The Federal Communications Commission (FCC) will soon vote on whether to repeal an Obama-era rule classifying Internet Service Providers (ISPs) as “common carriers.” That rule was put in place to achieve net neutrality, an attractive-sounding goal that many Americans—millennials especially—reflexively support.

Read the full piece here.

This week the FCC will vote on Chairman Ajit Pai’s Restoring Internet Freedom Order. Once implemented, the Order will rescind the 2015 Open Internet Order and return antitrust and consumer protection enforcement to primacy in Internet access regulation in the U.S.

Read the full piece here.

In response to the Federal Communication Commission’s (FCC) proposal to rescind its so-called “net neutrality” rules, U.S. Sen. Cory Booker tweeted that the FCC is “giving corporations power over the once neutral internet.”

Booker and his social media allies think the new rules will destroy the internet as we know it. Except, they won’t.

Read the full piece here.

 

Summary

Concern about both fake news and the size of Internet mega-platforms like Facebook is popular these days. In each case the concern is intuitively obvious yet the pathway by which it manifests into tangible harm ambiguous. There are clear examples of “fake news” being used for illegitimate purposes, as well as examples of platforms engaging in (or facilitating) alarming behavior – but it is challenging to draw a clean line between such problematic conduct and other non-problematic or even desirable conduct. Better understanding these delineations is a pressing task.

Fake news is largely distributed via social media platforms like Facebook. Indeed, the more malicious of such news is often designed specifically to take advantage of these platforms. It is reasonable to think that the concerns that we have about each may therefore be related – that fake news is a Facebook problem. This is the approach put forth in recent work by Sally Hubbard, who argues that fake news is an antitrust problem. Her basic thesis is that platforms with substantial market-share, such as Facebook, have pushed quality news organizations out of the market and that those news organizations would be better able to compete for consumer attention if there were more competition between platforms like Facebook.

It is a clever and provocative argument. But it is ultimately not a compelling one. Facebook isn’t what’s killing quality news – the Internet did that, and Facebook (and other social media) are merely the deformed phoenices that arose from the traditional media’s online ashes. Facebook and its ilk may be “killing news,” but it is not because these mega-platforms are harming competition – rather, the problem is that traditional media simply cannot effectively compete with social media in the winner-take-all marketplace for consumer attention. This may be a problem – it is certainly an issue that we as a society are and will continue to consider from law and policy perspectives – but it is not an antitrust problem.

I address these issues in more depth in the following three parts. I start by reviewing the evidence about what is killing the news (it’s not Facebook!). I then look at competition in the information economy and at the horizontal and vertical relationships between Facebook and the news media. I then turn the argument on its head, looking at how the problem we face – both with too little quality news and too much fake news – may be better addressed with less competition rather than more.

Throughout this discussion I will treat two recent articles as urtext: Hubbard’s piece in Forbes in which she explains “Why Fake news Is An Antitrust Problem,”2 and a follow-up interview on the topic that she did with Vox.3 I also note that throughout I will follow Hubbard’s lead and use Facebook as the poster-example of a significant social-media platform – though both she and I recognize that other tech platforms operate in this space. Indeed, the fact that Facebook, Twitter, and Google are all important platform-sources of news (fake and otherwise) demonstrates the most basic concern with the argument, that there is no lack of competition for information, true or otherwise.

Withholding Time Warner content from competitors would make no financial sense. AT&T has agreed to pay $85 billion for Time Warner. More than half of Time Warner’s revenue, $6 billion last year, comes from fees that distributors pay to carry its content. Because fewer than 15% of home-video subscriptions are on networks owned by AT&T (DirecTV, U-verse, and DirecTV Now), the bulk of that revenue comes from other providers.

Read the full piece here.

The last time I saw Fred McChesney was at a conference at Notre Dame where we both spoke, three years ago. We laughed heartily about how the stock market fools political observers. When a presidential candidate who will do terrible things to the economy is running, the stock market will tank as he appears to gain credibility as a successor, leading journalists and voters to blame the incumbent president for the market fall.

Read the full piece here.

Summary

This case raises significant questions about the thoroughness with which a court must review agency decisionmaking—or the extent to which a court may instead defer to that decisionmaking—when the agency has reversed a prior policy determination in the absence of a change in applicable law.

The Open Internet Order (“OIO”) issued by the Federal Communications Commission (“FCC” or “Commission”) presents such a policy reversal. The FCC ostensibly rooted the OIO in sufficient factual and legal analysis, but closer examination reveals that the OIO is based upon implausible factual assertions, questionable factual reinterpretations, and the strategic disavowal of long-defended statutory interpretation, all in support of a radical change in federal telecommunications policy that raises questions of vast economic and political significance.

Nevertheless, as discussed in Part I, the D.C. Circuit opinion affirming the OIO reflexively afforded substantial deference to the FCC, declining to consider serious questions about the reasonableness or permissibility of the FCC’s decisionmaking process. That decision is both in tension with this Court’s precedents and, more, raises exceptionally important and previously unaddressed questions about this Court’s precedents on judicial review of agency changes of policy.

As discussed in Part II, recent empirical work suggests that there are systematic problems with judicial review of agency changes in policy. These problems—respecting the substantive quality of agency and judicial decisions as well as judicial understanding of, or compliance with, this Court’s precedents governing such review—have led to consistently inconsistent review of agency policy changes in the circuit courts. Judicial review of agency policy changes thus presents a certiorari doublewhammy: there is a need for this Court to clarify existing precedent regarding judicial review of such policy changes and to address inconsistent application of that precedent, as well as for this Court to consider whether evidence of systematically problematic decisionmaking when agencies change policies militates in favor of a more searching standard of review.

Part III discusses how the D.C. Circuit and the Commission’s OIO implicate these concerns.

A new article by Professors Cass Sunstein and Adrian Vermeule highlights the exceptional significance of this issue. See Cass R. Sunstein & Adrian Vermeule, The Morality of Administrative Law, HARV. L. REV. (forthcoming 2018). In discussing empirical evidence collected by Professors Kent Barnett and Christopher Walker (discussed in Part II), Sunstein and Vermeule note that there is a “discrepancy between the law on the books and the law in action” when it comes to how courts review changes in agency policy. Id. (manuscript at 24) (https://papers.srn.com/abstract_id=3050722).

In National Cable & Telecommunications Ass’n v. Brand X Internet Services, 545 U.S. 967 (2005), this Court held that an agency’s alteration of policy is not grounds for heightened scrutiny. Id. at 981 (“Agency inconsistency is not a basis for declining to analyze the agency’s interpretation under the Chevron framework. Unexplained inconsistency is, at most, a reason for holding an interpretation to be an arbitrary and capricious change from agency practice under the Administrative Procedure Act.”). But, as Sunstein and Vermeule observe, “Brand X notwithstanding, the Court just isn’t particularly clear or consistent about the role of consistency under Chevron.” Sunstein & Vermeule, supra (manuscript at 23-24 n.159).

Indeed, “[a]t the level of individual cases, although no subsequent case has denied the rule expressly laid out in Brand X, opinions have occasionally adverted to consistency as a Chevron factor—including opinions for the Court.” Id. (ms. at 23). Moreover, contrary to the rule laid out in Brand X, “[a]t the level of large-N research, recent work by Chris Walker and Kent Barnett shows that judges in fact tend to defer more heavily to consistent agency interpretations.” Id. (ms. at 23-24).

In this instance, it seems likely that the policy under review will reverse course yet again, with the agency returning to the pre-OIO interpretation of the
law and issuing new rules consistent with that interpretation. Indeed, it must be acknowledged that the FCC could reverse the OIO as soon as December of this year. Under ordinary circumstances this would appear to moot, or at least substantially lessen, the concerns raised by petitioners here.

But the foreseeability of significant administrative policy changes—in this case and elsewhere—abetted by the precedent of substantial deference established in this case, militates in favor of the Court granting
certiorari. Should the FCC reverse the OIO, it is a foregone conclusion that supporters of the current order will challenge that reversal in a proceeding that will raise many of the same legal concerns currently at issue. The issuance of a new rule will thus not moot the issues in this case, but simply raise the precise issues yet again. Indeed, without clear guidance from this Court, there is every reason to believe the process will become an endless feedback loop—in the case of this regulation and others—at great cost not only to regulated entities and their consumers, but also to the integrity of the regulatory process.

Summary

Over the past 20 years, credit cards have become an increasingly popular means of paying for goods and services in Canada. Today nearly 90 percent of Canadian adults own a credit card and approximately 65 percent of all point of sale payments are made using credit cards.

The rise of credit cards has been driven by the benefits that accompany their use, including convenience, security, insurance, and warranties on purchases. But arguably the biggest driver has been the rewards that cards offer, such as cash back, Air Miles or Aeroplan rewards, or merchant-specific rewards. About 80 percent of Canadians with credit cards have at least one card that offers rewards for use, and owners of credit cards with rewards say that the rewards are the primary reason they use their rewards card for purchases.

The benefits provided by credit cards are paid for by the issuing bank through a combination of annual fees charged to cardholders and transaction fees charged to merchants. In closed-loop three-party card systems (primarily American Express, as well as international cards issued by Discover), the payment card provider charges both merchants and consumers directly. In four-party card systems (Visa and Mastercard), card issuers charge cardholders directly but the fees from merchants come via the acquirer (such as a merchant’s bank), which charges merchants a service charge. The largest portion of the merchant service charge is the interchange fee, which is passed on to issuing banks.

In spite of the higher annual fees on cards with more benefits, the vast majority of consumers report that they receive more benefits from their cards than the cost of the fees they carry. Middle class consumers are the major beneficiaries of credit card rewards. A consumer or household earning $40k might expect annual rewards valued at $450, while paying fees of $75, providing a net bene t of $375. Meanwhile, a consumer or household earning $90k might expect benefits of about $1350 while paying $225 in fees, providing a net bene t of around $1125.

Merchants, however, are less happy with the higher interchange fees. Apparently assuming that all of the bene t of rewards cards accrues to users, while merchants bear the added interchange cost, these merchants say that the increase has negatively affected their profitability. Of note, however, the number of merchants who accept credit cards, after falling in the early 2000s, has increased in the past decade – and appears to have risen more rapidly following the introduction of more generous rewards cards, in spite of a rise in accompanying interchange fees.

Some merchant groups have, in fact, called for the government to impose caps on interchange fees; in February 2016, a private member’s bill was introduced in Parliament seeking to do just that.

Interchange fee caps, like other price controls, tend to have predictable effects: as a rule, they result in other prices increasing, leading to a redistribution, but not a reduction, in overall costs. Several other countries have introduced caps on interchange fees, including, of particular relevance, the caps introduced in Australia in 2003. These caps resulted in a significant increase in the annual fees charged to cardholders and a substantial reduction in the rate at which card use earned rewards.

Using data on and analysis of the effect of Australia’s interchange fee caps, combined with publicly available and proprietary data on Canadian credit card use, household income and expenditure, and other economic variables, the authors of this report modelled the likely effects of introducing a cap on interchange fees in Canada. They estimate that, were an interchange fee cap imposed here, it would have significant negative consequences for Canadian consumers and the Canadian economy as a whole. Specifically, they estimate that if interchange fees were forcibly reduced by 40 percent:

1. On average, each adult Canadian would be worse off to the tune of between $89 and $250 per year due to a loss of rewards and increase in annual card fees:a For an individual or household earning $40,000, the net loss would be $66 to $187; andb for an individual or household earning $90,000, the net loss would be $199 to $562.
2. Spending at merchants in aggregate would decline by between $1.6 billion and $4.7 billion, resulting in a net loss to merchants of between $1.6 billion and $2.8 billion.
3. GDP would fall by between 0.12 percent and 0.19 percent per year.
4. Federal government revenue would fall by between 0.14 percent and 0.40 percent.

The authors estimate that a tighter cap on interchange fees would have a more dramatic negative effect on middle class households and the economy as a whole.

They also provide specific case studies for three typical middle class households, showing how a cap on interchange fees, along the lines of those imposed in Australia, would affect their household income and expenditure.

Continue reading at Macdonald-Laurier Institute

Summary

In its description of this workshop, the Commission notes that “consumers may suffer injury when information about them is misused,” and suggests that this workshop “will address questions such as how to best characterize these injuries, how to accurately measure such injuries,” and so on. While these are crucial questions, we offer these comments in order to address another set of questions that is missing from the event’s description: How should the Commission determine whether or not, in fact, the conduct leading to such injuries constitutes actionable “misuse[]?” The question is a fundamental one that must be addressed in order to evaluate how businesses, consumers, and the Commission itself do and should respond to purported informational injuries.

Fundamentally, there is a great deal of ambiguity about how consumer protection law should treat data and data breaches. When there is a data breach, the calculation of the extent of informational harm (if any) to consumers is a difficult one. This is complicated, of course, by the sometimes tenuous connection between conduct and injury. It is further complicated, even assuming that particularized harm can be accurately assessed, by the need to balance harms against the benefits conferred by decisions within the firm to optimize a product or service, to lower prices, or to promote other consumer-valued features, such as ease-of-use, performance, and so forth. Where the same conduct that may produce informational injury also produces consumer benefit, determining whether the net effect is, in fact, harmful or not is essential.

The Commission purports to evaluate injury (along with the other elements required by Section 5(n) of the FTC Act) under a so-called “reasonableness” standard. Superficially, at least, this seems sensible: Unfairness entails a balancing of risk, benefits, and harms, and a weighing of avoidance costs consistent with a negligence regime.3 Easily seen and arguably encompassed within this language are concepts from the common law of negligence such as causation, foreseeability and duty of care. The FTC collapses this into its “reasonableness” approach, specifically eschewing strict liability:

The touchstone of the Commission’s approach to data security is reasonableness: a company’s data security measures must be reasonable and appropriate in light of the sensitivity and volume of consumer information it holds, the size and complexity of its business, and the cost of available tools to improve security and reduce vulnerabilities…. [T]he Commission… does not require perfect security; reasonable and appropriate security is a continuous process of assessing and addressing risks; there is no one-size-fits-all data security program; and the mere fact that a breach occurred does not mean that a company has violated the law.

Giving purchase to a reasonableness approach under the Commission’s own guidance would seem to require establishing (i) a clear baseline of appropriate conduct, (ii) a company’s deviation from that baseline, (iii) proof that its deviation caused, or was significantly likely to cause, harm, (iv) substantial harm, (v) proof that the benefits of (e.g., the cost savings from) a company’s conduct didn’t outweigh the expected costs, and (vi) a demonstration that consumers’ costs of avoiding harm would have been greater than the cost of the harm.

Unfortunately, by eliding the distinct elements of a Section 5 unfairness analysis in the data security context, the FTC’s reasonableness approach risks ignoring Congress’ plain requirement that the Commission demonstrate duty, causality and substantiality, and perform a cost-benefit analysis of risk and avoidance costs.

While the FTC pays lip service to addressing these elements, its inductive, short-cut approach of attempting to define reasonableness by reference to the collection of practices previously condemned by its enforcement actions need not — and, in practice, does not — actually entail doing so. Instead, we “don’t know… whether… practices that have not yet been addressed by the FTC are ‘reasonable’ or not,” and we don’t know how the Commission would actually weigh them in an actual rigorous analysis.

At the root of this workshop is the implicit recognition that some, including the FTC itself, have asserted that the unauthorized exposure of private information may be, in and of itself, a harm to individuals, apart from any concrete economic consequences that may result from the exposure. In the FTC’s Opinion in LabMD, for instance, the Commission asserted that

the disclosure of sensitive health or medical information [that] causes additional harms that are neither economic nor physical in nature but are nonetheless real and substantial and thus cognizable under Section 5(n)… disclosure of the mere fact that medical tests were performed irreparably breached consumers’ privacy, which can involve “embarrassment or other negative outcomes, including reputational harm.”

We would contend, however, that defining and evaluating the types of “informational harms” that should be actionable in the case of a data breach, requires that the Commission also address fundamental problems with its overall approach to identifying cognizable injury and determining liability under Section 5.

As we discuss below and explain in detail in the attached paper, the FTC’s current “reasonableness standard” for liability under Section 5 runs the risk of being no standard at all. And it is impossible to escape the troubling conclusion that ultimately (and wrongly) the mere retention of data by a firm could be enough to violate Section 5 under this approach.

Such an approach does not comport with the scope of the Congressional grant of authority in Section 5, particularly as it was explicitly limited by Section 5(n). Instead, it converts what should be thought of fundamentally as a demanding cost-benefit requirement meant to limit the Commission’s discretion into a lenient strict liability standard. Before the Commission can understand how to fit different sorts of potential harms into its enforcement framework, it should clarify its approach, and ensure that it is in line with the text and intent of Section 5.