Tim Brennan finds the new shift in antitrust thought and enforcement connected to the Neo-Brandeisian movement to be flawed for the most part. However, he writes that a reinvigorated focus on tacit collusion, which some have blamed for the rise of prices for groceries and apartment rents, may deserve consideration and further study.
This article is part of a symposium studying the “paradigm shift” in antitrust scholarship and policy. Inspired by philosopher Thomas Kuhn’s work on progress in science, this symposium asks if and how the tenures of Federal Trade Commission Chair Lina Khan, Department of Justice Assistant Attorney General Jonathan Kanter, and scholarship associated with the antimonopoly or Neo-Brandeisian movement has changed how we understand the priorities of antitrust enforcement, evidence of anticompetitive harm, and the study and enforcement of antitrust more broadly. Over the next few days, we will publish contributions from Tim Brennan, Eleanor Fox, Daniel Francis, Andrew Gavil, Richard Markovits, John Mayo, Steven Salop, and Randy Stutz. You can read the previously published articles here. ProMarket encourages our readers to respond to the symposium and the ideas these scholars put forth with their own. Responses can be sent to ProMarket@chicagobooth.edu.
It is no headline at this time that antitrust enforcement policy has taken an acute turn. Drivers of this turn, or “paradigm shift,” include not only more active enforcement in some areas that proponents argue have seen too little attention, such as labor markets. These drivers also include the “Neo-Brandeisian” incorporation of values in antitrust behind consumer welfare and economic efficiency that have driven antitrust policy for the last half century.
This turn has received extensive criticism in some quarters, including from me. I will summarize some of those criticisms below. That said, it is responsible not to drain the bath water without checking for a baby: Might there be some practice to which the prior “consumer welfare” standard had provided too little attention? One candidate is “tacit collusion”—the claim that sellers (or buyers) in a market might collectively raise (or lower) prices without an explicit agreement, with the attendant reduction in sales (or purchases).
Any relative neglect of tacit collusion may be a consequence of the absence of a clear and unambiguous framework. With apologies for a slight theoretical excursion, I will briefly suggest a couple of reasons for this absence. I will then take a look at a few areas of current enforcement interest to see if invigorated attention to tacit collusion might be useful. I find that the answer is possibly yes, often no. Finally, I conclude with some thoughts on enforcement criteria and tactics.
Careening around the acute turn
Before getting to what might be right in the paradigm shift in antitrust thought and enforcement, briefly stated criticisms of that turn may be useful. The first two are general, the latter two apply to specific substantive areas.
Multiple objectives. A hallmark of the new turn is the belief that antitrust enforcement should be governed by a host of aims other than consumer welfare and economic efficiency, which have served as legal and policy guideposts for decades. In an Antitrust Bulletin article published before the current United States antitrust enforcers came into their positions, I found a dozen such objectives that critics of the traditional approach had proposed, including income equality, environmental sustainability, small business protection, and media veracity.
Many of these objectives may be quite worthy to pursue. However, antitrust cases are not an effective way to address broader societal concerns better addressed through taxation or regulation. Moreover, attempting to do so through antitrust is likely to dilute its ability to fulfill its unique role as promoting economic welfare through protecting the competitive process. A multiple objective approach also requires guidance as to how to balance these goals against each other. For example, how much higher should prices get by reducing competition from lower-cost large firms in order to promote small businesses? Failure to provide such guidance makes it impossible for judges and juries to decide cases. Proponents of the multiple objective approach could argue that there are no trade offs with consumer welfare, but if so, then they have no reason to advocate pursuing anything other than consumer welfare.
Politicization. An inevitable byproduct of adding objectives to antitrust is that which objectives should be pursued, and how those pursuits should be balanced, become a matter for political resolution. This can be good and, in many cases, appropriate, when policy contexts require consideration of multiple objectives. An example may be communications regulation, which may have to consider both competition and social values associated with ability to share and access information. However, antitrust is supposed to provide stable guideposts on permissible business conduct. Politicizing that process threatens to remove that stability, reducing overall economic welfare and quite possibly impeding the alternative objectives that led to politicization in the first place.
Vertical integration. Books have been written about this, so I can be brief. Long-standing criticisms of the turn toward a focus on vertical antitrust cases, in which firms that operate at different points of a supply chain merge or otherwise integrate, rest on two related premises. One is that vertical integration and other contracting practices affect how firms are organized, but not how they compete with each other. The second is that vertical combinations reduce incentives to raise prices, since doing so by one side reduces the profits of the other. In contrast, mergers of competitors (horizontal integration) increase incentives to raise prices because the profit effect goes in the opposite direction.
Certainly, there can be reasons to investigate cases of vertical integration. Where vertical integration enables a firm to exercise market power that they already had but could not exploit, they merit scrutiny; regulatory evasion is a good example. However, if a firm can already exploit that power directly, vertical integration adds no harm at the margin and may promote economic efficiency.
A final point is that to provide useful guidance to business, antitrust should recognize the virtue of simplicity. Allowing vertical integration that does not increase existing market power or facilitate exercise of existing but unexploited market power would promote simplicity with no apparent competitive harm.
Labor. A second specific area receiving added attention in the current antitrust policy environment is the possibility that a merger or practice could promote the ability to exercise monopsony power against labor, driving down employment and wages. To my knowledge, the traditional approach did not preclude enforcement predicated on this concern. Any reluctance to prosecute such cases may have sprung from an inability to delineate a labor market over which a set of firms might have market power, or difficulty in finding evidence that firms did or would reduce hiring in order to drive down price—necessary in the same way that a monopolist has to reduce output to raise prices.
Tacit collusion: A renewed consideration?
With all of these criticisms, one might just give up on justifying a paradigm shift in antitrust practice, but it is important to see if the traditional approach has missed something that the new perspective might highlight. One that comes to mind is tacit collusion. To understand the potential for addressing it, we need to begin with why it had been neglected.
An long-standing aphorism that “an economist [is someone] who, when told something works in practice, is concerned whether it works in theory.” While this looks like a joke about economists ignoring reality, it reflects the importance that theory gives to explaining whether something we think we see is in fact what we see.
Tacit collusion, in which firms coordinate to set prices or output without an explicit agreement, may fit this aphorism. It superficially is hard to explain because any competitor generally has a profit incentive to undercut its rivals and capture their business, leading to the competitive outcome. The major theory to support collusion without an agreement involves repeated games, where a decision to undercut rivals today would lead them to revert to that competitive outcome, with lower profits for all. This theory has two problems. The first, known as the “folk theorem,” is that virtually any outcome, not just the monopoly outcome, could come about with repeated games. The second, the “renegotiation problem,” is that the theory does not preclude the firms from reengaging in tacit collusion after one firm undercuts the others. The undercut firms may choose to reset prices or output to match the new level rather than punish the undercutting firm and themselves by reverting to the competitive outcome.
These theoretical flaws don’t mean that tacit collusion isn’t real, only that we do not have a good theory for it. Tacit collusion is the basis for enhanced concern with mergers to acquire and put out of business a “maverick” firm that routinely prices below others in a market. Tacit collusion could also explain why divestitures as merger remedies intended to restore competition may not have the intended benefits (although that evidence is not without debate).
Maybe; maybe not
Might tacit collusion be relevant in some potential areas of current concern?
Groceries. Food prices are on the minds of many, and some attribute high food prices to anticompetitive pricing by grocery chains. Without evidence of an explicit agreement, any such pricing would have to reflect tacit collusion. With the hundreds or thousands of products that groceries offer, and in many cases with rapidly changing ingredient and wholesale prices, a tacit (or explicit) agreement would be unlikely. A possibility, however, might be a tacit agreement among grocery chains not to open otherwise profitable stores close to stores of others, thus giving each store local monopolies to some degree. Those concerned with tacit collusion among grocers might look for evidence along these lines.
Pharmaceuticals. Another sector of concern regarding prices is pharmaceuticals. A particular concern is that the three leading pharmaceutical drug benefit managers (PBMs)—who set up pharmacy networks, negotiate drug prices, and determine drug availability for entities that provide pharmaceutical insurance—raise prices by favoring more expensive drugs over lower cost alternatives. Given the vast number of available drugs, like grocery items, tacit collusion does not seem likely. A more plausible anticompetitive tactic might be for a drug company to pay PBMs for exclusive or preferential treatment, making the drug company the potential monopolist.
Apartments. Another recent case involves apartment owners in an area using a common service to propose rental rates for their units based on the unit’s characteristics. An intriguing aspect about this case is that it may avoid the aforementioned flaws in the repeated game model. By producing a solution, the service’s calculation algorithm avoids the “folk theorem” indeterminacy. In addition, as I understand it, it is not easy for an apartment owner to leave the system and later return, perhaps preventing “restarting” that would otherwise eliminate the threat of punishment for undercutting. I need to note that I am not acquainted with either the facts or the potential benefits of having a service calculate rates. Nevertheless, the potential use of this service, and perhaps algorithmic pricing generally, could be a response to the shortcomings of repeated game theory.
Non-compete agreements. A final prominent area of concern is when an employer requires employers to agree not to work later for a competitor. It is not difficult to understand why an employer would not want an employee to take to a competitor the knowledge and connections the employer provided. Absent a non-compete agreement, the employer may be discouraged from providing employees with information that would benefit the employer, employee, and consumers, by enhancing service quality and competition. However, an explicit agreement among employers to refuse to hire each other’s employees could lead to monopsony against particular groups of workers. One might imagine that industry-wide non-compete agreements could be the result of tacit monopsony collusion, without having to come up with salaries for each category of worker that industry hires. It cannot be said enough, though, that situation-specific facts matter.
Enforcement suggestions
If tacit collusion is to become an explicit antitrust doctrine, two things may need to change. One is a potential gap in the antitrust law. Neither Section 1 or Section 2 of the Sherman Act may allow prosecutions based on tacit collusion. The former requires an “agreement,” but the defining characteristic of tacit collusion is the lack of an agreement. The latter applies to acts of “monopolization,” but that is usually taken to refer to a unilateral action to create or maintain market power, not something a group of firms does without an explicit and agreed upon action.
Tacit collusion could play a role in merger analysis, where a criterion for illegality in Section 7 of the Clayton Act is when a merger “may tend to inhibit competition.” However, we need some method for determining when a merger would make tacit collusion sufficiently likely to credibly threaten competition. Merger simulations will not be relevant, because they rely on firms individually maximizing profits rather than tacitly colluding. At this stage, enforcers are likely to require documentary evidence of tacit collusion in merger cases—and in Sherman Act cases as well if courts find they apply.
Authors’ Disclosures: the authors reports no conflicts of interest. You can read our disclosure policy here.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.
