This is a rather long piece – which you can find here – that tries to understand how antitrust should be applied in the context of the sharing economy.
I think the spur for this piece is the recent price-fixing case brought against Uber in New York. Regardless of the incentives for writing the paper, it tries to identify the various approaches that antitrust can adopt regarding digital platforms and to determine which one is better suited. The paper also argues that: “Unique to sharing economy enterprises is a structure that approaches a single entity while remaining a set of agreements among individual actors. This structure results in a sharing of economic risks among the participants in a sharing economy enterprise which can incentivize efficiencies in operation that ordinarily are found in a single entity. The article concludes that those efficiencies can overcome anticompetitive concerns about coordination on competitively sensitive matters.”
The paper begins by observing that: “antitrust law has evolved over more than a century to regulate a market place occupied by firms. It uses a complex set of rules to assess the economic effects of agreements between firms, but leaves conduct within a firm subject to little scrutiny. In a traditional firm, owners contribute capital, managers make business decisions and employees implement those decisions. The sharing economy confronts the antitrust law by disaggregating the roles usually played by actors within a firm.” This is important because antitrust is used to looking at some practices individually, and will likely now need to take into account the particular circumstances of digital platforms. For example: “Suppliers in a sharing economy enterprise reach agreement with the platform company about terms of service [and] nearly always include systems for making and receiving payment; impose requirements for suppliers’ relationships with consumers; and establish a means for consumer reviews.(…) Some terms are competitively neutral or procompetitive. Others, particularly including agreements on price, are typically considered matters of substantial antitrust concern.”
Antitrust law looks at such practices in a variety of different ways: “One approach is to consider the agreements between suppliers and the platform to be “vertical agreements” subject to a fairly lenient rule of reason. Another approach is to treat the agreements as a hub-and-spoke conspiracy, reflecting a horizontal agreement among suppliers to the enterprise orchestrated by the platform. Agreements on price, one of the most competitively sensitive terms, may even be automatically illegal.” [emphasis is mine]
In developing its argument, the paper is structured as follows:
- Part I seeks to describe the sharing economy and how it is distinct from the traditional economy. It begins by providing a comprehensive review of literature on the “sharing economy”, and discusses the seven characteristics that the FTC found in its report on the sharing economy: (1) Three players: buyers, sellers, and platform; (2) Atomistic buyers and sellers; (3) Sellers face minimal entry and exit barriers by avoiding specialized investment; (4) Thick markets; (5) Platforms provide assurance of safety and reliability; (6) Platform charges fee or commission; (7) Platforms commonly cause regulatory disruption. The paper then goes in depth into the contractual terms underpinning a number of digital economy players, such as Uber, Airbnb, Lyft, Couchsurfing and Task Rabbit.
Following this descriptive section, the paper looks at the economics of digital platforms. In particular, it finds that these platforms benefit from a number of efficiencies, including efficiencies of: the online platform; low search and transaction costs; end-running regulation designed to control centrally coordinated unitary firms; small scale. Ultimately, these efficiencies may approximate the market from the competitive ideal, to the point of eliminating the transaction costs that underpin the formation of firms under classic Coasean theory.
- Part II examines the antitrust principles under Section One of the Sherman Act for assessing agreements between firms. This is straightforward legal analysis, and it will be interesting mainly for those interested in getting a grasp of how Section 1 works. The most interesting elements of this analysis, for our purposes here, are that: (i) the prohibition does not extend to intra-firm transactions; and that (ii) cooperation through agreements that are subject to a per se rule are strongly discouraged since a court will not listen to any procompetitive arguments in support of such agreements.
- Part III seeks to determine how competition principles should be applied in sharing economy contexts. Theoretical insights are tested, throughout, by reference to Uber. I will review this section in a bit more detail, given its obvious relevance.
Agreements between drivers and Uber seem to be vertical contracts, which do not per se raise competition issues. While the structure of Uber could be seen as amounting to a hub-and-spoke mechanism of horizontal coordination between drivers, it is argued that Uber is merely a matching service that does not involve a set of vertical resale price maintenance agreements as in most hub-and-spoke cases. The authors also distinguish sharing economy platforms from other horizontal practices, such as membership of an organisation, joint ventures, and joint sales agency.
If one were to conclude that Uber drivers are in agreement with each other, a second step of the analysis would be to determine whether this amounts to prohibited price fixing. While this prohibition is usually applicable to cartels, there is very low tolerance towards any type of horizontal agreement which affects prices. Price-fixing seems nonetheless to be allowed when it is necessary for a product to exist at all (such as broadcast rights for some sports, or blanket licenses to a copyright pool), which is something that a platform may sometimes claim.
This exception is part of the wider doctrine of ancillarity – i.e. that restraints of trade (such as price fixing) should be lawful if they reasonably further the purposes of a lawful contract. For a service such as Uber: “the tougher question under the ancillarity doctrine is whether the agreement to charge the price determined by the Uber pricing algorithm is reasonably ancillary to the matching service provided by Uber”. The reason for this is that: “The reasonableness of the connection between a price agreement and lawful contractual purpose varies by context”. Assessing this reasonableness ultimately requires a counter-factual which is difficult to determine.
At this point, the paper goes into a fairly technical (US law) discussion of who bears the burden of proof if the question of the legality of Uber’s practices is to be assessed under a rule of reason. In short, the application of the rule of reason requires the shifting of burdens of proof between the plaintiff and the defendant on three questions. The plaintiff bears the initial burden of demonstrating the likelihood of anticompetitive effects. If the plaintiff establishes that likelihood, the burden shifts to the defendant to present a procompetitive justification. If the defendant succeeds in this, the burden shifts back to the plaintiff to show that the anticompetitive effects outweigh the procompetitive effects.
In short, and as a result, the authors conclude that when applying black-letter antitrust law to different sharing economy enterprises, the terms imposed by different platforms will influence their antitrust treatment.
- Part IV argues that the principles discussed in Part III should be altered for application in the sharing economy. It does this in two stages:
- Subpart A establishes a comprehensive framework for distinguishing between sharing economy enterprises based on two variables: the degree of risk sharing among economic agents, and the extent and subject matter of the coordination among them. To begin, the paper argues that antitrust rules categorically consider that conduct within one single economic entity (the firm) is exempt from antitrust scrutiny; and that some types of coordination between such entities is prohibited per se. However, this is said to be inappropriate for the sharing economy, which gave origin to business structures that break down the structure of the firm into smaller components, because it is more efficient for these components to cooperate between themselves than to aggregate themselves into a unitary corporate figure. Instead, the authors suggest that: “Coordination of behavior by potentially competing economic actors threatens the efficiency that competition promotes(…). On the other hand, risk sharing by economic actors has the potential to promote efficiency.” As a result, they argue that “the extent of coordination should be viewed as a matter of degree and compared to the extent of risk sharing. Even price fixing among competitors may be acceptable if the degree of risk sharing is sufficiently extensive. (…)
- Subpart B draws generalizable lessons regarding the application of antitrust to the sharing economy. In particular, and given platforms’ structures – which, as the authors admit, may impose more or less control on the transactions that take place on the platform – the authors consider that platforms should not be subject to per se treatment, but should instead be subject to quick look scrutiny under the rule of reason. “The closer in form to Uber the more searching the rule of reason analysis, with Lyft (with its established, but alterable, price terms) close to the Uber-level scrutiny and Airbnb (with its minimal control over transaction terms) close to the hands-off approach accorded to Task Rabbit. This rule of reason inquiry, a matter of degree rather than a categorical distinction, requires substantial judicial inquiry into the competitive sensitivity of the subject of the agreement.” Furthermore, part of this assessment will require, subsequent to an assessment of the competitive sensitivity of a particular set of coordinating agreements, a judgement concerning the degree of risk sharing among the suppliers.
This is a thoughtful piece, even if it focuses a bit too much on the Uber case at times. The authors’ focus on an individual case instead of on the general class of sharing economy businesses is reflected in the absence of conceptual clarity in the remedies the author propose – how are courts to identify sharing economy businesses? What are the elements, and normative benchmarks, for determining whether certain contractual arrangements infringe competition law or not (e.g. how do we measure risk, coordination, and their ration)? What is the relationship between risk-taking, coordination and consumer welfare (if any)?
While I am attracted to the boldness that the authors demonstrate in their attack of the conceptual foundations of antitrust enforcement (e.g. single economic entity, price fixing, etc.), I am not sure that such courage is required here. I fail to see why the ancillarity doctrine would not suffice in this context (even if I’m aware of the limitations that this doctrine has when circumscribing per se prohibitions). After all, many practices where price-fixing occurs do not fall foul of antitrust precisely because they are perceived to be pro-competitive on the whole. I don’t understand why a rule of reason approach would not conclude that such practices are ultimately welfare enhancing. And I surely don’t see how the benefits of the authors’ proposal can outweigh the uncertainty that their proposals would introduce into the law if adopted.
I do think, however, this is a very interesting contribution to work on the limitations of current antitrust doctrines – and on the potential need to update these doctrines – in the face of a proliferation of innovative business models. In particular, I am fond of the authors’ conceptual grid: it provides a very useful tool to think about the foundations of antitrust law, and about whether we may need to rethink those foundations.