Tragedy of the Successful Firm: Q&A with Konstantinos Stylianou
Dr. Konstantinos Stylianou is an Assistant Professor in Competition Law and Regulation and the Deputy Director of the Centre for Business Law and Practice at the University of Leeds. His areas of focus are communications networks, digital markets, blockchain, and high tech industries in general. Dr. Stylianou holds an S.J.D. from the University of Pennsylvania, an LL.M. from Harvard and both a Bachelor’s and Master’s from Aristotle University.
DisCo spoke with Professor Stylianou to discuss his article, The Tragedy of the Successful Firm, and we summarize that conversation in this blog post. We recommend reading the full article for a deeper dive into some of the topics discussed in this Q&A session.
DisCo: In your article, you focus on the recent decisions regarding Google Search and Google Android and how both of these decisions were critical opportunities for the EU to have provided some concrete guidance on how platform firms could stay compliant with antitrust laws. Which areas do you think European competition enforcers must address if they are to guide innovative firms toward success without fear of undue punishment?
Stylianou: Google Search and Google Android are such important decisions that present a great opportunity to revisit everything that’s wrong with EU competition law. That’s not to say that they are bad decisions, only that they are so far-reaching that they inevitably touch upon most chronic weaknesses of competition enforcement. And the kind of weaknesses I have in mind do not depend on whether one is conservative or neo-Brandeisian or other; they transcend ideology, and hopefully we can all agree that they need to be fixed. I think the four main weaknesses are: a) the lack of recognition of online platforms’ systemic features; b) the enduring influence of the nebulous concept of special responsibility; c) the circular definition of abuse of dominance; and d) the disconnect between the chosen goals of competition law — whatever those might be — and the theory of harm employed in each case.
DisCo: In your article, you argue that platform ecosystems in digital markets may present novel features that are uncommon in other sectors, and therefore an elevated measure of attention is required to acknowledge them. What are some features that are uncommon, and why do you think they deserve an elevated measure of attention?
Stylianou: One thing that has remained severely unappreciated in online platform systems is their systemic nature. There is a gigantic amount of literature in engineering, business, and science and technology studies that discusses the intricacies of so-called large technical systems (LTS) that comprise multiple highly interconnected parts, so that changes in one part may have unanticipated consequences for other parts and for the general operation of the system as a whole. This literature has remained absent in legal analyses. LTS often benefit from having an authority, an orchestrator that coordinates the system’s operation, maintenance and evolution. Think of platforms like Symbian and i-mode (in Europe) that collapsed in large part due to a lack of cohesion and coordination. And even systems that we tend to think of as open and decentralized, like Linux for example, have in fact central points of control — Linus Torvalds is affectionately known as a benevolent dictator. But it is this kind of pervasive control that antitrust authorities often fear, and since online platforms, and especially the big ones that antitrust authorities would care to go after, can well constitute LTS, we see them becoming antitrust targets. The fears may well be justified, but what is missing is at least a recognition of and a discussion around the benefits of pervasive control and the risks of upsetting the systemic balance of LTS. The Google Android case was a perfect testbed for that analysis, but sadly also a foregone opportunity.
DisCo: You also discuss the problem with the concept of “special responsibility” under EU competition law. In short, why is “special responsibility” such a controversial concept?
Stylianou: The controversy around special responsibility is well documented: the concept is ill-defined, it adds nothing to the rationale of the Commission’s and the Court’s case-law, it is conflated with the concept of abuse of dominance, and it is unfair to dominant firms. What the special responsibility doctrine asks is fundamentally irreconcilable with itself: firstly, it asserts that “as a result of the mere presence of a dominant undertaking competition is weakened” and then it imposes the special responsibility on dominant firms to not “further weaken competition”. What, then, can a dominant firm do if its mere existence weakens competition and therefore any action on its part will further weaken it, unless the firm indeed turns against itself? Obviously, this impasse was not the intention of the Court when it came up with the concept. But it is now its own special responsibility to clarify the concepts’ contours.
DisCo: Dominant firms only violate competition law if they abuse their position. What is problematic with the EU’s concept of abuse and how does this affect firms who utilize innovative business models?
Stylianou: If you look at all the definitions of abuse provided by the Commission and the Court you will notice that they are circular. The whole point of the concept of abuse of dominance is to help us determine when a practice harms competition in a way that competition law disapproves of. But to define abuse of dominance as employing practices that “are detrimental to [consumers] through their impact on an effective competition structure,” that “substantially fetter competition,” that constitute “recourse to methods different from those which condition normal competition” or “recourse to means other than those within the scope of competition on the merits” is tautological because these definitions contain the very elements the definitions are supposed to be used to prove, i.e. what is “detrimental,” what is “normal,” what is “substantially fetter competition” etc? And the more novel and experimental a business model is, which is often the case with online platforms, the more difficult it becomes to predict whether it will run afoul of any of these tautologies. Alternative formulations employed by US courts or by scholars could be adopted by the Court and the Commission but these too are not without their problems, and that’s a different story.
DisCo: The goals of competition law have been varied — including but not limited to efficiency, consumer welfare, the process of competition per se, and so forth — but you argue that these goals are not necessarily always aligned. How might conflicting goals affect businesses, and especially platform business models?
Stylianou: Perhaps the most famous quote in antitrust is Robert Bork’s opinion that “antitrust policy cannot be made rational until we are able to give a firm answer to one question: What is the point of the law — what are its goals? Everything else follows from the answer we give.” It doesn’t have to be the case that antitrust policy has only one goal; in fact, with the resurgence of a more social construction of antitrust, a plurality of goals tends to become the norm. The problem is that they can’t all be equally served at the same time, and at that point one has to pick, choose and combine. Obviously, it does not serve legal certainty to not know ex ante what goals a competition authority or court will seek to achieve. But, besides listing and ranking their priorities, enforcers and courts can and should also make sure that for every theory of harm they identify in any given case, they link it to the antitrust goal it seeks to achieve. Otherwise, it is impossible to determine whether antitrust law is applied pursuant to its stated goals, and if that’s not established, then the whole construct collapses. We haven’t traditionally seen this in cases, but then again, we never needed to as much as we do now, because it is the first time in history that antitrust has been associated with so many goals and it has matured enough to have clear analytical frameworks where the relevant goals can be inserted in and related to well-developed theories of harm.