Notice & Comment

What Are Networks, Platforms, and Utilities and What Should We Do with Them? by Josh Macey and Genevieve Lakier

*This is the seventh post in a symposium on Morgan Ricks, Ganesh Sitaraman, Shelley Welton, and Lev Menand’s “Networks, Platforms, and Utilities: Law and Policy.” For other posts in the series, click here.

Networks, Platforms and Utilities (NPU) is an ambitious book.It covers an enormous range of industries and regulatory frameworks—everything from banking to the postal system to online marketplaces. As others have noted, it examines not only the current organization of these fields, but the crucial, complex histories that created them. This would be enough of an achievement for one book. But the casebook’s ambitions do not stop there: as its authors indicate in their contribution to this symposium, the book puts these various industries, histories, and regulatory frameworks into conversation with one another in order to change how students, scholars, and policymakers think about the regulatory state and its possibilities. The book wants to bring back a way of thinking about the regulatory state that has “dropped out of the legal and political imagination.”

The fact that the casebook serves these dual purposes means that it is not enough to evaluate the book pedagogically. Because the book is also a normative project, we must also assess how well it works ideologically. What vision of the regulatory state does it offer? And is this an attractive, correct, or persuasive vision?

There is not enough space in this short response to answer these questions fully. But we think they are important to put on the table. In what follows, we aim to show why it might be worth thinking seriously about the book’s politics, some potential challenges with the theory that guides it, and the regulatory solutions it offers.

1. What is an NPU?

    As its name suggests, the casebook seeks to define a new field of law: the field of networks, platforms, and utilities. It does so in order to revive, if also reshape, a much older field of law—the law of regulated industries. This is an exciting goal.

    For too long, the law of regulated industries has been regarded as a turgid academic backwater—an area in which all the good ideas had been developed by 1945. But it was not always so. NPU law, in its previous incarnations, once played a core part of law school curricula. In 1914, Felix Frankfurter began his class on regulated industries, which he taught in his first year at Harvard Law School, by claiming “that no branch of the law is more important than this law of public calling.”[1] We do not disagree. Government regulation of banking, railroads, telecommunications and the other industries that were previously brought together under the rubric of regulated industries serves enormously important goals. It ensures (or attempts to ensure at least) equitable access to the public sphere. It protects against monopoly. And it promotes the reliability of fundamental, infrastructural services. By demonstrating the extent to which the pursuit of these goals has in the past, and continues today, to motivate and shape the project of American governance, the casebook does a great deal to prove Frankfurter’s point. And by dispensing with the old labels (“regulated industries,” “the law of public calling”), it is able to bring into conversation bodies of law that would never have fit into the earlier rubrics.

    As with any project of definition and re-definition, though, defining the scope of this sort-of-new-but-sort-of-old field of NPU law has its challenges. The definitions the casebook provides are broad. A network, the casebook tells us, is a “system[], often characterized by ‘links’ and ‘nodes’ that connect[s] users to each other or to different places.” A platform is a “centralized space[]—physical or virtual—[that is] designed to facilitate particular economic or social activities.” And a utility “provide[s] essential, general-purpose services to their users on a more or less continuous basis.”

    On their face, these definitions appear to encompass many more industries than those highlighted in the book. Consider, for example, the food industry. Today, an incredibly complex system of “links” and “nodes” connects producers, distributors and retailers to ensure that food moves from the farm to the supermarket (or, increasingly, the Dollar Store) and then to the table. A similarly complex system ensures the delivery of healthcare and insurance. Like the other examples highlighted in the casebook, all three of those industries provide access to what we might think of as infrastructural goods—indeed, some of the most fundamental infrastructural goods one can imagine. But none makes an appearance in the casebook or shows any sign of appearing in a later edition. Why?

    This may feel like an unfair question to ask of a casebook that is already many hundreds of pages long. No casebook can discuss everything, it is true. Part of the art of a good casebook is knowing what to leave out. But the economic, social, and political importance of the food, healthcare, and insurance industries make it worth pausing to consider why they might be absent. Why does the social media industry get a chapter, but not these industries? One reason might be historical—perhaps these were not industries that were traditionally recognized as public callings or regulated as public utilities. But that isn’t quite true. Insurance was one of the first industries regulated as a public utility and remains subject to public utility regulation in many states. Meanwhile, in the early twentieth century, the category of public utility was fluid and adaptable.

    So perhaps the answer is that there are no vocal calls to regulate these three industries as public utilities—in marked contrast, here, to social media. If so, that suggests that the scope of NPU law is not defined solely by the definitions of networks, platforms, and utilities the casebook provides in its first chapter, but is instead a product of our political past and present.

    That is not necessarily a bad thing. Law is a creature of politics after all! One can well understand why the authors of the casebook might want to concentrate their fire on industries in which the NPU concept might have some influence today. Nevertheless, one might worry that by focusing on the industries that are today primarily associated with the public utility concept, the casebook provides a less ambitious vision for the regulatory state than it might otherwise. Would it be possible to apply the tools of public utility to healthcare? Or to food? What would that look like? Conversely, does focusing the conversation about regulatory reform on the industries that have historically been regulated by what the authors of the casebook describe as the NPU toolkit bind us too much to the tools of the past? What, in other words, does thinking about networks, platforms, and utilities that have not historically fallen under the rubric of regulated industries tell us about the logic and limits of the regulatory reforms the authors highlight?

    To put it another way: the casebook presents NPU law as occupying a relatively stable domain and suggests that there is a discrete set of industries that qualify as NPUs. But is this true? What goods and services in our complex economy do not travel through systems organized around nodes and links? Similarly, it is by no means only the industries highlighted in this book that raise concern about monopolistic or oligopolistic domination, destructive competition, and shoddy service. If so, does it unduly limit the book’s ambitions to demarcate a special area of NPU law that is defined by its exceptional character? Is the domain of NPU law, and the range of its regulatory possibilities, a product of politics—and, as such, subject to political reinvention?

    Here, again, Frankfurter is instructive. When Frankfurter introduced his regulated industries class, he urged his students to work with him to reinvent this body of law so that it could effectively address the Progressive Era’s distinct economic and social challenges. He explained that

    The Course itself is in the making, not only because your teacher is new; but because your subject is fresh. In a very true sense therefore, we must work out this Course together; for the law is, (as to a great portion of it) still fluid, there are still many unsolved problems and new problems constantly arise. Their correct solution devolves upon you and your contemporaries as these problems will come to you, as practitioners, as legislators; as judges and administrators. It is perfectly plain that the old lawyers and judges are not equipped for the task that is at hand. Their viewpoints were largely fashioned in a generation that had different controlling social ideas, and on the whole, they cannot make the adjustment, they cannot bring the openmindedness necessary for the task of formulating a new and progressing body of laws.[2]

    The law of regulated industries has been around for a long time, now—much longer than when Frankfurter was writing. But, as this casebook makes clear, it is ripe for reinvention and reconceptualization. One might wonder whether a truly revived NPU law might require—or push us to reimagine—the scope of its boundaries. At the very least, we think that while the casebook compellingly shows why the industries it covers are included in the field of NPU law, it would benefit from spending more time explaining and exploring why the borders are set where they are.

    2. Is NPU Administrable?

    We also worry about the gap between theory and practice. The NPU tools are hard to administer. Consider the example of electricity. Public utility regulation of the electricity industry occurred in part because of industry support. The business magnate Samuel Insull began advocating for public utility regulation in 1898, when he said that “the best service at the lowest possible price can only be obtained, certainly in connection with the industry with which we are identified [electricity], by exclusive control of a given territory being placed in the hands of one undertaking.”[3] As President of the Chicago Edison Company, it was obviously in Insull’s interest to argue that the public interest would be best served if regulators protected his electric companies from competition. Even if it wasn’t in his interest, one might imagine that public utility regulation appeared to Insull and wealthy capitalists like him a reasonable compromise to state power, at a time when municipal control of electric companies was coming dangerously close to public control (and socialism!).

    But when Insull’s empire collapsed during the Great Depression, it became clear that public utility regulation had created opportunities for large holding companies to expose captive ratepayers to the business risks undertaken by large holding companies. This concern featured prominently in Franklin Delano Roosevelt’s 1932 presidential campaign. In a speech in Portland, Oregon, Roosevelt said, “The Insull failure has done more to open the eyes of the American public to the truth than anything that has happened. It shows us that the development of these financial monstrosities was such as to compel inevitable and ultimate ruin.”

    One might respond that the early-twentieth century energy regulators had failed to effectively administer the NPU toolkit. After all, it was only after Roosevelt became President that Congress passed the Public Utility Holding Companies Act, which limited utilities’ geographic service territories and prevented utilities from entering certain business activities. But Roosevelt himself was concerned “that private manipulation had outsmarted the slow-moving power of government.” That concern has proven prescient.

    Today, electric utilities are regulated in precisely the manner NPU suggests. They are perhaps the best example we have of utility regulation. Yet with few notable exceptions, these companies have emerged as vocal and committed opponents of decarbonization policies. The states whose power sectors are responsible for the most emissions are those that continue to treat all segments of the electricity industry as public utilities. Rate regulated utilities pay above-market prices to buy coal from their own affiliates. They guarantee pipeline debt to secure regulatory authorization to build new gas infrastructure. They invoke their exclusive franchises to block policies that would support widespread electrification, rooftop solar, and energy efficiency programs.

    Electric utility opposition to the clean energy transition should not be surprising. Why would a company that earns a steady return from fossil generators voluntarily transition to cheaper and cleaner fuels? But as an example of how public utilities can end up wielding power not in the public interest, it poses a significant challenge to the NPU model. Can NPU law prevent ensure that regulators remain sufficiently competent and well-resourced over time to outsmart the manipulations of private industry? We hope so! But here too, we think the casebook should do more to highlight the dangers.

    3. How Much Should We Trust the NPU Toolkit?

    The casebook expresses some amount of nostalgia for the old NPU toolkit. Indeed, its explicit aim, the authors tell us, “is to provide a modern, comprehensive volume that pushes the field back to the center of legal education and scholarship and reintroduces its tools and logic to policymakers.” But as the previous section suggests, NPU tools are just as capable of impeding progressive reforms as they are of promoting a progressive vision.

    Take, for example, the case of entry restrictions. Historically, the primary argument for entry restrictions is that they enable incumbent firms to provide cross-subsidies to equalize the costs of service to differently-positioned consumers. Imagine a railroad that operates one route from New York to Boston and another route from New York to Syracuse. It is expensive to build a railroad. But once the company does so, the cost of transporting each additional customer is low. Without internal cross-subsidies, the railroad would be able to charge customers traveling from New York to Boston a relatively low fare, since they could spread their fixed costs across a large number of customers. However, because the New York to Syracuse route has fewer customers than the New York to Boston route, the company would have to charge its Syracuse customers more. Now imagine that the railroad is required to charge the same price to all customers. In that case, it needs to use the profits from its New York to Boston route to subsidize its Syracuse customers. Early twentieth century regulators were concerned that a competitor would enter the New York to Boston market, offer cheaper prices, and render the whole cross-subsidy scheme unsustainable. This is known as cream-skimming. The authors suggest that entry restrictions are needed to prevent cream-skimming.[4]

    The problem with this account is that policymakers have successfully sustained cross-subsidies while relaxing entry and exit restrictions. In the electricity industry, for example, regulators usually designate one utility as the “provider of last resort.” That company is required to provide customers with electricity service, often at subsidized rates, if someone’s retail electric provider goes out of business. Recently, the regulatory solution has not been to eliminate retail competition, but to make sure that the provider of last resort is compensated for providing backup power. Often, other retail electric providers—the companies that directly compete with the provider of last resort—pay a surcharge that allows the provider of last resort to offer retail electric service if other retailers go out of business.[5] Similarly, bank regulators can support lending to low-income communities through direct subsidies or by requiring all banks to offer a certain number of low-interest loans to low-income communities. The point is that subsidized services can come from a variety of sources, and that relaxed entry restrictions are compatible with a market in which one or all firms provided subsidized service to certain classes of customers.

    The fact that entry restrictions are not always required to enable cross-subsidies to occur is something that some of the individual chapters make clear. One of the surprises about the casebook is that there are sometimes tensions between the theoretical framework laid out in the opening chapters (which embraces the tool of entry restrictions) and the specific examples discussed in the later chapters (which problematize them). Many of the casebook’s specific examples explore the fluidity and adaptability of the NPU toolkit and challenge its traditional reliance on tools such as entry restrictions. For example, in the chapter on electricity, the authors note that the decision to relax entry barriers was an experiment with open access that ended up improving on the traditional public utility approach.

    We nevertheless wonder if, in the framing section, the authors could emphasize more than they currently do the costs of some of the tools in the toolkit, as well as their ongoing reinvention. And we hope that scholars and teachers will explore whether–and to what extent–the complicated regulatory systems described in the substantive chapters rely on and adapt the tools laid out in the opening chapters.


    None of the critiques above are meant to detract from the achievement of the new casebook. In the past century, a small number of casebooks have not just organized existing doctrine, but also served as a reference point for an emerging field of legal study. We suspect Networks, Platforms, and Utilities will join that company, and deservedly so. It synthesizes a doctrinal area that has fallen out of fashion, describes the regulatory toolkit that has been used to manage those industries, and suggests how that toolkit might be used to mitigate serious contemporary social and economic problems. We assume it will influence a generation of teachers and students to view the regulatory landscape in the terms it provides.

    Precisely for this reason, however, we think it is important to critically interrogate how the casebook presents the rediscovered world it depicts. Our point is not that we should reject the NPU toolkit, or that the authors’ attempt to synthesize different strands of NPU law is misguided, or even that the authors are wrong on any particular issue. We also are aware that in the individual chapters, there are many instances in which the authors hint—or more than hint—at some of the questions we raise.

    Nevertheless, we worry that the casebook, in its effort to remind students and teachers of the many tools that have historically been available to regulators who wish to turn private industry to the public good, paints an overly rosy picture of a legal regime that was always a compromise between the needs of industry and the needs of the public. These tools promote important ends, but they also raise a number of challenges that we think students, professors, and policymakers should be pushed to grapple with today, at this moment of resurgent possibility for the NPU model.

    The first set of challenges is administrative. How much do we trust regulators? Is that constant over time and across industry? And how do we mitigate the regressive policies that arise from applying the NPU toolkit?

    The second set of challenges is definitional. Which industries fall under the NPU umbrella? We think that it is important to problematize the idea that there is a coherent set of industries that can be understood as NPUs, and we want to understand the stakes of the NPU label. It is possible that the label signifies a great deal. But it is also possible that NPU regulation refers to a set of regulatory interventions, the merits of which can only be assessed by taking a careful look at idiosyncratic features of idiosyncratic industries. Finally, while the authors mention public ownership, how do we know when that should be a viable option, and how does the regulatory tool of public ownership compare to, and fit in with, the rest of the NPU toolkit?

    These are the questions that this rich and provocative book implicitly raises but that it could do more to put on the table. We are grateful to the authors for this opportunity to remark on their casebook, for raising this set of questions, and for creating a space for an important scholarly debate.

    Josh Macey is Assistant Professor of Law at the University of Chicago.

    Genevieve Lakier is Professor of Law and Herbert and Marjorie Fried Teaching Scholar at the University of Chicago.

    [1] See Class Notes for Felix Frankurter’s 1914 Class on Regulated Industries, at 165 (Sept. 28, 1914), available in the Harvard Law Library.

    [2] Frankfurter Notes at 168.

    [3] See Samuel Insull, Speech before the Nat’l Electr. Light Assoc. (1898).

    [4] NPU at 29.

    [5] Alternatively, the regulator may directly subsidize the provider of last resort.

    Print Friendly, PDF & Email