DRAFT

STRUCTURING DIGITAL PLATFORM MARKETS: ANTITRUST AND

UTILITIES’ CONVERGENCE

Elettra Bietti1*

Abstract

The regulation of digital platforms is frequently framed as a legal and

institutional trade-off. Should policy makers “regulate” or should they
“break-up” Big Tech? Should they decentralize digital power or should they
transform companies like Google into accountable bottlenecks? These
dichotomies reflect an impoverished – and deregulatory – understanding of
the scope of antitrust law and the nature of regulation in the digital economy.
Antitrust, which includes remedies such as break-ups, is conceived as a body
of law which acts marginally to preserve pre-legal and decentralized market
processes. Utilities and other regulatory schemes are viewed as rigid modes
of intervention in production that interfere with free competition and limit
consumer choice and innovation.

The polarity between antitrust and regulation obscures a more nuanced

reality where decentralizing and centralizing efforts, which structure and
enable digital markets, overlap across legal domains. The Article defends a
conceptual move away from disciplinary siloes and discontinuous remedial
solutions and toward a joint approach to law in digital ecosystems. In
practice, antitrust and regulatory law are converging in a new way. Antitrust
cases are increasingly sensitive to the gatekeeping power of platform
intermediaries and digital market regulation is becoming consciously
procompetitive. As such, conventional justifications for the distinction
between antitrust and regulation, e.g. the preference for underenforcement
and ex-post intervention, are weakening. Antitrust is but one branch of law
that structures and enables competition. Regulation does not undermine but
instead can promote competition, innovation and consumer choice.

Relying on the case of Google and its regulation between 1998 and 2022, the
Article situates antitrust and public utility efforts as part of a spectrum of
coextensive regulatory approaches to digital markets. It configures the space
of regulatory possibility across ex ante and ex post, centralizing and
decentralizing strategies. Its aim is to guide a move away from siloed or a-

1*
Joint Postdoctoral Fellow at NYU Law and Cornell Tech, S.J.D. Harvard Law School.

Electronic copy available at: https://ssrn.com/abstract=4275143

2 DRAFT [Feb-23

contextual efficiency or deregulatory justifications and toward situated legal-

regulatory decisions about the collective needs and choices that markets like
search or online advertising can advance. The digital platform economy is a
place to begin experimenting with new ways for law to structure production
in line with the public interest. The question is not whether to break-up or
regulate Big Tech, it is what forms of competition, innovation and choice are
needed in a digital society.

Table of Contents
Introduction ..................................................................................................................... 3
I. Digital Platforms, Antitrust Exceptionalism and The Regulatory State ............. 6
A. Digital Platforms: Definition, Evolution and the Role of Law ...................... 6
1. What is a ‘Digital Platform’? .................................................................... 8
2. The Role of Law in the Digital Platform Economy ................................ 10
3. Interpreting Convergence ........................................................................ 13
B. Conventional Accounts of Antitrust and Economic Regulation .................. 15
1. Competition, monopoly and consumer choice ........................................ 16
2. Antitrust and regulation as friends and foes ............................................ 18
C. The Public Utility as Exception to the Free Market Default ........................ 21
1. Infrastructure ........................................................................................... 22
2. Defining Public Utility Regulation.......................................................... 23
D. Conclusions to Part I .................................................................................... 24
II. Addressing Google’s Power: From Laissez Faire to Antitrust and Regulation 25
A. Introducing Google ...................................................................................... 26
B. Underenforcement ........................................................................................ 29
1. Advertising .............................................................................................. 31
2. Search ...................................................................................................... 33
3. Under-enforcement.................................................................................. 34
C. A Turning Point............................................................................................ 36
D. The Infrastructural Turn in Digital Antitrust ............................................... 37
1. Search ...................................................................................................... 37
2. Advertising .............................................................................................. 38
3. Is Antitrust Changing? ............................................................................ 40
E. The Pro-Competitive Regulatory Turn ........................................................ 41
1. Google as a Common Carrier and Utility? .............................................. 41
2. Toward Pro-Competitive Regulation ...................................................... 43
3. “Light-Handed Pro-Competitive Regulation” ......................................... 44
F. Conclusions to Part II ................................................................................... 45
III. Envisioning and Constructing The Convergence of Antitrust and Utilities in
Digital Platform Markets...................................................................................................... 46
A. Envisioning the Convergence of Antitrust and Utilities .............................. 47
1. The Convergence of Antitrust and Utilities ............................................ 51
2. Lessons from ARD .................................................................................. 52
3. Stretching ARD ....................................................................................... 54
B. Pragmatism in the Digital Platform Economy: Integrating Remedies and
Values ...................................................................................................................... 57
1. Integrating Means: Remedies, Processes, Institutions ............................ 57
2. Integrating Ends: Values and Goals ........................................................ 59
C. Pragmatism in Practice: Markets and Society’s Dynamic Co-Dependence 60

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 3
1. Beyond Individual Choice in Digital Platform Markets: The Limits of
AICOA ................................................................................................................. 60
2. The FTC: Between Antitrust and Consumer Law ................................... 62
Conclusion .................................................................................................................... 65

INTRODUCTION

Digital platform regulation debates are frequently framed in terms of a

trade-off between “regulating” or “breaking-up” Google or Facebook, as if
these were mutually exclusive alternatives.2 Scholars and policy-makers have
distinguished between a Jeffersonian decentralizing and a Hamiltonian
centralizing attitude to digital platforms, or between a preference for
antitrust’s ex post and marginal role on the one hand, and regulation’s more
direct and structural role on the other.3 Framing digital governance in terms
of these or similar binaries is frequent, yet it often obscures the full breadth
and scope of legal possibility in digital markets while producing
impoverished responses to abuses of digital platform power and consumer
harm. It also fails to reflect a nuanced reality where antitrust and regulation
are converging. This Article makes sense of such nuances and introduces a
method for overcoming regulatory siloes between antitrust and other forms
of regulation.
Over the last 50 years, conventional accounts have conceived of antitrust
law as a force that merely corrects defects in naturally competitive
marketplaces, and of other branches of regulation, including utility
regulation, as drastic exceptions that apply to structurally inefficient markets
and industries.4 Today, antitrust agencies increasingly rely on the Sherman

2
See, e.g., Dipayan Ghosh, Don’t Break Up Facebook — Treat It Like a Utility, HARV.
BUS. REV., May 30, 2019, https://hbr.org/2019/05/dont-break-up-facebook-treat-it-like-a-
utility.
3 Frank Pasquale, Tech Platforms and the Knowledge Problem, Vol. II AMERICAN

AFFAIRS, Summer 2018, https://americanaffairsjournal.org/2018/05/tech-platforms-and-

the-knowledge-problem/ (emphasizing trends toward centralization and decentralization in
platform markets); William P. Rogerson & Howard Shelanski, Antitrust Enforcement,
Regulation and Digital Platforms, 168 PENN. L. REV. 1911 (2020) (arguing in favor of pro-
competitive regulation in digital platform markets); Herbert Hovenkamp, Antitrust and
Platform Monopoly, 130 YALE L. J. 1952 (2021) (arguing that antitrust, not regulation is
best suited to govern digital platform markets); OECD Directorate for Financial and
Enterprise Affairs Competition Committee, Ex-Ante Regulation and Competition in Digital
Markets – Note by BEUC (2 December 2021),
https://one.oecd.org/document/DAF/COMP/WD(2021)66/en/pdf (emphasizing ex ante and
ex post nature of enforcement).
4 Stephen G. Breyer, Antitrust, Deregulation, and the Newly Liberated Marketplace, 75

CAL. L. REV. 1005 (1987) (distinguishing antitrust from the notion of economic

Electronic copy available at: https://ssrn.com/abstract=4275143

4 DRAFT [Feb-23

and Clayton Acts to justify durable structural and infrastructural interventions

in digital platform markets.5 At the same time, legislators are devising
statutory frameworks such as the American Innovation and Choice Online
Act (AICOA) aimed at regulating self-preferencing and other anti-
competitive behavior and rendering digital markets more decentralized and
competitive.6 This convergence casts doubts on the frequently emphasized,
but misleading, opposition between ‘antitrust’ and ‘regulation.’ This Article
offers a way of framing the relationship between these areas of law that
overcomes conventional polarities and regulatory siloes, reflecting the
platform economy’s hybrid dynamism and the ubiquitous role of law in
digital settings.
Digital platforms challenge longstanding economic paradigms of
competition and natural monopoly, as well as efficiency-based and
deregulatory justifications for legal intervention (or the absence thereof).
Platforms are at once highly competitive and inherently monopolistic, they
afford a panoply of choices to consumers, but also severely limit persons’
ability to make rational plans and determine their digital lives. Platforms
manage and control marketplaces, defined as spaces for barter and exchange,
as well as infrastructures, defined as shared resources that facilitate the
production of other goods.7 Google, for example, has created valuable
infrastructural goods such as search or maps and manages them as private
marketplaces. Web search is an infrastructural good that provides universal
access to information, yet it is also a marketplace where advertisers, content
producers and users meet and transact, mediated by carefully architected
algorithms. Google’s advertising auctions are marketplaces where
advertising content and space are exchanged, and they are also an
infrastructure that currently sustains the free internet. As such, it is unclear
why regulating web search or online advertising would implicate either
antitrust or regulatory strategies. Regulating these markets requires both

regulation); Herbert Hovenkamp, Antitrust and the Regulatory Enterprise, 2004 COLUMBIA
BUS. L.REV. 335, 336 (2004) (arguing that antitrust and regulation are “complementary
products”).
5
See, e.g., State of New York et al. v. Facebook Inc., No. 1:2020cv03589 (U.S. District
Court for the District of Columbia, filed December 9th, 2020); FTC v. Facebook Inc., No.
1:2020cv03590 (U.S. District Court for the District of Columbia, filed December 9th,
2020); United States of America v. Google LLC, No. DC/1:20-cv-03010 (US District
Court for the District of Columbia, filed October 20, 2020); State of Colorado et al v.
Google LLC, No. 1:2020cv03715 (US District Court for the District of Columbia, filed
December 17, 2020); The State Of Texas, et al v. Google, LLC, No. 4:2020cv00957 (US
District Court for the Eastern District of Texas, filed December 16, 2020).
6
American Innovation and Choice Online Act, S.2992, 117th Cong. (2021-2022)
[hereinafter AICOA].
7
BRETT M. FRISCHMANN, INFRASTRUCTURE: THE SOCIAL VALUE OF SHARED RESOURCES
3-4 (2012) (defining infrastructure).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 5

antitrust and branches of law more attuned to the public dimension of digital
services.
Instead of focusing on doctrinal divides and endogenous disciplinary
concerns, asking for example whether antitrust is “the best tool for the job”
of promoting “competition in digital-platform markets,”8 regulators and
scholars should turn their attention to cross-cutting procedural and normative
questions: How to make sense of the convergence between a panoply of
remedies and resources available to courts and regulators in the digital
economy across and beyond disciplinary boundaries? What normative values
and social goals must drive regulatory design strategies in specific tech
platform contexts? Relying on antitrust and regulation’s ex ante and ex post
procedural affordances and their decentralizing and centralizing tendencies,
regulators might in practice opt for facilitating the decentralization of some
digital functions while centralizing other functions. They might act in both
an ex ante and ex post manner to capture the dynamic nature of digital
innovation. They might be faced with normative conundrums requiring that
they promote privacy, speech and security considerations which can
sometimes constrain competition, individual choice and innovation.9 The
task entails pooling from a range of different remedies and regulatory
resources at once and considering each digital function (e.g. “web search”)
as a situated problem that is not solved generically by appealing to one
remedy (“break-up”), body of law (“antitrust”), or goal (“promoting
competition”), and rejecting another. Further, an integrated approach
demonstrates that regulation does not simply disrupt competitive processes,
undermining choice and innovation. It in fact can promote and help them
flourish.
The Article presents a deconstructive argument followed by a
constructive one. Part I describes the arc of digital platform governance
efforts, offers a critique of the literature on the relation between antitrust and
regulation, and grounds an argument for greater regulatory experimentation
in the need to revisit the public utility as a more versatile concept. Part II
describes the evolution and convergence between antitrust and regulatory
efforts in digital settings, looking at the case of Google’s activities on web

8
Hovenkamp, Platform Monopoly, supra note 3.
9
These normative choices arise not only in relation to digital platforms like Google or
Facebook, discussed in this Article, but also as regards emerging technologies such as the
internet of things. See, e.g., Chris Jay Hoofnagle, Aniket Kesari & Aaron Perzanowski, The
Tethered Economy, 87 GEORG. WASH. L. REV. 782 (2019) (discussing, inter alia, data
flows in the Internet of Things); Peter Swire, The Portability and Other Required Transfers
Impact Assessment: Assessing Competition, Privacy, Cybersecurity, and Other
Considerations, Georgia Tech Scheller College of Business Research Paper No. 3689171,
https://ssrn.com/abstract=3689171 (discussing data portability and FRAND-like
obligations in the Internet of Things).

Electronic copy available at: https://ssrn.com/abstract=4275143

6 DRAFT [Feb-23

search and online advertising. It shows that antitrust and utility-like efforts
have not only always been intertwined, but they are also currently converging
in distinct new ways. In particular, existing lawsuits against Big Tech and
legislative proposals such as AICOA share a concern for pre-structuring
digital markets so as to make them more competitive and contestable. Part III
normatively evaluates the convergence of antitrust and regulatory strategies
in digital platform settings and starts the project of designing more integrated
policy efforts. After introducing a visual framework which illustrates how
antitrust and regulatory efforts are converging in digital settings, Part III
introduces a pragmatic methodology. This helps integrate goals and remedies
and cuts across legal domains such as antitrust, utilities and consumer
protection law. The Article concludes with two examples illustrating why the
design of more integrated institutional efforts is a normative imperative in the
digital economy.

I. DIGITAL PLATFORMS, ANTITRUST EXCEPTIONALISM AND THE

REGULATORY STATE

Asking whether to “break-up” or “regulate” Big Tech, the alleged tension

between antitrust and (utility) regulation, denotes a deregulatory bias. There
are two possible interpretations of this tension. At its weakest, positing a
dichotomy between the two assumes that digital markets are lawless spaces
and that legal intervention – centralizing or decentralizing – is now needed to
restore a prior (unregulated) status quo or to bring about a different regulated
one. At its strongest, the distinction rests on a view of antitrust as a body of
law that promotes, and does not interfere with, the autonomy of competitive
market processes, unlike other forms of regulation such as utilities. This
section presents the context and evolution of digital platform regulation
efforts and describes these two ways of interpreting the relationship between
antitrust and regulation. It also delves deeper into some conventional
accounts of the role of antitrust, the history of utilities and the role of
economic regulation. As argued, classical and neoclassical beliefs in perfect
competition have durably influenced the law’s conception of markets as self-
correcting devices. They have limited regulatory experimentation in digital
markets in ways that perpetuate certain harms.

A. Digital Platforms: Definition, Evolution and the Role of Law

At its inception, the internet was imagined as a utopian decentralized and

anarchic economy that could foster freedom and equality free from the

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 7

coercive power of the state.10 Today, that early promise dating from the 1990s
is demonstrably shortsighted. The digital economy has become a collection
of ‘walled gardens’ dominated by a handful of platform companies such as
Google/Alphabet, Facebook/Meta or Amazon.11 These gatekeepers leverage
their privileged access to data, attention and infrastructural capability to
enclose users and competitors in relations of dependency while offering them
services that are increasingly core to modern life. The Internet of today is
therefore very different from the Internet of the 1990s. It is far less “open”
and decentralized, and far less facially competitive. Small entrepreneurs have
fewer opportunities to compete and create successful digital businesses and
consumers have a more difficult time switching from one platform to a
competing one for the same service.
Contrary to prevalent beliefs, the transition from a relatively utopian
internet of decentralized networks to a relatively dystopian internet of
enclosed platforms, which happened over the course of one, or at best two,
decades, was not the result of law’s absence.12 As Robert Hale explained, law
orders markets through private contract and property law rules.13 Contract,

10
See, e.g., Reno v. American Civil Liberties Union, 521 U.S. 844, 851 (1997) (“a unique
medium-known to its users as ‘cyberspace’ - located in no particular geographical location
but available to anyone, anywhere in the world, with access to the Internet”); John Perry
Barlow, Selling Wine Without Bottles: The Economy of Mind on the Global Net, WIRED
MAGAZINE (1994), https://www.eff.org/pages/selling-wine-without-bottles-economy-mind-
global-net (arguing that digital artifacts should be freed from the constraints of copyright
law); John Perry Barlow, A Declaration of Independence of Cyberspace, Electronic Frontier
Foundation (1996), https://www.eff.org/cyberspace-independence (arguing that cyberspace
must be free from the pressures and laws of states); Yochai Benkler, A Political Economy of
Utopia? 18 DUKE L. & TECH. REV. 78 (2019) (discussing Barlow’s two essays in a
contemporary light).
11
See JONATHAN ZITTRAIN, THE FUTURE OF THE INTERNET – AND HOW TO STOP IT
(2008), Introduction (describing the increasingly gated and enclosed evolution of the
Internet, and asks how we might stop such enclosure trend).
12
See, generally, SHOSHANA ZUBOFF, THE AGE OF SURVEILLANCE CAPITALISM (2019)
(arguing that the growth of companies like Google was the result of an absence of law,
rather than the result of specific legal choices). For critiques of this perspective on law in
digital markets see, e.g., Amy Kapczynski, The Law of Informational Capitalism, Review
of Shoshana Zuboff, The Age of Surveillance Capitalism and Julie Cohen, Between Truth
and Power, The Legal Constructions Of Informational Capitalism, 129 YALE L. J. 1460
(2020), Evgeny Morozov, Capitalism’s New Clothes, THE BAFFLER, February 4, 2019,
https://thebaffler.com/latest/capitalisms-new-clothes-morozov.
13 See Robert L. Hale, Coercion and Distribution in a Supposedly Non-Coercive State, 38

POLITICAL SCIENCE QUARTERLY 470 (1923) (expanding the notion of coercion to show that
it is ubiquitous in property and contractual relations and is an often desirable way to
structure marketplaces). Also see BARBARA FRIED, THE PROGRESSIVE ASSAULT ON
LAISSEZ FAIRE: ROBERT HALE AND THE FIRST LAW AND ECONOMICS MOVEMENT (1998)
(describing Hale’s work and context). And see Yochai Benkler, An Unhurried View of
Private Ordering in Information Transactions, 53 VANDERBILT L. REV. 2063 (2000)

Electronic copy available at: https://ssrn.com/abstract=4275143

8 DRAFT [Feb-23

property and other laws and regulatory modalities indeed shape digital
markets constantly and in a durable way. Faith in individual choice, freedom,
market efficiency, openness and decentralization have not led to less law; in
practice they helped build the legal and intellectual case for new forms of
control.14 New experimental ways to enclose people as assets emerged
around the early 2000s, in response to new financial and legal requirements.
As a result of the dot-com bubble bursting, for example, open platforms such
as Google began to rely on decentralized information sharing infrastructures
to make money from advertising.15 Between 2000 and 2020, the internet then
grew increasingly concentrated, with the most successful platform models –
Google, Facebook, Amazon, Apple to name a few – benefiting from a
virtuous cycle of winner-take-all dynamics and vertical integration. Many of
these companies have become unparalleled transnational powers, deserving
the title “new governors.”16
This evolution prompts two questions. First, what are digital platforms?
And second, what role did law play in digital platforms’ growth and what is
law’s role today? Answering these two questions will surface some
complexities around the antitrust-regulation polarity and prepare the ground
for an experimental approach to integrating antitrust, regulation and utility
regulation in the digital economy.

1. What is a ‘Digital Platform’?

Information platforms are intermediaries that use “data-driven,

algorithmic methods and standardized, modular interconnection protocols to
facilitate digitally networked interactions and transactions among its users.”17

(arguing that enclosing information goods is not per se desirable); Julie E Cohen, Lochner
in Cyberspace: The New Economic Orthodoxy of “Rights Management”, 97 MICH. L. REV.
462 (1998) (showing the distributive politics of rights over information); Margaret J. Radin
& R. Polk Wagner, The Myth of Private Ordering: Rediscovering Legal Realism in
Cyberspace, 73 CHICAGO-KENT L. REV. 1295 (1998) (discussing the limits of laisser faire
and private ordering in digital settings).
14
See JULIE E. COHEN, BETWEEN TRUTH AND POWER 23 (2019) (hereafter TRUTH AND
POWER).
15
ZUBOFF, supra note 12, discussing this shift, however without emphasizing the role of
law.
16
Kate Klonick, The New Governors: The People, Rules, and Processes Governing Online
Speech, 131 HARV. L. REV. 598, 598 (2018). Also see, e.g., Anupam Chander,
Facebookistan, 90 NORTH CAROLINA L. REV. 1807 (2012) (arguing that platforms should
be considered self-standing jurisdictions with a power equivalent to that of states).
17
Julie E. Cohen, Tailoring Election Regulation: The Platform is the Frame, 4 GEO. L.
TECH. REV. 641, 656 (2020). Also see José van Dijck et al., Platformisation, 8 INTERNET
POL’Y REV. 1 (2019) (platforms are “digital infrastructures that facilitate and shape

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 9

The notion of a ‘platform’ encompasses an incredibly diverse array of

technological structures, types of organization and business models, which
range from very large to medium- and small-sized. In this Article, I focus
primarily on Google, and two of its activities: Search and online advertising.
The reason for this choice is simple. Google is a former start-up that
successfully grew to become one of the largest companies on earth. As such,
its existence raises several antitrust and regulatory conundrums, not least as
regards its control of search and advertising markets. Google’s business
model also illustrates a paradox that underlies many platform business
models. Google is at once an operator of marketplaces for private exchange,
a provider of collectively valuable public goods and a self-interested actor
motivated by its own, primarily ad-based, profits. It builds valuable
infrastructure that serves the public and at the same time it makes revenues
by intermediating transactions between advertisers, content and website
owners, and users. As such, Google challenges existing views on the
relationship between competitive markets, monopolies, and key public
resources or infrastructures.
Google possesses six characteristics in common with other ‘platform
companies’ such as Meta, Amazon, Apple, Microsoft, Uber and Twitter.
First, most of them originated in a context that was – materially, legally and
ideologically – radically permissionless, open and decentralized by default.18
Their structure and modes of operation were conditioned by this context.
Second, in spite of their decentralized origins, platform companies have
become centralized proprietary spaces and have transformed the advantages
of openness and decentralization into opportunities for centralization and
power concentration.19 Third, platform-based markets such as web search are
pervaded by network effects and prone to tipping.20 They are winner-take-all
environments that grow after they have reached a certain critical tipping
point. This means that platform business models are either very successful or
fail. There is hardly a competitive state of affairs in which several competing
platform models peacefully coexist. Companies engage in attempts to

personalized interactions among end-users and complementors, organized through the

systematic collection, algorithmic processing, monetization and circulation of data.”)
18
Kevin Werbach, Digital Tornado: the Internet and telecommunications policy, OPP
WORKING PAPER SERIES 17 (1997) (characterizing the internet as such).
19
Digital Platforms Project, Market Structure and Antitrust Subcommittee Report, GEORGE
J. STIGLER CENTER FOR THE STUDY OF ECONOMY AND THE STATE, 11-21 (2019),
https://spinup-000d1a-wp-offload-media.s3.amazonaws.com/faculty/wp-
content/uploads/sites/16/2019/09/marketstructurereport15may2019.pdf.
20
Nicolas Petit & Natalia M. Belloso, A Simple Way to Measure Tipping in Digital
Markets, PROMARKET.ORG, April 6, 2021,
https://www.promarket.org/2021/04/06/measure-test-tipping-point-digital-markets/
(discussing the ‘tipping’ phenomenon in digital markets).

Electronic copy available at: https://ssrn.com/abstract=4275143

10 DRAFT [Feb-23

undermine each other’s ability to compete, raise barriers and costs to entry or
otherwise conclude mutually beneficial but competitively dubious
agreements among themselves. Fourth, platform gatekeeping companies rely
on the capture, use, creation and monetization of data and its algorithmic
processing. This core reliance on data and algorithmic processing raises novel
surveillance issues.21 Fifth, these businesses create and leverage opacity as a
core business strategy.22 Opacity, in turn, can have implications from a
surveillance and non-discrimination perspective. Sixth, platform business
models transform infrastructural resources into markets and vice-versa.
Platform companies manage markets and infrastructures and leverage data,
asymmetries of knowledge and opacity to make money while offering critical
services to the public. They are simultaneously market-makers and
infrastructure providers.23
Each of these characteristics is highly relevant not only to an
understanding of how the platform economy evolved in the way it did, but
also to how one should regulate platforms today.

2. The Role of Law in the Digital Platform Economy

It is frequently thought that law failed to catch on with tech companies’

bottom-up growth, that it moved too slowly or that it is too rigid to address
harm in emerging markets. As Julie Cohen puts it, “[e]ncounters between
networked information technologies and law tend to be framed as examples
of what happens when an irresistible force meets an immovable object” – law

21
See generally Shoshana Zuboff, Big other: surveillance capitalism and the prospects of
an information civilization, 30 JOURNAL OF INFORMATION TECHNOLOGY (2015) (coining
the term surveillance capitalism to describe the exploitative practices of data-driven
business models); SAFIYA UMOJA NOBLE, ALGORITHMS OF OPPRESSION: HOW SEARCH
ENGINES REINFORCE RACISM (2018) (describing the discriminatory impacts of data-driven
recommender systems on people of color); Jose van Dijck, Datafication, dataism and
dataveillance: Big Data between scientific paradigm and ideology, 12 BIG DATA
SURVEILLANCE (2014) (introducing the ideas of dataism and dataveillance to explain the
phenomenon of ubiquitous data collection and life tracking and its discontents).
22
Johanna Gunawan et al., A Comparative Study of Dark Patterns Across Web and Mobile
Modalities, 5 PROC. ACM HUM.-COMPUT. INTERACT. (2021),
https://doi.org/10.1145/3479521 (mapping the use of dark patterns across 150 platforms on
the web and mobile); Gregory Day and Abbey Stemler, Are Dark Patterns
Anticompetitive? 72 ALA. L. REV. 1 (2020) (arguing that digital manipulation and opaque
design should constitute anticompetitive behavior)..
23
José van Dijck, David Nieborg & Thomas Poell, Reframing Platform Power, 8 INTERNET
POLICY REVIEW (2019), https://policyreview.info/articles/analysis/reframing-platform-
power (describing platform power in terms of infrastructural power and co-ependencies);
Elettra Bietti, Self-Regulating Platforms and Antitrust Justice, 101 TEXAS L. REV. 165, 173
(2022) (arguing that platforms are both monopolistic actors, marketplaces and market
makers).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 11

being the immoveable object.24 Yet, as Cohen demonstrated, law is far from
immoveable; it has been there all along, it enables new technologies and
marketplaces to emerge and evolves with them.
The story of law and digital platforms’ (co-)evolution is often told in the
following way.

a. The early internet

For a long time, the digital economy was a decentralized competitive
space that was efficient, fostered plenty of innovation and provided
consumers with many opportunities and choices which would not have been
available to them absent the internet.25 As such, there was not much need for
legal or regulatory intervention. Digital markets allowed consumers and
digital businesses, including platform gatekeepers, to interact and function
properly on their own. They were in fact an example of the way markets ought
to work. They were efficient, frictionless, allowed costless switching and
elicited plenty of new information that could be used to cater to consumers’
needs.26 These markets also fostered freedom and autonomy at unprecedented
scale.27 Today, the story goes, things are different. Platforms have grown and
become entrenched in a way that makes consumer switching very difficult
and that makes it very challenging for small startups and new entrants to
emerge and succeed.28 What were once competitive, free spaces that fostered
innovation and choice are now spaces of enclosure and control that impede
those goals. While previously a lawless and deregulatory approach to the

24
TRUTH AND POWER, Introduction, at 1.
25
Werbach, supra note 18, at 17.
26
Hal R. Varian, Computer Mediated Transactions, AMERICAN ECONOMIC REVIEW:
PAPERS & PROCEEDINGS 100 (May 2010) (showing that computer-mediated transactions
create efficiencies); Hal R. Varian, Beyond Big Data, 49 BUSINESS ECONOMICS (2013)
(exploring market changes due to big data developments); Hal R. Varian, Intelligent
Technology, 53 FINANCE & DEVELOPMENT (September 2016) (celebrating the virtues of
advances in AI and data-driven technologies). Also see VIKTOR MAYER-SCHONBERGER &
THOMAS RAMGE, REINVENTING CAPITALISM IN THE AGE OF BIG DATA (2018)(arguing that
big data brings about an era of perfect information and ideal marketplaces).
27
See Yochai Benkler, Siren Songs and Amish Children: Autonomy, Information, and Law,
76 N.Y.U. L. REV. 23 (2001); YOCHAI BENKLER, THE WEALTH OF NETWORKS (2006) (each
presenting an optimistic, yet cautious, view of the potential of networks to enhance
autonomy). Also see Kiel Brennan-Marquez & Daniel Susser, Privacy, Autonomy, and the
Dissolution of Markets, KNIGHT FIRST AMENDMENT INSTITUTE, August 11, 2022,
https://knightcolumbia.org/content/privacy-autonomy-and-the-dissolution-of-markets (a
revised account of digital markets’ capacity to foster autonomy).
28
See House Committee on the Judiciary, Investigation of Competition in the Digital
Marketplace: Majority Staff Report and Recommendations, October 6th, 2020,
https://judiciary.house.gov/news/documentsingle.aspx?DocumentID=3429 (hereafter
House Report).

Electronic copy available at: https://ssrn.com/abstract=4275143

12 DRAFT [Feb-23

digital economy was justified, today’s concentrated platform economy

requires that law and governments intervene to protect consumers and
competitors.

b. To ‘break-up’ or to ‘regulate’?
It is in this context, and as a sequel to this story, that views on the role of
antitrust and regulatory intervention in the digital platform economy have
recently developed. Law is plural, it comes in many forms. There are many
legal doctrines that can be relied on to regulate digital businesses. Among
these methods and strategies, however, two paths have been singled out as
particularly promising for fighting the growing and threatening power of tech
platform giants. The first consists in returning to the decentralized roots of
the digital economy, disaggregating and breaking-up platforms and enforcing
competition rules so that the digital economy can return to its natural, pre-
legal, and decentralized state, fostering innovation and choice without
requiring durable government involvement. The second consists instead in a
change of course. The deregulatory attitude of the 1990s contributed to a
centralized economy. Re-decentralizing platforms today could lead to
another wave of centralization. Thus, what is needed is a more hands-on
durable form of intervention that takes digital markets as structurally fragile.
Adequate levels of choice, innovation and competition can only be ensured,
under this view, through strong and durable regulatory structuring.
These two approaches seem antithetic because they entail two different
attitudes to digital markets. The first is friendly to the idea of self-correcting
digital markets, the other is skeptical of it. The disagreement on whether to
“break-up” or “regulate” Facebook or Google, at bottom, is a disagreement
on whether self-correcting competitive markets can alone or through punctual
facilitative interventions produce the benefits economic theorists impute to
them.29 The identification of antitrust with a deregulatory digital market
ideology has led technology scholars to cautious skepticism on antitrust and
“break-up” remedies. Scholars have claimed that “splitting Instagram and
WhatsApp off from Facebook… will not solve the most intractable issues
about how to moderate speech...”30 or that “antitrust-based approaches do not
align well with surveillance abuses.”31 One media scholar has even warned
“don’t expect American antitrust to accomplish anything to limit the power

29
See below discussion on perfect competition at I.B.
30
Evelyn Douek, Breaking Up Facebook Won’t Fix Its Speech Problems, SLATE, May 10,
2019, https://slate.com/technology/2019/05/chris-hughes-facebook-antitrust-speech.html.
31
Julie E. Cohen, How (Not) to Write a Privacy Law, KNIGHT FIRST AMENDMENT
INSTITUTE, March 23, 2021, https://knightcolumbia.org/content/how-not-to-write-a-
privacy-law.

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 13

of Facebook or Google any time soon.”32 These perspectives reflect a dismal

faith in the potential of antitrust, and a skeptical view of its remedies.

c. A convergence?
Recognizing the limits of these understandings of antitrust law means
coming to terms with the regulatory state in its complexity and perceiving the
beginnings of a third approach that sits between decentralizing and
centralizing, between antitrust and utility regulation, between bottom-up
competition and top-down regulation. Two parallel moves characterize it. On
the one hand, antitrust lawsuits, which for long were hands-off and timid in
tech platform markets, are increasingly aimed at Big Tech’s infrastructural
power and are designed to attract quasi-regulatory remedies such as
interoperability, break-ups, and non-discrimination obligations. On the other
hand, far from the monolithic conceptions of utilities law that prevail in
energy or transportation industries, regulation in digital markets is being
envisioned as “light-hand” and “pro-competitive.”33 Policy makers seek to
structure digital markets through sectoral law so that they can become more
decentralized, horizontal, and conducive to competition and consumer
choice.34 The convergence between antitrust and regulatory attitudes is both
a response to narrow conceptions of antitrust law and the symptom of a
transformed ‘free markets’ mentality in digital settings, one that recognizes
the imperative to structure markets to enable competitive disruption.

3. Interpreting Convergence

To argue that there is a convergence we must first define the notions of

‘antitrust’ and ‘regulation’. As a matter of positive law, antitrust law is the
universe of US antitrust laws that includes the Sherman Act,35 the Clayton
Act,36 the Federal Trade Commission Act,37 the Hart-Scott-Rodino Act,38 the
Robinson-Patman Act.39 Regulation is a more pluralistic concept with less
precise legal boundaries. It is found in numerous sectoral instruments across

32
Siva Vaidhyanathan, So the government’s antitrust lawsuit against Facebook failed.
Where now?, THE GUARDIAN, June 30, 2021,
https://www.theguardian.com/commentisfree/2021/jun/30/facebook-antitrust-lawsuit-big-
tech?ref=upstract.com&curator=upstract.com&utm_source=upstract.com.
33
See Rogerson & Shelanski, supra note 3.
34
Id.
35
15 U.S.C. § 1 & 2.
36
15 U.S.C. § 13 and following.
37
15 U.S.C. §§ 41 to 58.
38
15 U.S.C. § 18(a).
39
Amending section 2 of the Clayton Act, 15 U.S.C. §§ 13(a) to (f).

Electronic copy available at: https://ssrn.com/abstract=4275143

14 DRAFT [Feb-23

a plurality of industries including transportation, communications, electricity,

and many more.40 While this distinction is hard to contest as a matter of
positive law, the aspirational justifications for distinguishing antitrust from
regulation are more contestable. What does antitrust (or break-up remedies)’s
normative distinctiveness rest on?
Some accounts assume that networked market processes are lawless and
that antitrust and regulation are two different ways for law to insert itself into
networked processes, in pursuit of different, and possibly antithetic aims,
either to re-decentralize or to render these processes more accountable. The
problem is that networks are not in fact lawless. The literature on digital
regulation has demonstrated that digital platforms have grown and thrived
not in the absence of, but because of, norms, technological possibilities and
an intentional stance on law.41 Antitrust and regulatory intervention,
therefore, do not come into play in a pre-legal vacuum. Antitrust law
constitutes (digital) networks and operates as background structure to their
competitiveness together with other frameworks such as intellectual property,
speech or labor law.42 Networks are structured by many kinds of law, with
pre-existing centralizing and decentralizing effects, that are on a continuum
with antitrust and regulatory intervention. It ensues that conceiving of
antitrust and of regulation as ex post and exogenous forms of intervention that
aim at opposing and distinctive goals is misleading.
That is why the antitrust and economic literature engages in more
sophisticated justifications for the normative diction between the two. It
emphasizes the special ideal of competition and holds that competition
(understood as the distributed exchange of promises in unstructured settings)
can be facilitated but not fully shaped by law. The imperative is to let market

40
Some public utility statutes include various Post Office Acts, the Pacific Railroad Act of
1862, the Telegraph Act of 1866, the Federal Reserve Act of 1913, the Federal Water
Power Act, the Radio Act of 1927, the Communications Act of 1934, the Public Utility
Holding Company and the Federal Power Acts of 1935, the Federal Aviation Act of 1958,
the Telecommunications Act of 1996. “Regulation” can also be construed differently and
more expansively. See, e.g., Lawrence Lessig, The New Chicago School, 27 J. LEGAL
STUD. 661, 662 n.1 (1998) (“[By regulation] I mean the constraining effect of some action,
or policy, whether intended by anyone or not. In this sense, the sun regulates the day, or a
market has a regulating effect on the supply of oranges.”).
41
See TRUTH AND POWER (articulating the role of law in structuring and reacting to
technological phenomena in platform settings). Also see Kapczynski, supra note 12; Elettra
Bietti, A Genealogy of Digital Platform Regulation, 7 GEO. L. TECH. REV. 1, 1 (2022)
(describing the evolution of legal attitudes in platform markets).
42
See Sanjukta Paul, Antitrust as Allocator of Coordination Rights, 67 UCLA L. REV. 378
(2020) (framing antitrust law as a body positively shaping markets); Nathan Tankus &
Luke Herrine, Competition Law as Collective Bargaining Law, in CAMBRIDGE HANDBOOK
OF LABOUR IN COMPETITION LAW (2022) (arguing that competition law encompasses labor
laws).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 15

processes play out autonomously by minimizing or limiting legal

intervention. Unlike other forms of law, antitrust creates spaces where
individual preferences can be expressed in a distributed way, bringing about
unplanned yet collectively beneficial states of affairs. Antitrust is thus special
and different from other forms of “regulation” that pre-determine and
structure a market or industry.
This Article now turns to examining this conception of antitrust law’s
distinctiveness, one entrenched in economic law and technology policy
today. It then turns to an overview of infrastructural and utility regulation.
Influenced by classical and neoclassical ideas, scholars have long described
competitive processes as separate or separable from their social causes and
effects.43 Even when they acknowledge that antitrust and regulatory goals and
remedies converge in digital platform markets, many accounts of antitrust
and regulation remain heavily premised on free market optimism and the
need to promote self-correcting competitive processes while adopting a
minimalist stance on legal and regulatory intervention.44 In so doing, even
the most sophisticated accounts fail to recognize that regulation isn’t an
exception, but is the digital economy’s underlying default.

B. Conventional Accounts of Antitrust and Economic Regulation

Over the last 40 years economists and economic law scholars’ attention
has been on pursuing efficiency, competitiveness, innovation and choice.
This ‘free markets’ rationality has led to a decline in utility and sectoral
regulation.45 It has also led to the privatization of numerous industries, e.g.
airlines.46 In this context, antitrust and regulation have been frequently
construed as separate legal domains operating in separate descriptive and

43
Bietti, Self-Regulating Platforms, supra note 23, at 190 (showing that Chicago School
antitrust law in particular is committed to the separability of market efficiency and
distributive considerations).
44
Rogerson & Shelanski, supra note 3; Hovenkamp, Platform Monopoly, supra note 3.
45
WILLIAM NOVAK, NEW DEMOCRACY: THE CREATION OF THE MODERN AMERICAN STATE
112 (2022), (hereafter NEW DEMOCRACY) (“[T]he last half century or so has witnessed a
sustained effort … to undermine and undo the public utility idea. Perhaps aware of the
intimate connection between public utilities and the rise of the regulatory state, two
generations of critics of regulation have taken direct aim at almost every aspect of the
progressive public utility paradigm.”).
46
PAUL SABIN, PUBLIC CITIZENS: THE ATTACK ON BIG GOVERNMENT AND THE REMAKING
OF AMERICAN LIBERALISM (2021) (tracing the roots of 1980s Reaganism in earlier liberal
attitudes to institutions); PAUL STEPHEN DEMPSEY & ANDREW R. GOETZ, AIRLINE
DEREGULATION AND LAISSEZ-FAIRE MYTHOLOGY 179-87 (1992) (criticizing Alfred
Khan’s approach to airline deregulation). Also see Paul Stephen Dempsey, The Rise and
Fall of the Civil Aeronautics Board - Opening Wide the Floodgates of Entry, 11
TRANSPORT. L. J. (1979)(discussing the CAB’s role in the history of airline deregulation).

Electronic copy available at: https://ssrn.com/abstract=4275143

16 DRAFT [Feb-23

normative ways, over separate objects, and often having separate goals.
Dominant conceptions have interpreted antitrust in competitive markets and
privatized industries as a preferred default, and regulation as an exception to
be justified only in rare circumstances. Revisiting the plausibility of some of
these tendencies has significant repercussions for how we might regulate
digital platform ecosystems today and in the future.

1. Competition, monopoly and consumer choice

Microeconomic theory describes competition and monopoly as opposites.

Under conditions tending toward perfect competition, barriers to entry for
supplying a good (e.g. pens) are low or inexistent. Many suppliers can
produce pens and compete on selling them. As more pens get sold, the market
reaches a point where supply meets demand for pens. At that point the market
has reached its allocative potential, meaning that all pens have been optimally
distributed to match sellers and buyers’ preferences (or choices) in the
economy. At the polar opposite end, under natural monopoly, barriers to
entry are very high. Producing the good (e.g. electricity) entails very high
fixed costs which can only be recovered over time, by selling a large quantity
of electricity. Once one producer has incurred the high fixed costs of
producing electricity, it would be inefficient for other producers to enter the
electricity market and incur the same costs. This would drive some or all
producers off the market. It is most efficient for the market to be regulated as
a monopoly: a regulator can centrally ensure that there is one supplier of
electricity, that the supplier is able to recover its costs without charging
excessive prices, and that supply is fair and non-discriminatory.47
Economists have approached competitive markets as markets that
naturally tend toward allocative efficiency, a situation where each consumer
can make their preferred purchasing choice(s). The idea is that, when allowed
to operate competitively and in a decentralized way, markets produce
opportunities for optimal ex post allocation. The informal allocation of
preferences through bottom-up individual action is famously synthesized by
Adam Smith. When a man acts selfishly intending “only his own gain” he in
fact acts as if “led by an invisible hand to promote an end which was no part
of his intention,” which in turn benefits society.48 On the other hand, natural
monopolies are not conducive to significant consumer choice. In
monopolistic industries the government or regulator must determine which

47
William J. Baumol, On the Proper Cost Tests for Natural Monopoly in a Multiproduct
Industry, 67 THE AM. ECON. REV. 809 (Dec., 1977) (defining natural monopoly).
48
ADAM SMITH, THE WEALTH OF NATIONS (1776), Book I Chapter VII (describing the
“invisible hand” mechanism). Also see ADAM SMITH, THE THEORY OF MORAL SENTIMENTS
(1759), Part IV, Chapter 1.

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 17

goods or services must be produced (and by whom), and mandate or constrain

their production. Choice is limited in an ex ante way by what the regulator
has determined to be good for society.
It must be noted that the concepts of perfect competition and natural
monopoly are ideal simplified models and are contested among economists.49
Modern economic theory recognizes a multiplicity of nuanced hybrids that
sit in-between perfectly competitive and naturally monopolistic industry
structures.50 What matters for our purposes is that antitrust law and policy as
well as regulation and utility regulation have developed to reflect ideal
theories of competition and monopoly, often more starkly than economic
theories themselves.51 They have been strongly influenced by paradigms of
perfect competition and natural monopoly and have embedded these ideals
into real-world productive settings, including technology markets. In the last
50 years, antitrust and utility regulation have been claimed by a tradition of
classical and neoclassical economic policy that treats competition as the
default mode of production and resource allocation and monopoly as an
exception to be fought (or tolerated where efficient).52
The paradigmatic role of antitrust law was expressed in Northern Pacific
Railway,

[t]he Sherman Act was designed to be a comprehensive charter of economic liberty

aimed at preserving free and unfettered competition as the rule of trade. It rests on
the premise that the unrestrained interaction of competitive forces will yield the best
allocation of our economic resources, the lowest prices, the highest quality and the
greatest material progress, while at the same time providing an environment
conducive to the preservation of our democratic political and social institutions.53

Friedrich Hayek also famously defended free and unfettered competition

on the ground that public institutions lack the knowledge to design optimal

49
HAL R. VARIAN, INTERMEDIATE MICROECONOMICS: A MODERN APPROACH, 1 (8th Ed,
2010); and see generally CRANE & HOVENKAMP, infra note 51 (illustrating contestations on
the relationship between state and market as regards competition over the last century).
50
See, e.g., MASSIMO MOTTA, COMPETITION POLICY: THEORY AND PRACTICE 75-89 (2004)
(discussing network effects, non-physical barriers to entry and other characteristics that sit
in-between competitive markets and natural monopolies). WILLIAM BAUMOL, JOHN C.
PANZAR AND ROBERT D. WILLIG, CONTESTABLE MARKETS AND THE THEORY OF
INDUSTRIAL STRUCTURE (1982) (on the contestation of monopolies over time).
51
DANIEL A. CRANE AND HERBERT HOVENKAMP, THE MAKING OF COMPETITION POLICY:
LEGAL AND ECONOMIC SOURCES 1, 71, 125 (2013) (describing the influence of neoclassical
and other economic theories on competition policy over time).
52
Id. See also Sanjukta Paul, Beyond Neoclassical Antitrust, BOSTON REVIEW, June 23,
2022, https://bostonreview.net/forum_response/beyond-neoliberal-antitrust/ (denouncing
the over-emphasis on perfect competition economic law and policy).
53
N. Pac. Ry. Co. v. United States, 356 U.S. 1, 4 (1958).

Electronic copy available at: https://ssrn.com/abstract=4275143

18 DRAFT [Feb-23

regulatory interventions.54 Instead, distributed competitive markets enable

people to express their individualized needs and preferences free from
planned regulatory intervention or coercion, thus achieving the collective
good.55
Digital platform markets have been and continue to be shaped by a belief
in letting individual preferences govern otherwise autonomous competitive
markets. These beliefs have produced narrow constructions of antitrust and
regulatory intervention, and a minimalistic approach to law in productive
settings.

2. Antitrust and regulation as friends and foes

Influenced by neoclassical theories of competition and monopoly,

scholars interpret antitrust law and economic regulation, including public
utility regulation, as frenemies.56
The core case of public utility regulation, at least in theory, is narrow and
discrete: public utility regulation is required in areas where production
follows a natural monopoly model and competitive dynamics do not apply.57
Beyond this core case where competition and therefore antitrust allegedly
have no role to play, regulation, understood as an exogenous force that is
implanted into markets, is generally justified as something that can be layered
on or can replace a default competitive state of affairs only in cases of
extreme market inefficiencies or externalities. Former Supreme Court Justice
Stephen Breyer, for example, mentions possible market “defects” that justify
regulatory intervention while emphasizing competition and antitrust as the
default:

the classicist embraces regulation faute de mieux. Competition is more desirable,

and antitrust may help maintain competition. Yet for one reason or another, in these
special markets, competition cannot work or by itself is inadequate. Thus, one must
turn to regulation as a supplement or substitute.58

54
FRIEDRICH A. HAYEK, THE CONSTITUTION OF LIBERTY 29-30 (1960) (making the case
for deregulation).
55
Id.
56
For a different use of the word “frenemies,” see ARIEL EZRACHI & MAURICE E. STUCKE,
VIRTUAL COMPETITION, THE PROMISE AND PERILS OF THE ALGORITHM-DRIVEN ECONOMY
1, 147 (2016).
57
See ALFRED KHAN, THE ECONOMICS OF REGULATION: PRINCIPLES AND INSTITUTIONS, 2
(Vol. 1, 1970). However, see NOVAK, NEW DEMOCRACY, at 123, (citing Bruce Wyman,
The Law of Public Callings as a Solution to the Trust Problem, 17 HARV. L. REV. 157
(1904): “monopoly was just one of many important factors that could justify a public
utility”) in .
58
Breyer, supra note 4, at 1007. Also see Hovenkamp, supra note 4, at 339 (noting
information asymmetries as a key case of externality justifying regulatory intervention).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 19

There are two conventional ways of understanding the way antitrust

relates to other regulation in the economy. One approach sees them as
substitutes, the other as complements. The two approaches are not necessarily
in opposition and broadly reflect the way US law construes the relation
between antitrust and other regimes, as bodies of law that exist side-by-side
and sometimes pre-empt or impede each other’s applicability.59 Recently, the
tendency of courts has been to interpret antitrust and regulation as mutually
exclusive substitutes, even in the presence of an express antitrust “savings
clause”.60
The first approach is to view antitrust and regulation as competing
alternatives or substitutes that aim at the same goals, with antitrust generally
taking priority over regulation. Stephen Breyer describes classical theories of
antitrust and economic regulation as having similar pro-competitive goals
including prices close to incremental costs, efficient production and
innovation.61 One can include consumer choice among these goals.

The case of digital markets seems a case in point, see Rogerson & Shelanski, supra note 3,
at 1914 (arguing for supplementing general purpose digital antitrust actions with sectoral
regulation).
59
The relationship is governed by three main legal regimes in the United States. First, the
so-called Noerr-Pennington doctrine immunizes petitioners of legislation from antitrust
scrutiny. See R.R. Presidents Conference v. Noerr Motor Freight, Inc., 365 U.S. 127
(1961); United Mine Workers v. Pennington, 381 U.S. 657 (1965). Second, the state action,
or Parker, doctrine immunizes conduct governed by state and local regulations from
federal antitrust law. See Parker v. Brown, 317 U.S. 341 (1943). See also Hovenkamp,
supra note 4, at 347. (“[i]f the state passes a statute that unambiguously calls for potato
price regulation, and a potato agency created for this purpose actively administers the rate-
making process, then the antitrust court has no choice but to find the regulation immune
from the antitrust laws.”) Third, federal law can expressly exempt conduct in a given
industry from antitrust scrutiny through a preemption or immunity clause (e.g. Webb-
Pomerene Act, 15 U.S.C. §62 (2006), exempting collective export associations from
antitrust liability), it can expressly preserve antitrust scrutiny through a savings clause
(Telecommunications Act of 1996, 47 U.S.C. § 152 expressly preserving antitrust law), or
it can impliedly exempt antitrust. Regarding implied exemption, until 2004 the law
interpreted antitrust and regulation as compatible unless there was “plain repugnancy
between the two.” See United States v. Philadelphia Nat'l Bank, 374 U.S. 321, 350-1
(1963) (“repeals of the antitrust laws by implication from a regulatory statute are strongly
disfavored, and have only been found in cases of plain repugnancy between the antitrust
and regulatory provisions.”); and Otter Tail Power Co. v. United States 410 U.S. 366
(1973) (where the objectives of the Federal Power Act and of antitrust law were held to be
aligned). However, as argued in Howard Shelanski, The Case for Rebalancing Antitrust
and Regulation, 109 MICH. L. REV. 683 (2011), Trinko and Credit Suisse loosened the
standard of implied immunity making the joint application of antitrust and regulation more
difficult to justify. See Verizon Commc’ns Inc. v. L. Offs. of Curtis V. Trinko, LLP 540
U.S. 398 (2004); Credit Suisse Sec. LLC v. Billing, 551 U.S. 264 (2007).
60
See Trinko, 540 U.S. 398 (2004); Credit Suisse, 551 U.S. 264 (2007).
61
Breyer, supra note 4, at 1006.

Electronic copy available at: https://ssrn.com/abstract=4275143

20 DRAFT [Feb-23

Regulation aims at these goals directly, “bypass[ing] the competitive

process” and typically entails an administrative agency and “network[s] of
rules that determine the regulated firms' prices while, in principle, spurring
those firms toward innovation and production efficiency.”62 Antitrust instead
operates indirectly. “The antitrust laws set forth a few negatively phrased
directives, which are enforced by the courts or the Federal Trade
Commission. … [T]hey promote competition so that competition itself can
bring us its economic benefits.”63
The picture that emerges is one where antitrust and regulation have
alternative functions and antitrust is preferred:

[A]ntitrust is not another form of regulation. Antitrust is an alternative to regulation

and, where feasible, a better alternative. … the classicist first looks to the
marketplace to protect the consumer; he relies upon the antitrust laws to sustain
market competition. He turns to regulation only where free markets policed by
antitrust laws will not work—where he finds significant market "defects" that
antitrust laws cannot cure.64

Another tendency is to interpret antitrust and regulation as

complementary, that is as two separate bodies of doctrine with distinct
functions that can operate side-by-side, while tackling distinct harms or
objects of regulation. Herbert Hovenkamp, for example, makes the case for
antitrust and regulation as “complementary products.”65 He too argues that
most markets are efficient and tend toward competition, but some exceptional
markets “fail to clear at efficient levels” and require regulatory intervention.66
He then argues that antitrust should be understood as a “residual regulator”
that operates in the zone of free, unregulated or deregulated markets, in the
space where regulation is unnecessary or in retreat:

When the government makes rules about price or output, market forces no longer
govern. To that extent antitrust is shoved aside. A corollary is that as an industry
undergoes deregulation, or removal from the regulatory process, antitrust re-enters
as the residual regulator. … .67

62
Id.
63
Id.
64
Id, at 1007.
65
Hovenkamp, supra note 4, at 336; Herbert Hovenkamp, Antitrust and Regulation Over
Time, The Regulatory Review, Oct 1, 2020,
https://www.theregreview.org/2020/10/01/hovenkamp-antitrust-regulation-over-time/. Also
see Howard Shelanski, The Case for Rebalancing Antitrust and Regulation, 109 MICH. L.
REV. 683 (2011) (arguing against exemptions of antitrust in regulated sectors); Rogerson &
Shelanski, supra note 3, at 1914 (arguing for supplementing general purpose antitrust with
sectoral regulation in technology markets).
66
Hovenkamp, The Regulatory Enterprise, supra note 4, at 336.
67
Id. at 341.

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 21

Antitrust law takes a market's regulatory structure as given, warts and all, and tries
to prevent injuries to competition that the regulatory process leaves untended. …
[A]ntitrust's role is "residual." It picks up only where regulation leaves off.68
(emphasis added)

What emerges from Hovenkamp’s analysis is the idea that antitrust

operates as the law of the free market and operates in a pre-structured
marketplace. It intervenes to correct the operation of the “free market” as an
imagined structure that is whole and separable from other “regulated” aspects
of the economy. Harm to competition is not like any other market failure, it
is a specific kind of harm that requires a specific form of after-the-fact judicial
intervention. No matter whether they consider antitrust and sectoral economic
regulation as substitutes or as complements, these conventional approaches
emphasize the conceptual (or normative) separability and preference for
antitrust, over other forms of regulation such as utilities law. In so doing, they
favor an acontextual ideal of decentralized competitive marketplaces that
autonomously afford choices to consumers.

C. The Public Utility as Exception to the Free Market Default

The idea of the public utility is in tension with the conventional

preference for antitrust. The utility originates in a nineteenth century
bureaucratic experiment in democratizing control over industrial processes.69
The intent was to govern certain unaccountable industry segments, functions,
and private infrastructures in collectively accountable ways. Utilities
frontally challenge the classicist preference for deregulatory competition as a
vehicle for attaining consumer choice and innovation. In the case of Munn v.
Illinois, decided in 1876, the Supreme Court established a “constitutional
foundation for the… law of public utilities.”70 It upheld a fine against the
owners and operators of a grain warehouse who had violated the licensing
and rate-setting requirements imposed by an Illinois regulation.71 Speaking

68
Id. at 342.
69
WILLIAM NOVAK, NEW DEMOCRACY, at 121. Also see JOHN DEWEY, THE PUBLIC AND
ITS PROBLEMS 109 (1927) (the democratic state is dedicated to “the utilization of
government as the genuine instrumentality of an inclusive and fraternally associated
public.”); FELIX FRANKFURTER, PUBLIC AND ITS GOVERNMENT 81 (1930) (emphasizing the
importance of government’s role “in securing for society those essential services which are
furnished by public utilities.”).
70
William J. Novak, The Public Utility Idea and the Origins of Modern Business
Regulation, in N. R. LAMOREAUX AND W. J. NOVAK (EDS.), THE CORPORATION AND
AMERICAN DEMOCRACY (2017), at 170 (presenting a history of utilities as paradigms of
regulatory experimentation).
71
Munn v. Illinois, 94 U.S. 113 (1876).

Electronic copy available at: https://ssrn.com/abstract=4275143

22 DRAFT [Feb-23

for the majority, Justice Waite stated that:

Property does become clothed with a public interest when used in a manner to make
it of public consequence and affect the community at large. When, therefore, one
devotes his property to a use in which the public has an interest, he, in effect, grants
to the public an interest in that use, and must submit to be controlled by the public
for the common good, to the extent of the interest he has thus created.72

Classicist and neoclassicist influences led to 50 years of critiques of

utility regulation.73 For example, Richard Posner argued that regulating the
natural monopolies of today “is to gamble dangerously with the future.”74
Since the benefits of regulation were overall “dubious,” Posner advocated for
repealing most if not all public utility regulation and common carrier
legislation.75 Harold Demsetz similarly questioned natural monopoly
justifications for public utility regulation suggesting that observed
concentration did not always directly cause monopoly prices and could be
efficient.76 Subsuming many of these views, Rick Geddes noted in 2000 that
“[w]here once regulated or government-owned monopolies dominated
because of the belief that most utilities were ‘natural monopolies,’ there is
now a growing consensus that competition can perform a broader and more
effective role.”77
If anything, the effect of these critiques has been to limit appetite for
regulatory experimentation and to atrophy intellectual debates on law and
regulation’s role in the economy. As a result of these and other skepticisms,
public utility thinking today is confined to narrow segments of the economy
and the role of the regulatory state in governing collectively valuable digital
assets in under-explored. Critiques notwithstanding, the public utility
remains an important repository of strategies, remedies and ideas that should
not be dismissed in designing future digital platform regulation efforts.

1. Infrastructure

72
Munn, at 126.
73 NEW DEMOCRACY, at 112 (“[T]he last half century or so has witnessed a sustained effort
… to undermine and undo the public utility idea. Perhaps aware of the intimate connection
between public utilities and the rise of the regulatory state, two generations of critics of
regulation have taken direct aim at almost every aspect of the progressive public utility
paradigm.”).
74
Posner, infra note 154, at 635-6.
75
Id. at 638.
76
Harold Demsetz, Why Regulate Utilities?, 11 J. OF L. AND ECON. 55, 61 (1968) (the case
against utility regulation).
77
Rick Geddes, Public Utilities, in ENCYCLOPEDIA OF LAW AND ECONOMICS: THE HISTORY
AND METHODOLOGY OF LAW AND ECONOMICS 1162, 1163 (2000).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 23

Platform companies like Google are market-makers and infrastructure-

providers. Brett Frischmann defines infrastructure as “a large-scale physical
resource made by humans for public consumption”78 or as “shared means to
many ends.”79 Infrastructures are material facilities, often created or owned
by privates, that are managed so as to further collectively valuable ends.
Roads and electric power grids, for example, play essential roles as inputs
into a variety of downstream goods, as do less tangible resources such as
linguistic and scientific conventions.80 Some infrastructures, like public
parks, are managed through public provision; others, like Wikipedia, are
managed by members of a community or commons; yet others are privately
provided; and finally there is infrastructure like local electricity supply which
is managed as a utility by local utility commissions. Because building and
managing infrastructure can be costly, one way to ensure that it is built and
managed efficiently is to authorize a private entity to do so under regulatory
supervision, aka the public utility.

2. Defining Public Utility Regulation

The existing body of antitrust laws, passed by Congress to protect

competition and prevent monopolization,81 encompasses laws and policies
whose goal is “to perfect the operation of competitive markets”82 and “ensure
that competition in the marketplace is not restricted in a way that is
detrimental to society.”83 Public utilities law is a more open-ended ensemble
of arguably ex ante regulatory frameworks. It emanates from a plurality of
sectoral instruments across many industries including transportation,
communications, or electricity, and structures industrial processes more
durably.84
Thomas Hardman defines the utility through three primary
characteristics.85 First, there must be a naturally monopolistic business “in
the sense that competition cannot be relied upon to secure the public or social
interest in having reasonable rates and adequate service, service without

78
BRETT M. FRISCHMANN, INFRASTRUCTURE: THE SOCIAL VALUE OF SHARED RESOURCES
3 (2012) (defining infrastructures).
79
Id. at 4.
80
TRUTH AND POWER, at 23.
81
Standard Oil Co. v. FTC, 340 U.S. 231, 249 (1951) (“Congress was dealing with
competition, which it sought to protect, and monopoly, which it sought to prevent.”).
82
Frank H. Easterbrook, Limits of Antitrust, 63 TEX. L. REV. 1, 1 (1984) (delineating
antitrust law’s under-enforcement virtues and limits).
83
MOTTA, supra note 50, at 30.
84
See note 40.
85
Thomas P. Hardman, Public Utilities: I. The Quest for a Concept, 37 W. VA. L. REV. 250
(1931) (defining public utilities).

Electronic copy available at: https://ssrn.com/abstract=4275143

24 DRAFT [Feb-23

discrimination…”86 It must be more efficient or socially beneficial to have

one regulated provider rather than many competitors. Second, there must be
a holding out to serve the public at large which is different from a promise
under a private contractual agreement to serve a class of people. Third, the
business must be “essential” to the general welfare, that is “not merely useful
or advantageous … such as theatres, skating rinks, … but … reasonably
“necessary”” to a public.87
These features are neither necessary nor sufficient. From a legal
perspective it seems that neither the natural monopoly nor public purpose nor
essentiality are necessary or sufficient to determine the existence of public
utility regulation.88 Instead, a utility is a business that has de facto been
subjected to certain formalities such as administrative rulemaking and
adjudication on price or standards,89 permission or licensing to enter/exit an
industry,90 mandates to provide service on a fair and non-discriminatory
basis,91 and often the presence of a supervisory utility commission or agency.
Morgan Ricks and co-authors identify a list of features that characterize what
they call “Networks, Platforms and Utilities” law: non-discrimination rules,
rate setting, profit sharing, equal access rules, universal service requirements,
exit restrictions, entry restrictions, interconnection mandates, structural
separations, and ownership and control restrictions.92 As will be discussed,
many of these remedies also overlap with antitrust remedies and are very
useful to draw from in digital settings.

D. Conclusions to Part I

The analysis in Part I shows that the relationship between antitrust law

86
Id. at 251.
87
Id. at 251.
88
See, e.g., Pub. Util. Comm’ v. Texas Tel. Ass’n., 163 S.W.3d 204 (Tex. App. 2005)
where a “public interest” determination was held to be necessary to receive public funding
but not to compete in an industry. There are many non-naturally monopolistic industries
that are nonetheless the subject of heavy government regulation that serves public
purposes. Similarly, there are limits to what “public interest” regulation entails in a
naturally monopolistic industry, see, e.g., NAACP v. Federal Power Comm’n, 425 U.S.
662 (1976), New York Cent. Sec. Corp., v. U.S., 287 U.S. 12 (1932).
89
FPC v. Hope Nat. Gas Co., 320 U.S. 591 (1944); William Boyd, Just Price, Public
Utility, and the Long History of Economic Regulation in America, 35 YALE J. ON REG. 721
(2018) (a genealogy of price-setting and utilities law).
90
New State Ice v. Liebmann, 282 US 262 (1932).
91
See, e.g., American Hoechest Corps. v. Dept. of Pub. Utils., 379 Mass. 408 (1980).
Mountain States Legal Foundation v. Utah Pub. Serv. Comm’n, 636 P.2d 1047 (1981).
92
MORGAN RICKS, GANESH SITARAMAN, SHELLEY WELTON & LEV MENAND, NETWORKS,
PLATFORMS, AND UTILITIES LAW AND POLICY (2022), Chapter 1 (textbook on regulation).
Some of these remedies are discussed in greater detail below.

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 25

and other forms of regulatory intervention, such as utilities, in digital

platform markets is more contested than might first appear. A reductive
interpretation conceives of platforms as lawless spaces which require either
discrete re-decentralization through antitrust or durable restructuring through
utility regulation or central planning. This conception of antitrust and utility
regulation as discrete forces that are introduced after-the-fact into a pre-
structured economy is not tenable because both domains structure digital
platforms ubiquitously. A stronger interpretation that dominated the antitrust
and economic regulation literature in the last fifty years claims that antitrust
allows enforcers to promote competition indirectly, by facilitating markets’
autonomous operation while intervening only at the margins. Regulation
instead is a forceful type of direct intervention that disrupts the autonomy of
self-regulating market processes. This interpretation acknowledges that
antitrust and regulation both pre- and re-structure markets constantly. Yet it
also exhibits a deregulatory bias, viewing antitrust as superior to other
branches of law because of its light-handed, ex-post, marginal role. This
preference for competition, consumer choice and the autonomy of market
processes obscures the rich history of regulatory experimentation with
utilities in various industries and fails to reflect the regulatory default in
(technology) markets. As such, current conceptions of the relations between
antitrust, economic regulation and utilities law make digital regulators’ job
more limited and ineffective than it could otherwise be.
As will now be shown, antitrust enforcement and infrastructural or utility
regulation are “converging” in platform settings. Far from operating on
different subject-matters or from having different normative functions, they
both contribute to structuring digital markets in a durable, ex ante way.
Claims about the distinctive role of antitrust are thus, at bottom, normative
claims. They are attempts at keeping market processes separate from
deliberations about social needs, social harms and democratic values.

II. ADDRESSING GOOGLE’S POWER: FROM LAISSEZ FAIRE TO

ANTITRUST AND REGULATION

Part I showed that the frequently emphasized polarity between breaking-

up or regulating Big Tech, and between antitrust and utility regulation more
generally, is driven by a belief in antitrust’s special role in self-correcting
markets. The strongest justification for antitrust exceptionalism, indeed, rests
on faith in the separateness of competitive market processes from social
needs and values.
Relying on Google as a case study, Part II now discusses how antitrust
agencies, regulators and scholars have construed Google’s power, revealing
blind spots in the conception and governance of digital markets. Far from

Electronic copy available at: https://ssrn.com/abstract=4275143

26 DRAFT [Feb-23

antitrust exceptionalism, Google’s story illustrates a progressive move from

a laissez-faire approach to law in markets toward increasing consciousness
about controlling infrastructural power. The State’s role in digital
environments has evolved from that of a negative preserver of autonomous
markets and their inherent efficiencies to that of a conscious designer of
competitive and decentralized digital economies. This shift consciously
opens digital markets to competition and blurs the lines between the mission
of antitrust and that of other legal domains such as utilities law. Antitrust
lawsuits and sectoral reforms appear motivated by a similar concern to pre-
structure markets and enable distributed choice and competition. Whether
consciously or not, therefore, agencies and courts are engaging in a holistic
effort to advance competition and choice in digital environments.
Yet while there is broad consensus that law must promote choice and
competition,93 there is little agreement on these notions’ relation to other
values and policy goals such as privacy, equality, or healthy public sphere.
Yet if markets are structured by law, they must necessarily affect and be
affected by extrinsic social dimensions. Even at a time of regulatory
experimentation and appetite for fighting digital power, therefore, the
emphasis on choice and competition pulls us backward. It redirects policy
efforts away from creative digital policy responses and toward the classicist
preference for self-correcting markets.

A. Introducing Google

Google/Alphabet is one of the biggest internet companies in existence

today. Its operations span five continents and its numerous and diversified
products and services affect people’s daily pursuits, education, healthcare,
professional life and more. Google launched in 1998 as a general search
engine.94 By June 2000, Google had indexed 1 billion web pages and in
October of that same year it had launched AdWords, an online advertising
93
As will be discussed, many antitrust scholars are now recognizing the need for “pro-
competitive” regulation. See, e.g., Randal C. Picker, What Should We Do About The Big
Tech Monopolies?, 1 TECHREG CHRONICLE 27 (2021),
https://www.competitionpolicyinternational.com/wp-content/uploads/2022/02/TechREG-
INAUGURAL-EDITION-REGULATING-DIGITAL-ECONOMY-DECEMBER-2021.pdf
(defending regulation in platform settings); Rogerson & Shelanski, supra note 3 (arguing
for supplementing general purpose antitrust actions in technology markets with sectoral
regulation); Filippo Lancieri & Caio M.S. Pereira Neto, Designing Remedies for Digital
Markets: The Interplay Between Antitrust and Regulation, J. OF COMP. L. & ECON. 1, 32
(2021) (“the growing interplay between antitrust and regulation in digital markets will
likely lead to a system of overlapping jurisdictions that blurs … administrative
boundaries.”)
94
Google Inc., Registration Statement (Form S-1) 1 (Apr. 29, 2004),
https://www.sec.gov/Archives/edgar/data/1288776/000119312504073639/ds1.htm.

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 27

service that would soon become the backbone of its business model.95
Between 2000 and 2020, Google acquired more than 260 companies,96
including user-generated video company YouTube in 2006,97 adtech player
DoubleClick in 2007,98 Motorola Mobility in 2011,99 Waze100 and
DeepMind.101 In 2015, Google became a wholly owned subsidiary of its
parent company Alphabet.102 Today Google counts at least nine products with
more than a billion users each: Search, Gmail, Android, Chrome, Maps,
Drive, Photos, Play Store and YouTube.103 In 2019, Google reported a total
revenue of $160.7 billion, 83% of which was derived from ads.104 These
revenues were as high as Hungary’s GDP that same year, according to the
World Bank indicator,105 and have continued to grow in 2021.106

95
House Report, supra note 28, at 174.
96
Id. See also an updated list of publicly known acquisitions here:
https://en.wikipedia.org/wiki/List_of_mergers_and_acquisitions_by_Alphabet#cite_note-1.
97
Andrew Ross Sorkin and Jeremy W Peters, Google to Acquire YouTube for $1.65
Billion, THE NEW YORK TIMES, October 9, 2006,
https://www.nytimes.com/2006/10/09/business/09cnd-deal.html.
98
Louise Story and Miguel Helft, Google Buys DoubleClick for $3.1 Billion, THE NEW
YORK TIMES, April 14, 2007,
https://www.nytimes.com/2007/04/14/technology/14DoubleClick.html.
99
Hayley Tsukayama, Google agrees to acquire Motorola Mobility, THE WASHINGTON
POST, August 15, 2011, https://www.washingtonpost.com/blogs/faster-
forward/post/google-agrees-to-acquire-motorola-
mobility/2011/08/15/gIQABmTkGJ_blog.html.
100
Ingrid Lunden, Google Bought Waze For $1.1B, Giving A Social Data Boost To Its
Mapping Business, TECHCRUNCH, June 11, 2013, https://techcrunch.com/2013/06/11/its-
official-google-buys-waze-giving-a-social-data-boost-to-its-location-and-mapping-
business/.
101
Kwame Opam, Google buying AI startup DeepMind for a reported $400 million, THE
VERGE, January 26, 2014, https://www.theverge.com/2014/1/26/5348640/google-
deepmind-acquisition-robotics-ai.
102
Letter from Larry Page, CEO, Alphabet Inc., and Sundar Pichai, CEO, Google LLC
(2015), https://abc.xyz/investor/founders-letters/2015/index.html#2015-larry-alphabet-
letter.
103
Harry McCracken, How Google Photos joined the billion-user club, FAST CO, July 24,
2019, https://www.fastcompany.com/90380618/how-google-photos-joined-the-billion-
user-club.
104
Alphabet Inc., Annual Report (Form 10-K), February 3, 2020, at 26–30,
https://www.sec.gov/Archives/edgar/data/1652044/000165204420000008/goog10-
k2019.htm.
105
See the World Bank Indicator here:
https://data.worldbank.org/indicator/NY.GDP.MKTP.CD?year_high_desc=true.
106
Google/Alphabet’s total revenues in 2021 were 257.637 billion dollars. Google’s
Annual Report explains the figures as follows: “Revenues were $257.6 billion, an increase
of 41%. The increase in revenues was primarily driven by Google Services and Google
Cloud. The adverse effect of COVID-19 on 2020 advertising revenues also contributed to

Electronic copy available at: https://ssrn.com/abstract=4275143

28 DRAFT [Feb-23

My focus here is on what the company calls “Google Services”107 and

particularly on two Google businesses: web search and display advertising.
The reason for this choice is that web search is Google’s original business
segment and advertising is Google’s primary source of revenue. Both are key
sectors of activity and dominance for Google and loci of regulatory scrutiny
from an antitrust perspective and beyond.
Google’s Search business was built around Sergey Brin and Larry Page’s
“Page Rank” algorithm. It made web search possible by ranking search
results according to how much their were linked to on the web.108 The most
linked-to pages would feature more prominently at the top of search
results.109 Google’s success when it launched in 1998 was almost immediate.
Today, however, searching the web often results in lengthy lists of paid
results and advertisements. What was once a magical experience now “feels
monetized to death, soulless, and exhausting.”110 Google’s search business
has benefited consumers and society. Yet looking forward, this critical
societal function cannot be left solely in the unregulated hands of an ad-
driven private company.111
“Google has become the controlling node and the central authority for
advertising, which serves as the primary currency enabling a free and open
internet.”112 Online advertising is the primary mechanism through which
digital services, including search, are funded and is what allows for actors
across the digital economy to sustain themselves. If the online advertising
ecosystem fails, this could bring down the digital economy as we know it.
Online advertising markets are concentrated and opaque. Google dominates
online search and display advertising. On the search advertising market,

the year-over-year growth.” Alphabet Inc., Annual Report (Form 10-K), Feb. 1st, 2022, at
32, https://abc.xyz/investor/static/pdf/20220202_alphabet_10K.pdf?cache=fc81690.
107
Which includes ads, Search, Android, Chrome, Gmail, Maps and YouTube.
108
Sergey Brin & Larry Page, The anatomy of a large-scale hypertextual Web search
engine, 30 COMP. NETWORKS AND ISDN SYSTEMS (1998),
http://infolab.stanford.edu/pub/papers/google.pdf (detailing what would subsequently
become the Google "Search Rank” algorithm).
109 Id.
110
Charlie Warzel, The Open Secret of Google Search, THE ATLANTIC, June 20, 2022,
https://www.theatlantic.com/ideas/archive/2022/06/google-search-algorithm-
internet/661325/ (criticizing Google Search).
111
It is worth noting that Brin and Page themselves recognized that advertising would favor
economic interests over the interests of users. See Brin & Page, supra note 108, Appendix
A (“The goals of advertising business models do not always correspond to providing
quality search to users. … we expect that advertising funded search engines will be
inherently biased towards the advertisers and awat from the needs of consumers. … we
believe that [advertising] causes enough mixed incentives that it is crucial to have a
competitive search engine that is transparent and in the academic realm.”).
112
Second Amended Complaint at 7, In re Google Digital Advertising Antitrust Litigation,
State of Texas AG et al. v. Google LLC, No. 1:21-md-03010-PKC (S.D.N.Y. 2021).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 29

advertisers pay Google to link their advertised content to specific search

words or phrases so that advertised content is presented to users in response
to relevant search queries.113 On the display advertising market, Google
controls advertisers’ ability to place ads across the web. The process of
selling and placing ads on web and mobile includes real-time ad exchanges
(for large advertisers, relying on real-time bidding mechanisms) and ad
networks (for smaller advertisers). Google operates a large and vertically
integrated in-house trading platform which covers many aspects of this
process.114 Google intermediates on both the advertising and the publishing
sides, not only selling the ability to show ads on its suite of platforms
(YouTube, Maps, Google Sites, Android Apps, and more), but also
dominating advertising on other websites through a service called AdSense.115
In controlling advertising pipelines, Google provides critical internet
infrastructure which allows advertisers and web publishers to transact.116
As said, Google’s business model is illustrative of a paradox that
underlies many platform models. Like other gatekeepers, Google is at once
an operator of marketplaces for private exchange, a provider of collectively
valuable public goods and a self-interested actor motivated by its own, often
ad-based, profits. It builds valuable infrastructure that serves the public in
multiple ways, and at the same time makes revenues by intermediating
transactions between advertisers, content and website owners, and users.
Regulating Google requires ensuring that these three functions (intermediary,
public good, market actor) are performed without allowing Google’s interests
in profit maximization and growth to prevail over consumers, creators and
advertisers’ needs.

B. Underenforcement

Until recently, U.S. regulatory attitudes vis-à-vis Google were timid and

113
CMA, Online platforms and digital advertising market study, FINAL REPORT, July 1st
2020, at 222, https://www.gov.uk/cma-cases/online-platforms-and-digital-advertising-
market-study, [hereafter CMA Report].
114
This is thanks to a series of acquisitions including the acquisition of DoubleClick in
2007, of AdMob in 2009, of Invite Media in 2010, of AdMeld in 2011, and of Adometry in
2014. For more detail see CMA Report, at 272.
115
See Dina Srinivasan, Why Google Dominates Advertising Markets, 24 STAN. TECH. L.
REV. 55 (2020), https://ssrn.com/abstract=3500919 (an analysis of Google’s
anticompetitive behavior on ad-tech markets).
116
Chand Narendra-Colucci & Ethan Zuckerman, Forgetful Advertising: Imagining a More
Responsible Digital Ad System, ESSAYS ON A HEALTHY DIGITAL PUBLIC SPHERE ,
INFORMATION SOCIETY PROJECT, September 2022,
https://law.yale.edu/isp/publications/digital-public-sphere/healthy-digital-public-
sphere/forgetful-advertising-imagining-more-responsible-digital-ad-system (arguing that
advertising is infrastructure, and introducing a model for responsible online advertising).

Electronic copy available at: https://ssrn.com/abstract=4275143

30 DRAFT [Feb-23

dominated by the imperative to minimize interference with self-correcting

market mechanisms. In the European Union, regulators started a few high-
profile antitrust investigations against the company in the 2010s,117 and in the
Google Spain ruling, courts imposed a set of right to be forgotten obligations
on the search engine.118 In the U.S., many good faith efforts were eroded by
a culture dominated by deregulatory imperatives.119 Free speech and
technology law were long pervaded by a deregulatory bent that justified
regulatory apathy.120 For example, in the case of Search King v. Google in
2003, the District Court of Oklahoma held that search rankings were
“fundamentally subjective” opinions constitutionally protected under the
First Amendment and as such Google was immune from tortious liability.121
In connection to debates on antitrust in search markets,122 many scholars

117
These cases ended in decisions many years later. See Google Search (Shopping), Case
AT 39740, Commission Decision C/2017/4444 (2018) OJ C 9/11 (fining Google and
Alphabet €2.42 billion); Google Search (AdSense), Case AT 40411, Commission Decision
C/2019/2173 (2020) OJ C 369/04 (fining Google and Alphabet €1.49 billion); Google
Android, Case AT.40099, Commission Decision, C/2018/4761 (2018) OJ C 402/08 (fining
Google and Alphabet €4.34 billion).
118
Case 131/12, Google Spain SL, Google Inc. v. Agencia Española de Protección de
Datos and Mario Costeja González [2014] ECR I-000, nyr.
119
With some exceptions. I discuss antitrust below. On the consumer protection front, in
2002, the FTC’s Consumer Protection Bureau sent search engines letters asking them to
distinguish clearly between organic search results and paid search results or ads to avoid
misleading consumers. (FTC, Commercial Alert Response Letter, June 27, 2002,
https://www.ftc.gov/legal-library/browse/cases-proceedings/closing-letters/commercial-
alert-response-letter.) The guidance was then updated in 2013 when it was found that paid
results had become confusing and indistinguishable from organic results. (FTC Consumer
Protection Staff Updates Agency's Guidance to Search Engine Industry on the Need to
Distinguish Between Advertisements and Search Results, June 25, 2013,
https://www.ftc.gov/news-events/news/press-releases/2013/06/ftc-consumer-protection-
staff-updates-agencys-guidance-search-engine-industry-need-distinguish.) See also Data
Brokers: A Call For Transparency and Accountability: A Report of the Federal Trade
Commission, May 2014,
https://ec.europa.eu/commission/presscorner/detail/en/IP_18_4581.
120
Amanda Shanor, The New Lochner, WIS. L. REV. 133 (2016) (arguing that First
Amendment doctrine excessively constrains the administrative state); Genevieve Lakier,
The First Amendment's Real Lochner Problem, 87 U. CHICAGO L. REV. 1243 (2020)
(arguing that First Amendment doctrine advances a narrow conception of negative liberty).
121
Search King v. Google No. Civ-02-1457-M, 2003 WL 21464568 (W.D. Okla. May 27,
2003).
122
Id. See also Eric Goldman, Search Engine Bias and the Demise of Search Engine
Utopianism, 8 YALE J.L. & TECH. 188, 189 (2006) (arguing that search engines make
editorial choices); Eugene Volokh & Donald M. Falk, Google First Amendment Protection
for Search Engine Search Results, 8 J.L. ECON. & POLY 883, 886-894 (2012) (arguing that
search engine results are speech and as such protected under the First Amendment). Though
see more recent developments: Eugene Volokh, Social Media Platforms as Common
Carriers?, Draft Presented at Freedom of Expression Scholars Conference, April 2021

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 31

supported leaving Google’s search results alone.123 In some cases, research

suggesting that antitrust should not apply was funded by Google itself.124 As
a result, Google’s search, advertising and other practices have been
significantly under-scrutinized. Coupled with successful behind-the-scenes
lobbying,125 the deregulatory bent of American policy and commentary
meant that Google was left largely free to regulate its search results and
advertising marketplaces as it pleased. Today, even as the winds are
changing, big tech continues to strategize and divert regulatory efforts.126

1. Advertising

Google is unmistakably the dominant player on online advertising

markets. In 2007, the FTC dismissed (i.e., never started) an investigation into
Google’s proposed acquisition of DoubleClick, the dominant player in the
third-party ad serving market at the time. The FTC reasoned that even if the
transaction posed consumer privacy issues “the sole purpose of federal
antitrust review of mergers and acquisitions is to identify and remedy

(arguing that social media platforms should be regulated as neutral carriers of speech); Biden
v. Knight First Am. Inst. at Columbia Univ., __ S. Ct. __ (2021) (Thomas, J., concurring,
stating, obiter, that Twitter should be considered a common carrier); Netchoice LLC. v.
Paxton, 596 U.S. ___ (2022) (Justice Alito, with whom Justice Thomas and Justice Gorsuch
join, dissenting, arguing that Facebook ought to be regulated as a neutral carrier of speech).
123
Geoffrey Manne and Joshua D.Wright, Google and the Limits of Antitrust: The Case
Against the Antitrust Case Against Google, 34 HARV. J. OF L. AND PUB. POL.’Y (2011)
(arguing that Google Search does not violate antitrust); Geoffrey A. Manne and William
Rinehart, The Market Realities that Undermined the FTC’s Antitrust Case Against Google,
HARV. J. OF L. & TECH. OCCASIONAL PAPER SERIES (2013), (describing some errors on the
part of the FTC); Michael A. Salinger & Robert J. Levinson, Economics and the FTC’s
Google Investigation, 46 REV. IND. ORGAN. 25 (2015). But see Mark R. Patterson, Google
and Search Engine Market Power, Fordham Law Legal Studies Research Paper No.
2047047 (2012), https://ssrn.com/abstract=2047047 (arguing that asymmetries of
knowledge and control over information constitutes Google’s market power); JOHN E.
KWOKA JR. AND LAWRENCE J. WHITE, THE ANTITRUST REVOLUTION: ECONOMICS,
COMPETITION, AND POLICY (2018) (discussing the political context and history of the
FTC’s Google investigation).
124
Eugene Volokh & Donald M. Falk, First Amendment Protection For Search Engine
Search Results, April 20, 2012, https://volokh.com/wp-
content/uploads/2012/05/SearchEngineFirstAmendment.pdf (paper commissioned by
Google).
125
See ZUBOFF, supra note 12; Zephyr Teachout & Lina Khan, Market Structure And
Political Law: A Taxonomy Of Power, 9 DUKE J. OF CON. L. & PUB. POL.’Y 37 (2014)
(discussing industry lobbying and other forms of government capture by large firms).
126
Leah Nylen, Meta Seeks Recusal of FTC Chair Khan in Within Deal Challenge,
BLOOMBERG, Sept 23, 2022, https://www.bloomberglaw.com/bloomberglawnews/privacy-
and-data-security/XQFOEVO000000?bna_news_filter=privacy-and-data-security#jcite
(discussing Meta’s attempts to recuse Chair Khan).

Electronic copy available at: https://ssrn.com/abstract=4275143

32 DRAFT [Feb-23

transactions that harm competition.”127

In its decision to close the investigation, the FTC argued that ad server
markets were competitive, even if they were concentrated: “the ad
intermediation market is a … competitive market with dozens of
competitors.”128 Given the competitiveness of ad markets, the FTC found that
the merger would not be an obstacle to competition and should proceed. Its
predictions that the merger would not undermine competition in the long run
(not eliminate competitive constraints, not lead to exclusionary tying or
bundling on Google’s part, no risk of the merging parties’ databases merging
leading to an intermediation advantage) would later prove wrong.
In separate statements, two Commissioners emphasized notions of
consumer choice and privacy in platform settings. Both Commissioners
Leibovitz and Harbour emphasized the need to intervene to give consumers
choices regarding the merger’s privacy implications.129 Commissioner
Harbour also emphasized choice from an antitrust perspective, pointing to
inter-platform competition.130
Section 5 of the FTC Act entails distinct regimes for “unfair methods of
competition” (antitrust law) and “unfair and deceptive acts” (consumer law,
including privacy).131 In spite of these two opinions highlighting that the

127
Statement of the Federal Trade Commission Concerning Google/DoubleClick, Case n.
0710170, December 20, 2007, at 2.
128
Id. at 10.
129 Concurring Statement of Commissioner Jon Leibowitz in Google/DoubleClick, FTC

File No. 071-0170, at 3,

https://www.ftc.gov/sites/default/files/documents/public_statements/concurring-statement-
commissioner-jon-leibowitz-google/doubleclick-matter/071220leib_0.pdf (“Will industry
be willing and able to develop short, conspicuous, and intelligible notices so that
consumers can make informed choices about online tracking and targeted advertising?);
Dissenting Statement of Commissioner Pamela Jones Harbour, In the matter of
Google/DoubleClick, F.T.C. File No. 071-0170, at 12,
https://www.ftc.gov/sites/default/files/documents/public_statements/statement-matter-
google/doubleclick/071220harbour_0.pdf (“the combined firm is urged to state clearly and
unambiguously what kind of information it intends to gather”).
130
Commissioner Harbour, n.25, at 10 (“if network effects lead to a reduction in the
number of search competitors, consumers will suffer from a diversity of choice among
search engines, which will reduce the incentives of search firms to compete based on
privacy protections or related non-price dimensions”).
131
FTC, A Brief Overview of the Federal Trade Commission's Investigative, Law
Enforcement, and Rulemaking Authority, May 2021, https://www.ftc.gov/about-
ftc/mission/enforcement-authority. There is a surprising scarcity of literature on the
relationship between the two, but see Luke Herrine, The Folklore of Unfairness, 96 N.Y.U.
L. REV. 431, 454 (2021) (describing the joint history of unfair competition and unfair
commercial practices); Robert H. Lande and Neil W. Averitt, Consumer Sovereignty: A
Unified Theory of Antitrust and Consumer Protection Law, 65 ANTITRUST L. J. 713 (1997),
https://ssrn.com/abstract=1134798 (unifying antitrust and consumer protection around
the idea of consumer choice).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 33

FTC could take privacy and consumer choice considerations into account, the
majority concluded that data and privacy harms did not warrant intervention
or investigation from a competition law perspective. Privacy harms were of
no interest to regulators concerned with “unfair methods of competition”
only.
The FTC’s assumptions about the separateness of competition and the
free and competitive nature of advertising markets, the large diversity of
choices these markets afford and the need to let them operate autonomously
are illustrative of the early 2010s. Google’s speculations about its future post-
merger conduct, for example, were given weight in spite of their subsequently
demonstrated flaws.

2. Search

In the early 2010s, the FTC and various state AGs started an antitrust
investigation alleging Google self-preferencing on search results. The FTC
dropped the case in 2013 and did not start formal proceedings, in spite of 19
months spent reviewing over nine million pages of documents from Google
and others.132 The original allegation was that “Google unfairly preferences
its own content on the Google search results page and selectively demotes its
competitors’ content from those results.”133 To have a claim under section 2
of the Sherman Act, the FTC had to show an intent to acquire or maintain
monopoly power through exclusionary conduct.134 The question was thus
framed as “whether Google changed its search results primarily to exclude
actual or potential competitors and inhibit the competitive process, or on the
other hand, to improve the quality of its search product and the overall user
experience.”135 The FTC found that although Google effectively harmed and
demoted potential competitors, “in the main, [it] adopted the design changes
… to improve the quality of its search results, and … any negative impact on

132
Id. See also Leah Nylen, How Washington fumbled the future, POLITICO, March 16,
2021, https://www.politico.com/news/2021/03/16/google-files-ftc-antitrust-investigation-
475573#:~:text=The%20FTC%20spent%2019%20months,%2C%20Tripadvisor%2C%20F
acebook%20and%20Amazon.
133
Statement of the FTC Regarding Google’s Search Practices - In the Matter of Google
Inc.; January 3, 2013, at 1, https://www.ftc.gov/legal-library/browse/cases-
proceedings/public-statements/statement-federal-trade-commission-regarding-googles-
search-practices-matter-google-inc.
134
To state a monopolization claim under § 2 of the Sherman Act, a plaintiff must allege
that (1) the defendant possessed monopoly power in the relevant market; (2) the defendant
willfully acquired or maintained that power through exclusionary conduct; and (3) the
defendant's conduct caused antitrust injury. MetroNet Servs. Corp. v. Qwest Corp., 383
F.3d 1124, 1130 (9th Cir. 2004).
135
FTC, supra note 133, at 2.

Electronic copy available at: https://ssrn.com/abstract=4275143

34 DRAFT [Feb-23

actual or potential competitors was incidental to that purpose.”136 Showing

harm to competitors seemed an uphill battle. As argued by Geoffrey Manne
and William Rinehart, even if intent to exclude could be proved, there was
no plausible harm to competitors since being systematically denied the first
two spots in search was the norm for most websites.137 The FTC would not
likely have been able to show that competitors were harmed compared to the
counterfactual.
The fragility of the case’s section 2 claim should not obscure Google’s
infrastructural and information power at the time.138 Scrutinizing
anticompetitive conduct at the content level, in the form of ranking
discrimination or self-preferencing, omits that Google was already building
and controlling an infrastructural search and information access empire
through purchasing adtech companies, developing Android and expanding
into new markets. Beyond revealing an entrenched optimism regarding the
competitiveness of digital markets, the case highlights a larger problem. It
assumes that abuses of power in concentrated digital markets are an exception
to be proved in court by showing monopolization and harm to competitors,
and not a default systemic feature of the contemporary internet that requires
structural, durable and pre-emptive constraints.

3. Under-enforcement

Both Google/Doubleclick and the search preferencing apathy reflect a

climate of under-enforcement in digital platform markets. The assumptions
during this pre-2020 period, which remain vivid today, were at least two-fold.
The first is that competition between platforms is one click away due to low
entry costs, given that building digital infrastructure can be done at low cost
by anyone at any time.139 Because of the healthy competitiveness of these
markets, successful firms must have deployed an appropriate level of skill,

136
Id.
137
Manne & Rinehart, Market Realities, supra note 123, at 9.
138
Patterson, supra note 123; Marshall Steinbaum, Establishing Market and Monopoly
Power in Tech Platform Antitrust Cases, THE ANTITRUST BULLETIN 1, 9 (2022) (describing
market power in platform markets and critiquing the analysis in Ohio v. American Express,
585 U.S. ___ (2018)); Daniel Francis, Making Sense of Monopolization: Antitrust and the
Digital Economy 84 ANTITRUST L.J. (2022) (theorizing monopolization in a digital
context); John M. Newman, Antitrust in Digital Markets, 72(5) VAND. L. REV. 1497 (2019)
(arguing that dominant monopolization paradigms in antitrust law do not fit digital
markets).
139
But see John M. Newman, The Myth of Free, 86 GEO. WASH. L. REV. 513 (2018)
(rebutting that assumption).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 35

foresight or industry.140 Their behavior must thus be left untouched by courts

and authorities. Only a deregulatory attitude to antitrust and regulatory
intervention can protect the virtues of competitive markets, and their
consequent innovation and choice benefits.
The second assumption is that in the digital economy the interests of
platform gatekeepers such as Google are aligned with the interests of
consumers and competitors. In the case of Google/Doubleclick, the resulting
advertising ecosystem was seen as bringing benefits to consumers,
advertisers and customers in the form of expanded information and
transactional opportunities (maximizing output). In the case of web search,
Google’s rankings could offer website owners an opportunity for greater
visibility. In this sense, Google’s service expands rather than undermines
opportunities for choice, and for transactional and allocative efficiencies.
These two assumptions show that the predominant view until recently
was one of general optimism toward the competitiveness of digital platform
markets and a general skepticism for regulatory intervention including for
example the combination of antitrust with privacy concerns. The view has
long been that antitrust should be an ex-post instrument that is best under-
enforced. It should incentivize platforms to build products and experiences
that are innovative and bring apparent diversity and choice to consumers even
when this leads to consumer harms. Clear harm can be policed after-the-fact.
Many scholars and policy makers celebrated digital platforms as
hypercompetitive marketplaces that afforded consumers an unprecedented
menu of options and choices.141 They neglected that these business models
were acquiring increasing infrastructural control, the ability to shape (and not
just enhance) online choices, as well as the power to lock-in consumers,
foreclose competitors and privatize public goods. Law pre-2020 favored the
interests of private gatekeepers and neglected the interests of actors not
represented by self-correcting market forces: consumers, content producers,
advertisers, and possibly also competitors.142

140
United States v. Grinnell Corp., 384 U.S. 563, 570-71 (1966) (stating that a
monopolization offence requires improper conduct-as opposed to superior skill, foresight,
or industry).
141
THOMAS RAMGE AND VIKTOR MAYER-SCHÖNBERGER, REINVENTING CAPITALISM IN
THE AGE OF BIG DATA (2018) (celebrating the potential of data-driven markets). Also see
VILI LEHDONVIRTA, CLOUD EMPIRES: HOW DIGITAL PLATFORMS ARE OVERTAKING THE
STATE AND HOW WE CAN REGAIN CONTROL (2022) (discussing the early Internet as a
locus of promise and free interaction).
142
This confirms the thesis of sociologist Neil Fligstein, Markets as Politics: A Political-
Cultural Approach to Market Institutions, 61 AM. SOCI. REV. 656, 662, 664 (1996) (“Law
and accepted practices often reflext the interests of the most organized forces in society” at
662. “At the beginning of a new market, the largest firms are the most likely to be able to
create a conception of control and a political coalition to control competition” at 664).

Electronic copy available at: https://ssrn.com/abstract=4275143

36 DRAFT [Feb-23

In their optimism, many scholars distinguished digital platform markets

from the software and hardware markets at stake in Microsoft.143 In that case,
in response to the DOJ’s allegations of illegal monopolization on the personal
computer market, the court ordered Microsoft’s divestiture.144 The case
ended in a settlement where the company agreed to a series of remedies,
including allowing device manufacturers to unbundle their devices from
Microsoft’s products, and providing fair access to their APIs.145 Even in
Microsoft, however, the court’s under-enforcement bent is visible. The court,
for example, stated that novel business practices, particularly in technology
markets, should be presumed to be legal unless “precise harm” can be
shown.146

C. A Turning Point

Today, regulators are finally starting to move away from an

underenforcement logic while applying the lessons of Microsoft to digital
platform markets. While no Microsoft-like remedies have yet been imposed
on platforms like Google, Facebook or Amazon, faith in naturally super-
competitive and efficient marketplaces is giving way to an emphasis on
Google and other platforms’ infrastructural power and dominance. This
perspective reveals law’s role in shaping digital environments, including how
competitive they can be. As part of this evolution, the lines between the core
object of antitrust law, the promotion of self-correcting competitive markets
and distributed individual preferences, and the core object of regulation,
natural monopoly or other goals exogenous to competition and distributed
exchanges of preferences, and thus between antitrust and economic
regulation themselves, are increasingly blurred. Regulation and antitrust
thinking, their centralizing and decentralizing, over- and under-enforcement
as well as ex ante and ex post characteristics, are unmistakably converging,
though perhaps without anyone noticing.
In October 2020, the United States Committee on the Judiciary Antitrust

143
United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001).
144
United States v. Microsoft Corp., 97 F. Supp. 2d 59, 64-65 (D.D.C. 2000) (the District
Court for the District of Columbia issued an order requiring Microsoft to submit a proposed
plan of divestiture, with the company to be split into an operating systems business and an
applications business).
145
Microsoft, supra note 143. Department of Justice, Reach Effective Settlement on Antitrust
Lawsuit, PRESS RELEASE, November 2nd, 2001,
https://www.justice.gov/archive/atr/public/press_releases/2001/9463.htm; U.S. v. Microsoft
Corporation Information on the Settlement, November 6th, 2001,
https://www.justice.gov/atr/usdoj-antitrust-division-us-v-microsoft-corporation-
information-settlement.
146
Microsoft, supra note 143, at 91 (citing N. Pac. Ry. Co., 356 U.S. at 5).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 37

Subdivision issued a more than 400-page report which focuses on the

dominant role of four tech giants - Apple, Amazon, Google and Facebook.147
Its proposals, which include prohibitions on discrimination and self-
preferencing, presumptive prohibitions of mergers and a strengthened section
2 prohibition, represent a change of direction.148 The Report illustrates two
ongoing trends: the move in antitrust to increasingly address structural and
infrastructural aspects in digital platform markets, and the push to pass ex
ante statutory frameworks into law to decentralize and render digital markets
structurally more competitive. Notwithstanding convergence, the primacy of
ideals of “competition,” “choice” and “innovation” remains as strong as ever.

D. The Infrastructural Turn in Digital Antitrust

1. Search

Soon after the 2020 Report came out, the Department of Justice (DOJ)
accused Google of maintaining an anticompetitive monopoly in search and
search advertising through exclusionary agreements with actors controlling
devices, web browsers, and other search access points.149 In December of the
same year, several state attorneys general, including Colorado, started a
similar case against Google for alleged anticompetitive behavior on search
and search advertising markets.150 In both cases, antitrust plaintiffs go beyond
the 2013 focus on content rankings, and instead investigate a series of
exclusionary distribution agreements with Apple, Mozilla and others that
maintain Google’s dominance by ensuring that it remains the default on web
browsers, mobile and other devices, often in exchange of a share of
advertising profits. The complaints allege that, through these agreements,
Google achieved an overall 60% coverage of all general search queries, and
that nearly half the remaining queries (about 20% more) are funneled through
Google owned-and-operated properties such as Chrome.151 The complaint

147
House Report, supra note 28.
148
Id. at 20.
149
United States v. Google LLC, No. DC/1:20-cv-03010 (D.D.C. 2020); Department of
Justice, Justice Department Sues Monopolist Google For Violating Antitrust Laws, PRESS
RELEASE, October 20th, 2020, https://www.justice.gov/opa/pr/justice-department-sues-
monopolist-google-violating-antitrust-laws. Also see Cecilia Kang, David McCabe &
Daisuke Wakabayashi, U.S. Accuses Google of Illegally Protecting Monopoly, THE NEW
YORK TIMES, October 20, 2020, https://www.nytimes.com/2020/10/20/technology/google-
antitrust.html.
150
State of Colorado et al v. Google LLC, No. 1:2020cv03715 (D.D.C. 2020).
151
See, e.g., United States v. Google LLC, No. DC/1:20-cv-03010 (US District Court for
the District of Columbia, filed October 20, 2020), Amended Complaint, January 15, 2021,
para 5, at 4, https://www.justice.gov/atr/case-document/file/1428271/download.

Electronic copy available at: https://ssrn.com/abstract=4275143

38 DRAFT [Feb-23

also highlights that Google and Apple concluded a deal whereby Apple
agreed to preset Google as its default general search engine on Safari and
across Apple devices in exchange for a large percentage of Google’s
advertising revenues.152
Although the outcome of this case is uncertain, court documents already
reveal where antitrust enforcement in this area might be heading. First, the
focus of regulators between 2013 and now has moved from a concern with
discrimination on the content layer, at the level of search rankings, to an
infrastructural concern, i.e. worry about concentrated control over
infrastructural bottlenecks and awareness that search competition cannot be
achieved without contestability over search infrastructure. It must, in other
words, become possible for new entrants to build a competing ecosystem or
tap into the dominant ecosystem in a fair and transparent way. This goes
hand-in-hand with an emerging view that antitrust cannot merely intervene
ex post to correct self-preferencing behavior and deter future behavior. It
must help reconfigure the structure of search markets in a dynamic before-
the-fact way. Competitive processes, in other words, are not simply
something to constrain from the outside; they are processes constructed
through durable remedies over time.
Second, the focus in the 2013 case was not on antitrust law’s role in
enabling search diversity and including new entrants. The FTC and the States
failed to imagine effective remedies to such bias. Even if the case had been
successful, interventions targeting self-preferencing algorithms, alone, would
unlikely have displaced Google’s dominant position on search.153 The DOJ’s
current linking of search to advertising conduct instead denotes a better grasp
of the multi-sidedness of Google’s business model. In this sense, it seems that
antitrust is evolving toward a greater sensitivity for infrastructure and concern
for distrupting vertical integration, away from the prior tolerance of vertical
agreements and efficient monopolies – often imputed to Chicago School
antitrust.154

2. Advertising

Also in December 2020, ten state attorneys general including Texas

brought a suit against Google alleging anticompetitive conduct in advertising
markets, and in January 2023 the Department of Justice filed a second lawsuit

152
Id. para 86, at 28.
153
Google Search (Shopping) and Google Search (AdSense), supra note 117.
154
Richard A. Posner, Natural Monopoly and Its Regulation, 21 STAN. L. REV. 549, 564
(1969) (“the effort of a businessman to monopolize a market by producing at a cost so low
as to drive out his competitors and deter new entry or, the monopoly achieved, to improve
his return by lowering his costs still further is not at all reprehensible.”).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 39

against Google in these markets.155 In the first case’s unredacted amended

complaint, plaintiffs describe Google’s dominant position in advertising
markets showing that nearly all online publishers sell ad space through
Google and nearly every advertiser relies on Google to display
advertisements on the internet.156 Google is therefore the primary middleman,
bottleneck and monetization infrastructure of the internet.
The Complaint alleges anticompetitive conduct on various online
advertising market segments. It identifies tying arrangements and
exclusionary conduct as well as a price fixing agreement with Facebook.157
Against the grain of the FTC’s 2007 DoubleClick predictions, the complaint
shows that Google started bundling its services with DoubleClick’s soon after
acquiring it. Relying on its presence on the ad-buying and ad-serving
segments, as well as its monopoly over auctions and exchanges, Google
started routing traffic into its walled ecosystem, preventing advertisers and
publishers from benefiting from competitors’ lower prices. The complaint
also alleges that Google restricted publishers from selling inventory on more
than one exchange at a time and used its asymmetric access to information to
prevent publishers from communicating about their inventory with non-
Google exchanges. The complaint also describes anticompetitive and
deceptive practices concerning a competitive ad allocation mechanism called
“header bidding” which threatened Google’s dominant position. Google and
Facebook allegedly coordinated to “kill” header bidding by steering
customers away from it. In exchange for Facebook’s cooperation and help in
maintaining Google’s dominance on advertising intermediation, Google
promised Facebook advantages on its auctions, e.g. bidding priority.
It is still early stages in this and other litigation in adtech markets, and too
soon to know which, if any, of the allegations pleaded so far will proceed to
trial and what relief, if any, these case might lead to. The complaint is
nonetheless ambitious in its breadth and creative in its findings. The evolution
from the deregulatory perspective the FTC adopted in DoubleClick is visible
and the case is likely to open adtech exchanges to greater competition.

155
Complaint, Texas et al. v. Google LLC, Case No. 4:20cv957 (E.D. Tex. Dec. 16, 2020);
Complaint, United States et al. v Google LLC, Case 1:23-cv-00108 (E.D. Virg. Jan. 24,
2023). (Note to editors: this paragraph will need to be amended in light of the new DOJ
complaint.)
156
Second Amended Complaint at 4, Texas et al. v. Google LLC, Case No. 4:20-CV-957-
SDJ (E.D. Tex. Aug. 4, 2021). Taylor Hatmaker, Google asks a judge to dismiss Texas
antitrust lawsuit about its ad business, TECHCRUNCH, January 21st, 2022,
https://social.techcrunch.com/2022/01/21/google-asks-a-judge-to-dismiss-texas-antitrust-
lawsuit-about-its-ad-business/.
157
Id. at para 117.

Electronic copy available at: https://ssrn.com/abstract=4275143

40 DRAFT [Feb-23

3. Is Antitrust Changing?

To sum up those trends, one notices an increasing awareness of the need

for ongoing (or ex ante) intervention in digital markets, away from a
conception of digital markets as self-correcting and as requiring immunity
from exogenous interventions. At the same time, there is a greater interest in
the effects of antitrust remedies and their potential to address market power
at various levels of the internet’s stack. In this sense, antitrust is moving
toward a “direct” regulatory vision, less deferent to the idea of autonomous
competitive market forces.
This evolution suggests three normative imperatives for antitrust law in
technology platform markets. First, antitrust must incorporate considerations
that have been traditionally considered outside its scope including privacy,
unfair and deceptive acts and other consumer protection concerns. These
factors, which might be considered ‘regulation,’ have been taken into account
only sporadically until 2020, but a change of direction is underway.158
Second, the role of antitrust cannot remain purely negative, after-the-fact and
aimed at preserving competitive markets through under-enforcement.
Antitrust enforcers are beginning to embrace a positive, structuring role in
digital ecosystems. Third, while antitrust can do a lot to protect consumers, it
cannot do so alone, at least insofar as it continues to be construed as a
minimal, residual promoter of distributed transactions in marketplaces. A
range of complementary and ancillary strategies are needed which take
collective interests and public institutions as central, e.g. public utility
regulation, corporate chartering, cooperativism or public options. Measures
might include stronger regulatory powers to request information, carry out
investigations, and impose structural remedies.159 The extent to which these
strategies can be accommodated within the scope of antitrust law or require
alliances between antitrust and other regimes is an open question which will
be partially addressed in what follows. There is no reason to limit antitrust
possibilities before having experimented with them, especially in a political

158
Reuben Binns & Elettra Bietti, Dissolving Privacy One Merger at a Time: Competition,
Data and Third Party Tracking, 36 COMPUTER LAW & SECURITY REVIEW (2020) (showing
that agencies in the US and EU have not paid sufficient attention to data-sensitive mergers).
159
See Rory Van Loo, The Missing Regulatory State: Monitoring Businesses in an Age of
Surveillance, 72 VAND. L. REV. 1563 (2019) (arguing that online platforms need the kind
of regulatory monitoring seen in other industries); Rory Van Loo, Digital Market
Perfection, 117 MICH. L. REV. 815, 824 (2019) (suggesting friction for digital markets
analogous to that which exists in stock markets); Caleb Griffin, Systemically Important
Platforms, 107 CORN. L. REV. 445 (2022), (arguing for a broad array of oversight
mechanisms that would complement existing antitrust remedies including greater
regulatory access to middleware).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 41

climate where passing legislation is a slow and uncertain process.160

E. The Pro-Competitive Regulatory Turn

In parallel to antitrust’s transformation,161 a new regulatory attitude is

emerging that seeks to decentralize digital economies and structure
competitive processes by imposing obligations on platform bottlenecks.

1. Google as a Common Carrier and Utility?

Treating Google as a regulated entity challenges the idea of self-

correcting competitive processes. Attempts to do so have thus failed to
generate momentum at the policy level to this day. Early skepticism dates
from at least 2006 when Gigi Sohn, President of Public Knowledge, argued
that unlike the broadband network industry, search was a competitive
market.162 Today, digital markets including search are concentrated, and
network effects make it as difficult to compete with incumbent platforms as
with large broadband companies.
Already in 2008, Frank Pasquale and Oren Bracha argued in a series of
articles that search engines should in fact be treated as common carriers and
should be subject to search neutrality principles similar to those that applied
to telecommunications companies, because of the critical impact these
engines had on society.163 Showing that a “peaceable kingdom” of
competition between diverse search engines was “unlikely to develop,”164
their solution was a “baseline of accountability for dominant search
results.”165 A common carrier or utility framework could help ensure
universal and fair access as well as transparency and accountability in search.

160
Note that the American Innovation and Choice Online Act (AICOA) discussed in this
Article failed to be passed in Congress.
161
These include the cases mentioned supra at note 5.
162
Gigi Sohn, Another Red Herring, PUBLIC KNOWLEDGE (June 5, 2006), cited in Frank
Pasquale, Internet Nondiscrimination Principles: Commercial Ethics for Carriers and
Search Engines, 2008 UNIV. OF CHI. LEG. FORUM (2008), at fn 47.
163
Oren Bracha & Frank Pasquale, Federal Search Commission – Access, Fairness and
Accountability in the Law of Search, 93 CORNELL L. REV. 1149 (2008) (suggesting an
agency for regulating search engines); Frank Pasquale, Internet Nondiscrimination
Principles: Commercial Ethics for Carriers and Search Engines, 2008 UNIV. OF CHI. L.
FORUM 263 (2008) (suggesting common carrier principles for holding search engines to
account). See also FRANK PASQUALE, THE BLACK BOX SOCIETY: THE SECRET
ALGORITHMS THAT CONTROL MONEY AND INFORMATION 98–100 (2015) (discussing
Google and the analogy to Robert Hale’s writing on utilities).
164
Frank Pasquale, Rankings, Reductionism, and Responsibility, 54 CLEV. ST. L. REV. 115,
135 (2006) (showing that search rankings can be biased).
165
Id.

Electronic copy available at: https://ssrn.com/abstract=4275143

42 DRAFT [Feb-23

Beyond asking companies to clearly differentiate between organic and paid

results, as the FTC had done,166 Pasquale and Bracha proposed a set of top-
down procedural standards enforced by a regulator,167 possibly a Federal
Search Commission.168
More recently, Sabeel Rahman and Peter Swire, writing separately, have
argued that some Big Tech services should be regulated as utilities.169 Swire
argues that natural monopoly justifications for public utility law now apply
to digital platform markets which are characterized by a “high market share,
network effects, and difficulty for consumers to live without the service.”170
He also argues that in digital platform markets, price-setting translates into
content quality and standard-setting.171 Rahman proposes a multi-pronged
approach to utility regulation in digital markets. The first step is to identify
“infrastructural” goods and services, which arguably include many digital
platform infrastructures.172 The subsequent step consists in imagining how to
address infrastructural (digital) power through three sets of tools: regulatory
oversight by bodies such as the FTC or FCC; firewalls or structural separation
between infrastructural goods and services and other goods and services
offered by the same entity; and finally public options which include
nationalization, municipalization but also simply the government competing
with other businesses on given market segments.173
166
See FTC guidelines, supra note 119.
167
Pasquale, Nondiscrimination Principles, supra note 163.
168
Bracha & Pasquale, supra note 163.
169
Peter Swire, Should the Leading Online Tech Companies Be Regulated as Public
Utilities? LAWFARE, August 2, 2017, https://www.lawfareblog.com/should-leading-online-
tech-companies-be-regulated-public-utilities (arguing for treating Big Tech as public
utilities); Sabeel Rahman, Infrastructural Regulation and the New Utilities, 35 YALE J.
REG. 911 (2018) (arguing for a new approach to infrastructural regulation in tech markets);
K. Sabeel Rahman, Domination, Democracy, and Constitutional Political Economy in the
New Gilded Age: Towards a Fourth Wave of Legal Realism?, 94 TEXAS L. REV. 1329
(2016) (introducing an approach to regulation grounded in republican theories of non-
domination); K. Sabeel Rahman, The New Utilities: Private Power, Social Infrastructure,
and the Revival of the Public Utility Concept, 39 CARDOZO L. REV. 1621 (2018) (arguing
that utility-regulation is suited to new infrastructural industries such as technology); Josh
Simons and Dipayan Ghosh, Utilities for democracy: Why and how the algorithmic
infrastructure of Facebook and Google must be regulated, BROOKINGS INSTITUTION,
August, 2020, https://scholar.harvard.edu/dipayan/publications/utilities-democracy-why-
and-how-algorithmic-infrastructure-facebook-and-google (imagining the governance of
Big Tech as public utilities).
170
Swire, Utilities, supra note 169.
171
Id.
172
Sabeel Rahman, Infrastructural Regulation and the New Utilities, 35 YALE J. REG. 911
(2018); K. Sabeel Rahman, The New Utilities: Private Power, Social Infrastructure, and the
Revival of the Public Utility Concept, 39 CARDOZO L. REV. 1621 (2018).
173
See Jon D. Michaels, We the Shareholders: Government Market Participation in the
Postliberal U.S. Political Economy, 120 COL. L. REV. 465 (2020) (imagining public

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 43

These proposals are criticized by free market supporters. Arguments for

treating Google as a regulated entity have indeed been met with skeptical
reactions emphasizing digital markets’ competitiveness, dynamism and
susceptibility to disruptive bottom-up innovations.174 Critics claim that
“social media services are not physical resources with high fixed costs, and
they do not possess ‘bottlenecks’ in any conventional sense of the term.”175
These critiques are misleading. They assume a binary landscape where a
firm is either competitive or a natural monopoly, where subjecting a company
to regulation means treating it like a traditional electric utility, reducing
consumer choice and innovation for the foreseeable future. Yet not all
regulation is command-and-control regulation, and as such, not all regulation
undermines competition. Unlike traditional utilities, Google and other big
tech companies were not costly to build. Further, although digital platform
markets are winner-take-all markets prone to tipping, it can be efficient to
have more than one platform provider on a given market segment. Users,
customers or workers frequently engage in so-called “multi-homing,” using
various platforms at the same time. Regulating digital platform markets is not
like regulating static natural monopolies. It requires a dynamic approach to
aligning infrastructural facilities with the public interest while opening
opportunities for competition and disruption. Regulation, competition and
contestability can indeed go hand-in-hand in technology markets and
embedding them requires sectoral regulation that differs from the traditional
utility model. A new approach to regulation in technology markets is indeed
needed and is emerging.

2. Toward Pro-Competitive Regulation

To many, treating digital platforms as utilities goes too far and is

excessively disruptive of the dominant faith in free market competition.176
Yet the legitimacy and desirability of digital “regulation” has become widely
accepted. Randal Picker, for instance, acknowledges that the boundaries
between antitrust and regulation are becoming more blurred in platform
markets.177 Like in the AT&T case, antitrust and legislation can operate in
tandem, pursuing similar goals in similar manners.178 He recognizes that
“there is some overlap between how antitrust can operate and how new

capitalism as a post-liberal modality of market governance).

174
See, e.g., Adam Thierer, The Perils of Classifying Social Media Platforms as Public
Utilities, 21 COMMLAW CONSPECTUS 249 (2013) (arguing the case against treating Big
Tech as utilities).
175
Id. at 274.
176 Rogerson & Shelanski, supra note 3.
177
Picker, supra note 93, at 2.
178
Id.

Electronic copy available at: https://ssrn.com/abstract=4275143

44 DRAFT [Feb-23

platform legislation is likely to be framed.”179 William Rogerson and Howard

Shelanski express a general preference for ordinary common law-based
antitrust and for minimizing interferences with competition but recognize that
the digital platform context requires what they call “light-handed pro-
competitive” regulation.180 They distinguish light-handed regulation from
what they instead label “traditional cost-based regulation of public
utilities,”181 arguing that the latter reduces firms’ incentive to lower costs and
innovate whereas the former “aims to govern firms’ behavior and market
power by protecting and promoting competition.”182 Pierre Larouche also
interprets recent legislative efforts as procompetitive.183 He argues that digital
regulatory efforts on both sides of the Atlantic are best viewed as
procompetitive efforts.184

3. “Light-Handed Pro-Competitive Regulation”

A number of pro-competitive bills have been proposed in the United

States. Though their success remains uncertain, two have generated
momentum: the “American Innovation and Choice Online Act” (S. 2992)
(hereafter AICOA) proposed by Senators Amy Klobuchar (D-MN) and
Chuck Grassley (R-IA) and the “Open App Markets Act” (S. 2710) proposed
by Senators Richard Blumenthal (D-CT) and Marsha Blackburn (R-TN).
Both bills are motivated by a recognition that antitrust enforcement based on
the Sherman Act and other antitrust laws alone do not suffice to tackle the
infrastructural dominance of technology gatekeepers. They both seek to open
technology platform markets to greater contestability; facilitating intra-
platform and inter-platform competition through transparency,
interoperability and access remedies. AICOA in particular gives powers to
the FTC to investigate the conduct of “covered platform operators” that
engage in statutorily prohibited conduct. This is defined to include self-
preferencing, discrimination against customers on their platform, refusals to
deal, to access or to interoperate, tying and bundling, abuses and self-dealing
made possible based on privileged access to information about customers.185
The Bill, in other words, is a form of ‘regulation’ that confers more or clearer

179
Id.
180
Rogerson & Shelanski, supra note 3, at 1924.
181
Id.
182
Id. at 1925.
183
Pierre Larouche, Foreword: Legislation on digital platform giants: The future of
competition law? 1 CONCURRENCES 2 (2022) (arguing that the DMA is a competition law
framework). Also see Schweitzer, infra note 186 (also arguing that the DMA is a pro-
competitive regulatory framework).
184
Id.
185
AICOA.

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 45

powers to antitrust agencies to police anti-competitive conduct in digital

platform markets.
In Europe, the Digital Markets Act (hereafter DMA) is similarly and even
more explicitly aimed at introducing “contestability” and “fairness” in digital
platform markets.186 Together with the Digital Services Act, the DMA
conceives big tech platforms as bottlenecks that should let other smaller
businesses enter digital market segments and compete with incumbents. Part
of the motivation behind these proposals, which squarely apply to companies
such as Google, includes a geopolitical interest of EU member states in
regulating US platforms and enable home-grown innovation. The DMA was
introduced by the European Commission in mid-December 2020 and will
likely become enforceable in the Spring of 2023.187 It is addressed to
“gatekeepers,” defined as providers of core platform services,188 and lays
down rules ensuring that digital markets are “contestable and fair”.189 Beyond
imposing high penalties for violations,190 the DMA prohibits practices that
“limit contestability” or “are unfair,” such as the combination of a
gatekeeper’s data with third party data, bundling and discriminating between
customers. It also gives users and customers powers to uninstall pre-installed
products, switch platforms, request access and imposes transparency
measures.191 Although the European Commission is given powers to impose
fines and behavioral or structural remedies on digital platforms, in theory the
regime will be enforced separately from the EU Commission’s ordinary
competition law powers.192

F. Conclusions to Part II

Part I articulated the roots of the conceptual relationship between

186
DMA. For commentary, see, e.g., Heike Schweitzer, The Art to Make Gatekeeper
Positions Contestable and the Challenge to Know What is Fair: A Discussion of the Digital
Markets Act Proposal, 3 ZEUP (2021) (arguing that the DMA should be interpreted as a
competition policy measure).
187
Digital Services Act Package: https://ec.europa.eu/digital-single-market/en/digital-
services-act-package; Natasha Lomas, EU’s new rules for Big Tech will come into force in
Spring 2023, says Vestager, TECHCRUNCH, May 5, 2022,
https://techcrunch.com/2022/05/05/digital-markets-act-enforcement-margrethe-vestager/.
188
See DMA, Art. 3.
189
DMA, Art. 1.
190
Up to 10% of global turnover.
191
DMA, Art. 6.
192
DMA, Recital 10 (draft dated May 11, 2022) (“since this Regulation aims to
complement the enforcement of competition law, it should apply without prejudice to
Articles 101 and 102 TFEU, to the corresponding national competition rules and to other
national competition rules regarding unilateral conduct.”)
https://www.consilium.europa.eu/media/56086/st08722-xx22.pdf.

Electronic copy available at: https://ssrn.com/abstract=4275143

46 DRAFT [Feb-23

antitrust, utilities and other areas of law and regulation. Part II described a
shift in digital markets from a regulatory default which consisted in avoiding
interferences with digital platforms’ activities to a current phase of light-
handed regulation which addresses big tech platforms’ infrastructural
dominance while restoring competition through behavioral and structural
remedies. The shift from a deregulatory conception of digital markets as
naturally competitive and innovative to an institutionalist conception of
digital markets as spaces that require legal structuring suggests the need to
revisit the way scholars theorize antitrust’s relation to utilities and other
regulatory regimes in digital platform markets, descriptively and
normatively.

III. ENVISIONING AND CONSTRUCTING THE CONVERGENCE OF

ANTITRUST AND UTILITIES IN DIGITAL PLATFORM MARKETS

Antitrust and utility approaches to technology markets are converging.

Antitrust lawsuits are aiming to restructure tech markets, dislocate vertically
integrated firms, and place infrastructural power at the center. Regulatory
reforms are motivated by a pro-competitive vision that emphasizes inter- and
intra-platform competition through prohibitions on self-preferencing in
particular. The goals, purposes and remedies offered by antitrust in tech
markets increasingly overlap and coincide with the goals of top-down
regulatory regimes. This convergence suggests the possibility of a hybrid
approach to antitrust and utility regulation, sensitive to the dynamism of
digital ecosystems as well as to the function of platforms as intermediaries,
public goods and market actors.
What does convergence look like? What opportunities does it create and
what are its limits? Part III answers these questions, proceeding in three
steps.193 First, it introduces a visual device, the Antitrust & Regulation Device
or ARD. ARD shows that, in combination, antitrust and utility strategies pre-
structure digital markets while aiming to decentralize power and decision-
making. It also shows that legislators and agencies are continuing to
selectively prioritize self-correcting market processes and individual
preferences over the advancement of values such as a healthy digital public
sphere, privacy or equality.
Second, Part III introduces a pragmatic methodology that helps make
sense of the normative and functional synergies between antitrust and utility

193
Note that the history of antitrust law is replete with examples of overlaps between
sectoral and utility-like legislation and antitrust. See, e.g., AT&T, infra note 201; Microsoft,
supra note 143; Trinko, supra note 60.

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 47

approaches.194 This method allows the creative combination of pre-existing

bodies of doctrine and remedies. The quest to regulate a platform company
like Google is not about abstract trade-offs between antitrust and command-
and-control regulation; it is a contextual, dynamic and inescapably value-
oriented process that requires combining regulatory strategies, supervising
some market segments to open other segments to unsupervised competition.
After articulating this methodology, Part III considers two concrete cases: (a)
reform efforts such as AICOA and (b) synergies between unfair competition
and unfair commercial practices under section 5 of the FTC Act.

A. Envisioning the Convergence of Antitrust and Utilities

The relationship between antitrust and regulation has a long history. Until
the 1970s and 1980s, US scholars saw the two as complementary and
emphasized antitrust’s decentralizing effects, and utilities’ centralizing
effects.195 Today, in spite of 50 years of neoclassicist antitrust law which
tolerated ‘efficient’ market concentration, the emphasis on decentralization
and market structure remains pervasive among laypeople, but also lawyers
and scholars.196 Frank Pasquale describes the crossroads of antitrust and
regulatory visions of the internet as a tension between the decentralizing
“Jeffersonian” antitrust path and the centralizing “Hamiltonian”
regulatory/utilities path:197

Populist localizers want a new era of antitrust enforcement to break up giant firms.
These Jeffersonian critics of big tech firms, megabanks, and health care behemoths
are decentralizers. They believe that power is and ought to be distributed in a just
society. They promote strong local authorities to counterbalance the centripetal
accumulation of wealth and power in multinational firms.

Others have promoted gigantism as inevitable or desirable, and argue that we simply
need better rules to cabin abuses of corporate power. Today’s Hamiltonians argue
that massive stores of data are critical to the future of artificial intelligence—and
thus to the productive dynamism of the economy. They focus on improving the

194
The term “pragmatic” refers to John Dewey’s approach to regulation which consists in a
contextual definition and tailoring of means and ends. See DEWEY, supra note 69.
195
See ROBERT BORK, THE ANTITRUST PARADOX (1978) (arguing in 1978 that antitrust’s
“populist” goals of structural decentralization and protecting small firms should be
revisited).
196
Also see PHILLIP E. AREEDA AND HERBERT HOVENKAMP, AN ANALYSIS OF ANTITRUST
PRINCIPLES AND THEIR APPLICATION (2021), Chapter 1 (“When laypersons, including some
lawyers, speak of “competition” they think of a market containing numerous firms…
Protecting “competition” may then mean preserving a market structure that permits small
firms to enter freely and profit, even if this entails forcing larger firms not to compete too
strenuously lest the smaller ones be unable to survive.”).
197
Pasquale, supra note 3.

Electronic copy available at: https://ssrn.com/abstract=4275143

48 DRAFT [Feb-23

regulation of leading firms rather than on breaking them up.198

Further, antitrust’s ex post, case-by-case nature as a common law or

reactive branch of enforcement is often contrasted to utilities and other
regulatory interventions’ structural ex ante nature.199
Let us represent these polarities – centralizing/decentralizing and ex
ante/ex post – visually. I call the visual device the Antitrust & Regulation
Device or ARD (see Fig. 1).

Figure 1

The remedies traditionally associated with antitrust law and public utility
regulation can be represented along ARD’s two axes. The x axis is horizontal
and represents a spectrum of approaches to law that go from decentralizing
to centralizing. From legal interventions that directly aim to protect and
correct decentralized productive processes and individual transactions, or that
have this effect, to modes of intervention that intentionally preserve more
centralized infrastructures, such as utility regulation, or that have this effect
through court enforcement, such as copyright law. The x axis thus goes from
policy action that promotes uncoordinated invisible hand transactions and
preferences (e.g. some antitrust remedies) to action that preserves more
concentrated structures. These can include more structured coordination of
productive decisions, infusing pre-determined purposes directly into

198
Id.
199
OECD, supra note 3, at 4 (“Competition law … is fundamentally a case-by-case
instrument that focuses on one individual situation with a specific set of facts where one
specific undertaking—or a few undertakings in the case of an anti-competitive
agreement—is found to be in violation of competition rules. … Although the adoption of a
decision should have a precedent and deterrent effect on other undertakings not involved in
the specific case and discipline their future behaviour, this is far from certain and is
unlikely to be as effective as ex ante rules that can tackle recurrent and systemic problems
affecting a sector.”).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 49

productive structures, consolidating and restructuring firms internally,

imposing rules or prescribing conduct. The second axis, the y axis, is vertical:
it represents the spectrum from ex ante to ex post forms of regulation, from
structural and infrastructural regulation that is durable and applies to entire
business segments or industries, to after-the-fact fines and other court-based
remedies that are punctual, reactive and apply only to a specific defendant on
a case-by-case basis.
Figure 2 represents common conceptions of antitrust and utilities, at
least how these fields are imagined and described in scholarly and popular
debates.

Figure 2

On the top-left are classic ex post antitrust remedies. These include

remedies issued by a court in response to a complaint by a private plaintiff,
or by a regulator such as the Federal Trade Commission or a State Attorney
General, and they also include out-of-court settlement remedies such as those
imposed in the Microsoft or the AT&T cases.200 They include break-up
remedies. An illustrative case is the break-up of the AT&T telephone and
telephone equipment monopoly into seven telephone operating companies
and one long-distance carrier.201 They also include sanctions and fines such
as are frequently imposed by antitrust authorities in Europe,202 and can be

200
Microsoft, supra note 143. AT&T, infra note 201. On settlements more generally see
Douglas H. Ginsburg & Joshua D. Wright, Antitrust Settlements: The Culture of Consent,
in WILLIAM E. KOVACIC: AN ANTITRUST TRIBUTE 177 (2013) (articulating the virtues of
antitrust settlements); Joshua D. Wright & Douglas H. Ginsburg, The Economic Analysis of
Antitrust Consents, EURO. J.L. & ECON. (2018) (a tentative case against settlements);
Jonathan M. Jacobson, Issues in Antitrust Consent Decrees, Presented to the Department of
Justice Antitrust Division, April 26, 2018 (practicalities related to antitrust settlements).
201
See United States v. AT&T Co., 552 F. Supp. 131, 201, 224 (D.D.C. 1982).
202
See, e.g., Google Search (Shopping), supra note 117 (imposing a fine on Google of
EUR 2.42 billion for preferential treatment in comparison shopping). Google LLC,

Electronic copy available at: https://ssrn.com/abstract=4275143

50 DRAFT [Feb-23

requested by the FTC in exceptional circumstances, for example where the

terms of a Consent Order are violated.203 Even in their more radical forms,
these remedies fit the traditional view of antitrust law as a toolkit that does
not disrupt or change the structure of markets but mainly corrects
obstructions and inefficiencies after-the-fact, reflecting Hovenkamp’s views
on the proper scope of antitrust law as a tool that does not (pre-)structure
markets.204
On the bottom-right is traditional public utility regulation, characterized
as structural and ex ante and compatible with large concentrations of power
in digital markets. Here the tendency is to view legal and regulatory
intervention as something that is centrally planned and coordinated.
Remedies on this side of the spectrum include non-discrimination
obligations, which require a platform, bottleneck or incumbent to treat all
customers, users and suppliers on equal terms.205 In certain regulated
industries, ex ante measures also include the prospective rate-setting aimed
at avoiding price increases and lowering outputs by monopolists.206
Traditional ex ante utility measures also include entry and exit restrictions,
which, respectively, limit the number of players on a given industry segment
or limit the ability of companies offering essential services at a loss to reduce
or terminate those services.207 And they include measures aimed at ensuring
universal access on fair and non-discriminatory terms, such as common

formerly Google Inc. and Alphabet, Inc. v. European Commission, T-612/17

ECLI:EU:T:2021:763 (10 November 2021) (confirming the Commission’s decision).
203
Complaint for Civil Penalties, Injunction, and Other Relief, United States v. Facebook,
Inc., No. 91-cv-2184, 2019 WL 3318596 (D.D.C., July 24, 2019) (where FTC fined
Facebook $5 billion for non-compliance with a 2012 Consent Order and for other FTC Act
violations).
204
Hovenkamp, supra note 4, at 342.
205
For examples of non-discrimination obligations see for example the telecommunications
sector where common carriers’ tariffs, including tariffed terms and conditions and rates,
must be “just and reasonable and may not be unreasonably discriminatory”
Communications Act of 1934, Section 201(b) and 202(a), as amended.
206
FPC v. Hope Nat. Gas Co., 320 U.S. 591 (1944) (establishing that the standard for
courts’ review of rate-setting under the Natural Gas Act was that they must be “just and
reasonable”). Also see William Boyd, Just Price, Public Utility, and the Long History of
Economic Regulation in America, 35 YALE J. ON REG. 721 (2018) (tracing a history of
public utility rate-setting).
207
See, e.g., New State Ice v. Liebmann, 282 US 262 (1932) (entry restrictions imposed on
businesses manufacturing, selling and distributing ice held to violate the Fourteenth
Amendment’s due process clause because the business was private and not affected by a
public interest).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 51

carriage obligations,208 net neutrality mandates,209 and universal service

requirements.210

1. The Convergence of Antitrust and Utilities

Digital antitrust and regulatory attitudes are converging toward the

bottom left of ARD: decentralizing aims & ex ante regulation. Antitrust
enforcers are increasingly sensitive to antitrust remedies’ role in pre- and
restructuring digital ecosystems, and sectoral regulation is now aimed
proactively at decentralizing digital markets, enabling new entrants to
compete against incumbents, promoting consumer choice (see Fig. 3).

Figure 3

Lawsuits against Google and other big tech companies challenge power
over vertically integrated platform models such as the combination of
Android, Chrome and YouTube, and networks of agreements between
providers of upper-layer functionality (e.g. Google Search) and of lower-
layer operating systems (e.g. Apple’s iOS). These cases seek to impose
obligations and structural remedies on companies to enable long-term
competition. At the regulatory level, sectoral Bills such as AICOA or the

208
See, e.g., the Hepburn Rate Act of 1906. Section 1(4) imposed a common carrier
obligation on pipelines to “provide . . . transportation upon reasonable request therefor” and
section 1(5) required common carriers to publish their rates, which must “be just and
reasonable; and every unjust and unreasonable charge for such service or any part thereof is
prohibited and declared to be unlawful.”
209
See the Communications Act of 1934, as amended by the Telecommunications Act of
1996, (regulating telcos but not information-service providers as common carriers. Also
see Nat’l Cable & Telecomm. Assn. v. Brand X Internet Servs., 545 U.S. 967 (2005).
210
Communications Act of 1934, Pub. L. No. 73-416, para 1, 48 Stat. 1064, 1064 (the
industry should be regulated “so as to make available, so far as possible, to all the people of
the United States a rapid, efficient, Nation-wide, and world-wide wire and radio
communication service with adequate facilities at reasonable charges”).

Electronic copy available at: https://ssrn.com/abstract=4275143

52 DRAFT [Feb-23

Open App Markets Act advanced a vision of digital platform markets as more
competitive and contestable by default. They imposed statutory obligations
and liability on large bottlenecks and incumbents.
In the past, too, antitrust remedies have had long-lasting structural effects.
The Microsoft settlement is one example.211 The corpus of essential facilities
and duty-to-deal exceptions in antitrust is also forward-looking: it imposes
quasi-regulatory duties on certain firms.212 It has been argued that there are
reasons to revive the essential facilities doctrine in technology markets.213
Duties to license key telecommunications patents on “Fair Reasonable and
Non-Discriminatory” terms are another example of pre-emptive
procompetitive measure. These open the market to competitors and promote
the development of innovative products based on these patents.214 Overall,
ARD highlights that regulatory attitudes in digital markets are far from
“laissez-faire.” The core contestation is around the form that ex ante
procompetitive measures must take. This finding is significant for reasons I
now explain.

2. Lessons from ARD

The visual convergence of antitrust and regulation efforts in the bottom-

211
See, e.g., Lina Khan’s response to Kara Swisher, LIVE CNBC Capital Exchange: Big
Tech, big deals and a new era of antitrust — 1/19/22, YOUTUBE, at 20’50 (Discussing
Microsoft’s acquisition of Blizzard and Chinese competition.)
212
A number of cases have led to the demise of the essential facilities doctrine and its
construal as a very narrow exception. United States v. Terminal R.R Ass’n of St. Louis,
224 U.S. 383, 394 (1912); Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S.
585, 601–05 (1985); Trinko, 540 U.S. 398, 410– 11 (2004). See, especially, MCI
Commc’ns Corp. v. AT&T Co., 708 F.2d 1081, 1132–33 (7th Cir. 1983) (establishing the
essential facilities’ test: “(1) control of the essential facility by a monopolist; (2) a
competitor's inability practically or reasonably to duplicate the essential facility; (3) the
denial of the use of the facility to a competitor; and (4) the feasibility of providing the
facility”).
213
See Nikolas Guggenberger, Essential Platforms, 24 STAN. TECH. L. REV. 237 (2021)
(arguing that the essential facilities exception should be applied more expansively in tech
markets); Erik Hovenkamp, The Antitrust Duty to Deal in the Age of Big Tech, 131 YALE L.
J. 1483 (2022) (arguing that two separate lines of cases on duty to deal should be
distinguished in light of tech platform challenges.
214
See, e.g., Broadcom Corp. v. Qualcomm Inc., 501 F.3d 297 (3rd Cir. 2007); Microsoft
Corp. v. Motorola, Inc., 696 F.3d 872 , 875 (9th Cir. 2012); Apple Inc. v. Samsung Elecs.
Co., No. 11-CV-01846, 2012 WL 1672493 (N.D. Cal. May 14, 2012); FTC v. Qualcomm
Inc., 969 F.3d 974 (9th Cir. 2020). See also Mark A. Lemley, Intellectual Property Rights
and Standard-Setting Organizations, 90 CALIF. L . REV. 1889 (2002); Mark A. Lemley,
Timothy Simcoe, How Essential Are Standard-Essential Patents? 104 CORNELL L. REV.
607 (2019); Carl Shapiro & Mark A. Lemley, The Role of Antitrust in Preventing Patent
Holdup, 168 U. PA. L. REV. 2019 (2020); Peter Swire, supra note 9 (framing data sharing
obligations through the lens of FRAND).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 53

left quadrant of ARD illustrates a growing appetite for hybrid approaches that
combine faith in competition and consumer choice with recognition of the
public value of infrastructure. It carries more than one lesson.
First, perhaps most obviously, the convergence toward ex ante
decentralizing legal intervention in digital markets shows that antitrust and
utility regulation are not the polar opposites they are sometimes claimed to
be in popular or policy discourse. Instead, they are best understood as points
on a common spectrum or map. Both antitrust and regulatory measures can
centralize and decentralize, both can act in structuring or reactive ways, and
they can be combined in creative ways.
Second, the convergence toward ex ante decentralization crystallizes an
existing normative consensus in policy circles around promoting competition
and consumer choice in digital markets. This broad consensus is significant
in two ways. On the one hand, it crystallizes a recognition that markets, and
in particular competitive markets, are made and not found. Perfect
competition lives in textbooks. In real digital markets, competition requires
more than minimal ex post correction; it requires durable pre-emptive
structuring.215 This seems particularly salient in platform ecosystems where
markets are formed through conscious infrastructural choices. Without
holding those who control platform ecosystems in check, and ensuring
contestability over digital bottlenecks, digital markets cannot be
“competitive.” Competition requires creating options and standards,
imposing interoperability and non-discrimination obligations, it requires
enabling new entrants to compete with incumbents, and enabling the exercise
of individual choice and switching.
On the other hand, the overlap is also reflective of – one might say
“neoliberal” – faith in individual preferences in markets. Indeed, the vision
motivating reforms such as AICOA, as well as the vision underlying some
antitrust lawsuits, is one of perfect competition and contestability in digital
markets: a scenario where a multiplicity of platform providers can co-exist
and offer services to users and customers who in turn can switch iteratively
between platforms. The hope is that a mix of antitrust lawsuits and regulatory
reforms might create a radically more decentralized platform economy
governed by individual preferences and entrepreneurial freedoms, instead of
the whims of platform giants.
Structuring markets this way reflects the same faith in autonomous self-
correction that pervades the literature discussed in Part I. It reflects a
normative consensus between progressives and conservatives that
autonomous competitive markets must be shielded from social and political
pressures including questions of privacy or equality in an economy. This
215
But see Easterbrook, supra note 82, (arguing in favor of underenforcement and the ex-
post nature of antitrust).

Electronic copy available at: https://ssrn.com/abstract=4275143

54 DRAFT [Feb-23

consensus makes it difficult to incorporate normative concerns other than

negative freedom and efficiency into the regulation of platform markets.
Digital economies are not just “markets” where offer and demand meet; they
are also inputs into a variety of social and political activities. This suggests
that digital regulation should emphasize values other than choice and
entrepreneurial freedom. Antitrust and regulatory strategies could together
also advance goals considered exogenous to competition (e.g. fighting
surveillance, creating a plurality of options for engaging in public discourse,
promoting egalitarian content moderation rules).
Recognizing that the preference for decentralized marketplaces governed
by individual choices is only one among a plurality of possible approaches
opens competition policy to procedural and normative contestation and begs
for a new methodology.

3. Stretching ARD

ARD is a visual, explanatory device; it does not purport to explain the

role and goals of antitrust, regulation or competition law in contemporary
democratic societies. Although textbook views tend to place antitrust in the
top left quadrant and regulation on the bottom right, these bodies of norms
and remedies could be mapped differently. Remedies such as interoperability
are part of both antitrust and utilities law and can be deployed to serve a range
of different functions and ends.216 Far from obscuring this plurality of
functions and rationales, ARD elucidates that the province of these bodies of
law is irreducibly contestable.
There are ways in which antitrust law favors centralized power rather than
decentralized competition. As the Supreme Court noted in Trinko, the
opportunity to charge monopoly prices “is an important element of the free-
market system” and “attracts 'business acumen' in the first place; it induces
risk taking that produces innovation and economic growth.”217 Chicago
School proponents Robert Bork and Richard Posner have emphasized the
efficiencies of certain monopolies.218 Judge Easterbrook’s view of markets as

216
Note that a similar point is made by Filippo Lancieri & Caio M.S. Pereira Neto,
Designing Remedies for Digital Markets: The Interplay Between Antitrust and Regulation,
J. OF COMP. L. & ECON. 1 (2021) (arguing for synergies between antitrust and regulatory
methods).
217
Trinko, at 407.
218
ROBERT BORK, THE ANTITRUST PARADOX (1978) and see Posner, supra note 154, at 564
(“the effort of a businessman to monopolize a market by producing at a cost so low as to
drive out his competitors and deter new entry or, the monopoly achieved, to improve his
return by lowering his costs still further is not at all reprehensible.”). For a critical analysis
of this view see Rudolph J. Peritz, A Counter-History of Antitrust Law, 1990 DUKE L. J.
263 (1990) (an alternative to Chicago’s history of antitrust) and see Sanjukta Paul, Antitrust

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 55

inherently self-correcting also justified erring on the side of under-

enforcement, forgiving many acts of monopolization.219 In the 1930s, Ronald
Coase theorized the firm as a space immune from the rules of competition,
separate from the competitive order, where concentrated power and
hierarchical relations were acceptable and legitimate.220 Sanjukta Paul
diagnoses these trends and explains that “antitrust law tends to allocate
coordination rights, across doctrinal areas, according to criteria that
systematically prefer concentrated control over dispersed coordination or
cooperation.”221 According to her, antitrust law is far from an ex post body of
law with decentralizing effects, it in fact operates as an ex ante body of law
that centralizes market power.
Traditional ex ante centralizing utility measures are rare, if inexistent, in
technology markets. Processes of regulation do not take digital markets as
naturally monopolistic in a static sense. They are instead modular and
incremental. Ex ante rate-setting is infrequent because price is not necessarily
a central issue. Quality and technical standards as well as transparency rules
are often designed in an ex post dynamic and reactive manner. For example,
the modern internet is built on technical standards and protocols managed and
updated by the Internet Engineering Task Force (IETF), an association of
private network designers, operators, vendors, and researchers.222 Similarly,
the World Wide Web Consortium (W3C), an association of private
companies and other stakeholders, is responsible for developing technical
standards for the web.223 Besides technical standards, privacy and disclosure
rules also sit ambiguously across ex ante and ex post paradigms. The
California Consumer Privacy Act (CCPA), for example, establishes ex ante
obligations for businesses to reactively respond to data access and portability
requests

A business that receives a verifiable consumer request … to access personal

information shall promptly take steps to disclose and deliver, free of charge …

as Allocator, supra note 42 (re-interpreting antitrust law with a neo-Brandeisian

inclination). Note that the Harvard School of antitrust also followed a total welfare
standard, see Alan J. Meese, Debunking the Purchaser Welfare Account of Section 2 of the
Sherman Act: How Harvard Brought us a Total Welfare Standard and Why We Should
Keep it, 85 N. Y. U. L. Rev. 659, 690-698 (2010) (explaining the appeal of the total
welfare-productive efficiencies-based account of antitrust that favors firms over a direct
consideration of consumers’ well-being).
219
Easterbrook, supra note 82, at 2.
220
Ronald H. Coase, The Nature of the Firm, 4 ECONOMICA 386 (1937) (explaining the
nature of the firm in terms of information externalities).
221
Sanjukta Paul, Antitrust as Allocator, supra note 42, at 383.
222
About, INTERNET ENGINGEERING TASK FORCE, https://www.ietf.org/about/.
223
About W3C, THE WORLD WIDE WEB CONSORTIUM, https://www.w3.org/Consortium/.

Electronic copy available at: https://ssrn.com/abstract=4275143

56 DRAFT [Feb-23

the personal information required...224

Interoperability and data portability remedies are versatile.225 This is a

family of remedies that sits across antitrust and various regulatory regimes.
Interoperability in particular is a term that exists in both antitrust law and
regulatory law and many existences.226 The UK Furman Report states that
digital marketplaces’ tendency to tip could be remedied through measures
including what the report calls “data mobility,” portability, and
interoperability, whose overall promise is to create greater choice in digital
environments.227 Interoperability is “the ability to transfer and render useful
data and other information across systems, applications, or components”228
or “a technical mechanism for computing systems to work together – even if
they are from competing firms.”229
These remedies structure and react, centralize and decentralize productive
functions. They illustrate that the boundaries between antitrust, regulation,
and utilities are fluid. Mapping regulatory possibilities in the digital
economy, therefore, entails viewing existing remedies as part of contested
families of approaches whose precise design and application depends on
function, purpose and context.

224
California Consumer Privacy Act, section 1(d), Cal. Civ. Code. § 1798.100. (d)
(hereafter CCPA). The California Privacy Rights Act of 2020, coming into force in January
2023, has added protections to the CCPA.
225 Elettra Bietti, Explainer: Competition, Data and Interoperability in Digital Markets,

PRIVACY INTERNATIONAL, Aug. 20, 2020,

https://privacyinternational.org/explainer/4130/explainer-competition-data-and-
interoperability-digital-markets (a taxonomy of interoperability). Also see Jacques Crémer,
Yves-Alexandre de Montjoye & Heike Schweitzer, Competition Policy for the Digital Era,
PUBLICATIONS OFFICE OF THE EUROPEAN UNION 83 (2019) (describing different forms of
interoperability); Michael Kades & Fiona Scott Morton, Interoperability as a competition
remedy for digital networks, EQUITABLE GROWTH WORKING PAPER SERIES 2 (Sept. 2020)
https://equitablegrowth.org/working-papers/interoperability-as-a- competition-remedy-for-
digital-networks/ (arguing for interoperability as a remedy in tech platform settings).
226
See Swire, supra note 9.
227
Jason Furman et al., Unlocking digital competition, UK Government 5, 64 (Mar. 13,
2019) https://www.gov.uk/government/publications/unlocking-digital-competition-report-
of-the-digital-competition-expert-panel (a Report on digital challenges and competition).
228
Urs Gasser, Interoperability in the Digital Ecosystem, Berkman-Klein Center Research
Publication No. 2015-13 (July 6, 2015),
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2639210 (defining interoperability).
229
Ian Brown, Interoperability as a Tool for Competition Regulation, OPEN FORUM
ACADEMY REPORT 5 (Nov. 2020), (an overview of interoperability in the digital economy).

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 57

B. Pragmatism in the Digital Platform Economy: Integrating

Remedies and Values

Instead of assuming that antitrust has functional and normative

boundaries, ARD clarifies that it is coextensive with a panoply of
complementary regulatory strategies along functional and normative lines. A
joint approach to antitrust and utilities is not only versatile but can help
address harms that sit in the forgotten interstices between legal domains.
How, then, should regulators think about integrating antitrust and other legal
strategies?
What is needed is a method that reconfigures market regulation efforts
along functional, instead of doctrinal, axes.230 This method is pragmatic in
that it allows pooling bodies of doctrine and remedies in creative ways that
are more responsive to contextual conditions. The methodology has two
dimensions. First, it allows the production and integration of procedural and
remedial options (means) in ways that would not have been possible through
a doctrinally siloed approach to digital governance. For example, the
interactions between antitrust, consumer and privacy law, not least in the
DoubleClick case discussed above, could be significantly enriched and give
way to new procedural synergies between the FTC’s Bureau of Competition
and Bureau of Consumer Protection or the creation of a new agency. Second,
the method clarifies the normative stakes of policymaking (and scholarship)
on tech and market governance (ends). Envisioning digital regulation as a
process that integrates means and ends, the remedies/processes and the
values/interests that remedies advance or undermine, makes it adaptable to
context and better equipped to prevent future (and interstitial) harm.

1. Integrating Means: Remedies, Processes, Institutions

Filippo Lancieri and Caio Mario Pereira Neto have provided a helpful
error-cost framework which guides the adoption of remedies across different
areas of economic regulation including antitrust and utility-like regulation.231
They argue that “[r]emedy design […reflects] a continuum distribution of
possibilities that can and should be modulated in intensity: authorities must
choose one or a couple of solutions from a range of available
interventions.”232 They see regulation as a “learning process that enables the
reduction of over- and under-enforcement errors in remedy design and

230
The term “pragmatic” refers to John Dewey’s approach to regulation which consists in a
contextual definition and tailoring of means and ends. See DEWEY, supra note 69.
231
Lancieri & Neto, supra note 216.
232
Id. at 30.

Electronic copy available at: https://ssrn.com/abstract=4275143

58 DRAFT [Feb-23

implementation over time.”233 They suggest a framework that first identifies

the unlawful conduct, then assesses remedies functionally, and finally
compares different possible institutions and authorities able to address the
conduct or impose the remedy. They also mention a fourth step which
includes monitoring any remedies after they are imposed.
William Rogerson and Howard Shelanski also offer a valuable method
for combining antitrust and regulatory remedies.234 They identify some
virtues and limits of antitrust enforcement: it affords stability, develops
slowly, is case-by-case and ex post. Antitrust however leads to
underenforcement, it leaves some pressing issues, such as predatory pricing,
unexplored because case selection is driven by the market and not by what
expert regulators consider urgent.235 The “novel challenges” presented by
digital platforms suggest the superiority of an approach that relies on an
informed and coordinated regulator in this context.236 In particular, Rogerson
and Shelanski suggest a test to determine whether regulation is needed: (1)
does the monopoly conduct meet “criteria of economic harm, durability and
remediability”? and (2) are “available regulatory mechanisms … in fact likely
to remedy … harm[…] without creating equally harmful side-effects for
consumers”?237
These approaches offer a valuable cost-benefit lens through which to
examine questions of economic regulation in the platform economy. They
open the door to synergies between antitrust and other legal strategies in the
digital platform economy. For example, it might be most effective for
regulatory agencies to collaborate, e.g. by including privacy scrutiny in
merger investigations, or it might instead be preferable to create an agency
exclusively tasked with digital market regulation. What these approaches
miss, however, is that there is no neutral universal framework by reference to
which one defines “harm” and assesses “side-effects for consumers” so as to
determine appropriate remedies in a given context. Rogerson and Shelanski’s
view on what constitutes a “harm” in this context might not be the view of a
privacy or speech scholar considering the same market. While some harms
may be universally recognized as urgent and in need of regulation, others are
irreducibly contested. A methodological shift that overcomes doctrinal
boundaries in enforcement cannot stop at an integration of existing tools and
procedural contestation. The object of enforcement, the harms, values and
goals that are addressed or pursued through regulation must themselves
become contestable.

233 Id.
234
Rogerson and Shelanski, supra note 3.
235
Id. at 1917-21.
236
Id. at 1923-24.
237
Id. at 1924.

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 59

2. Integrating Ends: Values and Goals

Revisiting the normative premises of existing and future enforcement

approaches is key to pragmatic policymaking in the digital context.
Policymaking is not just a procedural endeavor, or a deliberation about
means. It is also a dynamic deliberation about values, goals, or ends. To
justify any integration effort and say that it is normatively appropriate
requires two steps. First, one must define the ends of regulation, that is the
interests, harms, goals or values that justify the integration of policy
approaches, remedies, constraints. Ends can be as specific as increasing the
variety of travel aggregator options available on Google Search, as broad as
limiting data collection in online advertising markets or increasing data
mobility between social media operators, or they can be more abstract such
as enhancing consumer choice and autonomy or advancing competition and
innovation. Second, one must return to the question of available remedies,
doctrines and institutions in defining the available means, i.e. the ensemble
of feasible remedies that, together or in the alternative, are appropriate to
advancing the ends. The more confined the ends, the more one might be
tempted to apply an existing body of doctrine or “means” to pursue it. The
broader the ends, the wider the range and combination of feasible means to
advance them. Investigating means need not be limited to existing remedies
and can consist in designing new remedies that better address given ends. The
more radical and novel one’s definition of ends, the more creativity it will
require as regards the means. Algorithmic disgorgement, for example, is a
remedy designed to address emerging algorithmic abuses.238
This methodology overcomes doctrinal siloes between antitrust and other
domains and addresses harms that arise in their interstices by generating
contestation on the ends of regulation and the means of pursuing them. The
method can guide the exercise that legislators, stakeholders and experts
engage in when determining a new policy line, reform package, bill, set of
guidelines or strategy for governing the digital platform economy. It can also
guide the way a court might think about the boundaries and relation between
existing doctrines and remedies, as well as how it interprets the distribution
of powers and competences among regulatory agencies.
To take an example, regulating online display advertising – advertising
across web and mobile – requires asking, first, what advertising should be for
on the web or on mobile. Is advertising about informing the public? Is it about
enabling free internet access? Is it simply aimed at creating addictive
attention ecosystems that benefit platform gatekeepers and advertisers? Is it
238
See In the Matter of Everalbum, Inc., 2021 WL 1923172 (2021); Tiffany C. Li,
Algorithmic Destruction, 75 SMU L. REV. 479 (2022) (discussing such remedies).

Electronic copy available at: https://ssrn.com/abstract=4275143

60 DRAFT [Feb-23

a system that ensures a fair distribution of economic advantages in the digital

economy? Clarifying the ends of advertising regulation necessarily
conditions one’s answer to the question of means, i.e. which remedies and
institutional strategies must one adopt to regulate those markets? Good
answers seek ends that strike a balance between the interests of users,
advertisers, content producers and platform gatekeepers. It is possible that, in
some contexts, “maximizing individual consumer choice” or “maximizing
consumer welfare” are the overall best answers to the question of ends.
However, one cannot assume that the same ends are worth pursuing across
industries and market segments. “Choice” or “welfare” are contested
concepts and must be specified by reflexively articulating their stakes in
specific contexts. Regulating digital markets like online display advertising,
in order words, cannot be divorced from the question of what these markets,
concretely, are for.
To sum-up, recognizing the antitrust-regulation convergence generates
new questions not only on the means but also on the ends of regulatory
intervention in digital platform markets. It prompts a departure from abstract
conventional slogans such as “competition”, “innovation” or “choice” in
favor of more concrete engagement with situated harms, weighted value
determinations and proportionate remedies.

C. Pragmatism in Practice: Markets and Society’s Dynamic Co-

Dependence

If applied in policymaking, the pragmatic method just introduced could

have significant implications in the short and longer term, proving that
antitrust and other domains of law such as utilities or consumer law change
by intersecting with and reacting to one another. I now turn to two examples.
My aim is to show that designing legislative or administrative action requires
a reflective approach to the ends being pursued and the means at hand: how
values including autonomy, privacy, free expression or self-determination
define and are defined by markets, and how policymakers consciously
construct markets and society.

1. Beyond Individual Choice in Digital Platform Markets: The Limits of

AICOA

As discussed, “light-hand procompetitive regulation” has recently been

considered in Congress. Key examples include AICOA as well as the
European DMA, which is AICOA’s EU counterpart. As noted, these break
with digital markets’ underenforcement past. They are ex ante instruments
intended to impose certain statutory obligations on platform gatekeepers like

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 61

Google so as to enable more competition and consumer choice.

The philosophy underlying these proposals is an optimistic vision of
competition in platform markets fostered by structural and behavioral
measures such as prohibitions on self-preferencing, discriminatory terms of
service, bundling and obligations to respond to interoperability and data
access requests. The ends or goal of these reforms is competition and
increasing entrepreneurs and consumers’ freedoms in digital markets, the
ability of businesses to enter new markets, and of consumers to move
between platform providers. The idea is to ensure enough digital options and
to give users the ability to switch between these options.
AICOA and similar reforms, however, have limits. The ultimate power to
determine the shape of platform infrastructures and the available consumer
options is left to entrepreneurs and private companies. These reforms remain
committed to a vision of self-correcting markets as autonomous efficient
devices guided by preference signals and divorced from other social and
structural preoccupations. As such they favor some interests over others.
AICOA continues to assume that there is a realm of ‘competition’ detachable
from the ordinary operation of law and politics that can be switched on then
left ‘alone’. What is missing is a realistic assessment of the values at play in
the construction and reconstruction of markets. What do consumers and
businesses need and want from these markets and infrastructures? Favoring
switching and the ability of individuals to opt-in or out of pre-existing
platform infrastructures is a normative choice. It is a political compromise
too. Against the grain of many predictions, the tendency of AICOA or DMA
to treat competition as a neutral valueless ideal hinders debate on what really
is at stake. More competitive digital platform ecosystems won’t afford more
empowered digital choices. They are likely to keep people locked-into
opaque ecosystems prone to surveillance and other extractive logics. AICOA-
like reforms may bring about more options, but these options are likely to
remain qualitatively limited, they will continue to be motivated by profit
rationales.
An antidote is to think about the ends of regulation in platform settings
differently. Instead of relying on the abstract terminology of individual
choice and competition, we should think of the goals of regulating each given
productive function. What values should be embedded or dynamically
infused in search or advertising ecosystems? What productive structures
would best advance these values? The antidote, in other words, is to think
beyond the doctrinally siloed and inflexible goals of antitrust law, utilities or
of data protection law, and instead to ask what terms like competition and
choice ought to mean and how they can be embedded in digital platform
settings, through a variety of institutional strategies and remedies.
Embedding them might entail combining institutional efforts that have so

Electronic copy available at: https://ssrn.com/abstract=4275143

62 DRAFT [Feb-23

far been understood as separate, or relying on structural (ex-ante), court-

based (ex-post), centralizing and decentralizing mechanisms at once to create
more accountable and competitive search or advertising regimes.

2. The FTC: Between Antitrust and Consumer Law

A helpful place to experiment with disciplinary convergences and

synergies in the short term could be the Federal Trade Commission. Section
5(a)(1) of the Federal Trade Commission Act has two parts, and both refer to
the notion of “unfairness”: “Unfair methods of competition in or affecting
commerce, and unfair or deceptive acts or practices in or affecting commerce,
are hereby declared unlawful.”239 The Federal Trade Commission’s Bureau
of Competition has powers to police “unfair methods of competition” which
include conduct prohibited by the Sherman and Clayton Acts.240 In 1938, the
FTC Act was amended to include “unfair and deceptive acts or practices.”241
The latter are addressed separately from “unfair methods of competition” by
the FTC’s Bureau of Consumer Protection.242 The division between FTC
bureaus is largely taken for granted in legal scholarship and policy circles.
Antitrust and consumer protection are interpreted as separate branches of law
that confer different investigatory and enforcement powers to the FTC and
respond to different normative concerns.
However, a few of these branches’ normative “ends” overlap and seem to
converge even beyond digital platform markets. The history of “unfairness”
at the FTC is an example that is far from Janus-faced.243 Both unfair methods
of competition and unfair or deceptive acts can be historically traced back to
the common law doctrine of “unfair competition” and linked to concerns
about abusive and harmful practices relating to industrialization and
commerce during the late XIXth and early XXth century.244 It is also
plausible that the notion of “unfair methods of competition” requires wide

239
15 U.S.C. § 45.
240
Bureau of Competition, https://www.ftc.gov/about-ftc/bureaus-offices/bureau-
competition.
241
Federal Trade Commission Act, Pub. L. No. 63-203, 38 Stat. 717, 719 (1914)
(empowering the FTC to prevent “unfair methods of competition” through regulation); and
see J. Howard Beales, The FTC's Use of Unfairness Authority: Its Rise, Fall, and
Resurrection, Speech at the Marketing and Public Policy Conference, Washington, D.C.,
May 30, 2003, https://www.ftc.gov/news-events/news/speeches/ftcs-use-unfairness-
authority-its-rise-fall-resurrection (on the FTC’s unfairness authority).
242
Bureau of Consumer Protection, https://www.ftc.gov/about-ftc/bureaus-offices/bureau-
consumer-protection.
243
See Luke Herrine, The Folklore of Unfairness, 96 N.Y.U. L. REV. 431, 454 (2021)
(describing the origins of “unfair methods of competition” as part of a longer history of
“unfair competition”).
244
Id.

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 63

construction that extends beyond the current bounds of the Sherman and
Clayton Acts and justifies reforming the existing antitrust laws.245 As argued
by Sandeep Vaheesan, in their antitrust caselaw “courts do not
indiscriminately protect competition, but rather restrict certain forms of
competition they deem unfair” meaning that they rely on a moral
understanding of unfairness as a guiding principle for enforcement.246
Reliance on unfairness as a moral principle suggests that antitrust and
consumer protection law are not so separate in their purposes. In Europe,
scholars have also argued that the concept of “fairness” is a point of
normative connection between disjointed branches of legal enforcement in
markets including competition and consumer protection law.247
Other normative synergies between consumer law and antitrust can be
articulated through the notion of consumer choice. Together, antitrust and
consumer law ensure that consumers have options and that they have the
ability to make active choices among them without being impaired in their
choosing by unfair or deceptive commercial practices.248 Competition and
antitrust law have as their mission “to protect the array of options in the
marketplace.”249 As noted by Robert Lande, “[t]he antitrust laws are intended
to ensure that the marketplace remains competitive so that worthwhile
options are produced, and this range of options is not to be significantly
impaired or distorted by anticompetitive practices.”250 Antitrust cases and
other procompetitive regulatory frameworks in the digital economy regularly

245
Sandeep Vaheesan, The Morality of Monopolization Law, 63 WILLIAM & MARY L. REV.
ONLINE 119 (2022) (layering a moral lens on competition law).
246
Sandeep Vaheesan, Antitrust Law’s Unwritten Rules of Unfair Competition,
PROMARKET, November 17, 2021, https://www.promarket.org/2021/11/17/antitrust-
enforcement-unfair-competition-ftc/ (discussing unfair competition as a normative guide
for antitrust).
247
Inge Graef, Damian Clifford & Peggy Valcke, Fairness and enforcement: bridging
competition, data protection, and consumer law, 8 INT. DATA PRIV.’Y L. 200 (2018)
(fairness as a common root of three legal areas); Francisco Costa-Cabral and Orla Lynskey,
Family Ties: The Intersection between Data Protection and Competition in EU Law, 54
COMM. MARKET L. REV. 11, 18 (2017) (synergies between antitrust and data protection
law).
248
Neil W. Averitt & Robert H. Lande, Using the “Consumer Choice” Approach to
Antitrust Law, 74 ANTITRUST L.J. 175, 182 (2007); Neil W. Averitt & Robert H. Lande,
Consumer Sovereignty: A Unified Theory of Antitrust and Consumer Protection Law, 65
ANTITRUST L.J. 713 (1997); Neil W. Averitt & Robert H. Lande, Consumer Choice: The
Practical Reason for Both Antitrust and Consumer Protection Law, 10 LOY. CONSUMER L.
REV. 44 (1998) (grounding antitrust and consumer protection law in consumer choice).
249
Neil W. Averitt & Robert H. Lande, Using the “Consumer Choice” Approach to
Antitrust Law, 74 ANTITRUST L.J. 175, 182 (2007) (defending a conception of antitrust
grounded in consumer choice).
250
Robert H. Lande, Consumer Choice as the Ultimate Goal of Antitrust, 62 U. PITT. L.
REV. 503, 503 (2001) (defending a conception of antitrust grounded in consumer choice).

Electronic copy available at: https://ssrn.com/abstract=4275143

64 DRAFT [Feb-23

refer to “consumer choice” as a goal or proxy for remedying consumer and

antitrust harm.251 But what is choice? Choice can be interpreted as a formal
ideal – the short-term satisfaction of consumer preferences – or as a
substantive qualitative ideal.252 For example, in the telecommunications and
mass media context, having the choice doesn’t just mean exercising the will
to make a choice. It means having a plurality of media options to choose
between, sufficient information to choose between them, and also having
sources that are sufficiently differentiated and of high quality to make it
meaningful to choose between them. Media pluralism and the promotion of
values unrelated to efficiency or individual preferences are pre-conditions to
ensuring choice in this context. In the digital platform economy, similarly,
consumer choice is pre-conditioned by the ability to access sufficiently good
options. In other words, consumer choice requires mobilization against
monopolistic tendencies and it requires promoting more pluralistic and
egalitarian spaces for public discourse.
When interpreted through the lens of fairness or consumer choice,
antitrust and consumer law do not appear as sharply distinct as their recent
doctrinal interpretations might suggest. The convergence of notions of
fairness and choice in technology platform markets seems an opportunity to
experiment with new procedural synergies between these two separate bodies
of law. Experimenting means articulating the notions of consumer choice and
fairness, and the remedies that advance them, with increasing specificity.
Shying away from the need to define ends and integrate means by relying on
the abstract language of competition or choice will not bring regulators closer
to the realities of digital platform markets. Instead, experimentation could
mean facilitating collaborations between the FTC’s Consumer Protection and
Competition Bureaus. These could jointly formulate policy objectives such
as fairness or consumer choice in specific digital industry segments. Their
outputs could include guidelines or rules delineating technical standards for
fair recommendation systems or advertising-related transactions.253 One
could alternatively be as ambitious as creating a federal agency with powers

251
First Complaint, FTC v. Facebook Inc., No. 1:2020cv03590 (D.D.C. 2020) (defining
harms to consumers in social media markets in terms of loss of consumer choice, and
harms to advertisers in terms of loss of advertising options).
252
Howard Shelanski discusses the distinction between two conceptions of the public
interest along these same lines in Howard A. Shelanski, Antitrust Law and Mass Media
Regulation: Can Merger Standards Protect the Public Interest? 94 CAL. L. REV. 371, 399
(2006) (arguing that “antitrust has no mechanism by which to evaluate or prioritize variety
and quality to suit particular social goals over the diversity and quality levels that
consumers demand”).
253
As regards rulemaking, see Rohit Chopra & Lina M. Khan, The Case for “Unfair
Methods of Competition” Rulemaking, 87 UNIV. CHI. L. REV. 357 (2020) (the case for the
FTC’s rulemaking authority)

Electronic copy available at: https://ssrn.com/abstract=4275143

Feb-23] DRAFT 65

to enforce consumer standards in digital markets in ways that could erode the
differences between antitrust and consumer protection while focusing
enforcement on the question of “fairness” in markets.

CONCLUSION

The ideal of self-correcting competitive markets as the economic and

policy default has hindered and delayed proper acknowledgement of new
manifestations of harm, control and knowledge asymmetries in technology
platform markets. Against the grain of many conventional accounts, antitrust
and economic regulation coexist by transforming each other dynamically.
They should be considered part of one single coextensive toolbox that
encompasses ex ante and ex post, as well as centralizing and decentralizing
regulatory strategies. Viewing the relationship between antitrust and
regulation as a relationship between two distinct domains is misleading.
Antitrust, regulation and utility regulation are part of a joint umbrella of
approaches to law and institutions whose boundaries are irremediably fluid.
Particularly in digital platform contexts, how we interpret their relationship
depends on our normative priors, and on whether we are open to viewing
markets and society as co-dependent. Any discussion of disciplinary
boundaries and remedies cannot be divorced from a normative contestation
of the values that the regulation of different markets or productive functions
can and ought to advance. Focusing on what markets or productive processes
are for, instead of whether they are efficient or what volume of transactions
they afford in the abstract, means re-aligning their functions to society’s more
pressing needs.

Electronic copy available at: https://ssrn.com/abstract=4275143