential role in forming the rules we apply in competition law, and for our understanding of these rules. But once we have rules based on sound economics, we should apply those “form based” rules. To reduce competition law to the application of cost-based tests leads, IMO, all to often to results that are neither appropriate nor, in practice, practicable/foreseeable. They have their place in some instances (namely, predatory pricing), but generally, I’m rather sceptical.
More fundamentally, as a German ordo-liberal (oh no, one of those!), I see competition law as the constitution of the free market, which is a basic subsystem of any free society. The value of freedom is not limited to maximising the output of food on everyones table (i.e. consumer wellfare). Competition law protects freedom of entrepreneurship as well. Without it, neither our economy nor our society is free. Yes, experience tells us that unfree societies are not very efficient. But even if they would be, I still wouldn’t want to live in a society which, while feeding me very efficiently, keeps me in chains nonetheless. From this perspective, the – admittedly very catchy – slogan of “protect competition, not competitors” is just silly (and very dangerous). It’s like saying, “protect the freedome of the press, not journalists”. We shouldn’t sacrifice the freedom of our undertakings to compete on the merits for reasons of (oftentimes: merely alleged) efficiency. It’s just another road to serfdom.
Moritz Am Ende
22 October 2018 at 4:48 pm
Moritz,
I think if you are in favour of economics-based rules, you should find yourself in the “substance legalist” box. The best illustration is the per se prohibition of hardcore cartels. The substance of this rule is informed by economics. But it is a “legalist” rule since it also prohibits cartels which fail to produce any harm (of which there may be more thana few, if you look at analyses of the effectiveness of legal cartels, e.g. export cartels under the US NIRA). In spite of that fact, that per se rule was not challenged by the Chicago school.
HK47
24 October 2018 at 1:01 pm
Thanks everybody for the comments! It is great that the post has generated some debate.
I would agree with HK47. If one accepts the role of economic analysis in formulating rules and standards, then one is not a formalist, but a ‘substance legalist’.
The rule on predatory pricing is another great example of a ‘substance legalist’ approach. It is a rule like the ones Moritz prefers: pricing below AVC is prohibited, full stop – no questions asked. This rule, however, is informed by economic analysis (pricing below AVC makes no economic sense if it is not for the exclusion of competition.
Since we talk about AKZO: the ‘as efficient competitor’ standard is already present in this case (see para 72). The principle according to which EU competition law is only concerned with the exclusion of efficient rivals is at the heart of the case law. You may dislike it, Moritz, but reneging on the ‘as efficient competitor’ standard means reneging on the case law (or advocating a substantial change in the law). From the perspective of positive competition law analysis, it is not simply a matter of opinion or personal preferences.
Once again: thanks very much! And apologies for joining the debate so late!
Pablo Ibanez Colomo
29 October 2018 at 5:26 pm
Dear Pablo,
Indeed this is a topic to raise the heartbeat of any devoted competition lawyer! 🙂 I think there are a lot of issues with terminology that muddy the debate, which is also why I like your matrix so much. Even if we still seem to disagree in which box I belong. 🙂
Since you are mentioning AKZO (which I had indeed in mind): I would argue that AKZO introduced cost based tests because one needs to draw the line between legitimate competition on the merits – and price competition is at the heart of it, even if it eliminates a competitor – and predatory conduct which goes beyond. AKZO is a great case to show what predatory pricing is about. And the rule established by the ECJ makes sense even for a “formal legalist”: Once a firm goes below AVC, there is very good reason to assume they did it (only) to eliminate a rival. What’s abusive here from my formal (?) legalist approach is that the dominant company doesn’t act for it’s own immediate benefit anymore (where the crushing of the rival is only a side effect), but primarily to hurt other, weaker market participants. In particular if the obvious aim pursued is not to make a profit, but to maintain or strenghten a dominant position. Still I could easily imagine a situation where a firm prices below average variable costs but is still acting in its own, immediate (legitimate) self-interest. Imagine if shutting down a production line and firing it up later would involve significant additional costs, and the undertaking has good reason to believe that prices will recover before the savings from avoided below cost sales exceed the additional costs linked to shutting down the production line. In such a situation, I would argue that the assumption (or, procedurally: presumption) of abusiveness of below AVC sales may be rebutted.
To me, there is a difference to having an as-efficient-competitor test for predatory pricing – where it makes a lot of sense to draw a sensible line between “good price competition to crush your rivals” and “bad price competition to crush your rivals” and other areas such as conditional rebates. From a legalist perspective, I don’t need an as-efficient-competitor test to make a case why loyality rebates from dominant companies are problematic (hint: the rebate is not the problem). I would still be open to (economic) arguments that in some instances, even dominant undertakings may have legitimate reasons to impose some sort of exclusivity. But then I would like to see those reasons in the actual case and that they are good enough to justify the conditionality. Objective justification all the way. It’s a very different argument than as-efficient-competitor (which I see as the core of what you call the “substantive legalist” approach).
Yes I know that the ECJ has gradually adopted the as-efficient-competitor test to areas of competition law other than predatory pricing, and has now, for all practical purposes, generally endorsed it in Intel. That’s a fact, I know. But IMO, a very deplorable one. I don’t have to agree with everything the ECJ says; just as the proponents of the so-called “more economic approach” didn’t agree with the case law before.
Anyway, your original point was the distinction between discretionalists and legalists, which is a very useful observation. And it seems that we all agree that we should stick with the legalist approach (either form or substance based). Jan and I only disagreed insofar as you argued that the “consumer welfare standard wars” are over. As you see, the resistance is willing to fight on ;).
Cheers,
Moritz
Moritz Am Ende
29 October 2018 at 7:53 pm
Thanks, Moritz for sharing your thoughts!
I would say it is important to distinguish between the ‘as efficient competitor test’ and the principle whereby competition law should only be concerned with the exclusion of equally efficient rivals.
I can see the arguments against applying the ‘as efficient competitor test’ to rebates; these arguments are compelling. At the same time, I am convinced that any legal test should be based on the principle, endorsed by the Court in Intel, that competition law is only concerned with rivals that are as efficient as the dominant firm.
As to the legal test that should apply to rebates: as I explain in the presentation, the case law before Intel was, first and foremost, problematic from the perspective of a legalist: it failed to provide legal certainty, it did not treat like practices alike and gave, in practice, discretion to authorities and courts. Something needed to change (and it was clear it would, sooner or later).
The case law could have moved in two directions. It could have moved in the direction you may prefer, i.e. by making all rebates prima facie prohibited (subject to an objective justification), or it could have moved (as it did eventually) in the direction of a case-by-case analysis (which does not necessarily mean that the ‘as efficient competitor test’ is applicable). The ‘substance legalist’ approach, which the Court embraced in Intel, is the second.
Pablo Ibanez Colomo
31 October 2018 at 11:30 am