The Journal of Things We Like (Lots)
Select Page

Third Party Standing Is Not for Sissies

Curtis Bradley & Ernest Young, Unpacking Third Party Standing, __ Yale L. J. __ (forthcoming), available at SSRN.

Justice Scalia once famously said: “Administrative Law is not for sissies.” His colorful rhetoric undoubtedly was based on the combination of opacity, complexity, ambiguity, and internal inconsistency that characterizes the field of administrative law. The law governing third party standing has similar characteristics. Curtis Bradley and Ernest Young do an excellent job of “unpacking” third party standing in Unpacking Third Party Standing, but it too would not be a good candidate for casual reading by “sissies.” I have read it twice now, and I am far short of having a complete understanding of the intricate analysis in the article. The quality of the analysis is so good, however, that I plan to read it several more times.

The article is extremely ambitious. It is an attempt to “unpack” and explain a doctrine that many fine scholars have been unable to explain in a coherent manner. The reasoning the Supreme Court has used when it has addressed the doctrine is often inconsistent, unhelpful, and incomplete. The authors attribute the failure of the Court and scholars to describe and explain the doctrine in a coherent manner to their attempt to describe it as a single doctrine.

The authors use the general patterns of the Court’s decisions and one test that the Court has long used for a particular narrow purpose—the zone of interests test—in their effort to explain third party standing doctrine. In their view, the doctrine should be understood to refer to three discrete circumstances in which third parties should or should not have standing—parties that are directly regulated, parties that are collaterally injured, and parties that are representing other parties. Before they can complete that ambitious project, however, the authors must first “unpack” first party standing and identify the existence and scope of many of the substantive rights that are frequently invoked in third party standing cases. After taking those two preliminary steps, the authors use the results as inputs in the process of identifying the three circumstances in which a party should or should not have standing to assert the rights of another and the prerequisites to standing that a party should be required to establish to qualify for third party standing in each of the three classes of cases.

The authors argue that standing should be relatively easy to demonstrate in cases in which the party that is asserting the rights of a third party is directly regulated by the rule that is being challenged. In such cases, the only possible limit on standing is prudential rather than constitutional. The petitioner should have standing to assert the third party’s rights if the challenged rule could plausibly violate the rights of the third party. The traditional requirements to establish third party standing—demonstrating that the petitioner has an adequate relationship with the third party and that the third party confronts obstacles to its participation—should not be prerequisites to standing in this category of cases.

By contrast, in the second category of cases, a party that is only collaterally injured by the challenged rule should be denied standing except in the rare cases in which the party can satisfy the relationship plus obstacle test.

The third category of cases—where a party is representing a third party—raises serious Article III questions. In that category of cases, the court must decide whether the party can invoke the injuries to the third party caused by the challenged rule as the basis for standing. The authors argue that courts should apply agency law rigorously to determine whether the relationship between the representing party and the third party is sufficient to allow the representing party to rely on injuries to the third party as the basis for its standing to assert the rights of the third party.

After they identify the three classes of cases in which courts should or should not grant third party standing, the authors apply the prerequisites to the availability of third party standing in each of those three classes of cases to four types of actions that courts have long entertained without seriously considering the propriety of allowing third party standing—class actions, actions by organizations on behalf of their members, multi-district litigation, and attempts to obtain injunctions with nationwide scope. They argue that the Court should reconsider the permissibility and scope of each of those common practices to ensure that it is applying the normatively defensible prerequisites to the availability of third party standing in each of the four contexts in which the Court has routinely allowed third parties to assert the rights of others.

I don’t know whether I agree with the authors. I would have to reread the article with care several more times and reread some of the many authorities they cite to decide whether I agree with them. I plan to undertake that daunting task, however. The article is well-written, well-reasoned and well-researched. The authors have cited every important scholarly article that is relevant to their project. Their discussion of the sources they cite provides solid evidence that they understand each. This is a must-read article for any scholar, judge, or practitioner who wants to try to understand third party standing.

Cite as: Richard Pierce, Third Party Standing Is Not for Sissies, JOTWELL (May 18, 2021) (reviewing Curtis Bradley & Ernest Young, Unpacking Third Party Standing, __ Yale L. J. __ (forthcoming), available at SSRN),

Artificial Intelligence Meets Simplexity

Joshua D. Blank & Leigh Osofsky, Automated Legal Guidance, 106 Cornell L. Rev. 179 (2020).

Agency reliance on subregulatory guidance to advise the public is a perennial topic of discussion among regulatory practitioners and administrative law scholars. We want agencies to be forthcoming in sharing their thoughts regarding the laws that they administer, yet we fret that they rely inappropriately on subregulatory guidance to avoid their procedural responsibilities, and we struggle to balance the two.

The use of artificial intelligence in the administration of government statutes and programs is another hot topic these days, and rightly so. Optimism abounds that agencies will be able to harness the machines to make administration fairer and more efficient, yet of course we should think critically as well about the problems that relying on computer algorithms to achieve administrative ends may raise. In Automated Legal Guidance, Joshua Blank and Leigh Osofsky extend their wonderful work on “simplexity” in tax administration to put these concepts together and offer a critique of government reliance on artificial intelligence to provide guidance to the public.

As Blank and Osofsky explained a few years ago in another article, Simplexity: Plain Language and the Tax Law, “simplexity occurs when the government presents clear and simple explanations of the law without highlighting its underlying complexity or reducing this complexity through formal legal changes.” As they documented, IRS publications routinely translate complicated tax concepts into plain language to make those concepts more accessible to the general public. The problem with this exercise is that it “(1) present[s] contested tax law as clear tax rules, (2) add[s] administrative gloss to the tax law, and (3) fail[s] to fully explain the tax law, including possible exceptions.” In short, there is a trade-off between linguistic simplicity and accuracy. That trade-off does not always favor taxpayers, and it can result in inequitable treatment of different types of taxpayers.

The simplexity concept is by no means unique to tax. Many agencies publish guidance that aims to simplify complicated statutory and regulatory concepts for general audiences. In Automated Legal Guidance, Blank and Osofsky extend their simplexity critique to guidance from other agencies, and particularly guidance that is automated. Their examples include the U.S. Citizenship and Immigration Services virtual assistant “Emma” as well as the “MISSI” system used by the Mississippi state government to help people determine which state agency or service might be able to help them with particular problems. Their primary focus, however, is the IRS’s Interactive Tax Assistant (ITA), described by the IRS as “a tax law resource that takes you through a series of questions and provides you with responses to tax law questions” and that “can determine if a type of income is taxable, if you’re eligible to claim certain credits, and if you can deduct expenses on your tax return.” Suffering from deep budget cuts, the IRS has cut back on access to human beings to provide the public with tax assistance in favor of steering them to ITA.

Using a series of basic tax hypotheticals, Blank and Osofsky tested the accuracy and biases of ITA’s answers. Some answers were consistent with a more sophisticated reading of the tax laws. Other answers deviated from tax law requirements and were taxpayer-favorable in doing so, which at first blush might seem like a good thing but also could subject taxpayers who followed ITA’s advice to IRS enforcement. Still other answers were inconsistent with tax law requirements in ways that would deprive taxpayers of benefits to which they were entitled.

Blank and Osofsky acknowledge that ITA is superior to static written guidance in many ways. ITA is more personalized than written guidance and provides answers to questions that are at least clear, if infected by simplexity. ITA often is a quicker way to get answers than reading through written guidance. Nevertheless, ITA—and other automated guidance—can be improved, and those who promote the use of artificial intelligence in government administration would be wise to heed the suggestions that Blank and Osofsky advocate. The most obvious is simply to be cognizant of the tradeoffs inherent in simplexity. With that, government officials also should be aware of their audience and ensure that they “more accurately target the right legal dictates to the right people in the right situations” by adjusting the programming accordingly.

Blank and Osofsky also approach the issue of simplexity in automated guidance from the perspective of administrative law doctrine. They conclude, probably rightly, that the guidance provided by ITA is not legislative but interpretative in character, and thus not subject to notice-and-comment rulemaking procedures. On the other hand, they note that “the automated nature of systems like ITA seem to exacerbate problems already endemic to the administrative guidance.” Accordingly, they suggest “some form of centralized oversight, review, and public comment, regardless of whether such automated guidance is classified as a legislative rule.” Recognizing that noncompliance with subregulatory guidance in the tax context can lead to the assessment of penalties for tax underpayments, they argue that taxpayers ought to be able to rely on guidance from ITA as a defense against such penalties. Finally, they suggest that, as automated legal guidance evolves, agencies ought to figure out how to reduce its reliance on simplexity.

In a world in which government agencies are expected to do more and more with less and less, artificial intelligence holds great promise for the efficient administration of complicated statutory and regulatory schemes. Optimism in this regard should not blind us, however, to the tradeoffs and drawbacks of this turn to automation. Drawing from the tax system, with which millions of ordinary people interact regularly, Blank and Osofsky tell an important cautionary tale for those of us who care about the efficacy and legitimacy of administrative governance.

Cite as: Kristin Hickman, Artificial Intelligence Meets Simplexity, JOTWELL (April 20, 2021) (reviewing Joshua D. Blank & Leigh Osofsky, Automated Legal Guidance, 106 Cornell L. Rev. 179 (2020)),

Illuminating The Problems of Electricity Regulation

Shelley Welton, Rethinking Grid Governance for the Climate Change Era, Calif. L. Rev. (forthcoming 2021), available at SSRN.

Energy law today should be everyone’s concern, and especially the concern of administrative law scholars. Scientists report that we are in a climate emergency. Policymakers agree that the electricity sector will be vital to the clean energy transition. A functional electric grid is also a public good that cannot be taken for granted, as the recent disaster in Texas underscores. State and federal agencies, in partnership with their sibling branches, will play a pivotal role in administering the energy solutions the nation adopts. It is an administrative law problem for the ages.

Yet the field of energy law can be an impenetrable slog (I say this as a once and future teacher of the class). Really grappling with energy administration requires excavating dense layers of complex science and technology, unusual regulatory structures, and endless insiders’ terminology to reveal the important problems that lie beneath. Professor Shelley Welton is here to help. In a series of articles, she has elegantly translated the core dilemmas of the clean energy transition. Problems that seemed hyper-technical emerge as familiar administrative law issues of accountability, institutional design, and allocations of power among public and private entities. I’ll focus on one article, Rethinking Grid Governance for the Climate Change Era, but I recommend the entire series, available on SSRN.

If you’re new to the administrative law of energy, here’s something you might not know: for roughly 2/3 of the country’s population, key decisions regarding electric power rest not with a state or federal agency, but with privatized administrators known as Regional Transmission Organizations or Independent System Operators (RTOs for short). As Welton describes them, RTOs are “private membership clubs in which incumbent industry members make the rules for electricity markets and the electricity grid through private mini-democracies—with voting privileges reserved for RTO members.” They also have an important public dimension: FERC incentivized and guided their creation, effectively bestowing them with power to file “tariffs” (energy law speak for rate schedules and related rules and policies), subject only to deferential FERC review pursuant to the Federal Power Act.

Welton’s core argument is that to make needed climate progress, it’s time to rethink RTOs’ structure and role in our electricity system. States around the country are increasingly adopting bold clean-energy mandates, but RTOs are standing in the way—yet escaping public scrutiny. Once brought to light, she argues, RTOs’ shortcomings are best understood as a product of their flawed design. Their structure and incentives enable protectionism by powerful fossil-fuel incumbents in energy markets, and that’s what has happened. We need a new model.

Administrative law scholars will find much to admire in Welton’s reconstruction of how RTOs came about. She traces their emergence as a product of the late 1990s intellectual milieu in regulatory governance: a focus on “new governance” and privatized strategies that would allow government to run more “like a business.” Early on in their existence, RTOs did very technical work akin to traffic direction, such that privatization seemed sensible. But over time, RTOs have gained responsibilities and real clout, transforming from “policy takers” to “policy makers.” Every RTO now runs (and sets the rules of) a set of electricity markets—markets where generators sell and utilities buy electricity for a region before eventually selling it to end users. RTOs have also expanded into transmission planning and transmission cost allocation. Still more, several RTOs now control “resource adequacy,” the question whether a region has enough capacity to meet projected demand; through capacity auctions, RTOs determine which energy resources will be financed in that region. In short, in much of the country, RTOs can shape the electric grid’s sources (renewable or not), capacity (how much), and payment (to whom and how much). Given this influence, Welton explains that “if the United States is to have any chance at decarbonizing at the rate necessary to avoid catastrophic climate change, then RTOs must play a pivotal role.”

The problem is that RTOs haven’t been cooperative in aiding decarbonization. Welton describes them as “inveterate stallers when it comes to integrating new resources that would improve their markets but threaten incumbents’ bottom line,” and “aggressive and misguided” in impeding market competition by renewables. These delays and obstructions are highly consequential, effectively blocking the clean energy transition. Yet RTOs have escaped sustained criticism, much less reform. Welton suspects this is partly because “these topics are so complicated that they confound efforts at media attention or civic engagement.” Read a few of the documents and filings that she brings to life in this article, and I’m confident you’ll agree.

Welton’s intervention is a direct response to this problem of unaccountability-through-complexity. She makes the underlying energy concepts clear and accessible. Take her explanation of how some RTOs have resisted the integration of renewable resources in their markets. One would be forgiven for staring blankly when an RTO asserts that “the market participation of resources that receive ‘state support’ results in ‘price suppression and thus negatively impact[s] the market’s ability to retain and justly compensate needed existing resources and to attract new, competitively compensated resources.’” But Welton breaks it down. “In plainer speak, natural gas generators are worried that the entry of substantial renewable resources into the market might lower prices enough to drive fossil fuel competitors out of business, or halt future construction of fossil fuel-fired generation.” And “[c]uriously,” she observes, these RTOs define “state support” to include only state policies that promote clean energy, not longstanding subsidies for fossil fuels. Yet these RTOs explain their renewables-resistance only using “vague worries about ‘market integrity’ or ‘investor confidence.’” (To those excuses, Welton chides: “there is no reason that fossil fuel generators should be confident when building new, polluting generation for states that do not want it or need it.”) The narrative of incumbent protectionism becomes easy to grasp. It turns out that energy law is full of familiar administrative law stories once we can see the field clearly.

So what went wrong, in terms of regulatory design? Welton takes this up in Part IV of the article. Here she goes further than many commentators on privatized governance structures, who cast privatization as offering the benefits of increased efficacy and efficiency, to be weighed against the costs of decreased accountability. That’s too generous here, Welton opines. “RTOs’ membership-club format has not led to entrepreneurial efficiency,” she argues—it has done the opposite, allowing the energy haves to “block cost-reducing reforms.” Most impressively, Welton dissects internal RTO governance (famous for its opacity) to show why these failures shouldn’t be that surprising. For example, RTOs supposedly use a sort of internal checks-and-balances model of weighted voting, in which the demand and supply sectors of the market have voting power that could offset the other. But that’s no check at all because, as Welton explains, both of those sectors prefer to build more traditional electricity infrastructure rather than support a shift to renewables. Only consumers and their designated advocates are “natural watchdog[s] against overbuilding,” but they have negligible voting power. As a result, reforms that would help consumers (and the planet) “often wither and die in committee.”

Welton concludes by proposing four solutions that might contribute to a new model of administering the electricity system. First, FERC could pare back RTOs’ authority, reverting them “to a more basic set of functions.” Second, FERC might accept RTOs’ expansive duties but increase public oversight and control accordingly. For example, FERC might require RTOs to give greater say (or even veto power) to the states in which they operate; FERC itself might use what power it has left (after judicial rulings narrowly construing its authority) to step in; or Congress might increase FERC’s oversight authority. Third, either FERC or Congress could take steps to alter underlying power dynamics by limiting utility power or scrutinizing the effect of corporate mergers on the electricity system. Finally, and most radically, FERC or Congress might drive a transition to either public ownership or public control of the grid, a model used in other nations.

In the end, Welton’s article is a twofold achievement. It is an expertly told excavation of a pressing policy problem—to which solutions “must be calculated, swift, and decisive if the United States is to achieve anything close to the clean energy transition demanded by atmospheric physics.” It is also an ingenious analysis of how institutional design choices, and especially choices regarding the structure of privatization, drive administrative performance. Welton connects the history of RTOs to trends in regulatory thought and links RTOs’ future to efforts to reclaim the concept of “public utility”—the storied and reemerging idea that concentrated corporate power might be steered toward the public interest. That effort may be promising in a variety of sectors today. But Welton’s closing pitch is that none is more important than the electricity sector, “which will either embrace the existential challenge of climate change or take us all down with it.”

Cite as: Miriam Seifter, Illuminating The Problems of Electricity Regulation, JOTWELL (March 30, 2021) (reviewing Shelley Welton, Rethinking Grid Governance for the Climate Change Era, Calif. L. Rev. (forthcoming 2021), available at SSRN),

Patent Fake News

Janet Freilich, Ignoring Information Quality, __ Fordham L.R. __ (forthcoming 2021), available at SSRN.

Complaints about the patent system are legion. Critics complain that it is too easy to get a patent, that it is too easy to challenge an existing patent, that many patent denials are rationally inexplicable, that aggressive enforcement of patents stifles innovation, that patent trolls abuse the system to extort money from innocent users of widespread technology, and that inventors leverage modest modifications of existing patents to extend the patent period beyond intended legislative limits. While Janet Freilich’s forthcoming article, Ignoring Information Quality, may not reveal the root of all patent evil, it illuminates an important problem in the U.S. patent system, namely that patent examiners rely on low quality information to make their ever-important decisions on patentability. This, according to Professor Freilich, leads examiners to grant patents based on dubious claims that undercut, rather than further, patent law’s purpose of encouraging useful innovation and to reject deserving patents based on an incorrect understanding of background information.

The attentive reader may wonder why this is an administrative law jot rather than an intellectual property one. The answer is simple—the Patent and Trademark Office (PTO), the agency that grants patents, is an administrative agency, and thus Professor Freilich’s article is a case study in the importance of high quality information across the spectrum of administrative law. Information quality problems like those that plague the patent system exist in many corners of administrative law where sensible policy decisions and predictions are possible only in light of high quality information. Professor Freilich’s paper shines a light on a problem in the patent system that is similar to problems that have been noticed in administrative rulemaking, where mountains of comments may overwhelm the capacity of agencies to separate the wheat from the chaff and in adjudications where subjects of administrative action in areas such as immigration enforcement may lack the capacity or knowledge to gather and present the facts relevant to their cases.

Through careful examination of numerous patent files and caselaw on patent validity, Professor Freilich illustrates that patent examiners are not “digging” (her word) into the quality of evidence that is key to whether an invention is patentable. This is because, as courts have acknowledged, examiners lack the capacity to do the scientific work that would be necessary to evaluate the information. Patents may be granted or denied based on a cursory look at information that, with appropriate inquiry, would be revealed to support the opposite conclusion. Freilich contends, quite persuasively, that examiners’ “failure to evaluate evidence . . . permeates every aspect of examiner behavior.” If true, and Freilich’s article seems to establish that it is true, what we have is an unreliable administrative system for making highly consequential determinations, something administrative law should not tolerate.

Even more startling, in my view, is Professor Freilich’s analysis of the PTO’s policies which seem to guarantee that patent awards will be based on low quality, biased information. One key element of patentability is that an invention must be “useful.” PTO rules tell examiners that ordinarily they should accept an applicant’s assertion that their invention is useful and should only “rarely” request additional evidence. Even further, the PTO’s rules specify that the claim that an invention is useful should be rejected mainly only when the claim violates a scientific principle. Professor Freilich summarizes this as a requirement that examiners “accept the applicant’s stated utility unless it is utterly impossible.” This extreme standard does not inspire confidence in the accuracy of patent determinations.

Professor Freilich supports her claim that examiners rarely inquire into the quality of the information contained in the patent application with research that indicates that when examiners reject patent applications, it is always because a required piece of information is absent, not because the information provided is of insufficient quality. This is a disturbing finding because patentability depends on the accuracy of the information provided by the applicant, not merely whether the applicant has completed all of the required sections of the application.

Professor Freilich proposes to address the pervasive problem of low quality information in the patent system in a number of ways, all of which could be applied across a wide spectrum of information-dependent administrative proceedings. Professor Freilich’s first, quite simple, proposal is to require applicants to provide scientific corroboration for their conclusory statements on the elements of patentability, such as lab notebooks, copies of scientific analyses and physical models, or detailed drawings. She also suggests that all parties to patent applications, including inventors and their attorneys, should be held rigorously to a duty of disclosure to the PTO, with penalties such as additional scrutiny applied to applicants and their agents who do not disclose all relevant information in a useful format. Finally, she observes briefly that automation in the form of artificial intelligence may be useful tools for improving PTO performance.

Another of Professor Freilich’s proposals raises a central issue in administrative law—deference. In light of the generally low quality of information underlying many patents, she proposes that courts should no longer presume that patents are valid, at least not with regard to those elements that are fact-dependent. In her view, the patent process should be viewed more like a registration system than an examination system. In this light, courts would presume only that the PTO’s formal requirements have been satisfied, while leaving the substantive question of patentability open to non-deferential judicial determination. Professor Freilich recognizes that post hoc judicial determination is far from ideal, preferring her proposed internal reforms, but she recognizes that until the PTO starts performing more reliably, it may be the only alternative.

Professor Freilich’s article should be of interest to two groups of scholars, most obviously patent scholars but more broadly administrative law scholars looking to broaden their perspective into agencies that are rarely considered in the general study of administrative law. We have seen in recent decades a judicial broadening of the scope of general principles of administrative law to agencies that previously seemed to function in their own, isolated, spheres. Perhaps the PTO will be next.

Cite as: Jack Beermann, Patent Fake News, JOTWELL (February 22, 2021) (reviewing Janet Freilich, Ignoring Information Quality, __ Fordham L.R. __ (forthcoming 2021), available at SSRN),

Defending “Universal Vacatur” — Nationwide Injunctions for Administrative Law Nuts

Mila Sohoni, The Power to Vacate a Rule, 88 Geo. Wash. L. Rev. 1121 (2020).

The nationwide injunction has seized the imagination of courts and law professors in recent years. Not surprisingly, JOTWELL’s pages screens have given it extensive attention. Recent jots have described important work by Samuel Bray (twice), Amanda Frost (also twice), Russell Weaver, and Alan Trammell that attacks, defends, or theorizes nationwide (or “universal”) injunctions. Jack Beermann, in praising Bray and Frost, did have one complaint: “As an administrative law nut, I wish they both grappled more with the meaning of the APA’s instruction that reviewing courts should ‘hold unlawful and set aside’ unlawful agency action.” Mila Sohoni has now filled that void. Sohoni convincingly shows that there just can be no question that in the Administrative Procedure Act Congress authorized—indeed, indicated a preference for and established a presumption in favor of—nationwide relief when a court finds a regulation defective. When APA § 706(2) authorizes a reviewing court to “set aside” an agency rule, it means exactly that.

In 2018, Attorney General Jeff Sessions distributed Litigation Guidelines instructing civil litigators in the Department of Justice (DOJ) to oppose universal injunctions always and everywhere. The memo’s seven sections gave seven reasons why such relief was beyond the pale, including the assertion that it was unconstitutional. Section VII was headed: “In APA Cases: Universal Vacatur is not Contemplated by the APA.” Sohoni’s article resoundingly contradicts this assertion.

The term “universal vacatur,” used by DOJ and adopted by Sohoni, is a neologism. No court has ever used it, and a Westlaw search of the secondary legal literature reveals only two usages: one in the current article and the other an article by . . . Mila Sohoni. When a court finds an agency regulation unlawful, the order typically “vacates” the rule; that action is “vacatur.” There is some dispute about whether and when a court can or should remand a defective rule to an agency without vacating it—a practice the Administrative Conference of the U.S. has endorsed though only in limited circumstances—which is known as “remand without vacatur.” But there has not been any disagreement about what “vacate” means in this setting: to “vacate” a rule or portion thereof means to set it aside or to invalidate it, period. Therefore “universal vacatur” at least borders on redundancy. Still, Sohoni adopts the term, explaining that while it “is relatively unfamiliar (and perhaps a bit loaded), it does crisply capture the concept of setting aside a rule not just as to the plaintiffs, but as to anyone.” Let it be some solace to DOJ that it has won the (terminological) battle even though it has lost the (substantive) war.

Sohoni begins by situating universal vacatur within the administrative law context. Challenges to statutes often arise in the context of the application of the statute to a particular person; to grant the litigating party relief does not require “setting aside” the statute universally. Challenges to regulations can arise in that setting, as, for example, when an entity challenges the validity of a regulation as a defense in an enforcement action. When that happens, the obvious and often sufficient relief is to block application of the regulation to the party. But the APA also authorizes a direct challenge to a regulation. Post Abbott Labs, that is the norm. And it does so through language that directly authorizes courts to consider the validity of the rule itself, not the validity of its application in particular circumstances. Accordingly, preconceptions about universal relief appropriate to challenges to statutes have to be left at the door.

With that groundwork laid, the article is a familiar and convincing exercise in statutory interpretation.

With regard to text, the APA’s grant of authority to “set aside” a regulation reads naturally as authorizing universal vacatur. The obvious synonym, or definition, Sohoni says, is “invalidate.” What else could “set aside” mean? DOJ does not deny that “set aside” means “set aside.” Instead, it contends that the thing that courts are authorized to set aside is not the regulation itself but only its application to the party challenging it. As Sohoni explains, that is a bizarre way of understanding the standard pre-enforcement challenge to an agency regulation.

Sohoni also offers a lengthy examination of equitable remedies against agencies in the pre-APA period, demonstrating that what we would today call a nationwide or universal injunction was well-established by 1946, so there is no reason not to take the text to mean what it says. The provenance of the phrase “set aside” is not completely clear, but she offers one compelling nugget from the 1941 Attorney General’s Report, which noted that a “judgment adverse to a regulation results in setting it aside” (emphasis added). That sure sounds like it is the regulation, not its application, that is invalidated. The direct legislative history from Congress’s drafting and consideration of the Act five years later is thin on this point, so Sohoni has little to say about it, but she does draw some supportive inferences from the history of § 705, regarding stays during litigation. (Sections 705 and 706 are mutually supportive; if a court can grant universal vacatur under the latter, it surely can grant a nationwide injunction against the enforcement pending litigation under the former, and vice versa.)

Finally, Sohoni describes the consistent post-1946 understanding (at least consistent until the recent brouhaha) that courts that set aside a regulation under § 706 have authority to grant relief in the form of universal vacatur.

This article is not and does not purport to be the last word on the nationwide injunction debate. (Of course, the last word is unlikely ever to be uttered, though the Supreme Court may weigh in this Term.) Larger constitutional issues loom in the background, and Sohoni perhaps brushes them off a little quickly. But The Power to Vacate a Rule is the definitive statement on the statutory issue of the legitimacy of universal relief in challenges to regulations under the APA.

Cite as: Michael E Herz, Defending “Universal Vacatur” — Nationwide Injunctions for Administrative Law Nuts, JOTWELL (January 19, 2021) (reviewing Mila Sohoni, The Power to Vacate a Rule, 88 Geo. Wash. L. Rev. 1121 (2020)),

The Nondelegation Doctrine and a Deep Dive Into Federal Taxation of Real Estate in 1798 That You Didn’t Even Know You Needed

Nicholas R. Parrillo, A Critical Assessment of the Originalist Case Against Administrative Regulatory Power: New Evidence from the Federal Tax on Private Real Estate in the 1790s, 131 Yale L.J. (forthcoming 2021), available at SSRN.

Not so long ago, teaching the nondelegation doctrine in Administrative Law class was straightforward. Assign students an excerpt from Whitman v. American Trucking Association. Maybe add a bit of Justice Scalia’s dissent in Mistretta. Discuss the emptiness of the intelligible-principle principle. Everyone in class agrees that, whether you like it or not, a nondelegation doctrine that can accommodate delegations to act in the “public interest” or set “fair and equitable” prices does not do very much to limit the scope of the modern regulatory state.

But change has been in the air for several years—as most clearly demonstrated by Gundy v. United States (2019). As readers of this website will likely recall, the Court in Gundy addressed a nondelegation challenge to the Sex Offender Registration and Notification Act (SORNA), which requires sex offenders to register before completing their sentences of imprisonment. Barring time travel, this requirement would have been hard to apply to persons who had completed their sentences before SORNA’s enactment. To deal with them, SORNA delegated to the Attorney General the authority to “specify the applicability” of registration requirements and to “prescribe rules for registration.” Justice Kagan, writing for a controlling plurality, found sufficient constraints in SORNA’s text, purpose, and history to reject the nondelegation challenge. Just as about a century’s worth of the Court’s precedents might lead one to expect.

Justice Alito, concurring, expressed willingness to reexamine the nondelegation doctrine in a future case if a majority proves willing to do so, but thought singling out SORNA would otherwise be a bit “freakish.”

Justice Gorsuch, in a dissent joined by the Chief Justice and Justice Thomas, condemned the intelligible-principle principle as a liberty-destroying “misadventure” that “has no basis in the original meaning of the Constitution, in history, or even in the decision from which it was plucked.” He explained that delegations of rulemaking authority to agencies to govern private conduct are permissible in just three circumstances. First, if Congress makes the policy decisions, agencies can “fill up the details.” Second, Congress can impose conditional rules that come into force only if an agency later finds some triggering fact to be true. Third, Congress does no harm when it “delegates” authority to the other branches that they already possess under the Constitution.

Take Gundy, mix in Justice Kavanaugh and maybe Justice Barrett, and we appear to have a solid majority of the Court interested in overhauling the nondelegation doctrine in a way that is at least consistent with the conclusion that much agency regulation of private conduct is unconstitutional.

In response to this prospect, three leading scholars of administrative law have recently produced two major law review articles that deploy originalist arguments to debunk the nondelegation doctrine. Professors Julian Mortenson and Nicholas Bagley, based on a wide-ranging survey of Anglo-American legal thought before and after 1789, conclude that “the overwhelming majority of Founders didn’t see anything wrong with delegations as a matter of legal theory.” (Delegation at the Founding, 121 Columbia L. Rev. (forthcoming 2021)) Turning to practice, they identify numerous delegations of generous grants of discretionary rulemaking authority to administrative authorities from the very early years of the Republic. In their view, squaring this history with a meaningful nondelegation doctrine requires too many gerrymandered exceptions to be remotely persuasive. Delegation at the Founding, which does not pull any punches, has already generated a great deal of discussion in law reviews and the blogosphere. For one of several rejoinders, you might check out, Ilan Wurman’s Nondelegation at the Founding, forthcoming in the Yale Law Journal.

The other major salvo, and the subject of the remainder of this jot, is Professor Nicholas Parrillo’s splendid article with the really long title, A Critical Assessment of the Originalist Case Against Administrative Regulatory Power: New Evidence from the Federal Tax on Private Real Estate in the 1790s, which will be published next year in the Yale Law Journal. This article, too, examines early congressional legislation to shed light on whether the original understanding of the Constitution demands a nondelegation doctrine with sharp teeth. Rather than go for the magisterial sweep in the fashion of Mortensen and Bagley, however, Parrillo instead goes for the deep dive, exploring for over one hundred pages the discretionary powers that Congress granted to administrators to implement a tax on real estate that Congress imposed in 1798.

To explain why this 1798 tax is so important, Professor Parrillo notes that proponents of a strong nondelegation doctrine often try to explain away early congressional delegations by arguing that they fall into exceptions that allow more constitutional room for rulemaking. For instance, greater administrative discretion is permissible for foreign and military affairs because they fall within the natural domain of the executive in any event. Also, Congress can grant rulemaking discretion to administrative authorities to determine how to allocate benefits, services, and privileges. On this view, the nondelegation doctrine does, however, bar Congress from delegating to administrative authorities the power to promulgate “coercive regulation[s] of private rights and private conduct.” (P. 9.)

One way to combat the nondelegation doctrine in this limited form is to contend, as Mortensen and Bagley do, that the exceptions represent an implausible effort to save the doctrine from the evidence of history. Another way is to find early congressional delegations that in point of fact did grant agencies coercive power over private rights. And, in the form of the 1798 federal real estate tax, Professor Parrillo has found just such a delegation.

He explains that, in its first decade, Congress relied on import duties and excise taxes to fund the federal government. In 1798, however, Congress needed another revenue stream to help pay for a military buildup. In the Lay and Collect Act, Congress therefore imposed a direct tax of 2 million dollars, which it apportioned across the states according to their populations. (P. 22.) This tax fell, in the first instance, on slaveowners, who were to pay a tax of 50 cents per slave. The remainder of the states’ quotas were to be covered first by a progressive tax on houses and then, if necessary, by a flat tax on land. (Pp. 22-23.)

The Lay and Collect Act instructed that taxes on real estate should be calculated according to the terms of the Valuation Act. This statute, which Congress had enacted just a week earlier, provided that real estate should be valued at its “worth in money.” (P. 22.) As Professor Parrillo deftly explains, the extreme vagueness of this standard was intentional.

The Valuation Act divvied up states into tax districts. Each district got a federal tax commissioner who had been nominated by the President and confirmed by the Senate. Together, the commissioners of a given state formed its federal tax board. There were 94 commissioners in all. They appointed, on a conservative estimate, over 2100 principal and assistant assessors to aid them in their work. For those keeping score at home, that’s a bigger administrative apparatus than the Post Office had at the time. (P. 27.)

To enable the boards to carry out their functions, Congress delegated significant rulemaking powers to them. For instance, the Valuation Act gave each board authority to “establish all such regulations, as to them, or a majority of them, shall appear suitable and necessary, for carrying this act into effect; which regulations shall be binding on each commissioner and assessor, in the performance of the duties enjoined by, or under this act.” (P. 27 (quoting Valuation Act).) The boards’ “greatest” rulemaking power, however, enabled them to ensure fair allocation of the tax burden across districts by revising all the valuations in a district at once by rule. To this end, boards “had the power to raise or lower the tax assessments of thousands of property owners all at once, by any percentage amount, so long as the change ‘shall appear to be just and equitable.’” (P. 30 (quoting Valuation Act).)

“Just and equitable.” That sure sounds like a lot of regulatory discretion.

After explaining the structure, operation, and powers of the federal tax boards, Professor Parrillo explores:

  • The informational and methodological difficulties of real estate valuation in the 1790s—and the wide discretion that the boards necessarily had in carrying out mass-revision rulemakings as a result;
  • The political and contentious nature of the boards’ authority to allocate tax burdens across different geographic areas with different economic interests (e.g., determining assessments for coastal areas versus backcountry);
  • The binding nature of the boards’ mass revisions, which were “final and absolutely binding on taxpayers” with no avenue for judicial review (P. 15); and
  • The wide and bipartisan acceptance of the boards’ powers—and the notable absence of constitutional objections to them in connection with the 1798 tax or to similar taxes that Congress imposed in later decades.

Professor Parrillo has, by poring over centuries-old records of an almost forgotten tax and placing them in historical context, produced a truly impressive piece of legal scholarship. His Critical Assessment should play an indispensable role in the post-Gundy evolution of the nondelegation doctrine.

(Also, did I mention it has 695 footnotes?)

Cite as: Richard Murphy, The Nondelegation Doctrine and a Deep Dive Into Federal Taxation of Real Estate in 1798 That You Didn’t Even Know You Needed, JOTWELL (December 15, 2020) (reviewing Nicholas R. Parrillo, A Critical Assessment of the Originalist Case Against Administrative Regulatory Power: New Evidence from the Federal Tax on Private Real Estate in the 1790s, 131 Yale L.J. (forthcoming 2021), available at SSRN),

Administrative Law All the Way Up

Kathryn E. Kovacs, Constraining the Statutory President, __ Wash. U. L. Rev. __ (forthcoming), available at SSRN.

There’s a legal black hole at 1600 Pennsylvania Avenue where there shouldn’t be—or so argues Professor Kathryn Kovacs in a spirited new article, Constraining the Statutory President. A “legal black hole,” as Adrian Vermeule defined it, exists when statutes or legal rules create a zone in which the rule of law, including the constraints of administrative procedure and the checks of judicial review, cannot penetrate. And, as Vermeule pointed out, the Court created just such a black hole in Franklin v. Massachusetts by flatly excluding the President from the ambit of the Administrative Procedure Act: “the President is not an agency within the meaning of the APA.” Franklin is probably not what most people would call a “fixed star in our constitutional constellation.” But it is now a fixed “black hole” in the (administrative) constiution—a holding subsequently reaffirmed by the Court, abided by lower courts, left undisturbed by Congress, and treated as a given by many scholars.

Kovacs dissents from this acceptance of Franklin. Her article is one of a noteworthy set of recent articles that have examined presidential power and the constraints that should be imposed upon it. This boomlet of scholarship has addressed, among other topics, frameworks for judicial review of presidential orders, the importance of presidential fact finding, and the internal White House process of crafting presidential proclamations, directives, and orders. Speaking on a more conceptual plane, one new article offers a comprehensive defense of the thesis that the “duality” of the Presidency—i.e., that the President is both a mortal man and a public office—is a “defining ambiguity” of public law.

Kovacs’s subject is a different Presidential “duality.” At times, the President acts as the constitutional President—for example, when he vetoes a bill or nominates a justice. At other times, the President acts as if he were an agency—for example, when he issues proclamations or orders under color of statutory authority. When the President acts like an agency, Kovacs argues, then he ought to be treated like an agency. (Pp. 7, 49.) The Statutory President, she contends, should comply with the same administrative procedural and substantive requirements with which Congress’s other statutory delegates (i.e., ordinary agencies) must comply when they issue rule-like regulatory commands.

Franklin now blocks that possibility, but Kovacs argues that Franklin was both wrong and harmful: wrong as a matter of text, legislative history, and constitutional structure, and harmful as a matter of policy and administrative law values. The APA did not expressly exclude the President from its capacious definition of agency—and, she argues, if the APA’s drafters did think that the President wasn’t an agency, then that is because they couldn’t and didn’t anticipate how much consequential regulatory action modern presidents are liable to produce. (P. 35.) Kovacs notes that reversing Franklin would leave a wide swath of presidential action outside the scope of APA review, including many actions involving treaties, tariffs, or military affairs. (Pp. 29-30.) But in other regulatory realms, including in the domain of some immigration policymaking, the APA’s constraints would apply. All in all, Franklin should be reversed by the Court, she concludes. (P. 36.) And if the Court does not do so, Kovacs urges that Congress take action, either by amending the APA to legislatively override Franklin (P. 60) or by refraining from entrusting more power to the unconstrained actor at the apex of the executive branch. (Pp. 60-61.)

What turns on the President being an “agency” within the meaning of the APA? Savvy lawyers today are often able to frame their suits to evade Franklin—for example, by seeking nonstatutory review or by suing a subordinate officer. But Kovacs is not satisfied by such routes around Franklin; she would prefer to deep-six it altogether. To Kovacs, the stakes appear high: “Failing to curb the Statutory President,” writes Kovacs, “will further enable our nation’s descent into authoritarianism.” (P. 9.) On the other hand, she argues, imposing administrative procedural constraints upon the President would promote a suite of virtues and values familiar to administrative law, among them greater deliberation, enhanced political accountability, more robust public participation, and increased transparency. (Pp. 37-46.)

One important element of Franklin was Justice Scalia’s concurrence in that case. Kovacs correctly observes that Scalia “believed that [Massachusetts] lacked standing” (P. 14), but his opinion said a bit more than just that. Scalia offered a full-throated rejection of the proposition that any federal court could enjoin the President in the performance of his official actions. Indeed, Scalia rebuked the notion that a federal court could issue even a declaratory judgment that runs directly against the President: to his mind, it was a sad commentary on the degraded state of legal understanding that the district court in Franklinentered this order against the President without blinking an eye.”

If the President had been held to be an agency within the meaning of the APA in Franklin, then obviously many more such decrees against Presidential agency action would have been in the offing—orders that (on Scalia’s view) would have been not just unconstitutional, but embarrassingly unconstitutional. Writing for a slim 5-4 majority in the relevant portion of Franklin, Justice O’Connor accommodated that concern by reading the APA as excluding the President on constitutional avoidance grounds. But O’Connor also hedged about the consequences of that holding by expressing the Court’s confidence “that the President and other executive and congressional officers would abide by” a district court’s “authoritative interpretation . . . even when though they would not be directly bound by such a determination.” If that assumption is unwarranted, then that would provide another reason to question Franklin’s outcome.

The chief reward of Kovacs’ article lies in its focused exploration of the road left untaken in Franklin. Additional rewards lie in the interesting nuggets she has scattered throughout her discussion. For example, as Kovacs notes, the Court decided Franklin in an unusually short span of time—just over three months elapsed between the initial notice of appeal and the decision. And I learned for the first time from Kovacs that Franklin has caused a circuit split to “hatch” between the D.C. Circuit and the Federal Circuit concerning the availability of nonstatutory review. (P. 21.) Finally, I always enjoy reading about instances in which the sitting Justices litigated a particular position in their earlier careers as government lawyers or in private practice—and, as it happens, Franklin was argued for the United States by one John G. Roberts, Esquire. (P. 12.)

Today, of course, whether the Court would decide this issue differently may depend in part on the vote of the former government lawyer—Chief Justice Roberts—who successfully argued the Franklin appeal. For that and other reasons, persuading the Court to reverse Franklin may be an uphill fight. And, as Kovacs recognizes, the prospect of a Congressional override of Franklin “seems like a long shot” as well. (P. 60.) If the end result would seem to be that Franklin, warts and all, is likely to remain a fixture in the firmament of our administrative law, then that should perhaps not come as a surprise. As any physicist (not to mention Adrian Vermeule) could tell you, it’s not easy to fill up a black hole.

Cite as: Mila Sohoni, Administrative Law All the Way Up, JOTWELL (November 6, 2020) (reviewing Kathryn E. Kovacs, Constraining the Statutory President, __ Wash. U. L. Rev. __ (forthcoming), available at SSRN),

How Commercial Secrets Become Government Secrets

Deepa Varadarajan, Business Secrecy Expansion and FOIA, 68 UCLA Law Review __ (forthcoming, 2021), available at SSRN.

The breadth of exemptions to mandatory disclosure of government records under the Freedom of Information Act (FOIA) has long been criticized. The exemptions for national security, the agency privileges, and privacy are among those that have rightly come under fire. Even Congress has tried to rein in some of the most sweeping of these exemptions with recent amendments to the statute itself, capping the applicability of one exemption, the deliberative process privilege, to records less than twenty-five years old. But FOIA usually stands alone. Rarely is there an analogous body of law to learn from or to which it compares.

Deepa Varadarajan’s terrific forthcoming article, Business Secrecy Expansion and FOIA, demonstrates that FOIA’s trade secrets exemption is an exception. Varadarajan traces the history of trade secrets litigation back to its common law origins, documenting its steadfast march toward an ever-broader understanding of what constitutes a trade secret. Now, she explains, trade secrecy law protects any information of commercial value not generally known in the industry and which the owner has taken measures to protect.

In parallel, however, Varadarajan explains that FOIA jurisprudence stubbornly took a narrower view when interpreting the exemption for “trade secrets” and “confidential” commercial information. The “trade secrets” prong was limited to information tied to a production process (like formulas, plans, or devices), and the “confidential commercial information” provision was limited to information the release of which would cause substantial competitive injury. While information that firms voluntarily provided to the government qualified as exempt from disclosure under a broader interpretation, the generally narrower definition led to litigation successes in obtaining a wide variety of records regarding, for example, government contracting, private prisons, and more.

That is, the narrow construction prevailed until the Supreme Court’s decision last year in Food Marketing Institute v. Argus Leader Media. (Disclosure: I was amicus curiae in FMI.) Varadarajan artfully describes how FMI discarded the stricter FOIA standard in favor of aligning FOIA with modern trade secrets litigation. In effect, it rendered exempt from the FOIA disclosure mandate any information submitted by a firm that the firm does not routinely share and provided to the government under an assurance of privacy.

While details of how this standard is applied are being hashed out in the lower courts now, I was struck by Varadarajan’s central observation about the relationship between this development and trade secrets litigation. Not only does it move FOIA towards modern trade secrecy, she theorizes critically important ways in which it may make FOIA’s exemption cover even more business information than is protected by trade secrecy. Since trade secrets litigation is typically brought against former employees, there are certain inherent limits to how it might apply. To begin, it cannot be applied to limit employees from using their general professional knowledge and skills and it cannot apply to that which is generally known within the industry (even if not known in the general public).  In addition, and maybe even more importantly, it is limited to that which has some actual commercial value to the trade secret owner. Yet none of those limitations appear in the FMI standard under FOIA or would arise in litigation between a member of the public and a government agency.

Varadarajan’s work is important in several respects, not the least of which is exposing the relationship between trade secrets litigation and the application of the trade secrets exemption under FOIA. Beyond that, she describes how government-held trade secrets are precisely the ones that should be more narrowly defined. After all, these claims are used to keep hidden from public view the activities of private companies performing essential government functions—everything from private prisons to algorithms used in law enforcement to pricing in publicly provided healthcare services. As more and more functions are privatized, this exemption grows in size. It may shield from the public information that is critical to FOIA’s oversight and accountability aims.

She ends with a compelling case that there are real paths forward, even after FMI. While various legislative fixes she proposes might be the obvious starting point, there are also interesting litigation strategies that can be employed. For example, her suggestion of using the 2016 amendment to FOIA requiring agencies to apply exemptions only when there is a “foreseeable harm” from release of the records could be interpreted to alter the substantive standard for trade secrecy claims.

Even more promising, she proposes government contracting terms that impose clear disclosure requirements, thereby negating any claim that the firm expected or was promised secrecy or was closely guarding that information. This sort of built in transparency on the front end has gained traction more broadly, and agency officials I have spoken with have sometimes referred to this concept as “FOIA-ready” record keeping, or other times “transparency by design.” The idea is if we make records in releasable forms at the outset, we can open more of the government to the public.

Varadarajan is exactly right when she writes that the relationship between trade secrecy and FOIA’s commercial exemption “has thus far gone unnoticed.” But her outstanding contribution provides the foundation that is greatly needed by scholars and advocates to imagine commercial secrets beyond the FMI decision.

Cite as: Margaret Kwoka, How Commercial Secrets Become Government Secrets, JOTWELL (October 6, 2020) (reviewing Deepa Varadarajan, Business Secrecy Expansion and FOIA, 68 UCLA Law Review __ (forthcoming, 2021), available at SSRN),

AI Agents in Federal Agencies

David Freeman Engstrom, Daniel E. Ho, Catherine M. Sharkey & Mariano-Florentino Cuéllar, Government by Algorithm: Artificial Intelligence in Federal Administrative Agencies, Report for the Administrative Conference of the United States (2020), available at SSRN.

The use of artificial intelligence is on the rise at federal agencies, and administrative law scholars aren’t paying enough attention. In a forthcoming article, Ryan Calo and Danielle Citron question whether this increasingly “automated administrative state” presents a legitimacy crisis. After all, legislatures delegate broad law-implementation authority to administrative agencies because of regulators’ “ability to accrue expertise and the prospect of flexible and nimble responses to complex problems.” Yet these agencies are increasingly subdelegating such implementation authority to “systems in which they hold no expertise, and which foreclose discretion, individuation, and reason-giving almost entirely.” Despite these concerns, Calo and Citron ultimately do not demand a deconstruction of the automated administrative state. Instead, they argue that “agencies should consciously select technology to the extent its new affordances enhance, rather than undermine, the [expertise] rationale that underpins the administrative state.”

The Calo-Citron article deserves its own Jot. But it also raises a more fundamental question: How automated is the federal administrative state today? Although some work has documented the use of artificial intelligence and machine learning (AI/ML) at a handful of federal agencies, we had lacked a system-wide study. Recognizing this deficiency, the Administrative Conference of the United States commissioned an all-star group of scholars to comprehensively examine this regulatory landscape. In February, those scholars—David Freeman Engstrom, Daniel E. Ho, Catherine M. Sharkey, and Mariano-Florentino Cuéllar—issued their 122-page report, entitled Government by Algorithm: Artificial Intelligence in Federal Administrative Agencies.

There is so much to like *lots* in this report—too much to cover in this short Jot. But I’ll flag a few highlights.

Part I of the report takes inventory of AI/ML across the federal bureaucracy. To do so, the researchers—including the report authors and a large team of law students, political scientists, and computer scientists—focus on the 142 most significant federal agencies. They find that 45% (64 agencies) have embraced AI/ML to some degree. As depicted in Figure 2 of the report, the most common use of AI/ML is for “regulatory resource, analysis, and monitoring,” followed by (in order) enforcement, public services and engagement, internal management, and adjudication. Roughly half of AI/ML use cases (84 of 157) were developed in house. Part I also breaks down the data by agency, subject matter, and implementation stage, among other things. A 43-page online appendix provides even more granularity, including details on all 157 use cases identified at these agencies. In reviewing the findings, I was struck by the fact that most agencies (though not all) experimenting with AI/ML have substantial resources. Many under-resourced agencies, such as agencies focused on immigration and veterans affairs, don’t seem to be doing much AI/ML innovation. I’ll return to this observation later.

After presenting the big picture in Part I, Part II details eight fascinating case studies into how federal agencies have utilized AI/ML. The case studies were not chosen at random, but as illustrations of how agencies utilize AI/ML in various regulatory settings. To illustrate agency enforcement practices, the report has case studies on the Securities and Exchange Commission and Customs and Border Protection. On regulatory analysis, it presents case studies on two pilot programs at the Food and Drug Administration. The report includes three case studies on citizen engagement—at the Consumer Financial Protection Bureau, the Federal Communications Commission, and the U.S. Postal Service.

On administrative adjudication, the report presents case studies on the Social Security Administration and the U.S. Patent and Trademark Office. As Cary Coglianese and Lavi Ben Dor have observed, federal agencies—compared to Article III federal courts—seem much more willing to embrace and experiment with AI/ML when adjudicating. Such experimentation in adjudicating individual claims may seem particularly concerning to many observers. The report explores how the Social Security Administrative has tried to improve the quality of administrative adjudication through the use of AI/ML—by clustering appeals by issue to be decided by specialized appellate adjudicators, by accelerating appeals based on predicted likelihood of success, and by leveraging natural language processing for quality assurance. Similarly, in its case study of informal adjudication at the U.S. Patent and Trademark Office, the report explores how the agency has utilized AI/ML with respect to patent classification, patent prior art search, trademark classification, and prior trademark search. In the patent prior art search context, for instance, the agency developed an in-house tool called Sigma. Yet it ultimately did not implement Sigma agency-wide because the pilot program revealed that the tool “improve[d] efficiency only for examiners with a computer science background.”

Each case study could be the subject of a standalone law review article. Indeed, for scholars looking for research agenda ideas in this area (or law students for student note topics), check out the implications section of each case study, which flags some of the most pressing issues related to that use of AI/ML.

The final part of the report (Part III) zooms back out to examine several cross-cutting implications of the report’s findings. The authors focus on six categories of implications and recommendations:

(1) building internal capacity;

(2) transparency and accountability;

(3) bias, disparate treatment, and disparate impact;

(4) hearing rights and algorithmic governance;

(5) gaming and adversarial learning; and

(6) the external sourcing challenge.

Similar to the case studies in Part II, these half-dozen implications categories could each be the subject of a full-length article. Consider, for instance, the first category of building internal capacity. The report recognizes that federal agencies generally face a “make-or-buy” decision when it comes to incorporating AI/ML in their regulatory activities. The procurement and development of AI/ML will continue to be a challenge for federal agencies. Outsourcing development has important advantages, as those outside of the agency likely have much deeper expertise in developing these AI/ML tools. Yet, as the report underscores, subject-matter expertise can be critical in developing the AI/ML, such that outsourcing the entire process to outside groups would be problematic.

Outsourcing in some circumstances may also be in tension with core values of administrative governance. As Calo and Citron argue, AI/ML developments at federal agencies should be

oriented toward the furthering of substantive commitments and values, such as access, quality, and self-assessment. They are not designed simply to save costs (and in the process undermine procedural commitments without garnering more efficiency) but rather to enhance the capabilities of the administrative state itself—both agencies and officials—to engage in more effective and fair governance. In general, they would not outsource agency functions requiring expertise and discretion to third parties whose software and hardware deliver neither. These efforts have potential to enhance the justification of the bureaucratic state by, ideally, generating knowledge, enhancing expertise, tailoring outcomes, and increasing responsiveness—the purported reasons Congress created agencies to carry out its will in the first place.

Not surprisingly, the report authors strongly recommend internal development even when an agency procures AI/ML from outside the agency. This internal development includes: (1) increasing technical infrastructure and data capacity; (2) developing staff capacity and competency; (3) investing in AI/ML strategy and “regulatory sandboxes”; and (4) incorporating “accountability and transparency by design.” The remaining implications categories in Part III similarly explore how AI/ML should be developed across the administrative state to address some of the dangers of automating regulatory activities.

A critical challenge to developing AI/ML internally, as I hinted above, is that federal agencies face significant budget constraints. Buying off-the-shelf solutions may thus be more tempting for agencies with limited resources, yet such outsourcing often carries greater risks from a Calo-Citron legitimacy perspective. And some agencies that may benefit the most from AI/ML innovations face greater budgetary challenges than others, such that buying AI/ML is not even possible. In the final implications category, the report wisely observes that make or buy are not the only options. Federal agencies can also borrow. These approaches include collaborating with outside non-commercial entities to develop AI/ML, sponsoring AI/ML innovation competitions, and collaborating with other administrative agencies. This borrowing category deserves much more attention in the literature and in the real world. Indeed, we may find similar parallels and helpful lessons from the history of military software procurement, as my colleague Bryan Choi explores in Software as a Profession, where the government pursued such a hybrid development and procurement strategy.

Government by Algorithm is a massive contribution to the administrative law literature, and it lays the foundation for an ambitious research agenda for this emerging subfield of administrative law. The report could not have come at a better time, as federal agencies turn more and more to AI/ML to regulate.

Cite as: Christopher Walker, AI Agents in Federal Agencies, JOTWELL (September 4, 2020) (reviewing David Freeman Engstrom, Daniel E. Ho, Catherine M. Sharkey & Mariano-Florentino Cuéllar, Government by Algorithm: Artificial Intelligence in Federal Administrative Agencies, Report for the Administrative Conference of the United States (2020), available at SSRN),

When the Government Breaks Its Financial Promises

Matthew B. Lawrence, Disappropriation, 120 Colum. L. Rev. 1 (2020).

What happens when Congress enacts a permanent commitment to pay for some ongoing activity, but then fails to appropriate the funds necessary to do so? Until recently, this phenomenon was almost unheard of. But contemporary examples abound. For example, Congress failed to fund different two different provisions of the Affordable Care Act that had committed the federal government to pay insurers through the “risk corridors” program and through “cost-sharing reductions”; failed to fund tribes who had elected under the Indian Self-Determination and Education Act to provide services that had previously been provided by the federal government; and failed to fund the Children’s Health Insurance Program, a complex cooperative federalism program that is deeply entrenched, albeit in different ways, in states’ health care coverage for children. These examples either reflect or generate inter-branch conflict, implicating the executive branch, and often courts, in what may at first glance seem to be simply an intra-legislative breakdown.

We need a vocabulary to identify and describe this phenomenon and a framework through which to understand and assess it. Matthew Lawrence’s new article Disappropriation does just that.1

Disappropriation, as Lawrence defines it, is “legislative failure to appropriate funds necessary to honor a government commitment in time to honor that commitment.” Disappropriation results from “the dissonance between Congress’s legislative power and its appropriations power.” In other words, Congress can enact laws committing the federal government to pay for something, but unless it also designates a source of funds, it has not provided an actual appropriation, as required by the Appropriations Clause.

Sometimes, Congress enacts legislation that harmonizes these powers, providing both a commitment to pay and an ongoing permanent appropriation. Think, for example, of Social Security and Medicare. Other times, however, Congress enacts legislation that reflects dissonance between these powers, committing the government to pay but depending for funding on the annual appropriations process. Food stamps and Medicaid, as “appropriated entitlements,” fall into this category.

When the dissonance between Congress’s legislative and appropriations powers results in a disappropriation, or even comes close to it, negative consequences can ensue. For example, those relying on the government’s anticipated payments can face devastating financial losses, with a particularly burdensome disparate impact on less well-capitalized smaller businesses and individuals without the means to survive the delay associated with a legal or political battle over payment. The rule of law also suffers, in that the executive branch is forced to violate a statutory command because it is without the means to do so. Disappropriation can also result in a problematic kind of delegation, where the executive branch is left to pick and choose how to dole out whatever minimal funds exist to support the commitment in the absence of an intelligible principle, because the law never contemplated this kind of executive discretion.

At the same time, as Lawrence points out, strategic exploitation of dissonance between Congress’s legislative and appropriations powers can result in a threat of disappropriation that may have salutary consequences. While permanent commitments shift inter-branch powers towards the executive branch to implement those commitments, annual appropriations pull some of the power back towards Congress. This fact may strengthen Congress’s oversight function. For example, Congress can use the threat of disappropriation to extract executive commitments to produce documents or testify in committee. The threat of disappropriation can also empower blocks in Congress other than party leadership, so that issues of importance to other groups (members of the minority, committees) can rise to salience.

Is there a way to prevent the negative consequences of dissonance that result from an actual disappropriation while retaining the benefits that accrue from strategic exploitation of the threat of disappropriation? Lawrence suggests that there is a category of rules that does just this: rules that “promote durability (the likelihood that a policy will stay in place and so its capacity to engender reliance) but not entrenchment (the difficulty of changing policy for a majority that wishes to do so).” The key, he argues, is to prevent “bargaining failure,” where a disappropriation results because of either “uncertainty surrounding the impact of legislative action or inaction” or “private information about the consequences of disappropriation.”

In light of this goal, he proposes that courts adopt an interpretive presumption against disappropriation of unambiguous legal commitments; welcome lawsuits by civil servants to enforce disappropriation as a superior alternative to resolving questions about legislative standing; offer expedited declaratory (rather than injunctive) relief when there is a dispute about whether a disappropriation has occurred; and “seek to minimize interference with the political branches and reduce uncertainty surrounding the potential impacts of future disappropriations” when considering claims for damages in any individual disappropriations case. Lawrence’s careful analysis of each of these suggestions provides a critical tool for courts and litigants in thinking through the long-term, systemic implications of different paths for resolving discrete legal controversies.

Lawrence’s contribution is a welcome addition to the growing body of administrative law scholarship on federal funding. In a world where inter-branch disputes increasingly involve money—building a wall at the southern border with funds that far exceed Congress’s appropriations, placing a hold on aid to Ukraine while asking for investigations of political rivals, limiting oversight over Coronavirus relief—insights like the ones Lawrence offers in this excellent article are vital.

  1. I commented on a draft of the article for the 2019 Administrative Law New Scholarship Roundtable.
Cite as: Eloise Pasachoff, When the Government Breaks Its Financial Promises, JOTWELL (July 20, 2020) (reviewing Matthew B. Lawrence, Disappropriation, 120 Colum. L. Rev. 1 (2020)),