Thibault Schrepel-The Theory of Granularity PDF

THE THEORY OF GRANULARITY
A Path for Antitrust in Blockchain Ecosystems
Dr. Thibault Schrepel 1
Assistant Professor, Utrecht University School of Law

Faculty Associate, Harvard University’s Berkman Klein Center
INTRODUCTORY THOUGHTS .............................................................................................. 3

I. THE THEORY OF THE FIRM: A PILLAR OF ANTITRUST ANALYSIS
JEOPARDIZED BY BLOCKCHAIN ................................................................................ 6
A. The Theory of the Firm as a Pillar for Antitrust Law ................................................... 6
1. The Theory of The Firm: Not a Marginal Revolution ............................................. 6
2. The Theory of the Firm in Modern Antitrust Law ................................................ 10
B. The Theory of the Firm Vis-à-Vis Blockchain ............................................................. 15
1. Ronald Coase’s Logic in Light of Blockchain ....................................................... 16
2. Blockchain as an Escape to Antitrust and Competition Law ................................... 24
II. THE THEORY OF GRANULARITY: A NEW PATH FOR APPLYING
ANTITRUST TO BLOCKCHAIN ECOSYSTEMS ........................................................... 27
A. The Theory of Granularity as a Theoretical and Legal Framework for Blockchain ........... 27
1. A Theorization of Blockchain as an Institution..................................................... 27
2. Transposition to Antitrust and Competition Law: the Blockchain Nucleus ................. 35
B. The Theory of Granularity in Action........................................................................ 42
1. The Theory of Granularity in Support of Market Analysis ..................................... 42
2. The Theory of Granularity in Support of the Analysis of Anti-Competitive Practices .......... 45
CONCLUDING THOUGHTS ................................................................................................. 48
1 Assistant Professor at Utrecht University School of Law, Faculty Associate at Harvard
University’s Berkman Klein Center for Internet & Society, Associate Researcher at
University of Paris 1 Panthéon-Sorbonne and Invited Professor at Sciences Po Paris. I
would like to thank Vitalik Buterin (creator of the Ethereum) for our numerous
conversations about public blockchains; Oliver Hart (Professor of Economics at Harvard
University and recipient of the 2016 Nobel Prize in Economic Sciences) for our exchanges
about economic theory; Karl Friston (Professor of Imaging Neuroscience at UCL) for his
comments and reading recommendations about free energy; Nicolas Petit (Professor of
Competition Law at the European University Institute) for his suggestions regarding the
theory of the firm; Nathan Kaiser (Faculty Associate at the Berkman Center) for his
comments on blockchain-related sections, and Rhys Lindmark (MIT Media Lab) for his
suggestions regarding institutional cryptoeconomics. Thank you to Thymo Burgemeester,
Charlotte van Oirsouw, Chiara Pescetto and Christos Beloulis for their research assistance.
No outside funding was received or relied upon for this paper. I also do not have any
financial interest at stake in blockchain.
ABSTRACT
Modern antitrust and competition law relies extensively on the firm as

defined by Ronald Coase: a hierarchy reducing transaction costs thanks to
vertical control, where such control defines the firm’s boundaries.
Meanwhile, the governance of public permissionless blockchains is
horizontal. Transaction costs are minimized thanks to specific
characteristics that are singular to these blockchains and do not depend on
the verticality of relationships. The absence of vertical control to direct the
resources holds antitrust and competition in check.
Against this background, the present article introduces the “theory of

granularity,” which permits analysis of the roles played by each (group of)
participant in the horizontal governance of public permissionless
blockchains. On this basis, one may identify a “blockchain nucleus,” i.e., a
set of participants collaborating to ensure and maximize the blockchain
survival by “controlling” it all together. Antitrust and competition law
becomes applicable again as the nucleus serves as the basis for the definition
of the relevant market and market power, the assessment of practices’
legality, and liability assignment.
2
INTRODUCTORY THOUGHTS
T he theory of the firm is antitrust and competition law most fundamental

pillar. Initiated in an article authored by Ronald Coase in 1937 and
entitled “The Nature of the Firm,”2 that theory sought to answer a question
previously ignored: if markets are indeed efficient, why are firms emerging?
Ronald Coase responded in a simple and elegant way, stressing the idea that
firms make it easier to organize certain exchanges. Without naming it, Coase
introduced the concept of transaction costs—referring to all the costs that must
be incurred by the parties to complete a transaction—and explained that firms
emerge and extend into any place where they can minimize these costs.
The reason for this cost reduction, Ronald Coase said, is that firms are
hierarchically organized. Orders and directions are given from the top and
trickle down the hierarchy. As such, firms use “command and control”
mechanisms, allowing for the orderly coordination of the firm’s employees.
This control ultimately leads to reduced transaction costs associated with
activities such as researching the right partner, negotiating and drafting
contracts, etc. On the contrary, this kind of efficiency is not found on the
market where free economic agents compete with each other. The boundary
between the firm and the market is thus defined: where control stops, the
firm’s perimeter stops as well.
Antitrust and competition law did not wait for Ronald Coase to exist and
develop.3 Yet, as we intend to show, all modern case law and regulations are
built on the basis of the above-mentioned article. For this reason, studying
the “The Nature of the Firm” relevance over the decades has proven essential.
Ronald Coase’s explanations help in understanding the extent to which firm
expansion is a pro-competitive phenomenon, and conversely, which
practices implemented by such companies are anti-competitive.
In this study, we propose to analyze Coase’s article in light of public

permissionless blockchains at the level of the platform layer,4 which is
generally described under the generic term of “blockchain.”5 The platform
2 Ronald H. Coase, The Nature of the Firm, 4 ECONOMICA 386 (1937).

3 The Sherman Act is dated July 2nd, 1890.
4 The “platform layer” is here intended to describe the base layer, such as the Ethereum,
on which decentralized applications are built on. Several scholars are also calling it the
“blockchain lawyer,” see, for instance, Primavera de Filippi, Greg Mcmullen, Governance of
Blockchain systems – Governance of and by distributed infrastructures, HAL.ARCHIVES-OUVERTES.FR
13 (Feb. 22, 2019), https://hal.archives-ouvertes.fr/hal-02046787/document
[https://perma.cc/RU5F-QXPJ].
5 In this article, the word “blockchain” refers to “public permissionless blockchains,”
3
layer allows for software to be built on top of it, thus providing participants
with different applications.6 The rules and constraints exercised on the
platform layer then determine the use of the software layers built on top of
it.7 As such, they have a great economic potential.
In the meantime, one cannot apply the legal solutions designed for
centralized platforms to the realm of blockchain. Blockchain is indeed an
“institutional technology,” i.e., a new way for coordinating economic
activities that does not have the characteristics of a firm or a market.8 The
applicability of antitrust and competition law (and, more generally, of
business law) is called into question insofar as the firm, which is the starting
point of competitive analysis, gives way to entities without any power of
command and control, and, consequently, without clearly defined
boundaries.
Public permissionless blockchains are in fact as in the public domain:9 no

access request is needed before consulting or exploiting the information they
contain; it is also not necessary to obtain authorization to register information
or implement transactions on them. There is no entrance or exit to public
permissionless blockchains; they are there. One must, therefore, question the
applicability of antitrust and competition law to such entities. We investigate
that issue in Part I of this Article, showing how the absence of inside and
outside the blockchain defeats antitrust law, which, traditionally, focuses on
effects outside the firm (I).
unless otherwise expressly stated. The platform layer allows any participant to read the data,
to register transactions, and to mine new blocks. Access to a blockchain defines whether
blockchains are public or private. When anyone can access it, the blockchain is public. On
the contrary, the blockchain is private when authorization is needed. On top of the public
or private nature of blockchain, writing permissions on the ledger define whether a
blockchain is permissioned and permissionless. When anyone can write on the blockchain,
it is permissionless. The blockchain is permissioned when only certain users can write on it.
It follows that public blockchain can be permissionless or permissioned while private
blockchain are, by definition, always permissioned.
6 It is therefore essential to focus first and foremost on the platform layer, which this
Article proposes to do, because “the market cap of the protocol always grows faster than
the combined value of the applications built on top, since the success of the application layer
drives further speculation at the protocol layer,” see Joel Monegro, Fat Protocols, UNION
SQUARE VENTURES (Aug. 8, 2016), https://www.usv.com/writing/2016/08/fat-
protocols/ [https://perma.cc/KGS6-7HLN]. For more on the distinction between the
platform layer and application layers, see Thibault Schrepel, Is Blockchain the Death of Antitrust
Law? The Blockchain Antitrust Paradox, 3 GEO. L. TECH. REV. 281, 295 (2019).
7 DE FILIPPI, supra note 4, at 4.
8 Sinclair Davidson, Primavera de Fillippi et al., Blockchains and the Economic Institutions of
Capitalism, 14 J. I. ECON 639 (2018).

9 Helen Eenmaa-Dimitrievaa & Maria José Schmidt-Kessen, Creating Markets in No-Trust
Environments: The Law and Economics of Smart Contracts, 35 COMPUT. LAW SECUR. REV. 69, 72
(2019) (calling public blockchains “a truly public space”); Mark A. Engelhardt, Hitching
Healthcare to the Chain: An Introduction to Blockchain Technology in the Healthcare Sector, 7 TECHNOL.
INNOV. MANAG. REV. 22, 28 (2017).
4
Part II of the present Article is dedicated to opening the blockchain black

box, similar to Coase opening of the firm’s black box in the 1930s, with the
intention to find a way to make antitrust and competition law applicable
again10 (II). The “theory of granularity” is introduced to this end with the
hope of facilitating the understanding the role being played by the smallest
element in a blockchain, a necessity due to the horizontal nature of
blockchain governance. We show how one can deduce the existence of a
“blockchain nucleus,” creating as such a legal fiction to which apply antitrust
and competition law. We finally demonstrate the utility of that theory at all
stages of antitrust and competition law analyses, whether it is in terms of
analyzing relevant markets and market power or evaluating anti-competitive
practices and assigning liability.
10 Dealing not with the question of antitrust and competition law (theoretical)
application, but its practical applicability, see Thibault Schrepel, Is Blockchain the Death of
Antitrust Law? The Blockchain Antitrust Paradox, 3 GEO. L. TECH. REV. 281 (2019); Thibault
Schrepel, Collusion by Blockchain and Smart Contracts, 33 HARV. J.L. & TECH. 117, 127 (2019).
5
I. THE THEORY OF THE FIRM:

A PILLAR OF ANTITRUST ANALYSIS JEOPARDIZED BY
BLOCKCHAIN
The firm, as defined by Ronald Coase, is the theoretical pillar on which

antitrust and competition law takes grip (A). Blockchain, a new institutional
way to transact, creates legal challenges to the extent that it does not have
similar characteristics allowing antitrust and competition law to be applied (B).
A. The Theory of the Firm as a Pillar for Antitrust Law
The firm theory is a pillar on which modern antitrust and competition law
is based, whether in the United States or in Europe (2). It represents the
advent of a new paradigm in economic theory that is characterized by
transaction costs and which has been subsequently incorporated into the
competitive analysis (1).
1. The Theory of The Firm: Not a Marginal Revolution
The theory of the firm, as first defined by Ronald Coase, emphasizes the
importance of transaction costs which are being reduced thanks to the
control exercised from the top of the firm (a). That theory has proven to be
the basis of modern microeconomics. It stands in its entirety to this day (b).
a. Highlights of Ronald Coase’s Article
In 1937, Ronald Coase published “The Nature of the Firm,” an article that he
wrote when he was twenty-one years old. His article, which contains no
mathematics and is just twenty pages long, remains to this day one of the
most cited publications in economic theory.11 Its impact cannot be
overestimated.12
11 Google Scholar counts about 45.000 citations, although the paper “had little or no
influence for thirty or forty years” according to R. H. Coase, The Nature for the Firm: Influence,
4 J. L. ECON. & ORG. 33 (1988). Explaining how Judge Bork helped rediscover Ronald
Coase’s article, see Alan J. Meese, Robert Bork’s Forgotten Role in The Transaction Cost Revolution,
79 ANTITRUST L.J. 30. The Swedish Academy eventually called it Ronald Coase’s “first
major study,” see The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 1991,
NOBELPRIZE.ORG (Aug. 29, 2019), https://www.nobelprize.org/prizes/economic-
sciences/1991/press-release/ [https://perma.cc/LUK3-HAAZ].
12 This is all the more true since the article findings’ are still valid, indeed, as Ronald H.
Coase explained himself, the firm has not changed since then, see Russ Roberts, Coase on
Externalities, the Firm, and the State of Economics, ECONTALK.ORG, (Aug. 30, 2019),
6
In it, Ronald Coase questions the existence of firms if, indeed, markets work
and provide economic agents with what they need. His explanation lies in
transaction costs.13 Ronald Coase lays down the idea that the conclusion of
a transaction involves costs, whether it is the costs of finding economic agents
on the market, the costs of negotiation, the costs of drafting a contract, etc.
The firm helps in reducing some of these costs, the reason why they emerge
and prosper. As such, firms are seen for the first time as an institutional
device.14 Ronald Coase just opened the black box of the firm.15
Ronald Coase then proceeds to explain why firms reduce these costs. His
explanations come down to the power of command and control that is exercised
from the top to the bottom of the firm.16 Transactions are just to be executed.
In Ronald Coase’s words, “in place of the complicated market structure with
exchange transactions is substituted the entrepreneur-co-ordinator, who
directs production,”17 and from this substitution follows cost reductions as
“by forming an organisation and allowing some authority (an ‘entrepreneur’)
to direct the resources, certain marketing costs are saved.18” He thus defines
the firm as “the system of relationships which comes into existence when the
direction of resources is dependent on an entrepreneur.”19
Ronald Coase particularly emphasizes the firm ability to deal with

contingencies during the performance of the contract. While firms manage
long-term relationships, the market mainly permits short-term contracts
based on the price mechanism.20 Thus, Ronald Coase argues, “it seems
http://www.econtalk.org/coase-on-externalities-the-firm-and-the-state-of-economics/
[https://perma.cc/SVJ7-9WWA].
13 Ronald Coase did not specifically name “transaction costs” as such, in fact, the term
had been first introduced by Carl J. Dahlman, The Problem of Externality, 22 J.L. & ECON.
141 (1979).
14 Boudreaux on Coase, ECONTALK.ORG (Oct. 28, 2013),
http://www.econtalk.org/boudreaux-on-coase/ [https://perma.cc/2T9V-AK63].
15 See Oliver E. Williamson, Opening the Black Box of Firm and Market Organization: Antitrust,
in THE MODERN FIRM, CORPORATE GOVERNANCE, AND INVESTMENT 11, 33 (Per-Olof

Bjeggren & Denis C. Mueller eds., 2009).
16 According to Cambridge Dictionary, command and control refers to “a situation in
which managers tell employees everything that they should do, rather than allowing them
to decide some things for themselves,” see the Cambridge English Dictionary, “Command
and Control.”
17 COASE, supra note 2, at 388.
18 Id. at 392.
19 Id. at 394. It should be noted that an alternative theory defends that the firm does not
more power over its employees than over independent contractors, see Armen Alchian &
Harold Demsetz, Production, Information Costs, and Economic Organization, 62 AMER. ECON.
REV. 777 (1972).
20 COASE, supra note 2, at 391-392 (arguing that a firm is likely to emerge in those cases
“where a very short term contract would be unsatisfactory”); Id. at 389 (defending that the
distinguishing mark of the firm is the supersession of the price mechanism); also Id. at 390
(explaining that the main reason why it is profitable to establish a firm is that using the price
mechanism has a cost).
7
improbable that a firm would emerge without the existence of uncertainty”21

on the market. This assumption is based on the theory of incomplete
contracts, according to which the contracting parties cannot anticipate all
the situations that may arise during the performance of their contract.22 In
this regard, the firm helps in creating a way to settle down disputes, which,
as a consequence, reduces all the upfront costs related to the management of
potential disputes. Here again, Ronald Coase puts the firm ability to exercise
control at the center of his demonstration. He was awarded the 1991 Nobel
Prize in Economics “for his discovery and clarification of the significance of
transaction costs and property rights for the institutional structure and
functioning of the economy.”23
b. The impact of Coase’s Article on economics
Ronald Coase’s Article has put transactions at the center of modern

economics, making it “the ultimate unit of microeconomic analysis.”24
Although Ronald Coase complained in 1988 that the concept was “largely
absent from current economic theory,”25 it has transformed the firm’s
perception, from a production function into a governance structure.26
In the tradition of Ronald Coase, the work of Oliver Williamson stands out
among others.27 Mainly, Williamson has built predictability in the event
where transaction costs using the firm, or the market, could not be compared

22 See generally Oliver Hart & John Moore, Incomplete Contracts and Renegotiation, 56(4)
ECONOMETRICA, 755-785 (1988).
23 The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 1991,
NOBELPRIZE.ORG, https://www.nobelprize.org/prizes/economic-
sciences/1991/coase/facts/ [https://perma.cc/65EN-ASUU].
24 OLIVER E. WILLIAMSON, MARKETS AND HIERARCHIES: ANALYSIS AND ANTITRUST
IMPLICATIONS 20 (Free Press 1975). For a broader perspective, see Richard A. Posner, Nobel
Laureate: Ronald Coase and Methodology, 7 J. ECON. PERSPECT 195, 206 (1993) (arguing that
Ronald Coase rejected a place for formal theory in economics). Finally, underlining the fact
that Ronald Coase has established the continuing legitimacy of the neoclassical model, see
Herbert J. Hovenkamp, The Antitrust Movement and the Rise of Industrial Organization, 09-34
UNIVERSITY OF IOWA LEGAL STUDIES RESEARCH PAPER 105, 121 (2009).
25 RONALD H. COASE, THE FIRM, THE MARKET, AND THE LAW 6 (University of
Chicago Press 1988).

26 Scott E. Masten, About Oliver E. Williamson, in FIRMS, MARKETS, AND HIERARCHIES:
THE TRANSACTION COST PERSPECTIVE 42 (Glenn R. Carroll & David J. Teece eds., 1999).
27 One could also underline the importance of the work of Armen Alchian & Harold
Demsetz who have linked the issue of transaction costs to the definition and specification of
property rights, see ALCHIAN & DEMSETZ, supra note 19, also, Benjamin Klein & Andres V.
Lerner, The Firm in Economics and Antitrust Law, 1 ISSUES IN COMPETITION LAW AND POLICY
249 (2008).
8
directly.28 He has also developed his research on optimal firm design.29 As such,
Williamson has helped to open the firm’s black box further, putting the firm’s
“control instruments”30 and the “means by which to infuse order”31 at the very
center of his analysis. He was awarded the 2009 Nobel Prize in Economics.
Alternative theories to those of Ronald Coase have also developed. One may
quote, primarily, the incentive theory, according to which the firm is an
incentive system that uses various instruments combining authority,
ownership, and compensation to ensure that all employees contribute their
best to the firm’s interest.32 Here, the firm’s survival is only dependent on its
ability to create such an incentive to survive. It is as a nexus of written and
unwritten contracts between different economic actors in which each
contractual relationship is an agency relationship whose optimal
configuration must be found. Fundamentally, there is no difference in nature
between the firm and the market, as relationships within the firm do not
imply the exercise of any authority or control. Although it is not used in
antitrust and competition law,33 that theory will nonetheless prove to be
useful for the subsequent analyses in the present Article.
To this day, the road ahead seems to lie in two directions. First, transaction
costs could help foster our understanding of anti-competitive behaviors,
notably, online practices in two-sided markets.34 Studying such practices is
not the focus of this Article. Second, transaction costs economics could help
28 MASTEN, supra note 26, at 38. Williamson was awarded the Nobel prize in 2009 “for
his analysis of economic governance, especially the boundaries of the firm,” The Sveriges
Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2009, NOBELPRIZE.ORG,
https://www.nobelprize.org/prizes/economic-sciences/2009/williamson/facts/
[https://perma.cc/S7FK-SVZ3].
29 WILLIAMSON, supra note 24, at 117 (analyzing transaction costs within the firm);
MASTEN, supra note 26, at 48. Generally speaking, discussing Williamson’s indirect
contribution to the field of corporate law, see Roberta Romano, Corporate Law and Corporate
Governance, 5 INDUSTRIAL AND CORPORATE CHANGE 277 (1996); Oliver E Williamson, The
Economics of Internal Organization: Exit and Voice in Relation to Markets and Hierarchies, 66 AM.
ECON. REV 369 (1976) (assessing that internal firms structure matters in assessing the
properties of administrative organization).
30 Oliver E. Williamson, The Vertical Integration of Production: Market Failure Considerations,
61(2) AM. ECON. REV 112, 113 (1971).

31 See WILLIAMSON, supra note 15, at 34; Oliver E. Williamson, The Theory of the Firm as
Governance Structure: From Choice to Contract, 16(3) J. ECON. PERSPECT 180 (2002).
32 Developing on these instruments, see Bengt Holmström & Paul Milgrom, The Firm as
an Incentive System, 84(4) AMERICAN ECONOMIC REVIEW 972-991 (1994). Michael C. Jensen
is at the origin of this alternative theory of the firm, see Michael C. Jensen, Theory of the Firm:
Managerial Behavior, Agency Costs and Ownership Structure, 3(4) J. FINANC. ECON 78 (1976).
33 In fact, that theory had been expressly rejected by courts, as described below.
34 Herbert J. Hovenkamp, Harvard, Chicago and Transaction Cost Economics in Antitrust
Analysis, 55 ANTITRUST BULLETIN 613 (2010) (revisiting Sherman Act Section 2 offenses
through the lens of transaction costs). One must underline that Oliver Williamson has
already paved the way in using transaction costs to study anti-competitive behaviors,
although not in relation with digital issues, see Oliver E. Williamson, The Economics of Antitrust:
Transaction Cost Considerations, 122 U. PA. L. REV. 1439 (1974) (analyzing price discrimination
through transaction costs).
9
in developing new tools to understand the firm35 and market dynamics.

Ronald Coase already envisaged this path in 1988, underlining that “so
much that happens in the economic system is designed either to reduce
transaction costs or to make possible what their existence prevents,” putting
the concept at the very center for further research.36 The present Article
precisely focuses on understanding the dynamism between firms, markets,
and blockchain. In that sense, it results entirely from Ronald Coase’s logic.
2. The Theory of the Firm in Modern Antitrust Law
The theory of the firm, such as defined by Ronald Coase, is at the heart of
modern antitrust and competition law. Despite the absence of court decisions
expressly mentioning the work of Ronald Coase in defining the firm’s
boundaries, his theory has impacted decades of modern case law. It notably
helps in pointing out where control is being exercised, and therefore, where
are the firm’s boundaries (a). Antitrust and competition law applies to all
entities defined accordingly (b).
a. The firm’s boundaries in antitrust and competition law
The Sherman Act in the United States and the Treaty on the Functioning of
the European Union in Europe are both the subject of extensive case law.
The vast majority of the jurisprudence is hardly concerned with the question
of the firm; in other words, the person who is the subject of antitrust and
competition law. The firm’s structure has transformed very little since the
introduction of these two texts; it has become more complex but has not
changed in nature. For that reason, litigation generally involves other issues
that are subject to further disagreement. Nevertheless, the emergence of
public permissionless blockchain forces us to reassess the definition of the
firm to analyze whether decentralized groups can be captured by antitrust
and competition law as currently conceived, or if blockchains should be
captured thanks to another theory.
In the United States, antitrust provisions apply to all “persons”37 affecting trade
and commerce by way of unlawful restraints and monopolies.38 According
to Section 7 of the Sherman Act, “the word ‘person,’ or ‘persons,’ wherever
35 Oliver E. Williamson, Why Law, Economics, and Organization?, 1 ANNU. REV. LAW. SCI.
369, 392 (2005).

37 See 15 U.S. C. § 1 (1890); 15 U.S. C. § 2 (1890).
38 See S. 1, An Act to protect trade and commerce against unlawful restraints and monopolies (Sherman
Antitrust Act), May 13, 1890, U.S. CAPITOL VISITOR CENTER,

https://www.visitthecapitol.gov/exhibitions/artifact/s-1-act-protect-trade-and-
commerce-against-unlawful-restraints-and-monopolies [https://perma.cc/8L5X-4FD5].
10
used in sections 1 to 7 of this title shall be deemed to include corporations

and associations existing under or authorized by the laws of either the United
States, the laws of any of the Territories, the laws of any State, or the laws of
any foreign country.”39 The jurisprudence does not further define the term
person; it simply defines exemption regimes for which antitrust is not
applicable, mainly, agencies and instrumentalities of the federal
government.40
Defining the firm’s boundaries in antitrust law is essential as one person

cannot conspire with itself.41 In that regard, the Supreme Court has ruled
out in Copperweld that an agreement between two factions of the same entity
could be found to be illegal as “the parent may assert full control at any
moment if the subsidiary fails to act in the parent’s best interests,”42 a ruling
that has been called “consistent with Ronald Coase's theory of the firm.”43
Control is then central in delimiting the boundaries of an entity,44 and “when
making a single-entity determination, courts must examine whether the
conduct in question deprives the marketplace of the independent sources of
economic control that competition assumes.”45 Only when “general
39 See 15 U.S. C. § 7 (1890).

40 See Eleanor M. Fox & Deborah Healey, When the State Harms Competition – the Role for
Antitrust and Competition Law, 13-11 LAW & ECONOMICS RESEARCH PAPER SERIES 769, 784
(2014).
41 Defining what is a firm in antitrust law proves to be particularly crucial if one
considers the fact that civil enforcement actions are generally brought against entities and
not individuals. Things are however different for criminal enforcement actions that may be
brought against firms and individuals.
42 See Copperweld Corp. et al. v. Independence Tube Corp., 467 US 752 (1984), in
which the Supreme Court held that a single entity cannot be defined on the sole interests
of the parties because “[t]he coordinated activity of a parent and its wholly owned
subsidiary must be viewed as that of a single enterprise,” at 771.
43 RUDOLPH J. R. PERITZ, COMPETITION POLICY IN AMERICA, 1888-1992: HISTORY,
RHETORIC, LAW, 356 (OUP, 1996); see also, KLEIN & LERNER, supra note 27, at 252.
44 The presumption according to which the components of the firm “act to maximize
the firm’s profits” does not hold in rare cases when “the intrafirm agreements may simply
be a formalistic shell for ongoing concerted action,” see 130 S.Ct. 2201, 2215 (2010), citing
Topco Associates, Inc., 405 U.S., at 609; Sealy, 388 U.S., at 352-354. In such cases, the
two entities are then seen as two separate firms as no control is being exercised.
45 American Needle Inc. v. National Football League, 538 F.3d 736, 742 (7th Cir. 2008),
confirmed on this point by American Needle, Inc. v. National Football League, 560 U.S.
183 (2010) (focusing on “control a single aggregation of economic power”). It has been said
in Copperweld Corp. et al. v. Independence Tube Corp., 467 U.S. 752, 771 (1984), that
when one entity controls another, it is similar to “a multiple team of horses drawing a vehicle
under the control of a single driver,” and therefore, they form a single firm. Judge
Easterbrook further called “silly” the fact of taking into consideration the sole interests of
the parties without first analysing control. In his views, “conflicts are [indeed] endemic in
any multi stage firm,” see Chicago Prof'l Sports Ltd. P'ship v. NBA, 95 F.3d 593 (7th Cir.
1996). Generally speaking, US courts hold parent companies liable for the behavior of their
subsidiaries (1) if the parent company itself was actively involved in the practice, or (2) if the
plaintiff managed to meet the requirements of corporate law to pierce the corporate veil of
the subsidiary, on the subject, see Carsten Koenig, Comparing Parent Company Liability in EU
and US Antitrust And Competition Law, 41 WORLD COMPET. 69, 80 (2018) (stressing that
holding parent companies liable is harder in the United States than it is in Europe, notably,
because courts require proof that the control has been exercised in practice, and don’t
11
corporate actions are guided or determined” by “separate corporate

consciousnesses,” two entities could be seen as two separate firms for the
purpose of antitrust law.46 One must, therefore, make no mistake about it;
only control makes the firm and helps in defining its scope.47
In Europe, the theory of the firm such as defined by Coase is also the basis of
modern competition law.48 Article 1 of Protocol 22 to the European
Economic Area Agreement defines the firm as being “any entity carrying out
activities of a commercial or economic nature,” but the concept is not
properly defined in the E.U. Treaty. The case-law further designates as
“undertaking” “every entity engaged in an economic activity, regardless of
the legal status of the entity and the way in which it is financed.”49 It follows
that the legal form of the entity offering the economic activity does not
matter.50 In fact, the Court of Justice has made clear in Shell that
“undertakings” are economic units rather than legal units,51 and as it has
been underlined by Wouter Wils, the concept of undertaking takes as a
starting point the existence Ronald Coase’s pathbreaking article.52
That definition of the firm is yet incomplete, as it does not permit to define
its boundaries. Indications in this regard come from the Imperial Chemical
Industries case in which the Court has ruled that carrying out “the instructions
given” by a company was essential in analyzing the independence of a
subsidiary, and that “where a subsidiary does not enjoy real autonomy in
determining its course of action in the market,” the prohibitions set out in
TFEU Article 101 were inapplicable.53 The Court further held in Akzo Nobel
simply rely on a theoretical control-based liability).

46 Copperweld, 467 U. S., at 771.
47 For instance, in Copperweld v. Independence Tube, 467 U.S. 752, 769 (1984), the
Court has focused its analysis on the existence of “independent centers of decision making.”
It has been confirmed in American Needle, Inc. v. National Football League, 560 U.S. 183
(2010).
48 Florence Thépot, The Firm in Antitrust and Competition Law, in THE INTERACTION
BETWEEN COMPETITION LAW AND CORPORATE GOVERNANCE 33 (2019).

49 Case C-41/1990, Klaus Höfner and Fritz Elser v Macrotron GmbH, 1991 I-01979,
para. 21. Please note that the term undertaking has been previously dealt with in Case
170/83, Hydrotherm Gerätebau GmbH v Compact del Dott. Ing. Mario Andreoli & C.
Sas., 1984 European Court Reports 1984-02999 para. 11.
50 Joined Cases C-264/01, C-306/01, C-354/01 and C-355/01I, AOK
Bundesverband; Bundesverband der Betriebskrankenkassen; Bundesverband der
Innungskrankenkassen; Bundesverband der Landwirtschafts Krankenkassen; Verband der
Angestelltenkrankenkassen eV; Verband der Arbeiter-Ersatzkassen; Bnndesknappschaft;
See-Krankenkasse v. Ichthyol Company Cordes; Mundipharma GmbH; Go ̈deke Share
Company; Intersan, 2003 ECR I-2493, Opinion of AG Jacobs (2003), para. 25.
51 See Case T-11/89 Shell v. Commission [1992] E.C.R. II-884, para. 311. One may
want to reiterate that Ronald’s Coase has been the first to explain that economic units exist
thanks to control with the aim to reducing transaction costs.
52 Wouter P.J. Wils, The Undertaking as Subject of E.C. Competition Law and the Imputation of
Infringements to Natural or Legal Persons, 25 EUR. LAW REV. 99, 102 (2000)
53 Case 48/69, Imperial Chemical Industries Ltd. v. Commission of the European
Communities, 1972 E.C.R. 619 (1972), para. 133-134. See also, Case C-73/95 P, Viho
12
that “the actual exercise of decisive influence”54 defines firm limits in

competition law and that “it is sufficient for the Commission to prove that
the subsidiary is wholly owned by the parent company in order to presume
that the parent exercises a decisive influence over the commercial policy of
the subsidiary.55” In the end, entities are defined as all the elements over
which control is exercised, as in the United States.56 For instance, in
Hydrotherm, the Court of justice has found that a natural person, a limited
partnership and another undertaking constituted a single economic unit
when they were all controlled by the same natural person.57 That logic results
from Ronald Coase’s Nature of the Firm.58
Europe BV v Commission of the European Communities, 1996 ECR. l-5457 (1996), in

which the Court held that control can be inferred from subsidiaries not enjoying “real
autonomy in determining their course of action in the market, but carry out the instructions
issued to them by the parent company controlling them.” Case C-73/95 P, Viho Europe
BV v Commission of the European Communities, 1996 ECR l-5457 (1996), para. 16, Case
48/69, Imperial Chemical Industries Ltd. v. Commission of the European Communities,
1972 ECR 619 (1972), para. 133-134; Case 15/74, Centrafarm BV and Adriaan De Peijper
v. Sterling Drug Inc., 1974 ECR 1147 (1974); Case 16/74, Centrafarm BV and Adriaan
De Peijper v. Winthrop BV, 1974 ECR 1183 (1974), para. 32; Case 30/87, Corinne Bodson
v. Pompes Funebres des Regions Liberees SA, 1988 ECR 2479 (1988), para. 19; Case
66/86 Ahmed Saeed Flugreisen and Others v Zentrale zur Bekaempfung Unlauteren
Wettbewerbs 1989 ECR 803 (1989), paragraph 35; Case 22/71, Béguelin Import Co vs.
SAGL Import Export 1971 ECR 949 (1971), para. 8.
54 Akzo Nobel and Others v Commission, C-516/15 P, EU:C:2017:314, para 55.
55 Akzo Nobel and Others v Commission, C-516/15 P, EU:C:2017:314, para 61.
Indeed, the European Commission may imply analyzing shares holding or other factual
elements. Studying who is supporting financial risks is one of them, see Case C–309/99,
J.C.J. Wouters and Others v. Algemene Raad van de Nederlandse Orde van Advocaten,
2002 ECR 1577 (2002), para. 48; see also THÉPOT, supra note 48, at 37 (defending that
analyzing how the risk of an activity is spread indicates whether the relationship or activity
at stake is inside or outside an undertaking).
56 In Case 170/83, Hydrotherm Gerätebau GmbH v Compact del Dott. Ing. Mario
Andreoli & C. Sas., 1984 ECR 02999 (1984), the Court has found that when legal entities
have “identical interests and [are] controlled by the same natural person,” then they form
“a single economic entity,” Case 170/83, Hydrotherm Gerätebau GmbH v Compact del
Dott. Ing. Mario Andreoli & C. Sas., 1984 ECR 02999 (1984), para 11. Despite this
confusing wording, here again, the Court ended up analyzing the control exercised as the
sole determinant of economic interests, see IOANNIS LIANOS, VALENTINE KORAH & PAOLO
SICILIANI, COMPETITION LAW: ANALYSIS, CASES, & MATERIALS, 329 (OUP, 2019) (“The
concept of control plays an important role in the process of defining the scope of the
competition law intervention against an economic entity as it determines the tangible or
intangible assets that constitute the core of the undertaking, define its boundaries and are
thus presumed to be under the authority of the undertaking’s agents, which may engage
through anti-competitive strategies the undertaking’s liability”).
57 Case 170/83, Hydrotherm Gerätebau GmbH v Compact del Dott. Ing. Mario
Andreoli & C. Sas., 1984 European Court Reports 1984-02999 para. 12.
58 KLEIN & LERNER, supra note 27, at 266. Also, discussing the case law showing how
central the element of “control” is, see ALISON JONES, BRENDA SUFRIN & NIAMH DUNNE,
EU COMPETITION LAW: TEXT, CASES, AND MATERIALS 157 (OUP 7th ed. 2019). See for
instance the case C-22/98 Becu 1999 ECR 1-5665 (1999), para. 89-91, in which the Court
of Justice has held that dock workers, performing work under the direction of their
employers, were to be incorporated into the undertaking concerned.
13
b. The firm as a pillar of antitrust and competition law
The definition of the firm’s boundaries helps three fundamental steps of

antitrust and competition law: (i) determining whether the law should apply,
(ii) assessing practices, (iii) and assigning liability.
First, establishing the firm’s boundaries helps in determining the extent to

which antitrust and competition law is applicable. U.S. antitrust law provides
for several exemptions to certain types of entities, which requires first to
identify the firm and second to understand its activities. European
competition law only applies when the firm carries out an economic activity.
Once again, it is then necessary to identify the firm’s boundaries to determine
the type of activities carried out within them.
Second, analyzing the firm’s boundaries permits to assess the legality of

practices.59 In terms of collusion, as previously discussed, U.S. and European
courts have recognized that two legal entities that are part of the same
economic unit cannot be held guilty of collusion as one cannot agree with
itself.60 Furthermore, antitrust and competition law prohibits different forms
of cooperation outside the firms, while cooperation within the firm is always
permitted. The logic is similar in terms of monopolization and abuse of
dominant position. As a company cannot abuse its market power against
itself, abuses of power are only illegal when they manifest outside the firm.
Above all, assessing the firm boundaries is essential in evaluating market
power and, accordingly, whether Sherman Act Section 2 or TFUE Article
102 applies in the first place.
59 Frank H. Easterbrook, Allocating Antitrust Decisionmaking Tasks, 76 GEO. L.J. 305, 314
(1987) (doubting the ability of courts to evaluate practices, even in light of transaction costs).
Also, underlining that transaction costs economics “assumes away the critical question of
who transacts with whom,” yet, that “the question of who can trade is at least as important
as the question of the terms of trading,” see Herbert J. Hovenkamp, Harvard, Chicago, and
Transaction Cost Economics in Antitrust Analysis, 55 ANTITRUST BULLETIN 626 (2010).
60 In the United States, see for instance Zachair v. Driggs 141 F 3d 1162 (4th Circ 1998);
467 US 752, 769, 767–72 (1984); Copperweld Corp. et al. v. Independence Tube Corp.,
467 U.S 752, 769, 767–72 (1984); for an analysis, Herbert J. Hovenkamp, Federal Antitrust
Policy, The Law of Competition and its Practice 246 (5th ed. 2016). Herbert Hovenkamp, Federal
Antitrust Policy, The Law of Competition and its Practice 246 (5th ed. 2016); also, Spencer Weber
Waller, Corporate Governance and Competition Policy, 18 GEO. MASON L. REV. 833, 841 (2011).
In Europe, see for instance the Case C-73/95 P, Viho Europe BV v Commission of the
European Communities, 1996 ECR l-5457 (1996); Case 170/83, Hydrotherm Gerätebau
GmbH v Compact del Dott. Ing. Mario Andreoli & C. Sas., 1984 European Court Reports
1984-02999 para. 11, 22. Case C-266/93, Bundeskartellamt v Volkswagen and VAG
Leasing GmbH, 1995 ECR I-3477 (1995); for an analysis, see Alison Jones, The Boundaries of
an Undertaking in EU Antitrust and Competition Law, 8 EUR. COMPET. J. 301, 308 (2012).
14
Finally, identifying the firm boundaries helps in attributing liability.61 Liability

for anti-competitive practices rests with the parent company that ultimately
controls other entities, to the extent that such control has been exercised.62
This stems from the classic distinction between ownership and control.63
Generally speaking, it is safe to assume that the broad definition of antitrust

and competition law applicability will capture blockchain participants at the
individual level.64 Nevertheless, analyzing what constitutes a firm within the
meaning of antitrust and competition law remains essential as it permits to
analyze anti-competitive practices that are carried out beyond the simple
framework of the individual. Defining the firm’s boundaries is, in fact, a
necessary step in understanding competitive dynamics, in analyzing
practices, and eventually, in assigning antitrust liability at an upper level. For
that reason, one must be concerned with the elements taken into account to
define the firm in antitrust and competition law. In the United States as in
Europe, only one matters: control. This reasoning is problematic when it
comes to blockchain.
B. The Theory of the Firm Vis-à-Vis Blockchain
We have established that antitrust and competition law is built on the theory
of the firm as defined by Ronald Coase. We must now analyze the extent to
which blockchain permits to organize economic activities outside the firm,
or differently put, outside of antitrust and competition law main pillar. To
this end, we show that blockchain introduces a new way to transact (1) that
escapes antitrust and competition law because no power of command control
can be exercised within it (2).
61 Case C-204/00 P, Aalborg Portland and Others v Commission, Opinion of AG Ruiz-

Jarabo Colomerpara, 2004 ECR I-123 (2004), para. 68.
62 In the United States, see In re Digital Music Antitrust Litigation, 812 F. Supp. 2d 390
(2011) (rejecting “a joint liability theory to put entire corporate families on the hook for
antitrust violations”); Cupp v. Alberto-Culver USA, Inc., 310 F. Supp. 2d 963 (2004)
(holding that the mere existence of a corporate relationship does not implicate a parent in
its subsidiary’s actions). In Europe, see for Case C‐625/13, Villeroy & Boch AG v European
Commission, 2017 Reports of Cases before the Court of Justice and the Court of First
Instance 1 (2017). para 146; Case C‐625/13, Villeroy & Boch AG v European Commission,
2017 Reports of Cases before the Court of Justice and the Court of First Instance 1 (2017).
para 146; Case C-155/14 P, Evonik Degussa GmbH and AlzChem AG v. European
Commission, 2016 Reports of Cases before the Court of Justice and the Court of First
Instance 1 (2016), para 27; Joined Cases C-93/ 13 P & C-123/13 P, Comm’n and Others
v. Versalis and Others, 2015 Reports of Cases before the Court of Justice and the Court of
First Instance 1 (2015), para 40. For an analysis, see Alison Jones, The Boundaries of an
Undertaking in EU Competition Law, 8 EUR. COMPET. J. 301, 328-331 (2012).
63 On the distinction, see Spencer Weber Waller, Corporate Governance and Competition Policy,
18 GEO. MASON L. REV. 833, 835 (2011).

64 When antitrust and competition law is applied to individuals, the question of revealing
their real-life identities becomes crucial, see SCHREPEL, supra note 6, at 322.
15
1. Ronald Coase’s Logic in Light of Blockchain
Ronald Coase explains the creation of firms through transaction costs. In the
following section, we analyze the extent to which blockchain can be used to
reduce some of these costs and carry out transactions outside the firms and
markets (a). We further argue that singular features of blockchain are making
it an attractive institution, and that they are to be integrated as a new variable
in the Coasian’s equation (b).65
a. The blockchain as a third institutional way
One may argue that blockchain is a new institution, allowing for transactions
thanks to characteristics other than those of the firms and markets.
Measuring the extent to which that institution could take over the two others
is nonetheless challenging.
Distinction from other institutions. In Ronald Coase’s theory of the

firm, transaction costs explain which transactions take place within or outside
of the firm. These costs may also explain the extent to which blockchain may
constitute a third way to carry out transactions. In fact, because blockchain
is yet a technology in the process of being broadly adopted,66 two schools
postulate the validity of their conclusion without being able to substantiate
them. One claims that blockchain will help to strengthen companies, and,
ultimately, will enable them to expand their borders; the other defends that
blockchain will give rise to the emergence of an alternative to the firm and
markets.
The first school argues that blockchain could allow a reduction in transaction
costs within the firm thanks to smart contracts, i.e., automated transactions
on the blockchain.67 By executing transactions when certain specified
65 See Prateek Goorha, The Return of ‘The Nature of the Firm’: The Role of the Blockchain, 1 THE
JBBA 71 (2018) (arguing that “nothing dares alter Coasian logic, but the space of
cryptographic stigmergy does add nuance to his argument”).
66 One may argue that blockchain is very much on the verge of being broadly adopted,
see for instance Daily Bitcoin Transactions Have Increased by 57% Since the Beginning of 2019,
LONGHASH, https://www.longhash.com/en/news/2470/Daily-Bitcoin-Transactions-
Have-Increased-by-57-Since-the-Beginning-of-2019 [https://perma.cc/4ASC-C873];
Chauhan, Om Prakash Malviya, Madhav Verma & Tejinder Singh Mor, Blockchain and
Scalability, in IEEE International Conference on Software Quality, RELIABILITY AND
SECURITY COMPANION, 2018 (underlining that number of transactions are also increasing
in an exponential manner, including for Bitcoin and Ethereum).
67 See, for instance, David Yermack, Corporate Governance and Blockchains, REVIEW OF
FINANCE (forthcoming 2017); Ian Grigg, Triple Entry Accounting, IANG (Dec. 25, 2005),
http://iang.org/papers/triple_entry.html [https://perma.cc/S84X-LLAL]; Bernard
Lunn, Blockchain Economy is the 1937 Coase Theorem brought to life and that worries Big Corp, DAILY
FINTECH (Mar. 10, 2018), https://dailyfintech.com/2018/03/10/blockchain-economy-is-
the-1937-coase-theorem-brought-to-life-and-that-worries-big-corp/
[https://perma.cc/B9FC-GK5X].
16
conditions are met, it seems that the costs to ensure the execution of
transactions could indeed be reduced. As a consequence, innovation could
increase by allowing firms to focus on their core activities rather than the
implementation of transactions.68 As such, the firm as a “nexus of contracts”
would extend its scope. That hypothesis does not tell, however, why
independent economic agents on the market could not also use blockchain
and smart contracts to reduce the transaction costs they have to bear. Indeed,
because these agents do not benefit from the trust that may exist within the
firm regarding the proper execution of internal transactions, blockchain
could help them in this regard. To this extent, the reasons given to explain a
possible expansion of the firm’s scope seems to be insufficiently substantiated
as they are limited to analyzing the firm as an institution that would not suffer
from competition with other institutional ways to transact.
Alternatively, the second school argues that blockchain could allow the
creation of a third way alongside firms and markets because it has certain
systematic advantages over the two.69 In this sense, blockchain would
compete with these two institutions rather than changing them.70 To start
68 Christian Catalini & Joshua S Gans, Some Simple Economics of the Blockchain, 5191-16
MIT SLOAN RESEARCH PAPER (2019); Carl J. Schramm, Blockchain: Leveraging a Trust
Technology in Expeditionary Economics, 12 INNOV. TECHNOL. GOV. GLOB. 28, 33 (2019)
(claiming that blockchain will also increase firms’ competence of extracting value from
innovation); more generally, see Ramsi Woodcock, What am I missing? - The Fundamental Unit
of Competition Is Not the Firm, ZEPHYRANTH.PW (Jul. 21, 2019),
https://zephyranth.pw/2019/07/21/the-fundamental-unit-of-competition-is-not-the-
firm/ [https://perma.cc/7YJD-XDF3] (arguing that “it is reasonable to suppose that
sometimes, the best way to promote competition, and reap its benefits in terms of greater
innovation, will come not by making it easier for startups to enter a market, but by insisting
that the large firms that dominate the market organize themselves internally in ways that
promote innovation,” what the blockchain could permit by refocusing on the participants
rather than the entity itself). More broadly, see Ian Lee, Enterprise Blockchains 3.0: Third Time’s
a Charm?, Medium (2019), https://medium.com/swlh/enterprise-blockchains-3-0-third-
times-a-charm-e0fc3c7d66f4 [https://perma.cc/PF6K-XZ6A] (discussing enterprise-led
blockchains and guessing “Sponsored Networks,” “where established enterprises sponsor — yet
don’t own or control — external, independent teams to develop or build on existing open
source protocols and public networks for the corporation and general use” could scale rapidly).
69 Generally, see Yochai Benkler, Coase’s Penguin, or, Linux and The Nature of the Firm, 112
YALE L.J. 369, 444 (2002) (arguing that “peer production has a systematic advantage over
markets and firms in matching the best available human capital to the best available
information inputs in order to create information products”); Sinclair Davidson, Primavera
De Filippi & Jason Potts, Economics of Blockchain, SSRN (Mar. 9, 2016),
http://www.ssrn.com/abstract=2744751 [https://perma.cc/THV4-86JF]. Also, see
DAVIDSON, supra note 8, at 641 (defending the idea that there are six ways for transactions
to take place, “firms, markets, commons, clubs, relational contracts and governments”), and
Taylor Pearson, Markets Are Eating The World, RIBBONFARM (2019),
https://www.ribbonfarm.com/2019/02/28/markets-are-eating-the-world/
[https://perma.cc/JK95-92Y6] (explaining how public blockchain may disrupt firm and
governments which, in the end, are forms of ledgers); finally, Hanna Halaburda, Natalia
Levina & Semi Min, Understanding Smart Contracts as a New Option in Transaction Cost Economics
(2019).
70 DAVIDSON, supra note 8, at 653 (arguing that blockchain technology may erode the
margin of the comparative efficiency of firms).
17
with, that school of thoughts argues that blockchain indeed differs from the firm
for two reasons that are ultimately related to the fact that it allows self-
organization rather than top-down planning.71
Firstly, blockchain leads to economizing on certain transaction costs due to

singular features.72 Such reductions may occur for small and large
transactions in lowering, among others, change search costs, contracting
costs, and coordination costs.73 Regarding small scale activities, blockchain
may help in transacting without a central authority since it creates an
architectural trust between the parties.74 The smaller the firm, the more
intensely it feels transaction costs.75 For that reason, blockchain may open
new perspectives for small businesses. As per more significant scale activities,
blockchain could be used to cut the costs of having a central authority
coordinating numerous transactions,76 and as such, could extend the
boundaries of efficient contracting into what was once the exclusive province
of the firm.77 Moreover, blockchain could allow the creation of new types of
transactions.78 Blockchain could indeed cause specific transaction costs not
71 ConsenSysMedia, Joe Lubin - Nature Of The Firm, v2.0 Keynote from EtherealNY #Blockchain
Conference 2018, YOUTUBE.COM (May 13, 2018),

https://www.youtube.com/watch?v=SQbcGhnv4jw [https://perma.cc/CB6T-R7E5]
(defending that “top-down command and control has taken us a long way as a society,” but
that “we can do much better” thanks to blockchain).
72 See Max Raskin, The Law and Legality of Smart Contracts, 1 GEO. L. TECH. REV 304, 305,
336 (2017) (opposing blockchain “radicals” with those arguing that smart contracts help in
reducing the firm’s transaction costs). Also, Christian Catalini, The Firm as a Nexus of Smart
Contracts? How Blockchain and Cryptocurrencies Can Transform the Digital Economy, YALE J. ON
REG.: NOTICE & COMMENT (June 7, 2017), http://yalejreg.com/nc/the-firm-as-a-nexus-
of-smart-contracts-how-blockchain-and-cryptocurrencies-can-transform-the-digital-
economy-by-christian-catalini/ [https://perma.cc/KM4Y-963N].
73 Patrick Waelbroeck, An Economic Analysis of Blockchains, 6893 CESIFO WORKING
PAPER 1, 14 (2018).
74 Clay Shirky, Here Comes Everybody: The Power of Organizing Without Organizations 41 (2011)
(arguing that when there is a small decrease in transaction costs, “small companies become
more effective, doing more business at lower cost than the same company does in a world
of high transaction costs”); Catherine Mulligan, Blockchain will kill the traditional firm,
IMPERIAL COLLEGE BUSINESS SCHOOL BLOGS (Oct. 16, 2017),
https://www.imperial.ac.uk/business-school/knowledge/finance/blockchain-will-kill-the-
traditional-firm/ [https://perma.cc/VZ5M-LKWM]; LUBIN, supra note 71 (defending that
“tokenizing everything will create fluid hyper efficient markets, and we will all be able to
wrap ourselves in api's and be very effective single elements”).
75 Jeremy M. Sklaroff, Smart Contracts and the Cost of Inflexibility, 166 U. PA. L. REV. 263,
297 (2017). In fact, there is a natural economic limit on how small firms can ultimately
shrink. Blockchain may eliminate this limit to the point of having free agents contracting
with one another.
76 DAVIDSON, supra note 8, at 649. Also, SHIRKY, supra note 74, at 41 (arguing that
management challenges grow faster than organizational size).

77 Tom Glocer, Blockchain, Coase and the Theory of the Firm, TOM GLOCER’S BLOG (October
23, 2017) http://www.tomglocer.com/2017/10/23/blockchain-coase-and-the-theory-of-

the-firm/ [https://perma.cc/2UQ4-Z9JR].
78 DAVIDSON, supra note 8, at 653-656 (defending that the mass adoption of blockchain
“may lead to an evolution of the economic institutions of capitalism itself”); Nick Tomaino,
The Slow Death of the Firm, THECONTROL.CO (Oct. 21, 2017), https://thecontrol.co/the-
slow-death-of-the-firm-1bd6cc81286b [https://perma.cc/TC42-HD3G] (arguing that
18
only falling moderately, but collapsing.79 As such, blockchain could permit

loosely coordinated groups to operate transactions that were previously out
of reach, even within the firm, as it would have implied engaging too high
transaction costs.
Secondly, Williamson argued that firms help in minimizing opportunism by

exploiting incomplete contracts. The firm provides the parties with a
framework to find a solution80 in the event of a litigious situation that has not
been contractually framed. On the contrary, blockchain is immutable,81 and
in that regard, it does not permit to manage unforeseen situations. Therefore,
to the extent that the writing of complete contracts is a utopia, the firm would
always be preferred to blockchain.82 In reality, blockchain manages
opportunism in different ways. Blockchain indeed creates trust by integrating
market mechanisms into a closed and guaranteed payment systems through
cryptographic enforcement and execution,83 therefore minimizing
opportunism. On top of that, several smart contracts could be combined84
and subordinate to each other to manage situations in which cryptographic
features are proved to be insufficient. As a result, establishing the absolute
superiority of firms on blockchains (or vice versa) for certain types of
transactions seems unrealistic at the moment, but the two can already be
distinguished from one another.
blockchain helps in the creation of new products that could not exist be created by firms);
Sid Venkateswaran, Ronald Coase and the Nature of the Blockchain, MEDIUM (Dec. 18, 2017),
https://medium.com/@sidmvenkat/ronald-coase-and-the-nature-of-the-blockchain-
e60021c036c9 [https://perma.cc/F7NS-FDPC].
79 SHIRKY, supra note 74, at 44; James Hazard, Odysseas Sclavounis & Harald Stieber,
Are Transaction Costs Drivers of Financial Institutions? Contracts Made in Heaven, Hell, and the Cloud
in Between, in BANKING BEYOND BANKS AND MONEY (Springer 2016).
80 Irving Wladawsky-Berger, Blockchain and the Future of the Firm, WSJ BLOG (Sep. 29, 2017
1:40 PM), https://blogs.wsj.com/cio/2017/09/29/blockchain-and-the-future-of-the-

firm/ [https://perma.cc/ZDG7-6LRW].
81 Differently put, blockchain is generally seen as an economic world of complete
contracts while firms are a nexus of incomplete contracts. On that, see DAVIDSON, supra
note 69. On how the firms can help in managing incomplete contracts, see Oliver Hart &
John Moore, Property Rights and the Nature of the Firm, 98(6) JOURNAL OF POLITICAL
ECONOMY 1119 (1990); OLIVER E. WILLIAMSON, THE ECONOMIC INSTITUTIONS OF
CAPITALISM 32 (Free Press 1985).
82 Oliver Hart, An Economist’s Perspective on the Theory of the Firm, 89 COLUMBIA LAW
REVIEW 1757 (1989); see DAVIDSON, supra note 69. After all, the actual execution of
transactions is “a secondary part of life; the primary problem or function is deciding what
to do and how to do it,” see FRANK H. KNIGHT, RISK, UNCERTAINTY AND PROFIT 267,
268 (Hart, 1921).
83 DAVIDSON, supra note 8, at 649-650.
84 For an explanation of smart contracts library, see Thibault Schrepel, Collusion by
Blockchain and Smart Contracts, 33 HARV. J.L. & TECH. 117, 143 (2019).
19
It is also argued that blockchain differs from the market in several ways85 that are
linked to the possibility of creating a “cryptographic stigmergy;” i.e., allowing
large groups of individuals to interact with one another in a stable
environment while the market requires a strong social trust between the
parties involved in a transaction.86 First, the technology chains blocks of
verified transactions, and as such, provide more information as the basis of
transaction than the sole price.87 Second, public permissionless blockchains
are open access, which means that the price of discovering the conditions for
any transaction is reduced compared to what they could be on the market.
Third, the concept of tokenisation can help in creating an efficient market
place because the platform is managed in perpetuity by decentralized nodes
rewarded for their efforts in upholding the network, as we will further
explain.88 Fourth, blockchain forces its users to internalize the costs of
verifying transactions.89 Differently put, blockchain not only reduces specific
transaction costs, but it also makes them part of what creates trust: the fact
that users are engaging in the costly activity of verifying transactions is what
makes blockchains valuable.90
The nature of blockchain as an institution. If blockchain is indeed a

third way to transact, alongside firms and markets, it is explained by the fact
that technologies can be more than a weak form of transaction determinants.
According to John R. Commons and Oliver Williamson, transactions rather
than technology “mainly determine the efficacy of exchange by one mode of
organization as compared with another.”91 In that sense, transactions are
85 GOORHA, supra note 65, at 73 (arguing that blockchain is based on the use of the price
mechanism thanks to which coordination across the cryptographic space is made possible).
86 Id. at 72.
87 Id.; Udo Pesch & Georgy Ishmaev, Fictions and Frictions: Promises, Transaction Costs and
the Innovation of Network Technologies, 49 SOC. STUD. SCI 264, 269 (2019).
88 Jeremy Liu, Blockchain, Decentralisation, and the ‘Theory of the Firm,’ MEDIUM (Dec. 13,
2017) https://medium.com/the-pointy-end/blockchain-decentralisation-and-the-theory-
of-the-firm-92649c62350d [https://perma.cc/96X8-WNAU].
89 GOORHA, supra note 65. Also, Floris F. Seuren, Exploring the Applicability of Blockchain in
Lowering Transaction Costs in the Commercial Real Estate Due Diligence Process: A Case Study Research,
SEMANTICSCHOLAR.ORG (2018),
https://www.semanticscholar.org/paper/1-Exploring-the-applicability-of-blockchain-
in-in-%3A-Loris-S./990a9a92800d02cfa372fb0fb6be974d5b925806#references
[https://perma.cc/KH9W-QPLP].
90 Richie Etwaru, Blockchain: Trust Companies: Every Company Is at Risk of Being Disrupted by
a Trusted Version of Itself, 185 (2017) (arguing that Coase did not consider the improvement
to market efficiencies from increased trust in transactions). More broadly, see Qiao Wang,
Cryptonetworks and the Theory of the Firm, TOKEN DAILY,
https://www.tokendaily.co/blog/cryptonetworks-and-the-theory-of-the-firm
[https://perma.cc/9ZUK-54QM] (explaining that cryptonetworks differs from the market
because they benefit from network effects), and Jesse Walden, Past, Present, Future: From Co-
ops to Cryptonetworks, ANDREESSEN HOROWITZ (2019),
https://a16z.com/2019/03/02/cooperatives-cryptonetworks/ [https://perma.cc/349H-
KUT2] (explaining that traditional platforms first collaborate with their users and end up
competing with them, while blockchain platforms may sustain continued cooperation).
91 WILLIAMSON, supra note 24, at 248.
20
“the ultimate unit of economic investigation.”92 Williamson further

underlines that although technologies are “non-unimportant,” they explain,
“only relatively small and specialized organizational structures.”93 As such,
technology would be only a pure semi-weak form of transaction determinants
where technology “delimits the set of feasible organizations” without being
“neither fully determinative of nor irrelevant to economic organization.”94
In short, Williamson views technologies as being indecisive when deciding
where transactions take place and the shaping of institutions.
If one may agree with Williamson’s analysis regarding past technologies, one
may also want to describe blockchain as a semi-strong form of transaction
determinants,95 meaning that “but for monopoly or related price theoretic
reasons, technology is determinative of economic organization.”96 Contrary
to Williamson’s belief, according to which technology can only be used
within a firm,97 one may contend that blockchain allows transactions
between free agents relying on more information than they have on the
market. As previously explained, blockchain may also be used to minimize
opportunism through cryptographic enforcement and execution. As such,
blockchain attractiveness does not entirely rely on reducing transaction costs,
but it is also explained by different characteristics that may be of interest to
its participants. The rise of several public permissionless blockchains, some
of them being used for several hundred thousand transactions per day,
support the idea that the blockchain is indeed an institution which is arousing
the interest of many economic agents.98
92 JOHN R. COMMONS, INSTITUTIONAL ECONOMICS 6 (Madison: University of

Wisconsin Press, 1934).
94 Oliver E. Williamson, Technology and Transaction Cost Economics: A Reply, 10(3) JOURNAL
OF ECONOMIC BEHAVIOR & ORGANIZATION 356, 357 (1988). He also underlines that
technologies are only helpful in choosing “between feasible organizing modes” while
transaction costs are determining feasibility, see Oliver E. Williamson, The Organization of
Work: A Comparative Perspective, 1 J. ECON. BEHAV. ORGAN. 5, 11-12 (1980).
95 See Ernest J. Englander, Technology and Oliver Williamson’s Transaction Cost Economics,
10(3) J. ECON. BEHAV. ORGAN. 339-353 (1988) (arguing that the second industrial
revolution occurred thanks to technologies as a semi-strong form of transaction
determinacy).
96 WILLIAMSON, supra note 94 (arguing that technological changes result in the necessity
to change institutions). Also, see Vernon L. Smith, Economic Theory and Its Discontents, 64 AM.
ECON. REV. 320, 321 (1974) (defending that a new micro theory requiring “a more
sophisticated treatment of the technology of transacting” is needed); lastly, DOUGLASS C.
NORTH, STRUCTURE AND CHANGE IN ECONOMIC HISTORY 9 (WW Norton & Co 1981)
(arguing that organizational innovations are needed to realize the potential of the new
technology).
97 Oliver E. Williamson, Technology & Transaction Cost Economics, 10(3) J. ECON. BEHAV.
ORGAN. 356 (1988) (analyzing the impact of the technology on transactions within the
boundaries of the firm).
98 Confirmed Transactions Per Day, BLOCKCHAIN.COM,
https://www.blockchain.com/charts/n-transactions [https://perma.cc/WDK3-YZLF].
See also CryptoCurrency, REDDIT,
https://www.reddit.com/r/CryptoCurrency/comments/e8qpn6/two_years_ago_less_th
21
The limits of the distinction between institutions. Blockchain does

not permit a complete elimination of all transaction costs.99 One may, in fact,
list the new transaction costs associated with the use of smart contract in three
categories.100 The first, search costs, encompasses the costs of verifying smart
contract features, of establishing confidence in smart property, and of
learning about cryptographic tools. The second, bargaining costs, are the
costs engaged in drafting contracts and translating them into computer
code.101 The third, commitment costs, are engaged when paying fees to full
nodes102 or information services, when protecting against coercion and theft
to digital assets, and when managing the risk of smart contracts failure.
As a result, depending on blockchain attractiveness, the distinction between

the three institutional ways to transact could be narrower than it appears at
first sight.103 It might be that only a few transactions will be optimized if they
occur in one of these three institutions, while the nature and frequency of
other transactions will be mostly indifferent to the institution in which they
take place. For now, no empirical analysis has compared the costs and
features of implementing different types of transactions within these three
institutions.104 One can only refute the Coasian’s assumption according to
which markets and the firms are the only two “alternative methods of
coordinating production” by stressing that the number of transactions
an_10_million_in_ether_was/ [https://perma.cc/JZB4-J8Z9] (“two years ago, less than

$10 million in ether was used in decentralized finance (defi) en Ethereum. Today, it's more
than $650 million”). Also, Chris Berg, Sinclair Davidson & Jason Potts, The Blockchain
Economy: A Beginner’s Guide To Institutional Cryptoeconomics, MEDIUM (2017)
https://medium.com/cryptoeconomics-australia/the-blockchain-economy-a-beginners-
guide-to-institutional-cryptoeconomics-64bf2f2beec4 [https://perma.cc/US7E-KYC4]
(for an introduction to institutional cryptoeconomics and the idea that “blockchains can be
used by firms, but they can also replace firms).
99 PESCH & ISHMAEV, supra note 87, at 271.
100 Ben Garfinkel, Recent Developments in Cryptography and Possible Long-Run Consequences 92
(2019); SKLAROFF, supra note 75, at 297.

101 DON TAPSCOTT & ALEX TAPSCOTT, BLOCKCHAIN REVOLUTION: HOW THE
TECHNOLOGY BEHIND BITCOIN IS CHANGING MONEY, BUSINESS, AND THE WORLD 103
(Portfolio 2016); SKLAROFF, supra note 75, at 298 (arguing that “the pressure to construct
larger and longer-term agreements that can change flexibly in response to unpredictable
events, will exist even in a world of smart contracts”).
102 Full nodes are blockchain participants storing an entire copy of the blockchain ledger
on their computers, for more, see infra Part II.A.1.a.

103 Craig S. Wright, Proof of Work as it Relates to the Theory of the Firm, SSRN 2,
https://www.ssrn.com/abstract=2993312 [https://perma.cc/A984-BE62] (Last revised:

Jun 29, 2017) (contending that the distinction between blockchain and other institutional
ways is narrower than one may expect, and this, regardless of the protocol that is being
implemented).
104 Discussing the idea, see DAVIDSON, supra note 8, at 640 (discussing the idea that
blockchain is an institutional technology); also, Vitalik Buterin, Visions, Part 1: The Value of
Blockchain Technology, ETHEREUM BLOG (Apr. 12, 2015),
https://blog.ethereum.org/2015/04/13/visions-part-1-the-value-of-blockchain-
technology/ [https://perma.cc/BBF8-MK6U].
22
implemented on blockchain continues to grow.105 For the rest, the impact of

blockchain as an institution will depend on the appeal created by its different
features (besides reducing transaction costs) such as managing opportunism
or being unstoppable.106
b. Integrating blockchain into the Coasian’s equation
Ronald Coase’s analysis of the firm is not static. From the very first pages of
his 1937’s Article, Coase is careful not to predict the firm’s systematic
prevalence over the market, underlining that “the degree to which the price
mechanism is superseded varies greatly.”107 He concludes by stressing that
“it is clear that the dynamic factors are also of considerable importance, and
an investigation of the effect changes have on the cost of organizing within
the firm and on marketing costs generally will enable one to explain why
firms get larger and smaller.”108 In a sense, Ronald Coase recognizes that
transaction costs vary considerably depending on circumstances that cannot
always be predicted and that the firm’s future depends on them. Yet, Ronald
Coase only admits dynamism between two institutions, namely, the firm and
the market, which are “alternative methods of co-ordinating production.”109
He further emphasizes that “a firm will tend to expand until the costs of
organizing an extra transaction within the firm become equal to the costs of
carrying out the same transaction by means of an exchange on the open
market or the costs of organizing in another firm.”110 The market is, in that
sense, the institution in which transactions take place by default. This
dichotomy between the firm and the market stems from the idea that
technological innovations are meant “to bring factors of production nearer
together,” in other words, ways “to increase the size of the firm.111”
As such, the Coasian equation may be represented as follows: where X

represents all the transaction costs associated with carrying out a transaction
within the firm, and Y represents all the transaction costs associated with
carrying out a transaction through the price mechanism (market), then,
where X < Y, the transaction is carried out within the firm. Conversely,
where X > Y, the transaction is carried out on the market.
105 LONGHASH, supra note 66; CHAUHAN, supra note 66 (underlining that number of
transactions are also increasing in an exponential manner, including for Bitcoin and
Ethereum).
106 SCHREPEL, supra note 84, at 120 (discussing the idea that blockchain code is
unstoppable).
108 Id. at 405.
109 Id. at 404 (underlining that economic activities can be organized within the firm or
thanks to the price mechanism, i.e., within the market).

110 Id. at 395.
111 Id. at 397.
23
However, one may argue that the emergence of blockchain as a new

institutional way to transact modifies Coase’s equation. The size of the firm
is no longer solely determined by the “marketing costs (that is, the costs of
using the price mechanism), and the costs of organizing of different
entrepreneurs,”112 but also by the costs of operating the transaction through
the blockchain. Thus, when B represents all the transaction costs associated
with completing a transaction using a blockchain, then, where X < Y and X
< B, the transaction is completed within the firm. Conversely, where Y < X
and Y < B, the transaction is carried out on the market. Finally, where B <
X and B < Y, the transaction is carried out through a blockchain. We further
propose to study the extent to which antitrust and competition law are
equipped to deal with the legal issues encountered in this last hypothesis.
2. Blockchain as an Escape to Antitrust and Competition Law
We have found that blockchain opens a third transactional way. One must
now analyze the attributes of that institution to determine to which extent
antitrust and competition law can be applied to it. After showing that
blockchain does not rely on the same form of command and control that the
firm (a), we analyze the extent to which blockchain participants may escape
antitrust and competition law (b).
a. Blockchain emergence and horizontality
The firm is an economic reality with a legal translation to which the law is
applied.113 Public permissionless blockchains as institutions are technological
and economic realities with no proper legal translation. Finding a legal
framework encapsulating blockchains reality is not an easy task as they are
spontaneous organizations114 using the price mechanisms of markets, the
governance properties of a common, and, to some extent, the constitutional
properties of a nation-state.115 The existence and survival of public
permissionless blockchain do not rely on one individual, and the decision of
a few individuals cannot stop them.116
112 Id. at 403.

113 Discussing whether the firm is a legal fiction or not, see ERIC W. ORTS, BUSINESS
PERSONS: A LEGAL THEORY OF THE FIRM 29 (Oxford 2013).
114 Karl J. Friston & Christopher D. Frith, Active Inference, Communication and Hermeneutics,
68 CORTEX 129 (2015) (speaking to the notion of generalized synchrony as a universal

phenomenon which this Article applies to public permissionless blockchain).
115 See DAVIDSON, supra note 69. Several scholars, for that reason, have stressed that
blockchains are catallaxies in the meaning of Friedrich Hayek, “a special kind of

spontaneous order,” see FRIEDRICH HAYEK, LAW, LEGISLATION AND LIBERTY 109
(Routledge 1973).
116 See Jameson Lopp, Who Controls Bitcoin Core?, CYPHERPUNK COGITATIONS (2018),
https://blog.lopp.net/who-controls-bitcoin-core-/ [https://perma.cc/ET5E-JPB4]
24
One may also underline that they allow for stigmergic coordination, whereby
“certain species of animals—such ants, termites, birds, etc.—create complex
social structures that do not rely on any hierarchical structure.”117 Indeed,
blockchain governance relies on different incentives that do not include
direct orders over others.118 Differently put, firms have different vertical
structures, while blockchains have different horizontal structures.119 As a
result, it follows that blockchains are platforms without being intermediaries.
Contrary to the firm which is controlling all transactions, blockchain
participants are not involved in each of them.
In short, blockchain is an emergent and horizontal form of transaction

determinacy. As a result, the theory of the firm initiated by Ronald Coase,
which is centered around direct top-down control to reduce transaction costs,
cannot be transposed to blockchain. One must, therefore, assess the extent
to which antitrust and antitrust and competition law can be applied to
blockchain in which the concept and exercise of control, as interpreted in
antitrust and competition law, cannot not found.
b. No control, no borders, no antitrust
The lack of command and control within blockchains create two legal issues.
First, the principle of individual liability implies that entities cannot possibly
be punished for practices over which they have no theoretical control. As a
result, one may wonder how to attribute liabilities within blockchain. For
instance, anyone may decide to use a public permissionless blockchain to
carry out transactions; in that regard, blockchain greatly differs from the firm
where only the employees can carry out transactions within it. Additionally,
blockchains may have different layers on top of one another,120 and it might
not be possible to control them nor to stop the addition of new layers.121
(arguing that “the definition (control) of Bitcoin the protocol is like the definition of a
language (…) No one is forced to agree with the definition of a given word in a dictionary,
neither are they forced to agree with code in a given Bitcoin implementation by running it.
(…) No one controls Bitcoin. No one controls the focal point for Bitcoin development”).
117 See DAVIDSON, supra note 69.
118 As indicated in the introduction to this Article, our study focuses on public
blockchains, not private ones where the structure is more vertical.

119 Several technologies can introduce verticality into blockchain, see for example
Backfeed which is creating a “driven by a uniform hierarchy of ends, and which incorporate
its own economic or monetary system,” on the subject, see DAVIDSON, supra note 69. This,
however, remains the exception to the principle of horizontal governance in public
blockchains.
120 Primavera De Filippi, No Blockchain Is an Island, COINDESK,
https://www.coindesk.com/no-blockchain-island [https://perma.cc/BSG8-L9ND] (last
updated Mar. 1, 2018) (discussing the nature of layers in the blockchain sphere).
121 For instance, see Bitcoin to Get Smart Contracts, TRUSTNODES (2019),
https://www.trustnodes.com/2019/08/07/bitcoin-to-get-smart-contracts
[https://perma.cc/N585-4S59].
25
Second, from the absence of command and control results the absence of
well-defined borders. Public permissionless blockchains are indeed at the
disposal of whoever wants, without entry and exit. As a result, their frontiers
are vague, if not indeterminable. Assessing the legality of practices
(cooperation being mainly permitted within entities, and not outside of it) as
well as their effects (antitrust and competition being concerned with the
effects of a practice outside the firm) is challenging.122
It thus appears that blockchain is a new institution that does present similar
characteristics to those of the firm. To the extent that antitrust and
competition law relies on these features, they become mostly inoperable in
the face of blockchain.123 The bigger blockchain will get as a transactional
institution, the bigger the problem will be. A new theory must be found so to
create a new legal fiction to which the law can be (re)applied.
122 See infra II.B.2.

123 Geoffrey Moore, The Nature of the Firm—75 Years Later, OPENMIND 8 (2014),
https://www.bbvaopenmind.com/wp-content/uploads/2015/02/BBVA-OpenMind-
The-Nature-of-the-Firm-75-Years-Later-Geoffrey-Moore.pdf.pdf
[https://perma.cc/CY4G-WHE7] (“No doubt this will create a new generation of liability
cases focused on determining the boundaries of accountability, and I do not envy the
adjudicators of these cases as those boundaries are inherently fuzzy.)
26
II. THE THEORY OF GRANULARITY:

A NEW PATH FOR APPLYING ANTITRUST TO
BLOCKCHAIN ECOSYSTEMS
We have seen in Part I that blockchain is a new institution to which the

theory of the firm cannot be transposed. For that reason, we now engage in
conceptualizing a new theory to create a legal fiction that defines and delimits
public permissionless blockchains. Section A introduces the “theory of
granularity.” (A). Section B demonstrates how that theory allows a complete
and fair application of antitrust and competition law while matching the
transactional and economic reality of blockchain (B).
A. The Theory of Granularity as a Theoretical

and Legal Framework for Blockchain
To make the law applicable to public permissionless blockchains, one must

conceptualize them.124 We introduce the “theory of granularity” to this end
(1) and outline how it might be transposed to antitrust and competition law
(2). Transactional by nature, that theory aims to explain the existence of public
permissionless blockchain beyond the simple cost-reduction framework. It
seeks to translate accurately the economic reality of such blockchains.
1. A Theorization of Blockchain as an Institution
The theory of granularity, to which one may want to provide a semantic

explanation (a), aims at capturing blockchain governance as a new
transactional institution (b). It thus fills the gap created by the inability to
transpose the theory of the firm to public permissionless blockchains.
a. The theory of granularity: semantic explanation
In The Nature of the Firm, Ronald Coase made an essential distinction between
organizations and organisms.125 While firms are organizations, blockchains
are clusters of organisms that, by nature, are spontaneous. Their functioning
124 The law cannot be (properly) applied to what is not understood by the legislator.
125 COASE, supra note 2, at 387 (explaining that an organism economic “works itself”).
Friedrich Hayek, to which Ronald Coase refers to, called organisms “an order which is not
made by anybody but which forms itself,” see FRIEDRICH HAYEK, KINDS OF ORDER IN
SOCIETY (1964) (Indianapolis: Liberty Fund, 2013).
27
must be analyzed and understood this way so that antitrust and competition
law can be applied to them, when necessary.
This present Article introduces the theory of granularity to this end.

Generally speaking, the notion of granularity defines the size of the smallest
element in a system, i.e., an organism. This theory, therefore, aims analyzing
the role played by each element of a blockchain. Unlike the firm where a
vertical and pyramidal control is exercised over its components, blockchains
are made of horizontal governance mechanisms.126 It reinforces the
importance of each organism as one cannot merely assume that they will
follow one coordinated direction.127 One must, therefore, study blockchain
smallest organisms, the role they play, and their dynamism.128 Only by
attaining the granularity level129 blockchain governance can be appropriately
understood, and antitrust law can be applied to its different life forms.
b. The theory of granularity: analyzing blockchain governance structure
Blockchain is a space in which different forms of powers are being exercised.

However, unlike the firm in which a power of command and control is
exercised top-down, no single actor has the power to control a public
permissionless blockchain entirely.130 In the words of Vitalik Buterin, creator
of the Ethereum, blockchains “are logically centralized (there is one
commonly agreed state and the system behave like a single computer) […]
politically decentralized (no one controls them) and architecturally
decentralized (no infrastructural central point of failure).”131
As a result, multiple interests can compete within the same blockchain; they
may even be opposed. In a sense, blockchain “contribute[s] to the realization
of a number of individual objectives which no one knows in their totality.”132
For that reason, one must study the different types of power that are generally
found within public permissionless blockchains to understand which interests
may eventually prevail over others. A distinction between three categories of
126 One may find different degrees of horizontality, see infra II.B.2.
127 For an analysis of the role played by each blockchain participants, see infra II.A.1.
128 See David S. Evans, Economic Aspects of Bitcoin and Other Decentralized Public-Ledger
Currency Platforms, 685 UNIVERSITY OF CHICAGO COASE-SANDOR INSTITUTE FOR LAW &
ECONOMICS RESEARCH PAPER 1, 16 (2014) (explaining that the analysis of the public
ledger platforms is “much more complicated” than typical open project projects).
129 When the granularity level is reached, it is then impossible to break down the
elements in smaller pieces.

130 See CATALINI & GANS, supra note 68, at 16 (defending that in the blockchain, each
participant “can theoretically shape its evolution in a way that is proportional to its stake in
the platform - e.g. in terms of computing power, storage, labor or capital dedicated to it”).
131 Vitalik Buterin, The Meaning of Decentralization, MEDIUM (Feb. 6, 2017),
https://medium.com/@VitalikButerin/the-meaning-of-decentralization-a0c92b76a274
[https://perma.cc/L6E3-PU2E].
132 HAYEK, supra note 115.
28
blockchain participants may be helpful in this regard, namely, between

founders or core developers, users and miners. We show that although each
blockchain has its specificities, the above-mentioned groups will generally use
the same mechanisms to express their preferences,133 and they will encounter
the same type of limits if they act on their own. Eventually, their powers may
suffer from four constraints, namely: law, markets, social norms, and
architecture.134
The power of founders and core developers.135 Blockchain founders

and core developers are in charge of implementing the original rules of a
blockchain.136 They are, in fact, designing the code software, thereby also
determining which consensus protocol will be used.137
In general, one may stress that the architecture of blockchain limits their
power as they lose any form of direct control on other participants once the
blockchain is put online.138 Indeed, founders and core developers cannot
133 Jean Bacon, Johan David Michels, Christopher Millard & Jatinder Singh, Blockchain
Demystified: A Technical and Legal Introduction to Distributed and Centralised Ledgers, XXV(1) RICH.
J.L. & TECH. 34 (2018).
134 See LAWRENCE LESSIG, CODE: AND OTHER LAWS OF CYBERSPACE, VERSION 2.0,
124 (Basic Books 2006) (“Norms constrain through the stigma that a community imposes;
markets constrain through the price that they exact; architectures constrain through the
physical burdens they impose, and law constrains through the punishment it threatens).
These constraints are “distinct, yet, they are plainly interdependent. Each can support or
oppose the others,” see Id. at 123. Also, DE FILIPPI, supra note 120 (for an application of
these four constraints to blockchain).
135 Founders and core developers are here put in the same category, although one may
want to make a distinction between them, as they are eventually in charge of the blockchain
core code.
136 Vlad Zamfir, Blockchain Governance 101, GOODAUDIENCE BLOG (Sep. 30, 2018),
https://blog.goodaudience.com/blockchain-governance-101-eea5201d7992
[https://perma.cc/T4BY-NFE7]. In short, core developers, also called “protocol
developers,” are to be differentiated from the developers building decentralized applications
on top of the blockchain. Core developers are indeed “reviewing the code, proposing
conceptual changes to the code, reviewing changes proposed by other coders, drafting new
code and revising existing code, security-testing new code, compiling code into new releases,
and communicating about the project with other developers, among others” according to
Angela Walch, The Fiduciaries of Public Blockchains, SSRN 9,
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3203198
[https://perma.cc/XJ3B-G6LH] (last revised Feb. 23, 2019). See also HALABURDA &
MUELLER-BLOCH, supra note 136. Being in charge of the blockchain’s original design give
them the opportunity, of course, to design it in a way to exercise more or less influence over
the entire blockchain. For example, if core developers grant themselves the power to
validate each transaction, they will exercise more control over the blockchain than under a
proof of work mechanism.
137 See M. Rauchs, A. Glidden, B. Gordon, G. Pieters, M. Recanatini, F. Rostand, K.
Vagneur, B. Zhang, Distributed Ledger Technology Systems: A Conceptual Framework, CAMBRIDGE

CENTER FOR ALTERNATIVE FINANCE 55 (2018) (describing the different forms of protocol
design and calling them anarchic, hierarchical, federal, plutocratic or democratic).
138 DE FILIPPI, supra note 4, at 21; Raina S. Haque, Rodrigo Seira Silva-Herzog et al.,
Blockchain Developers and Fiduciary Duty: An Ill-Fitting Framework, CLS BLUE SKY BLOG (Jun.
21, 2019), http://clsbluesky.law.columbia.edu/2019/06/21/fiduciary-duty-and-
blockchain-developers-an-ill-fitting-framework/ [https://perma.cc/WE5Q-QAGJ]
29
impose changes on a blockchain139 or control who may propose protocol

updates.140 They do not further control who can use the blockchain at the
platform layer141 or build applications on top of it142. As a result, blockchain
founders and core developers do not control the blockchain code, interface,
application, data, or benefice.143 They are more or less influential
(“These developers are structurally unable to make decisions to impose changes on

participants in a network”).
139 In fact, founders and core developers may simply submit changes to the majority of
participants, see HAQUE, supra note 138. One may suggest that analyzing how core
developers’ proposals to update a blockchain protocol are being welcomed may prove useful
in determining the degree of their power over other participants.
140 On some blockchains, the core developers are the only ones with the power to
actually propose core code updates to the network. This is the case with Bitcoin Core, in
which “maintainers” “have commit access and are responsible for merging patches from
contributors. They perform a janitorial role merging patches that the team agrees should
be merged. They also act as a final check to ensure that patches are safe and in line with
the project goals. The maintainers’ role is by agreement of project contributors,” see About,
Bitcoin CORE, https://bitcoincore.org/en/about/ [https://perma.cc/U696-6Y2J]. It can
also be that only the core developers can send emergency message to all the nodes on the
network to ask them to adopt updates, as it is the case on the Bitcoin blockchain, see WALCH,
supra note 136, at 4 & 13; see also Arthur Gervais et al., Is Bitcoin a Decentralized Currency?,
EPRINT.IACR.ORG, http://eprint.iacr.org/2013/829.pdf [https://perma.cc/T6H7-
AZ75]. Nevertheless, Bitcoin core developers cannot impose these changes unilaterally, see
Nick Tomaino, The Governance of Blockchains, MEDIUM (Feb. 28, 2017),
https://thecontrol.co/the-governance-of-blockchains-5ba17a4f5da6
[https://perma.cc/MKW4-PHS7]. The same is true for the Ethereum core developers, see
BACON, supra note 133, at 35. In short, protocol developers are not agents of another part
nor delegated with power/authority by other participants. There exists no agency
relationship between developers and any other participants in the network. Protocol
developers lack the ability to bind any other network participant, and no other network
participant has the power to control the voluntary actions of protocol developers. They
have no power to speak authoritatively for the community as a whole. On that, see Raina S.
Haque, Rodrigo Seira Silva-Herzog, Brent A. Plummer & Nelson M. Rosario, Blockchain
Development and Fiduciary Duty, 2 STAN. J. BLOCKCHAIN L. & POL. 139, 178 (2019). See Yessi
Bello Perez, The Controversies of Blockchain Governance and Rough Consensus, TNW HARD FORK
(Jan. 25. 2019), https://thenextweb.com/hardfork/2019/01/25/the-controversies-of-
blockchain-governance-and-rough-consensus/ [https://perma.cc/45W8-LXT7]
(explaining that changes to the core code are submitted to nodes. They may vote for
implementing them, or for rejecting them). See also S Azouvi, M Maller & S. Meiklejohn,
Egalitarian Society or Benevolent Dictatorship: The State of Cryptocurrency Governance, SMEIKLEJ.COM
(2018), https://smeiklej.com/files/bitcoin18b.pdf [https://perma.cc/S2GR-UJD5]
(explaining that discussions regarding blockchain updates can, nonetheless, be centralized).
141 Only permissioned blockchains permit to do that, see HALABURDA & MUELLER-
BLOCH, supra note 136. As for permissionless blockchain, as Tapscott & Tapscott put it,
“nothing passes through a central third party; nothing is stored on a central server,”
TAPSCOTT & TAPSCOTT, supra note 101, at 35.
142 CATALINI, supra note 72 (arguing that “the token can bootstrap the development of
an entire innovation ecosystem where anyone can build novel applications on top of the
underlying protocol without requiring permission from a network operator or
intermediary”).
143 Nick Tomaino, The Slow Death of the Firm - The Control, THECONTROL.CO (Oct. 21,
2017), https://thecontrol.co/the-slow-death-of-the-firm-1bd6cc81286b
[https://perma.cc/H59E-WMZY]. On Bitcoin, for instance, “developers who propose
changes to the software can let the development team know. If it is a simple noncontroversial
change then they will adopt it. If not, the developer is supposed to post the change and it will
be adopted if there is a broad consensus in the community that it should be,” EVANS, supra
note 128, at 1 & 16. Also, Peder Østbye, The Adequacy of Competition Policy for Cryptocurrency
30
“advisors,”144 but their influence is always limited by the fact that blockchain
participants will almost always maximize their personal benefit.145 Social
norms further limit them to the extent that they may fear a hard fork146 if
they do not manage to obtain a sufficiently broad consensus among the group
of users and minors.147 It limits all the more so their willingness to act against
the interests of other participants.148
One could underline, nevertheless, that these limits to their power are linked
to the decentralized nature of blockchain governance, which is not a
necessary feature but needs to be enacted.149 New blockchains may appear
in which greater power will be given to the founders and core developers.150
Two inherent limits to such blockchain must nonetheless be noticed from the
get-go. First, the extent to which (re)centralized blockchain could thrive
remains to be seen.151 Indeed, such blockchain could create mistrust by
confining power in the hands of a few, and, as such, could disincentivize users
to join them. Second, (re)centralized blockchains would to some degree
function less efficiently than truly decentralized blockchains because all of its
participants would not have a personal interest in improving it any longer.
This lack, of efficiency, even if it only concerns certain types of transactions,
Markets, SSRN 15 (Aug. 24, 2017) https://ssrn.com/abstract=3025732

[https://perma.cc/2DQ5-7PFU] (stressing that blockchain’s creators inability to alter the
protocol or manipulate the transactions on the ledger is a key characteristic that should be
taken into account in modeling competition law).
144 TAPSCOTT & TAPSCOTT, supra note 101, at 89; also, HAQUE, supra note 138 (“A
protocol developers’ ability to influence the welfare of the cryptocurrency holder - i.e., the
value of the cryptocurrency - is therefore limited to proposing and advocating for a solution
that the community may or may not adopt”). One may stress, nonetheless, that when these
“advisors” are all working in the same space, they may collude to impose these changes
more easily, see BUTERIN, supra note 131.
145 TOMAINO, supra note 143. Generally, on the importance of product design as a way
to control the use of the product, see LESSIG, supra note 134. This explains why, after the
DAO Ethereum fork in 2015, the Ethereum and Ethereum classic have survived: miners
have switched from one to the other, depending on the gains they could get. For more on
the Ethereum majors forks, see Ashley Viens, Mapping the Most Important Ethereum Forks,
VISUAL CAPITALIST (Nov. 26, 2019) https://www.visualcapitalist.com/mapping-major-
ethereum-forks/ [https://perma.cc/5BT7-ASWF].
146 BACON, supra note 133, at 36.
147 See TechCrunch, Blockchain Governance with Vlad Zamfir (Ethereum Foundation) at Ethereum
Meetup 2018, YOUTUBE.COM, https://www.youtube.com/watch?v=bk3shfEYWN4

[https://perma.cc/QZG6-R6XN] (explaining that when a decision is considered to be
illegitimate because it does not follow the norms, there is a high risk of forking)
148 See Dirk A. Zetzsche, Ross P. Buckley & Douglas W. Arner, The Distributed Liability of
Distributed Ledgers: Legal Risks of Blockchain, 2018 U. ILL. L. REV. 1361 (2018) (explaining that
core developers may want to collaborate with other participants).
149 See HALABURDA & MUELLER-BLOCH, supra note 136.
150 For instance, one can imagine a blockchain in which the consensus mechanism could
involve the blockchain foundation in the validation of all transactions.

151 EVANS, supra note 128, at 1 & 16 (stressing that blockchains with flexible and
centralized governance systems would not necessarily secure the same amount of interest
on the part of laborers and users).
31
would be a brake on these blockchains, which probably explains why, to this

day, they fail to prosper.
The power of users.152 On permissionless public blockchains, users

propose new transactions. Any participant can choose to become a user.153
They exercise power over the blockchain to the extent that their decision to
use it (or not) is at the core of the blockchain economic value.154 Their
influence goes from influencing transaction fees155 to providing additional
value by developing applications running on top of the platform layer.156
They can also force hard forks on the blockchain.157 However, their power
is limited by the fact that they cannot exercise coordinated control as their
behaviors are highly decentralized and spontaneous in nature.158 It creates
an architectural limit and makes their behavior primarily dependent on
prices.159
The power of miners.160 On permissionless public blockchains, miners

validate transactions assembled into blocks. Any participant can choose to
become miner.161 They may influence the blockchain by realizing a 51%
152 See Zetzsche, supra note 148, at 1384 (making a distinction between qualified users,
such as intermediaries, and simple users). One may want to mention that users may or may
not be coin holders.
154 SCHREPEL, supra note 6, at 296 (studying network effects on public and private
blockchains). Also, Mappo, Blockchain Governance 101, MEDIUM (Jan. 14, 2019),
https://medium.com/aelfblockchain/blockchain-governance-101-eb2d769e85c5
[https://perma.cc/2LB8-32NV] (explaining that users may indirectly hinder network
effect by implementing unpopular decisions).
155 Hanna Halaburda & Christoph Mueller-Bloch, Will We Realize Blockchain’s Promise of
Decentralization, HARVARD BUSINESS REVIEW (Sep. 4, 2019),

https://hbr.org/2019/09/will-we-realize-blockchains-promise-of-decentralization
[https://perma.cc/PM6L-TT7H].
156 See Vitalik Buterin, Notes on Blockchain Governance, VITALIK.CA (Dec. 17, 2017),
https://vitalik.ca/general/2017/12/17/voting.html [https://perma.cc/ADP8-UCG4]
(stressing that users’ influence is limited by their potential lack of technical knowledge). Also,
Balázs Bodó & Alexandra Giannopoulou, The Logics of Technology Decentralization – The Case of
Distributed Ledger Technologies, 2019-05 AMSTERDAM LAW SCHOOL RESEARCH PAPER 1, 8 (2019)
(developing that idea that core developers may use their technical skills to influence others).
157 Jake Frankenfield, Hard Fork (Blockchain) Definition, INVESTOPEDIA (Nov. 25, 2019)
https://www.investopedia.com/terms/h/hard-fork.asp [https://perma.cc/M9G2-
7NVE].
158 BUTERIN, supra note 131. In the words of Vitalik Buterin, blockchain communities
are “using coordination problems” to their advantage, preventing one group of users from
taking over, see Vitalik Buterin, Engineering Security Through Coordination Problems, VITALIK.CA
(May 8, 2017) https://vitalik.ca/general/2017/05/08/coordination_problems.html
[https://perma.cc/C3U5-KE9S].
159 LESSIG, supra note 134, at 3.
160 See EVANS, supra note 128, at 1 & 2 (explaining that “the open-source public ledger
platforms require another class of participants—the laborers who perform transaction

processing—that require pecuniary compensation as an integral element of the protocol.”
These laborers are called miners within the blockchain sphere).
32
attack162 and forcing a soft fork of the blockchain163. The risk is higher when
miners are grouped into mining pools.164 In such a scenario, the blockchain
protocol is changed to loosen the rule-set enforced by full nodes.165 Such a
change occurs when enough hashing power, or energy expended to mine a
cryptocurrency, is devoted to it.166 The power of miners to initiate soft forks
is nonetheless limited by the blockchain architecture and social norms
combined, namely, by the fact that they must convince blockchain
participants operating as nodes to run the new version of the software167 and
users to transact on the new fork.168 Miners also suffer from market
constraints as initiating a soft fork may decrease the value of the tokens they
own, considering the fact that their value depends on the blockchain
robustness.169 The price mechanism also mainly guides their action, creating
a robust market-related constraint.
162 See Primavera De Filippi & Benjamin Loveluck, The Invisible Politics of Bitcoin:
Governance Crisis of a Decentralised Infrastructure, 5 INTERNET POL'Y REV 1, 10–11 (2016).
163 BUTERIN, supra note 131 (arguing that the risk is particularly high when there is a
strong community spirit); Vitalik Buterin, Engineering Security Through Coordination Problems,
VITALIK.CA (May 8, 2017)
https://vitalik.ca/general/2017/05/08/coordination_problems.html
[https://perma.cc/C3U5-KE9S] (underlining that “it is theoretically possible for miners to
switch 99% of their hashpower to a chain with new rules” but in practice “quite hard to
do”). More generally, see Taylor Pearson, The Blockchain Man, RIBBONFARM (Oct. 10, 2017),
https://www.ribbonfarm.com/2017/10/10/the-blockchain-man/
[https://perma.cc/76FC-EK42] (contending that “appeals to authority, and perhaps
violence, will be replaced by forking. If you disagree with a decision, you can fork a new
blockchain”).
164 Mining pools may indeed facilitate collusion, for instance, if all miners are in the
same space. Nonetheless, whether there is a mining pool or not does not shift the paradigm
as, in any case, such pools only reflect the will of their members. On the subject, see
WRIGHT, supra note 103. Generally, individual miners do not have enough computing
power to mine a block in widely used blockchains successfully, hence the existence of mining
pools, see Philipp Hacker, Corporate Governance for Complex Cryptocurrencies? A Framework for
Stability and Decision Making in Blockchain-Based Organizations, SSRN 32,
[https://perma.cc/DYZ2-THLJ] (last revised Sep. 14, 2019). Miners may nonetheless,
even at their individual level, prioritize transactions and exercise a form of power as such,
see N. Malik, M. Aseri, et al., Why Bitcoin Will Fail to Scale?, SSRN,
[https://perma.cc/TGB5-FJR6] (last revised Nov. 15, 2019).
165 Full nodes are blockchain participants storing an entire copy of the blockchain ledger
on their computers, for more, see supra Part II.A.1.a.

166 Kristian Soltes, The First Blockchain Antitrust Case. Or Is It?, CONSTANTINE CANON
BLOG (May 29, 2019), https://constantinecannon.com/2019/05/29/the-first-blockchain-

antitrust-case-or-is-it/ [https://perma.cc/RVB9-V96J].
167 BACON, supra note 133, at 37 (showing that there is no guarantee that nodes will
agree to fork).
168 John Light, The differences between a hard fork, a soft fork, and a chain split, and what they mean
for the future of bitcoin, MEDIUM (Sep. 25, 2017), https://medium.com/@lightcoin/the-

differences-between-a-hard-fork-a-soft-fork-and-a-chain-split-and-what-they-mean-for-
the-769273f358c9 [https://perma.cc/ABQ3-GF7L]. In the case of a soft fork, the
blockchain remains however compatible with previous versions despite the modification to
its protocol, see HAQUE, supra note 138 (new transactions and new blocks can still be
processed on the blockchain, as long as they do not violate the new rules of the protocol).
169 See Satoshi Nakamoto, Bitcoin: A Peer-to-Peer Electronic Cash System, 4 BITCOIN.ORG
33
This overall balance of power, common to all blockchains, is the general

analytical framework (as illustrated below) within which to analyze whether one
of these groups, on a case-by-case basis, has sufficient influence to be
qualified as control under antitrust or competition law. One may underline
once more that these constraints mainly apply when these groups act on their
own. When blockchain participants are collaborating, they can evade many
of these constraints and, as a result, take control over the blockchain.
Appendix n°1: A representation of blockchain power game
Finally, one may want to mention that core developers, users and miners,
may also store a copy of the blockchain ledger. When doing so, their
computers are labeled as light nodes if they only store a subset of the
blockchain ledger, and full nodes if they store a copy of the entire
https://bitcoin.org/bitcoin.pdf [https://perma.cc/KJT5-KJ43]. Also, ZAMFIR, supra note

136 (arguing that in the case of a fork, users support a “loss of network effects as a result of
lack of consensus makes the trademark lose value”).
34
blockchain.170 Although these nodes are not actors on the blockchain, but a
tool ensuring its proper functioning, storing the blockchain copy comes with
power. First, blockchain participants which are nodes may alter the copy of
the blockchain altogether.171 Second, they may also (threaten to) validate
blocks in which there is double-spending172 as they are in charge of
preventing users from spending the same coin twice by checking the
proposed transaction against a list of previous unspent transaction outputs.
In short, these blockchain participants are ensuring the reliability of
blockchains, and in turn, their value. Their power is however mainly limited
by the fact that they cannot either control or influence transactions, which is
an architectural limitation.
2. Transposition to Antitrust and Competition Law:

the Blockchain Nucleus
So far, the theory of granularity has allowed us to determine what are the
different forms of power enjoyed by blockchain participants. One must now
detail how to identify a legal fiction exercising a form of control to which
antitrust and competition law can be applied.173 To this end, we explain what
a blockchain nucleus is and analyze the influence it is exercising over
blockchain participants (a). We then detail how to define that nucleus in
antitrust and competition law analysis (b).
a. The blockchain nucleus: principles, usefulness, and challenges
The nucleus. None of the three types of blockchain participants can impose
their power on other groups to the point of taking complete control over the
blockchain. Blockchains are indeed decentralized and prevent the exercise
of vertical power, which remains a significant difference with the firm in
which a group, or sometimes even an individual, can control the other
participants and force them to collaborate.

171 HACKER, supra note 164 (explaining that blockchain participants operating as nodes
are confirming authenticity through consensus). See WAELBROEK, supra note 73, at 22
(blockchain participants operating as nodes have voting power on changes in the
communication protocol). Lastly, see Felix Irresberger, Coin concentration of Proof-of-Stake
blockchains, 19-04 LEEDS UNIVERSITY BUSINESS SCHOOL WORKING PAPER 1, 3 (2019)
(explaining that nodes with more coins have more voting power).
172 In such a scenario, blockchain participants operating as nodes would add these blocks
to their local copy of the blockchain and broadcasts it to other nodes on the network.
173 The usefulness of antitrust and competition law is here presumed. The debate
regarding antitrust legitimacy until it “can be provided with a basis for proceeding in a more
self-confident way,” is left apart. On it, see Oliver E. Williamson, Allocative Efficiency and the
Limits of Antitrust, 59 THE AMERICAN ECONOMIC REVIEW 105, 116 (1969).
35
That being said, even with horizontal and decentralized governance, a group
of participants may achieve a form of direct control over the blockchain by
collaborating, and as such, by circumventing (some of) the constraints
imposed on them.174 We contend that such a group exists for each
blockchain,175 and we call it the nucleus. The nucleus includes all the
participants who have a personal interest (albeit transiently) in always
collaborating with each other.176 They do not compete, and although their
short-term interests may diverge, they seek to achieve the same long-term
goal: ensuring the blockchain survival.177 Here lies the nucleus’ raison d'être.
In its absence, the blockchain existence is random and cannot be ensured or
maximized.
Usefulness. Assessing which participants are part of the nucleus permits to

determine which of them have decided to join forces to escape constraints
with the aim of influencing, and ultimately, controlling the blockchain.
Differently put, it leads to the identification of the participants who can be
held liable for a breach of antitrust and competition law when it is shown
that they have used such influence.178 Identifying the nucleus amounts to the
creation of a legal fiction to which the law can be applied, such as:
174 See the appendix n°1.

175 See ZAMFIR, supra note 136 (for a blockchain to be successful, collaboration is needed).
176 Within the firm, all participants have an interest in its survival. Within the
blockchain, only those within the nucleus have a similar interest. The work of Karl Friston
is helpful, by analogy, in understanding how organisms establish their form and the
“interest” they also have in staying together, see for instance Karl Friston et al., Knowing One’s
Place: A Free-Energy Approach to Pattern Regulation, 12 J. R. SOC. INTERFACE 1 (2015). A parallel
may indeed be drawn between the principle of free energy, explaining how biological
systems maintain their order, and the blockchain nucleus, dealing with the question of how
public permissionless blockchain maintain their existence. For further explanations on the
idea of free energy and how cells stay together in the real-space, see Shaun Raviv, The Genius
Neuroscientist Who Might Hold the Key to True AI, WIRED, https://www.wired.com/story/karl-
friston-free-energy-principle-artificial-intelligence/ [https://perma.cc/9M44-45TG].
177 Again, blockchain is not “a multiple team of horses drawing a vehicle under the
control of a single driver,” see Copperweld Corp. et al. v. Independence Tube Corp., 467
U.S. 771 (1984). Horses (participants) may go any direction they want, except if they have
an interest not to, i.e., if there are part of the nucleus.
178 In a sense, identifying the blockchain nucleus relates to the theory of incentives as
developed by HOLMSTRÖM & MILGROM, supra note 32, at 972. The nucleus allows
anticipating the extent to which the participants cooperate with each other, or conversely,
compete with each other.
36
Appendix n°2: A representation of the blockchain nucleus
Courts and antitrust or competition authorities will have to decide on the

nucleus size, differently put, on which participants they wish to include in
this legal fiction to be held liable for anti-competitive practices. Of course,
the further away a participant will be from the nucleus very center, the more
difficult it will become to prove that he could have influenced the practice of
other participants.
Absent the identification of a blockchain nucleus, Section 1 of the Sherman

Act and Article 101 of the TFEU would also be held in check.179 Holding
that all users of a blockchain are individual firms would prevent the
qualifications of collusive agreements at the level of public permissionless
179 See supra I.A.2.
37
blockchains. The same goes for Section 2 of the Sherman Act and Article
102 of the TFEU as it would create a significant legal loophole for all
strategies at a more general level than the one of individuals. In short,
determining the nucleus size is central in identifying which participants are
collaborating to escape the constraints imposed on them to control the
blockchain, and potentially, if these participants are abusing their control for
anti-competitive purposes.
Challenge. Defining precisely the nucleus size cannot be done using all the
criteria that are usually considered when studying the governance of an
organization, namely, “administrative control and the contract law
regime.”180
Regarding administrative control, blockchain makes the separation between

ownership and control obsolete insofar as whoever holds a token is the only
person who can exercise control over it.181 The presumption used in antitrust
and competition law that a parent company can be held liable for all of the
anti-competitive practices implemented by its subsidiary, on the sole idea
that owning that subsidiary confers a power of command and control, is
therefore obsolete.182 Indeed, the absence of a clearly defined hierarchy in
public permissionless blockchain makes it impossible to impose a joint
direction on all participants. Blockchain direction emerges from its various
actors who act within the framework of what smart contracts permit.183

181 See Melvyn Weeks, The Evolution and Design of Digital Economies 9 (2018); also, see Hidden
Forces Podcast, Joseph Lubin | ConsenSys and the Nature of the Firm in a Decentralized Economy,
SOUNDCLOUD.COM (Aug. 20, 2018), https://soundcloud.com/hiddenforces/joseph-lubin-
consensys-and-the [https://perma.cc/2FB4-29L4] (arguing that the mere possession of a
token is equivalent to the possession of the proposed service). One may want to underline
that blockchain participants stores their tokens in their personal wallet, and as such, bear
their own risks.
182 Roger D. Blair & David L. Kaserman, Ownership and Control in the Modern Corporation:
Antitrust Implications, 11 JOURNAL OF BUSINESS RESEARCH 333, 338 (1983) (generally

speaking, the accountability of practices in antitrust and competition law rests with those
who hold ownership, although there are individual sanctions that can be imposed on
managers who hold control over the activity). Because the notion of “firm” is defined
according to the control that is held over an activity, antitrust and competition law
reconciles ownership and control by assuming that the one who holds ownership also holds
control over that activity. For this reason, antitrust and competition law is not primarily
focused on firms’ internal structure, see WALLER, supra note 63, at 838.
183 Directions emerge within blockchain without suffering from the tragedy of the
commons. Generally, see Garrett Hardin, The Tragedy of the Commons, 62 SCIENCE 1243
(1968) (explaining that the term is used to describe the situation in which individuals have
an incentive to damage the collective good); also, BENKLER, supra note 69, at 378 (applying
the tragedy of the commons to open source software); Mike Maples Jr, Crypto Commons,
MEDIUM (2019), https://blog.usejournal.com/crypto-commons-da602fb98138
[https://perma.cc/F8FH-362X] (discussing “Crypto Commons”), and Oliver E.
Williamson, Markets and Hierarchies: Some Elementary Considerations, 63(2) THE AMERICAN
ECONOMIC REVIEW 321 (1973) (underlining the issue of the commons within peer groups).
Blockchain creates a decentralized governance system without direct “mutual coercion,
mutually agreed upon” and without transferring commons ownership to a few individuals,
38
Regarding contracts, one must underline that transactions implemented on

blockchain are not primarily guided by contract law, but by cryptographic
rules.184 The governance of blockchain is, for that other reason, different
from the firms’. The main challenge, in this regard, is then to identify the
criteria to assess each participant’s capacity to act independently, i.e., the
distance from the nucleus, without creating type 1 or 2 errors.
b. Defining the nucleus size: a primer
Determining the nucleus size to serve antitrust and competition law analysis
implies to establish which participants have an interest in the blockchain
survival.185 To this end, the nucleus size must be assessed on the basis of
concrete and quantifiable elements to ensure legal certainty and to limit legal
errors. At the same time, the analysis of the selected criteria must be
comprehensible enough so that blockchain participants and public
authorities can evaluate them without incurring excessive costs.
Within the firm, a top-down power of command and control ensures the
entity’s survival by reducing transaction costs. Within blockchains, the
nucleus protects its own interest using various means. Because blockchain is
horizontal and decentralized, it is necessary to consider several elements that,
taken together, make it possible to assess the capacity of control. Only then
creating a legal fiction around the participants exercising such control—the
nucleus—would be justified. To this end, we propose evaluating the technical
capacity to enjoy a restricted and horizontal power of command and control,
the capacity to interfere with the blockchain economic value, and the one to
influence the blockchain norms.
but through cryptographically secure constitutional smart contracts. Specifically, on the

interaction between blockchain and the commons see DAVIDSON, supra note 69, at 13
(arguing that blockchain “provides a technical solution (cryptographic consensus) to the
problem of cooperation in joint or group production at scale while still maintaining the
benefits of commons-type (i.e., polycentric) institutional governance”). As a consequence,
neither the core developers, the users or the miners can exercise direct power. They all take
part in insufflating a direction to the blockchain. They do so to different degrees, according
to their distance from the nucleus.
184 See Primavera de Filippi & Aaron Wright, Blockchain and the Law: The Rule of Code (2018)
(defining cryptographic rules as Lex Cryptographia), also see Thibault Schrepel, Anarchy,
State, and Blockchain Utopia: Rule of Law versus Lex Cryptographia, SSRN (Nov. 22, 2019),
[https://perma.cc/5HSS-NW9B].
185 Of course, the nucleus is constantly evolving. Some participants who are part of it
may exit it over time, and vice versa. Identifying the nucleus’ at a given moment remains
useful for antitrust and competition law purposes, as we intend to show.
39
The first element in determining which participants are part of the nucleus is
the technical ability to enjoy a restricted and horizontal power of
command and control. To this end, one must assess the architectural
characteristics of each blockchain to assess the ability of a few users to impose
decisions on others. The more a group of users can control others, the more
they may achieve the blockchain survival and, accordingly, enter the nucleus.
In fact, the original design of a blockchain may give more or less extensive
power to one of the three groups of users, whether it is to put them in charge
of implementing the execution of transactions, to designate them as miners
in certain specific cases, or to change the blockchain design unilaterally.
Besides, some blockchains may also use several mechanisms based on the
platform layer to create governance, off-chain or side-chain, in a more
traditional sense of the term.186 In a sense, this first element of defining the
blockchain nucleus is tightly linked to the Lex Cryptographia—rules administered
through self-executing smart contracts and decentralized mechanisms.187 It
allows analyzing the extent to which a group of users controls part of the
blockchain and can thus reduce conflicts to ensure its profits.188
The second element for assessing the nucleus size is the ability to interfere
with the blockchain economic value.189 Indeed, when some users
govern the pricing structures and may change incentives, they have indirect
control over the blockchain to the extent that, although the creation of most
public permissionless blockchains are ideological by nature, their users very
often seek to achieve profits.190 The identification of the participants holding
such capacity is made thanks to previous and actual investments in the
blockchain.191 Generally speaking, the more a participant has invested in the
186 On the subject, see ZAMFIR, supra note 136 (arguing against the idea that blockchain
nodes should automatically upgrade when an on-chain governance process decides on an

upgrade because “blockchain governance is too important for us to let a small handful of
cryptocurrency whales make arbitrary decisions”).
187 Primavera de Filippi & Aaron Wright, Decentralized Blockchain Technology and the Rise of
Lex Cryptographia, SSRN 1, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664

[https://perma.cc/KAP7-AVPP].
188 This analysis, in other words, focuses on the three elements of the Heuristic Model
of Firm as Governance Structure as defined by Williamson, see WILLIAMSON, supra note 31,
at 180 (namely conflict, mutuality and order).
189 Our analysis, here, borrows from the theory of incentive system developed about the
firm by JENSEN, supra note 32, at 78; see our developments in part I. A. 1. b.
190 The sense of community is not a strong transaction determinant as one may read, for
instance, in Sebastien Meunier & Danni Zhao-Meunier, Bitcoin, Distributed Ledgers and the Theory
of the Firm, SSRN (2019), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3327971
[https://perma.cc/8XZ6-55CQ]. Profits very often take over that sense of community, see,
for instance, what happened after the BCH hard fork, Nick James, Mining Bitcoin Forks Has
Become More Profitable Than BTC Itself, Miners Pulling Out, ETHEREUM WORLD NEWS (Nov.
27, 2018), https://ethereumworldnews.com/mining-bitcoin-forks-has-become-more-
profitable-than-btc-itself-miners-pulling-out/ [https://perma.cc/9LVR-YBLZ].
191 These investments can be of a financial nature but can also be measured in terms of
time and trust devoted to the blockchain. For instance, a participant with numerous smart
contracts running on a blockchain has a strong interest in its survival. For an explanation of
40
blockchain, the more that user has an incentive in finding a way of

controlling its economic value.192 As such, identifying which participants
have created or improved the blockchain193 may show an interest in
capturing control to ensure the blockchain survival. The same goes for those
who are holding a significant number of tokens.194 As for the actual exercise
of such power, it may be shown by any interference with the blockchain
attractiveness and incentives. For instance, changing the size of each block
may alter the number and types of transactions, but also, the blockchain
overall value.195 The same goes for proposing modifications to the core code,
which may further attract new participants.
The third element for determining which participants are part of the nucleus is
the ability to influence the blockchain norms.196 Here, norms are
defined as the “constraints imposed not through the organized or centralized
actions of a state, but through the many slight and sometimes forceful
sanctions that members of a community impose on each other,”197 in other
words, the unwritten rules that one often feels compelled to follow.198 The
more a participant can incentivize others to behave in a certain way, or else
they will be rejected from the community, the more they exercise control
over the blockchain general direction.199 For instance, when core developers
can get all of their proposed modifications to the core code adopted by other
participants, they constraint these participants to maximize the blockchain
survival, potentially to the detriment of their particular interests.
Overall, one may contend that only the analysis of the above-mentioned
abilities makes it possible to analyze who is ultimately controlling the
blockchain to ensure its survival.200
why having invested (in something) in the past creates a strong incentive to keep on investing,
in particular, because of sunk costs and path dependence, Hai R. Arkes & Catherine Blumer,
The Psychology of Sunk Costs, 35 ORGANIZATIONAL BEHAV. AND HUM. DECISION PROCESSES
124, 124 (1984) (explaining that “the sunk cost effect is manifested in a greater tendency to
continue an endeavor once an investment in money, effort, or time has been made.”).
192 For that reason, core developers tend to be closer to the nucleus than simple users.
193 These are usually the core developers.
194 Core developers, users and miners may all be coin or token holders, the reason why
this Article does not distinguish coin holders as a fourth independent users’ category.
195 See SOLTES, supra note 166.
196 See ZAMFIR, supra note 136 (discussing norms in blockchains).
197 LESSIG, supra note 134, at 340.
198 See ZAMFIR, supra note 136 (blockchain governance is not merely an issue of design, it is
also a result of politics). Also, Ying-Ying Hsieh, Jean-Philippe Vergne & Sha Wang, The Internal
Land External Governance of Blockchain-Based Organizations, in BITCOIN AND BEYOND:
CRYPTOCURRENCIES, BLOCKCHAINS AND GLOBAL GOVERNANCE (Routledge, 2017) (norms
may also come from outside the blockchain, for instance, from media or general public interest).
199 See Nigel Dodd, The Social Life of Bitcoin, SAGE JOURNAL 10 (Dec. 17, 2017),
https://journals.sagepub.com/doi/abs/10.1177/0263276417746464?journalCode=tcsa
[https://perma.cc/TP49-P9LG] (studying the influence of norms in the Bitcoin blockchain).
200 Often, it may prove necessary to combine the analysis of these three different
capacities to establish a body of corroborating evidence. Such methodology is often used in
41
B. The Theory of Granularity in Action
The theory of granularity leads to identifying what constitutes the blockchain

nucleus. It permits the creation of a legal fiction to which antitrust and
competition law can be applied. Thanks to it, the analysis of relevant markets
and market power becomes possible (1). The theory of granularity also helps
in qualifying anti-competitive practices and assigning liability (2).
1. The Theory of Granularity in Support of Market Analysis
Public permissionless blockchains are horizontal and decentralized. The

absence of borders makes it very difficult to classify them, and as such, to
delimit relevant markets (a). The analyze of market power, as a consequence,
is also precluded (b). The theory of granularity helps in overcoming this
difficulty on both fronts.
a. The determination of relevant market thanks to the theory of granularity
Centralized platforms are at the very center of a vast majority of research

concerning two-sided markets.201 While the definition of the relevant market
in which these platforms operate is still subject to debate,202 decentralized
platforms raise new issues that also need to be urgently analyzed.
Generally speaking, it is considered that there is a single market

encompassing both sides of two-sided transaction platforms and that the
demand side, i.e., the service proposed by the platform, is key in determining
the relevant market for its entirety.203 Product markets are then defined first
antitrust and competition law, see, for instance, Case C-407/08 P, Knauf Gips v
Commission, 2010 I-06375, para. 65.; Monsanto Co. v. Spray-Rite Service Corp., 465 US
752, 768 (1984) (stating that circumstantial evidence that reasonably tends to prove that the
parties had a conscious commitment to a common scheme designed to achieve an unlawful
objective could be used).
201 The first article on the subject was Jean-Charles Rochet & Jean Tirole, Platform
Competition in Two-Sided Markets, 1 J. EUR. ECON. ASSOC. 990 (2003). It has been cited 4483
times according to Google Scholar, including 2 400 times since the 1st January of 2015. See
also, WEEKS, supra note 181, at 27 (explaining that “economists and computer scientists are
learning how to adapt the tools developed to build and understand the dynamics of
centralised marketplaces” to decentralised markets).
202 See Lapo Filistrucchi, Damien Geradin, Eric van Damme & Pauline Affeldt, Market
Definition in Two-Sided Markets: Theory and Practice, 10(2) J. Comp. L. & Econ. 293, 293 (2014)
(introducing a conceptual framework for defining relevant markets in two-sided markets).
More generally, see Louis Kaplow, Why (Ever) Define Markets?, 124 HARV. L. REV. 81 (2010).
For an answer, see Gregory J. Werden, Why (Ever) Define Markets: An Answer to Professor Kaplow,
78 ANTITRUST L.J. 729 (2012).
203 For an analytical framework, see FILISTRUCCHI, supra note 202, at 296. For case-
law, see EC Decision Case AT.40099, Google Android, para 213: “for the definition of the
relevant market, demand-side substitution constitutes the most immediate and effective
disciplinary force on the suppliers of a given product,” although that case concerned a two-
42
by identifying the nature of each product or service, and then by carrying

out a substitutability test.204 Such an analysis is made possible by the fact that
centralized platforms are often centered on one or a few products or services.
It opens a way for distinguishing between them, and also, for comparing
them with each other. For instance, Uber and Airbnb offer one easily
identifiable service that determines their relevant product or service market.
Such a two-step reasoning is not easily transposable to public permissionless

blockchains. When looking at the type of services they offer to categorize
them, one may distinguish between three major types of blockchains.205 The
first, blockchain 1.0, allows only one type of application, namely, crypto-
currencies. They compete with each other. Blockchain 2.0 permits the
execution of smart contracts. Smart contracts can be used to automate an
almost infinite number of transactions, and for that reason, it is not possible
to consider that all blockchains 2.0 compete with each other as they may
automate transactions of a potentially very different nature. It is also not
possible, a priori, to categorize one blockchain into a single product market.
Finally, blockchains 3.0 make it possible to do everything a computer can do,
but in a decentralized way.206 The creators of such blockchains have no
control over which products and services can be offered. For that reason,
such blockchains cannot be considered as part of a single relevant market.
For example, unlike smartphone OS or computer operating systems,
blockchains 3.0 are not intended to work on predefined types of products
(such as smartphones). They can operate wherever the Internet is available,
and they may even allow the subsequent development of operating systems.
For that, they may not be categorized into a single relevant product market.
As a consequence, public permissionless blockchain poses new challenges in

terms of relevant markets that cannot be addressed by transposing the
analyzes of centralized two-sided platforms to it. The theory of granularity,
on the contrary, makes it possible to identify a blockchain nucleus and to
provide both a theoretical and practical answer to the issue of determining
the relevant market.
sided non-transaction platform. In the United Stated, the 2010 Horizontal Merger
Guidelines assert that “market definition focuses solely on demand substitution factors,” see
US Dep’t of Justice and FTC, Horizontal Merger Guidelines (2010) § 4. Demand-side
substitution is, however, not the only factor taken into account. Arguing that “competition
cannot be accurately assessed by looking at only one side of the platform in isolation” in
two-sided transaction markets, see Ohio et al. v American Express Co. et al., 585 U.S. ___;
138 S. Ct. 2274 (2018). It remains central.
204 EC Decision Case AT.40099, Google Android, para 213 (“for the definition of the
relevant market, demand-side substitution constitutes the most immediate and effective
disciplinary force on the suppliers of a given product.”); also, JON BAKER, THE ANTITRUST
PARADIGM: RESTORING A COMPETITIVE ECONOMY 185 (Harvard Press 2019).
205 MELANIE SWAN, BLOCKCHAIN: BLUEPRINT FOR A NEW ECONOMY, Preface, IX
(O’Reilly Media 2015).

206 Id.
43
Once the blockchain nucleus has been identified, one may indeed determine
the type of activities around which it is built, i.e., all activities ensuring the
blockchain survival. This way, an analysis of substitutable blockchains on the
demand side becomes possible in practice. Although a vast majority of
blockchains 2.0 and 3.0 can, from a theoretical point of view, compete with
each other, analyzing real alternatives on the demand-side leads to defining
the relevant market. Of course, blockchains may operate in several relevant
markets, just as it is the case for centralized platforms. Nevertheless, the
theory of granularity makes it possible to consolidate each blockchain’s
primary use(s) around which the nucleus converges. Focusing on the
blockchain nucleus also makes it possible to delimit the geographical market
by identifying, from the user’s point of view, all the substitutable blockchains
for the provision of a product or service in a given area.
b. The determination of market power thanks to the theory of granularity
For many years, a number of academic articles have denounced a disconnect

between static legal analysis and the dynamic nature of digital markets.207
Public permissionless blockchains exacerbate this problem insofar as their
borders are not as stable as the firms’. Public permissionless blockchains have
no entry or exit door, freeing its participants to move from one blockchain
to another. Such movability must lead to rethinking antitrust and
competition law analyses in terms of market power.208
The theory of granularity helps in overcoming this difficulty as it permits to

center antitrust and competition law on the participants gathered together
within the same legal fiction. Comparing the different nuclei in the same
relevant market leads to establishing their market power.
Courts, antitrust and competition authorities will have, at this stage (i.e.,
market power determination), to establish the nucleus size accurately.
Defining too large a size would result in including participants who can
exercise no control over others. The size of the nucleus is then to be delimited
207 For instance, see J. Gregory Sidak & David J. Teece, Dynamic Competition in Antitrust
Law, 5 J. COMPETITION L. & ECON 581, 582 (2009) (underlining that antitrust and
competition law has been static for several decades already), also, Jan Krämer & Michael
Wohlfarth, Market Power, Regulatory Convergence, and the Role of Data in Digital Markets, 42
TELECOMMUN. POLICY 154, 168 (2017) (underlining that, for instance, the dynamic nature
of digital markets is per se in conflict with static approaches to define the relevant market);
Ariel Ezrachi, The Goals of EU Competition Law and the Digital Economy, BEUC DISCUSSION
PAPER 11 (2018) (stressing that one of the challenges for enforcement in the digital age
“pertains to the difficulties in apprehending dynamic changes”).
208 Generally speaking, as it has been stressed by Lawrence Lessig, “an unmovable, and
unmoving, target of regulation, then, is a good start toward regulability,” LESSIG, supra note
134, at 139.
44
in such a way as to include only the participants who are genuinely seeking
the blockchain survival and holds the power to achieve that result.
2. The Theory of Granularity in Support of

the Analysis of Anti-Competitive Practices
The theory of granularity helps in understanding the effects of practices by

recreating an inside and outside (the nucleus), which is required in antitrust
and competition analysis (a). Ultimately, it helps in assigning liability (b).
a. The assessment of anti-competitive practices using the theory of granularity
Firms decide internally of collusive agreements or monopolization practices,

and antitrust and competition law is mostly concerned with their external
effects.209 That distinction between the effects inside and outside the firm
guides the analyses of all potential practices, but public permissionless
blockchains hold it in check. Indeed, the absence of clearly defined
boundaries for such blockchains at the platform layer prevents the distinction
between what’s inside or outside of it. Analyzing practices is immensely
complex for that reason.
The delimitation of a blockchain nucleus reintroduces the possibility of

analyzing internal and external effects since it recreates borders—namely,
inside and outside the legal fiction. It is easily understood through the
example of collusion. As entities cannot collude with themselves, one needs
to delimit their boundaries to analyze whether a collusive practice has
occurred. In that regard, only agreements between two different nuclei
should be worrisome. Agreements between blockchain participants outside
of any nucleus, de facto lacking any ability to control, should not trigger much
competitive concerns.
The same goes for monopolization and abuse of dominance cases as the
theory of granularity makes it possible to define a legal fiction whose market
power will be assessed in relation to others. It subsequently permits to analyze
to what extent practices may constitute an abuse of this power by analyzing
their external effects. Such analysis is not possible when blockchains are only
seen from a distance because blurry decentralized entities in which no
command and control power is exercised do not have visible frontiers.
209 Thibault Schrepel, Antitrust Conversations with Nobel Laureates, CONCURRENTIALISTE
REVIEW, 14 (2018).
45
b. The assignment of liability using the theory of granularity
Jean-Charles Rochet and Jean Tirole have stressed that “platforms are
controlled by competing entities, either profit-maximizing firms or not-for-
profit associations.”210 Decentralized platforms, on the contrary, work
differently since a few users cannot unilaterally decide on a strategy to achieve
their results. In that regard, blockchain analysis differs significantly from the
model introduced by Rochet and Tirole as practices emerge on it rather than
being imposed.211
The decentralized and horizontal nature of blockchain raises two issues in

antitrust and competition law, namely, characterizing anti-competitive
practices, and identifying the person(s) liable for them. The theory of
granularity introduced in this article provides an answer to these two issues.
We have shown that the behaviors of participants in public permissionless

blockchains may be influenced in three ways by the nucleus: the technical
ability to enjoy a restricted and horizontal power of command and control,
the ability to interfere with the blockchain economic value, and the ability to
influence the blockchain norms.212 As a result, the participants in the nucleus
are to be held liable for all illegal conducts committed within the nucleus or
the perimeter it can control or influence to a great degree.213 In practice,
agencies and courts will have to decide on the width of that perimeter on a
case-by-case basis.214 On top of that, by analogy with the American and
210 ROCHET & TIROLE, supra note 201, at 1000 (emphasize added).
211 Blockchains are no “islands of conscious power,” D. H. ROBERTSON, CONTROL OF
INDUSTRY: CAMBRIDGE ECONOMIC HANDBOOKS IV 85 (London: Nisbet, 1923), coining
the terms to describe firms. No overall conscious strategy can be imposed on blockchain. It
is problematic to the extent that objective intent is usually inferred from the implementation
of practices, which no longer seems possible here. Generally, on the role of intent in antitrust
and competition law, see Marina Lao, Aspen Skiing and Trinko: Antitrust Intent and Sacrifice, 73
ANTITRUST L.J. 171, 208 (2005) (discussing the role of intent in the Aspen Skiing and
Trinko cases); Nicolo Zingales, Antitrust Intent In An Age of Algorithmic Nudging, 7 J. ANTITRUST
ENFORCEMENT 386, 387 (2019) (proving the existence of a subjective intent is not required
in antitrust and competition law, but objective intent is).
212 See supra II.A.2.a.
213 One must distinguish between different degrees of distance with the nucleus, i.e.,
different degrees of collaboration or competition with it. This is equivalent to analyzing one
participant’s ability to compete with other participants by moving away from the nucleus
when desired. This capacity is a key criterion in antitrust and competition law for
determining the extent to which economic agents are undertaking, see Joined Cases 40–48,
50, 54–56, 111, 113 & 114/73, Coöperatieve Vereniging Suiker Unie v. Commission,
[1975] ECR. 1663 (1975), para 173. The ECJ has frequently repeated this rule, see, for
instance, Case C-8/08, T-Mobile, 2009 ECR I-4529 (2009), para 32, Case107/82,
Allgemeine Elektrizitäts-Gesellschaft AEG-Telefunken AG v. Commission, 1983 ECR
3151 (1983), para 49. As such, the further a participant is from the nucleus, the less that
nucleus will be able to control it.
214 Some participants may indeed collaborate to ensure the blockchain survival in a
majority of cases—they constitute the nucleus’s nearby perimeter—while other participants

may compete regularly with the nucleus, especially if they have a personal interest in the
46
European liability regimes for parent companies,215 agencies will have to

bring the proof that the nucleus not only had the power to influence other
participants, but that it has actually exercised that influence.216
The nucleus participants, however, should not be held liable for the practices
committed outside of the perimeter they can influence.217 This absence of
liability is not problematic because, in any case, blockchain participants
outside the perimeter of influence are being submitted to different constraints
that deprive them of any power to coordinate their actions effectively to
change the blockchain direction, or to abuse their power with that objective.
Finally, about fine calculation, the nucleus will be subject to the same rules
applied to firms. The turnover realized by the nucleus, based on its
transactions, will serve as a basis for calculating sanctions. Other transactions
the nucleus could not have influenced will be excluded.218
survival of another blockchain.

215 See KOENIG, supra note 45 (explaining that in the United States, courts rely less on
assumptions to infer liability to parent companies than they do in Europe).

216 Case T-24/05, Alliance One International, Inc., formerly Standard Commercial
Corp. and Others v European Commission, 2010 II-05329 (2010).

217 Vitalik Buterin, Control as Liability, VITALIK.CA (May. 9, 2019),
https://vitalik.ca/general/2019/05/09/control_as_liability.html
[https://perma.cc/SR7C-9M43] (“every bit of control you have is a liability”). Liability in
blockchain has to stop where control stops.
218 For instance, in 2018, the total value of bitcoin transactions was $3.2 trillion, Joseph
Young, $3.2 Trillion in Bitcoin Payments Processed in 2018, Better Version of Gold?, CCN.COM
(2019), https://www.ccn.com/3-2-trillion-in-bitcoin-payments-processed-in-2018-is-the-
cryptocurrency-a-better-version-of-gold/ [https://perma.cc/U9ZY-BRDH]. Even a small
percentage of this value represents a very significant and dissuasive sum of money.
47
CONCLUDING THOUGHTS
The theory of the firm, as first introduced by Ronald Coase, is at the heart
of modern antitrust and competition law. Antitrust and competition rules are
being applied to firms when practices are implemented at a broader scale
than the individual level. The firm’s scope and liability are deemed to expend
wherever it can exercise a top-down control to reduce transaction costs. The
reduction of these costs is often achieved by ordering collaboration between
employees, while market participants outside the firm are compelled to compete.
We have shown that, in the absence of vertical control, the theory of the firm
fails in capturing public permissionless blockchain, as well as delimiting their
frontiers. As a result, antitrust and competition laws are inapplicable to such
blockchains. This article puts a solution forward by introducing the “theory
of granularity.” By allowing the analysis of interactions between blockchain
participants, the theory of granularity determines to the extent to which a
form of control—different from that of the firm—can be exercised. It leads
in that regard to the identification of a nucleus that has power over the
blockchain, i.e., a legal fiction to which antitrust and competition law can be
applied. The theory of granularity is, as such, the gateway to antitrust and
competition law in the face of public permissionless blockchains.
Other issues remain to be explored. First, the interactions between non-

public or non-permissionless blockchain participants must be clarified.
Further research is needed in that field to determine whether the forms of
control exercised within these blockchain resembles the firms’, or if the
theory of granularity could also help in this regard.
The second unsolved issue concerns the applicability of antitrust and

competition law to the software layer. For instance, several blockchain
applications permit the introduction of mechanisms, creating a more
traditional form of governance at the level of the platform layer,219 raising
the question of whether blockchains running these apps are then getting
closer to firms. Other applications also aim at the creation of entirely
decentralized organizations. Labeled decentralized autonomous
organizations (“DAO”)220 or decentralized collaborative organizations
219 This is the case, for example, of Backfeed, see Backfeed - Wiki, Golden,
https://golden.com/wiki/Backfeed [https://perma.cc/76Y7-8K7T]. See also PEREZ, supra

note 140 (explaining that on-chain governance takes power away from the miners because
it lies with the blockchain participants operating as nodes).
220 RASKIN, supra note 72, at 336 (a DAO is an organization where the rules of
management are predetermined and run on computers”); also, GARFINKEL, supra note 100,
at 63 (explaining that no leader of a DAO can prevent the relevant smart contracts from
executing in a specified way).
48
(“DCO”),221 they combine a set of smart contracts to imitate firms’

governance; although they are not necessarily recognized as proper firms
under corporate law.222 Moreover, when these decentralized organizations
are combined with firms in the traditional sense of the term,223 new corporate
law issues arise.
Third, research is needed regarding blockchain applications intended for end-

consumers. The software layer is, indeed, influenced by the platform layer
on top of which it functions.224 All blockchain applications remain subject to
the general characteristics of the platform layer, and ultimately, depend on the
three categories of participants that allow the proper execution of smart
contracts and the maintenance of blockchain integrity.225 For that reason,
better understanding the interdependence of all the blockchain layers
appears to be a priority for future work.
Finally, the theory of granularity raises questions about the nature of antitrust
and competition law. This body of law has always had the objective of
ensuring the proper functioning of markets by restricting certain practices
implemented by firms,226 particularly, unnecessary collaborations between
them.227 It has always been concerned with the external effects of firms’
practices. Thanks to the theory of granularity, antitrust and competition law
enters into public permissionless blockchains internal structure and daily
operations. It becomes subatomic.228 This theory also leads to legitimizing
the collaboration of blockchain participants in the nucleus, where it would
have been prohibited in the theory’s absence. For these reasons, the theory
of granularity leads to a complete questioning of what antitrust and
competition law is.
221 DAVIDSON, supra note 69 (explaining that DCO consist of a number of individuals
contributing out of their own free will to a collaborative project); also, Law ReImagined - Demo
- Automated and Autonomous Legal BBLLC, YOUTUBE.COM (Jun. 10, 2019)
https://www.youtube.com/watch?v=UjayMQ_9M9U [https://perma.cc/X492-
HAMK] (for a view of a DAO functioning).
222 See Gravel & Shea, dOrg Launches First Limited Liability DAO, GRAVEL & SHEA BLOG
(Jun. 2019), https://www.gravelshea.com/2019/06/dorg-launches-first-limited-liability-

dao/[https://perma.cc/PQ5N-799S].
223 See GARFINKEL, supra note 100, at 64; also, Eric A. Posner & E. Glen Weyl, Quadratic
Voting as Efficient Corporate Governance, 643 UNIVERSITY OF CHICAGO COASE-SANDOR

INSTITUTE FOR LAW & ECONOMICS RESEARCH PAPER 1 (Oct. 3, 2013); Alexandra
Andhov, Corporations on Blockchain: Opportunities & Challenges, 2019-85 UNIVERSITY OF
COPENHAGEN FACULTY OF LAW RESEARCH PAPER 1, 15 (2019); EVANS, supra note 128, at
1 & 15 (underlining that a for-profit entity could, in principle, operate a public ledger
platform using blockchain).
224 See BUTERIN, supra note 146 (the bottom layer is always the ultimate deciding layer).
225 This is the case, for example, of applications such as Arcade city
(https://arcade.city/) [https://perma.cc/V3WR-85BJ], which operates on the basis of the

Ethereum and which, in fact, does not control its transactions as a firm controls its own.
226 Edward B. Rock, Corporate Law Through an Antitrust Lens, 92 COLUM. L. REV. 497, 497 (1992).
227 Id. at 498.
228 WALLER, supra note 63, at 888 (underlining that, traditionally, corporate governance
is subatomic in nature).
49

Thibault Schrepel-The Theory of Granularity PDF

Uploaded by

Copyright:

Available Formats

You might also like

Thibault Schrepel-The Theory of Granularity PDF

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Thibault Schrepel-The Theory of Granularity PDF

Uploaded by

Copyright:

Available Formats

THE THEORY OF GRANULARITY

A Path for Antitrust in Blockchain Ecosystems

Dr. Thibault Schrepel 1

Assistant Professor, Utrecht University School of Law

INTRODUCTORY THOUGHTS .............................................................................................. 3

1 Assistant Professor at Utrecht University School of Law, Faculty Associate at Harvard

Modern antitrust and competition law relies extensively on the firm as

Against this background, the present article introduces the “theory of

T he theory of the firm is antitrust and competition law most fundamental

In this study, we propose to analyze Coase’s article in light of public

2 Ronald H. Coase, The Nature of the Firm, 4 ECONOMICA 386 (1937).

Public permissionless blockchains are in fact as in the public domain:9 no

Capitalism, 14 J. I. ECON 639 (2018).

Part II of the present Article is dedicated to opening the blockchain black

I. THE THEORY OF THE FIRM:

The firm, as defined by Ronald Coase, is the theoretical pillar on which

A. The Theory of the Firm as a Pillar for Antitrust Law

1. The Theory of The Firm: Not a Marginal Revolution

a. Highlights of Ronald Coase’s Article

Ronald Coase particularly emphasizes the firm ability to deal with

in THE MODERN FIRM, CORPORATE GOVERNANCE, AND INVESTMENT 11, 33 (Per-Olof

improbable that a firm would emerge without the existence of uncertainty”21

b. The impact of Coase’s Article on economics

Ronald Coase’s Article has put transactions at the center of modern

21 COASE, supra note 2, at 392.

Chicago Press 1988).

61(2) AM. ECON. REV 112, 113 (1971).

in developing new tools to understand the firm35 and market dynamics.

2. The Theory of the Firm in Modern Antitrust Law

a. The firm’s boundaries in antitrust and competition law

369, 392 (2005).

Antitrust Act), May 13, 1890, U.S. CAPITOL VISITOR CENTER,

used in sections 1 to 7 of this title shall be deemed to include corporations

Defining the firm’s boundaries in antitrust law is essential as one person

39 See 15 U.S. C. § 7 (1890).

corporate actions are guided or determined” by “separate corporate

simply rely on a theoretical control-based liability).

BETWEEN COMPETITION LAW AND CORPORATE GOVERNANCE 33 (2019).

that “the actual exercise of decisive influence”54 defines firm limits in

Europe BV v Commission of the European Communities, 1996 ECR. l-5457 (1996), in

b. The firm as a pillar of antitrust and competition law

The definition of the firm’s boundaries helps three fundamental steps of

First, establishing the firm’s boundaries helps in determining the extent to

Second, analyzing the firm’s boundaries permits to assess the legality of

Finally, identifying the firm boundaries helps in attributing liability.61 Liability

Generally speaking, it is safe to assume that the broad definition of antitrust

B. The Theory of the Firm Vis-à-Vis Blockchain

61 Case C-204/00 P, Aalborg Portland and Others v Commission, Opinion of AG Ruiz-

18 GEO. MASON L. REV. 833, 835 (2011).

1. Ronald Coase’s Logic in Light of Blockchain

a. The blockchain as a third institutional way

Distinction from other institutions. In Ronald Coase’s theory of the

margin of the comparative efficiency of firms).

Firstly, blockchain leads to economizing on certain transaction costs due to

Conference 2018, YOUTUBE.COM (May 13, 2018),

management challenges grow faster than organizational size).

23, 2017) http://www.tomglocer.com/2017/10/23/blockchain-coase-and-the-theory-of-