Archives For regulation

A debate has broken out among the four sitting members of the Federal Trade Commission (FTC) in connection with the recently submitted FTC Report to Congress on Privacy and Security. Chair Lina Khan argues that the commission “must explore using its rulemaking tools to codify baseline protections,” while Commissioner Rebecca Kelly Slaughter has urged the FTC to initiate a broad-based rulemaking proceeding on data privacy and security. By contrast, Commissioners Noah Joshua Phillips and Christine Wilson counsel against a broad-based regulatory initiative on privacy.

Decisions to initiate a rulemaking should be viewed through a cost-benefit lens (See summaries of Thom Lambert’s masterful treatment of regulation, of which rulemaking is a subset, here and here). Unless there is a market failure, rulemaking is not called for. Even in the face of market failure, regulation should not be adopted unless it is more cost-beneficial than reliance on markets (including the ability of public and private litigation to address market-failure problems, such as data theft). For a variety of reasons, it is unlikely that FTC rulemaking directed at privacy and data security would pass a cost-benefit test.

Discussion

As I have previously explained (see here and here), FTC rulemaking pursuant to Section 6(g) of the FTC Act (which authorizes the FTC “to make rules and regulations for the purpose of carrying out the provisions of this subchapter”) is properly read as authorizing mere procedural, not substantive, rules. As such, efforts to enact substantive competition rules would not pass a cost-benefit test. Such rules could well be struck down as beyond the FTC’s authority on constitutional law grounds, and as “arbitrary and capricious” on administrative law grounds. What’s more, they would represent retrograde policy. Competition rules would generate higher error costs than adjudications; could be deemed to undermine the rule of law, because the U.S. Justice Department (DOJ) could not apply such rules; and innovative efficiency-seeking business arrangements would be chilled.

Accordingly, the FTC likely would not pursue 6(g) rulemaking should it decide to address data security and privacy, a topic which best fits under the “consumer protection” category. Rather, the FTC presumably would most likely initiate a “Magnuson-Moss” rulemaking (MMR) under Section 18 of the FTC Act, which authorizes the commission to prescribe “rules which define with specificity acts or practices which are unfair or deceptive acts or practices in or affecting commerce within the meaning of Section 5(a)(1) of the Act.” Among other things, Section 18 requires that the commission’s rulemaking proceedings provide an opportunity for informal hearings at which interested parties are accorded limited rights of cross-examination. Also, before commencing an MMR proceeding, the FTC must have reason to believe the practices addressed by the rulemaking are “prevalent.” 15 U.S.C. Sec. 57a(b)(3).

MMR proceedings, which are not governed under the Administrative Procedure Act (APA), do not present the same degree of legal problems as Section 6(g) rulemakings (see here). The question of legal authority to adopt a substantive rule is not raised; “rule of law” problems are far less serious (the DOJ is not a parallel enforcer of consumer-protection law); and APA issues of “arbitrariness” and “capriciousness” are not directly presented. Indeed, MMR proceedings include a variety of procedures aimed at promoting fairness (see here, for example). An MMR proceeding directed at data privacy predictably would be based on the claim that the failure to adhere to certain data-protection norms is an “unfair act or practice.”

Nevertheless, MMR rules would be subject to two substantial sources of legal risk.

The first of these arises out of federalism. Three states (California, Colorado, and Virginia) recently have enacted comprehensive data-privacy laws, and a large number of other state legislatures are considering data-privacy bills (see here). The proliferation of state data-privacy statutes would raise the risk of inconsistent and duplicative regulatory norms, potentially chilling business innovations addressed at data protection (a severe problem in the Internet Age, when business data-protection programs typically will have interstate effects).

An FTC MMR data-protection regulation that successfully “occupied the field” and preempted such state provisions could eliminate that source of costs. The Magnuson–Moss Warranty Act, however, does not contain an explicit preemption clause, leaving in serious doubt the ability of an FTC rule to displace state regulations (see here for a summary of the murky state of preemption law, including the skepticism of textualist Supreme Court justices toward implied “obstacle preemption”). In particular, the long history of state consumer-protection and antitrust laws that coexist with federal laws suggests that the case for FTC rule-based displacement of state data protection is a weak one. The upshot, then, of a Section 18 FTC data-protection rule enactment could be “the worst of all possible worlds,” with drawn-out litigation leading to competing federal and state norms that multiplied business costs.

The second source of risk arises out of the statutory definition of “unfair practices,” found in Section 5(n) of the FTC Act. Section 5(n) codifies the meaning of unfair practices, and thereby constrains the FTC’s application of rulemakings covering such practices. Section 5(n) states:

The Commission shall have no authority . . . to declare unlawful an act or practice on the grounds that such an act or practice is unfair unless the act or practice causes or is likely to cause substantial injury to consumers which is not reasonably avoidable by consumers themselves and not outweighed by countervailing benefits to consumers or to competition. In determining whether an act or practice is unfair, the Commission may consider established public policies as evidence to be considered with all other evidence. Such public policy considerations may not serve as a primary basis for such determination.

In effect, Section 5(n) implicitly subjects unfair practices to a well-defined cost-benefit framework. Thus, in promulgating a data-privacy MMR, the FTC first would have to demonstrate that specific disfavored data-protection practices caused or were likely to cause substantial harm. What’s more, the commission would have to show that any actual or likely harm would not be outweighed by countervailing benefits to consumers or competition. One would expect that a data-privacy rulemaking record would include submissions that pointed to the efficiencies of existing data-protection policies that would be displaced by a rule.

Moreover, subsequent federal court challenges to a final FTC rule likely would put forth the consumer and competitive benefits sacrificed by rule requirements. For example, rule challengers might point to the added business costs passed on to consumers that would arise from particular rule mandates, and the diminution in competition among data-protection systems generated by specific rule provisions. Litigation uncertainties surrounding these issues could be substantial and would cast into further doubt the legal viability of any final FTC data protection rule.

Apart from these legal risk-based costs, an MMR data privacy predictably would generate error-based costs. Given imperfect information in the hands of government and the impossibility of achieving welfare-maximizing nirvana through regulation (see, for example, here), any MMR data-privacy rule would erroneously condemn some economically inefficient business protocols and disincentivize some efficiency-seeking behavior. The Section 5(n) cost-benefit framework, though helpful, would not eliminate such error. (For example, even bureaucratic efforts to accommodate some business suggestions during the rulemaking process might tilt the post-rule market in favor of certain business models, thereby distorting competition.) In the abstract, it is difficult to say whether the welfare benefits of a final MMA data-privacy rule (measured by reductions in data-privacy-related consumer harm) would outweigh the costs, even before taking legal costs into account.

Conclusion

At least two FTC commissioners (and likely a third, assuming that President Joe Biden’s highly credentialed nominee Alvaro Bedoya will be confirmed by the U.S. Senate) appear to support FTC data-privacy regulation, even in the absence of new federal legislation. Such regulation, which presumably would be adopted as an MMR pursuant to Section 18 of the FTC Act, would probably not prove cost-beneficial. Not only would adoption of a final data-privacy rule generate substantial litigation costs and uncertainty, it would quite possibly add an additional layer of regulatory burdens above and beyond the requirements of proliferating state privacy rules. Furthermore, it is impossible to say whether the consumer-privacy benefits stemming from such an FTC rule would outweigh the error costs (manifested through competitive distortions and consumer harm) stemming from the inevitable imperfections of the rule’s requirements. All told, these considerations counsel against the allocation of scarce FTC resources to a Section 18 data-privacy rulemaking initiative.

But what about legislation? New federal privacy legislation that explicitly preempted state law would eliminate costs arising from inconsistencies among state privacy rules. Ideally, if such legislation were to be pursued, it should to the extent possible embody a cost-benefit framework designed to minimize the sum of administrative (including litigation) and error costs. The nature of such a possible law, and the role the FTC might play in administering it, however, is a topic for another day.

The European Commission this week published its proposed Artificial Intelligence Regulation, setting out new rules for  “artificial intelligence systems” used within the European Union. The regulation—the commission’s attempt to limit pernicious uses of AI without discouraging its adoption in beneficial cases—casts a wide net in defining AI to include essentially any software developed using machine learning. As a result, a host of software may fall under the regulation’s purview.

The regulation categorizes AIs by the kind and extent of risk they may pose to health, safety, and fundamental rights, with the overarching goal to:

  • Prohibit “unacceptable risk” AIs outright;
  • Place strict restrictions on “high-risk” AIs;
  • Place minor restrictions on “limited-risk” AIs;
  • Create voluntary “codes of conduct” for “minimal-risk” AIs;
  • Establish a regulatory sandbox regime for AI systems; 
  • Set up a European Artificial Intelligence Board to oversee regulatory implementation; and
  • Set fines for noncompliance at up to 30 million euros, or 6% of worldwide turnover, whichever is greater.

AIs That Are Prohibited Outright

The regulation prohibits AI that are used to exploit people’s vulnerabilities or that use subliminal techniques to distort behavior in a way likely to cause physical or psychological harm. Also prohibited are AIs used by public authorities to give people a trustworthiness score, if that score would then be used to treat a person unfavorably in a separate context or in a way that is disproportionate. The regulation also bans the use of “real-time” remote biometric identification (such as facial-recognition technology) in public spaces by law enforcement, with exceptions for specific and limited uses, such as searching for a missing child.

The first prohibition raises some interesting questions. The regulation says that an “exploited vulnerability” must relate to age or disability. In its announcement, the commission says this is targeted toward AIs such as toys that might induce a child to engage in dangerous behavior.

The ban on AIs using “subliminal techniques” is more opaque. The regulation doesn’t give a clear definition of what constitutes a “subliminal technique,” other than that it must be something “beyond a person’s consciousness.” Would this include TikTok’s algorithm, which imperceptibly adjusts the videos shown to the user to keep them engaged on the platform? The notion that this might cause harm is not fanciful, but it’s unclear whether the provision would be interpreted to be that expansive, whatever the commission’s intent might be. There is at least a risk that this provision would discourage innovative new uses of AI, causing businesses to err on the side of caution to avoid the huge penalties that breaking the rules would incur.

The prohibition on AIs used for social scoring is limited to public authorities. That leaves space for socially useful expansions of scoring systems, such as consumers using their Uber rating to show a record of previous good behavior to a potential Airbnb host. The ban is clearly oriented toward more expansive and dystopian uses of social credit systems, which some fear may be used to arbitrarily lock people out of society.

The ban on remote biometric identification AI is similarly limited to its use by law enforcement in public spaces. The limited exceptions (preventing an imminent terrorist attack, searching for a missing child, etc.) would be subject to judicial authorization except in cases of emergency, where ex-post authorization can be sought. The prohibition leaves room for private enterprises to innovate, but all non-prohibited uses of remote biometric identification would be subject to the requirements for high-risk AIs.

Restrictions on ‘High-Risk’ AIs

Some AI uses are not prohibited outright, but instead categorized as “high-risk” and subject to strict rules before they can be used or put to market. AI systems considered to be high-risk include those used for:

  • Safety components for certain types of products;
  • Remote biometric identification, except those uses that are banned outright;
  • Safety components in the management and operation of critical infrastructure, such as gas and electricity networks;
  • Dispatching emergency services;
  • Educational admissions and assessments;
  • Employment, workers management, and access to self-employment;
  • Evaluating credit-worthiness;
  • Assessing eligibility to receive social security benefits or services;
  • A range of law-enforcement purposes (e.g., detecting deepfakes or predicting the occurrence of criminal offenses);
  • Migration, asylum, and border-control management; and
  • Administration of justice.

While the commission considers these AIs to be those most likely to cause individual or social harm, it may not have appropriately balanced those perceived harms with the onerous regulatory burdens placed upon their use.

As Mikołaj Barczentewicz at the Surrey Law and Technology Hub has pointed out, the regulation would discourage even simple uses of logic or machine-learning systems in such settings as education or workplaces. This would mean that any workplace that develops machine-learning tools to enhance productivity—through, for example, monitoring or task allocation—would be subject to stringent requirements. These include requirements to have risk-management systems in place, to use only “high quality” datasets, and to allow human oversight of the AI, as well as other requirements around transparency and documentation.

The obligations would apply to any companies or government agencies that develop an AI (or for whom an AI is developed) with a view toward marketing it or putting it into service under their own name. The obligations could even attach to distributors, importers, users, or other third parties if they make a “substantial modification” to the high-risk AI, market it under their own name, or change its intended purpose—all of which could potentially discourage adaptive use.

Without going into unnecessary detail regarding each requirement, some are likely to have competition- and innovation-distorting effects that are worth discussing.

The rule that data used to train, validate, or test a high-risk AI has to be high quality (“relevant, representative, and free of errors”) assumes that perfect, error-free data sets exist, or can easily be detected. Not only is this not necessarily the case, but the requirement could impose an impossible standard on some activities. Given this high bar, high-risk AIs that use data of merely “good” quality could be precluded. It also would cut against the frontiers of research in artificial intelligence, where sometimes only small and lower-quality datasets are available to train AI. A predictable effect is that the rule would benefit large companies that are more likely to have access to large, high-quality datasets, while rules like the GDPR make it difficult for smaller companies to acquire that data.

High-risk AIs also must submit technical and user documentation that detail voluminous information about the AI system, including descriptions of the AI’s elements, its development, monitoring, functioning, and control. These must demonstrate the AI complies with all the requirements for high-risk AIs, in addition to documenting its characteristics, capabilities, and limitations. The requirement to produce vast amounts of information represents another potentially significant compliance cost that will be particularly felt by startups and other small and medium-sized enterprises (SMEs). This could further discourage AI adoption within the EU, as European enterprises already consider liability for potential damages and regulatory obstacles as impediments to AI adoption.

The requirement that the AI be subject to human oversight entails that the AI can be overseen and understood by a human being and that the AI can never override a human user. While it may be important that an AI used in, say, the criminal justice system must be understood by humans, this requirement could inhibit sophisticated uses beyond the reasoning of a human brain, such as how to safely operate a national electricity grid. Providers of high-risk AI systems also must establish a post-market monitoring system to evaluate continuous compliance with the regulation, representing another potentially significant ongoing cost for the use of high-risk AIs.

The regulation also places certain restrictions on “limited-risk” AIs, notably deepfakes and chatbots. Such AIs must be labeled to make a user aware they are looking at or listening to manipulated images, video, or audio. AIs must also be labeled to ensure humans are aware when they are speaking to an artificial intelligence, where this is not already obvious.

Taken together, these regulatory burdens may be greater than the benefits they generate, and could chill innovation and competition. The impact on smaller EU firms, which already are likely to struggle to compete with the American and Chinese tech giants, could prompt them to move outside the European jurisdiction altogether.

Regulatory Support for Innovation and Competition

To reduce the costs of these rules, the regulation also includes a new regulatory “sandbox” scheme. The sandboxes would putatively offer environments to develop and test AIs under the supervision of competent authorities, although exposure to liability would remain for harms caused to third parties and AIs would still have to comply with the requirements of the regulation.

SMEs and startups would have priority access to the regulatory sandboxes, although they must meet the same eligibility conditions as larger competitors. There would also be awareness-raising activities to help SMEs and startups to understand the rules; a “support channel” for SMEs within the national regulator; and adjusted fees for SMEs and startups to establish that their AIs conform with requirements.

These measures are intended to prevent the sort of chilling effect that was seen as a result of the GDPR, which led to a 17% increase in market concentration after it was introduced. But it’s unclear that they would accomplish this goal. (Notably, the GDPR contained similar provisions offering awareness-raising activities and derogations from specific duties for SMEs.) Firms operating in the “sandboxes” would still be exposed to liability, and the only significant difference to market conditions appears to be the “supervision” of competent authorities. It remains to be seen how this arrangement would sufficiently promote innovation as to overcome the burdens placed on AI by the significant new regulatory and compliance costs.

Governance and Enforcement

Each EU member state would be expected to appoint a “national competent authority” to implement and apply the regulation, as well as bodies to ensure high-risk systems conform with rules that require third party-assessments, such as remote biometric identification AIs.

The regulation establishes the European Artificial Intelligence Board to act as the union-wide regulatory body for AI. The board would be responsible for sharing best practices with member states, harmonizing practices among them, and issuing opinions on matters related to implementation.

As mentioned earlier, maximum penalties for marketing or using a prohibited AI (as well as for failing to use high-quality datasets) would be a steep 30 million euros or 6% of worldwide turnover, whichever is greater. Breaking other requirements for high-risk AIs carries maximum penalties of 20 million euros or 4% of worldwide turnover, while maximums of 10 million euros or 2% of worldwide turnover would be imposed for supplying incorrect, incomplete, or misleading information to the nationally appointed regulator.

Is the Commission Overplaying its Hand?

While the regulation only restricts AIs seen as creating risk to society, it defines that risk so broadly and vaguely that benign applications of AI may be included in its scope, intentionally or unintentionally. Moreover, the commission also proposes voluntary codes of conduct that would apply similar requirements to “minimal” risk AIs. These codes—optional for now—may signal the commission’s intent eventually to further broaden the regulation’s scope and application.

The commission clearly hopes it can rely on the “Brussels Effect” to steer the rest of the world toward tighter AI regulation, but it is also possible that other countries will seek to attract AI startups and investment by introducing less stringent regimes.

For the EU itself, more regulation must be balanced against the need to foster AI innovation. Without European tech giants of its own, the commission must be careful not to stifle the SMEs that form the backbone of the European market, particularly if global competitors are able to innovate more freely in the American or Chinese markets. If the commission has got the balance wrong, it may find that AI development simply goes elsewhere, with the EU fighting the battle for the future of AI with one hand tied behind its back.

In current discussions of technology markets, few words are heard more often than “platform.” Initial public offering (IPO) prospectuses use “platform” to describe a service that is bound to dominate a digital market. Antitrust regulators use “platform” to describe a service that dominates a digital market or threatens to do so. In either case, “platform” denotes power over price. For investors, that implies exceptional profits; for regulators, that implies competitive harm.

Conventional wisdom holds that platforms enjoy high market shares, protected by high barriers to entry, which yield high returns. This simple logic drives the market’s attribution of dramatically high valuations to dramatically unprofitable businesses and regulators’ eagerness to intervene in digital platform markets characterized by declining prices, increased convenience, and expanded variety, often at zero out-of-pocket cost. In both cases, “burning cash” today is understood as the path to market dominance and the ability to extract a premium from consumers in the future.

This logic is usually wrong. 

The Overlooked Basics of Platform Economics

To appreciate this perhaps surprising point, it is necessary to go back to the increasingly overlooked basics of platform economics. A platform can refer to any service that matches two complementary populations. A search engine matches advertisers with consumers, an online music service matches performers and labels with listeners, and a food-delivery service matches restaurants with home diners. A platform benefits everyone by facilitating transactions that otherwise might never have occurred.

A platform’s economic value derives from its ability to lower transaction costs by funneling a multitude of individual transactions into a single convenient hub.  In pursuit of minimum costs and maximum gains, users on one side of the platform will tend to favor the most popular platforms that offer the largest number of users on the other side of the platform. (There are partial exceptions to this rule when users value being matched with certain typesof other users, rather than just with more users.) These “network effects” mean that any successful platform market will always converge toward a handful of winners. This positive feedback effect drives investors’ exuberance and regulators’ concerns.

There is a critical point, however, that often seems to be overlooked.

Market share only translates into market power to the extent the incumbent is protected against entry within some reasonable time horizon.  If Warren Buffett’s moat requirement is not met, market share is immaterial. If XYZ.com owns 100% of the online pet food delivery market but entry costs are asymptotic, then market power is negligible. There is another important limiting principle. In platform markets, the depth of the moat depends not only on competitors’ costs to enter the market, but users’ costs in switching from one platform to another or alternating between multiple platforms. If users can easily hop across platforms, then market share cannot confer market power given the continuous threat of user defection. Put differently: churn limits power over price.

Contrary to natural intuitions, this is why a platform market consisting of only a few leaders can still be intensely competitive, keeping prices low (down to and including $0) even if the number of competitors is low. It is often asserted, however, that users are typically locked into the dominant platform and therefore face high switching costs, which therefore implicitly satisfies the moat requirement. If that is true, then the “high churn” scenario is a theoretical curiosity and a leading platform’s high market share would be a reliable signal of market power. In fact, this common assumption likely describes the atypical case. 

AWS and the Cloud Data-Storage Market

This point can be illustrated by considering the cloud data-storage market. This would appear to be an easy case where high switching costs (due to the difficulty in shifting data among storage providers) insulate the market leader against entry threats. Yet the real world does not conform to these expectations. 

While Amazon Web Services pioneered the $100 billion-plus market and is still the clear market leader, it now faces vigorous competition from Microsoft Azure, Google Cloud, and other data-storage or other cloud-related services. This may reflect the fact that the data storage market is far from saturated, so new users are up for grabs and existing customers can mitigate lock-in by diversifying across multiple storage providers. Or it may reflect the fact that the market’s structure is fluid as a function of technological changes, enabling entry at formerly bundled portions of the cloud data-services package. While it is not always technologically feasible, the cloud storage market suggests that users’ resistance to platform capture can represent a competitive opportunity for entrants to challenge dominant vendors on price, quality, and innovation parameters.

The Surprising Instability of Platform Dominance

The instability of leadership positions in the cloud storage market is not exceptional. 

Consider a handful of once-powerful platforms that were rapidly dethroned once challenged by a more efficient or innovative rival: Yahoo and Alta Vista in the search-engine market (displaced by Google); Netscape in the browser market (displaced by Microsoft’s Internet Explorer, then displaced by Google Chrome); Nokia and then BlackBerry in the mobile wireless-device market (displaced by Apple and Samsung); and Friendster in the social-networking market (displaced by Myspace, then displaced by Facebook). AOL was once thought to be indomitable; now it is mostly referenced as a vintage email address. The list could go on.

Overestimating platform dominance—or more precisely, assuming platform dominance without close factual inquiry—matters because it promotes overestimates of market power. That, in turn, cultivates both market and regulatory bubbles: investors inflate stock valuations while regulators inflate the risk of competitive harm. 

DoorDash and the Food-Delivery Services Market

Consider the DoorDash IPO that launched in early December 2020. The market’s current approximately $50 billion valuation of a business that has been almost consistently unprofitable implicitly assumes that DoorDash will maintain and expand its position as the largest U.S. food-delivery platform, which will then yield power over price and exceptional returns for investors. 

There are reasons to be skeptical. Even where DoorDash captures and holds a dominant market share in certain metropolitan areas, it still faces actual and potential competition from other food-delivery services, in-house delivery services (especially by well-resourced national chains), and grocery and other delivery services already offered by regional and national providers. There is already evidence of these expected responses to DoorDash’s perceived high delivery fees, a classic illustration of the disciplinary effect of competitive forces on the pricing choices of an apparently dominant market leader. These “supply-side” constraints imposed by competitors are compounded by “demand-side” constraints imposed by customers. Home diners incur no more than minimal costs when swiping across food-delivery icons on a smartphone interface, casting doubt that high market share is likely to translate in this context into market power.

Deliveroo and the Costs of Regulatory Autopilot

Just as the stock market can suffer from delusions of platform grandeur, so too some competition regulators appear to have fallen prey to the same malady. 

A vivid illustration is provided by the 2019 decision by the Competition Markets Authority (CMA), the British competition regulator, to challenge Amazon’s purchase of a 16% stake in Deliveroo, one of three major competitors in the British food-delivery services market. This intervention provides perhaps the clearest illustration of policy action based on a reflexive assumption of market power, even in the face of little to no indication that the predicate conditions for that assumption could plausibly be satisfied.

Far from being a dominant platform, Deliveroo was (and is) a money-losing venture lagging behind money-losing Just Eat (now Just Eat Takeaway) and Uber Eats in the U.K. food-delivery services market. Even Amazon had previously closed its own food-delivery service in the U.K. due to lack of profitability. Despite Deliveroo’s distressed economic circumstances and the implausibility of any market power arising from Amazon’s investment, the CMA nonetheless elected to pursue the fullest level of investigation. While the transaction was ultimately approved in August 2020, this intervention imposed a 15-month delay and associated costs in connection with an investment that almost certainly bolstered competition in a concentrated market by funding a firm reportedly at risk of insolvency.  This is the equivalent of a competition regulator driving in reverse.

Concluding Thoughts

There seems to be an increasingly common assumption in commentary by the press, policymakers, and even some scholars that apparently dominant platforms usually face little competition and can set, at will, the terms of exchange. For investors, this is a reason to buy; for regulators, this is a reason to intervene. This assumption is sometimes realized, and, in that case, antitrust intervention is appropriate whenever there is reasonable evidence that market power is being secured through something other than “competition on the merits.” However, several conditions must be met to support the market power assumption without which any such inquiry would be imprudent. Contrary to conventional wisdom, the economics and history of platform markets suggest that those conditions are infrequently satisfied.

Without closer scrutiny, reflexively equating market share with market power is prone to lead both investors and regulators astray.  

In a constructive development, the Federal Trade Commission has joined its British counterpart in investigating Nvidia’s proposed $40 billion acquisition of chip designer Arm, a subsidiary of Softbank. Arm provides the technological blueprints for wireless communications devices and, subject to a royalty fee, makes those crown-jewel assets available to all interested firms. Notwithstanding Nvidia’s stated commitment to keep the existing policy in place, there is an obvious risk that the new parent, one of the world’s leading chip makers, would at some time modify this policy with adverse competitive effects.

Ironically, the FTC is likely part of the reason that the Nvidia-Arm transaction is taking place.

Since the mid-2000s, the FTC and other leading competition regulators (except for the U.S. Department of Justice’s Antitrust Division under the leadership of former Assistant Attorney General Makan Delrahim) have intervened extensively in licensing arrangements in wireless device markets, culminating in the FTC’s recent failed suit against Qualcomm. The Nvidia-Arm transaction suggests that these actions may simply lead chip designers to abandon the licensing model and shift toward structures that monetize chip-design R&D through integrated hardware and software ecosystems. Amazon and Apple are already undertaking chip innovation through this model. Antitrust action that accelerates this movement toward in-house chip design is likely to have adverse effects for the competitive health of the wireless ecosystem.

How IP Licensing Promotes Market Access

Since its inception, the wireless communications market has relied on a handful of IP licensors to supply device producers and other intermediate users with a common suite of technology inputs. The result has been an efficient division of labor between firms that specialize in upstream innovation and firms that specialize in production and other downstream functions. Contrary to the standard assumption that IP rights limit access, this licensing-based model ensures technology access to any firm willing to pay the royalty fee.

Efforts by regulators to reengineer existing relationships between innovators and implementers endanger this market structure by inducing innovators to abandon licensing-based business models, which now operate under a cloud of legal insecurity, for integrated business models in which returns on R&D investments are captured internally through hardware and software products. Rather than expanding technology access and intensifying competition, antitrust restraints on licensing freedom are liable to limit technology access and increase market concentration.

Regulatory Intervention and Market Distortion

This interventionist approach has relied on the assertion that innovators can “lock in” producers and extract a disproportionate fee in exchange for access. This prediction has never found support in fact. Contrary to theoretical arguments that patent owners can impose double-digit “royalty stacks” on device producers, empirical researchers have repeatedly found that the estimated range of aggregate rates lies in the single digits. These findings are unsurprising given market performance over more than two decades: adoption has accelerated as quality-adjusted prices have fallen and innovation has never ceased. If rates were exorbitant, market growth would have been slow, and the smartphone would be a luxury for the rich.

Despite these empirical infirmities, the FTC and other competition regulators have persisted in taking action to mitigate “holdup risk” through policy statements and enforcement actions designed to preclude IP licensors from seeking injunctive relief. The result is a one-sided legal environment in which the world’s largest device producers can effectively infringe patents at will, knowing that the worst-case scenario is a “reasonable royalty” award determined by a court, plus attorneys’ fees. Without any credible threat to deny access even after a favorable adjudication on the merits, any IP licensor’s ability to negotiate a royalty rate that reflects the value of its technology contribution is constrained.

Assuming no change in IP licensing policy on the horizon, it is therefore not surprising that an IP licensor would seek to shift toward an integrated business model in which IP is not licensed but embedded within an integrated suite of products and services. Or alternatively, an IP licensor entity might seek to be acquired by a firm that already has such a model in place. Hence, FTC v. Qualcomm leads Arm to Nvidia.

The Error Costs of Non-Evidence-Based Antitrust

These counterproductive effects of antitrust intervention demonstrate the error costs that arise when regulators act based on unverified assertions of impending market failure. Relying on the somewhat improbable assumption that chip suppliers can dictate licensing terms to device producers that are among the world’s largest companies, competition regulators have placed at risk the legal predicates of IP rights and enforceable contracts that have made the wireless-device market an economic success. As antitrust risk intensifies, the return on licensing strategies falls and competitive advantage shifts toward integrated firms that can monetize R&D internally through stand-alone product and service ecosystems.

Far from increasing competitiveness, regulators’ current approach toward IP licensing in wireless markets is likely to reduce it.

[TOTM: The following is part of a digital symposium by TOTM guests and authors on the legal and regulatory issues that arose during Ajit Pai’s tenure as chairman of the Federal Communications Commission. The entire series of posts is available here.

Jerry Ellig was a research professor at The George Washington University Regulatory Studies Center and served as chief economist at the Federal Communications Commission from 2017 to 2018. Tragically, he passed away Jan. 20, 2021. TOTM is honored to publish his contribution to this symposium.]

One significant aspect of Chairman Ajit Pai’s legacy is not a policy change, but an organizational one: establishment of the Federal Communications Commission’s (FCC’s) Office of Economics and Analytics (OEA) in 2018.

Prior to OEA, most of the FCC’s economists were assigned to the various policy bureaus, such as Wireless, Wireline Competition, Public Safety, Media, and International. Each of these bureaus had its own chief economist, but the rank-and-file economists reported to the managers who ran the bureaus – usually attorneys who also developed policy and wrote regulations. In the words of former FCC Chief Economist Thomas Hazlett, the FCC had “no location anywhere in the organizational structure devoted primarily to economic analysis.”

Establishment of OEA involved four significant changes. First, most of the FCC’s economists (along with data strategists and auction specialists) are now grouped together into an organization separate from the policy bureaus, and they are managed by other economists. Second, the FCC rules establishing the new office tasked OEA with reviewing every rulemaking, reviewing every other item with economic content that comes before the commission for a vote, and preparing a full benefit-cost analysis for any regulation with $100 million or more in annual economic impact. Third, a joint memo from the FCC’s Office of General Counsel and OEA specifies that economists are to be involved in the early stages of all rulemakings. Fourth, the memo also indicates that FCC regulatory analysis should follow the principles articulated in Executive Order 12866 and Office of Management and Budget Circular A-4 (while specifying that the FCC, as an independent agency, is not bound by the executive order).

While this structure for managing economists was new for the FCC, it is hardly uncommon in federal regulatory agencies. Numerous independent agencies that deal with economic regulation house their economists in a separate bureau or office, including the Securities and Exchange Commission, the Commodity Futures Trading Commission, the Surface Transportation Board, the Office of Comptroller of the Currency, and the Federal Trade Commission. The SEC displays even more parallels with the FCC. A guidance memo adopted in 2012 by the SEC’s Office of General Counsel and Division of Risk, Strategy and Financial Innovation (the name of the division where economists and other analysts were located) specifies that economists are to be involved in the early stages of all rulemakings and articulates best analytical practices based on Executive Order 12866 and Circular A-4.

A separate economics office offers several advantages over the FCC’s prior approach. It gives the economists greater freedom to offer frank advice, enables them to conduct higher-quality analysis more consistent with the norms of their profession, and may ultimately make it easier to uphold FCC rules that are challenged in court.

Independence.  When I served as chief economist at the FCC in 2017-2018, I gathered from conversations that the most common practice in the past was for attorneys who wrote rules to turn to economists for supporting analysis after key decisions had already been made. This was not always the process, but it often occurred. The internal working group of senior FCC career staff who drafted the plan for OEA reached similar conclusions. After the establishment of OEA, an FCC economist I interviewed noted how his role had changed: “My job used to be to support the policy decisions made in the chairman’s office. Now I’m much freer to speak my own mind.”

Ensuring economists’ independence is not a problem unique to the FCC. In a 2017 study, Stuart Shapiro found that most of the high-level economists he interviewed who worked on regulatory impact analyses in federal agencies perceive that economists can be more objective if they are located outside the program office that develops the regulations they are analyzing. As one put it, “It’s very difficult to conduct a BCA [benefit-cost analysis] if our boss wrote what you are analyzing.” Interviews with senior economists and non-economists who work on regulation that I conducted for an Administrative Conference of the United States project in 2019 revealed similar conclusions across federal agencies. Economists located in organizations separate from the program office said that structure gave them greater independence and ability to develop better analytical methodologies. On the other hand, economists located in program offices said they experienced or knew of instances where they were pressured or told to produce an analysis with the results decision-makers wanted.

The FTC provides an informative case study. From 1955-1961, many of the FTC’s economists reported to the attorneys who conducted antitrust cases; in 1961, they were moved into a separate Bureau of Economics. Fritz Mueller, the FTC chief economist responsible for moving the antitrust economists back into the Bureau of Economics, noted that they were originally placed under the antitrust attorneys because the attorneys wanted more control over the economic analysis. A 2015 evaluation by the FTC’s Inspector General concluded that the Bureau of Economics’ existence as a separate organization improves its ability to offer “unbiased and sound economic analysis to support decision-making.”

Higher-quality analysis. An issue closely related to economists’ independence is the quality of the economic analysis. Executive branch regulatory economists interviewed by Richard Williams expressed concern that the economic analysis was more likely to be changed to support decisions when the economists are located in the program office that writes the regulations. More generally, a study that Catherine Konieczny and I conducted while we were at the FCC found that executive branch agencies are more likely to produce higher-quality regulatory impact analyses if the economists responsible for the analysis are in an independent economics office rather than the program office.

Upholding regulations in court. In Michigan v. EPA, the Supreme Court held that it is unreasonable for agencies to refuse to consider regulatory costs if the authorizing statute does not prohibit them from doing so. This precedent will likely increase judicial expectations that agencies will consider economic issues when they issue regulations. The FCC’s OGC-OEA memo cites examples of cases where the quality of the FCC’s economic analysis either helped or harmed the commission’s ability to survive legal challenge under the Administrative Procedure Act’s “arbitrary and capricious” standard. More systematically, a recent Regulatory Studies Center working paper finds that a higher-quality economic analysis accompanying a regulation reduces the likelihood that courts will strike down the regulation, provided that the agency explains how it used the analysis in decisions.

Two potential disadvantages of a separate economics office are that it may make the economists easier to ignore (what former FCC Chief Economist Tim Brennan calls the “Siberia effect”) and may lead the economists to produce research that is less relevant to the practical policy concerns of the policymaking bureaus. The FCC’s reorganization plan took these disadvantages seriously.

To ensure that the ultimate decision-makers—the commissioners—have access to the economists’ analysis and recommendations, the rules establishing the office give OEA explicit responsibility for reviewing all items with economic content that come before the commission. Each item is accompanied by a cover memo that indicates whether OEA believes there are any significant issues, and whether they have been dealt with adequately. To ensure that economists and policy bureaus work together from the outset of regulatory initiatives, the OGC-OEA memo instructs:

Bureaus and Offices should, to the extent practicable, coordinate with OEA in the early stages of all Commission-level and major Bureau-level proceedings that are likely to draw scrutiny due to their economic impact. Such coordination will help promote productive communication and avoid delays from the need to incorporate additional analysis or other content late in the drafting process. In the earliest stages of the rulemaking process, economists and related staff will work with programmatic staff to help frame key questions, which may include drafting options memos with the lead Bureau or Office.

While presiding over his final commission meeting on Jan. 13, Pai commented, “It’s second nature now for all of us to ask, ‘What do the economists think?’” The real test of this institutional innovation will be whether that practice continues under a new chair in the next administration.

[TOTM: The following is part of a digital symposium by TOTM guests and authors on the legal and regulatory issues that arose during Ajit Pai’s tenure as chairman of the Federal Communications Commission. The entire series of posts is available here.

Mark Jamison is the Gerald L. Gunter Memorial Professor and director of the Public Utility Research Center at the University of Florida’s Warrington College of Business. He’s also a visiting scholar at the American Enterprise Institute.]

Chairman Ajit Pai will be remembered as one of the most consequential Federal Communications Commission chairmen in history. His policy accomplishments are numerous, including the repeal of Title II regulation of the internet, rural broadband development, increased spectrum for 5G, decreasing waste in universal service funding, and better controlling robocalls.

Less will be said about the important work he has done rebuilding the FCC’s independence. It is rare for a new FCC chairman to devote resources to building the institution. Most focus on their policy agendas, because policies and regulations make up their legacies that the media notices, and because time and resources are limited. Chairman Pai did what few have even attempted to do: both build the organization and make significant regulatory reforms.

Independence is the ability of a regulatory institution to operate at arm’s length from the special interests of industry, politicians, and the like. The pressures to bias actions to benefit favored stakeholders can be tremendous; the FCC greatly influences who gets how much of the billions of dollars that are at stake in FCC decisions. But resisting those pressures is critical because investment and services suffer when a weak FCC is directed by political winds or industry pressures rather than law and hard analysis.

Chairman Pai inherited a politicized FCC. Research by Scott Wallsten showed that commission votes had been unusually partisan under the previous chairman (November 2013 through January 2017). From the beginning of Reed Hundt’s term as chairman until November 2013, only 4% of commission votes had divided along party lines. By contrast, 26% of votes divided along party lines from November 2013 until Chairman Pai took over. This division was also reflected in a sharp decline in unanimous votes under the previous administration. Only 47% of FCC votes on orders were unanimous, as opposed to an average of 60% from Hundt through the brief term of Mignon Clyburn.

Chairman Pai and his fellow commissioners worked to heal this divide. According to the FCC’s data, under Chairman Pai, over 80% of items on the monthly meeting agenda had bipartisan support and over 70% were adopted without dissent. This was hard, as Democrats in general were deeply against President Donald Trump and some members of Congress found a divided FCC convenient.

The political orientation of the FCC prior to Chairman Pai was made clear in the management of controversial issues. The agency’s work on net neutrality in 2015 pivoted strongly toward heavy regulation when President Barack Obama released his video supporting Title II regulation of the internet. And there is evidence that the net-neutrality decision was made in the White House, not at the FCC. Agency economists were cut out of internal discussions once the political decision had been made to side with the president, causing the FCC’s chief economist to quip that the decision was an economics-free zone.

On other issues, a vote on Lifeline was delayed several hours so that people on Capitol Hill could lobby a Democratic commissioner to align with fellow Democrats and against the Republican commissioners. And an initiative to regulate set-top boxes was buoyed, not by analyses by FCC staff, but by faulty data and analyses from Democratic senators.

Chairman Pai recognized the danger of politically driven decision-making and noted that it was enabled in part by the agency’s lack of a champion for economic analyses. To remedy this situation, Chairman Pai proposed forming an Office of Economics and Analytics (OEA). The commission adopted his proposal, but unfortunately it was with one of the rare party-line votes. Hopefully, Democratic commissioners have learned the value of the OEA.

The OEA has several responsibilities, but those most closely aligned with supporting the agency’s independence are that it: (a) provides economic analysis, including cost-benefit analysis, for commission actions; (b) develops policies and strategies on data resources and best practices for data use; and (c) conducts long-term research. The work of the OEA makes it hard for a politically driven chairman to pretend that his or her initiatives are somehow substantive.

Another institutional weakness at the FCC was a lack of transparency. Prior to Chairman Pai, the public was not allowed to view the text of commission decisions until after they were adopted. Even worse, sometimes the text that the commissioners saw when voting was not the text in the final decision. Wallsten described in his research a situation where the meaning of a vote actually changed from the time of the vote to the release of the text:

On February 9, 2011 the Federal Communications Commission (FCC) released a proposed rule that included, among many other provisions, capping the Universal Service Fund at $4.5 billion. The FCC voted to approve a final order on October 27, 2011. But when the order was finally released on November 18, 2011, the $4.5 billion ceiling had effectively become a floor, with the order requiring the agency to forever estimate demand at no less than $4.5 billion. Because payments from the fund had been decreasing steadily, this floor means that the FCC is now collecting hundreds of billions of dollars more in taxes than it is spending on the program. [footnotes omitted]

The lack of transparency led many to not trust the FCC and encouraged stakeholders with inside access to bypass the legitimate public process for lobbying the agency. This would have encouraged corruption had not Chairman Pai changed the system. He required that decision texts be released to the public at the same time they were released to commissioners. This allows the public to see what the commissioners are voting on. And it ensures that orders do not change after they are voted on.

The FCC demonstrated its independence under Chairman Pai. In the case of net neutrality, the three Republican commissioners withstood personal threats, mocking from congressional Democrats, and pressure from Big Tech to restore light-handed regulation. About a year later, Chairman Pai was strongly criticized by President Trump for rejecting the Sinclair-Tribune merger. And despite the president’s support of the merger, he apparently had sufficient respect for the FCC’s independence that the White House never contacted the FCC about the issue. In the case of Ligado Networks’ use of its radio spectrum license, the FCC stood up to intense pressure from the U.S. Department of Defense and from members of Congress who wanted to substitute their technical judgement for the FCC’s research on the impacts of Ligado’s proposal.

It is possible that a new FCC could undo this new independence. Commissioners could marginalize their economists, take their directions from partisans, and reintroduce the practice of hiding information from the public. But Chairman Pai foresaw this and carefully made his changes part of the institutional structure of the FCC, making any steps backward visible to all concerned.

Congressman Buck’s “Third Way” report offers a compromise between the House Judiciary Committee’s majority report, which proposes sweeping new regulation of tech companies, and the status quo, which Buck argues is unfair and insufficient. But though Buck rejects many of the majority’s reports proposals, what he proposes instead would lead to virtually the same outcome via a slightly longer process. 

The most significant majority proposals that Buck rejects are the structural separation to prevent a company that runs a platform from operating on that platform “in competition with the firms dependent on its infrastructure”, and line-of-business restrictions that would confine tech companies to a small number of markets, to prevent them from preferencing their other products to the detriment of competitors.

Buck rules these out, saying that they are “regulatory in nature [and] invite unforeseen consequences and divert attention away from public interest antitrust enforcement by our antitrust agencies.” He goes on to say that “this proposal is a thinly veiled call to break up Big Tech firms.”

Instead, Buck endorses, either fully or provisionally, measures including revitalising the essential facilities doctrine, imposing data interoperability mandates on platforms, and changing antitrust law to prevent “monopoly leveraging and predatory pricing”. 

Put together, though, these would amount to the same thing that the Democratic majority report proposes: a world where platforms are basically just conduits, regulated to be neutral and open, and where the companies that run them require a regulator’s go-ahead for important decisions — a process that would be just as influenced lobbying and political considerations, and insulated from market price signals, as any other regulator’s decisions are.

Revitalizing the essential facilities doctrine

Buck describes proposals to “revitalize the essential facilities doctrine” as “common ground” that warrant further consideration. This would mean that platforms deemed to be “essential facilities” would be required to offer access to their platform to third parties at a “reasonable” price, except in exceptional circumstances. The presumption would be that these platforms were anticompetitively foreclosing third party developers and merchants by either denying them access to their platforms or by charging them “too high” prices. 

This would require the kind of regulatory oversight that Buck says he wants to avoid. He says that “conservatives should be wary of handing additional regulatory authority to agencies in an attempt to micromanage platforms’ access rules.” But there’s no way to avoid this when the “facility” — and hence its pricing and access rules — changes as frequently as any digital platform does. In practice, digital platforms would have to justify their pricing rules and decisions about exclusion of third parties to courts or a regulator as often as they make those decisions.

If Apple’s App Store were deemed an essential facility such that it is presumed to be foreclosing third party developers any time it rejected their submissions, it would have to submit to regulatory scrutiny of the “reasonableness” of its commercial decisions on, literally, a daily basis.

That would likely require price controls to prevent platforms from using pricing to de facto exclude third parties they did not want to deal with. Adjudication of “fair” pricing by courts is unlikely to be a sustainable solution. Justice Breyer, in Town of Concord v. Boston Edison Co., considered this to be outside the courts’ purview:

[H]ow is a judge or jury to determine a ‘fair price?’ Is it the price charged by other suppliers of the primary product? None exist. Is it the price that competition ‘would have set’ were the primary level not monopolized? How can the court determine this price without examining costs and demands, indeed without acting like a rate-setting regulatory agency, the rate-setting proceedings of which often last for several years? Further, how is the court to decide the proper size of the price ‘gap?’ Must it be large enough for all independent competing firms to make a ‘living profit,’ no matter how inefficient they may be? . . . And how should the court respond when costs or demands change over time, as they inevitably will?

In practice, infrastructure treated as an essential facility is usually subject to pricing control by a regulator. This has its own difficulties. The UK’s energy and water infrastructure is an example. In determining optimal access pricing, regulators must determine the price that weighs competing needs to maximise short-term output, incentivise investment by the infrastructure owner, incentivise innovation and entry by competitors (e.g., local energy grids) and, of course, avoid “excessive” pricing. 

This is a near-impossible task, and the process is often drawn out and subject to challenges even in markets where the infrastructure is relatively simple. It is even less likely that these considerations would be objectively tractable in digital markets.

Treating a service as an essential facility is based on the premise that, absent mandated access, it is impossible to compete with it. But mandating access does not, on its own, prevent it from extracting monopoly rents from consumers; it just means that other companies selling inputs can have their share of the rents. 

So you may end up with two different sets of price controls: on the consumer side, to determine how much monopoly rent can be extracted from consumers, and on the access side, to determine how the monopoly rents are divided.

The UK’s energy market has both, for example. In the case of something like an electricity network, where it may simply not be physically or economically feasible to construct a second, competing network, this might be the least-bad course of action. In such circumstances, consumer-side price regulation might make sense. 

But if a service could, in fact, be competed with by others, treating it as an essential facility may be affirmatively harmful to competition and consumers if it diverts investment and time away from that potential competitor by allowing other companies to acquire some of the incumbent’s rents themselves.

The HJC report assumes that Apple is a monopolist, because, among people who own iPhones, the App Store is the only way to install third-party software. Treating the App Store as an essential facility may mean a ban on Apple charging “excessive prices” to companies like Spotify or Epic that would like to use it, or on Apple blocking them for offering users alternative in-app ways of buying their services.

If it were impossible for users to switch from iPhones, or for app developers to earn revenue through other mechanisms, this logic might be sound. But it would still not change the fact that the App Store platform was able to charge users monopoly prices; it would just mean that Epic and Spotify could capture some of those monopoly rents for themselves. Nice for them, but not for consumers. And since both companies have already grown to be pretty big and profitable with the constraints they object to in place, it seems difficult to argue that they cannot compete with these in place and sounds more like they’d just like a bigger share of the pie.

And, in fact, it is possible to switch away from the iPhone to Android. I have personally switched back and forth several times over the past few years, for example. And so have many others — despite what some claim, it’s really not that hard, especially now that most important data is stored on cloud-based services, and both companies offer an app to switch from the other. Apple also does not act like a monopolist — its Bionic chips are vastly better than any competitor’s and it continues to invest in and develop them.

So in practice, users switching from iPhone to Android if Epic’s games and Spotify’s music are not available constrains Apple, to some extent. If Apple did drive those services permanently off their platform, it would make Android relatively more attractive, and some users would move away — Apple would bear some of the costs of its ecosystem becoming worse. 

Assuming away this kind of competition, as Buck and the majority report do, is implausible. Not only that, but Buck and the majority believe that competition in this market is impossible — no policy or antitrust action could change things, and all that’s left is to regulate the market like it’s an electricity grid. 

And it means that platforms could often face situations where they could not expect to make themselves profitable after building their markets, since they could not control the supply side in order to earn revenues. That would make it harder to build platforms, and weaken competition, especially competition faced by incumbents.

Mandating interoperability

Interoperability mandates, which Buck supports, require platforms to make their products open and interoperable with third party software. If Twitter were required to be interoperable, for example, it would have to provide a mechanism (probably a set of open APIs) by which third party software could tweet and read its feeds, upload photos, send and receive DMs, and so on. 

Obviously, what interoperability actually involves differs from service to service, and involves decisions about design that are specific to each service. These variations are relevant because they mean interoperability requires discretionary regulation, including about product design, and can’t just be covered by a simple piece of legislation or a court order. 

To give an example: interoperability means a heightened security risk, perhaps from people unwittingly authorising a bad actor to access their private messages. How much is it appropriate to warn users about this, and how tight should your security controls be? It is probably excessive to require that users provide a sworn affidavit with witnesses, and even some written warnings about the risks may be so over the top as to scare off virtually any interested user. But some level of warning and user authentication is appropriate. So how much? 

Similarly, a company that has been required to offer its customers’ data through an API, but doesn’t really want to, can make life miserable for third party services that want to use it. Changing the API without warning, or letting its service drop or slow down, can break other services, and few users will be likely to want to use a third-party service that is unreliable. But some outages are inevitable, and some changes to the API and service are desirable. How do you decide how much?

These are not abstract examples. Open Banking in the UK, which requires interoperability of personal and small business current accounts, is the most developed example of interoperability in the world. It has been cited by former Chair of the Council of Economic Advisors, Jason Furman, among others, as a model for interoperability in tech. It has faced all of these questions: one bank, for instance, required that customers pass through twelve warning screens to approve a third party app to access their banking details.

To address problems like this, Open Banking has needed an “implementation entity” to design many of its most important elements. This is a de facto regulator, and it has taken years of difficult design decisions to arrive at Open Banking’s current form. 

Having helped write the UK’s industry review into Open Banking, I am cautiously optimistic about what it might be able to do for banking in Britain, not least because that market is already heavily regulated and lacking in competition. But it has been a huge undertaking, and has related to a relatively narrow set of data (its core is just two different things — the ability to read an account’s balance and transaction history, and the ability to initiate payments) in a sector that is not known for rapidly changing technology. Here, the costs of regulation may be outweighed by the benefits.

I am deeply sceptical that the same would be the case in most digital markets, where products do change rapidly, where new entrants frequently attempt to enter the market (and often succeed), where the security trade-offs are even more difficult to adjudicate, and where the economics are less straightforward, given that many services are provided at least in part because of the access to customer data they provide. 

Even if I am wrong, it is unavoidable that interoperability in digital markets would require an equivalent body to make and implement decisions when trade-offs are involved. This, again, would require a regulator like the UK’s implementation entity, and one that was enormous, given the number and diversity of services that it would have to oversee. And it would likely have to make important and difficult design decisions to which there is no clear answer. 

Banning self-preferencing

Buck’s Third Way would also ban digital platforms from self-preferencing. This typically involves an incumbent that can provide a good more cheaply than its third-party competitors — whether it’s through use of data that those third parties do not have access to, reputational advantages that mean customers will be more likely to use their products, or through scale efficiencies that allow it to provide goods to a larger customer base for a cheaper price. 

Although many people criticise self-preferencing as being unfair on competitors, “self-preferencing” is an inherent part of almost every business. When a company employs its own in-house accountants, cleaners or lawyers, instead of contracting out for them, it is engaged in internal self-preferencing. Any firm that is vertically integrated to any extent, instead of contracting externally for every single ancillary service other than the one it sells in the market, is self-preferencing. Coase’s theory of the firm is all about why this kind of behaviour happens, instead of every worker contracting on the open market for everything they do. His answer is that transaction costs make it cheaper to bring certain business relationships in-house than to contract externally for them. Virtually everyone agrees that this is desirable to some extent.

Nor does it somehow become a problem when the self-preferencing takes place on the consumer product side. Any firm that offers any bundle of products — like a smartphone that can run only the manufacturer’s operating system — is engaged in self-preferencing, because users cannot construct their own bundle with that company’s hardware and another’s operating system. But the efficiency benefits often outweigh the lack of choice.

Self-preferencing in digital platforms occurs, for example, when Google includes relevant Shopping or Maps results at the top of its general Search results, or when Amazon gives its own store-brand products (like the AmazonBasics range) a prominent place in the results listing.

There are good reasons to think that both of these are good for competition and consumer welfare. Google making Shopping results easily visible makes it a stronger competitor to Amazon, and including Maps results when you search for a restaurant just makes it more convenient to get the information you’re looking for.

Amazon sells its own private label products partially because doing so is profitable (even when undercutting rivals), partially to fill holes in product lines (like clothing, where 11% of listings were Amazon private label as of November 2018), and partially because it increases users’ likelihood to use Amazon if they expect to find a reliable product from a brand they trust. According to Amazon, they account for less than 1% of its annual retail sales, in contrast to the 19% of revenues ($54 billion) Amazon makes from third party seller services, which includes Marketplace commissions. Any analysis that ignores that Amazon has to balance those sources of revenue, and so has to tread carefully, is deficient. 

With “commodity” products (like, say, batteries and USB cables), where multiple sellers are offering very similar or identical versions of the same thing, private label competition works well for both Amazon and consumers. By Amazon’s own rules it can enter this market using aggregated data, but this doesn’t give it a significant advantage, because that data is easily obtainable from multiple sources, including Amazon itself, which makes detailed aggregated sales data freely available to third-party retailers

Amazon does profit from sales of these products, of course. And other merchants suffer by having to cut their prices to compete. That’s precisely what competition involves — competition is incompatible with a quiet life for businesses. But consumers benefit, and the biggest benefit to Amazon is that it assures its potential customers that when they visit they will be able to find a product that is cheap and reliable, so they keep coming back.

It is even hard to argue that in aggregate this practice is damaging to third-party sellers: many, like Anker, have built successful businesses on Amazon despite private-label competition precisely because the value of the platform increases for all parties as user trust and confidence in it does.

In these cases and in others, platforms act to solve market failures on the markets they host, as Andrei Hagiu has argued. To maximize profits, digital platforms need to strike a balance between being an attractive place for third-party merchants to sell their goods and being attractive to consumers by offering low prices. The latter will frequently clash with the former — and that’s the difficulty of managing a platform. 

To mistake this pro-competitive behaviour with an absence of competition is misguided. But that is a key conclusion of Buck’s Third Way: that the damage to competitors makes this behaviour harmful overall, and that it should be curtailed with “non-discrimination” rules. 

Treating below-cost selling as “predatory pricing”

Buck’s report equates below-cost selling with predatory pricing (“predatory pricing, also known as below-cost selling”). This is mistaken. Predatory pricing refers to a particular scenario where your price cut is temporary and designed to drive a competitor out of business, so that you can raise prices later and recoup your losses. 

It is easy to see that this does not describe the vast majority of below-cost selling. Buck’s formulation would describe all of the following as “predatory pricing”:

  • A restaurants that gives away ketchup for free;
  • An online retailer that offers free shipping and returns;
  • A grocery store that sells tins of beans for 3p a can. (This really happened when I was a child.)

The rationale for offering below-cost prices differs in each of these cases. Sometimes it’s a marketing ploy — Tesco sells those beans to get some free media, and to entice people into their stores, hoping they’ll decide to do the rest of their weekly shop there at the same time. Sometimes it’s about reducing frictions — the marginal cost of ketchup is so low that it’s simpler to just give it away. Sometimes it’s about reducing the fixed costs of transactions so more take place — allowing customers who buy your products to return them easily may mean more are willing to buy them overall, because there’s less risk for them if they don’t like what they buy. 

Obviously, none of these is “predatory”: none is done in the expectation that the below-cost selling will drive those businesses’ competitors out of business, allowing them to make monopoly profits later.

True predatory pricing is theoretically possible, but very difficult. As David Henderson describes, to successfully engage in predatory pricing means taking enormous and rising losses that grow for the “predatory” firm as customers switch to it from its competitor. And once the rival firm has exited the market, if the predatory firm raises prices above average cost (i.e., to recoup its losses), there is no guarantee that a new competitor will not enter the market selling at the previously competitive price. And the competing firm can either shut down temporarily or, in some cases, just buy up the “predatory” firm’s discounted goods to resell later. It is debatable whether the canonical predatory pricing case, Standard Oil, is itself even an example of that behaviour.

Offering a product below cost in a multi-sided market (like a digital platform) can be a way of building a customer base in order to incentivise entry on the other side of the market. When network effects exist, so additional users make the service more valuable to existing users, it can be worthwhile to subsidise the initial users until the service reaches a certain size. 

Uber subsidising drivers and riders in a new city is an example of this — riders want enough drivers on the road that they know they’ll be picked up fairly quickly if they order one, and drivers want enough riders that they know they’ll be able to earn a decent night’s fares if they use the app. This requires a certain volume of users on both sides — to get there, it can be in everyone’s interest for the platform to subsidise one or both sides of the market to reach that critical mass.

The slightly longer road to regulation

That is another reason for below-cost pricing: someone other than the user may be part-paying for a product, to build a market they hope to profit from later. Platforms must adjust pricing and their offerings to each side of their market to manage supply and demand. Epic, for example, is trying to build a desktop computer game store to rival the largest incumbent, Steam. To win over customers, it has been giving away games for free to users, who can own them on that store forever. 

That is clearly pro-competitive — Epic is hoping to get users over the habit of using Steam for all their games, in the hope that they will recoup the costs of doing so later in increased sales. And it is good for consumers to get free stuff. This kind of behaviour is very common. As well as Uber and Epic, smaller platforms do it too. 

Buck’s proposals would make this kind of behaviour much more difficult, and permitted only if a regulator or court allows it, instead of if the market can bear it. On both sides of the coin, Buck’s proposals would prevent platforms from the behaviour that allows them to grow in the first place — enticing suppliers and consumers and subsidising either side until critical mass has been reached that allows the platform to exist by itself, and the platform owner to recoup its investments. Fundamentally, both Buck and the majority take the existence of platforms as a given, ignoring the incentives to create new ones and compete with incumbents. 

In doing so, they give up on competition altogether. As described, Buck’s provisions would necessitate ongoing rule-making, including price controls, to work. It is unlikely that a court could do this, since the relevant costs would change too often for one-shot rule-making of the kind a court could do. To be effective at all, Buck’s proposals would require an extensive, active regulator, just as the majority report’s would. 

Buck nominally argues against this sort of outcome — “Conservatives should be wary of handing additional regulatory authority to agencies in an attempt to micromanage platforms’ access rules” — but it is probably unavoidable, given the changes he proposes. And because the rule changes he proposes would apply to the whole economy, not just tech, his proposals may, perversely, end up being even more extensive and interventionist than the majority’s.

Other than this, the differences in practice between Buck’s proposals and the Democrats’ proposals would be trivial. At best, Buck’s Third Way is just a longer route to the same destination.

[TOTM: The following is part of a blog series by TOTM guests and authors on the law, economics, and policy of the ongoing COVID-19 pandemic. The entire series of posts is available here.

This post is authored by Jacob Grier, (Freelance writer and spirits consultant in Portland, Oregon, and the author of The Rediscovery of Tobacco: Smoking, Vaping, and the Creative Destruction of the Cigarette).]

The COVID-19 pandemic and the shutdown of many public-facing businesses has resulted in many sudden shifts in demand for common goods. The demand for hand sanitizer has drastically increased for hospitals, businesses, and individuals. At the same time, demand for distilled spirits has fallen substantially, as the closure of bars, restaurants, and tasting rooms has cut craft distillers off from their primary buyers. Since ethanol is a key ingredient in both spirits and sanitizer, this situation presents an obvious opportunity for distillers to shift their production from the former to the latter. Hundreds of distilleries have made this transition, but it has not without obstacles. Some of these reflect a real scarcity of needed supplies, but other constraints have been externally imposed by government regulations and the tax code.

Producing sanitizer

The World Health Organization provides guidelines and recipes for locally producing hand sanitizer. The relevant formulation for distilleries calls for only four ingredients: high-proof ethanol (96%), hydrogen peroxide (3%), glycerol (98%), and sterile distilled or boiled water. Distilleries are well-positioned to produce or obtain ethanol and water. Glycerol is used in only small amounts and does not currently appear to be a substantial constraint on production. Hydrogen peroxide is harder to come by, but distilleries are adapting and cooperating to ensure supply. Skip Tognetti, owner of Letterpress Distilling in Seattle, Washington, reports that one local distiller obtained a drum of 34% hydrogen peroxide, which stretches a long way when diluted to a concentration of 3%. Local distillers have been sharing this drum so that they can all produce sanitizer.

Another constraint is finding containers in which to the put the finished product. Not all containers are suitable for holding high-proof alcoholic solutions, and supplies of those that are recommended for sanitizer are scarce. The fact that many of these bottles are produced in China has reportedly also limited the supply. Distillers are therefore having to get creative; Tognetti reports looking into shampoo bottles, and in Chicago distillers have re-purposed glass beer growlers. For informal channels, some distillers have allowed consumers to bring their own containers to fill with sanitizer for personal use. Food and Drug Administration labeling requirements have also prevented the use of travel-size bottles, since the bottles are too small to display the necessary information.

The raw materials for producing ethanol are also coming from some unexpected sources. Breweries are typically unable to produce alcohol at high enough proof for sanitizer, but multiple breweries in Chicago are donating beer that distilleries can bring up to the required purity. Beer giant Anheuser-Busch is also producing sanitizer with the ethanol removed from its alcohol-free beers.

In many cases, the sanitizer is donated or sold at low-cost to hospitals and other essential services, or to local consumers. Online donations have helped to fund some of these efforts, and at least one food and beverage testing lab has stepped up to offer free testing to breweries and distilleries producing sanitizer to ensure compliance with WHO guidelines. Distillers report that the regulatory landscape has been somewhat confusing in recent weeks, and posts in a Facebook group have provided advice for how to get through the FDA’s registration process. In general, distillers going through the process report that agencies have been responsive. Tom Burkleaux of New Deal Distilling in Portland, Oregon says he “had to do some mighty paperwork,” but that the FDA and the Oregon Board of Pharmacy were both quick to process applications, with responses coming in just a few hours or less.

In general, the redirection of craft distilleries to producing hand sanitizer is an example of private businesses responding to market signals and the evident challenges of the health crisis to produce much-needed goods; in some cases, sanitizer represents one of their only sources of revenue during the shutdown, providing a lifeline for small businesses. The Distilled Spirits Council currently lists nearly 600 distilleries making sanitizer in the United States.

There is one significant obstacle that has hindered the production of sanitizer, however: an FDA requirement that distilleries obtain extra ingredients to denature their alcohol.

Denaturing sanitizer

According to the WHO, the four ingredients mentioned above are all that are needed to make sanitizer. In fact, WHO specifically notes that it in most circumstances it is inadvisable to add anything else: “it is not recommended to add any bittering agents to reduce the risk of ingestion of the handrubs” except in cases where there is a high probably of accidental ingestion. Further, “[…] there is no published information on the compatibility and deterrent potential of such chemicals when used in alcohol-based handrubs to discourage their abuse. It is important to note that such additives may make the products toxic and add to production costs.”

Denaturing agents are used to render alcohol either too bitter or too toxic to consume, deterring abuse by adults or accidental ingestion by children. In ordinary circumstances, there are valid reasons to denature sanitizer. In the current pandemic, however, the denaturing requirement is a significant bottleneck in production.

The federal Tax and Trade Bureau is the primary agency regulating alcohol production in the United States. The TTB took action early to encourage distilleries to produce sanitizer, officially releasing guidance on March 18 instructing them that they are free to commence production without prior authorization or formula approval, so long as they are making sanitizer in accordance with WHO guidelines. On March 23, the FDA issued its own emergency authorization of hand sanitizer production; unlike the WHO, FDA guidance does require the use of denaturants. As a result, on March 26 the TTB issued new guidance to be consistent with the FDA.

Under current rules, only sanitizer made with denatured alcohol is exempt from the federal excise tax on beverage alcohol. Federal excise taxes begin at $2.70 per gallon for low-volume distilleries and reach up to $13.50 per gallon, significantly increasing the cost of producing hand sanitizer; state excise taxes can raise these costs even higher.

More importantly, denaturing agents are scarce. In a Twitter thread on March 25, Tognetti noted the difficulty of obtaining them:

To be clear, if I didn’t have to track down denaturing agents (there are several, but isopropyl alcohol is the most common), I could turn out 200 gallons of finished hand sanitizer TODAY.

(As an additional concern, the Distilled Spirits Council notes that the extremely bitter or toxic nature of denaturing agents may impose additional costs on distillers given the need to thoroughly cleanse them from their equipment.)

Congress attempted to address these concerns in the CARES Act, the coronavirus relief package. Section 2308 explicitly waives the federal excise tax on distilled spirits used for the production of sanitizer, however it leaves the formula specification in the hands of the FDA. Unless the agency revises its guidance, production in the US will be constrained by the requirement to add denaturing agents to the plentiful supply of ethanol, or distilleries will risk being targeted with enforcement actions if they produce perfectly usable sanitizer without denaturing their alcohol.

Local distilleries provide agile production capacity

In recent days, larger spirits producers including Pernod-Ricard, Diageo, and Bacardi have announced plans to produce sanitizer. Given their resources and economies of scale, they may end up taking over a significant part of the market. Yet small, local distilleries have displayed the agility necessary to rapidly shift production. It’s worth noting that many of these distilleries did not exist until fairly recently. According to the American Craft Spirits Association, there were fewer than 100 craft distilleries operating in the United States in 2005. By 2018, there were more than 1,800. This growth is the result of changing consumer interests, but also the liberalization of state and local laws to permit distilleries and tasting rooms. That many of these distilleries have the capacity to produce sanitizer in a time of emergency is a welcome, if unintended, consequence of this liberalization.

[TOTM: The following is part of a blog series by TOTM guests and authors on the law, economics, and policy of the ongoing COVID-19 pandemic. The entire series of posts is available here.

This post is authored by Julian Morris, (Director of Innovation Policy, ICLE).]

SARS-CoV2, the virus that causes COVID-19, is now widespread in the population in many countries, including the US, UK, Australia, Iran, and many European countries. Its prevalence in other regions, such as South Asia, much of South America, and Africa, is relatively unknown. The failure to contain the virus early on has meant that more aggressive measures are now necessary in order to avoid overwhelming healthcare systems, which would cause unacceptable levels of mortality. (Sadly, Italy’s health system has already been overwhelmed, forcing medical practitioners to engage in the most awful triage decisions.) Many jurisdictions, ranging from cities to entire countries, have chosen to implement mandatory lockdowns. These will likely have the desired effect of slowing transmission in the short term, but they cannot be maintained indefinitely. The challenge going forward is how to contain the spread of the virus without destroying the economy. 

In this post I will outline the elements of a proposal that I hope might do that. (I’ve been working on this for about a week and in the meantime some of the ideas have been advanced by others. E.g. this and this. Great minds clearly think alike.)

1. Identify those who have had COVID-19 and have recovered — and allow them to go back to work

While there are some reports of people who have had COVID-19 becoming reinfected, this seems to be very rare (a recent primate study implies reinfection is impossible) and the alleged cases may have been a result of false negative tests followed by relapse by patients. The general presumption is that having the disease is likely to confer immunity for several months at least. Moreover, people with immunity who no longer show symptoms of the disease are very unlikely to transmit the disease. Allowing those people to go back to work will lessen the burden of the lockdown without appreciably increasing the risk of infection

One group of such people is readily identifiable, though small: Those who tested positive for COVID-19 and subsequently recovered. Those people should be permitted to go back to work immediately.

2. Where possible, test, trace, treat, isolate

The town of Vo in Northern Italy, the site of the first death in the country from COVID-19, appears to have stopped the disease from spreading in about three weeks. It did so through a combination of universal testing, two weeks of strict lockdown, and quarantine of cases.  Could this be replicated elsewhere? 

Vo has a population of 3,300, so universal testing was not the gargantuan exercise it would be in, say, the continental US. Some larger jurisdictions have had similar success without resorting to universal testing and lockdown. South Korea managed to contain the spread of SARS-CoV2 relatively quickly through a combination of: social distancing (including closing schools and restricting large gatherings), testing anyone who had COVID-19 symptoms (and increasingly those without symptoms), tracing and testing of those who had contact with those symptomatic individuals, treating those with severe symptoms, quarantining those who tested positive but had no or only mild symptoms (the quarantine was monitored using a phone app and strictly enforced), and publicly sharing detailed information about the known incidence of the virus. 

A study of 181 cases in China published in the Annals of Internal Medicine found that the mean incubation period for COVID-19 is just over 5 days and only about 1 in 100 cases take longer than 14 days. By implication, if people have been strictly following the guidelines on avoiding contact with others, washing/sanitizing hands, sanitizing other objects, and avoiding hand-to-face contact, it should be possible, after two weeks of lockdown, to identify the vast majority of people who are not infected by testing everyone for the presence of SARS-CoV2 itself.

But that’s a series of big ifs. Since it takes a few days for the virus to replicate in the body to the point at which it is detectable, people who have recently been infected might test negative. Also, it is unlikely to be feasible logistically to test a significant proportion of the population for SARS-CoV2 in a short period of time. Existing tests require the use of RT-PCR, which is expensive and time consuming, not least because it can only be done at a lab, and while the capacity for such tests is increasing, it is likely around 50,000 per day in the entire US. 

Test, trace, treat, and isolate may be a feasible option for towns and even cities that currently have relatively low incidence of SARS-CoV2. However, given the lethargic progress of testing in places such as the US, UK and India, and hence poor existing knowledge of the extent of infection, it will not be a universal panacea.

3. Test as many people as possible for the presence of antibodies to SARS-CoV2

Outside those few places that have dramatically ramped up testing, it is likely that many more people have had COVID-19 than have been tested, either because they were asymptomatic or because they did not require clinical attention. Many, perhaps most of those people will no longer have the virus in their system but they should still have antibodies (indicating immunity). In order to identify those people, there should be widespread testing for antibodies to SARS-CoV2. 

Antibody tests are inexpensive, quick, and some can be done at home with minimal assistance. Numerous such tests have already been produced or are in development (see the list here). For example, Chinese manufacturer Innovita has produced a test that appears to be effective; in a clinical trial of 447 patients, it identified the presence of antibodies to SARS-CoV2 in 87.3 % of clinically confirmed cases of COVID-19 (i.e. there were approximately 13% false negatives) but zero false positives. Innovita’s test was approved by China’s equivalent of the FDA and has been used widely there. 

Scanwell Health, a San Francisco-based startup, has an exclusive license to produce Innovita’s test in the U.S. and has already begun the process for obtaining approval from the US FDA under its Emergency Use Authorization. Scanwell estimates that the total cost of the test, including overnight shipping of the kit and support from a doctor or nurse practitioner from Lemonaid Health, will be around $70. One downside to Scanwell Health’s offering, however, is that it expects it to take 6-8 weeks to begin shipping testing kits once it receives authorization from the FDA

So far, the FDA has approved at least one SARS-CoV2 antibody test, produced by Aytu Bioscience in Colorado. But Aytu’s test is designed for use by physicians, not at home. In Europe, at least one antibody test, produced by German company PharmactACT, is already available. (That test has similar characteristics to Innovita’s.) Another has been approved by the MHRA in the UK for physician use and is awaiting approval for home use; the UK government has ordered 3.5 million of these tests, with the aim of distributing 250,000 per day by the end of April. 

Unfortunately, some people who have antibodies to SARS-CoV2 will also still be infectious. However, because different antibodies develop at different times during the course of infection, it may be possible to distinguish those who are still infectious from those who are no longer infectious. Specifically, immunoglobulin (Ig) M is present in larger amounts while the viral load is still present, while IgG is present in larger amounts later on (see e.g. this and the figure below). So, by testing for the presence of both IgM and IgG it should be possible to identify a large proportion of those who have had COVID-19 but are no longer infectious. (The currently available antibody tests result in about 13 percent false negatives, making them inappropriate as a means of screening out those who do not have COVID-19. But they produce zero false positives, making them ideal for identifying those who definitely have or have had COVID-19). In essence, people whose IgG test is positive but IgM test is negative can then go back to work. In addition, people who have had COVID-19 symptoms, are now symptom-free, and test positive for antibodies, should be allowed to go back to work.

4. Test for SARS-Cov2 among those who test negative for antibodies — and ensure that everyone who tests positive remains in isolation

Those people who test negative for SARS-CoV2 using the quick antibody immunoassay, as well as those who are positive for both IgG and IgM (indicating that they may still be infectious) should then be tested for SARS-CoV2 using the RT-PCR test described above. And those who test negative for SARS-CoV2 should then be permitted to go back to work. But those who test positive should be required to remain in isolation— and seek treatment if necessary.

5. Repeat steps 3 and 4 until nobody tests positive for COVID-19

By repeating steps 3 and 4, it should be possible gradually to enable the vast majority of the population to return to work, and thence to a life of greater normalcy, within a matter of weeks.

6. Some (possibly rather large) caveats

All of this relies on: (a) the ability rapidly to expand testing and (b) widespread compliance with isolation requirements. Neither of these conditions is by any means guaranteed, not least because the rules effectively discriminate in favor of people who have had COVID-19, which may create a perverse incentive to violate not only the isolation requirements but all the recommended hygiene practices — and thereby intentionally become infected with SARS-CoV2 on the presumption that they will then be able to go back to work sooner than otherwise. So, before this is rolled out, it is important to ensure that there will be widespread testing for COVID-19 in a timeframe shorter than the likely total time for contracting and recovering from COVID-19.

In addition, if test results are to be used as a means of establishing a person’s ability to travel and work while others are still under lockdown, it is important that there  be a means of verifying the status of individuals. That might be possible through the use of an app, for example; such an app might also provide policymakers to make better resources allocation decisions too. 

Also, at-risk individuals should be strongly advised to remain in isolation until there is no further evidence of community transmission. 

7. The Mechanics of Testing

Given that there are not currently sufficient tests available for everyone to be tested in most locations, one obvious question is: who should be tested? As noted above, it makes sense initially to target those who have had COVID-19 symptoms and have recovered. Since only those people who have had such symptoms—and possibly their physician if they presented with their symptoms—will know who they are, this will rely largely on trust. (It’s possible that self-reporting apps could help.) 

But it may make sense initially to target tests more narrowly. The UK is initially targeting the antibody detection kits to healthcare and other key workers—people who are essential to the continued functioning of the country. That makes sense and could easily be applied in other places. 

Assuming that key workers can be supplied with antibody detection kits quickly, distribution should then be opened up more widely. No doubt insurance companies will be making decisions about the purchase of testing kits. Ideally, however, individuals should be able to buy kits such as Scanwell’s without going through a bureaucratic process, whether that be their insurance company or the NHS. And vendors should be free to price kits as they see fit, without worrying about the prospect of being subject to price caps such as those imposed by Medicaid or the VA, which have the perverse effect of incentivising vendors to increase the list price. Finally, in order to increase the supply of tests as rapidly as possible, regulatory agencies should be encouraged to issue emergency approvals as quickly as possible. Having more manufacturers with a diverse array of tests available will increase access to testing more quickly and likely lead to more accurate testing too. Agencies such as the FDA should see this as their absolute priority right now. If the Mayo clinic can compress 6 months’ product development into a month, the FDA can surely do its review far more quickly too. Lives—and the economy—depend upon it.

[TOTM: The following is part of a blog series by TOTM guests and authors on the law, economics, and policy of the ongoing COVID-19 pandemic. The entire series of posts is available here.

This post is authored by Brent Skorup, (Senior Research Fellow, Mercatus Center, George Mason University).]

One of the most visible economic effects of the COVID-19 spread is the decrease in airline customers. Alec Stapp alerted me to the recent outrage over “ghost flights,” where airlines fly nearly empty planes to maintain their “slots.” 

The airline industry is unfortunately in economic freefall as governments prohibit and travelers pull back on air travel. When the health and industry crises pass, lawmakers will have an opportunity to evaluate the mistakes of the past when it comes to airport congestion and airspace design.

This issue of ghost flights pops up occasionally and offers a lesson in the problems with government rationing of public resources. In this case, the public resource are airport slots: designated times, say, 15 or 30 minutes, a plane may takeoff or land at an airport. (Last week US and EU regulators temporarily waived the use-it-or-lose it rule for slots to mitigate the embarrassing cost and environmental damage caused by forcing airlines to fly empty planes.)

The slots at major hubs at peak times of day are extremely scarce–there’s only so many hours in a day. Today, slot assignment are administratively rationed in a way that favors large, incumbent airlines. As the Wall Street Journal summarized last year,

For decades, airlines have largely divided runway access between themselves at twice-yearly meetings run by the IATA (an airline trade group).

Airport slots are property. They’re valuable. They can be defined, partitioned, leased, put up as collateral, and, in the US, they can be sold and transferred within or between airports.

You just can’t call slots property. Many lawmakers, regulators, and airline representatives refuse to acknowledge the obvious. Stating that slots are valuable public property would make clear the anticompetitive waste that the 40-year slot assignment experiment generates. 

Like many government programs, the slot rationing began in the US as a temporary program decades ago as a response to congestion at New York airports. Slots are currently used to ration access at LGA, JFK, and DCA. And while they don’t use formal slot rationing, the FAA also rations access at four other busy airports: ORD, Newark, LAX, and SFO.

Fortunately, cracks are starting to form. In 2008, at the tailend of the Bush administration, the FAA proposed to auction some slots in New York City’s three airports. The plan was delayed by litigation from incumbent airlines and an adverse finding from the GAO. With a change in administration, the Obama FAA rescinded the plan in 2009.

Before the Obama FAA recission, the mask slipped a bit in the GAO’s criticism of the slot auction plan: 

FAA’s argument that slots are property proves too much—it suggests that the agency has been improperly giving away potentially millions of dollars of federal property, for no compensation, since it created the slot system in 1968.

Gulp.

Though the GAO helped scuttle the plan, the damage has been done. The idea has now entered public policy discourse: giving away valuable public property is precisely what’s going on. 

The implicit was made explicit in 2011 when, despite spiking the Bush FAA plan, the Obama FAA auctioned two dozen high-value slots. (The reversal and lack of controversy is puzzling to me.) Delta and US Airways wanted to swap some 160 slots at New York and DC airports. As a condition of the mega-swap, the Obama FAA required they divest 24 slots at those popular airports, which the agency auctioned to new entrants. Seven low-fare airlines bid in the auction and Jetblue and WestJet won the divested slots, paying about $90 million combined

The older fictions are rapidly eroding. There is an active secondary market in slots in some nations and when prices are released it becomes clear that the legacy rationing amounts to public property setasides to insiders. In 2016 it leaked, for instance, that an airline paid £58 million for a pair of take-off and landing slots at Heathrow. Other slot sales are in the tens of millions of dollars.

The 2011 FAA auctions and the loosening of rules globally around slot sales signal that the competition benefits from slot markets are too obvious to ignore. Competition from new entry drives down airfare and increases the number of flights.

For instance, a few months ago researchers used a booking app to scour 50 trillion flight itineraries to see new entrants’ effect on airline ticket prices between 2017 and 2019. As the Wall Street Journal reported, the entry of a low-fare carrier reduced ticket prices by 17% on average. The bigger effect was on output–new entry led to a 30% YoY increase in flights.

It’s becoming harder to justify the legacy view, which allow incumbent airlines to dominate the slot allocations via international conferences and national regulations that require “grandfather” slot usage. In a separate article last year, the Wall Street Journal reported that airlines are reluctantly ceding more power to airports in the assignment of slots. This is another signal in the long-running tug-of-war between airports and airlines. Airports generally want to open slots for new competitors–incumbent airlines do not.

The reason for the change of heart? The Journal says,

Airlines and airports reached the deal in part because of concerns governments should start to sell slots.

Gulp. Ghost flights are a government failure but a rational response to governments withholding the benefits of property from airlines. The slot rationing system encourages flying uneconomical flights, smaller planes, and excess carbon emissions. The COVID-19 crisis allowed the public a glimpse at the dysfunctional system. It won’t be easy, but aviation regulators worldwide need to assess slots policy and airspace access before the administrative rationing system spreads to the emerging urban air mobility and drone delivery markets.

The following is the first in a new blog series by TOTM guests and authors on the law, economics, and policy of the ongoing COVID-19 pandemic. The entire series of posts is available at https://truthonthemarket.com/symposia/the-law-economics-of-the-covid-19-pandemic/.

Continue Reading...

Yesterday was President Trump’s big “Social Media Summit” where he got together with a number of right-wing firebrands to decry the power of Big Tech to censor conservatives online. According to the Wall Street Journal

Mr. Trump attacked social-media companies he says are trying to silence individuals and groups with right-leaning views, without presenting specific evidence. He said he was directing his administration to “explore all legislative and regulatory solutions to protect free speech and the free speech of all Americans.”

“Big Tech must not censor the voices of the American people,” Mr. Trump told a crowd of more than 100 allies who cheered him on. “This new technology is so important and it has to be used fairly.”

Despite the simplistic narrative tying President Trump’s vision of the world to conservatism, there is nothing conservative about his views on the First Amendment and how it applies to social media companies.

I have noted in several places before that there is a conflict of visions when it comes to whether the First Amendment protects a negative or positive conception of free speech. For those unfamiliar with the distinction: it comes from philosopher Isaiah Berlin, who identified negative liberty as freedom from external interference, and positive liberty as freedom to do something, including having the power and resources necessary to do that thing. Discussions of the First Amendment’s protection of free speech often elide over this distinction.

With respect to speech, the negative conception of liberty recognizes that individual property owners can control what is said on their property, for example. To force property owners to allow speakers/speech on their property that they don’t desire would actually be a violation of their liberty — what the Supreme Court calls “compelled speech.” The First Amendment, consistent with this view, generally protects speech from government interference (with very few, narrow exceptions), while allowing private regulation of speech (again, with very few, narrow exceptions).

Contrary to the original meaning of the First Amendment and the weight of Supreme Court precedent, President Trump’s view of the First Amendment is that it protects a positive conception of liberty — one under which the government, in order to facilitate its conception of “free speech,” has the right and even the duty to impose restrictions on how private actors regulate speech on their property (in this case, social media companies). 

But if Trump’s view were adopted, discretion as to what is necessary to facilitate free speech would be left to future presidents and congresses, undermining the bedrock conservative principle of the Constitution as a shield against government regulation, all falsely in the name of protecting speech. This is counter to the general approach of modern conservatism (but not, of course, necessarily Republicanism) in the United States, including that of many of President Trump’s own judicial and agency appointees. Indeed, it is actually more consistent with the views of modern progressives — especially within the FCC.

For instance, the current conservative bloc on the Supreme Court (over the dissent of the four liberal Justices) recently reaffirmed the view that the First Amendment applies only to state action in Manhattan Community Access Corp. v. Halleck. The opinion, written by Trump-appointee, Justice Brett Kavanaugh, states plainly that:

Ratified in 1791, the First Amendment provides in relevant part that “Congress shall make no law . . . abridging the freedom of speech.” Ratified in 1868, the Fourteenth Amendment makes the First Amendment’s Free Speech Clause applicable against the States: “No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law . . . .” §1. The text and original meaning of those Amendments, as well as this Court’s longstanding precedents, establish that the Free Speech Clause prohibits only governmental abridgment of speech. The Free Speech Clause does not prohibit private abridgment of speech… In accord with the text and structure of the Constitution, this Court’s state-action doctrine distinguishes the government from individuals and private entities. By enforcing that constitutional boundary between the governmental and the private, the state-action doctrine protects a robust sphere of individual liberty. (Emphasis added).

Former Stanford Law dean and First Amendment scholar, Kathleen Sullivan, has summed up the very different approaches to free speech pursued by conservatives and progressives (insofar as they are represented by the “conservative” and “liberal” blocs on the Supreme Court): 

In the first vision…, free speech rights serve an overarching interest in political equality. Free speech as equality embraces first an antidiscrimination principle: in upholding the speech rights of anarchists, syndicalists, communists, civil rights marchers, Maoist flag burners, and other marginal, dissident, or unorthodox speakers, the Court protects members of ideological minorities who are likely to be the target of the majority’s animus or selective indifference…. By invalidating conditions on speakers’ use of public land, facilities, and funds, a long line of speech cases in the free-speech-as-equality tradition ensures public subvention of speech expressing “the poorly financed causes of little people.” On the equality-based view of free speech, it follows that the well-financed causes of big people (or big corporations) do not merit special judicial protection from political regulation. And because, in this view, the value of equality is prior to the value of speech, politically disadvantaged speech prevails over regulation but regulation promoting political equality prevails over speech.

The second vision of free speech, by contrast, sees free speech as serving the interest of political liberty. On this view…, the First Amendment is a negative check on government tyranny, and treats with skepticism all government efforts at speech suppression that might skew the private ordering of ideas. And on this view, members of the public are trusted to make their own individual evaluations of speech, and government is forbidden to intervene for paternalistic or redistributive reasons. Government intervention might be warranted to correct certain allocative inefficiencies in the way that speech transactions take place, but otherwise, ideas are best left to a freely competitive ideological market.

The outcome of Citizens United is best explained as representing a triumph of the libertarian over the egalitarian vision of free speech. Justice Kennedy’s opinion for the Court, joined by Chief Justice Roberts and Justices Scalia, Thomas, and Alito, articulates a robust vision of free speech as serving political liberty; the dissenting opinion by Justice Stevens, joined by Justices Ginsburg, Breyer, and Sotomayor, sets forth in depth the countervailing egalitarian view. (Emphasis added).

President Trump’s views on the regulation of private speech are alarmingly consistent with those embraced by the Court’s progressives to “protect[] members of ideological minorities who are likely to be the target of the majority’s animus or selective indifference” — exactly the sort of conservative “victimhood” that Trump and his online supporters have somehow concocted to describe themselves. 

Trump’s views are also consistent with those of progressives who, since the Reagan FCC abolished it in 1987, have consistently angled for a resurrection of some form of fairness doctrine, as well as other policies inconsistent with the “free-speech-as-liberty” view. Thus Democratic commissioner Jessica Rosenworcel takes a far more interventionist approach to private speech:

The First Amendment does more than protect the interests of corporations. As courts have long recognized, it is a force to support individual interest in self-expression and the right of the public to receive information and ideas. As Justice Black so eloquently put it, “the widest possible dissemination of information from diverse and antagonistic sources is essential to the welfare of the public.” Our leased access rules provide opportunity for civic participation. They enhance the marketplace of ideas by increasing the number of speakers and the variety of viewpoints. They help preserve the possibility of a diverse, pluralistic medium—just as Congress called for the Cable Communications Policy Act… The proper inquiry then, is not simply whether corporations providing channel capacity have First Amendment rights, but whether this law abridges expression that the First Amendment was meant to protect. Here, our leased access rules are not content-based and their purpose and effect is to promote free speech. Moreover, they accomplish this in a narrowly-tailored way that does not substantially burden more speech than is necessary to further important interests. In other words, they are not at odds with the First Amendment, but instead help effectuate its purpose for all of us. (Emphasis added).

Consistent with the progressive approach, this leaves discretion in the hands of “experts” (like Rosenworcel) to determine what needs to be done in order to protect the underlying value of free speech in the First Amendment through government regulation, even if it means compelling speech upon private actors. 

Trump’s view of what the First Amendment’s free speech protections entail when it comes to social media companies is inconsistent with the conception of the Constitution-as-guarantor-of-negative-liberty that conservatives have long embraced. 

Of course, this is not merely a “conservative” position; it is fundamental to the longstanding bipartisan approach to free speech generally and to the regulation of online platforms specifically. As a diverse group of 75 scholars and civil society groups (including ICLE) wrote yesterday in their “Principles for Lawmakers on Liability for User-Generated Content Online”:

Principle #2: Any new intermediary liability law must not target constitutionally protected speech.

The government shouldn’t require—or coerce—intermediaries to remove constitutionally protected speech that the government cannot prohibit directly. Such demands violate the First Amendment. Also, imposing broad liability for user speech incentivizes services to err on the side of taking down speech, resulting in overbroad censorship—or even avoid offering speech forums altogether.

As those principles suggest, the sort of platform regulation that Trump, et al. advocate — essentially a “fairness doctrine” for the Internet — is the opposite of free speech:

Principle #4: Section 230 does not, and should not, require “neutrality.”

Publishing third-party content online never can be “neutral.” Indeed, every publication decision will necessarily prioritize some content at the expense of other content. Even an “objective” approach, such as presenting content in reverse chronological order, isn’t neutral because it prioritizes recency over other values. By protecting the prioritization, de-prioritization, and removal of content, Section 230 provides Internet services with the legal certainty they need to do the socially beneficial work of minimizing harmful content.

The idea that social media should be subject to a nondiscrimination requirement — for which President Trump and others like Senator Josh Hawley have been arguing lately — is flatly contrary to Section 230 — as well as to the First Amendment.

Conservatives upset about “social media discrimination” need to think hard about whether they really want to adopt this sort of position out of convenience, when the tradition with which they align rejects it — rightly — in nearly all other venues. Even if you believe that Facebook, Google, and Twitter are trying to make it harder for conservative voices to be heard (despite all evidence to the contrary), it is imprudent to reject constitutional first principles for a temporary policy victory. In fact, there’s nothing at all “conservative” about an abdication of the traditional principle linking freedom to property for the sake of political expediency.