Archives For

In our first post, we discussed the weaknesses of an important theoretical underpinning of efforts to expand vertical merger enforcement (including, possibly, the proposed guidelines): the contract/merger equivalency assumption.

In this post we discuss the implications of that assumption and some of the errors it leads to — including some incorporated into the proposed guidelines.

There is no theoretical or empirical justification for more vertical enforcement

Tim Brennan makes a fantastic and regularly overlooked point in his post: If it’s true, as many claim (see, e.g., Steve Salop), that firms can generally realize vertical efficiencies by contracting instead of merging, then it’s also true that they can realize anticompetitive outcomes the same way. While efficiencies have to be merger-specific in order to be relevant to the analysis, so too do harms. But where the assumption is that the outcomes of integration can generally be achieved by the “less-restrictive” means of contracting, that would apply as well to any potential harms, thus negating the transaction-specificity required for enforcement. As Dennis Carlton notes:

There is a symmetry between an evaluation of the harms and benefits of vertical integration. Each must be merger-specific to matter in an evaluation of the merger’s effects…. If transaction costs are low, then vertical integration creates neither benefits nor harms, since everything can be achieved by contract. If transaction costs exist to prevent the achievement of a benefit but not a harm (or vice-versa), then that must be accounted for in a calculation of the overall effect of a vertical merger. (Dennis Carlton, Transaction Costs and Competition Policy)

Of course, this also means that those (like us) who believe that it is not so easy to accomplish by contract what may be accomplished by merger must also consider the possibility that a proposed merger may be anticompetitive because it overcomes an impediment to achieving anticompetitive goals via contract.

There’s one important caveat, though: The potential harms that could arise from a vertical merger are the same as those that would be cognizable under Section 2 of the Sherman Act. Indeed, for a vertical merger to cause harm, it must be expected to result in conduct that would otherwise be illegal under Section 2. This means there is always the possibility of a second bite at the apple when it comes to thwarting anticompetitive conduct. 

The same cannot be said of procompetitive conduct that can arise only through merger if a merger is erroneously prohibited before it even happens

Interestingly, Salop himself — the foremost advocate today for enhanced vertical merger enforcement — recognizes the issue raised by Brennan: 

Exclusionary harms and certain efficiency benefits also might be achieved with vertical contracts and agreements without the need for a vertical merger…. It [] might be argued that the absence of premerger exclusionary contracts implies that the merging firms lack the incentive to engage in conduct that would lead to harmful exclusionary effects. But anticompetitive vertical contracts may face the same types of impediments as procompetitive ones, and may also be deterred by potential Section 1 enforcement. Neither of these arguments thus justify a more or less intrusive vertical merger policy generally. Rather, they are factors that should be considered in analyzing individual mergers. (Salop & Culley, Potential Competitive Effects of Vertical Mergers)

In the same article, however, Salop also points to the reasons why it should be considered insufficient to leave enforcement to Sections 1 and 2, instead of addressing them at their incipiency under Clayton Section 7:

While relying solely on post-merger enforcement might have appealing simplicity, it obscures several key facts that favor immediate enforcement under Section 7.

  • The benefit of HSR review is to prevent the delays and remedial issues inherent in after-the-fact enforcement….
  • There may be severe problems in remedying the concern….
  • Section 1 and Section 2 legal standards are more permissive than Section 7 standards….
  • The agencies might well argue that anticompetitive post-merger conduct was caused by the merger agreement, so that it would be covered by Section 7….

All in all, failure to address these kinds of issues in the context of merger review could lead to significant consumer harm and underdeterrence.

The points are (mostly) well-taken. But they also essentially amount to a preference for more and tougher enforcement against vertical restraints than the judicial interpretations of Sections 1 & 2 currently countenance — a preference, in other words, for the use of Section 7 to bolster enforcement against vertical restraints of any sort (whether contractual or structural).

The problem with that, as others have pointed out in this symposium (see, e.g., Nuechterlein; Werden & Froeb; Wright, et al.), is that there’s simply no empirical basis for adopting a tougher stance against vertical restraints in the first place. Over and over again the empirical research shows that vertical restraints and vertical mergers are unlikely to cause anticompetitive harm: 

In reviewing this literature, two features immediately stand out: First, there is a paucity of support for the proposition that vertical restraints/vertical integration are likely to harm consumers. . . . Second, a far greater number of studies found that the use of vertical restraints in the particular context studied improved welfare unambiguously. (Cooper, et al, Vertical Restrictions and Antitrust Policy: What About the Evidence?)

[W]e did not have a particular conclusion in mind when we began to collect the evidence, and we… are therefore somewhat surprised at what the weight of the evidence is telling us. It says that, under most circumstances, profit-maximizing, vertical-integration decisions are efficient, not just from the firms’ but also from the consumers’ points of view…. We therefore conclude that, faced with a vertical arrangement, the burden of evidence should be placed on competition authorities to demonstrate that that arrangement is harmful before the practice is attacked. (Francine Lafontaine & Margaret Slade, Vertical Integration and Firm Boundaries: The Evidence)

[Table 1 in this paper] indicates that voluntarily adopted restraints are associated with lower costs, greater consumption, higher stock returns, and better chances of survival. (Daniel O’Brien, The Antitrust Treatment of Vertical Restraint: Beyond the Beyond the Possibility Theorems)

In sum, these papers from 2009-2018 continue to support the conclusions from Lafontaine & Slade (2007) and Cooper et al. (2005) that consumers mostly benefit from vertical integration. While vertical integration can certainly foreclose rivals in theory, there is only limited empirical evidence supporting that finding in real markets. (GAI Comment on Vertical Mergers)

To the extent that the proposed guidelines countenance heightened enforcement relative to the status quo, they fall prey to the same defect. And while it is unclear from the fairly terse guidelines whether this is animating them, the removal of language present in the 1984 Non-Horizontal Merger Guidelines acknowledging the relative lack of harm from vertical mergers (“[a]lthough non-horizontal mergers are less likely than horizontal mergers to create competitive problems…”) is concerning.  

The shortcomings of orthodox economics and static formal analysis

There is also a further reason to think that vertical merger enforcement may be more likely to thwart procompetitive than anticompetitive arrangements relative to the status quo ante (i.e., where arrangements among vertical firms are by contract): Our lack of knowledge about the effects of market structure and firm organization on innovation and dynamic competition, and the relative hostility to nonstandard contracting, including vertical integration:

[T]he literature addressing how market structure affects innovation (and vice versa) in the end reveals an ambiguous relationship in which factors unrelated to competition play an important role. (Katz & Shelanski, Mergers and Innovation)

The fixation on the equivalency of the form of vertical integration (i.e., merger versus contract) is likely to lead enforcers to focus on static price and cost effects, and miss the dynamic organizational and informational effects that lead to unexpected, increased innovation across and within firms. 

In the hands of Oliver Williamson, this means that understanding firms in the real world entails taking an organization theory approach, in contrast to the “orthodox” economic perspective:

The lens of contract approach to the study of economic organization is partly complementary but also partly rival to the orthodox [neoclassical economic] lens of choice. Specifically, whereas the latter focuses on simple market exchange, the lens of contract is predominantly concerned with the complex contracts. Among the major differences is that non‐standard and unfamiliar contractual practices and organizational structures that orthodoxy interprets as manifestations of monopoly are often perceived to serve economizing purposes under the lens of contract. A major reason for these and other differences is that orthodoxy is dismissive of organization theory whereas organization theory provides conceptual foundations for the lens of contract. (emphasis added)

We are more likely to miss it when mergers solve market inefficiencies, and more likely to see it when they impose static costs — even if the apparent costs actually represent a move from less efficient contractual arrangements to more efficient integration.

The competition that takes place in the real world and between various groups ultimately depends upon the institution of private contracts, many of which, including the firm itself, are nonstandard. Innovation includes the discovery of new organizational forms and the application of old forms to new contexts. Such contracts prevent or attenuate market failure, moving the market toward what economists would deem a more competitive result. Indeed, as Professor Coase pointed out, many markets deemed “perfectly competitive” are in fact the end result of complex contracts limiting rivalry between competitors. This contractual competition cannot produce perfect results — no human institution ever can. Nonetheless, the result is superior to that which would obtain in a (real) world without nonstandard contracting. These contracts do not depend upon the creation or enhancement of market power and thus do not produce the evils against which antitrust law is directed. (Alan Meese, Price Theory Competition & the Rule of Reason)

Or, as Oliver Williamson more succinctly puts it:

[There is a] rebuttable presumption that nonstandard forms of contracting have efficiency purposes. (Oliver Williamson, The Economic Institutions of Capitalism)

The pinched focus of the guidelines on narrow market definition misses the bigger picture of dynamic competition over time

The proposed guidelines (and the theories of harm undergirding them) focus upon indicia of market power that may not be accurate if assessed in more realistic markets or over more relevant timeframes, and, if applied too literally, may bias enforcement against mergers with dynamic-innovation benefits but static-competition costs.  

Similarly, the proposed guidelines’ enumeration of potential efficiencies doesn’t really begin to cover the categories implicated by the organization of enterprise around dynamic considerations

The proposed guidelines’ efficiencies section notes that:

Vertical mergers bring together assets used at different levels in the supply chain to make a final product. A single firm able to coordinate how these assets are used may be able to streamline production, inventory management, or distribution, or create innovative products in ways that would have been hard to achieve though arm’s length contracts. (emphasis added)

But it is not clear than any of these categories encompasses organizational decisions made to facilitate the coordination of production and commercialization when they are dependent upon intangible assets.

As Thomas Jorde and David Teece write:

For innovations to be commercialized, the economic system must somehow assemble all the relevant complementary assets and create a dynamically-efficient interactive system of learning and information exchange. The necessary complementary assets can conceivably be assembled by either administrative or market processes, as when the innovator simply licenses the technology to firms that already own or are willing to create the relevant assets. These organizational choices have received scant attention in the context of innovation. Indeed, the serial model relies on an implicit belief that arm’s-length contracts between unaffiliated firms in the vertical chain from research to customer will suffice to commercialize technology. In particular, there has been little consideration of how complex contractual arrangements among firms can assist commercialization — that is, translating R&D capability into profitable new products and processes….

* * *

But in reality, the market for know-how is riddled with imperfections. Simple unilateral contracts where technology is sold for cash are unlikely to be efficient. Complex bilateral and multilateral contracts, internal organization, or various hybrid structures are often required to shore up obvious market failures and create procompetitive efficiencies. (Jorde & Teece, Rule of Reason Analysis of Horizontal Arrangements: Agreements Designed to Advance Innovation and Commercialize Technology) (emphasis added)

When IP protection for a given set of valuable pieces of “know-how” is strong — easily defendable, unique patents, for example — firms can rely on property rights to efficiently contract with vertical buyers and sellers. But in cases where the valuable “know how” is less easily defended as IP — e.g. business process innovation, managerial experience, distributed knowledge, corporate culture, and the like — the ability to partially vertically integrate through contract becomes more difficult, if not impossible. 

Perhaps employing these assets is part of what is meant in the draft guidelines by “streamline.” But the very mention of innovation only in the technological context of product innovation is at least some indication that organizational innovation is not clearly contemplated.  

This is a significant lacuna. The impact of each organizational form on knowledge transfers creates a particularly strong division between integration and contract. As Enghin Atalay, Ali Hortaçsu & Chad Syverson point out:

That vertical integration is often about transfers of intangible inputs rather than physical ones may seem unusual at first glance. However, as observed by Arrow (1975) and Teece (1982), it is precisely in the transfer of nonphysical knowledge inputs that the market, with its associated contractual framework, is most likely to fail to be a viable substitute for the firm. Moreover, many theories of the firm, including the four “elemental” theories as identified by Gibbons (2005), do not explicitly invoke physical input transfers in their explanations for vertical integration. (Enghin Atalay, et al., Vertical Integration and Input Flows) (emphasis added)

There is a large economics and organization theory literature discussing how organizations are structured with respect to these sorts of intangible assets. And the upshot is that, while we start — not end, as some would have it — with the Coasian insight that firm boundaries are necessarily a function of production processes and not a hard limit, we quickly come to realize that it is emphatically not the case that integration-via-contract and integration-via-merger are always, or perhaps even often, viable substitutes.

Conclusion

The contract/merger equivalency assumption, coupled with a “least-restrictive alternative” logic that favors contract over merger, puts a thumb on the scale against vertical mergers. While the proposed guidelines as currently drafted do not necessarily portend the inflexible, formalistic application of this logic, they offer little to guide enforcers or courts away from the assumption in the important (and perhaps numerous) cases where it is unwarranted.   

[TOTM: The following is part of a symposium by TOTM guests and authors on the 2020 Vertical Merger Guidelines. The entire series of posts is available here.

This post is authored by Geoffrey A. Manne (President & Founder, ICLE; Distinguished Fellow, Northwestern University Center on Law, Business, and Economics ); and Kristian Stout (Associate Director, ICLE).]

As many in the symposium have noted — and as was repeatedly noted during the FTC’s Hearings on Competition and Consumer Protection in the 21st Century — there is widespread dissatisfaction with the 1984 Non-Horizontal Merger Guidelines

Although it is doubtless correct that the 1984 guidelines don’t reflect the latest economic knowledge, it is by no means clear that this has actually been a problem — or that a new set of guidelines wouldn’t create even greater problems. Indeed, as others have noted in this symposium, there is a great deal of ambiguity in the proposed guidelines that could lead either to uncertainty as to how the agencies will exercise their discretion, or, more troublingly, could lead courts to take seriously speculative theories of harm

We can do little better in expressing our reservations that new guidelines are needed than did the current Chairman of the FTC, Joe Simons, writing on this very blog in a symposium on what became the 2010 Horizontal Merger Guidelines. In a post entitled, Revisions to the Merger Guidelines: Above All, Do No Harm, Simons writes:

My sense is that there is no need to revise the DOJ/FTC Horizontal Merger Guidelines, with one exception…. The current guidelines lay out the general framework quite well and any change in language relative to that framework are likely to create more confusion rather than less. Based on my own experience, the business community has had a good sense of how the agencies conduct merger analysis…. If, however, the current administration intends to materially change the way merger analysis is conducted at the agencies, then perhaps greater revision makes more sense. But even then, perhaps the best approach is to try out some of the contemplated changes (i.e. in actual investigations) and publicize them in speeches and the like before memorializing them in a document that is likely to have some substantial permanence to it.

Wise words. Unless, of course, “the current [FTC] intends to materially change the way [vertical] merger analysis is conducted.” But the draft guidelines don’t really appear to portend a substantial change, and in several ways they pretty accurately reflect agency practice.

What we want to draw attention to, however, is an implicit underpinning of the draft guidelines that we believe the agencies should clearly disavow (or at least explain more clearly the complexity surrounding): the extent and implications of the presumed functional equivalence of vertical integration by contract and by merger — the contract/merger equivalency assumption.   

Vertical mergers and their discontents

The contract/merger equivalency assumption has been gaining traction with antitrust scholars, but it is perhaps most clearly represented in some of Steve Salop’s work. Salop generally believes that vertical merger enforcement should be heightened. Among his criticisms of current enforcement is his contention that efficiencies that can be realized by merger can often also be achieved by contract. As he discussed during his keynote presentation at last year’s FTC hearing on vertical mergers:

And, finally, the key policy issue is the issue is not about whether or not there are efficiencies; the issue is whether the efficiencies are merger-specific. As I pointed out before, Coase stressed that you can get vertical integration by contract. Very often, you can achieve the vertical efficiencies if they occur, but with contracts rather than having to merge.

And later, in the discussion following his talk:

If there is vertical integration by contract… it meant you could get all the efficiencies from vertical integration with a contract. You did not actually need the vertical integration. 

Salop thus argues that because the existence of a “contract solution” to firm problems can often generate the same sorts of efficiencies as when firms opt to merge, enforcers and courts should generally adopt a presumption against vertical mergers relative to contracting:

Coase’s door swings both ways: Efficiencies often can be achieved by vertical contracts, without the potential anticompetitive harms from merger

In that vertical restraints are characterized as “just” vertical integration “by contract,” then claimed efficiencies in problematical mergers might be achieved with non-merger contracts that do not raise the same anticompetitive concerns. (emphasis in original)

(Salop isn’t alone in drawing such a conclusion, of course; Carl Shapiro, for example, has made a similar point (as have others)).

In our next post we explore the policy errors implicated by this contract/merger equivalency assumption. But here we want to consider whether it makes logical sense in the first place

The logic of vertical integration is not commutative 

It is true that, where contracts are observed, they are likely as (or more, actually)  efficient than merger. But, by the same token, it is also true that where mergers are observed they are likely more efficient than contracts. Indeed, the entire reason for integration is efficiency relative to what could be done by contract — this is the essence of the so-called “make-or-buy” decision. 

For example, a firm that decides to buy its own warehouse has determined that doing so is more efficient than renting warehouse space. Some of these efficiencies can be measured and quantified (e.g., carrying costs of ownership vs. the cost of rent), but many efficiencies cannot be easily measured or quantified (e.g., layout of the facility or site security). Under the contract/merger equivalency assumption, the benefits of owning a warehouse can be achieved “very often” by renting warehouse space. But the fact that many firms using warehouses own some space and rent some space indicates that the make-or-buy decision is often unique to each firm’s idiosyncratic situation. Moreover, the distinctions driving those differences will not always be readily apparent, and whether contracting or integrating is preferable in any given situation may not be inferred from the existence of one or the other elsewhere in the market — or even in the same firm!

There is no reason to presume in any given situation that the outcome from contracting would be the same as from merging, even where both are notionally feasible. The two are, quite simply, different bargaining environments, each with a different risk and cost allocation; accounting treatment; effect on employees, customers, and investors; tax consequence, etc. Even if the parties accomplished nominally “identical” outcomes, they would not, in fact, be identical.

Meanwhile, what if the reason for failure to contract, or the reason to prefer merger, has nothing to do with efficiency? What if there were no anticompetitive aim but there were a tax advantage? What if one of the parties just wanted a larger firm in order to satisfy the CEO’s ego? That these are not cognizable efficiencies under antitrust law is clear. But the adoption of a presumption of equivalence between contract and merger would — ironically — entail their incorporation into antitrust law just the same — by virtue of their effective prohibition under antitrust law

In other words, if the assumption is that contract and merger are equally efficient unless proven otherwise, but the law adopts a suspicion (or, even worse, a presumption) that vertical mergers are anticompetitive which can be rebutted only with highly burdensome evidence of net efficiency gain, this effectively deputizes antitrust law to enforce a preconceived notion of “merger appropriateness” that does not necessarily turn on efficiencies. There may (or may not) be sensible policy reasons for adopting such a stance, but they aren’t antitrust reasons.

More fundamentally, however, while there are surely some situations in which contractual restraints might be able to achieve similar organizational and efficiency gains as a merger, the practical realities of achieving not just greater efficiency, but a whole host of non-efficiency-related, yet nonetheless valid, goals, are rarely equivalent between the two

It may be that the parties don’t know what they don’t know to such an extent that a contract would be too costly because it would be too incomplete, for example. But incomplete contracts and ambiguous control and ownership rights aren’t (as much of) an issue on an ongoing basis after a merger. 

As noted, there is no basis for assuming that the structure of a merger and a contract would be identical. In the same way, there is no basis for assuming that the knowledge transfer that would result from a merger would be the same as that which would result from a contract — and in ways that the parties could even specify or reliably calculate in advance. Knowing that the prospect for knowledge “synergies” would be higher with a merger than a contract might be sufficient to induce the merger outcome. But asked to provide evidence that the parties could not engage in the same conduct via contract, the parties would be unable to do so. The consequence, then, would be the loss of potential gains from closer integration.

At the same time, the cavalier assumption that parties would be able — legally — to enter into an analogous contract in lieu of a merger is problematic, given that it would likely be precisely the form of contract (foreclosing downstream or upstream access) that is alleged to create problems with the merger in the first place.

At the FTC hearings last year, Francine LaFontaine highlighted this exact concern

I want to reemphasize that there are also rules against vertical restraints in antitrust laws, and so to say that the firms could achieve the mergers outcome by using vertical restraints is kind of putting them in a circular motion where we are telling them you cannot merge because you could do it by contract, and then we say, but these contract terms are not acceptable.

Indeed, legal risk is one of the reasons why a merger might be preferable to a contract, and because the relevant markets here are oligopoly markets, the possibility of impermissible vertical restraints between large firms with significant market share is quite real.

More important, the assumptions underlying the contention that contracts and mergers are functionally equivalent legal devices fails to appreciate the importance of varied institutional environments. Consider that one reason some takeovers are hostile is because incumbent managers don’t want to merge, and often believe that they are running a company as well as it can be run — that a change of corporate control would not improve efficiency. The same presumptions may also underlie refusals to contract and, even more likely, may explain why, to the other firm, a contract would be ineffective.

But, while there is no way to contract without bilateral agreement, there is a corporate control mechanism to force a takeover. In this institutional environment a merger may be easier to realize than a contract (and that applies even to a consensual merger, of course, given the hostile outside option). In this case, again, the assumption that contract should be the relevant baseline and the preferred mechanism for coordination is misplaced — even if other firms in the industry are successfully accomplishing the same thing via contract, and even if a contract would be more “efficient” in the abstract.

Conclusion

Properly understood, the choice of whether to contract or merge derives from a host of complicated factors, many of which are difficult to observe and/or quantify. The contract/merger equivalency assumption — and the species of “least-restrictive alternative” reasoning that would demand onerous efficiency arguments to permit a merger when a contract was notionally possible — too readily glosses over these complications and unjustifiably embraces a relative hostility to vertical mergers at odds with both theory and evidence

Rather, as has long been broadly recognized, there can be no legally relevant presumption drawn against a company when it chooses one method of vertical integration over another in the general case. The agencies should clarify in the draft guidelines that the mere possibility of integration via contract or the inability of merging parties to rigorously describe and quantify efficiencies does not condemn a proposed merger.

It might surprise some readers to learn that we think the Court’s decision today in Apple v. Pepper reaches — superficially — the correct result. But, we hasten to add, the Court’s reasoning (and, for that matter, the dissent’s) is completely wrongheaded. It would be an understatement to say that the Court reached the right result for the wrong reason; in fact, the Court’s analysis wasn’t even in the same universe as the correct reasoning.

Below we lay out our assessment, in a post drawn from an article forthcoming in the Nebraska Law Review.

Did the Court forget that, just last year, it decided Amex, the most significant U.S. antitrust case in ages?

What is most remarkable about the decision (and the dissent) is that neither mentions Ohio v. Amex, nor even the two-sided market context in which the transactions at issue take place.

If the decision in Apple v. Pepper hewed to the precedent established by Ohio v. Amex it would start with the observation that the relevant market analysis for the provision of app services is an integrated one, in which the overall effect of Apple’s conduct on both app users and app developers must be evaluated. A crucial implication of the Amex decision is that participants on both sides of a transactional platform are part of the same relevant market, and the terms of their relationship to the platform are inextricably intertwined.

Under this conception of the market, it’s difficult to maintain that either side does not have standing to sue the platform for the terms of its overall pricing structure, whether the specific terms at issue apply directly to that side or not. Both end users and app developers are “direct” purchasers from Apple — of different products, but in a single, inextricably interrelated market. Both groups should have standing.

More controversially, the logic of Amex also dictates that both groups should be able to establish antitrust injury — harm to competition — by showing harm to either group, as long as it establishes the requisite interrelatedness of the two sides of the market.

We believe that the Court was correct to decide in Amex that effects falling on the “other” side of a tightly integrated, two-sided market from challenged conduct must be addressed by the plaintiff in making its prima facie case. But that outcome entails a market definition that places both sides of such a market in the same relevant market for antitrust analysis.

As a result, the Court’s holding in Amex should also have required a finding in Apple v. Pepper that an app user on one side of the platform who transacts with an app developer on the other side of the market, in a transaction made possible and directly intermediated by Apple’s App Store, should similarly be deemed in the same market for standing purposes.

Relative to a strict construction of the traditional baseline, the former entails imposing an additional burden on two-sided market plaintiffs, while the latter entails a lessening of that burden. Whether the net effect is more or fewer successful cases in two-sided markets is unclear, of course. But from the perspective of aligning evidentiary and substantive doctrine with economic reality such an approach would be a clear improvement.

Critics accuse the Court of making antitrust cases unwinnable against two-sided market platforms thanks to Amex’s requirement that a prima facie showing of anticompetitive effect requires assessment of the effects on both sides of a two-sided market and proof of a net anticompetitive outcome. The critics should have been chastened by a proper decision in Apple v. Pepper. As it is, the holding (although not the reasoning) still may serve to undermine their fears.

But critics should have recognized that a necessary corollary of Amex’s “expanded” market definition is that, relative to previous standing doctrine, a greater number of prospective parties should have standing to sue.

More important, the Court in Apple v. Pepper should have recognized this. Although nominally limited to the indirect purchaser doctrine, the case presented the Court with an opportunity to grapple with this logical implication of its Amex decision. It failed to do so.

On the merits, it looks like Apple should win. But, for much the same reason, the Respondents in Apple v. Pepper should have standing

This does not, of course, mean that either party should win on the merits. Indeed, on the merits of the case, the Petitioner in Apple v. Pepper appears to have the stronger argument, particularly in light of Amex which (assuming the App Store is construed as some species of a two-sided “transaction” market) directs that Respondent has the burden of considering harms and efficiencies across both sides of the market.

At least on the basis of the limited facts as presented in the case thus far, Respondents have not remotely met their burden of proving anticompetitive effects in the relevant market.

The actual question presented in Apple v. Pepper concerns standing, not whether the plaintiffs have made out a viable case on the merits. Thus it may seem premature to consider aspects of the latter in addressing the former. But the structure of the market considered by the court should be consistent throughout its analysis.

Adjustments to standing in the context of two-sided markets must be made in concert with the nature of the substantive rule of reason analysis that will be performed in a case. The two doctrines are connected not only by the just demands for consistency, but by the error-cost framework of the overall analysis, which runs throughout the stages of an antitrust case.

Here, the two-sided markets approach in Amex properly understands that conduct by a platform has relevant effects on both sides of its interrelated two-sided market. But that stems from the actual economics of the platform; it is not merely a function of a judicial construct. It thus holds true at all stages of the analysis.

The implication for standing is that users on both sides of a two-sided platform may suffer similarly direct (or indirect) injury as a result of the platform’s conduct, regardless of the side to which that conduct is nominally addressed.

The consequence, then, of Amex’s understanding of the market is that more potential plaintiffs — specifically, plaintiffs on both sides of a two-sided market — may claim to suffer antitrust injury.

Why the myopic focus of the holding (and dissent) on Illinois Brick is improper: It’s about the market definition, stupid!

Moreover, because of the Amex understanding, the problem of analyzing the pass-through of damages at issue in Illinois Brick (with which the Court entirely occupies itself in Apple v. Pepper) is either mitigated or inevitable.

In other words, either the users on the different sides of a two-sided market suffer direct injury without pass-through under a proper definition of the relevant market, or else their interrelatedness is so strong that, complicated as it may be, the needs of substantive accuracy trump the administrative costs in sorting out the incidence of the costs, and courts cannot avoid them.

Illinois Brick’s indirect purchaser doctrine was designed for an environment in which the relationship between producers and consumers is mediated by a distributor in a direct, linear supply chain; it was not designed for platforms. Although the question presented in Apple v. Pepper is explicitly about whether the Illinois Brick “indirect purchaser” doctrine applies to the Apple App Store, that determination is contingent on the underlying product market definition (whether the product market is in fact well-specified by the parties and the court or not).

Particularly where intermediaries exist precisely to address transaction costs between “producers” and “consumers,” the platform services they provide may be central to the underlying claim in a way that the traditional direct/indirect filters — and their implied relevant markets — miss.

Further, the Illinois Brick doctrine was itself based not on the substantive necessity of cutting off liability evaluations at a particular level of distribution, but on administrability concerns. In particular, the Court was concerned with preventing duplicative recovery when there were many potential groups of plaintiffs, as well as preventing injustices that would occur if unknown groups of plaintiffs inadvertently failed to have their rights adequately adjudicated in absentia. It was also concerned with avoiding needlessly complicated damages calculations.

But, almost by definition, the tightly coupled nature of the two sides of a two-sided platform should mitigate the concerns about duplicative recovery and unknown parties. Moreover, much of the presumed complexity in damages calculations in a platform setting arise from the nature of the platform itself. Assessing and apportioning damages may be complicated, but such is the nature of complex commercial relationships — the same would be true, for example, of damages calculations between vertically integrated companies that transact simultaneously at multiple levels, or between cross-licensing patent holders/implementers. In fact, if anything, the judicial efficiency concerns in Illinois Brick point toward the increased importance of properly assessing the nature of the product or service of the platform in order to ensure that it accurately encompasses the entire relevant transaction.

Put differently, under a proper, more-accurate market definition, the “direct” and “indirect” labels don’t necessarily reflect either business or antitrust realities.

Where the Court in Apple v. Pepper really misses the boat is in its overly formalistic claim that the business model (and thus the product) underlying the complained-of conduct doesn’t matter:

[W]e fail to see why the form of the upstream arrangement between the manufacturer or supplier and the retailer should determine whether a monopolistic retailer can be sued by a downstream consumer who has purchased a good or service directly from the retailer and has paid a higher-than-competitive price because of the retailer’s unlawful monopolistic conduct.

But Amex held virtually the opposite:

Because “[l]egal presumptions that rest on formalistic distinctions rather than actual market realities are generally disfavored in antitrust law,” courts usually cannot properly apply the rule of reason without an accurate definition of the relevant market.

* * *

Price increases on one side of the platform likewise do not suggest anticompetitive effects without some evidence that they have increased the overall cost of the platform’s services. Thus, courts must include both sides of the platform—merchants and cardholders—when defining the credit-card market.

In the face of novel business conduct, novel business models, and novel economic circumstances, the degree of substantive certainty may be eroded, as may the reasonableness of the expectation that typical evidentiary burdens accurately reflect competitive harm. Modern technology — and particularly the platform business model endemic to many modern technology firms — presents a need for courts to adjust their doctrines in the face of such novel issues, even if doing so adds additional complexity to the analysis.

The unlearned market-definition lesson of the Eighth Circuit’s Campos v. Ticketmaster dissent

The Eight Circuit’s Campos v. Ticketmaster case demonstrates the way market definition shapes the application of the indirect purchaser doctrine. Indeed, the dissent in that case looms large in the Ninth Circuit’s decision in Apple v. Pepper. [Full disclosure: One of us (Geoff) worked on the dissent in Campos v. Ticketmaster as a clerk to Eighth Circuit judge Morris S. Arnold]

In Ticketmaster, the plaintiffs alleged that Ticketmaster abused its monopoly in ticket distribution services to force supracompetitve charges on concert venues — a practice that led to anticompetitive prices for concert tickets. Although not prosecuted as a two-sided market, the business model is strikingly similar to the App Store model, with Ticketmaster charging fees to venues and then facilitating ticket purchases between venues and concert goers.

As the dissent noted, however:

The monopoly product at issue in this case is ticket distribution services, not tickets.

Ticketmaster supplies the product directly to concert-goers; it does not supply it first to venue operators who in turn supply it to concert-goers. It is immaterial that Ticketmaster would not be supplying the service but for its antecedent agreement with the venues.

But it is quite relevant that the antecedent agreement was not one in which the venues bought some product from Ticketmaster in order to resell it to concert-goers.

More important, and more telling, is the fact that the entirety of the monopoly overcharge, if any, is borne by concert-goers.

In contrast to the situations described in Illinois Brick and the literature that the court cites, the venues do not pay the alleged monopoly overcharge — in fact, they receive a portion of that overcharge from Ticketmaster. (Emphasis added).

Thus, if there was a monopoly overcharge it was really borne entirely by concert-goers. As a result, apportionment — the complexity of which gives rise to the standard in Illinois Brick — was not a significant issue. And the antecedent transaction that allegedly put concertgoers in an indirect relationship with Ticketmaster is one in which Ticketmaster and concert venues divvied up the alleged monopoly spoils, not one in which the venues absorb their share of the monopoly overcharge.

The analogy to Apple v. Pepper is nearly perfect. Apple sits between developers on one side and consumers on the other, charges a fee to developers for app distribution services, and facilitates app sales between developers and users. It is possible to try to twist the market definition exercise to construe the separate contracts between developers and Apple on one hand, and the developers and consumers on the other, as some sort of complicated version of the classical manufacturing and distribution chains. But, more likely, it is advisable to actually inquire into the relevant factual differences that underpin Apple’s business model and adapt how courts consider market definition for two-sided platforms.

Indeed, Hanover Shoe and Illinois Brick were born out of a particular business reality in which businesses structured themselves in what are now classical production and distribution chains. The Supreme Court adopted the indirect purchaser rule as a prudential limitation on antitrust law in order to optimize the judicial oversight of such cases. It seems strangely nostalgic to reflexively try to fit new business methods into old legal analyses, when prudence and reality dictate otherwise.

The dissent in Ticketmaster was ahead of its time insofar as it recognized that the majority’s formal description of the ticket market was an artifact of viewing what was actually something much more like a ticket-services platform operated by Ticketmaster through the poor lens of the categories established decades earlier.

The Ticketmaster dissent’s observations demonstrate that market definition and antitrust standing are interrelated. It makes no sense to adhere to a restrictive reading of the latter if it connotes an economically improper understanding of the former. Ticketmaster provided an intermediary service — perhaps not quite a two-sided market, but something close — that stands outside a traditional manufacturing supply chain. Had it been offered by the venues themselves and bundled into the price of concert tickets there would be no question of injury and of standing (nor would market definition matter much, as both tickets and distribution services would be offered as a joint product by the same parties, in fixed proportions).

What antitrust standing doctrine should look like after Amex

There are some clear implications for antitrust doctrine that (should) follow from the preceding discussion.

A plaintiff has a choice to allege that a defendant operates either as a two-sided market or in a more traditional, linear chain during the pleading stage. If the plaintiff alleges a two-sided market, then, to demonstrate standing, it need only be shown that injury occurred to some subset of platform users with which the plaintiff is inextricably interrelated. The plaintiff would not need to demonstrate injury to him or herself, nor allege net harm, nor show directness.

In response, a defendant can contest standing by challenging the interrelatedness of the plaintiff and the group of platform users with whom the plaintiff claims interrelatedness. If the defendant does not challenge the allegation that it operates a two-sided market, it could not challenge standing by showing indirectness, that plaintiff had not alleged personal injury, or that plaintiff hasn’t alleged a net harm.

Once past a determination of standing, however, a plaintiff who pleads a two-sided market would not be able to later withdraw this allegation in order to lessen the attendant legal burdens.

If the court accepts that the defendant is operating a two-sided market, both parties would be required to frame their allegations and defenses in accordance with the nature of the two-sided market and thus the holding in Amex. This is critical because, whereas alleging a two-sided market may make it easier for plaintiffs to demonstrate standing, Amex’s requirement that net harm be demonstrated across interrelated sets of users makes it more difficult for plaintiffs to present a viable prima facie case. Further, defendants would not be barred from presenting efficiencies defenses based on benefits that interrelated users enjoy.

Conclusion: The Court in Apple v. Pepper should have acknowledged the implications of its holding in Amex

After Amex, claims against two-sided platforms might require more evidence to establish anticompetitive harm, but that business model also means that firms should open themselves up to a larger pool of potential plaintiffs. The legal principles still apply, but the relative importance of those principles to judicial outcomes shifts (or should shift) in line with the unique economic position of potential plaintiffs and defendants in a platform environment.

Whether a priori the net result is more or fewer cases and more or fewer victories for plaintiffs is not the issue; what matters is matching the legal and economic theory to the relevant facts in play. Moreover, decrying Amex as the end of antitrust was premature: the actual affect on injured parties can’t be known until other changes (like standing for a greater number of plaintiffs) are factored into the analysis. The Court’s holding in Apple v. Pepper sidesteps this issue entirely, and thus fails to properly move antitrust doctrine forward in line with its holding in Amex.

Of course, it’s entirely possible that platforms and courts might be inundated with expensive and difficult to manage lawsuits. There may be reasons of administrability for limiting standing (as Illinois Brick perhaps prematurely did for fear of the costs of courts’ managing suits). But then that should have been the focus of the Court’s decision.

Allowing standing in Apple v. Pepper permits exactly the kind of legal experimentation needed to enable the evolution of antitrust doctrine along with new business realities. But in some ways the Court reached the worst possible outcome. It announced a rule that permits more plaintiffs to establish standing, but it did not direct lower courts to assess standing within the proper analytical frame. Instead, it just expands standing in a manner unmoored from the economic — and, indeed, judicial — context. That’s not a recipe for the successful evolution of antitrust doctrine.

On Tuesday, August 28, 2018, Truth on the Market and the International Center for Law and Economics presented a blog symposium — Is Amazon’s Appetite Bottomless? The Whole Foods Merger After One Year — that looked at the concerns surrounding the closing of the Amazon-Whole Foods merger, and how those concerns had played out over the last year.

The difficulty presented by the merger was, in some ways, its lack of difficulty: Even critics, while hearkening back to the Brandeisian fear of large firms, had little by way of legal objection to offer against the merger. Despite the acknowledged lack of an obvious legal basis for challenging the merger, most critics nevertheless expressed a somewhat inchoate and generalized concern that the merger would hasten the death of brick-and-mortar retail and imperil competition in the grocery industry. Critics further pointed to particular, related issues largely outside the scope of modern antitrust law — issues relating to the presumed effects of the merger on “localism” (i.e., small, local competitors), retail workers, startups with ancillary businesses (e.g., delivery services), data collection and use, and the like.

Steven Horwitz opened the symposium with an insightful and highly recommended post detailing the development of the grocery industry from its inception. Tracing through that history, Horwitz was optimistic that

Viewed from the long history of the evolution of the grocery store, the Amazon-Whole Foods merger made sense as the start of the next stage of that historical process. The combination of increased wealth that is driving the demand for upscale grocery stores, and the corresponding increase in the value of people’s time that is driving the demand for one-stop shopping and various forms of pick-up and delivery, makes clear the potential benefits of this merger.

Others in the symposium similarly acknowledged the potential transformation of the industry brought on by the merger, but challenged the critics’ despairing characterization of that transformation (Auer, Manne & Stout, Rinehart, Fruits, Atkinson).

At the most basic level, it was noted that, in the immediate aftermath of the merger, Whole Foods dropped prices across a number of categories as it sought to shore up its competitive position (Auer). Further, under relevant antitrust metrics — e.g., market share, ease of competitive entry, potential for exclusionary conduct — the merger was completely unobjectionable under existing doctrine (Fruits).

To critics’ claims that Amazon in general, and the merger in particular, was decimating the retail industry, several posts discussed the updated evidence suggesting that retail is not actually on the decline (although some individual retailers are certainly struggling to compete) (Auer, Manne & Stout). Moreover, and following from Horwitz’s account of the evolution of the grocery industry, it appears that the actual trajectory of the industry is not an either/or between online and offline, but instead a movement toward integrating both models into a single retail experience (Manne & Stout). Further, the post-merger flurry of business model innovation, venture capital investment, and new startup activity demonstrates that, confronted with entrepreneurial competitors like Walmart, Kroger, Aldi, and Instacart, Amazon’s impressive position online has not translated into an automatic domination of the traditional grocery industry (Manne & Stout).  

Symposium participants more circumspect about the merger suggested that Amazon’s behavior may be laying the groundwork for an eventual monopsony case (Sagers). Further, it was suggested, a future Section 2 case, difficult under prevailing antitrust orthodoxy, could be brought with a creative approach to market definition in light of Amazon’s conduct with its marketplace participants, its aggressive ebook contracting practices, and its development and roll-out of its own private label brands (Sagers).

Skeptics also picked up on early critics’ concerns about the aggregation of large amounts of consumer data, and worried that the merger could be part of a pattern representing a real, long-term threat to consumers that antitrust does not take seriously enough (Bona & Levitsky). Sounding a further alarm, Hal Singer noted that Amazon’s interest in pushing into new markets with data generated by, for example, devices like its Echo line could bolster its ability to exclude competitors.

More fundamentally, these contributors echoed the merger critics’ concerns that antitrust does not adequately take account of other values such as “promoting local, community-based, organic food production or ‘small firms’ in general.” (Bona & Levitsky; Singer).

Rob Atkinson, however, pointed out that these values are idiosyncratic and not likely shared by the vast majority of the population — and that antitrust law shouldn’t have anything to do with them:

In short, most of the opposition to Amazon/Whole Foods merger had little or nothing to do with economics and consumer welfare. It had everything to do with a competing vision for the kind of society we want to live in. The neo-Brandesian opponents, who Lind and I term “progressive localists”, seek an alternative economy predominantly made up of small firms, supported by big government and protected from global competition.

And Dirk Auer noted that early critics’ prophecies of foreclosure of competition through “data leveraging” and below-cost pricing hadn’t remotely come to pass, thus far.

Meanwhile, other contributors noted the paucity of evidence supporting many of these assertions, and pointed out the manifest value the merger seemed to be creating by pressuring competitors to adapt and better respond to consumers’ preferences (Horwitz, Rinehart, Auer, Fruits, Manne & Stout) — in the process shoring up, rather than killing, even smaller retailers that are willing and able to evolve with changing technology and shifting consumer preferences. “For all the talk of retail dying, the stores that are actually dying are the ones that fail to cater to their customers, not the ones that happen to be offline” (Manne & Stout).

At the same time, not all merger skeptics were moved by the Neo-Brandeisian assertions. Chris Sagers, for example, finds much of the populist antitrust objection more public relations than substance. He suggested perhaps not taking these ideas and their promoters so seriously, and instead focusing on antitrust advocates with “real ideas” (like Sagers himself, of course).

Coming from a different angle, Will Rinehart also suggested not taking the criticisms too seriously, pointing to the evolving and complicated effects of the merger as Exhibit A for the need for regulatory humility:

Finally, this deal reiterates the need for regulatory humility. Almost immediately after the Amazon-Whole Foods merger was closed, prices at the store dropped and competitors struck a flurry of deals. Investments continue and many in the grocery retail space are bracing for a wave of enhancement to take hold. Even some of the most fierce critics of deal will have to admit there is a lot of uncertainty. It is unclear what business model will make the most sense in the long run, how these technologies will ultimately become embedded into production processes, and how consumers will benefit. Combined, these features underscore the difficulty, but the necessity, in implementing dynamic insights into antitrust institutions.

Offering generous praise for this symposium (thanks, Will!) and echoing the points made by other participants regarding the dynamic and unknowable course of competition (Auer, Horwitz, Manne & Stout, Fruits), Rinehart concludes:

Retrospectives like this symposium offer a chance to understand what the discussion missed at the time and what is needed to better understand innovation and competition in markets. While it might be too soon to close the book on this case, the impact can already be felt in the positions others are taking in response. In the end, the deal probably won’t be remembered for extending Amazon’s dominance into another market because that is a phantom concern. Rather, it will probably be best remembered as the spark that drove traditional retail outlets to modernize their logistics and fulfillment efforts.  

For a complete rundown of the arguments both for and against, the full archive of symposium posts from our outstanding and diverse group of scholars, practitioners, and other experts is available at this link, and individual posts can be easily accessed by clicking on the authors’ names below.

We’d like to thank all of the participants for their excellent contributions!

 

What actually happened in the year following the merger is nearly the opposite: Competition among grocery stores has been more fierce than ever. “Offline” retailers are expanding — and innovating — to meet Amazon’s challenge, and many of them are booming. Disruption is never neat and tidy, but, in addition to saving Whole Foods from potential oblivion, the merger seems to have lit a fire under the rest of the industry.
This result should not be surprising to anyone who understands the nature of the competitive process. But it does highlight an important lesson: competition often comes from unexpected quarters and evolves in unpredictable ways, emerging precisely out of the kinds of adversity opponents of the merger bemoaned.

Continue Reading...

At this point, only the most masochistic and cynical among DC’s policy elite actually desire for the net neutrality conflict to continue. And yet, despite claims that net neutrality principles are critical to protecting consumers, passage of the current Congressional Review Act (“CRA”) disapproval resolution in Congress would undermine consumer protection and promise only to drag out the fight even longer.

The CRA resolution is primarily intended to roll back the FCC’s re-re-classification of broadband as a Title I service under the Communications Act in the Restoring Internet Freedom Order (“RIFO”). The CRA allows Congress to vote to repeal rules recently adopted by federal agencies; upon a successful CRA vote, the rules are rescinded and the agency is prohibited from adopting substantially similar rules in the future.

But, as TechFreedom has noted, it’s not completely clear that a CRA on a regulatory classification decision will work quite the way Congress intends it and could just trigger more litigation cycles, largely because it is unclear what parts of the RIFO are actually “rules” subject to the CRA. Harold Feld has written a critique of TechFreedom’s position, arguing, in effect, that of course the RIFO is a rule; TechFreedom responded with a pretty devastating rejoinder.

But this exchange really demonstrates TechFreedom’s central argument: It is sufficiently unclear how or whether the CRA will apply to the various provisions of the RIFO, such that the only things the CRA is guaranteed to do are 1) to strip consumers of certain important protections — it would take away the FCC’s transparency requirements for ISPs, and imperil privacy protections currently ensured by the FTC — while 2) prolonging the already interminable litigation and political back-and-forth over net neutrality.

The CRA is political theater

The CRA resolution effort is not about good Internet regulatory policy; rather, it’s pure political opportunism ahead of the midterms. Democrats have recognized net neutrality as a good wedge issue because of its low political opportunity cost. The highest-impact costs of over-regulating broadband through classification decisions are hard to see: Rather than bad things happening, the costs arrive in the form of good things not happening. Eventually those costs work their way to customers through higher access prices or less service — especially in rural areas most in need of it — but even these effects take time to show up and, when they do, are difficult to pin on any particular net neutrality decision, including the CRA resolution. Thus, measured in electoral time scales, prolonging net neutrality as a painful political issue — even though actual resolution of the process by legislation would be the sensible course — offers tremendous upside for political challengers and little cost.  

The truth is, there is widespread agreement that net neutrality issues need to be addressed by Congress: A constant back and forth between the FCC (and across its own administrations) and the courts runs counter to the interests of consumers, broadband companies, and edge providers alike. Virtually whatever that legislative solution ends up looking like, it would be an improvement over the unstable status quo.

There have been various proposals from Republicans and Democrats — many of which contain provisions that are likely bad ideas — but in the end, a bill passed with bipartisan input should have the virtue of capturing an open public debate on the issue. Legislation won’t be perfect, but it will be tremendously better than the advocacy playground that net neutrality has become.

What would the CRA accomplish?

Regardless of what one thinks of the substantive merits of TechFreedom’s arguments on the CRA and the arcana of legislative language distinguishing between agency “rules” and “orders,” if the CRA resolution is successful (a prospect that is a bit more likely following the Senate vote to pass it) what follows is pretty clear.

The only certain result of the the CRA resolution becoming law would be to void the transparency provisions that the FCC introduced in the RIFO — the one part of the Order that is pretty clearly a “rule” subject to CRA review — and it would disable the FCC from offering another transparency rule in its place. Everything else is going to end up — surprise! — before the courts, which would serve only to keep the issues surrounding net neutrality unsettled for another several years. (A cynic might suggest that this is, in fact, the goal of net neutrality proponents, for whom net neutrality has been and continues to have important political valence.)

And if the CRA resolution withstands the inevitable legal challenge to its rescision of the rest of the RIFO, it would also (once again) remove broadband privacy from the FTC’s purview, placing it back into the FCC’s lap — which is already prohibited from adopting privacy rules following last year’s successful CRA resolution undoing the Wheeler FCC’s broadband privacy regulations. The result is that we could be left without any broadband privacy regulator at all — presumably not the outcome strong net neutrality proponents want — but they persevere nonetheless.

Moreover, TechFreedom’s argument that the CRA may not apply to all parts of the RIFO could have a major effect on whether or not Congress is even accomplishing anything at all (other than scoring political points) with this vote. It could be the case that the CRA applies only to “rules” and not “orders,” or it could be the case that even if the CRA does apply to the RIFO, its passage would not force the FCC to revive the abrogated 2015 Open Internet Order, as proponents of the CRA vote hope.

Whatever one thinks of these arguments, however, they are based on a sound reading of the law and present substantial enough questions to sustain lengthy court challenges. Thus, far from a CRA vote actually putting to rest the net neutrality issue, it is likely to spawn litigation that will drag out the classification uncertainty question for at least another year (and probably more, with appeals).

Stop playing net neutrality games — they aren’t fun

Congress needs to stop trying to score easy political points on this issue while avoiding the hard and divisive work of reaching a compromise on actual net neutrality legislation. Despite how the CRA is presented in the popular media, a CRA vote is the furthest thing from a simple vote for net neutrality: It’s a political calculation to avoid accountability.

Although not always front page news, International Trade Commission (“ITC”) decisions can have major impacts on trade policy and antitrust law. Scott Kieff, a former ITC Commissioner, recently published a thoughtful analysis of Certain Carbon and Alloy Steel Products — a potentially important ITC investigation that implicates the intersection of these two policy areas. Scott was on the ITC when the investigation was initiated in 2016, but left in 2017 before the decision was finally issued in March of this year.

Perhaps most important, the case highlights an uncomfortable truth:

Sometimes (often?) Congress writes really bad laws and promotes really bad policies, but administrative agencies can do more harm to the integrity of our legal system by abusing their authority in an effort to override those bad policies.

In this case, that “uncomfortable truth” plays out in the context of the ITC majority’s effort to override Section 337 of the Tariff Act of 1930 by limiting the ability of the ITC to investigate alleged violations of the Act rooted in antitrust.

While we’re all for limiting the ability of competitors to use antitrust claims in order to impede competition (as one of us has noted: “Erecting barriers to entry and raising rivals’ costs through regulation are time-honored American political traditions”), it is inappropriate to make an end-run around valid and unambiguous legislation in order to do so — no matter how desirable the end result. (As the other of us has noted: “Attempts to [effect preferred policies] through any means possible are rational actions at an individual level, but writ large they may undermine the legal fabric of our system and should be resisted.”)

Brief background

Under Section 337, the ITC is empowered to, among other things, remedy

Unfair methods of competition and unfair acts in the importation of articles… into the United States… the threat or effect of which is to destroy or substantially injure an industry in the United States… or to restrain or monopolize trade and commerce in the United States.

In Certain Carbon and Alloy Steel Products, the ITC undertook an investigation — at the behest of U.S. Steel Corporation — into alleged violations of Section 337 by the Chinese steel industry. The complaint was based upon a number of claims, including allegations of price fixing.

As ALJ Lord succinctly summarizes in her Initial Determination:

For many years, the United States steel industry has complained of unfair trade practices by manufacturers of Chinese steel. While such practices have resulted in the imposition of high tariffs on certain Chinese steel products, U.S. Steel seeks additional remedies. The complaint by U.S. Steel in this case attempts to use section 337 of the Tariff Act of 1930 to block all Chinese carbon and alloy steel from coming into the United States. One of the grounds that U.S. Steel relies on is the allegation that the Chinese steel industry violates U.S. antitrust laws.

The ALJ dismissed the antitrust claims (alleging violations of the Sherman Act), however, concluding that they failed to allege antitrust injury as required by US courts deciding Sherman Act cases brought by private parties under the Clayton Act’s remedial provisions:

Under federal antitrust law, it is firmly established that a private complainant must show antitrust standing [by demonstrating antitrust injury]. U.S. Steel has not alleged that it has antitrust standing or the facts necessary to establish antitrust standing and erroneously contends it need not have antitrust standing to allege the unfair trade practice of restraining trade….

In its decision earlier this year, a majority of ITC commissioners agreed, and upheld the ALJ’s Initial Determination.

In comments filed with the ITC following the ALJ’s Initial Determination, we argued that the ALJ erred in her analysis:

Because antitrust injury is not an express requirement imposed by Congress, because ITC processes differ substantially from those of Article III courts, and because Section 337 is designed to serve different aims than private antitrust litigation, the Commission should reinstate the price fixing claims and allow the case to proceed.

Unfortunately, in upholding the Initial Determination, the Commission compounded this error, and also failed to properly understand the goals of the Tariff Act, and, by extension, its own role as arbiter of “unfair” trade practices.

A tale of two statutes

The case appears to turn on an arcane issue of adjudicative process in antitrust claims brought under the antitrust laws in federal court, on the one hand, versus antitrust claims brought under the Section 337 of the Tariff Act at the ITC, on the other. But it is actually about much more: the very purposes and structures of those laws.

The ALJ notes that

[The Chinese steel manufacturers contend that] under antitrust law as currently applied in federal courts, it has become very difficult for a private party like U.S. Steel to bring an antitrust suit against its competitors. Steel accepts this but says the law under section 337 should be different than in federal courts.

And as the ALJ further notes, this highlights the differences between the two regimes:

The dispute between U.S. Steel and the Chinese steel industry shows the conflict between section 337, which is intended to protect American industry from unfair competition, and U.S. antitrust laws, which are intended to promote competition for the benefit of consumers, even if such competition harms competitors.

Nevertheless, the ALJ (and the Commission) holds that antitrust laws must be applied in the same way in federal court as under Section 337 at the ITC.

It is this conclusion that is in error.

Judging from his article, it’s clear that Kieff agrees and would have dissented from the Commission’s decision. As he writes:

Unlike the focus in Section 16 of the Clayton Act on harm to the plaintiff, the provisions in the ITC’s statute — Section 337 — explicitly require the ITC to deal directly with harms to the industry or the market (rather than to the particular plaintiff)…. Where the statute protects the market rather than the individual complainant, the antitrust injury doctrine’s own internal logic does not compel the imposition of a burden to show harm to the particular private actor bringing the complaint. (Emphasis added)

Somewhat similar to the antitrust laws, the overall purpose of Section 337 focuses on broader, competitive harm — injury to “an industry in the United States” — not specific competitors. But unlike the Clayton Act, the Tariff Act does not accomplish this by providing a remedy for private parties alleging injury to themselves as a proxy for this broader, competitive harm.

As Kieff writes:

One stark difference between the two statutory regimes relates to the explicit goals that the statutes state for themselves…. [T]he Clayton Act explicitly states it is to remedy harm to only the plaintiff itself. This difference has particular significance for [the Commission’s decision in Certain Carbon and Alloy Steel Products] because the Supreme Court’s source of the private antitrust injury doctrine, its decision in Brunswick, explicitly tied the doctrine to this particular goal.

More particularly, much of the Court’s discussion in Brunswick focuses on the role the [antitrust injury] doctrine plays in mitigating the risk of unjustly enriching the plaintiff with damages awards beyond the amount of the particular antitrust harm that plaintiff actually suffered. The doctrine makes sense in the context of the Clayton Act proceedings in federal court because it keeps the cause of action focused on that statute’s stated goal of protecting a particular litigant only in so far as that party itself is a proxy for the harm to the market.

By contrast, since the goal of the ITC’s statute is to remedy for harm to the industry or to trade and commerce… there is no need to closely tie such broader harms to the market to the precise amounts of harms suffered by the particular complainant. (Emphasis and paragraph breaks added)

The mechanism by which the Clayton Act works is decidedly to remedy injury to competitors (including with treble damages). But because its larger goal is the promotion of competition, it cabins that remedy in order to ensure that it functions as an appropriate proxy for broader harms, and not simply a tool by which competitors may bludgeon each other. As Kieff writes:

The remedy provisions of the Clayton Act benefit much more than just the private plaintiff. They are designed to benefit the public, echoing the view that the private plaintiff is serving, indirectly, as a proxy for the market as a whole.

The larger purpose of Section 337 is somewhat different, and its remedial mechanism is decidedly different:

By contrast, the provisions in Section 337[] are much more direct in that they protect against injury to the industry or to trade and commerce more broadly. Harm to the particular complainant is essentially only relevant in so far as it shows harm to the industry or to trade and commerce more broadly. In turn, the remedies the ITC’s statute provides are more modest and direct in stopping any such broader harm that is determined to exist through a complete investigation.

The distinction between antitrust laws and trade laws is firmly established in the case law. And, in particular, trade laws not only focus on effects on industry rather than consumers or competition, per se, but they also contemplate a different kind of economic injury:

The “injury to industry” causation standard… focuses explicitly upon conditions in the U.S. industry…. In effect, Congress has made a judgment that causally related injury to the domestic industry may be severe enough to justify relief from less than fair value imports even if from another viewpoint the economy could be said to be better served by providing no relief. (Emphasis added)

Importantly, under Section 337 such harms to industry would ultimately have to be shown before a remedy would be imposed. In other words, demonstration of injury to competition is a constituent part of a case under Section 337. By contrast, such a demonstration is brought into an action under the antitrust laws by the antitrust injury doctrine as a function of establishing that the plaintiff has standing to sue as a proxy for broader harm to the market.

Finally, it should be noted, as ITC Commissioner Broadbent points out in her dissent from the Commission’s majority opinion, that U.S. Steel alleged in its complaint a violation of the Sherman Act, not the Clayton Act. Although its ability to enforce the Sherman Act arises from the remedial provisions of the Clayton Act, the substantive analysis of its claims is a Sherman Act matter. And the Sherman Act does not contain any explicit antitrust injury requirement. This is a crucial distinction because, as Commissioner Broadbent notes (quoting the Federal Circuit’s Tianrui case):

The “antitrust injury” standing requirement stems, not from the substantive antitrust statutes like the Sherman Act, but rather from the Supreme Court’s interpretation of the injury elements that must be proven under sections 4 and 16 of the Clayton Act.

* * *

Absent [] express Congressional limitation, restricting the Commission’s consideration of unfair methods of competition and unfair acts in international trade “would be inconsistent with the congressional purpose of protecting domestic commerce from unfair competition in importation….”

* * *

Where, as here, no such express limitation in the Sherman Act has been shown, I find no legal justification for imposing the insurmountable hurdle of demonstrating antitrust injury upon a typical U.S. company that is grappling with imports that benefit from the international unfair methods of competition that have been alleged in this case.

Section 337 is not a stand-in for other federal laws, even where it protects against similar conduct, and its aims diverge in important ways from those of other federal laws. It is, in other words, a trade protection provision, first and foremost, not an antitrust law, patent law, or even precisely a consumer protection statute.

The ITC hamstrings Itself

Kieff lays out a number of compelling points in his paper, including an argument that the ITC was statutorily designed as a convenient forum with broad powers in order to enable trade harms to be remedied without resort to expensive and protracted litigation in federal district court.

But, perhaps even more important, he points to a contradiction in the ITC’s decision that is directly related to its statutory design.

Under the Tariff Act, the Commission is entitled to self-initiate a Section 337 investigation identical to the one in Certain Alloy and Carbon Steel Products. And, as in this case, private parties are also entitled to file complaints with the Commission that can serve as the trigger for an investigation. In both instances, the ITC itself decides whether there is sufficient basis for proceeding, and, although an investigation unfolds much like litigation in federal court, it is, in fact, an investigation (and decision) undertaken by the ITC itself.

Although the Commission is statutorily mandated to initiate an investigation once a complaint is properly filed, this is subject to a provision requiring the Commission to “examine the complaint for sufficiency and compliance with the applicable sections of this Chapter.” Thus, the Commission conducts a preliminary investigation to determine if the complaint provides a sound basis for institution of an investigation, not unlike an assessment of standing and evaluation of the sufficiency of a complaint in federal court — all of which happens before an official investigation is initiated.

Yet despite the fact that, before an investigation begins, the ITC either 1) decides for itself that there is sufficient basis to initiate its own action, or else 2) evaluates the sufficiency of a private complaint to determine if the Commission should initiate an action, the logic of the decision in Certain Alloy and Carbon Steel Products would apply different standards in each case. Writes Kieff:

There appears to be broad consensus that the ITC can self-initiate an antitrust case under Section 337 and in such a proceeding would not be required to apply the antitrust injury doctrine to itself or to anyone else…. [I]t seems odd to make [this] legal distinction… After all, if it turned out there really were harm to a domestic industry or trade and commerce in this case, it would be strange for the ITC to have to dismiss this action and deprive itself of the benefit of the advance work and ongoing work of the private party [just because it was brought to the ITC’s attention by a private party complaint], only to either sit idle or expend the resources to — flying solo that time — reinitiate and proceed to completion.

Odd indeed, because, in the end, what is instituted is an investigation undertaken by the ITC — whether it originates from a private party or from its own initiative. The role of a complaining party before the ITC is quite distinct from that of a plaintiff in an Article III court.

In trade these days, it always comes down to China

We are hesitant to offer justifications for Congress’ decision to grant the ITC a sweeping administrative authority to prohibit the “unfair” importation of articles into the US, but there could be good reasons that Congress enacted the Tariff Act as a protectionist statute.

In a recent Law360 article, Kieff noted that analyzing anticompetitive behavior in the trade context is more complicated than in the domestic context. To take the current example: By limiting the complainant’s ability to initiate an ITC action based on a claim that foreign competitors are conspiring to keep prices artificially low, the ITC majority decision may be short-sighted insofar as keeping prices low might actually be part of a larger industrial and military policy for the Chinese government:

The overlooked problem is that, as the ITC petitioners claim, the Chinese government is using its control over many Chinese steel producers to accomplish full-spectrum coordination on both price and quantity. Mere allegations of course would have to be proven; but it’s not hard to imagine that such coordination could afford the Chinese government effective surveillance and control over  almost the entire worldwide supply chain for steel products.

This access would help the Chinese government run significant intelligence operations…. China is allegedly gaining immense access to practically every bid and ask up and down the supply chain across the global steel market in general, and our domestic market in particular. That much real-time visibility across steel markets can in turn give visibility into defense, critical infrastructure and finance.

Thus, by taking it upon itself to artificially narrow its scope of authority, the ITC could be undermining a valid congressional concern: that trade distortions not be used as a way to allow a foreign government to gain a more pervasive advantage over diplomatic and military operations.

No one seriously doubts that China is, at the very least, a supportive partner to much of its industry in a way that gives that industry some potential advantage over competitors operating in countries that receive relatively less assistance from national governments.

In certain industries — notably semiconductors and patent-intensive industries more broadly — the Chinese government regularly imposes onerous conditions (including mandatory IP licensing and joint ventures with Chinese firms, invasive audits, and obligatory software and hardware “backdoors”) on foreign tech companies doing business in China. It has long been an open secret that these efforts, ostensibly undertaken for the sake of national security, are actually aimed at protecting or bolstering China’s domestic industry.

And China could certainly leverage these partnerships to obtain information on a significant share of important industries and their participants throughout the world. After all, we are well familiar with this business model: cheap or highly subsidized access to a desired good or service in exchange for user data is the basic description of modern tech platform companies.

Only Congress can fix Congress

Stepping back from the ITC context, a key inquiry when examining antitrust through a trade lens is the extent to which countries will use antitrust as a non-tariff barrier to restrain trade. It is certainly the case that a sort of “mutually assured destruction” can arise where every country chooses to enforce its own ambiguously worded competition statute in a way that can favor its domestic producers to the detriment of importers. In the face of that concern, the impetus to try to apply procedural constraints on open-ended competition laws operating in the trade context is understandable.

And as a general matter, it also makes sense to be concerned when producers like U.S. Steel try to use our domestic antitrust laws to disadvantage Chinese competitors or keep them out of the market entirely.

But in this instance the analysis is more complicated. Like it or not, what amounts to injury in the international trade context, even with respect to anticompetitive conduct, is different than what’s contemplated under the antitrust laws. When the Tariff Act of 1922 was passed (which later became Section 337) the Senate Finance Committee Report that accompanied it described the scope of its unfair methods of competition authority as “broad enough to prevent every type and form of unfair practice” involving international trade. At the same time, Congress pretty clearly gave the ITC the discretion to proceed on a much less-constrained basis than that on which Article III courts operate.

If these are problems, Congress needs to fix them, not the ITC acting sua sponte.

Moreover, as Kieff’s paper (and our own comments in the Certain Alloy and Carbon Steel Products investigation) make clear, there are also a number of relevant, practical distinctions between enforcement of the antitrust laws in a federal court in a case brought by a private plaintiff and an investigation of alleged anticompetitive conduct by the ITC under Section 337. Every one of these cuts against importing an antitrust injury requirement from federal court into ITC adjudication.

Instead, understandable as its motivation may be, the ITC majority’s approach in Certain Alloy and Carbon Steel Products requires disregarding Congressional intent, and that’s simply not a tenable interpretive approach for administrative agencies to take.

Protectionism is a terrible idea, but if that’s how Congress wrote the Tariff Act, the ITC is legally obligated to enforce the protectionist law it is given.

The FTC will hold an “Informational Injury Workshop” in December “to examine consumer injury in the context of privacy and data security.” Defining the scope of cognizable harm that may result from the unauthorized use or third-party hacking of consumer information is, to be sure, a crucial inquiry, particularly as ever-more information is stored digitally. But the Commission — rightly — is aiming at more than mere definition. As it notes, the ultimate objective of the workshop is to address questions like:

How do businesses evaluate the benefits, costs, and risks of collecting and using information in light of potential injuries? How do they make tradeoffs? How do they assess the risks of different kinds of data breach? What market and legal incentives do they face, and how do these incentives affect their decisions?

How do consumers perceive and evaluate the benefits, costs, and risks of sharing information in light of potential injuries? What obstacles do they face in conducting such an evaluation? How do they evaluate tradeoffs?

Understanding how businesses and consumers assess the risk and cost “when information about [consumers] is misused,” and how they conform their conduct to that risk, entails understanding not only the scope of the potential harm, but also the extent to which conduct affects the risk of harm. This, in turn, requires an understanding of the FTC’s approach to evaluating liability under Section 5 of the FTC Act.

The problem, as we discuss in comments submitted by the International Center for Law & Economics to the FTC for the workshop, is that the Commission’s current approach troublingly mixes the required separate analyses of risk and harm, with little elucidation of either.

The core of the problem arises from the Commission’s reliance on what it calls a “reasonableness” standard for its evaluation of data security. By its nature, a standard that assigns liability for only unreasonable conduct should incorporate concepts resembling those of a common law negligence analysis — e.g., establishing a standard of due care, determining causation, evaluating the costs of and benefits of conduct that would mitigate the risk of harm, etc. Unfortunately, the Commission’s approach to reasonableness diverges from the rigor of a negligence analysis. In fact, as it has developed, it operates more like a strict liability regime in which largely inscrutable prosecutorial discretion determines which conduct, which firms, and which outcomes will give rise to liability.

Most troublingly, coupled with the Commission’s untenably lax (read: virtually nonexistent) evidentiary standards, the extremely liberal notion of causation embodied in its “reasonableness” approach means that the mere storage of personal information, even absent any data breach, could amount to an unfair practice under the Act — clearly not a “reasonable” result.

The notion that a breach itself can constitute injury will, we hope, be taken up during the workshop. But even if injury is limited to a particular type of breach — say, one in which sensitive, personal information is exposed to a wide swath of people — unless the Commission’s definition of what it means for conduct to be “likely to cause” harm is fixed, it will virtually always be the case that storage of personal information could conceivably lead to the kind of breach that constitutes injury. In other words, better defining the scope of injury does little to cabin the scope of the agency’s discretion when conduct creating any risk of that injury is actionable.

Our comments elaborate on these issues, as well as providing our thoughts on how the subjective nature of informational injuries can fit into Section 5, with a particular focus on the problem of assessing informational injury given evolving social context, and the need for appropriately assessing benefits in any cost-benefit analysis of conduct leading to informational injury.

ICLE’s full comments are available here.

The comments draw upon our article, When ‘Reasonable’ Isn’t: The FTC’s Standard-Less Data Security Standard, forthcoming in the Journal of Law, Economics and Policy.

Over the weekend, Senator Al Franken and FCC Commissioner Mignon Clyburn issued an impassioned statement calling for the FCC to thwart the use of mandatory arbitration clauses in ISPs’ consumer service agreements — starting with a ban on mandatory arbitration of privacy claims in the Chairman’s proposed privacy rules. Unfortunately, their call to arms rests upon a number of inaccurate or weak claims. Before the Commissioners vote on the proposed privacy rules later this week, they should carefully consider whether consumers would actually be served by such a ban.

FCC regulations can’t override congressional policy favoring arbitration

To begin with, it is firmly cemented in Supreme Court precedent that the Federal Arbitration Act (FAA) “establishes ‘a liberal federal policy favoring arbitration agreements.’” As the Court recently held:

[The FAA] reflects the overarching principle that arbitration is a matter of contract…. [C]ourts must “rigorously enforce” arbitration agreements according to their terms…. That holds true for claims that allege a violation of a federal statute, unless the FAA’s mandate has been “overridden by a contrary congressional command.”

For better or for worse, that’s where the law stands, and it is the exclusive province of Congress — not the FCC — to change it. Yet nothing in the Communications Act (to say nothing of the privacy provisions in Section 222 of the Act) constitutes a “contrary congressional command.”

And perhaps that’s for good reason. In enacting the statute, Congress didn’t demonstrate the same pervasive hostility toward companies and their relationships with consumers that has characterized the way this FCC has chosen to enforce the Act. As Commissioner O’Rielly noted in dissenting from the privacy NPRM:

I was also alarmed to see the Commission acting on issues that should be completely outside the scope of this proceeding and its jurisdiction. For example, the Commission seeks comment on prohibiting carriers from including mandatory arbitration clauses in contracts with their customers. Here again, the Commission assumes that consumers don’t understand the choices they are making and is willing to impose needless costs on companies by mandating how they do business.

If the FCC were to adopt a provision prohibiting arbitration clauses in its privacy rules, it would conflict with the FAA — and the FAA would win. Along the way, however, it would create a thorny uncertainty for both companies and consumers seeking to enforce their contracts.  

The evidence suggests that arbitration is pro-consumer

But the lack of legal authority isn’t the only problem with the effort to shoehorn an anti-arbitration bias into the Commission’s privacy rules: It’s also bad policy.

In its initial broadband privacy NPRM, the Commission said this about mandatory arbitration:

In the 2015 Open Internet Order, we agreed with the observation that “mandatory arbitration, in particular, may more frequently benefit the party with more resources and more understanding of the dispute procedure, and therefore should not be adopted.” We further discussed how arbitration can create an asymmetrical relationship between large corporations that are repeat players in the arbitration system and individual customers who have fewer resources and less experience. Just as customers should not be forced to agree to binding arbitration and surrender their right to their day in court in order to obtain broadband Internet access service, they should not have to do so in order to protect their private information conveyed through that service.

The Commission may have “agreed with the cited observations about arbitration, but that doesn’t make those views accurate. As one legal scholar has noted, summarizing the empirical data on the effects of arbitration:

[M]ost of the methodologically sound empirical research does not validate the criticisms of arbitration. To give just one example, [employment] arbitration generally produces higher win rates and higher awards for employees than litigation.

* * *

In sum, by most measures — raw win rates, comparative win rates, some comparative recoveries and some comparative recoveries relative to amounts claimed — arbitration generally produces better results for claimants [than does litigation].

A comprehensive, empirical study by Northwestern Law’s Searle Center on AAA (American Arbitration Association) cases found much the same thing, noting in particular that

  • Consumer claimants in arbitration incur average arbitration fees of only about $100 to arbitrate small (under $10,000) claims, and $200 for larger claims (up to $75,000).
  • Consumer claimants also win attorneys’ fees in over 60% of the cases in which they seek them.
  • On average, consumer arbitrations are resolved in under 7 months.
  • Consumers win some relief in more than 50% of cases they arbitrate…
  • And they do almost exactly as well in cases brought against “repeat-player” business.

In short, it’s extremely difficult to sustain arguments suggesting that arbitration is tilted against consumers relative to litigation.

(Upper) class actions: Benefitting attorneys — and very few others

But it isn’t just any litigation that Clyburn and Franken seek to preserve; rather, they are focused on class actions:

If you believe that you’ve been wronged, you could take your service provider to court. But you’d have to find a lawyer willing to take on a multi-national telecom provider over a few hundred bucks. And even if you won the case, you’d likely pay more in legal fees than you’d recover in the verdict.

The only feasible way for you as a customer to hold that corporation accountable would be to band together with other customers who had been similarly wronged, building a case substantial enough to be worth the cost—and to dissuade that big corporation from continuing to rip its customers off.

While — of course — litigation plays an important role in redressing consumer wrongs, class actions frequently don’t confer upon class members anything close to the imagined benefits that plaintiffs’ lawyers and their congressional enablers claim. According to a 2013 report on recent class actions by the law firm, Mayer Brown LLP, for example:

  • “In [the] entire data set, not one of the class actions ended in a final judgment on the merits for the plaintiffs. And none of the class actions went to trial, either before a judge or a jury.” (Emphasis in original).
  • “The vast majority of cases produced no benefits to most members of the putative class.”
  • “For those cases that do settle, there is often little or no benefit for class members. What is more, few class members ever even see those paltry benefits — particularly in consumer class actions.”
  • “The bottom line: The hard evidence shows that class actions do not provide class members with anything close to the benefits claimed by their proponents, although they can (and do) enrich attorneys.”

Similarly, a CFPB study of consumer finance arbitration and litigation between 2008 and 2012 seems to indicate that the class action settlements and judgments it studied resulted in anemic relief to class members, at best. The CFPB tries to disguise the results with large, aggregated and heavily caveated numbers (never once actually indicating what the average payouts per person were) that seem impressive. But in the only hard numbers it provides (concerning four classes that ended up settling in 2013), promised relief amounted to under $23 each (comprising both cash and in-kind payment) if every class member claimed against the award. Back-of-the-envelope calculations based on the rest of the data in the report suggest that result was typical.

Furthermore, the average time to settlement of the cases the CFPB looked at was almost 2 years. And somewhere between 24% and 37% involved a non-class settlement — meaning class members received absolutely nothing at all because the named plaintiff personally took a settlement.

By contrast, according to the Searle Center study, the average award in the consumer-initiated arbitrations it studied (admittedly, involving cases with a broader range of claims) was almost $20,000, and the average time to resolution was less than 7 months.

To be sure, class action litigation has been an important part of our system of justice. But, as Arthur Miller — a legal pioneer who helped author the rules that make class actions viable — himself acknowledged, they are hardly a panacea:

I believe that in the 50 years we have had this rule, that there are certain class actions that never should have been brought, admitted; that we have burdened our judiciary, yes. But we’ve had a lot of good stuff done. We really have.

The good that has been done, according to Professor Miller, relates in large part to the civil rights violations of the 50’s and 60’s, which the class action rules were designed to mitigate:

Dozens and dozens and dozens of communities were desegregated because of the class action. You even see desegregation decisions in my old town of Boston where they desegregated the school system. That was because of a class action.

It’s hard to see how Franken and Clyburn’s concern for redress of “a mysterious 99-cent fee… appearing on your broadband bill” really comes anywhere close to the civil rights violations that spawned the class action rules. Particularly given the increasingly pervasive role of the FCC, FTC, and other consumer protection agencies in addressing and deterring consumer harms (to say nothing of arbitration itself), it is manifestly unclear why costly, protracted litigation that infrequently benefits anyone other than trial attorneys should be deemed so essential.

“Empowering the 21st century [trial attorney]”

Nevertheless, Commissioner Clyburn and Senator Franken echo the privacy NPRM’s faulty concerns about arbitration clauses that restrict consumers’ ability to litigate in court:

If you’re prohibited from using our legal system to get justice when you’re wronged, what’s to protect you from being wronged in the first place?

Well, what do they think the FCC is — chopped liver?

Hardly. In fact, it’s a little surprising to see Commissioner Clyburn (who sits on a Commission that proudly proclaims that “[p]rotecting consumers is part of [its] DNA”) and Senator Franken (among Congress’ most vocal proponents of the FCC’s claimed consumer protection mission) asserting that the only protection for consumers from ISPs’ supposed depredations is the cumbersome litigation process.

In fact, of course, the FCC has claimed for itself the mantle of consumer protector, aimed at “Empowering the 21st Century Consumer.” But nowhere does the agency identify “promoting and preserving the rights of consumers to litigate” among its tools of consumer empowerment (nor should it). There is more than a bit of irony in a federal regulator — a commissioner of an agency charged with making sure, among other things, that corporations comply with the law — claiming that, without class actions, consumers are powerless in the face of bad corporate conduct.

Moreover, even if it were true (it’s not) that arbitration clauses tend to restrict redress of consumer complaints, effective consumer protection would still not necessarily be furthered by banning such clauses in the Commission’s new privacy rules.

The FCC’s contemplated privacy regulations are poised to introduce a wholly new and untested regulatory regime with (at best) uncertain consequences for consumers. Given the risk of consumer harm resulting from the imposition of this new regime, as well as the corollary risk of its excessive enforcement by complainants seeking to test or push the boundaries of new rules, an agency truly concerned with consumer protection would tread carefully. Perhaps, if the rules were enacted without an arbitration ban, it would turn out that companies would mandate arbitration (though this result is by no means certain, of course). And perhaps arbitration and agency enforcement alone would turn out to be insufficient to effectively enforce the rules. But given the very real costs to consumers of excessive, frivolous or potentially abusive litigation, cabining the litigation risk somewhat — even if at first it meant the regime were tilted slightly too much against enforcement — would be the sensible, cautious and pro-consumer place to start.

____

Whether rooted in a desire to “protect” consumers or not, the FCC’s adoption of a rule prohibiting mandatory arbitration clauses to address privacy complaints in ISP consumer service agreements would impermissibly contravene the FAA. As the Court has made clear, such a provision would “‘stand[] as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress’ embodied in the Federal Arbitration Act.” And not only would such a rule tend to clog the courts in contravention of the FAA’s objectives, it would do so without apparent benefit to consumers. Even if such a rule wouldn’t effectively be invalidated by the FAA, the Commission should firmly reject it anyway: A rule that operates primarily to enrich class action attorneys at the expense of their clients has no place in an agency charged with protecting the public interest.

There must have been a great gnashing of teeth in Chairman Wheeler’s office this morning as the FCC announced that it was pulling the Chairman’s latest modifications to the set-top box proposal from its voting agenda. This is surely but a bump in the road for the Chairman; he will undoubtedly press ever onward in his quest to “fix” a market that is flooded with competition and consumer choice. But, as we stop to take a breath for a moment while this latest FCC adventure is temporarily paused, there is a larger issue worth considering: the lack of transparency at the FCC.

Although the Commission has an unfortunate tradition of non-disclosure surrounding many of its regulatory proposals, the problem has seemingly been exacerbated by Chairman Wheeler’s aggressive agenda and his intransigence in the face of overwhelming and rigorous criticism.

Perhaps nowhere was this attitude more apparent than with his handling of the Open Internet Order, which was plagued with enough process problems to elicit a call for a delay of the Commission’s vote on the initial rules from Democratic Commissioner Rosenworcel, and a strong rebuke from the Chairman of the House Oversight Committee prior to the Commission’s vote on the final rules (which were not disclosed to the public until after the vote).

But the same cavalier dismissal of public and stakeholder input has plagued the Chairman’s beleaguered set-top box proposal, as well.

As Commissioner Pai noted before Congress in March:

The FCC continues to choose opacity over transparency. The decisions we make impact hundreds of millions of Americans and thousands of small businesses. And yet to the public, to Congress, and even to the Commissioners at the FCC, the agency’s work remains a black box.

Take this simple proposition: The public should be able to see what we’re voting on before we vote on it. That’s how Congress works, as you know. Anyone can look up any pending bill right now by going to congress.gov. And that’s how many state commissions work too. But not the FCC.

Exhibit A in Commissioner Pai’s lament was the set-top box proceeding:

Instead, the public gets to see only what the Chairman’s Office deigns to release, so controversial policy proposals can be (and typically are) hidden in a wave of media adulation. That happened just last month when the agency proposed changes to its set-top-box rules but tried to mislead content producers and the public about whether set-top box manufacturers would be permitted to insert their own advertisements into programming streams.

Now, although the Chairman’s initial proposal was eventually released, we have only a fact sheet and an op-ed by Chairman Wheeler on which to judge the purportedly substantial changes embodied in his latest version.

Even Democrats in Congress have recognized the process problems that have plagued this proceeding. As Senator Feinstein (D-CA) urged in a recent letter to Chairman Wheeler:

Given the significance of this proceeding, I ask that you make public the new proposal under consideration by the Commission, so that all interested stakeholders, members of Congress, copyright experts, and others can comment on the potential copyright implications of the new proposal before the Commission votes on it.

And as Senator Heller (R-NV) wrote in a letter to Chairman Wheeler this week:

I believe it is unacceptable that the FCC has not released the text of this proposal before Thursday’s vote. A three-page fact sheet does not provide enough details for Congress to conduct proper oversight of this rulemaking that will significantly impact both consumers and industry…. I encourage you to release the text immediately so that the American public has a full understanding of what is being considered by the Commission….

Of course, this isn’t a new problem at the FCC. In fact, before he supported Chairman Wheeler’s efforts to impose Open Internet rules without sufficient public disclosure, then-Senator Obama decried then-Chairman Martin’s efforts to enact new media ownership rules with insufficient process in 2007:

Repealing the cross ownership rules and retaining the rest of our existing regulations is not a proposal that has been put out for public comment; the proper process for vetting it is not in closed door meetings with lobbyists or in selective leaks to the New York Times.

Although such a proposal may pass the muster of a federal court, Congress and the public have the right to review any specific proposal and decide whether or not it constitutes sound policy. And the Commission has the responsibility to defend any new proposal in public discourse and debate.

And although you won’t find them complaining this time (because this time they want the excessive intervention that the NPRM seems to contemplate), regulatory advocates lamented just exactly this sort of secrecy at the Commission when Chairman Genachowski proposed his media ownership rules in 2012. At that time Free Press angrily wrote:

[T]he Commission still has not made public its actual media ownership order…. Furthermore, it’s disingenuous for the FCC to suggest that its process now is more transparent than the one former Chairman Martin used to adopt similar rules. Genachowski’s FCC has yet to publish any details of its final proposal, offering only vague snippets in press releases… despite the president’s instruction to rulemaking agencies to conduct any significant business in open meetings with opportunities for members of the public to have their voices heard.

As Free Press noted, President Obama did indeed instruct “agencies to conduct any significant business in open meetings with opportunities for members of the public to have their voices heard.” In his Memorandum on Transparency and Open Government, his first executive action, the president urged that:

Public engagement enhances the Government’s effectiveness and improves the quality of its decisions. Knowledge is widely dispersed in society, and public officials benefit from having access to that dispersed knowledge. Executive departments and agencies should offer Americans increased opportunities to participate in policymaking and to provide their Government with the benefits of their collective expertise and information.

The resulting Open Government Directive calls on executive agencies to

take prompt steps to expand access to information by making it available online in open formats. With respect to information, the presumption shall be in favor of openness….

The FCC is not an “executive agency,” and so is not directly subject to the Directive. But the Chairman’s willingness to stray so far from basic principles of transparency is woefully inconsistent with the basic principles of good government and the ideals of heightened transparency claimed by this administration.

As Commissioner Wheeler moves forward with his revised set-top box proposal, and on the eve of tomorrow’s senate FCC oversight hearing, we would do well to reflect on some insightful testimony regarding another of the Commission’s rulemakings from ten years ago:

We are living in a digital gold age and consumers… are the beneficiaries. Consumers have numerous choices for buying digital content and for buying devices on which to play that content. They have never had so much flexibility and so much opportunity.  

* * *

As the content industry has ramped up on-line delivery of content, it has been testing a variety of protection measures that provide both security for the industry and flexibility for consumers.

So to answer the question, can content protection and technological innovation coexist?  It is a resounding yes. Look at the robust market for on-line content distribution facilitated by the technologies and networks consumers love.

* * *

[T]he Federal Communications Commission should not become the Federal Computer Commission or the Federal Copyright Commission, and the marketplace, not the Government, is the best arbiter of what technologies succeed or fail.

That’s not the self-interested testimony of a studio or cable executive — that was Gigi Sohn, current counsel to Chairman Wheeler, speaking on behalf of Public Knowledge in 2006 before the House Energy and Commerce Committee against the FCC’s “broadcast flag” rules. Those rules, supported by a broad spectrum of rightsholders, required consumer electronics devices to respect programming conditions preventing the unauthorized transmission over the internet of digital broadcast television content.

Ms. Sohn and Public Knowledge won that fight in court, convincing the DC Circuit that Congress hadn’t given the FCC authority to impose the rules in the first place, and she successfully urged Congress not to give the FCC the authority to reinstate them.

Yet today, she and the Chairman seem to have forgotten her crucial insights from ten years ago. If the marketplace for video content was sufficiently innovative and competitive then, how can it possibly not be so now, with audiences having orders of magnitude more choices, both online and off? And if the FCC lacked authority to adopt copyright-related rules then, how does the FCC suddenly have that authority now, in the absence of any intervening congressional action?

With Section 106 of the Copyright Act, Congress granted copyright holders the exclusive rights to engage in or license the reproduction, distribution, and public performance of their works. The courts are the “backstop,” not the FCC (as Chairman Wheeler would have it), and section 629 of the Communications Act doesn’t say otherwise. All section 629 does is direct the FCC to promote a competitive market for devices to access pay-TV services from pay-TV providers. As we noted last week, it very simply doesn’t allow the FCC to interfere with the license arrangements that fill those devices, and, short of explicit congressional direction, the Commission is simply not empowered to interfere with the framework set forth in the Copyright Act.

Chairman Wheeler’s latest proposal has improved on his initial plan by, for example, moving toward an applications-based approach and away from the mandatory disaggregation of content. But it would still arrogate to the FCC the authority to stand up a licensing body for the distribution of content over pay-TV applications; set rules on the terms such licenses must, may, and may not include; and even allow the FCC itself to create terms or the entire license. Such rules would necessarily implicate the extent to which rightsholders are able to control the distribution of their content.

The specifics of the regulations may be different from 2006, but the point is the same: What the FCC could not do in 2006, it cannot do today.

Imagine if you will… that a federal regulatory agency were to decide that the iPhone ecosystem was too constraining and too expensive; that consumers — who had otherwise voted for iPhones with their dollars — were being harmed by the fact that the platform was not “open” enough.

Such an agency might resolve (on the basis of a very generous reading of a statute), to force Apple to make its iOS software available to any hardware platform that wished to have it, in the process making all of the apps and user data accessible to the consumer via these new third parties, on terms set by the agency… for free.

Difficult as it may be to picture this ever happening, it is exactly the sort of Twilight Zone scenario that FCC Chairman Tom Wheeler is currently proposing with his new set-top box proposal.

Based on the limited information we have so far (a fact sheet and an op-ed), Chairman Wheeler’s new proposal does claw back some of the worst excesses of his initial draft (which we critiqued in our comments and reply comments to that proposal).

But it also appears to reinforce others — most notably the plan’s disregard for the right of content creators to control the distribution of their content. Wheeler continues to dismiss the complex business models, relationships, and licensing terms that have evolved over years of competition and innovation. Instead, he offers  a one-size-fits-all “solution” to a “problem” that market participants are already falling over themselves to provide.

Plus ça change…

To begin with, Chairman Wheeler’s new proposal is based on the same faulty premise: that consumers pay too much for set-top boxes, and that the FCC is somehow both prescient enough and Congressionally ordained to “fix” this problem. As we wrote in our initial comments, however,

[a]lthough the Commission asserts that set-top boxes are too expensive, the history of overall MVPD prices tells a remarkably different story. Since 1994, per-channel cable prices including set-top box fees have fallen by 2 percent, while overall consumer prices have increased by 54 percent. After adjusting for inflation, this represents an impressive overall price decrease.

And the fact is that no one buys set-top boxes in isolation; rather, the price consumers pay for cable service includes the ability to access that service. Whether the set-top box fee is broken out on subscribers’ bills or not, the total price consumers pay is unlikely to change as a result of the Commission’s intervention.

As we have previously noted, the MVPD set-top box market is an aftermarket; no one buys set-top boxes without first (or simultaneously) buying MVPD service. And as economist Ben Klein (among others) has shown, direct competition in the aftermarket need not be plentiful for the market to nevertheless be competitive:

Whether consumers are fully informed or uninformed, consumers will pay a competitive package price as long as sufficient competition exists among sellers in the [primary] market.

Engineering the set-top box aftermarket to bring more direct competition to bear may redistribute profits, but it’s unlikely to change what consumers pay.

Stripped of its questionable claims regarding consumer prices and placed in the proper context — in which consumers enjoy more ways to access more video content than ever before — Wheeler’s initial proposal ultimately rested on its promise to “pave the way for a competitive marketplace for alternate navigation devices, and… end the need for multiple remote controls.” Weak sauce, indeed.

He now adds a new promise: that “integrated search” will be seamlessly available for consumers across the new platforms. But just as universal remotes and channel-specific apps on platforms like Apple TV have already made his “multiple remotes” promise a hollow one, so, too, have competitive pressures already begun to deliver integrated search.

Meanwhile, such marginal benefits come with a host of substantial costs, as others have pointed out. Do we really need the FCC to grant itself more powers and create a substantial and coercive new regulatory regime to mandate what the market is already poised to provide?

From ignoring copyright to obliterating copyright

Chairman Wheeler’s first proposal engendered fervent criticism for the impossible position in which it placed MVPDs — of having to disregard, even outright violate, their contractual obligations to content creators.

Commendably, the new proposal acknowledges that contractual relationships between MVPDs and content providers should remain “intact.” Thus, the proposal purports to enable programmers and MVPDs to maintain “their channel position, advertising and contracts… in place.” MVPDs will retain “end-to-end” control of the display of content through their apps, and all contractually guaranteed content protection mechanisms will remain, because the “pay-TV’s software will manage the full suite of linear and on-demand programming licensed by the pay-TV provider.”

But, improved as it is, the new proposal continues to operate in an imagined world where the incredibly intricate and complex process by which content is created and distributed can be reduced to the simplest of terms, dictated by a regulator and applied uniformly across all content and all providers.

According to the fact sheet, the new proposal would “[p]rotect[] copyrights and… [h]onor[] the sanctity of contracts” through a “standard license”:

The proposed final rules require the development of a standard license governing the process for placing an app on a device or platform. A standard license will give device manufacturers the certainty required to bring innovative products to market… The license will not affect the underlying contracts between programmers and pay-TV providers. The FCC will serve as a backstop to ensure that nothing in the standard license will harm the marketplace for competitive devices.

But programming is distributed under a diverse range of contract terms. The only way a single, “standard license” could possibly honor these contracts is by forcing content providers to license all of their content under identical terms.

Leaving aside for a moment the fact that the FCC has no authority whatever to do this, for such a scheme to work, the agency would necessarily have to strip content holders of their right to govern the terms on which their content is accessed. After all, if MVPDs are legally bound to redistribute content on fixed terms, they have no room to permit content creators to freely exercise their rights to specify terms like windowing, online distribution restrictions, geographic restrictions, and the like.

In other words, the proposal simply cannot deliver on its promise that “[t]he license will not affect the underlying contracts between programmers and pay-TV providers.”

But fear not: According to the Fact Sheet, “[p]rogrammers will have a seat at the table to ensure that content remains protected.” Such largesse! One would be forgiven for assuming that the programmers’ (single?) seat will surrounded by those of other participants — regulatory advocates, technology companies, and others — whose sole objective will be to minimize content companies’ ability to restrict the terms on which their content is accessed.

And we cannot ignore the ominous final portion of the Fact Sheet’s “Standard License” description: “The FCC will serve as a backstop to ensure that nothing in the standard license will harm the marketplace for competitive devices.” Such an arrogation of ultimate authority by the FCC doesn’t bode well for that programmer’s “seat at the table” amounting to much.

Unfortunately, we can only imagine the contours of the final proposal that will describe the many ways by which distribution licenses can “harm the marketplace for competitive devices.” But an educated guess would venture that there will be precious little room for content creators and MVPDs to replicate a large swath of the contract terms they currently employ. “Any content owner can have its content painted any color that it wants, so long as it is black.”

At least we can take solace in the fact that the FCC has no authority to do what Wheeler wants it to do

And, of course, this all presumes that the FCC will be able to plausibly muster the legal authority in the Communications Act to create what amounts to a de facto compulsory licensing scheme.

A single license imposed upon all MVPDs, along with the necessary restrictions this will place upon content creators, does just as much as an overt compulsory license to undermine content owners’ statutory property rights. For every license agreement that would be different than the standard agreement, the proposed standard license would amount to a compulsory imposition of terms that the rights holders and MVPDs would not otherwise have agreed to. And if this sounds tedious and confusing, just wait until the Commission starts designing its multistakeholder Standard Licensing Oversight Process (“SLOP”)….

Unfortunately for Chairman Wheeler (but fortunately for the rest of us), the FCC has neither the legal authority, nor the requisite expertise, to enact such a regime.

Last month, the Copyright Office was clear on this score in its letter to Congress commenting on the Chairman’s original proposal:  

[I]t is important to remember that only Congress, through the exercise of its power under the Copyright Clause, and not the FCC or any other agency, has the constitutional authority to create exceptions and limitations in copyright law. While Congress has enacted compulsory licensing schemes, they have done so in response to demonstrated market failures, and in a carefully circumscribed manner.

Assuming that Section 629 of the Communications Act — the provision that otherwise empowers the Commission to promote a competitive set-top box market — fails to empower the FCC to rewrite copyright law (which is assuredly the case), the Commission will be on shaky ground for the inevitable torrent of lawsuits that will follow the revised proposal.

In fact, this new proposal feels more like an emergency pivot by a panicked Chairman than an actual, well-grounded legal recommendation. While the new proposal improves upon the original, it retains at its core the same ill-informed, ill-advised and illegal assertion of authority that plagued its predecessor.