Blockchain Technology and Smart Contract PDF

BLOCKCHAIN TECHNOLOGY AND
SMART CONTRACTS FROM A

FINANCIAL LAW PERSPECTIVE
BLOCKCHAIN TECHNOLOGY IN FINANCIAL TRANSACTIONS:

REGULATING A REVOLUTION?
Maurice Demeyer
Student number: 01310663
Supervisor: Prof. Dr. Michel TISON

Commissary: Mr. Evariest CALLENS
A dissertation submitted to Ghent University in partial fulfillment of the requirements for the
degree of Master of Laws.
Academic year: 2017 - 2018

ACKNOWLEDGMENTS
This Master’s dissertation demarks the final moments of my law studies at Ghent University
and therefore, it is the ideal opportunity to express my sincere gratitude to everyone who
participated in making the past five years exciting, intellectually enriching and simply
exceptionally beautiful.
I would like to acknowledge the Institute of Financial Law at Ghent University for decisively
inciting my interest through the various advanced legal courses offered on capital markets,
private banking law and corporate law. I am especially thankful to my thesis supervisor,
professor Michel TISON, for granting me the opportunity to immerse myself in the intriguing
topic of distributed ledger technology and for his continuous guidance, advice and
enthusiasm.
My parents have always given me all chances and freedom, and my family has steadfastly
encouraged me during stressful periods of deadlines and exams. They play a crucial role in
my life and I am very grateful for their love and support.
I would also like to thank my friends in computer engineering, Wout, Nathan and Dries, for
answering my many questions and assessing my statements, for proofreading my work and
for all enlightening discussions we had together.
Finally, I wish to acknowledge the friendly people of Living Tomorrow, Becon, Larcier,
Intersentia and the IFR for hosting blockchain and fintech conferences and for kindly
welcoming me at these outstanding events, where I have gathered numerous important
insights for my research.
Maurice Demeyer
Ghent, 15 May 2018
iii
ABSTRACT (ENGLISH AND DUTCH)
Through extensive study of literature, this Master’s dissertation aims to identify the
opportunities created by the advent of first and second generation distributed ledger
technology in the context of the financial sector. It also assesses the regulatability of various
distributed ledger configurations, and their relation to the pre-existing legal framework.
Finally, it identifies a number of priorities for suggested future law changes.
Our research leads to a categorization of possible distributed ledger technology applications,
based on a varying degree of decentralization: from highly decentralized native crypto-asset
blockchains to permissioned interbank payment and settlement ledgers; from irreversible
smart contracts that are meant to embody the legal agreement to smart contracts with a point
of entry for judicial intervention and which co-exist with a prevailing traditional contract.
This dissertation establishes that all degrees of decentralization are or can be regulated by
one way or another and that certain changes to the existing legal framework are desirable to
maximally benefit from the unique capacities of distributed ledger technology.
Deze Masterscriptie is erop gericht om, op basis van een uitgereide literatuurstudie, de
mogelijkheden te identificeren die “gedeeldgrootboektechnologie” (distributed ledger
technology) van de eerste en tweede generatie biedt in de financiële sector. Het evalueert
bovendien de vatbaarheid voor regulering van verscheidene distributed ledger constellaties,
alsook hun verhouding tot het positief recht. Ten slotte duidt deze scriptie een aantal
prioriteiten aan voor de voorgestelde toekomstige wetswijzigingen.
Ons onderzoek leidde tot een categorisering van de mogelijke distributed ledger
toepassingen, gebaseerd op hun verschillende graad van decentralisatie: gaande van uiterst
gedecentralizeerde native crypto-activa netwerken tot beperkt toegankelijke interbank
betaal- en afwikkelingssystemen; en gaande van onomkeerbare smart contracts die beogen
de onderliggende juridische overeenkomst te vervangen tot smart contracts die rechterlijke
tussenkomst toelaten en die naast een primerend traditioneel contract bestaan.
Deze dissertatie stelt vast dat alle gradaties van decentralisatie aan regulering onderhevig
zijn of op de één of andere manier voor regulering vatbaar zijn, en dat bepaalde
wetswijzigingen wenselijk zijn teneinde maximaal maatschappelijk voordeel te halen uit de
unieke capaciteiten van distributed ledger technology.
v
TABLE OF CONTENTS
ACKNOWLEDGMENTS ....................................................................................................................... iii
ABSTRACT (ENGLISH AND DUTCH) .................................................................................................. v
TABLE OF CONTENTS ........................................................................................................................ vii
TITLE 1. INTRODUCTION................................................................................................................... 11
TITLE 2. BLOCKCHAIN EXPLAINED ............................................................................................... 15
A. INFORMATION TECHNOLOGY ............................................................................................... 15
B. PRINCIPAL CATEGORIZATIONS............................................................................................. 18
1. Network access ......................................................................................................................... 18
2. Stages of development .............................................................................................................. 23
3. Consensus mechanisms ............................................................................................................. 26
C. DLT APPLICATIONS ................................................................................................................. 29
TITLE 3. CRYTPO-ASSETS.................................................................................................................. 31
CHAPTER 1. CRYPTOCURRENCIES ................................................................................................ 31
A. HIGHLY DECENTRALIZED AND PUBLIC – electronic cash ................................................... 32
1. Qualification by an economist ................................................................................................... 32
2. Qualification by a lawyer .......................................................................................................... 34
3. Regulation?............................................................................................................................... 36
B. PERMISSIONED AND PRIVATE – blockchain-based interbank payment solutions .................... 38
1. Interbank financial communication network .............................................................................. 39
2. Interbank money remittance network ......................................................................................... 41
3. Regulation?............................................................................................................................... 42
a. Individual nodes.................................................................................................................... 44
b. Blockchain as an entity ......................................................................................................... 45
c. Transactions.......................................................................................................................... 47
C. CONCLUSION ............................................................................................................................ 48
CHAPTER 2. CRYPTOSECURITIES................................................................................................... 50
A. THE MODERN SECURITIES MARKET – intermediaries, origins and problems ........................ 50
1. Trade lifecycle .......................................................................................................................... 50
vii
2. Origins ..................................................................................................................................... 52
3. Deficiencies of the intermediated infrastructure ......................................................................... 54
4. The blockchain alternatives ....................................................................................................... 55
B. HIGHLY DECENTRALIZED AND PUBLIC – token sales, ICOs and … ..................................... 56
1. vs. cryptocurrencies? ................................................................................................................. 56
2. Regulation? .............................................................................................................................. 59
a. Issuance and market infrastructure......................................................................................... 59
b. Transactions ......................................................................................................................... 61
3. Conclusion................................................................................................................................ 62
C. PERMISSIONED AND PRIVATE – blockchain-based interbank securities clearing and settlement

......................................................................................................................................................... 64
D. CONCLUSION............................................................................................................................ 68
TITLE 4. SMART CONTRACTS .......................................................................................................... 71
CHAPTER 1. ORIGINS........................................................................................................................ 71
CHAPTER 2. CODERS’ APPROACH – code is law............................................................................. 73
CHAPTER 3. SMART LEGAL CONTRACT – law is law, and code is code ......................................... 74
A. RELEVANT DISTINCTIONS ..................................................................................................... 76
1. Operational vs. non-operational contract clauses ........................................................................ 76
2. Automatable vs. non-automatable contractual rights and obligations .......................................... 78
3. Strong vs. weak smart contract code .......................................................................................... 78
4. External vs. internal smart contract models ................................................................................ 79
B. IN PRACTICE ............................................................................................................................. 79
1. External smart legal contract ..................................................................................................... 80
a. Setup and use ........................................................................................................................ 80
b. Non-operational and non-automatable distortions .................................................................. 81
c. Smart bugs: disputes and unwanted output............................................................................. 82
2. Internal smart legal contract ...................................................................................................... 83
a. Setup and relevance .............................................................................................................. 83
b. Non-operational and non-automatable distortions .................................................................. 84
i. General non-operational clauses ....................................................................................... 84
viii
ii. Ambiguous clauses ........................................................................................................... 85
iii. Non-automatable elements............................................................................................... 86
c. Smart bugs: disputes and unwanted output ............................................................................. 86
i. Strong internal smart contract ........................................................................................... 86
ii. Weak internal smart contract ............................................................................................ 88
C. CONCLUSION ............................................................................................................................ 88
TITLE 5. CONCLUDING CONSIDERATIONS ................................................................................... 93
BIBLIOGRAPHY ................................................................................................................................... 95
ix
TITLE 1. INTRODUCTION
Liberty in cyberspace will not come from the absence of the state. Liberty there, as
anywhere, will come from a state of a certain kind.1
– L. LESSIG, Code 2.0
1. In 2009, when Satoshi NAKAMOTO first introduced Bitcoin, the world was absorbed
by the revolutionary idea of a ‘virtual currency.2, 3 Besides physical money, funds held in
accounts in commercial banks and electronic money, a new way to transfer, receive and
even create value was born. The virtual currencies appeared to be groundbreaking because
of their advantages regarding transaction speed around the globe, security, low cost and high
privacy. ‘Bitcoin’ was and still is a buzzword. In fact, the expectations it created and the
2017 ‘investor’ excitement overshadowed the relatively simple but ingenious design of the
coin’s software. Nearly ten years after its release, Bitcoin and its many varieties are still too
volatile to lastingly reshape the landscape of established currencies. Focus now shifted to
the implementation behind the virtual currency transactions: a distributed system of
information ledgers called the ‘blockchain’.4 The variety of this technology’s potential
applications is virtually unlimited and stretches far beyond the original idea of coins and
currencies. Some authors predict that this method to exchange and record information will
innovate almost every aspect of our daily life and economic activities. Accordingly, PETERS
and PANAYI claim that blockchain is the latest disruptive technology.5 The view presented
by Gartner is more cautious but this organization’s well-known ‘Hype Cycle’ positions
1
L. LESSIG, Code version 2.0, New York, Basic Books, 2006, 4. Available at
http://codev2.cc/download+remix/Lessig-Codev2.pdf, last accessed 3 February 2018.
2
S. NAKAMOTO, “Bitcoin: A Peer-to-Peer Electronic Cash System”, 2008. Available at
https://bitcoin.org/bitcoin.pdf, last accessed 6 March 2017.
3
‘Satoshi NAKAMOTO’ is a pseudonym adopted by the individual or group behind the original Bitcoin concept.
4
Blockchain technology is also referred to as distributed ledger technology (DLT) or as a decentralized ledger,
decentralized network. However, to be precise, blockchain is the first fully functional manifestation of DLT.
A DLT is a distributed database and blockchain is a DLT which is organized as a chain of blocks that contain
transactional data.
5
G.W. PETERS and E. PANAYI, “Understanding Modern Banking Ledgers Through Blockchain Technologies:
Future of Transaction Processing and Smart Contracts on the Internet of Money” in P. TASCA, T. ASTE et al.
(eds), Banking Beyond Banks and Money. New Economic Windows, Cham, Springer, 2016, 276.
11
blockchain past the peak of inflated expectations with a mainstream adoption horizon of five
to ten years.6
2. Large numbers of new articles and insights on the subject are published on a weekly
basis. However, this literature and the ideas described therein are characterized by the
frequent use of phrases expressing hesitation like “could be”, “might become”, “would
mean” … This dissertation has the ambition to identify whether and by what means the
distributed ledgers truly justify the use of the word “disruptive”, and how the law answers
to it. Because developments are generally first driven by market dynamics and later
sanctioned by the law, we believe that a legal approach particularly suitable in order to
identify blockchain applications that are more likely to actually cause substantial transitions
in the near to middle long future. This research focusses on how these transitions thrive in a
pre-existing legal framework and how both the innovation and the law mature together, in a
process based on dialogue, growing insight and experience. Indeed, new internet technology
like DLT does not simply emerge in a legal vacuum.7 The finding that its architecture relies
on libertarian ideas about abolishing the middleman (and other external forms of control)
does not affect this legal reality.
3. One domain of law where the role of the middleman is of particular relevance, is
financial law. The modern banking infrastructure is continuously under development
because of the consistently increasing volume, speed, global spread and complexity of
executed transactions. While monetary payments are executed through their own electronic
settlement architectures (e.g. the TARGET2 service operated by the European Central Bank
(ECB) or the Continued Linked Settlement (CLS) system for foreign exchange), the
dematerialization on securities allowed for increased liquidity on securities markets as well.8
Nowadays however, the degree of liquidity in this context is challenged by efficiency limits.
The dematerialized security assets are held and transferred through an international
settlement network characterized by its patchiness and its complex pyramidal structure. Its
operation is based on an intricate array of accounts and registers as well as an interplay of
6
X., “Gartner Hype Cycle for Emerging Technologies”, 2017. Available at
https://blogs.gartner.com/smarterwithgartner/files/2017/08/Emerging-Technology-Hype-Cycle-for-
2017_Infographic_R6A.jpg, last accessed 20 February 2018.
7
8
For the sake of clarity, we will be using the term “securities” interchangeably to refer to different kinds of
negotiable financial assets; only making distinctions where it is essential.
12
many intermediaries whose regulation and supervision are shared amongst a corresponding
number of jurisdictions. The introduction of DLT opened the door for a new massive step
in the ongoing development of the securities market architecture: after dematerialization
comes cryptonization.9 Analogous to the Bitcoin procedure, todays dematerialized securities
could be converted to assets represented by bits and bytes in a network that permits direct,
speedy and inexpensive transfers from one user to another. This network may substitute both
the registers and the role of the intermediaries thanks to its revolutionary technical design.
Such innovation poses challenges for – or, if you like: is challenged by – many aspects of
the existent, dense financial regulation. We believe that the abundant number of important
policy considerations renders financial law the ideal subject for legal research on how
disruptive DLT really is.
4. The next paragraphs provide a brief summary of how this research endeavour is
structured. First, we inevitably need to explain the crucial aspects of what blockchain is, and
how and why it works. This section on computer science is kept to the essential information
required for the reader to further understand the impact of the technology from a financial-
legal point of view. It also sets out some of the key technical categorizations of blockchain
varieties existing thus far. Indeed, the individual characteristics of the code at the basis of a
certain blockchain software application have important repercussions on its legal analysis.
Subsequently, this paper is organised in a way that follows the evolutionary steps of
blockchain itself. As discussed in paragraphs 16 to 19, three development stages or
‘generations’ of distributed ledger technology currently coexist: crypto-assets, smart
contracts and decentralized autonomous organizations. The reader will observe that the red
line running through these chapters exists in the question how these novelties may be useful
to contemporary financial transactions and actors; and, how such applications fit in the
existing legal framework. As the development of the third category depends on the gradually
maturing groundwork of the first two categories, we will highly limit our analysis to the
legal questions and challenges faced by crypto-assets and cryptosecurities. We will thereby
9
This relatively new term has been used in relation to cryptocurrencies (see for example O. J. MANDENG,
“Cryptocurrencies, monetary stability and regulation: Germany’s nineteenth century private banks of issue”,
LSE Institute of Global Affairs 2018. Available at http://www.lse.ac.uk/iga/assets/documents/research-and-
publications/LSE-IGA-WP-5-2018-Ousmene-Mandeng.pdf, last accessed 11 March 2018.), not so much in
relation to securities. Its conceptual meaning is equally applicable and forms a logical next step after the
dematerialization of various kinds of assets.
13
try to establish to which extent new blockchain-based financial procedures assimilate with
contemporary financial processes and traditional legal concepts; and whether a need exists
for specific regulations of a whole new nature: the domain Lex Cryptographia.10
10
For the origin of this term, see: A. WRIGHT and P. DE FILIPPI, “Decentralized Blockchain Technology and
the Rise of Lex Cryptographia”, 2015. Available at
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last accessed 7 March 2017.
14
TITLE 2. BLOCKCHAIN EXPLAINED
A. INFORMATION TECHNOLOGY
5. The outset of this research will explain the main characteristics of the distributed
ledger’s unique setup. We will commence with a simplified practical description of the
blockchain, followed a brief theoretical clarification. We will then lay out some important
classifications based on the technical particularities of the existing blockchain varieties.
Finally, this chapter will describe several interesting blockchain applications to further
illustrate the technology and its broad potential for practical use.
6. A blockchain network is in fact nothing but a collective, chronological and public

bookkeeping of information via the internet.11 This information can be anything – for
example, it may consist of numerous changes of possession regarding an asset that is
represented by a digital file of data. Since digital goods are essentially easy to copy, they are
particularly subject to the risk of double spending.12 Public bookkeeping of all transactions
in a database helps to mitigate this risk as it enables other actors to verify who most recently
acquired the digital asset. It is however not always easy to preserve the authenticity of the
information in the books, especially in situations where everybody can access them. Hence,
in a blockchain, all transactions are recorded in one shared ledger which is simultaneously
accessed, saved and updated locally by every user participating in the network, which makes
it a ‘distributed’ ledger. This distributed arrangement is what makes the blockchain novel in
comparison to a traditional database. The different participants in a certain network are
called ‘nodes’.13 Users who wish to take in upholding the network can become a node
themselves and start verifying and saving the entire bookkeeping.
11
This comparison is inspired by one of the earliest easy-to-understand explanatory videos on blockchain:
DUTCHCHAIN, “The real value of bitcoin and crypto currency technology - The Blockchain explained”, video,
2014. Available at https://www.youtube.com/watch?v=YIVAluSL9SU, last accessed 6 March 2017.
See also, a more recent video: WESTPAC BANKING, “Blockchain Demystified”, video, 2016. Available at
https://youtu.be/LdOcXXB48fI, last accessed 6 March 2017.
12
D. DRESCHER, Blockchain Basics: A Non-Technical Introduction in 25 Steps, New York, Springer
Science+Business Media, 2017, 51.
13
We will only consider users who participate in the validating DLT transactions to ‘run a node’. In our
definition, users who take part in these transactions do not necessarily need to be a node themselves.
15
7. In order to secure the integrity of the contents of the database, it is structured as a
chronological and interconnected chain of packages of data, the blockchain. The software
will only allow an irreversible addition of new information to the decentralized ledger if the
nodes reach consensus on the correctness of the new data. Control over validating a new
addition is essentially shared among nodes in a blockchain – it is not allocated to any central
authority. Consequentially, trust is factually distributed, decentralized in a distributed
ledger, preventing malevolent nodes or intruders from unilaterally manipulating its contents.
8. The particular procedure that governs the approval of a transaction by the network is
called a ‘consensus mechanism’. It does not necessarily rely on singular voting rights per
node – it depends on the individual software which is built to meet the needs and purpose of
a certain application. The influence or ‘weight’ of each node is calculated in accordance
with the standards of the specific blockchain. In situations where nodes do not know each
other, and therefore, trust is absent, this factor is often derived from calculations performed
by the nodes to the network: ‘one-CPU-one-vote’.14 Bitcoin is the prime example of a
blockchain based on this mechanism, called ‘proof-of-work’. Each node automatically
collects the new transactions that are to be confirmed in a package of data, a block, then
starts confirming this block. The Bitcoin nodes invest considerable amounts of
computational power (hence ‘one-CPU-one-vote’) to solve the mathematical ‘hash puzzles’
generated based on the data in the block and on the solution of the puzzle of the preceding
block of transactions.15 The complexity of the puzzle is adapted to the desirable completion
time; in case of Bitcoin, this is ten minutes for each block of transactions. It is difficult to
predict who will solve the puzzle first – this procedure is often compared to a lottery; but
once the block of transactions has been verified, it is easy for the rest of the network to verify
the correctness of the solution to its puzzle. The solution, a ‘hash’, is then attached to the
block. It functions as the block’s unique signature or fingerprint and it makes the information
tamper-proof: any change to the transactions would invalidate the correspondence between
the block and its signature.16 Even if it were possible to manipulate the complexly generated
signature instantaneously, the coherence with the preceding and consecutive block in the
14
S. NAKAMOTO, “Bitcoin: A Peer-to-Peer Electronic Cash System”, 2008, 3. Available at
15
This puzzle is called a ‘hash function’ and its output is a ‘hash’ – the data inside is literally hashed, mixed-
up until it is no longer intelligible.
16
16
chain would be damaged and the interference would be obvious.17 Moreover, the puzzle is
a cryptographic hash function, making it impossible to trace the input based on the output.18
The finalized new block of data is subsequently broadcasted to all other nodes in the
network. They can verify the correctness of the hash instantaneously and they will add the
new block to their local copy of the blockchain and the new, longest chain of transactions
will be deemed authoritative. A rewards-based system incentivizes the nodes of a blockchain
network to participate in supporting the calculations. In the Bitcoin context, this exists in
rewarding virtual currency to the participant solving the puzzle – he ‘mines’ a (fraction of
a) new coin. Besides the proof-of-work consensus mechanism described above, other voting
systems exist which may be more suitable to meet the requirements of different blockchain
appliances, e.g. proof-of-stake.19, 20
9. Note that, thanks to its unique structure, the distinctive features of a blockchain and
DLT systems in general are its:
- Distributed trust: new transactions are confirmed by the aggregate of all network
participants pursuant to the blockchain’s consensus mechanism. There is no need
for a trusted third party.
- Immutability: the data structure of interlinked blocks, each with a cryptographic
signature attached to it, makes it very difficult to secretly change the ledger’s
contents.
- Durability: all nodes hold a local copy of the blockchain – the number of back-
ups of the data is equal to or greater than the number of nodes in the network.
These features are essential to every blockchain or DLT, but the underlying architecture and
algorithms differ to meet the specific needs of every application’s individual purpose. The
next paragraphs will discuss the most important variations.
17
Ibid., 86.
18
For more details about cryptographic hash functions, see S. FRIEDL, “An Illustrated Guide to Cryptographic
Hashes”, 2005. Available at http://www.unixwiz.net/techtips/iguide-crypto-hashes.html, last accessed 15
April 2018.
19
The proof-of-work consensus mechanism now takes approximately. 7 minutes to validate a Bitcoin
transaction, although now it is coded to confirm a block in around 10 minutes. It takes substantially more time
to confirm more sizable packages of information through proof-of-work (like coded contracts, infra, 18).
20
Infra, 22.
17
B. PRINCIPAL CATEGORIZATIONS
10. In order to increase the reader’s insight in the diversity of blockchain designs, this
section highlights three prevailing structural blockchain classifications. The distinctions are
based on the access to the network, the transactional complexity and the used consensus
mechanism. We invite the reader to bear this typology in mind as it will facilitate
understanding the impact of a blockchain’s individual design on the legal analysis in the
subsequent chapters.
1. Network access
11. Distributed ledger networks can firstly be classified based on who is allowed to
access or participate in the network. These are two distinct questions – users can access the
contents of the ledger and submit new information to it while network participants are nodes,
responsible for validating new information thus executing the consensus mechanism.
Network access as a user is determined by the distinction between public and private
networks whereas access as a node depends on the permissioned or permissionless character
of the network.21, 22 In many blockchains, there is an overlap between both criteria because
users also have an interest in keeping the network up and running. The distinction between
public and private ledgers is however independent from the question whether a network is
permissionless or not.
12. Most of the well-known blockchains in existence are of the permissionless kind. This
means that any aspiring participant can become a node in the network straight away, without
the need for prior authorisation. As it is the case in the Bitcoin network, anyone with a
computer can start supporting the network as a node, by verifying new transactions, thus
start mining new coins.
21
Note that a singular interpretation of these discerning characteristics does not yet exist among authors. Many
sources use private, public and permissioned, permissionless in an intertwined manner (see for example J.D.
CAYTAS, “Developing Blockchain Real-Time Clearing and Settlement in the EU, U.S. and Globally”,
Columbia Journal of European Law: Preliminary Reference 2016, 6-7). We hereby establish an important
distinction between both concepts, which we consistently rely on throughout this research.
22
18
Some enthusiasts argue that this designed openness and impartiality is the only sensible
option for a true blockchain, in line with its anti-institutional, libertarian philosophy.23 The
ultimate goal is to exclude any sort of ‘corruption’ – either by distortion of what the parties
really wanted to achieve, or in the traditional meaning of the word – in transactions between
parties, caused by an intervening and authoritative actor.24 This view is grounded on the
conviction that restricted access to the network undermines the functioning of its specific
processes. The architecture depends on the combined computational power of the
participants, which is essential to ensure trust among them. The absence of a process
confirming the legitimacy of new participants also allows for a pseudonymous blockchain
environment.25 Users in the network need not reveal their true identity – the transactions
they engage in are attributed to their pseudonymous identity which they personally access
through their private key and which is consistently represented by their public key.26 The
actor saves the private key locally, electronically or even by writing it down, depending on
its complexity. The nodes and other users can only verify this actor’s pseudonymous identity
through the presence of the unique public key. Hence, the private key to an account is
cryptographically attached to its public key. The lack of a central authority contributes to
the state-remote character of open and permissionless networks. They are in fact self-
governing by design – the internal processes are deemed to make any intervention of a
central authority unnecessary and undesirable.27 It is more challenging to regulate a network
when the network fundamentally governs itself.
13. Contrary to the above, permissioned blockchains only include trusted party nodes,
who were verified and admitted to the network by a preselected central authority. The
European Securities and Markets Association (ESMA) expressed its preference for these
closed systems in the context of securities markets for reasons of governance, scale and risk
23
An example is the point of view presented by T. MARCKX (co-founder TheLedger.be) at “Blockchain:
disrupting markets & governments”, Ghent, 25 October 2017.
24
E. TJONG TJIN TAI, “Smart contracts en het recht”, NJB 2017, 92(3), 179.
25
P. HACKER and C. THOMALE, “Crypto-Securities Regulation: ICOs, Token Sales and Cryptocurrencies under
EU Financial Law”, 2017, 31. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3075820, last
accessed 12 February 2018.
26
A user signs a transaction with his individual public key, which is only possible by using the correct private
key. Nodes can verify whether a transaction is signed with a matching pair of keys. Furthermore, a user can
create different addresses to transact from, all of them connected to a single pair of cryptographic keys.
27
P. PAECH, “The Governance of Blockchain Financial Networks”, Mod. L. Rev. 2017, 1084.
19
of illicit activities.28 Likewise, the Belgian bank KBC only considers a permissioned setup
in the development of the bank’s very own blockchain-based solution.29 Indeed, both
varieties of the technology serve entirely different purposes and accordingly, they are
characterized by a different spirit. The networks developed by for-profit market players and
the fintech start-ups they collaborate with mean to facilitate existing processes and to weed
out inefficient business practices among a defined group market participants among whom
a certain level of trust already exists. 30, 31 These ledgers are purpose-built in accordance with
the specific needs of the commissioning institutions. The practice of whitelisting selected
actors to a permissioned blockchain platform may for instance be helpful to administer a
financial institution’s know your customer policy.32
14. Clearly, there is an important split between the open and closed approaches:
institutions whose business currently exists in playing an intermediary role do not act in
pursuit of the libertarian philosophy of permissionless blockchains. Both varieties of the
technology serve entirely different needs and a permissioned design is more relevant to
facilitate trusted information exchanges among established financial market participants
with a real-word reputation.33 Nevertheless, exceptions exist: Ripple (in its standard version)
is one of the leading ‘altcoins’ (i.e. alternatives to Bitcoin and open to a broad audience) but
it is open to discussion whether or not it can be considered a real cryptocurrency.34 Ripple
is public – everybody is eligible to obtain an individual private key as a user and thus to take
part in transactions on the network – but it is built on a blockchain that is de facto
permissioned.35 Consequently, transaction speed is higher than on the permissionless Bitcoin
28
ESMA, “The distributed ledger technology applied to securities markets”, report, 2017, 4. Available at
https://www.esma.europa.eu/sites/default/files/library/dlt_report_-_esma50-1121423017-285.pdf, last
accessed 7 April 2018.
29
C. MUYLDERMANS (counsel regulatory affairs KBC) at the “Becon Blockchain Conference”, Brussels, 8
March 2017.
30
P. PAECH, “The Governance of Blockchain Financial Networks”, Mod. L. Rev. 2017, 1091. See also: R.
VERHAERT, Blockchain, de overheid aan of in de ketting?, Brugge, Vanden Broele, 2016.
31
D. MILLS, K. WANG et al., “Distributed ledger technology in payments, clearing, and settlement”, Finance
and Economics Discussion Series 2016, 12. Available at
https://www.federalreserve.gov/econresdata/feds/2016/files/2016095pap.pdf, last accessed 17 April 2018.
32
T. SWANSON, “Consensus-as-a-service: a brief report on the emergence of permissioned, distributed ledger
systems”, 2015, 5. Available at http://www.ofnumbers.com/wp-content/uploads/2015/04/Permissioned-
distributed-ledgers.pdf, last accessed 7 April 2018.
33
Ibid., 44.
34
Ripple, https://ripple.com/, last accessed 7 April 2018.
35
The ECB even categorizes Ripple as a “centralised virtual currency scheme”, see ECB, “Virtual currency
schemes – a further analysis”, report, 2015, 15. Available at
https://www.ecb.europa.eu/pub/pdf/other/virtualcurrencyschemesen.pdf, last accessed 15 April 2018.
20
blockchain.36 It has its own cryptocurrency, but its network is also open for use to transfer
any other virtual or (non-native) traditional currency through a specific system of gateways.
15. For the sake of clarity, the following scheme summarizes the definitions we adhere
to when we label a certain DLT network private or public, permissioned or permissionless.
36
D. SCHWARTZ, N. YOUNGS and A. BRITTO, “The Ripple Protocol Consensus Algorithm”, Ripple Labs Inc
2014, 8. Available at https://ripple.com/files/ripple_consensus_whitepaper.pdf, last accessed 15 April 2018.
21
Permissionless Permissioned
Public - Anyone can run a node and thus - A designated authority (or: the
participate in executing the consensus collectivity of the nodes) decides on who
algorithm. Incentive to do so exists in is allowed to run a node as a trusted
financial benefit. party. Incentive may exist in a financial
benefit or in supporting the common
interest.
- Anyone is eligible to create an - Anyone is eligible to create an
individual private and public key pair individual private and public key pair
and an address, and thus is accepted to and an address, and thus is accepted to
execute transactions in the network; as execute transactions in the network; as
well as to access data in the network. well as to access data in the network.
- e.g. Bitcoin. - e.g. Ripple.
Private - Anyone can run a node and thus - A designated authority decides on who
participate in executing the consensus runs a node as a trusted party (likely: the
algorithm. Incentive to do so exists in a actors establishing the DLT). Incentive
financial benefit. exists in supporting a common interest.
- A designated authority decides on - A designated authority decides on
granting a private and public key pair granting a private and public key pair
and an individual address to applicants and an individual address to applicants
who wish to transact in the network and who wish to transact in the network and
access its data. access its data. (likely: the actors
establishing the DLT).
- e.g. a fintech bank that entirely - e.g. a consortium of banks that agrees
distributes trust by giving nodes a to establish a DLT network to settle
financial incentive (e.g. cryptosecurities payments among each other. All
issued by the bank itself) and that offers participating banks run a node in the
hyper-secure, virtually risk-free savings network.
accounts to its paying clients. - e.g. 2: Corda.37
37
R3 Corda is specifically designed to meet the needs of the financial industry, https://docs.corda.net/key-
concepts-ecosystem.html, last accessed 5 May 2018.
22
2. Stages of development
16. Another important typology concerns the different evolutionary steps in the
complexity of the DLT and its capacities. Most authors generally discern two generations
of the technology: crypto-assets and smart contracts. A third category, decentralized
autonomous organizations (DAOs38), is often seen as a combination of smart contracts.
However, as Jane ZHANG righteously suggested, DAOs are fundamentally more far-reaching
and thus deserve to be ranked as a third stage of development.39
17. The first generation does not need an extensive introduction. Basic virtual currencies
like Bitcoin belong to the first materialization of blockchain and DLT. These networks
facilitate almost instantaneous and cost-efficient remittance of digital or crypto-assets.40 All
reassignments from one node to another are immutably logged to the database. This
information makes it possible to consistently identify the public key belonging to the final
holder of the rights regarding the value or the crypto-asset.41 A pure first-generation
blockchain thus only contains data on transfers, thereby identifying the most recent asset
holder and banning the double spending risk.42 In a more advanced configuration, this
transaction data can be represented in a complete list of individual accounts.43
Further distinction can be made between native and non-native crypto-assets: while native
crypto-assets only virtually exist the blockchain register’s data, the non-native entries
represent a separate asset in the world outside the ledger. These assets can range from
traditional currencies to intellectual property rights or property titles. Bitcoin is an example
of a native blockchain – the coins only exist by virtue of the data on the distributed ledger.
38
Not to be confused with “The” DAO, that is, a specific DAO-application which was set up as an investment
vehicle. Due to a loophole in the blockchain’s coding, one node was able to drain 70 million dollars of value
from The DAO. The consecutive “hard fork” to reverse these malevolent transactions caused major polemics
among the platform’s users. See also E. TJONG TJIN TAI, “Smart contracts en het recht”, NJB 2017, 92(3), 180.
English commentary on the event: P. HACKER and C. THOMALE, “Crypto-Securities Regulation: ICOs, Token
Sales and Cryptocurrencies under EU Financial Law”, 2017, 31. Available at
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3075820, last accessed 12 February 2018.
39
I concur with the viewpoint presented by J. ZHANG (founder and CEO at Shellpay) at “Becon Blockchain
Conference”, Brussels, 8 March 2017.
40
Cost-efficient to the user – some consensus algorithms are disproportionally energy-consuming. Infra, 19.
41
42
Supra, 6.
43
Ethereum, for instance. E. TJONG TJIN TAI, “Smart contracts en het recht”, NJB 2017, 92(3), 178.
23
18. More developed applications that go far beyond this first cryptocurrency function
have appeared more recently. A second-generation blockchain does not just serve as a
decentralized logbook of transfers, it also supports the performance of encoded computation
on the network platform.44 A widely known example is Ethereum project, which serves as a
smart contract programming environment.45 This means that the transactions regarding the
virtual currency can be tailored to the needs of the user: two nodes can agree that a
remittance of value between them will only occur contingent on the fulfilment of certain
pre-determined conditions. These platforms feature their own extended script language and
they make it possible to construct self-enforcing programs that are computed and executed
on the blockchain, so-called smart contracts.46 The purpose of the distributed ledger remains
the same as with first-generation varieties. Registering the source code of the smart contract
to the blockchain makes its clauses immutable and their execution, in principle, irreversible.
Thus, the self-executing program runs without anyone having the power to alter or to revoke
it. Parties may however choose to deliberately design for flexibility and adaptability by
enabling the processing of external inputs to the contract, or even by encoding a self-destruct
function.47
Potential smart contract applications are wide-reaching and go beyond traditional two-party-
contracts. They may for instance be used to construct self-enforcing voting and dividend
rights for the holders of ‘digital shares’ of a company.48 Thanks to the immutability of the
blockchain, nodes can always retrieve and recycle completed smart contracts that already
proved their worth. They can be reused as building blocks and combined together to encode
new and even more sophisticated smart contracts.49
44
45
Ethereum, https://www.ethereum.org/, last accessed 7 April 2018.
46
N. SZABO, “Smart Contracts”, 1994. Available at
http://www.fon.hum.uva.nl/rob/Courses/InformationInSpeech/CDROM/Literature/LOTwinterschool2006/sz
abo.best.vwh.net/smart.contracts.html, last accessed 24 April 2018.
47
G. JACCARD, “Smart Contracts and the Role of Law”, Justletter IT 2017, 23, 7; and infra, 68 and 77-78.
48
See for example stock issuance by Overstock.com as described in C. VAN DER ELST and A. LAFARRE,
“Blockchain and the 21st Century Annual General Meeting”, European Company Law Journal 2017, 4, 173.
49
24
19. Further sophistication and interconnection of smart contracts may eventually
develop a third and currently final stage of blockchain implementations: DAOs. A variety
of coded contracts can be bound together in a coordinated manner in order to create a
decentralized organization by itself.50 This concept corresponds with JENSEN’s and
MECKLING’s agency theory of the firm, stating that corporations essentially exist in the
accumulation of their internal and external contractual and other relationships.51
Accordingly, a DAO is defined by and operates pursuant to the smart contracts it consists
of. Such arrangement permits to do business through the set of coordinated smart contracts,
without incorporating into a traditional business entity without being owned or controlled
by a single person, yet these DAOs can theoretically operate in the market.52 In the context
of the emerging Internet of Things, people may even become redundant and a business could
just be made up of the DAO and the machines it administers.
Yet, these futuristic perspectives remain unfeasible for the time being. Both legal and
technical impediments will need to be tackled before fully independent DAOs can possibly
materialize. The necessary tools for a DAO to directly participate in the traditional business
landscape are currently lacking as it is unclear how these entities would be recognized by
law and how they can be governed; for instance, it is necessary to appoint identifiable
‘nearest persons’ to make it possible for the DAO to be held liable.53 We will not go into
further details on this topic because we believe that these questions can be traced back to
analogous problems in relation to crypto-assets and smart contracts – DAOs exist in
intertwined combinations of these building blocks and the governance of DAOs depends on
the governance of first and second generation blockchains. A problem of an entirely
different nature exists in the scalability issues connected to computation-intensive consensus
mechanisms, and therefore mainly faced by permissionless blockchain setups. These
blockchain structures ensure trust among nodes through processes that require heavy and
50
A. WRIGHT and P. DE FILIPPI, “Decentralized Blockchain Technology and the Rise of Lex Cryptographia”,
2015, 15. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last accessed 7 March
2017.
51
M. JENSEN and W. MECKLING, “Theory of the Firm: Managerial Behavior, Agency Costs and Ownership
Structure”, Journal of Financial economics, 1976, 3(4), 311.
52
53
S. BLEMUS, “Law and Blockchain: a legal perspective on current regulatory trends worldwide”, RTDF 2017,
4, 15; and, A. WRIGHT and P. DE FILIPPI, “Decentralized Blockchain Technology and the Rise of Lex
Cryptographia”, 2015, 55. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last
accessed 7 March 2017.
25
competitive computation on all network nodes.54 This process is very energy-consuming
which prevents mass adoption. Recent reports even estimate that the Bitcoin network
exceeds Ireland’s total energy consumption per year.55 Some authors are optimistic and
expect future advances in computer processing speed to deal with scalability bottlenecks.56
We believe however that the solution does not exist in stronger computers; instead,
ingenious new consensus mechanisms and network architectures will empower widespread
usage of blockchain networks.
3. Consensus mechanisms
20. The aforementioned proof-of-work procedure undoubtedly is the most recognized

consensus mechanism, a logical consequence of Satoshi NAKAMOTO applying it to the
original Bitcoin concept.57 It also is the mechanism responsible for the scalability issues. We
earlier explained how consensus is the instrument to create trust among participants in the
network. Consequently, the recommended standard for this mechanism depends on the
relationship between the nodes. As the Belgian researcher Kristof VERSLYPE labels it, trust
is decentralized in a permissioned network whereas it is distributed in a permissionless
network.58 The principle of adopting a voting mechanism to share the responsibility of
validating a block of new data is identical but its scale is different. If anyone can freely join
the operation of the network, trust is indeed distributed to anyone who asks for it – who
guarantees that nobody will register a million accounts to seize control? To the contrary,
when authority is merely decentralized among a closed group of individually identified and
approved participants, it is easier to assert that at least half of them will remain diligently
updating the chain in the common interest.
54
55
A. HERN, “Bitcoin mining consumers more electricity a year than Ireland”, The Guardian, 27 November
2017. Available at https://www.theguardian.com/technology/2017/nov/27/bitcoin-mining-consumes-
electricity-ireland, last accessed 8 March 2018.
56
E. KARAINDROU, “Distributed ledger technology and the future of payment services”, 2017, 26. Available
at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3078523, last accessed 10 March 2018.
57
Supra, 1; and for an earlier comment on proof-of-work: supra, 8. For good understanding: even though
proof-of-work is built on the cryptographic process of determining a block’s hash, the consensus mechanism
is independent from the blockchain’s immutable data structure. The consensus algorithm is the
operationalization of the distributed trust to validate new blocks.
58
As pointed out by K. VERSLYPE during his lecture “Bitcoin, Blockchain & Smart Contracts – Inleiding voor
juristen”, Ghent, 26 October 2017.
26
21. Complicated consensus procedures based on rewarding nodes for actively validating
blocks in an honest manner, like the ones based on computational power, are appropriate
only if trust is distributed. In a closed, permissioned blockchain environment with identified
participants, it may be more sensible to implement a one-node-one-vote standard. It has no
impact on the immutable character of the blockchain, but it removes the competition to
outperform one another in the confirmation process. Hence, there is no need to reward the
nodes validating the transaction – the blockchain will work fine as long as half of the
participants remains acting in good faith. Absent the need for rewards, the blockchain can
in certain circumstances exist without having its own cryptocurrency, token or other carrier
of value.
However, this approach is insufficient if the purpose of the distributed ledger requires a
permissionless setup. A more sophisticated consensus protocol is then essential to protect it
against nodes trying to add erroneous transactions to the chain. Such protocol will basically
demand some sort of effort from the nodes to prove that they belong to the network of
trustworthy participants backing the right information. This mechanism raises the barrier for
an intruder to distort the update of the blockchain – the intruder will have to exceed fifty
percent of the total effort contributed by all network participants in order to succeed. Besides
the effective but inefficient proof-of-work, new solutions are continuously being explored
and built. Below we will discuss both an increasingly common and a particularly interesting
alternative algorithm.
22. Proof-of-stake an alternative algorithm and its importance is growing since the
leading permissionless second generation blockchain platform Ethereum adopted a similar
consensus mechanism named Casper. 59, 60
Nodes are expected to ‘stake’ funds while
confirming new transactions, at the risk of losing their stake if the confirmed transaction
turns out to be inaccurate. The higher the value of a node’s stake, the higher the chances that
this particular node will finally confirm the transaction and generate a new block. He will
then be rewarded with a fixed commission fee. The cost to acquire an influential stake of
the ledger’s value makes it disproportionally expensive to attack its contents, parallel to how
59
Casper, https://blog.ethereum.org/2015/08/01/introducing-casper-friendly-ghost/, last accessed 8 March
2018.
60
An example of a successful first-generation ledger entirely relying on proof-of-stake is NXT, see
https://nxtwiki.org/wiki/Nxt_Wiki, last accessed 8 March 2018.
27
an attack on a proof-of-work ledger would require a disproportionate investment in
computative power – indeed, both mechanisms rely on a rational cost-benefit analysis. A
cheating participant will be deterred from compromising and jeopardizing the network he
himself invested so much in.
While proof-of-stake is not as energy-costly as proof-of-work, it also features some major

drawbacks. Since more wealth means more votes, hoarding is encouraged and this tendency
risks to ultimately cause the centralization of the blockchain’s value. Major stakeholders
accumulating influence over time may eventually obtain control over the network,
undermining the distributed trust among nodes.61 The formation of such monopoly position
inherently defeats the basic purpose of a distributed ledger. Proof-of-work ledgers risk
facing the same problem in the long run because certain pools are gradually acquiring a
majority of CPU power.
23. An interesting approach to deal with these issues is found in the Obelisk algorithm.
It is the consensus mechanism supporting the advanced Skycoin blockchain and it is built on
the web-of-trust concept. Nodes connect themselves to other nodes they personally trust,
and the density of a node’s interconnections determines its voting power in the validation of
new transactions. Each participant’s activity and influence on the ledger is transparent which
enables the community to react to malicious nodes. The creators describe this process as a
continuous audit by the Skycoin community.62
24. The reader should now better understand the basics of how the blockchain works.
An overview of DLT categorizations brings us to the conclusion that there is no such thing
as ‘the blockchain’ – the specific features of every DLT application are determined by its
unique underlying architecture. Importantly, even though the invention of Bitcoin and the
technology concept it is built on are often assimilated, it is necessary to consider that the
Bitcoin use case is a very specific emanation of DLT; and that blockchain networks with an
entirely different setup may support many other applications with each a very different,
individual purpose and spirit. We will briefly introduce some of these other use cases for
61
L. LEE, “New Kids on the Blockchain: How Bitcoin’s Technology Could Reinvent the Stock Market”,
Hastings Business Law Journal 2016, 12(2), 108.
62
Obelisk, https://www.skycoin.net/blog/statement/obelisk-the-skycoin-consensus-algorithm/, last accessed
11 March 2018.
28
illustration purposes before commencing a more detailed legal analysis of financial DLT
applications.
C. DLT APPLICATIONS
25. The following three promising use cases mean to further demonstrate what
blockchain technology is, among others, meant for. We mainly aim to establish that possible
applications go far beyond cryptocurrencies and monetary matters in general – we invite the
reader to bear in mind the diversity of this technology’s potential, even though the scope of
this thesis is limited to financial law.
Land registries – The Swedish land registry authority, the Lantmäteriet, started an
experiment in June 2016, aiming to put real estate transaction such as transfers of property
and deeds of mortgage on a blockchain.63 The project now entered its second phase where a
testbed-platform is being built.64 By developing a land registry in a decentralized ledger, the
Swedes intend to reduce paperwork, costs and fraud, and to speed up the transaction
processes.
Documenting a supply chain – An interesting private initiative is Modum, a Switzerland-

based startup that makes supply chain processes more efficient by combining blockchain
technology with another IT-revolution, the Internet of Things (IoT).65, 66
The company
currently operates an application in pharma logistics. T-Mining is a Belgian example
implementing blockchain technology to improve the efficiency of international transport and
logistics.67 In the future, this startup believes that a decentralized network allows for a safer
and faster online administration of documents related to international shipments such as bills
of lading and letters of credit.
63
G. CHAVEZ-DREYFUSS, “Sweden tests blockchain technology for registry”, Reuters, 16 June 2016. Available
at http://www.reuters.com/article/us-sweden-blockchain-idUSKCN0Z22KV, last accessed 10 March 2017.
64
LANTMÄTERIET et al., “The Land Registry in the blockchain – testbed”, report, 2017. Available at
https://chromaway.com/papers/Blockchain_Landregistry_Report_2017.pdf, last accessed 9 April 2017.
65
Modum, https://modum.io/about-us/, last accessed 9 April 2017.
66
IoT refers to the networked interconnection of everyday objects, which are often equipped with ubiquitous
intelligence.
67
N. WAUTERS (co -founder and CEO T-Mining) at “Becon Blockchain Conference”, Brussels, 8 March 2017.
29
Copyrights – A decentralized network permits to exclude central intermediation in a variety
of situations where the role of the middleman is crucial until today. The copyright sector is
an example of such a densely intermediated environment. Jaak is a second generation
blockchain project built on Ethereum and by storing rights and smart contracts to the
blockchain, it aims to facilitate and automate copyright management and royalty payments
for using the content.68
26. The virtually unlimited variety of blockchain possibilities demands for a firm
demarcation of the research. For the reasons discussed above and in view of our specific
area of interest, we will analyze the legal issues possibly arising from the two fundamental
generations of blockchain applications for financial purposes, so-called fintechs. 69 We will
thereby establish whether traditional legal concepts and methods of regulation are adequate
to govern these fintech innovations and to ensure their reliability in the context of the
contemporary financial industry.
68
Jaak, see http://musically.com/2017/02/20/blockchain-startup-jaak-unveils-m%CE%BEta-plans-pilots/,
last accessed 7 March 2018.
69
Supra, 3.
30
TITLE 3. CRYTPO-ASSETS
CHAPTER 1. CRYPTOCURRENCIES
27. Financial crypto-assets consist of both cryptocurrencies and cryptosecurities. We

will treat these subjects separately because of the distinct legal questions raised. Note
however that practice shows that cryptocurrencies are often used for speculative objectives,
as if they were investment assets.70 David YERMACK compares the behavior of average
Bitcoin buyers to the speculative investment pattern targeting internet stocks in the late
1990s.71 In the light of the first part of our research however, we will draw the line in
accordance with their initial apparent purpose – Satoshi NAKAMOTO, for instance, referred
to his creation as ‘electronic cash’ from the very beginning.72, 73 While the electronic cash
component serves holders to pay for external goods and services, the utility component of
cryptosecurities or ‘tokens’ is limited to accessing an internal function provided by their
issuer.74
70
S. ATHEY, I. PARASHKEVOV, V. SARUKKAI and J. XIA, “Bitcoin Pricing, Adoption, and Usage: Theory and
Evidence”, working paper, 2016, 30. Available at https://www.gsb.stanford.edu/faculty-research/working-
papers/bitcoin-pricing-adoption-usage-theory-evidence, last accessed 14 April 2018.
71
D. YERMACK, “Is Bitcoin a real currency? An economic appraisal”, working paper, National Bureau of
Economic Research Working Paper Series 2013, 2. Available at http://www.nber.org/papers/w19747, last
72
S. NAKAMOTO, “Bitcoin: A Peer-to-Peer Electronic Cash System”, 2008. Available at
73
See infra, 49, for a discussion of distinctive characteristics inherent to cryptosecurities.
74
P. HACKER and C. THOMALE, “Crypto-Securities Regulation: ICOs, Token Sales and Cryptocurrencies under
EU Financial Law”, 2017, 12. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3075820, last
accessed 12 February 2018.
31
A. HIGHLY DECENTRALIZED AND PUBLIC75 – electronic
cash
28. Some authors state that cryptocurrencies compete with traditional central bank-
issued fiat money.76, 77
In this regard, note that a performant cryptocurrency must be
constructed as a public blockchain. Assets on a private blockchain can only be transferred
to a selected group of users and as we will discuss below, a currency is fundamentally
characterized by a high degree of versatility and acceptance. The following paragraphs will
discuss to what extent a cryptocurrency indeed succeeds at fulfilling the functions and the
distinguishing requirements of a currency – consecutively from an economical and from a
legal point of view. We will then analyze how the underlying highly decentralized
blockchain application relates to the EU legal framework regarding payment services.
1. Qualification by an economist
29. From an economical perspective, money is defined by the following functions it

fulfils:
- a measure of value and a unit of account;

- a medium of exchange; and,
- a store of value or wealth.78
Looking at most established virtual currencies at the time of writing, their capacity to meet
these purposes is – at least – highly debatable.79 The main explanation for this failure lies in
their extreme volatility; which has been a consistent characteristic for all cryptocurrencies
since their surge in 2009. The Bitcoin volatility index varied between 4 and 10% over the
75
While many cryptocurrencies are permissionless, some (e.g. Ripple) have a particular consensus mechanism
which is not carried out by all users but by a select group of nodes with a uniquely operational task. These
permissioned varieties are still highly decentralized due to the unlimited access for users (i.e. public).
76
H. NABILOU and A. PRÜM, “Ignorance, Debt and Cryptocurrencies: The Old and the New in the Law and
Economics of Concurrent Currencies”, 2018, 5. Available at
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3121918, last accessed 11 April 2018.
77
ECB, “Virtual currency schemes – a further analysis”, report, 2015, 21. Available at
78
C. PROCTOR, Mann on the Legal aspects of money, Oxford, Oxford University Press, 2012, 10.
79
As also presented by J. RICHELLE (Lawsquare) at “Studiedagen Instituut Financieel Recht: Fintech”,
Brussels, 16 January 2018.
32
past eight years, reaching peaks of as much as 15%.80 These fluctuations are mainly caused
by the absence of a credible obligor behind the cryptocurrency (a central bank authority, for
instance) combined with the non-existent intrinsic value of the private keys to the blockchain
wallets. The fixed supply schedule intrinsic to the rewards-based mining procedure of most
cryptocurrencies today adds to the price volatility.81 As a consequence, cryptocurrencies are
currently very unwieldy as a measure of value or as a means of exchange – their value
compared to other currencies swings up and down every day and it would be very costly for
traders to bear or mitigate these risks. Since consumers always need to double check the
currency’s actual value in order to understand prices, it is also very confusing in the context
of payments.82 The extreme volatility furthermore prevents the cryptocurrency’s capacity to
act as a store of value. For comparison purposes, gold serves as a reasonable alternative way
to store value. The 30-days volatility index of this precious metal ranged between 0.50 and
1.50% over the past eight years, with an occasional peak of 2.50% in 2011 – far less than
Bitcoin’s variation between 4 and 10%.83 Holding cryptocurrencies, for even a very brief
period of time, is very risky, which collides with the function of storing value.84
Additionally, it is very difficult to protect the key to a person’s funds of cryptocurrencies
against operational risks like loss and theft.85 The key is in fact the ‘weak link’ of a
cryptocurrency wallet: it is nothing more than a numerical code and the owner has the
responsibility to keep and protect it – this includes the need for a very robust IT security
when accessing the cryptocurrency account. Absent a legal claim on an intermediary, one
80
Based on the 30-days BTC-USD volatility index; a measure of the Bitcoin value variation compared to US
dollars. Available at https://www.buybitcoinworldwide.com/nl/volatiliteits-index/, last accessed 13 April
2018. Note that, while the example of Bitcoin is by far the most important cryptocurrency in terms of market
capitalization, the other top five varieties (Ethereum, Ripple, Bitcoin Cash and Litecoin) face the same problem
of extreme volatility. See also http://cryx.io/ (last accessed 13 April 2018) and
https://www.sifrdata.com/cryptocurrency-volatility-index/ (last accessed 13 April 2018) on their volatility
indices.
81
H. NABILOU and A. PRÜM, “Ignorance, Debt and Cryptocurrencies: The Old and the New in the Law and
Economics of Concurrent Currencies”, 2018, 26. Available at
82
J.L. VERHELST, “Zijn cryptomunten munten? Een analyse van Bitcoin” in M.E. STORME and F. HELSEN
(eds), Innovatie en disruptie in het economisch recht, Antwerpen, Intersentia, 2017, 55-56 & 67; see also: D.
YERMACK, “Is Bitcoin a real currency? An economic appraisal”, working paper, National Bureau of Economic
Research Working Paper Series 2013, 11. Available at http://www.nber.org/papers/w19747, last accessed 14
March 2018.
83
Based on the 30-days Gold-USD volatility index. Available at
https://www.buybitcoinworldwide.com/nl/volatiliteits-index/, last accessed 13 April 2018.
84
J.L. VERHELST, “Zijn cryptomunten munten? Een analyse van Bitcoin” in M.E. STORME and F. HELSEN
(eds), Innovatie en disruptie in het economisch recht, Antwerpen, Intersentia, 2017, 68.
85
33
cannot rely on a bank to assume the security risk of a digital wallet analogous how bank
accounts for traditional currencies are held.86
Apart from the volatility issue affecting the three functions of money and the practical
problem with the required safety of a store of value, two more impediments of a practical
nature exist that restrict the functions of means of exchange and unit of account. In the first
place, cryptocurrencies do not have a single straightforward value. They can be bought and
sold from one of the numerous different online exchange platforms called ‘markets’ and
each of them applies an individual ‘current market price’. The difference between the
highest and lowest individual prices quoted by the active Bitcoin platforms with the highest
trading volumes usually ranges from 3 to 10%.87 This disparity makes it almost impossible
to find a single reference point. Instead, the value of most cryptocurrencies is usually
assessed through opaque 24-hours price aggregations that fail to indicate the true cost of
procuring or selling the currency at the present time. The second and last cumbersome factor
pertains to the relatively high cost of a single unit of the currently established
cryptocurrencies compared to the price of most ordinary products and services. Retail prices
quoted in two or even four or more (Bitcoin) decimals are simply puzzling. The mathematics
are transparent, but consumers tend to fail finding reference points due to the presence of
many leading zeros.88
2. Qualification by a lawyer
30. The economic reality discussed above seems to coincide with the legal reality, both
in literature and in the dominant interpretation by the EU lawmakers. There is no universal
legal definition of money, nor is there a pragmatic functional approach like the one applied
by economists. MANN established that different ‘theories of money’ exist that are used to
demark the legal concept.89 The most important ones are the ‘State theory of money’, which
bases the quality of money on the role of the State establishing it in its monetary system;
86
D. YERMACK, “Is Bitcoin a real currency? An economic appraisal”, working paper, National Bureau of
Economic Research Working Paper Series 2013, 14. Available at http://www.nber.org/papers/w19747, last
87
Ibid., 12. See https://www.cryptocompare.com/coins/btc/markets/USD (last accessed 14 April 2018) to
consult current Bitcoin prices quoted by the different markets. Given its volatility, it is important to only
consider the markets involved in recent transactions (cf. left column).
88
Ibid., 13.
89
C. PROCTOR, Mann on the Legal aspects of money, Oxford, Oxford University Press, 2012, 13.
34
and the ‘Societary theory of money’, where it is the attitude of society rather than the
authority of the state that defines money through its public acceptance and the confidence
of the people.90 The ‘Institutional theory of money’ is more recent and inspired by the
creation of the euro. It focuses the role of the central bank that controls the money rather
than on the state acting through legal tender laws.91 Cryptocurrencies, to the contrary, are
essentially built on a principle of disintermediation. They have no ties with states or central
banks; nor are they currently publicly accepted by any society. Consequentially,
cryptocurrencies do not fit in any of the legal definitions stipulated by MANN.
Two reports by the ECB are consistent with our foregoing analysis founded on MANN’s legal
definitions. In 2012, the ECB distinguished cryptocurrencies from electronic money.92 The
latter is defined in the Electronic Money Directive, as a monetary value represented by a
claim on the issuer which is issued on receipt of funds for the purpose of making payment
transactions and accepted by entities other than the electronic money issuer.93 The ECB
points out that cryptocurrencies are only accepted within the issuing virtual community – it
is not a claim on the issuer.94 In addition, cryptocurrency is not always issued on receipt of
funds since varieties built on reward-based consensus mechanisms also assign value to
mining nodes.95 Later, in its 2015 report, the ECB established that an interpretation of
cryptocurrencies may not even include the word ‘money’ because of the lacking high degree
of liquidity and the minimal current level of acceptance as a means of payment.96 These
considerations lead to ECB’s refined definition of cryptocurrencies as a digital
representation of value, not issued by a central bank, credit institution or e-money institution,
which, in some circumstances, can be used as an alternative to money.97 The European Court
of Justice (ECJ) however seemed to adhere to a diverging approach in Skatteverket v David
90
Ibid., 15-25.
91
Ibid., 27.
92
The ECB refers to cryptocurrencies as ‘decentralized bi-directional virtual currency schemes’.
93
Article 2 (2) Directive 2009/110/EC of the European Parliament and of the Council of 16 September 2009
on the taking up, pursuit and prudential supervision of the business of electronic money institutions.
94
ECB, “Virtual currency schemes”, report, 2012, 16. Available at
http://www.ecb.europa.eu/pub/pdf/other/virtualcurrencyschemes201210en.pdf, last accessed 15 April 2018.
95
Supra, 8.
96
97
Ibid., 25. See also: P. ATHANASSIOU, “Impact of digital innovation on the processing of electronic payments
and contracting: an overview of legal risks”, ECB Legal Working Paper Series 2017, 19. Available at
https://www.ecb.europa.eu/pub/pdf/scplps/ecb.lwp16.en.pdf?344b9327fec917bd7a8fd70864a94f6e, last
35
Hedqvist.98 This case involved a tax dispute between the Swedish Tax Authority and a
Swedish national exchanging traditional currency for Bitcoin and the other way around. The
national court was uncertain whether such exchange activities fall within the scope of the
Common VAT Directive exemption for ‘transactions of negotiation, concerning currency,
bank notes and coins used as legal tender…’.99 The ECJ considered that although Bitcoin is
no legal tender, it is a means of payment accepted between parties. Moreover, the court finds
that in the underlying relations, Bitcoin has no other purpose than to be a means of
payment.100 The ECJ decided not to classify the cryptocurrency as anything that is statutorily
recognized, and the court seemingly intentionally avoided to call it money. Instead it
classifies cryptocurrencies under a narrower category of liquid payment instruments like
electronic and bank account money. 101 This does not mean that Bitcoin has this same status.
While the euro as a currency takes the form of – besides scriptural money, banknotes and
coins – widely accepted electronic money, cryptocurrencies use their own denominations,
which are currently not backed by the law or declared legal tender in any state.102, 103 Hence,
it appears that the ECJ chose to solve the tax dispute pragmatically without leaving a
decisive impact on the cryptocurrency typology.104 Lastly, the ECB notes that even though
cryptocurrencies currently cannot be labeled money, it remains possible that a different
analysis applies to safer, more efficient and more reliable cryptocurrencies that may be
developed in the future.105
3. Regulation?
31. Despite the fact that the existent cryptocurrencies are incompatible with both the
economical and legal demarcations of the money concept, a number of transactions still
98
Judgment of 22 October 2015, Skatteverket v David Hedqvist, C-264/14, EU:C:2015:718, paragraphs 50-
57.
99
Article 135 (1) (e) Council Directive 2006/112/EC of 28 November 2006 on the common system of value
added tax.
100
Hedqvist, op. cit., paragraph 52.
101
P. HACKER and C. THOMALE, “Crypto-Securities Regulation: ICOs, Token Sales and Cryptocurrencies
under EU Financial Law”, 2017, 30. Available at
102
103
Hedqvist, op. cit., paragraph 49.
104
A. A. GIKAY, “Regulating decentralized cryptocurrencies under payment services law: lessons from
European Union law”, Journal of Law, Technology & the Internet 2018, 9, 23.
105
36
involve a payment through a cryptocurrency. This finding leads to the question whether the
relevant blockchain networks could in any way be subject to law in relation to payment
services. A preliminary problem however concerns the question how to connect a national
set of rules to a decentralized payment ledger. States developing such regulatory framework
will likely want to govern the operation of all cryptocurrency blockchain applications
accessible for users within their jurisdiction. The cryptocurrencies are designed to surmount
these very jurisdictions and the permissionless blockchain platforms as such are often built
by anonymous individuals. If, for this reason, an important cryptocurrency platform would
indeed escape necessary regulation, a state can resort to blocking access to the internet
webpages through which the blockchain application operates.106
32. The practical challenge of subjecting a decentralized network to national laws aside,
it is questionable whether any existing laws relating to payment systems can apply to
cryptocurrencies, given their very limited resemblance to traditional currencies. From an EU
perspective, the second European Payment Services Directive (PSD2) harmonizes the
European payments market by regulating payment service providers by submitting them to
the requirement of retrieving prior authorization, supervision by the competent home
member state authority and various rules of a prudential nature.107 These rules mean to
ensure these actors’ reliability. Cryptocurrency blockchain applications however fall out of
the scope of these rules.108 The currently dominant refusal to view cryptocurrencies as a
form of funds excludes them from the definition of payment services as applied by PSD2.109
Note that the ECJ cautiously labeled Bitcoin a ‘means of payment’ instead of a ‘payment
instrument’, which is a defined concept under PSD2, in the above-cited Hedqvist case.110
The scope of PSD2 is intentionally broad compared to the initial PSD dating from 2007, in
order to ensure equivalent operating conditions to existing players and to new market
106
2015, 55-56. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last accessed 7
March 2017.
107
Articles 5-37 Directive (EU) 2015/2366 of the European Parliament and of the Council of 25 November
2015 on payment services in the internal market (PSD2).
108
For a more elaborate analysis of the status of Bitcoin under PSD; see: C. HAUBEN, “Bitcoin en EU-recht:
de virtuele vreemde eend in de bijt” in M.E. STORME and F. HELSEN (eds), Innovatie en disruptie in het
economisch recht, Antwerpen, Intersentia, 2017, 83.
109
Ibid., annex I.
110
Hedqvist, op. cit., paragraph 50. See also: A. A. GIKAY, “Regulating decentralized cryptocurrencies under
payment services law: lessons from European Union law”, Journal of Law, Technology & the Internet, 2018,
9, 23.
37
entrants who introduce technically innovative payment products or services.111 However, it
does also not apply to cryptocurrencies. Among other changes, PSD2 thus created a
regulatory level playing field for both the banks and the fintechs, notably the payment
initiation and account information service providers.112, 113 These new actors create tools that
enable the consumer to rely on payment instruments provided by service providers other
than the bank where he or she holds an account in order to make online, immediate credit
transfers. These providers will also be able to offer to consumers a consolidated view of
their different payments accounts – by consequence, consumers would only need to possess
one payment service account. Cryptocurrencies are a different type of fintech, even though
they could serve a similar purpose of effectuating direct online payments without the need
to rely on the service provided by an established bank. In fact, cryptocurrencies are way
more far-reaching than the third-party service providers regulated by PSD2. Whereas PSD2
permits the payment initiation and account information services to compete with the banks’
payment services by establishing cooperation between traditional and new actors, the
philosophy behind cryptocurrencies is grounded on a radical split from established actors.
The ECB confirms this analysis in its latest report on cryptocurrencies. The phenomenon is
still new and evolving which makes it difficult to create tailor-made legislation.
Additionally, the ECB states that such efforts would not be proportional given the limited
usage of cryptocurrencies today.114 It must be noted however that the above analysis only
applies to pure cryptocurrencies, i.e. crypto-assets that are solely used as a means of payment
and that do not serve an investment-related purpose to the buyer, nor to the issuer – other
regulations may apply if these conditions are not fulfilled.115
B. PERMISSIONED AND PRIVATE – blockchain-based interbank

payment solutions
33. An approach fundamentally different from peer-to-peer transferable

cryptocurrencies is reflected by initiatives exploring the possibilities to adopt blockchain-
111
Recitals 3-6 PSD2.
112
P. BERGER (Baker McKenzie) at “Studiedagen Instituut Financieel Recht: Fintech”, Brussels, 16 January
2018.
113
Annex I and article 4 (3) PSD2.
114
115
Infra, 53-58.
38
based solutions to improve the current interbank payment infrastructure. Contrary to
cryptocurrencies, such ledgers only play an indirect role in payment transactions; they are
to be used by the actors underpinning payment services and not directly by the payors and
payees involved. They serve as a tool to enhance the operation of the modern financial
system.116 Such blockchain is thus likely constructed as a private and permissioned interbank
network – other options are theoretically possible but unfeasible at the time of writing due
to reasons of system integrity and legal compliance.117 In the current, centralized
constellation, accounts on different levels can be involved in the transfer of one and the same
asset, from the commercial bank accounts held by the payor and payee to, in some instances,
the accounts held with an intermediating central clearing bank at the top of the pyramid. The
payment data is validated and reconciliated – sometimes even manually – by all different
actors involved. This is a labor-intensive and costly process and moreover, it is prone to
errors.118 We discern two scenarios for the adoption of distributed ledger technology in the
interbank payment infrastructure: a more basic and a radical one.119 The first scenario
involves the introduction of decentralized technology to support or replace an interbank
telecommunication network that facilitates clearing between financial institutions. With a
network of more than 11,000 financial institutions, SWIFT currently is the leading actor
providing this service.120 The second option involves the replacement of the complete
interbank payment infrastructure. It does not only support the clearing function, but it also
provides for interbank settlement, like the ECB’s TARGET 2 system does for transfers
denominated in Euro.
1. Interbank financial communication network
34. A blockchain that merely serves as an interbank data transmission network does not
contain information on deposits or accounts. It provides for the transmittance of standardized
messages between banks. One of these message types, the MT103, is used to transmit
payment instructions from one bank to another, in the clearing process of a payment between
116
P. ATHANASSIOU, “Impact of digital innovation on the processing of electronic payments and contracting:
an overview of legal risks”, ECB Legal Working Paper Series 2017, 26. Available at
117
Ibid., 28.
118
Ibid., 27.
119
E. KARAINDROU, “Distributed ledger technology and the future of payment services”, 2017, 14-18.
Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3078523, last accessed 10 March 2018.
120
SWIFT, https://www.swift.com/about-us/discover-swift/messaging-standards, last accessed 15 April 2018.
39
their respective customers. Such message is merely an instruction and entails no immediate
transfer of funds from one account to another. This happens later through debiting and
crediting of the accounts held by the banks themselves, usually with an intermediary central
bank they rely on. In a 2016 whitepaper, SWIFT states that the discerning characteristics of
distributed ledger technology have the potential to be advantageous for its services.121 These
strengths include, according to the paper, the efficient propagation to all nodes in near to
real-time, the full traceability and immutability of transactions recorded in the chain and
possibly automated reconciliation. Also, SWIFT notes that the decentralized setup of a
blockchain guarantees trust in the integrity of the ledger and accuracy of the recorded
transactions. This last advantage implies a highly decentralized blockchain setup in which,
for instance, every participating financial institution is a node. This prompts the question
how SWIFT sees the actual implementation of the network, and what its own role in this
network is. The paper serves as a preliminary assessment and it does not yet provide details
on this matter. It only cautiously states that “DLT-based services could be provided by
SWIFT, our community or third parties”.122 Nevertheless, this organization emphasizes its
status as an industry cooperative and declares that thanks to its governance model, it is best
positioned to control the access to the permissioned network by taking up the task of
preselected central authority.123, 124 The whitepaper furthermore stresses the importance of a
high degree of standardization in order to allow for straight-through processing and
interoperability.125 Again, SWIFT refers to its business knowledge and experience with ISO
standards, establishing its qualities to realize an industry-standard distributed ledger.126 The
interim results of a proof of concept distributed ledger, launched by SWIFT and thirty-three
leading global banks are said to be “encouraging for this business use case” in a recent
report.127 Strengthened by this initial project, SWIFT reasserts it is in the industry’s interest
that the distributed ledger innovation is integrated in the contemporary interbank
infrastructure, so it can be further developed based on agreed standards.128
121
SWIFT, “SWIFT on distributed ledger technologies”, position paper, 2016, 4. Available at
https://www.swift.com/resource/swift-distributed-ledger-technologies, last accessed 15 April 2018.
122
Ibid., 3.
123
Ibid., 17.
124
Supra, 13.
125
Ibid., 9.
126
Ibid., 14.
127
SWIFT, “gpi Nostro Proof of Concept”, interim report, 2017, 3. Available at https://www.swift.com/news-
events/press-releases/swift-tests-show-blockchain-has-potential-for-global-liquidity-optimisation, last
128
Ibid., 21.
40
35. Despite SWIFT’s efforts and experience, it has also been argued that it is outdated
and unable to adopt the wider changes the financial telecommunications sector needs.129 In
2016, the above-mentioned initiative of Ripple engaged in a collaboration with a number of
banks to build an interbank network on the Ripplenet blockchain.130 This endeavor could
perform the services offered by SWIFT but it would be designed as a more accessible
application; and its capacity is not necessarily limited to processing payment information, it
would also be able to provide settlement services for any transaction with regard to digitally
represented assets. 131, 132
This larger prospect leads us to the second and more radical
scenario for the adoption of distributed ledger technology in the interbank payment
infrastructure, i.e. in a full-fledged interbank money remittance network.
2. Interbank money remittance network
36. A distributed ledger application that aims to also replace the interbank settlement
infrastructure operates as a master-record, held by all banks in the network. Unlike the data
transmission network described in the previous paragraph, this blockchain application is
comprised of entire accounts and balances but contrary to the cryptocurrency schemes, it
would not compete with traditional currencies. Indeed, rather than to replace commercial
money, its purpose is to represent money held in accounts with commercial banks in one
common distributed ledger.133 The ledger is continuously updated with all transactions or
balance adjustments and every bank in the network has direct access, eliminating all labor-
intensive reconciliation processes. The participants may choose to use a second distributed
ledger to increase internal efficiency, a blockchain within a node that is compatible with the
ledgers operated by other banks and with the master-ledger.134 Despite the absence of a
trusted third party responsible for the processing and monitoring of transactions in this
master-ledger, the state of the accounts held therein is regarded accurate and thus
129
C. SKINNER, “Will the blockchain replace SWIFT?”, American Banker, 8 March 2016. Available at
https://www.americanbanker.com/opinion/will-the-blockchain-replace-swift, last accessed 16 April 2018.
130
Global Payments Steering Group (CPSG), see https://ripple.com/insights/announcing-ripples-global-
payments-steering-group/, last accessed 15 April 2018.
131
K. JEMIMA, “Blockchain insiders tell us why we don’t need blockchain”, Financial Times, 2 May 2018.
Available at https://ftalphaville.ft.com/2018/05/02/1525253799000/Blockchain-insiders-tell-us-why-we-don-
t-need-blockchain/, last accessed 2 May 2018.
132
133
Ibid., 14.
134
Ibid., 15.
41
authoritative.135 This confidence requires a carefully designed consensus mechanism that
either relies on a limited number of nodes that are deemed to be especially trustworthy (to
the example of Ripple) or which encourages all participating banks to diligently remain
updating the chain in the network’s common interest.
The solution where a comprehensive blockchain-based application replaces the entire

contemporary back-end payment infrastructure is in fact the conceptual midpoint between a
society widely adopting pure, permissionless cryptocurrency blockchains (and thereby
dismissing banks) on the one hand and a group of banks giving their interbank
communication system the ‘blockchain facelift’ on the other hand. According to Elli
KARAINDROU, this is the most plausible and interesting scenario – she considers the initial
libertarian ideology behind the blockchain bygone and she claims that “one might say that
the question is no longer how DLT will disrupt the regulated sector, but rather how it will
become part of it.”136 This insight presumes a scenario where not only the nodes, but also
the distributed ledger itself is regulated.
3. Regulation?
37. The regulatability of the permissioned decentralized payment network is indeed a

crucial ingredient for its success. Philipp PAECH trusts that blockchain applications in the
financial industry will operate within the reach of relevant laws, regulators and courts.137
Given the densely regulated character of the financial sector, it is difficult to create a viable
interbank payment network that escapes the regulatory framework of the states of which the
financial industry is affected. Measures of a globally relevant nature like know your
customer (KYC) and anti-money laundering (AML) rules as well as tax-related requirements
are particularly relevant in this context because their enforcement may be threatened by the
135
136
137
42
pseudonymous character of a primitive blockchain.138, 139 Additionally, the 2008 financial
crisis revealed major vulnerabilities of the global financial integration and to some extent,
liberalization.140 International organizations like the Basel Committee and their member
states responded by promoting rules that would reinforce market integrity and financial
stability. The EU, for instance, adopted a more stringent regulatory framework with
qualitative and quantitative prudential standards for, and adequate supervision on, financial
institutions.141 The extensive set of KYC, AML and tax-related laws as well as the principles
bolstering the financial sector would likely also apply to an interbank blockchain network;
most certainly if it covers the EU and US. While these laws put substantial compliance duties
on the shoulders of the nodes, failing to abide by them risks to trigger draconian measures
by affected states, like forcing the internet service providers to block data passing through
the distributed ledger, thus effectively disabling the application.142 PAECH proposes that
alternatively, an interbank blockchain network could be set up among participants in
jurisdictions that barely or not regulate financial institutions and payment services, thereby
excluding the EU and the US.143 This fallback option is inadequate as it clashes with the
main objective of blockchain-based payment services; that is, executing cheap, swift and
secure international payments by eliminating the intermediary steps in the process. It is clear
that the reality of regulation is inevitable for a blockchain-based payment network with a
global reach. Less clear is the issue of how to control a body that lacks any form of central
power and that is identically geographically present at the address of every node. LESSIG
summarizes that “To regulate well, you need to know (1) who someone is, (2) where they
138
2017.
139
140
S. CLAESSENS and L. KODRES, “The Regulatory Responses to the Global Financial Crisis: Some
Uncomfortable Questions”, IMF Working Paper 2014, 6-8. Available at
https://www.imf.org/external/pubs/ft/wp/2014/wp1446.pdf, last accessed 16 April 2018.
141
Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the
activity of credit institutions and the prudential supervision of credit institutions and investment firms; and,
Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential
requirements for credit institutions and investment firms.
142
2017.
143
43
are, and (3) what they’re doing.”144 Yet, the blockchain is and will remain regulatable.145 In
the following paragraphs, we analyze how jurisdictions may identify, localize and govern a
permissioned blockchain-based money remittance application at the level its nodes, the
distributed ledger as an entity itself and the transactions it processes. While the first two
targets are to be subject to public laws of a regulatory nature, e.g. minimum capital
requirements, reporting duties and a reliable internal organization; the third category
comprises provisions of private law relating to the validity and enforceability of transactions
executed on the ledger.
a. Individual nodes
38. The nodes in an interbank money remittance network are to be existing banks and
financial institutions, hence, these actors are already regulated and both governing and
supervising them is nothing new. A payment ledger that relies on a cooperation of authorized
payment service providers should not pose great regulatory problems.146 It is possible that
the network also contains non-bank nodes with the only purpose of validating new
transactions.147 While KYC, AML, tax laws and minimum capital requirements are
irrelevant for these participants with a purely auxiliary role, it might make sense to subject
their activities and internal organization to some sort of prudential regulation and
supervision, according to the services they provide.148 Helpers of this kind are responsible
for bringing the blockchain’s consensus algorithm into practice and after all, this mechanism
is the backbone of the payment network’s reliability and security. Regarding the nodes that
take up an active financial role in the network, the single element demanding some attention
is the requirement to adequately guard the access gates to the network – that is, only financial
institutions that are indeed properly authorized to unfold banking activities can be allowed
144
See L. LESSIG, Code version 2.0, New York, Basic Books, 2006, 23. Available at
145
2017.
146
147
148
Ibid., 29; and, E. KARAINDROU, “Distributed ledger technology and the future of payment services”, 2017,
20. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3078523, last accessed 10 March 2018.
44
to participate.149 This can only be achieved with a permissioned, private blockchain
administered by a platform supervisor that has a good communication line with the
competent regulator or regulators. Note that every node can, regardless of its geographic
location, potentially access and impact any transaction executed over the ledger – be it, in
accordance with the rules set forth by the consensus algorithm; hence the necessity of a
tailor-made and performant mechanism to validate transactions. The shared control among
all nodes explains why every state where a participating bank is located has an actual interest
in regulating the entire network including all of its nodes. Yet, the resultant conflicts of laws
between nodes in different states render doing so practically unfeasible.150 A solution hinged
on one focal point and on some degree of mutual recognition makes more sense.151 For
instance, the home jurisdiction of a bank could make the legitimacy of participating in a
certain blockchain network subject to the existence of adequate regulation and supervision
by the jurisdiction appointed to govern the blockchain.152 Countries may want to create a
harmonized level playing field to ensure a common quality standard for this particular
regulatory task.153 Along these lines, we arrive at the question of how to regulate the network
as an entity.
b. Blockchain as an entity
39. The issue of governing the distributed ledger itself boils down to finding the right
anchor point to connect law with regulated substance. While it is almost impossible to
identify the person or persons controlling the software in a public, permissionless virtual
currency scheme like Bitcoin, this problem would not exist in relation to a modern interbank
money remittance blockchain.154 At this stage, well-established financial institutions and
even central banks are openly launching proof of concepts for this kind of application.
149
Or, in the EU context, institutions that enjoy a single license since the adoption of the Second Council
Directive 89/646/EEC of 15 December 1989 on the coordination of laws, regulations and administrative
provisions relating to the taking up and pursuit of the business of credit institutions (no longer in force).
150
151
Ibid., 20.
152
Alternatively, absent a uniform solution, countries can still enforce their AML, KYC and tax laws on nodes
under their jurisdiction by imposing full transparency of transactions and identification of transaction parties.
See A. WRIGHT and P. DE FILIPPI, “Decentralized Blockchain Technology and the Rise of Lex Cryptographia”,
2017.
153
154
Ibid., 1093.
45
PAECH calls the entity behind the development of the technology the ‘platform provider’.155
Contrastingly, according to a working paper published by the US Federal Reserve Board,
this entity has ‘ownership rights’ over the distributed platform.156 Both wordings have a
different connotation and it seems that PAECH’s term better complies with the underlying
philosophy of a blockchain application, regardless of its permissioned setup.157
Nevertheless, both authors agree that this entity is the single appropriate addressee for
regulation. It bears the responsibility for complying with financial regulation, supervision
and reporting obligations and it should oversee that the network does not allow for
transactions with non-regulated entrants or other ledgers lacking governance standards of a
comparable level.158, 159 Implementation requires policymakers and financial regulators to
tailor the existing regulatory framework accordingly in order to bring these new actors
within their scope in an adequate manner.160 From the perspective of the regulated entity,
efforts could be made to align the decentralized nature of the blockchain’s architecture with,
and to extend it to the responsibilities originating from the legal reality. Hence, the
individual financial institutions engaging to participate in the network as a node may want
to share the regulatory burden by engaging to equitably support the platform provider in its
governance tasks – this is where we believe that the term ‘platform owner’ fails to convey
this spirit of mutuality. Such collaboration arrangement is justified by the fact that, in the
event this blockchain application is successfully operationalized among the partnering
banks, it is expected to save them a substantial amount of operational costs; and, these
institutions will likely benefit from simplified compliance processes within their own
organization.161 Besides having a compliant internal organization put in place, the platform
provider should also ensure that its digital environment only allows for the execution of
155
Ibid., 1093 and 1101.
156
157
A relationship of ownership is appropriate vis-à-vis a traditional database, whereas a solution based on
distributed ledger technology is unique in its decentralized governance.
158
159
Acting in the capacity of the permissioned network’s preselected central authority, supra, 13.
160
161
Santander bank estimates that the adoption of blockchain technology may reduce transaction and
compliance costs for banks by up to 20 billion USD a year. See SANTANDER, INNOVENTURES et al., “The
Fintech 2.0 Paper: rebooting financial services”, 2016, 8. Available at http://santanderinnoventures.com/wp-
content/uploads/2015/06/The-Fintech-2-0-Paper.pdf, last accessed 17 April 2018.
46
transactions that are pursuant to the applicable rules of private law, in order to guarantee the
enforceability of the network participants’ rights.162 This consideration leads us to the final
and most challenging regulatory mystery: which law applies to cross-border payment
transactions registered on the ledger?
c. Transactions
40. The adoption of an interbank decentralized money remittance ledger as described in

paragraph 36 is particularly economically advantageous in the context of cross-border
transactions, executed and validated by nodes that are located in more than one jurisdiction.
Absent any adequate contractual arrangements, such transaction would potentially be
governed by the laws of every jurisdiction where a network node is located.163 A similar
analysis can be made regarding court jurisdiction: in the event a dispute arises, for instance
regarding a fraudulent or erroneously executed transaction; courts in every jurisdiction
where a node exists may simultaneously claim jurisdiction over the dispute.164 A clear
connecting factor between a transaction and a given territory does not exist due to the lack
of a single account or register where the funds involved are held.165 In the contemporary
legal landscape, the nodes in an interbank blockchain-based payment network would
transact under legally uncertain conditions.166 Hence, the network participants would incur
higher costs to mitigate this risk, partially defeating the advantage of decreased total
transaction cost.167 Such level of uncertainty is incompatible with the profoundly
coordinated blockchain application proposed in these paragraphs. The applicable law should
instead be defined from the outset, for all transactions occurring on the network. It may be
linked to the jurisdiction governing the platform provider as well as the network itself, as
presented in the previous paragraph; or it can be derived from a choice of law made by the
platform provider, or by the collaborating financial institutions setting up the network and
162
163
164
Ibid., 31.
165
Ibid., 31, see also P. PAECH, “The Governance of Blockchain Financial Networks”, Mod. L. Rev. 2017,
1106.
166
P. PAECH, “Securities, intermediation and the blockchain: an inevitable choice between liquidity and legal
certainty?”, Unif. L. Rev. 2016, 21, 622.
167
47
thus establish the provider.168 This choice of law should however be restricted to exclude
forum shopping, which would possibly weigh on the trustworthiness of distributed ledger
solutions from the perspective of the end-users.169 In the EU legal context, the Settlement
Finality Directive (SFD) limits this choice to the Member States where at least one of the
network participants has its head office.170
The blockchain-based configuration of the money remittance network should entail near to
real-time settlements. This quality may be weakened by erroneous implementations of the
technology or, in an early stage of development, by an inadequate linkage between the
accounts presented in the application’s master-ledger on the one hand and the concrete
financial situation of a bank on the other hand. In this situation, the law applicable to the
transactions is particularly relevant in the event one of the nodes should become insolvent
and other nodes seek to enforce their rights vis-à-vis the insolvent network participant.171
This enforceability is usually determined by the laws of the state where the debtor’s main
interests are located. This is also the approach followed by the EU Insolvency Regulation,
which in principal connects both the issue of international jurisdiction and applicable law to
this location-based factor, insofar it is situated within an EU Member State.172 Given the
multitude of potentially applicable insolvency laws this rule entails, it would once again be
very costly to achieve legal certainty for the participating banks.173 We suggest that an
analogous solution as presented above may resolve this obstacle, i.e. the countries involved
should simultaneously link the law applicable to any eventual insolvency dispute to the
qualified jurisdiction governing the platform provider itself.
C. CONCLUSION
41. We have identified several configurations for how DLT innovations may affect
money as we know it and the operation of the financial institutions underpinning our
payment systems. The pure, highly decentralized cryptocurrencies in their current form fail
168
Ibid., 1106.
169
Ibid., 1106.
170
Article 2 (a) Directive 98/26/EC of the European Parliament and of the Council of 19 May 1998 on
settlement finality in payment and securities settlement systems (SFD).
171
P. PAECH, “The Governance of Blockchain Financial Networks”, Modern Law Review, 2017, 1107.
172
Articles 3 and 7 Regulation (EU) 2015/848 of the European Parliament and of the Council of 20 May 2015
on insolvency proceedings.
173
P. PAECH, “The Governance of Blockchain Financial Networks”, Modern Law Review, 2017, 1107.
48
to meet the conditions set forth in economic and legal definitions of money, their use for
payment purposes is still very limited and therefore, extensive regulation is not yet present
nor is it urgent. While these initiatives further develop, cryptosecurity holders and users
should be conscious about the differences with the money held in bank accounts; of which
the most important difference pertains to the lack of a legal claim vis-à-vis the blockchain
entity.
42. Alongside the further development of pure cryptocurrencies, it is likely that

established financial institutions will adopt DLT to a certain degree to improve the efficient
administration and resilience of bank accounts and payment systems. Such innovation may
take place on the level of the existing interbank financial communication network or it may
even have the ambition to entirely replace the contemporary mechanisms by putting in place
a detailed and reliable master-record for all accounts held by the financial institutions in the
permissioned network. The second option will indeed bring about the greatest leaps of
increased efficiency, but it will require an enormous effort to convince and unite enough
participants to match the size of for instance the SWIFT systems and thus to create a relevant
global master-ledger; as well as to guarantee its interoperability with the existing computer
programs of all banks involved. Consequentially, the initiative undertaken by SWIFT to
enhance the legacy financial communication network is a more feasible scenario in the
nearer future, thanks to SWIFT’s established participation level in the financial industry and
its experience and standardization.
The deployment of an interbank DLT platform gives rise to a number of important policy
considerations as to the governance of the participating financial institutions, the interbank
platform itself and the law applicable to the transactions potentially effectuated thereon. In
the most workable approach to handle these three distinct legal questions, an important role
is reserved for the platform provider at the basis of the relevant DLT application. We expect
the platform provider to be a new and specialized actor established by the cooperating
financial institutions, responsible for a specific range of tasks regarding access control as
well as governance and compliance of the platform, its nodes and its users. In the context of
the appropriate international regulation, the platform provider can also serve as a reference
point to hinge applicable law and international jurisdiction on, with regard to the legal
relationships on the DLT.
49
CHAPTER 2. CRYPTOSECURITIES
43. While blockchain technology may considerably improve efficiency in

disintermediated peer-to-peer payments as well as in interbank payment mechanisms, such
improvement may be even greater if blockchain innovation is applied to the practice of
issuing, holding and transferring securities. Likewise, the legal challenges this transition
entails are also considerably greater, given the tremendously complex set of rules
surrounding these particular financial activities. We will highlight several aspects of the
current market infrastructure at the outset of this part, followed by an exploration of various
potential blockchain modernizations in this field.
A. THE MODERN SECURITIES MARKET – intermediaries,

origins and problems
44. The contemporary organization of holding and trading securities has become a prime
example of ultimate intermediation, especially after the implementation of some global and
EU responses to the 2008-2009 financial crisis. The following paragraphs explain the
functions and origins of the current framework, followed by some adverse effects of this
densely intermediated system.
1. Trade lifecycle
45. The lifecycle of a trade transaction involves numerous actors who all have their
specific tasks and responsibilities regarding the trading, clearing or settlement of the
securities. The number of relevant actors is determined by the complexity and risk innate to
the financial instrument at hand. The following brief overview of a transaction aims to
provide a general image of the industry, thereby revealing all steps and actors that possibly
play a role in a given trade.174
174
G.W. PETERS and E. PANAYI, “Understanding Modern Banking Ledgers Through Blockchain
Technologies: Future of Transaction Processing and Smart Contracts on the Internet of Money” in P. TASCA,
T. ASTE et al. (eds), Banking Beyond Banks and Money. New Economic Windows, Cham, Springer, 2016, 270-
272.
50
Trading – Investors place their orders to buy or sell a financial instrument with a trading
member, which is either a bank who centralizes these orders among a group of banks in an
order driven market, or a broker who internalizes these trades by determining an individual
buy or sell price for a certain financial instrument, in a quote driven market. The first
situation prevails in continental Europe whereas brokerage is typical in the US and UK
trading infrastructure. The banks or brokers will rely on an exchange that is responsible to
match the accumulated buy and sell orders. Hence, their task is to find counterparties for the
investors willing to buy or sell financial instruments.
Post-trading: clearing and settlement – After the banks or brokers establish a purchase
agreement between a buyer and a seller through the exchange, the clearing process is
initiated, which involves updating the accounts and arranging for the transfer of the
securities and money. Parties will clear the transaction, either bilaterally or through a third
party. If bilateral clearing is allowed by law, the contracting parties can engage in a direct
contractual relationship with each other. Given the considerable risks this entails, the use of
risk mitigation techniques like collateralization is essential. Alternatively, parties can seek
intermediation by a clearing house, whose role is to facilitate the settlement of the
agreement. Not all trading members have direct access to a clearing house, some need to
take an additional step through transacting with a clearing member. Clearing houses receive
the details on matched orders from their members. They standardize all steps leading up to
a net settlement of the aggregated orders; thereby reducing costs and the operational risk of
clearing and settling the transactions among multiple parties.
Central clearing – In the EU financial markets, the clearing of certain financial assets is
subject to a central clearing obligation.175 Central clearing serves as an additional layer of
risk insulation, built on the intermediation by a clearing house which operates as a Central
Counterparty (CCP). The CCP steps in between the original buyer and seller, thereby
becoming the buyer to the seller and the seller to the buyer of the original transaction. Hence,
through novation, it internalizes and virtually eliminates the counterparty risk the original
parties would otherwise incur vis-à-vis each other. Since CCPs have the important role to
guarantee the execution of the transactions, these entities are strongly regulated. The
175
E.g. certain classes of over-the-counter derivatives, determined in accordance with article 5 Regulation
(EU) No 648/2012 of the European Parliament and of the Council of 4 July 2012 on OTC derivatives, central
counterparties and trade repositories (EMIR).
51
intervention of central counterparties causes a shift of systemic risk from the trading banks
to the CCPs; hence, it is imperative for these entities to be robust enough to manage this risk
concentration. Therefore, CCPs in the EU are subject to prior authorization (or recognition,
for non-EU CCPs) and supervision by the ESMA and by national supervision bodies.
Settlement – The financial assets are held by custodian banks that ensure the safekeeping of
the investor’s securities and cash to mitigate risk of loss or theft. These banks hold their
clients’ assets in centralized custody at a national or international Central Securities
Depository (CSD or ICSD). This entity serves as a first point of entry into the market for
newly issued financial instruments, and usually all securities in circulation are deposited at
the same CSD at issuance. The custodian banks will send the trade details to the CSD, who
then validates and executes the transfer. While the custodian banks of buyer and seller are
not necessarily the same entity, the responsible CSD remains unchanged in the context of a
trade. For CSDs located within the EU, the CSDR and SFD establish a set of common
requirements.176 The CSDR aims at creating harmonized regulatory and operational
conditions for EU CSDs in order to enhance cross-border settlements. On the other hand,
the SFD plays an important role in mitigating the systemic risk related to settlement systems
in general, therefore it also applies to CSDs. Its goal is to guarantee the enforceability of the
transaction, even in the event of a participant’s insolvency.
2. Origins
46. The intermediated handling of transactions in the financial industry as displayed

above results from a number of historical efforts to increase transferability and thus liquidity
in financial markets; complemented by more recent initiatives to bolster their stability.
From a historical perspective, both shares and bonds have evolved from being nothing more
than a bundle of mutual personal obligations between issuer and investor. These primitive
financial instruments were difficult to transfer to a second acquirer because of three reasons:
the object of the relevant transfer was difficult to assess, given the personal nature of the
obligation; it was hard to ascertain whether or not the seller was empowered to dispose over
176
Regulation (EU) No 909/2014 of the European Parliament and of the Council of 23 July 2014 on improving
securities settlement in the European Union and on central securities depositories (CSDR) and the SFD.
52
the instruments and; lastly, the method to transfer these elementary shares and bonds, i.e.
assignment, was inadequate, given the fact that obligations (unlike rights) cannot be
assigned without the consent of the other party to the obligation.177 To resolve these
limitations, the market was in need of standardization – reducing various personal rights and
obligations to straightforward and fungible financial assets is the solution to avoid the excess
due diligence needs caused by the uncertainties in the initial situation.
To allow for smoother transferability, the market developed two new concepts: transfer of
financial assets by delivery of a certificate and transfer through entries in a registry.178 Rights
and obligations were incorporated in standardized paper, resulting in the certificate being
the security, allowing for transfer by delivery of a clearly identifiable asset – certificated
securities were treated as negotiable instruments, protecting the subsequent acquirer.179
Transfer through entry in a register by the issuer also promoted clarity and standardization;
and, additionally, it allowed for the issuer to, by changing the register, expressly accept the
transfer of obligations represented by the security, thus consenting to novation.180
While these solutions sufficiently addressed the lack of transferability, they were not future-
proof. The continuously growing financial markets in the industrialized world required more
liquidity and transfers of certificates or manual entries in registers did not support a high
frequency of transactions. This is where intermediation stepped in, along with two
innovations for increased liquidity: immobilization and dematerialization of the assets.181
Immobilized physical certificated securities were safely put in central securities depositories
and the financial institutions set up an infrastructure of interlinked accounts among them,
allowing them to keep record of all investors’ security holdings and transactions.182 Over
time, both physical certificates and changing the issuer’s registers became irrelevant and the
cascade of accounts set up by the financial industry replaced their functions.183 The issuance
177
178
Ibid., 616.
179
L. THÉVENOZ, “Intermediated Securities, Legal Risk, and the International Harmonisation of Commercial
Law”, Stan. J.L. Bus. & Fin. 2008, 13, 2.
180
181
182
183
Ibid., 618.
53
of securities had moved to the virtual, now computerized world, largely dominated by
trading members, clearing members and clearing houses, CCPs and CSDs.184
47. Among many other measures taken in the aftermath of the 2008 financial crisis, the
G20 decided to pursue a safer, more transparent and responsible financial system by
establishing a new and global financial policy. It instructed regulators to support effective
regulation and supervision to mitigate potentially systemic risks and the ESMA helped
implementing the G20 commitments in the EU.185 Part of this plan focused on the role
assigned to CCPs to centralize counterparty risk in the OTC derivatives trading business, as
explained earlier. This policy change drastically increased the systemic importance of CCPs
in the EU. While these entities are fundamentally different from banks in terms of their
business model, ESMA chairman Steven MAIJOOR estimates that the impact of a CCP failure
could exceed the systemic impact of the failure of a large bank.186 A similar systemic
important role was attributed to CSDs and accordingly, their operation and structure was
also harmonized and reinforced in the EU.187 These efforts illustrate the continued and
growing care dedicated to intermediaries that emerged from the immobilization and
dematerialization of securities. Certain intermediaries, especially banks, CCPs and CSDs,
have evolved from just necessary tools in meeting the need for more liquidity in the financial
market, into systemic important entities at the focal point of global financial policy
considerations attempting to decrease the risk of future market failures. Hence, their
functions in the modern day financial system have diversified and fulfilling these functions
adequately remains more relevant than ever before.
3. Deficiencies of the intermediated infrastructure
48. According to PAECH, the emergence of intermediation also entailed major

drawbacks, especially from the perspective of the investor.188 The dematerialization of
securities undermined the evidencing function of bearer securities and investors are not even
184
185
S. MAIJOOR, “Regulatory measures to prevent another crisis”, transcript, 19 November 2014, 2. Available
at https://www.esma.europa.eu/sites/default/files/library/2015/11/2014-1382.pdf, last accessed 22 April 2018.
186
Ibid., 9.
187
Recitals 1-5 CSDR.
188
certainty?”, Unif. L. Rev. 2016, 21, 617-619.
54
registered by their own name in the records that replace the certificates – only the
intermediaries are identified, and they hold the aggregate of the securities in bulk, pursuant
to the sequence of legal relations in the holding cascade. The former erga omnes claim held
by the investor vis-à-vis the issuer is reduced to a chain of stacked contractual relations,
mirroring each other from intermediary to intermediary.189 It is argued that the modern
transactional framework has weakened the position of the individual investor, especially in
the event one of the intermediaries becomes insolvent.190 The risks involved are enlarged
given the considerable amount of time it takes to complete the traditional settlement
process.191 The lack of harmonized legal approaches to determine the rights and duties in
the global intermediated financial infrastructure has worsened the problem.192
4. The blockchain alternatives
49. Various DLT-based solutions have been suggested to resolve the inefficiencies and
risks inherent to the modern infrastructure. These ideas differ in their degree of promoted
disintermediation, similar to the classification of the different strategies regarding the
implementation of blockchain technology in payment transactions. Some authors advocate
a Bitcoinization of the securities market.193 They propose complete disintermediation
through the establishment of a public distributed ledger serving as an independent
cryptosecurity record, where trades are concluded on a peer-to-peer stock exchange. The
issuance of these native cryptosecurities is called an Initial Coin Offering (ICO), i.e. the
blockchain version of an IPO.194 The cryptosecurities exist separately from the
intermediated securities market; however, in an alternative version of the native
cryptosecurity scenario, pre-existing intermediated securities would also be moved to the
189
Ibid., 619.
190
E. MICHELER and L. VON DER HEYDE, “Holding, clearing and settling securities through
blockchain/distributed ledger technology: creating an efficient system by empowering investors”, Journal of
International Banking & Financial Law 2016, 653.
191
The EU financial market has moved to a two days settlement cycle; in the US, this takes three days. See
also G.W. PETERS and E. PANAYI, “Understanding Modern Banking Ledgers Through Blockchain
T. ASTE et al. (eds), Banking Beyond Banks and Money. New Economic Windows, Cham, Springer, 2016, 270.
192
certainty?”, Unif. L. Rev. 2016, 21, 619-624.
193
E.g. LEE, PAECH and, to some extent, HACKER and THOMALE.
194
Note that an ICO should be distinguished from the hypothesis where pre-existing traditional securities are
moved to a blockchain environment.
55
blockchain environment, thereby gradually phasing out the current infrastructure.195 Others
envision a back-office adoption of blockchain technology to enhance the existing
intermediated approach.196 As is the case for cryptocurrencies, both tracks for blockchain
innovation are not mutually exclusive, yet they require a distinct regulatory approach and a
different focus.
B. HIGHLY DECENTRALIZED AND PUBLIC – token sales, ICOs

and …
1. vs. cryptocurrencies?
50. We already established that practice shows that the majority of cryptocurrency
holders is driven by speculative objectives rather than by an ambition to use the coins in
their account for day-to-day payments.197 Speculative investors of this kind anticipate an
appreciation of the cryptocurrency’s value, purely based on their artificial, coded scarcity
combined with an expected widespread adoption as a means of payment in the future.198
Besides this form of pseudo-investment, other blockchain-based assets have emerged that
bear a closer resemblance to traditional financial investment assets, for instance by
incorporating a dividend or interest rights, based on the expectation of future cash flows to
be generated by the issuer. However, it is often difficult to make the distinction between
cryptocurrencies and cryptosecurities since many crypto-assets exhibit specific
characteristics of both categories.199 Yet, the exact qualification of the tokens is important
in order to identify the potentially relevant legal framework.200 HACKER and THOMALE have
suggested a litmus test in case of doubt. According to these authors, a crypto-asset is to be
195
So-called trans-cryptocurrencies. See P. PAECH, “Securities, intermediation and the blockchain: an
inevitable choice between liquidity and legal certainty?”, Unif. L. Rev. 2016, 21, 634.
196
E.g. PETERS and PANAYI, MICHELER and VON DER HEYDE, ESMA.
197
Supra, 27.
198
For instance, the number of newly supplied Bitcoins decreases ensuring a final total of slightly less than 21
million coins in circulation. This upper limit will be reached around 2040, depending on the evolution of the
mining power in the network. See ECB, “Virtual currency schemes”, report, 2012, 25. Available at
http://www.ecb.europa.eu/pub/pdf/other/virtualcurrencyschemes201210en.pdf, last accessed 15 April 2018;
see also: J.L. VERHELST, “Zijn cryptomunten munten? Een analyse van Bitcoin” in M.E. STORME and F.
HELSEN (eds), Innovatie en disruptie in het economisch recht, Antwerpen, Intersentia, 2017, 34.
199
200
Infra, 54-57.
56
considered a cryptosecurity if at least one of the following circumstances applies: the issuers
raised significant expectations of profits through their communication, or, the given crypto-
asset is mostly bought with the intention to resell it with profit while the issuers are aware
of this incentive.201 Determining the nature of the crypto-assets based on the motifs of the
average token investor moreover corresponds with the scope of EU and US securities rules,
respectively embodied in the Prospectus Regulation and the US Howey-test. 202 Besides the
specific underlying objectives of investor and issuer, it is difficult to differentiate
cryptocurrencies from cryptosecurities.
51. Indeed, the technical setup of the fully disintermediated cryptosecurities strongly
resembles the construction of cryptocurrencies. In the context of the broader ‘tokenization
of assets’, companies have started issuing tokenized securities, thereby using public
blockchain platforms like Ethereum or Ripple.203 Just like cryptocurrency coins, these
securities exist only by entry in a public distributed ledger and ownership as well as all
transactions are recorded in this ledger; hence, the cryptosecurities are native crypto-
assets.204 Trading occurs on a public blockchain to prevent double-spending, the tokenized
assets are freely transferable upon initiative of the holder of the associated private key. Given
its public nature, a performant and future-proof consensus mechanism is indispensable for a
trustworthy cryptosecurities market. Tokens can be accurately designed to represent any
type of debt or equity financial instrument, possibly implementing smart contracts to
facilitate the execution of rights by the asset holder, e.g. automated dividend distribution to
shareholders.205
52. Similar to how publicly transferable native cryptocurrencies may compete with
traditional money, one could say that cryptosecurities compete with traditional means of
201
202
Ibid., 18-19 & 34.
203
In this regard, cryptocurrencies can also be seen as a specific kind of tokenized assets. See for example the
clarification published by the MONETARY AUTHORITY OF SINGAPORE (MAS), “MAS clarifies regulatory
position on the offer of digital tokens in Singapore”, press release, 1 August 2017. Available at
http://www.mas.gov.sg/News-and-Publications/Media-Releases/2017/MAS-clarifies-regulatory-position-on-
the-offer-of-digital-tokens-in-Singapore.aspx, last accessed 16 April 2018.
204
C. VAN DER ELST and A. LAFARRE, “Blockchain and the 21st Century Annual General Meeting”, European
Company Law Journal 2017, 4, 173.
205
Smart contracts, infra, 64 sqq.
57
issuing and trading financial assets.206 The blockchain solution has its strengths: it increases
stock market stability by, among others, banning the practice of spoofing and naked short
selling; and it allows for more transparency, fast settlement, twenty-four hours of trading
per day and reduced transaction costs.207 Peer-to-peer transactions or direct trading through
the exchange renders intermediation by a broker unnecessary and since the rights of all
investors are recorded in the distributed ledger, transfer agents or registrars are completely
obviated.208 Furthermore, these native cryptocurrencies are also well-positioned to both
meet the needs for higher liquidity in the financial markets without bringing along the
drawbacks caused by a chain of mirrored rights between numerous intermediaries.209 The
realization of the underlying objectives of dematerialization is preserved – tokenization
allows for even greater transferability and thus liquidity – while the cumbersome cascade of
contractual relations is eliminated along with the risks and uncertainty it caused. The
blockchain-based settlement system aims at increasing the direct relationship between
investors and the issuer, reducing the degree of dependency on intermediating actors.210
53. A specific aspect of native cryptosecurities is their initial emission to the market.
Unless they are meant to replace pre-existing traditional shares, new cryptosecurities are
distributed through an ICO. Note that an ICO can also be launched to create a new
cryptocurrency or any other kind of innovative blockchain platform. Many ICOs however
concern coins that serve an investment objective, for instance to raise money for a specific
and possibly lucrative project. For the time being, most established issuers of financial
instruments seek funding through the traditional infrastructure as described above, in order
to ensure legal certainty while the industry awaits regulatory changes to support blockchain
alternatives. For startups and young companies however, native cryptosecurities are
regarded an attractive and accessible new way of financing; hence, the startup environment
is currently the most important source of native cryptosecurities. They potentially offer
higher returns for the investors and, for the issuing company, it is a less burdensome way to
obtain financing than the venture capitalist alternative. Some even refer to ICOs as
206
Supra, 28 sqq.
207
Hastings Business Law Journal 2016, 12(2), 119-122.
208
Ibid., 123.
209
210
Consequentially, it may fulfill the role of ‘future innovations’ as envisioned by Charles MOONEY, see C.W.
MOONEY, “Beyond negotiability: a new model for transfer and pledge of interests in securities controlled by
intermediaries”, Cardozo L. Rev. 1990, 12, 414.
58
“unregulated issuances of crypto coins”, implying that cryptosecurity issuers avoid the
entirety of regulation surrounding a financing undertaking on the capital markets.211 The
statement is a misconception based on the finding that, currently, almost no regulation exists
that specifically addresses ICOs and other blockchain-based solutions for holding financial
assets. It must be noted that, depending on the governing jurisdiction, general financial
regulations possibly apply to these schemes, regardless of their underlying technology.212
2. Regulation?
a. Issuance and market infrastructure
54. Before assessing whether any legal provisions with a general scope may apply to the
issuance, trading and the (minimal) infrastructure surrounding cryptosecurities, we need to
establish how to appoint the jurisdiction that governs the issuer and the blockchain platform
it operates through. In the first place, concerning laws that create obligations for the issuer,
it is logical that an issuer will have to obey the rules that exist in the market where the
financial instruments are sold. As the platform will be accessible for investors through a
website, the issuer can be expected to use this layer to control in which countries the relevant
webpage is accessible, by blocking IP-addresses located elsewhere. Consequentially, the
issuer will, for instance, be responsible to obey prospectus rules that apply in the market or
markets where he offers the cryptosecurities through the blockchain platform’s website.213
Secondly, the general rules that apply to the blockchain application itself are more difficult
to enforce or to hold the issuer accountable for. However, while pure cryptocurrencies – in
contrast to their permissioned, private counterparts – lack any reference point for
enforcement, it might be more feasible to find a highly decentralized cryptosecurities
blockchain platform (or, its establishers) willing to actively comply with the relevant
laws.214 Indeed, a securities issuer seeking investments will likely choose a platform that is
trusted in the investor’s market. A platform that complies with financial regulations
211
See for example the wording used in I. KAMINSKA and P. MURPHY, “Bitcoin’s surge fuels fears of asset
bubble”, Financial Times, 14 May 2017. Available at https://www.ft.com/content/ce3ef54e-371b-11e7-bce4-
9023f8c0fd2e, last accessed 20 April 2018.
212
213
Ibid., 17.
214
See the preliminary question raised concerning the applications of payment service laws, supra, 31.
59
presumably bears such a ‘trusted’ seal. Moreover, a rational issuer will minimize its own
risk and ensure the reliability of the investment capital by choosing a trusted blockchain
platform. Accordingly, the success of this type of blockchain platform is to be contingent
on its compliance with the laws of a country that has a solid financial legal framework in
place, e.g. an EU member state; which is why it is in their interest to take such rules into
account.
55. In the event the spatial scope of EU or US security regulations would be established
to extend to a certain issuance of cryptosecurities and the market they are traded on, different
general sets of rules can be identified that do not take into account the technical
particularities of a given financial instrument. Consequentially, these laws contain
provisions that potentially apply to the activities on a market that is built on a highly
decentralized, public blockchain platform; and to the operation of such platform.215
Regarding the applicability of the general existing EU financial laws, the ESMA has already
stressed that the Prospectus Regulation, MiFID II, the AIFM Directive and the AML
Directive should be complied with by organizations issuing cryptosecurities. 216, 217 Other
relevant EU legal instruments generally applicable to securities, thus potentially applicable
215
Financial Times also recognized this reality after the misleading wording used medio 2017. See H. MURPHY
and B. THOMPSON, “Law firms look to capitalise on initial coin offering boom”, Financial Times, 26 March
2018. Available at https://www.ft.com/content/2ae9154c-1d56-11e8-aaca-4574d7dabfb6, last accessed 20
April 2018.
216
Article 1 (1) Regulation (EU) 2017/1129 of the European Parliament and of the Council of 14 June 2017
on the prospectus to be published when securities are offered to the public or admitted to trading on a regulated
market (Prospectus Regulation) (in force, but not yet entirely applicable); and,
Article 4 (1) (15) Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on
markets in financial instruments (MiFID II); and,
Article (4) (1) (n) Directive 2011/61/EU of the European Parliament and of the Council of 8 June 2011 on
Alternative Investment Fund Managers (AIFM Directive); and,
Article 3 (2) and article 2 (1) (2) Directive (EU) 2015/849 of the European Parliament and of the Council of
20 May 2015 on the prevention of the use of the financial system for the purposes of money laundering or
terrorist financing (AML Directive).
217
ESMA, “ESMA alerts firms involved in Initial Coin Offerings (ICOs) to the need to meet relevant
regulatory requirements”, statement, 13 November 2017. Available at
https://www.esma.europa.eu/sites/default/files/library/esma50-157-828_ico_statement_firms.pdf, last
60
to certain cryptosecurities, are the MAR, the UCITS Directive as well as the EMIR.218, 219
Note that rules on prospectus obligations and the rules regarding the trading and market
infrastructure are only applicable to securities, thereby explicitly excluding instruments of
payment such as pure cryptocurrencies.220 Determining the degree of this legal framework’s
applicability to ICOs and other tokenized financial instruments is however beyond the scope
of our current research. An in-depth analysis concerning the impact of notably the new
European Prospectus Regulation on ICOs is presented in the cited article by HACKER and
THOMALE.221 This research specifically addresses the subject of public blockchain
applications for registering and transferring financial assets in the EU legal context.
Similarly, in the US, Larissa LEE expects that most of the existing financial legal framework
will remain the same with a public cryptosecurities market. In a recent research paper, she
analyses the implications of the development of a cryptosecurities market, for the legal status
of traditional financial market actors under the Securities and Exchange Act. LEE concludes
that, while some intermediaries partially or wholly lose their relevance, the individual
responsibilities of issuers, purchasers and platforms that serve as an exchange would not
change.222
b. Transactions
56. Another question relates to determining the law applicable to the nature of the rights
associated with the cryptosecurity and the enforceability of their acquisition and disposition.
In the contemporary intermediated context, measures have been taken to promote the Place
218
Article 2 (1) Regulation (EU) No 596/2014 of the European Parliament and of the Council of 16 April 2014
on market abuse (MAR); and,
Article 1 (2) Directive 2009/65/EC of the European Parliament and of the Council of 13 July 2009 on the
coordination of laws, regulations and administrative provisions relating to undertakings for collective
investment in transferable securities (UCITS Directive); and,
Article 1 (3) EMIR.
219
220
Article 4 (1) (44) MiFID II and article 2 (a) Prospectus Regulation.
221
under EU Financial Law”, 2017. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3075820,
last accessed 12 February 2018.
222
Hastings Business Law Journal 2016, 12(2), 123-127.
61
of Relevant Intermediary Approach (PRIMA) in this regard.223 It is the answer to the conflict
of laws problems that emerged from the dematerialization and pooling of financial
instruments and it presupposes the existence of accounts held with intermediary actors.
However, excluding such intervention happens to be the very purpose of setting up a
blockchain for the issuance and trading of cryptosecurities. A possible alternative is to
determine the applicable law based on the location of the issuer, i.e. the lex societatis.224
This solution would lead to a situation where an investor with an internationally diverse
portfolio has to consider the rules of all countries where an issuer is incorporated.
57. Instead, it would be more manageable if all cryptosecurities issued through the same
blockchain platform were governed by one and the same jurisdiction, the lex systematis.225
This choice of law would have to coincide with the jurisdiction the platform itself complies
with for the sake of clarity. Thus, this mechanism once again implies the willingness of the
individuals or organization behind the blockchain platform to be governed by the laws of a
certain country, but as described above, this undoubtedly is in the interest of the issuers of
financial instruments.226
3. Conclusion
58. It is clear that the emergence of pure cryptosecurities has opened a door for issuers
to attract investments more fluently and in a less burdensome way. The investment
component of these tokenized assets means an important difference with cryptocurrencies,
arising from the fact that pure means of payment are exempt from many sets of rules that do
apply to financial instruments, as well as to their issuance and the infrastructure they are
traded on. As the relevant legal instruments are underdeveloped and only apply in general
terms, legal uncertainty still prevails at this experimental stage of cryptosecurities. This
explains why ICOs are, for the time being, mainly embraced by startups and smaller
223
See for example the aforementioned SFD and the Convention of 5 July 2006 on the Law Applicable to
Certain Rights in Respect of Securities held with an Intermediary (The Hague Securities Convention).
224
225
Ibid., 636.
226
Supra, 54.
62
innovative companies.227 Certain countries are willing to promote the pure cryptosecurity
emergence by taking some first steps towards the legal recognition of this innovation. France
recently passed a law that accommodates ICOs in the existing legal framework, by explicitly
allowing the use of DLT for the issuance and transmission of unlisted securities.228 The
issuer can choose to register the securities to a blockchain platform instead of to a CSD. The
French legislator ascertained that the targeted financial instruments are out of the scope of
EU rules regarding listed companies, regulated markets and organized trading facilities,
thereby avoiding a collision with mandatory EU rules. Similarly, the Monetary Authority in
Singapore invites fintech firms that develop blockchain platforms for financial services to
experiment with their innovations in the controlled environment of a proverbial ‘regulatory
sandbox’.229 This measure is in sharp contrast with efforts in China and South Korea to ban
ICOs.230
It is unlikely that the legal framework surrounding the established intermediated financial
markets would change overnight, especially not on the matters of heavily regulated subjects
such as listed companies and organized trading facilities. The ICOs and token trades will
not entirely replace the intermediated financial landscape anytime soon, however, there is a
tendency to grant the ICOs the room to further develop independently from the traditional
financial markets; while controlling potential risks. It is mainly important for investors to
receive information and education on the practical and legal differences between holding
shares through the intermediated market or by means of a blockchain-based issuance. LEE
compares the finding that both models for financial markets can exist next to each other to
the advent of the email and continued relevance of traditional post services.231 However,
while the modern market infrastructure must not necessarily be replaced, that does not mean
227
H. MURPHY and P. STAFFORD, “Blockchain explainer: a revolution in its infancy”, Financial Times (2
February 2018). Available at https://www.ft.com/content/6c707162-ffb1-11e7-9650-9c0ad2d7c5b5, last
228
Ordonnance n° 2017-1674 du 8 décembre 2017 relative à l'utilisation d'un dispositif d'enregistrement
électronique partagé pour la représentation et la transmission de titres financiers. Available at
https://www.legifrance.gouv.fr/affichTexte.do?cidTexte=JORFTEXT000036171908, last accessed 1 May
2018.
229
MAS, “Fintech Regulatory Sandbox Guidelines”, 2016. Available at
http://www.mas.gov.sg/~/media/Smart%20Financial%20Centre/Sandbox/FinTech%20Regulatory%20Sandb
ox%20Guidelines.pdf, last accessed 20 April 2018.
230
D. WEINLAND and L. LUCAS, “Asian investors flock to initial coin offerings despite bans”, Financial Times,
14 November 2017. Available at https://www.ft.com/content/05b2748c-c903-11e7-aa33-c63fdc9b8c6c, last
231
Hastings Business Law Journal 2016, 12(2), 127.
63
that DLT may not impact its back-end transaction mechanisms. The following paragraphs
will discuss how blockchain-technology may improve the efficiency and coordination across
all actors currently involved in the securities trade cycle.
C. PERMISSIONED AND PRIVATE – blockchain-based

interbank securities clearing and settlement
59. Financial institutions in the regulated securities market industry are actively
investing in the development of DLT-based solutions.232 Their incentive is two-fold: in the
first place, established actors likely want to guarantee the continuity of their own business
in the future; and additionally, the adoption of DLT can drastically increase the efficiency
of the infrastructure that currently underpins the financial markets. Pursuing the first
ambition demands innovative changes to keep offering competitive services – therefore,
both incentives go hand in hand. Likewise, the ESMA stipulated in a report to the industry
that DLT in financial markets should be of the permissioned, private kind.233 Policy and
governance arguments relating to the regulatability aspect of the technology explain this
stance. Analogous to the reasoning presented in paragraphs 37 to 40, it is manageable to
adequately subject emergent permissioned, private blockchain networks to a large part of
existing financial regulation and supervision; while highly decentralized ledgers need tailor-
made regulation to a greater extent.
60. DLT is capable of simplifying the contemporary regulated trade lifecycle in various
ways. It is suggested to adopt blockchain technology for each step of the trade settlement
process separately, so-called ‘consortium blockchains’.234 The existing actors situated at
every level of the cycle will then benefit from moving on to DLT. Consequentially, the
enhanced trade lifecycle works as follows. A securities transaction, once established by the
232
For example, the projects launched by clearing houses like the Australian Securities Exchange (ASX) and
the US Depository Trust & Clearing Corporation (DTCC). See M. ARNOLD, “Five ways banks are using
blockchain”, Financial Times, 16 October 2017. Available at https://www.ft.com/content/615b3bd8-97a9-
11e7-a652-cde3f882dd7b, last accessed 20 April 2018.
233
234
64
banks or brokers acting on behalf of the investors, is securely registered to and verified on
the first blockchain network; to which access is permissioned to the banks or brokers. By
doing so, validly recorded orders can easily be accumulated to facilitate communication with
the entities at the clearing level in the chain. Secondly, clearing members can set up a
distributed clearing house. They are the nodes in a blockchain to which the transaction is
uploaded in the form of a smart contract that represents the bilateral relations between the
buyer’s and seller’s end of the transactions. Consequentially, there is no need for an
intermediary entity cutting the contractual relation in two halves. Risk mitigation
mechanisms are to be built into the standardized smart contracts created at this level of the
chain – while a securities settlement blockchain network still primarily serves to store the
transfers of the financial asset, second generation innovations are introduced to expand its
functionality. Once this smart contract is verified, the execution is carried out by transmitting
the relevant information to the custodian or CSD level. Settlement entities, thirdly, replace
the contemporary accounts by the blockchain substitute. The custodian banks are the nodes
who are running this ledger. This innovation enhances the safety and permanence of the data
kept by custodian entities, which, after all, is the core purpose of custodians. This account
data can be complemented by smart contracts that support and automate other services
offered to clients and investors. An example of such application is the tool for shareholder
voting at an issuer’s annual general meeting, as described by VAN DER ELST and LAFARRE.235
In a scenario of full DLT adoption at the settlement level, these accounts will represent the
actual custody of all dematerialized financial assets in circulation. Upon issuance of these
securities, they will be directly recorded to the settlement ledger. Consequentially, the assets
registered on the blockchain application at this third and final stage of the settlement cycle
are native crypto-assets, contrary to the intermediary transactions that only occur on the DLT
systems established at phase one and two of the trade sequence.
The configuration described above is an example of how different DLTs can enhance the
performance of all contemporary intermediaries in the settlement cycle, while these entities
fulfill their individual purposes independently. Additionally, the advantages of reduced risk
and costs would be shared among all actors on every level of the procedure, including the
end-users and investors. Indeed, the interconnectedness of the three layers of the DLT and
the streamlined processes within every level leads to a shorter settlement cycle, which, by
235
C. VAN DER ELST and A. LAFARRE, “Blockchain and the 21st Century Annual General Meeting”, European
Company Law Journal 2017, 4, 174.
65
definition, also brings about a lower level of credit and counterparty risk.236 Nevertheless,
the effectuation of this adoption scheme must acknowledge various elements of legal reality
and the crucial role of certain intermediaries, especially in scenarios where a central clearing
obligation exists. In particular, in the EU, this process needs to be made compatible with the
standards stipulated by the ESMA for emerging DLT solutions on the securities markets.237
61. In 2016, the ESMA has launched a discussion paper with the ambition to engage in
a dialogue with the market of DLT fintech companies.238 The 2017 report is based on the
findings from this dialogue and it serves as a set of guidelines for innovative financial
players with the ambition to develop DLT solutions that are meant to operate within the
boundaries of the regulated EU financial market. Given the premature stage of various
projects, the ESMA chose to follow a general approach in the discussion of DLT
opportunities and caveats, without detailed consideration of where the DLT role in the
settlement cycle is envisioned. A plausible explanation is that the ESMA tries to avoid
unnecessarily limiting the innovative creativity; additionally, the DLT potential should be
accessible for the stakeholders at all levels of the current settlement cycle. Nevertheless, the
report heavily focuses on post-trading processes and the ESMA authors promote the concept
of a single ‘golden record’ blockchain which is shared across all market participants and
which centralizes all information related to a given financial instrument. Such scheme would
in fact allow for near to instantaneous clearing and settlement with a single step trade
confirmation, affirmation, allocation and settlement.239 This disruptive vision is somewhat
contrary to the ESMA’s conviction regarding the incessant relevance of regulated, pre-
authorized and supervised entities, even in the context of DLT clearing. It also seems
inconsistent with the concerns expressed in relation to a possible disturbance to the fair
competition between market participants who are or are not granted access to the single
distributed ledger. The blockchain is set up by private entities and as its success is largely
236
237
Ibid., 7-19.
238
ESMA, “ESMA Assesses usefulness of distributed ledger technologies”, press release, 2 June 2016.
Available at https://www.esma.europa.eu/press-news/esma-news/esma-assesses-usefulness-distributed-
ledger-technologies, last accessed 9 April 2018.
239
66
determined by the network effects of its extensive adoption, the market configuration risks
to be prone to monopolistic effects.240
62. The ESMA stresses that counterparty and credit risk cannot be excluded by
decentralized ledgers in their current development stage; especially not in relation to
financial instruments with a term structure. Accordingly, the execution of transactions that
are in the scope of the EMIR and MiFIR clearing obligation should, if conducted on a
blockchain, be compliant with the relevant provisions. More specifically, an authorized CCP
must join this network or the application itself would have to meet the requirements and
acquire authorization. An analogous reasoning pertains to the settlement level: if the
settlement transactions fall within the scope of the CSDR, the risk mitigating mechanisms
apply and hence a licensed CSD must be involved in the blockchain or it needs to obtain a
CSD authorization itself. A scenario where clearing and settlement are transformed into a
single, near to instantaneous transaction which is conducted on a single ‘golden record’, that
accommodates any trade or financial asset regardless of its individual features and needs,
leads to an unduly consolidation of the specific features attributed to a CCP c.q. a CSD –
that is, to the extent the activities are subject to a CCP clearing obligation. After all, a CCP
entity centralizes and accepts risk whereas a CSD is meant to avoid risk.241 On the other
hand, the DLT will not be deployed across the entire securities market all at once. The first
blockchain-based settlement systems that are designed to facilitate the existing
intermediated procedure in the EU could target a more accessible market segment with
straightforward financial products and lower risks; thereby steering away from the central
clearing obligation and, possibly, other regulations.242
240
Ibid., 11-15.
241
The DLT platform Corda, launched by R3, would however allow for the implementation of CCP and CSD
functions, and to meet the relevant legal requirements. See M. HEARN, “Corda: A distributed ledger”,
whitepaper, 2016, 28. Available at https://docs.corda.net/_static/corda-technical-whitepaper.pdf, last accessed
5 May 2018.
242
As also indicated by many of the respondents to the ESMA discussion paper, see ESMA, “The distributed
ledger technology applied to securities markets”, report, 2017, 23. Available at
67
D. CONCLUSION
63. Regardless of how a blockchain that administers the issuance, holding and
transactions of financial instruments relates to the contemporary, intermediated securities
markets; users and developers should be aware of general financial regulations that may not
have been created with DLT solutions in mind but that do apply to the activities this
innovation is capable to facilitate. The most suitable path depends on a wide range of factors:
the type of financial assets involved, the targeted average investor, the issuer’s quality and
strategy… Accordingly, interesting DLT applications will emerge both on a highly
decentralized level and in permissionless setups; both to directly accommodate native
crypto-assets and to facilitate the execution of an intermediary step through validating non-
native transaction data. The great variety of blockchain arrangements is capable to meet an
equal variety of market needs; while all different setups benefit from the core blockchain
features regarding its immutability, resilience and transparency within the network.243
Hence, we find it beneficial to promote the co-existence of both the disintermediated
scenarios (notably ICOs) and the traditional financial markets with DLT-based clearing and
settlement. An enhanced legal framework should focus on advising both users and investors
on the differences and on leading them to the most suitable scenario, especially where more
considerable risks are involved. Establishing such clarity would be significantly more
challenging if securities that pre-exist in the intermediated financial system are partly moved
to the pure DLT environment, thereby creating a category of trans-cryptosecurities.244
Furthermore, lawmakers will closely have to monitor DLT while the technology further
develops, and new regulatory action should bear the diversity of possible applications in
mind. While some legal novelties may overlap for the different sorts of crypto-assets, e.g.
regarding the law applicable to transactions involving native crypto-assets; other laws will
need to be tailor-made for an individual scenario, e.g. a set of regulatory authorization
conditions specifically meant for the activities of a DLT-based clearing solution. Given the
fact that the global nature of securities markets will only extend further due to the advent of
blockchain technology, EU and international cooperation and engagement will be crucial
243
T. ASTE et al. (eds), Banking Beyond Banks and Money. New Economic Windows, Cham, Springer, 2016, 259-
260.
244
68
for the development of an adequate legal framework that finds a balance between mitigating
risks and promoting valuable innovation.245 DLT is expected to rapidly develop, hence,
substantial efforts will be required to modify the legal reality accordingly. Or, as
appropriately phrased by Joanna Diane CAYTAS, “the current speed of global constitutional,
legislative, regulatory and adjudicatory systems not only should but must accelerate by
quantum leaps commensurate to those of the underlying technologies they relate to.”246
245
246
J.D. CAYTAS, “Developing Blockchain Real-Time Clearing and Settlement in the EU, U.S. and Globally”,
Columbia Journal of European Law: Preliminary Reference 2016, 11.
69
TITLE 4. SMART CONTRACTS
Le langage est source de malentendus.247

– A. DE SAINT-EXUPÉRY, Le Petit Prince
CHAPTER 1. ORIGINS
64. In the previous chapters, we already came across several instances where the use of
smart contracts is suggested to further enhance the functionality of a crypto-asset DLT
application, e.g. for the implementation of shareholder voting rights.248 While the recent
advent of blockchain technology provides the ideal platform to more efficiently
operationalize the concept of a smart contract, a considerable lapse of time exists between
the first appearance of both individual novelties. Nick SZABO first defined a smart contract
back in 1994 as follows:
A smart contract is a computerized transaction protocol that executes the terms of a

contract. The general objectives of smart contract design are to satisfy common
contractual conditions (such as payment terms, liens, confidentiality, and even
enforcement), minimize exceptions both malicious and accidental, and minimize the
need for trusted intermediaries. Related economic goals include lowering fraud loss,
arbitration and enforcement costs, and other transaction costs.249
According to this initial understanding, a smart contract is mainly characterized by the

computer protocol that determines its effectuation. SZABO built his idea on the theory that
many common contractual clauses are capable to be embedded in the hardware and software
humans interact with.250 As such, the machine incorporates the actual realization of the
meeting of the minds that precedes and constitutes the contract. For instance, the fulfillment
247
A. DE SAINT-EXUPÉRY, Le Petit Prince, Paris, Gallimard, 1943, 80.
248
Supra, 18 and 60.
249
N. SZABO, “Smart Contracts”, 1994. Available at
abo.best.vwh.net/smart.contracts.html, last accessed 24 April 2018.
250
N. SZABO, “Formalizing and Securing Relationships on Public Networks”, First Monday 1997.
71
of a sales contract is translated into the design of a vending machine.251 This primitive
example only includes one standard contract but with the development of computers came
the rise of a formal language to design more intricate and custom-made ‘vending machines’
and to operationalize specific financial contractual clauses that constitute collaterals, bonds,
futures, derivatives…252, 253 The aggregate of formal computer language is deemed to be
able to process data correctly in order to reach the outcome envisaged by the traditional
contract counterpart. Correspondingly, a smart contract is an execution-focused tool. The
result of embedding an agreement into hardware and software is that it becomes increasingly
and sometimes prohibitively expensive for parties to breach the terms of the contract; or,
more precisely, to breach the terms of the coded smart contract version of it.254 In conclusion,
“a smart contract is an agreement whose execution is automated.”255
65. The emergent blockchain technology and the concept of smart contracting are
different yet complementary. DLT extends the capacities of formal computer language, in a
way that it allows to guarantee the effective completion of the smart contract script between
individuals. It promotes the protocol in its ability to indeed achieve its execution through
the terms and mechanisms of the contract on the DLT itself without recourse to third
parties.256 By virtue of the employment of blockchain technology, both parties necessarily
run the same ‘golden’ version of the contract code which is anchored in the distributed
ledger; rather than both parties installing an individual instance of the program on their own
computers and relying on each other’s willingness to the run the code agreed on.257
Likewise, the smart contract concept extends the functionality of the initial, first generation
blockchain technology. It allows for the use of the consensus mechanism to process and
validate computation rather than to just validate data; and to cement the running of a
computer program in the blockchain rather than just adding blocks of plain transactional
251
Ibid.
252
Ibid., and N. SZABO, “A Formal Language for Analyzing Contracts”, 2002. Available at
abo.best.vwh.net/contractlanguage.html, last accessed 24 April 2018.
253
For a practical example of an elaborated bond issuance smart contract, see R3 LIMITED, “Writing a
contract”, 2018. Available at https://docs.corda.net/tutorial-contract.html, last accessed 26 April 2018.
254
N. SZABO, “Formalizing and Securing Relationships on Public Networks”, First Monday 1997.
255
M. RASKIN, “The Law and Legality of Smart Contracts”, Geo. L. Tech. Rev. 2017, 309.
256
Ibid., 321.
257
ISDA and LINKLATERS, “Whitepaper: Smart Contracts and Distributed Ledger – A Legal Perspective”,
2017, 9. Available at https://www.isda.org/a/6EKDE/smart-contracts-and-distributed-ledger-a-legal-
perspective.pdf, last accessed 24 April 2018.
72
data to the chain.258 The establishment of platforms that effectively host and execute smart
contract formal language has brought the smart contract concept into the mainstream.259 One
of the most successful DLT platforms supporting the use of smart contract language is
Ethereum, which follows the model of Bitcoin – it is permissionless and it relies on an
elaborate consensus mechanism – and it accommodates its own pure cryptocurrency to
perform contractual payment obligations with, ether.260 This platform has its own
programming language, designed specifically for the implementation in smart contracts:
Solidity.261
CHAPTER 2. CODERS’ APPROACH – code is law
66. The currently prevailing view on smart contracts represents a mere technical-
theoretical approach – what is the combination of smart contracting language and DLT
capable of? It represents the viewpoint of the computer engineer rather than how a law
professional examines it. While SZABO’s definition highlights the automated execution
aspect, little consideration is given to the establishment of a legally binding contract. By
describing a self-executing transaction protocol that realizes the terms of a (legal) contract
as a tool that, among others, decreases enforcement costs, it seems that execution and
enforceability are near to confounded. The definition presented by CLACK et al. even puts
both concepts on the same level:
A smart contract is an agreement whose execution is both automatable and

enforceable. Automatable by computer, although some parts may require human
input and control. Enforceability by either legal enforcement of rights and
obligations or tamper-proof execution.262 (emphasis added)
258
See also: D. DRESCHER, Blockchain Basics: A Non-Technical Introduction in 25 Steps, New York: Springer
259
260
Ethereum, https://www.ethereum.org/, last accessed 7 April 2018.
261
Solidity, http://solidity.readthedocs.io/en/latest/, last accessed 7 April 2018.
262
C. D. CLACK, V. A. BAKSHI and L. BRAINE, “Smart Contract Templates: foundations, design landscape and
research directions”, position paper, 2016, 2. Available at https://arxiv.org/pdf/1608.00771v2.pdf, last
73
Correspondingly, these authors differentiate between ‘traditional’ and ‘non-traditional’
means of enforcing an agreement. The traditional method comes down to enforcement by a
body of law that is backed by the power of a government.263 The non-traditional method
depends on the use of tamper-proof blockchain technology. This type of enforcement is
based on the assumption that it is possible to create a perfect computer language
implementation of an agreement which entirely excludes the possibilities of wrong-
performance and non-performance of the agreement.264
67. In effect, the computer science approach assumes that the computer code’s
automation of a transaction between parties, when embedded in the solid framework of a
DLT network, is equivalent to the legally binding agreement between these parties; because
the factual consequences they lead to are, or are expected to be, the same. The law sensu
lato, being the underlying legal agreement and the broader context of the legislation it exists
under, become irrelevant. This ideal considers that programs uploaded to a DLT exclusively
govern the arrangements between parties; therefore, it does not have to operate under the
shadow of the law.265 Lawrence LESSIG’S famous quote “code is law” corresponds with this
modern viewpoint.266
CHAPTER 3. SMART LEGAL CONTRACT – law is law,

and code is code267
68. A different way to look at smart contracts is by acknowledging the difference

between the legal, enforceable obligations on the one hand and the tool to wholly or partially
operationalize these obligations on the other hand.268 Indeed, for the purposes of a legal
analysis, we cannot consider the non-traditional method of smart contract enforcement
consistent with the legal notion of enforcement because this is inevitably linked to
263
Ibid., 4.
264
Ibid., 4.
265
E. TJONG TJIN TAI, “Smart contracts en het recht”, NJB 2017, 92(3), 179.
266
267
G. JACCARD, “Smart Contracts and the Role of Law”, Justletter IT 2017, 23, 9.
268
74
obligations recognized by law, and, therefore based on a valid agreement.269 The tamper-
proof computer code merely imitates the law and the law remains the only source of the
binding legal nature of the agreement.270 Generally speaking, the core condition for the
establishment of a legally enforceable contract is the expressed and valid consent by parties,
the consensus on the contents of an agreement with a valid object and causa – to put it in
Belgian civil law terms. While the smart contract is a very straightforward product of human
coding activity, the contents of the actual agreement are only to be found in the meeting of
the minds of the parties involved.
Nevertheless, it cannot be denied that once a protocol has been executed through the above-
mentioned non-traditional method, it is indeed ‘enforced’ in some sense. For the lawyer, a
so executed contract does not equal or guarantee a legally enforceable contract.271 Absent
the fulfillment of the constituent conditions underpinning a valid contract, the costs arising
from the ultimate enforcement of a potentially wrongfully executed contract may surpass
the normal enforcement costs – thereby nullifying the desired cost reduction benefit of the
technology.
69. This approach resonates the most with legal professionals and practitioners since it
detaches the computer protocol from the traditional contract that precedes and justifies the
implementation of the computer code.272 The readjusted concept that resulted from this
viewpoint was first labeled a ‘smart legal contract’ by Josh STARK.273 By distinguishing
between the code and the agreement it imitates, we have the opportunity to zoom in on the
interaction – or lack thereof – between both separate concepts. The next paragraphs will
briefly discuss this interplay between law and code in relation to the operational nature of
the contractual clauses, the degree of automatability of the rights and obligations they
contain, the ability for a court body to intervene and the internal or external positioning of
269
Ibid., 6.
270
271
272
As also presented by K. VERSLYPE during his lecture “Bitcoin, Blockchain & Smart Contracts – Inleiding
voor juristen”, Ghent, 26 October 2017.
273
J. STARK, “Making Sense of Blockchain Smart Contracts”, 2016. Available at
https://www.coindesk.com/making-sense-smart-contracts/, last accessed 22 April 2018. On the notion of smart
legal contracts, see also: G. JACCARD, “Smart Contracts and the Role of Law”, Justletter IT 2017, 23, 21-24;
and, S. BLEMUS, “Law and Blockchain: a legal perspective on current regulatory trends worldwide”, RTDF
2017, 4, 13.
75
the code vis-à-vis the legal agreement, the ‘dumb’ contract.274 While the third disparity
concerns the relationship between code and courts, the others appertain to the ties between
code and traditional contract.
A. RELEVANT DISTINCTIONS
1. Operational vs. non-operational contract clauses
70. Not all clauses in a traditional legal contract are capable of being expressed in a
language that can be readily executed by a computer. Operational clauses are mainly
concerned with the actual execution of the contract – although certain remedies can also be
expressed in operational terms – while non-operational or denotational terms are important
for the legal interpretation of the entire agreement.275 The operational clauses are
characterized by their straightforward conditional logic: if X, then Y; therefore, it is possible
to translate these provisions into pure Boolean logic.276, 277 For instance, if (when) a bond
reaches its maturity date, then the issuer repays the initial investment increased with payable
interests. Non-operational clauses do not embody such logic, they determine the wider
relationship between parties.278 For instance, a clause that determines the law applicable to
the contract does not have a conditional structure. This does not exclude the possibility that
an advanced computer would be able to read the structure and understand its semantics to
produce the desired result. For instance, such computer could recognize a standard
representation stating that parties are “duly organized and validly existing under the laws of
the jurisdiction of its organization or incorporation”, and automatically consult the electronic
274
As conceptually different from the ‘smart’ contract – I have borrowed this alternative term for the
traditional, legal agreement from C. LIM, T.J. SAW and C. SARGEANT, “Smart Contracts: Bridging the Gap
Between Expectations and Reality”, blog, 2016. Available at https://www.law.ox.ac.uk/business-law-
blog/blog/2016/07/smart-contracts-bridging-gap-between-expectation-and-reality, last accessed 2 May 2018.
275
276
For an interesting proposal for a machine-readable format of legal semantics creating obligations,
permissions and rights; see: H. LAM, M. HASHIMI and B. SCOFIELD, “Enabling Reasoning with LegalRuleML”
in J. J. ALFERES, L. BERTOSSI et al. (eds), Rule Technologies: Research, Tools, and Applications: 10th
International Symposium, proceedings, 6-9 July 2016, 241-257.
277
278
Ibid., 11.
76
registers in order to determine whether the agreement can indeed be executed.279
Additionally, there is a growing interest for the subject of creating a dedicated legal
computer language that also captures such non-operational statements.280
A specific category of clauses with an absent or low degree of operational character are
provisions that contain a certain subjectivity, which is per definition incompatible with
computer language.281 The conditional logic in such clause is rendered ineffectual because
of the ambiguous way it is expressed. This is for instance the case with all obligations to
carry out a task “in good faith”, to “produce reasonable efforts”, or to be “liable for
foreseeable damages only”; which are all circumstances to be assessed through human
judgment.282, 283
71. In some instances, the triggering factor that determines the outcome of an operational
or non-operational clause relies on data which is not capable of being readily built into the
contract itself. This is notably the case for events or values that are to be discovered in the
world outside of the contract code, like interest rates, stock market data or party decisions.
The smart contract can interface with this real-world information through determinations
provided by so-called ‘oracles’, which are preselected trusted sources of data that remain
available to the contract code while it is executing.284 We will further differentiate between
objective oracles, i.e. trusted sources of objective data; and, subjective oracles, which serve
as a point of entry for some sort of human judgment into the smart contract.
279
Ibid., 12.
280
research directions”, position paper, 2016, 11. Available at " https://arxiv.org/pdf/1608.00771v2.pdf, last
281
282
Also labelled “complex evaluative rules” by E. TJONG TJIN TAI, “Formalizing contract law for smart
contracts”, Tilburg private law working paper series 2017, 5. Available at
283
Alternatively, in a far future (cfr. the aforementioned Gartner Hype Cycle), by an artificial intelligence
application of which the design is based on machine learning technology and big data, by processing a
collection of human decision-making and experience of how to deal with various potential problems during
the life of a contract. See S. HASSAN and P. DE FILIPPI, “The expansion of algorithmic governance: from Code
is Law to Law is Code” Field Actions Science Reports 2017, 90; also mentioned by K. VERSLYPE during his
lecture “Bitcoin, Blockchain & Smart Contracts – Inleiding voor juristen”, Ghent, 26 October 2017.
284
accessed 24 April 2018; and, ISDA and LINKLATERS, “Whitepaper: Smart Contracts and Distributed Ledger –
A Legal Perspective”, 2017, 18. Available at https://www.isda.org/a/6EKDE/smart-contracts-and-distributed-
ledger-a-legal-perspective.pdf, last accessed 24 April 2018.
77
2. Automatable vs. non-automatable contractual rights and obligations
72. In order to allow the smart legal contract to imitate the content of a traditional
contract, it is essential that at least some contents of it, mainly the obligations incorporated
therein, are automatable in one way or another. Automatization is understood in a way that
the execution of the contents can be achieved by means of one or more computers, or by any
electronic means in general.285 For instance, a contractual obligation for a person to
personally carry out a task of a physical nature, like painting a wall, will not be automatable;
it may however become automatable in a future of advanced robotics and Internet of Things.
However, the majority of common obligations in a contract of a financial nature are indeed
automatable by a computer since many are already initiated by computer systems. Examples
are netting transactions between banks or bilateral clearing in DLT-based security trade
lifecycle, as presented earlier in this research.286
Even though certain contractual provisions might be automatable, it can be undesirable to

automatically trigger the contractually provided consequences for a given, contractually
specified event – this is especially relevant regarding certain rights emerging from the
contract. For instance, the contract may provide for early termination in the event of default
by any of the parties. However, the choice to actually execute this early termination should
remain at the discretion of the non-defaulting party, whose decision may well be influenced
by commercial or other considerations external to the agreement.287
3. Strong vs. weak smart contract code
73. A further important distinction was presented by Max RASKIN. It expresses the
degree of rigidity affiliated with the contract’s relevant enforcement method. The perfect
implementation of the above-mentioned non-traditional enforcement through integral
dependence on blockchain technology is regarded a strong smart contract because such a
configuration brings about prohibitive costs for a court of law or arbitration institute to alter
285
accessed 24 April 2018; and, E. TJONG TJIN TAI, “Smart contracts en het recht”, NJB 2017, 92(3), 179.
286
Supra, 59-62.
287
78
the contract post factuum.288, 289 A smart contract setup which somehow allows for a court
to influence the contract with relative ease and without incurring disproportional costs, after
it has been initiated, is labeled a weak smart contract.290 One way of constructing such state
is by establishing a subjective oracle of which the source has a certain degree of
discriminatory decision power; or which specifically enables for human judgment by an
external party.291 Likewise, all agreements that contain obligations regarding some sort of
alterable, non-computable behavior fall within this category, i.e. all non-automatable
obligations.292
4. External vs. internal smart contract models
74. Lastly, the smart legal contract approach allows us to discern two arrangements with
a different degree of interconnectedness between both code and contract, represented by two
distinct implementation schemes. These schemes are named the external model and the
internal model in a recent whitepaper by the ISDA and LINKLATERS.293 Whereas in the
internal model, the smart part of the contract is part of the legal agreement; it is kept outside
the scope of the legal agreement in the external setup.294 External smart contract code is
hierarchically secondary to the legal contract terms.295
B. IN PRACTICE
75. In the following paragraphs, we will examine some important aspects of the external
and internal variety of the smart contract with greater detail in the context of a hypothetical
example. We will thereby dedicate special attention to the obstacles posed by non-
288
289
Note that a strong smart contract setup requires that the assets involved in the transaction are crypto-assets
in order to exclude real-world interference.
290
Ibid., 310.
291
2015, 50. Available at " https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last accessed 7 March
2017.
292
293
294
Ibid., 14.
295
S. BLEMUS, “Law and Blockchain: a legal perspective on current regulatory trends worldwide”, RTDF
2017, 4, 14.
79
operational and non-automatable elements in the legal contract; as well as by potential
disputes or erroneous coding.
Imagine a consumer loan agreement for the purchase of a car. At the creditor’s side of the
agreement, core common obligations are providing the funds and assessing the client’s
financial situation and creditworthiness, as well as adequately informing and advising the
client. The main obligations that bear on the borrower, on the other side, consist of repaying
the funds in monthly instalments, providing accurate and honest information to the creditor
and, usually, providing some sort of collateral. The next parts will operationalize this
agreement in an external and an internal setup in order to identify the strength and potential
problems in both scenarios.
1. External smart legal contract
a. Setup and use
76. Parties may agree in the legal agreement to partially enhance the performance of
specific contract obligations by embedding them in a piece of computer code uploaded to a
DLT platform, specifying that the code is subordinated to the terms of the primary contract.
For instance, the external smart contract may automate the release of funds by the creditor
to the lender, or to his car dealership; and it may also automate the monthly repayments by
the debtor, linked to an appointed (crypto-) bank account. Such constellation does not add a
lot of certainty for the creditor since similar contemporary systems already allow for such
automation and because it is still the lender’s responsibility to keep the connected account
sufficiently pre-funded to answer to the recurring payment obligations.296 In the hypothesis
of rapidly maturing first generation blockchain technologies, however; the smart contract
code may be able to prompt performance from the lender’s account in the ‘golden’ crypto-
asset ledger, on which the external smart contract imposes some sort of restrictions regarding
the lender’s right of free disposal. Additionally, the external smart contract may govern the
handling of the collateral or the right to withhold performance in the event of breach; for
instance, by embedding a function for the execution of some sort of charge on the debtor’s
296
The solution to block the funds in such account comes at a cost of liquidity loss. See E. TJONG TJIN TAI,
“Smart contracts en het recht”, NJB 2017, 92(3), 179.
80
car.297 Eric TJONG TJIN TAI has thought out a simple program architecture to operationalize
such mechanism.298 Note that, in the external smart contract setup, the actual loan agreement
undergoes little change compared to modern loan agreements – it exists in a human language
document that stipulates all rights and obligations but that provides for partial and specific
execution by way of smart contract, subordinated to the main legal contract.
77. An external smart contract arrangement helps the creditor to streamline its processes
and to decrease potential losses caused by operational risks, human mistakes or
administrative delays. Given the fact that the piece of smart contract code is uploaded to a
DLT, both parties can trust that the automatization of their performances is administered in
a highly secure and resilient manner; and that all individual coded obligations are carried
out in a necessarily conjunct manner. Borrowers may benefit from reduced loan costs due
to the increased certainty, clarity and efficiency in the lender’s processes.
b. Non-operational and non-automatable distortions
78. The legal loan agreement may contain non-operational elements because of the lack
of Boolean logic (“This agreement is governed by the laws of Belgium.”) or because they
give expression to a certain ambiguity (“The lender has used reasonable efforts to
recommend a loan arrangement that adequately matches the borrower’s financial situation,
objectives and demands.”). The external smart contract is supplementary to the main legal
contract; hence, non-operational clauses can just be kept in the ‘dumb’ part of the contract
and the designer of the external smart component must ensure that the code corresponds
with these provisions. A similar assessment is to be made for non-automatable elements;
they can simply be kept outside the scope of the smart contract and remain executed in the
contemporary way. Consequentially, computer code is reserved for solidly automating
straightforward and fully operational processes; and non-operational as well as non-
automatable distortions cause little to no problems for external smart contract setups.
297
M. RASKIN, “The Law and Legality of Smart Contracts”, Geo. L. Tech. Rev. 2017, 329-333.
298
E. TJONG TJIN TAI, “Formalizing contract law for smart contracts”, Tilburg private law working paper
series 2017, 7. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3038800, last accessed 8
April 2018.
81
c. Smart bugs: disputes and unwanted output
79. A number of issues may arise throughout the duration of the agreement; parties may
want to renegotiate the contract terms based on changes in circumstances, parties may
become involved in a dispute regarding the external smart contract’s execution or parties
may simply find themselves confronted with a buggy computer program of which the output
does not at all reflect the meeting of the minds behind the traditional contract. Given the
external organization of the smart component, and thus its hierarchically secondary status
vis-à-vis the main legal contract; the code must be alterable at all times after the smart
contract’s launch, to align it with the authentic contract terms. Hence, external smart contract
code must, by definition, adopt the structure of a weak smart legal contract – there must be
a way to change the blockchain in exceptional circumstances.299
80. Implementing a weak smart contract setup requires adequate care not to undermine
the smart contract’s security. There are various functions that can be pre-programmed into
a smart contract to allow for later changes or for complete rescission by the rightful actor or
actors: (1) the code can contain certain variables that are designated to be changed (e.g. the
interest rate), (2) the code can provide for the disabling of certain functions in the smart
contract, even after they were uploaded to the DLT; as well as for the addition of new ones,
and (3) the smart contract can contain an auto-destruct function.300 Such functions must be
wrapped in a conditional statement that only grants access when called by the rightful
exerciser.301 The rightful exerciser is recognized and identified by its address or public key
on the DLT, which is only accessible by means of its unique private key.302 Consequentially,
firm security measures need to be taken to safeguard the private key of an actor who can
rescind or alter the contents of the smart contract.
In order to make a smart contract qualify as weak, the tools mentioned above, and
particularly the third tool, need to be accessible for a state-backed judicial body. Indeed,
courts should be assigned some sort of oracle-like status in the weak smart contracts
299
300
B. MARINO and A. JUELS, “Setting Standards for Altering and Undoing Smart Contracts” in J. J. ALFERES,
L. BERTOSSI et al. (eds), Rule Technologies: Research, Tools, and Applications: 10th International Symposium,
proceedings, 6-9 July 2016, 158-164.
301
Ibid., 158-159.
302
Supra, 12 on the subject of cryptographic key pairs.
82
concluded under their jurisdiction.303 Such mechanism requires that (1) every jurisdiction
runs a node on the relevant smart contract platform(s), which is controlled by the
jurisdiction’s judicial body; and, (2) the external and thus necessarily weak smart contract
incorporates data on the competent jurisdiction for any technical or juridical issues with the
smart contract, as established in the traditional legal contract it relates to. The relevant
jurisdiction may be directly appointed by the statement of this node’s public key or address;
or the smart contract may rely on a trusted third-party oracle to obtain this data from,
allowing for a higher degree of flexibility.304
81. Lastly, parties may want to expand the ‘weak’ nature of their external smart contract
by appointing more private key holders besides the judicial nodes to be the rightful exerciser
of specific tools to alter or terminate the smart contract. Hence, parties can provide for
conditional modification or termination rights (e.g. the borrower can execute an early
termination upon the repayment of the outstanding funds and an automatically calculated
reinvestment fee); or for a possibility to modify or to terminate the smart contract upon
mutual agreement.305 Accordingly, the wrapping around the relevant functions can identify
certain preconditions for its execution, as well as multiple public key holders or addresses
who must act jointly to trigger the function.
2. Internal smart legal contract
a. Setup and relevance
82. The consequences of an internal setup are more extensive and so are the questions
raised concerning the smart contract’s governance. In the internal smart contract version of
our loan agreement example, the computer protocol is agreed to constitute an integral
303
304
Additionally, states may want to engage in specific conflict of law treaties to expand and align the principles
of governing law and competent jurisdiction to smart contracts; and, to create a commitment to, if necessary,
alter the smart contract on each other’s behalf in the event the competent jurisdiction does not match the in the
(smart) contract nominated jurisdiction.
305
B. MARINO and A. JUELS, “Setting Standards for Altering and Undoing Smart Contracts” in J. J. ALFERES,
L. BERTOSSI et al. (eds), Rule Technologies: Research, Tools, and Applications: 10th International Symposium,
proceedings, 6-9 July 2016, 153-154 and 156-157.
83
component of the legal contract.306 Maintaining a natural language version of the clauses
transferred to the smart contract is irrational since code and contract language have the same
legal value.307
The internal version of the smart consumer loan agreement would therefore almost entirely
exist of computer code, including the precontractual steps – a particular application will
standardize the mandatory exchange of information between client and creditor, and it will
automatically produce correct advice and judgment on the appropriate loan contract terms;
thereby minimizing future conflicts over the validity of the loan agreement.308
83. Clearly, an internal smart contract generates the benefits mentioned above, along
with an even higher degree of certainty since the auto-execution is expanded to the entire
legal contract. To the contrary, it also makes it more challenging to deal with the issues of
non-operational and non-automatable elements, as well as with code deficiencies or
potential disputes.
b. Non-operational and non-automatable distortions
i. General non-operational clauses
84. We believe that most non-operational clauses will not pose any problems in the long
run. Some of these clauses could be artificially transformed into a more operational phrasing
– it is not because a typical governing law clause is not construed in a conditional manner,
that it does not comprise any conditional logic. Other general non-operational clauses, e.g.
an entire agreement clause, could be recognized and given effect to through the use of
advanced semantic analysis; or captured by custom-made computer language.309
306
307
Ibid., 16.
308
309
Supra, 70.
84
ii. Ambiguous clauses
85. Absent any futuristic artificial intelligence and machine learning innovations, the
solution suggested above is not workable when the non-operational character of a contract
clause exists in a certain subjectivity, therefore requiring human assessment. Some authors
accept this feature as one of the smart contract’s main strengths.310 Contracting parties are
obliged to construct their legal relation in a precise manner and by doing so, they ensure the
agreement’s effective execution. The accurate literalism inherent to computer code
eliminates the potential, frustrating incapability of vague human linguistics and it decreases
the need for ex post assessment.311 Others however call attention to the utility of somewhat
subjective contract language and the flexibility it allows for.312 Ambiguity may be desirable
from a pragmatic point of view: the nuance may make it easier for both sides of a negotiation
to agree and sign a contract.313 Moreover, the idea that it is possible to entirely strip a legal
relation of all its ambiguity may be utopic, especially in the context of more complex
business dealings. 314 Contracting parties will have to decide to either insert or ban
ambiguous language by making the cost-benefit analysis based on the complexity and
economic importance of the agreement they are concluding.
Parties who wish to take advantage of such flexibility however will likely have to accept a
weak smart contract configuration. Indeed, to allow for the subjective evaluation of the
adequate fulfillment of contractual clauses or conditions, the smart contract will somehow
have to embed an oracle which obtains data from human judgment.315 As a subjective oracle
ultimately relies on alterable human behavior, any smart contract that contains such oracle
310
S. HASSAN and P. DE FILIPPI, “The expansion of algorithmic governance: from Code is Law to Law is
Code” Field Actions Science Reports 2017, 89. Available at https://ssrn.com/abstract=3117630, last accessed
3 May 2018.
311
M. RASKIN, “The Law and Legality of Smart Contracts”, Geo. L. Tech. Rev. 2017, 326; and, A. WRIGHT
and P. DE FILIPPI, “Decentralized Blockchain Technology and the Rise of Lex Cryptographia”, 2015, 24-25.
Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last accessed 7 March 2017
312
313
2017; and, ISDA and LINKLATERS, “Whitepaper: Smart Contracts and Distributed Ledger – A Legal
Perspective”, 2017 13. Available at https://www.isda.org/a/6EKDE/smart-contracts-and-distributed-ledger-a-
legal-perspective.pdf, last accessed 24 April 2018.
314
Ibid., 13.
315
series 2017, 5. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3038800, last accessed 8
April 2018.
85
is of the weak variety – that is, if the appointed individuals behind the oracle are identified
to the court.316, 317
iii. Non-automatable elements
The distinction between internal and external smart contracts is irrelevant for the issue of
non-automatable rights and obligations – actions that are not somehow initiated by a
computer preclude the implementation in a true internal smart contract. This does however
not constitute a real impediment for applications in the scope of this research, as computer-
based transactions are ubiquitous in the financial sector.
c. Smart bugs: disputes and unwanted output
86. Internal smart contracts, once uploaded to a DLT platform and launched, make it
much more difficult to deal with the potential issues throughout the duration of the contract.
Absent any hierarchical relation to the traditional agreement, the parties to an internal smart
contract have to abide with the output of the computer program and both the track of a weak
or a strong constellation is theoretically possible.
i. Strong internal smart contract
87. Parties may consent to have their contractual relations governed by an internal smart
contract without any kind of built-in recourse mechanism. There are however multiple
disincentives for this approach. In the first place, it is conceptually difficult to harmonize
the principles of an internal smart contract on the one hand with the representation of the
parties’ intentions on the other hand. Unlike the external configuration, there is no entire,
legally existing or recognized agreement to fall back on to prove the meaning of the contract
in natural language. Given the freedom of the form, the legal issue exists in the alleged
consensus between parties and not in the mere use of computer code as a form.318 Indeed,
the opacity of computer language – as it exists today – raises doubts about the validity of
316
Even if judicial bodies would not have direct access to an internal smart contract; a court could issue an
injunction vis-à-vis the identified subjective oracle in the event of dispute. Supra, 71.
317
318
86
the parties’ consent to the terms of a smart contract, and it increases the risk of erroneous
consent.319 Especially in more complex and long-term arrangements, it is highly uncertain
whether a contracting party can foresee, conceive and indeed intend all consequences of the
many possible chains of commands an internal smart contract consists of.320, 321
Consequentially, internal, strong smart contracts are conceptually very close to the computer
science approach of “code is law”, therefore outside the scope of the law. The obligations
expressed therein risk to be grounded only on computer code; while the only authentic
source of enforceable legal obligations is society and its legal normativity.322
A creative solution to bring the internal, strong smart contract within the scope of the law
nevertheless, is by regarding the computer code as a third party ruling which bindingly
shapes the contractual relations between parties; thereby transforming potentially
unconceivable contractual obligations into determinable obligations, based on objective
standards. The difference is solely artificial in nature and since the ‘third party’, being the
computer program, is in fact built by the contracting parties, it seems indefensible to
consider it an independent entity.
Additionally, even if parties would verifiably grasp the substance of the code, and therefore
it would be possible to fit a choice for a strong, internal setup into the concept of a smart
legal contract; it still seems inadvisable to enter in such protocol for meaningful transactions.
The combination of the still relatively immature technology and equally undeveloped legal
framework produces a perfect climate for scams and unexpected, destructive loopholes.323
The choice for a strong, internal smart contract would therefore put all contractually bound
accounts of crypto-assets at risk, without any recourse in order to reverse harmful results.324
319
M. GIANCASPRO, “Is a ‘smart contract’ really a smart idea? Insights from a legal perspective”, Computer
Law & Security Review 2017, 33, 831; and, E. TJONG TJIN TAI, “Smart contracts en het recht”, NJB 2017,
92(3), 181.
320
JACCARD compares the consent to launch such computer protocol to consent in the context of general terms
in click-wrap agreements, of which the validity is equally doubtful. See G. JACCARD, “Smart Contracts and
the Role of Law”, Justletter IT 2017, 23, 22.
321
322
323
See for instance A. SAMSON, “SEC accuses crypto-bank of fraud, halts $1bn initial coin offering”, Financial
Times, 30 January 2018. Available at https://www.ft.com/content/678a75dc-05cc-11e8-9650-9c0ad2d7c5b5,
last accessed 5 May 2018; and, the aforementioned hack of The DAO, supra, 16.
324
87
ii. Weak internal smart contract
88. Parties who are concerned about safeguarding the continuous congruence between
the authentic content and meaning of their agreement and the implementation in an internal
smart contract, will likely wish to embed a tool that allows for judicial recourse by providing
backstop functions to an individually nominated node.325 By doing so, parties benefit from
the full certainty of execution offered by the DLT platform and they can rest assured that
the object of their actual consent will prevail in the event of manifest fraud, clear technical
defaults or proven illegalities. Note that judicial intervention will likely be limited to
marginal review of the contract’s legitimacy, given the freedom of contract and taken in
consideration the fact that the competent authority will have to rely on only the obvious
general principles of the internal smart contract to understand the parties’ intentions.
Parties may also wish to equip the weak internal smart contract with the necessary tools to
allow for modifications by mutual agreement, as described earlier.326
C. CONCLUSION
89. Based on all previous considerations, we can display the various possible smart legal
contract configurations in the following schematic overview:
325
As explained earlier, supra, 79-80. Alternatively, parties may opt to allow for subjective remedies without
resorting to a weak smart contract by creating some sort of ad hoc arbitration tool, which summons and
incentivizes arbitrators from around the internet to assess the situation; however, this idea ultimately also relies
on the perfection of the contract code and no such system has been successfully implemented so far.
Nonetheless, it may be a pragmatic solution for a first instance review of non-urgent internal smart contract
problems, supplementing and narrowing down the role of the state-backed judicial body, which would then
serve as an appeal forum. See A. WRIGHT and P. DE FILIPPI, “Decentralized Blockchain Technology and the
Rise of Lex Cryptographia”, 2015, 50. Available at
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last accessed 7 March 2017; and, I. KAMINSKA,
“Decentralized courts and blockchains”, Financial Times, 29 April 2016. Available at
https://ftalphaville.ft.com/2016/04/29/2160502/decentralised-courts-and-blockchains/, last accessed 5 May
2018.
326
Supra, 81.
88
Strong Weak
Internal - Potentially difficult to understand. - Potentially difficult to understand.
- Questionable status as a smart legal - High degree of total execution
contract; possibly unlawful. certainty.
- Incapable of accommodating - Incapable of accommodating non-
ambiguous and non-automatable automatable elements; ambiguities may
elements. be assessed by a subjective oracle.
- High risks in the event of undesirable - Backstop judicial recourse against
consequences. manifest fraud, legally invalid contract
formation or clear bugs.
External - Logical fallacy. - Natural language document as basis.
- High degree of flexibility regarding
the automated execution.
- Flexibility to provide for non-
operational and non-automatable
elements in the (‘dumb’) legal contract.
- Broader judicial recourse against
irregularities and against all non-
conformities with the (‘dumb’) legal
contract.
90. It is clear that the strong internal smart contract is conceptually the closest to the
archetypical “code is law” philosophy of breach-resistant coded agreements between
individuals, without any need for recourse to third parties. By building a scheme which is
immune for third party interference, it is also in principle immune for the adequate
application of the law. Such configuration entails considerable risks since many agreements
inevitably need some sort of mechanism to correct mistakes that may arise during the life of
a contract, including obvious matters such as erroneous consent, fraud other illegalities.327
The strong approach seems to overlook that it is precisely these issues that contract law has
created solutions for, over centuries of aggregated knowledge and practical experience.328
327
Supra, 87.
328
series 2017, 2 and 4. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3038800, last accessed
8 April 2018.
89
Therefore, a strong internal smart contract can arguably be considered a smart legal contract;
and, if transactions in crypto-assets become more widely adopted, it is possible that states
want to protect their citizens against these contracts by resorting to drastic measures aimed
at banning strong smart contracts; by pressurizing internet service providers, human-run
online entities like search machines or DLT platforms, software developers and hardware
manufacturers.329 The potential risk of course depends on the complexity and duration of
the agreement, the (crypto-)expertise of the contracting parties and the quantity of crypto-
asset funds potentially involved.
As mentioned earlier, strong external contracts cannot exist since the external arrangement
necessitates the possibility to enforce the prevailing ‘dumb’ agreement and thus to influence
the contract code. Hence, a strong external contract is a contradictio in terminis.330
91. The weak varieties, on the other hand, combine the unique characteristics of a DLT
smart contract with the protection offered by the law; to a larger or lesser extent depending
on the choice for an internal or external arrangement of the smart contract. Parties benefit
from the resilient auto-execution and durability offered by blockchain technology, as long
as it is consistent with the actual legal agreement they wish to automate and as long as it is
not somehow legally invalid – indeed, the intention to also execute the contract outside of
these circumstances would likely not constitute ‘acting in good faith’.
Whereas the code is fully scrutinized against the entire content of the ‘dumb’ contract in the
external setup, the internal smart contract solely allows for judicial review of the validity of
the contract formation and manifest technical defects – i.e., could both parties have intended
and foreseen the contract code to produce this output?
92. In conclusion, by launching an internal or external weak smart legal contract, parties
agree to rely on the code in the first place and to shift any potential enforcement of the law
to an ex post phase.331 It is the pinnacle of freedom of contracting regarding the execution
modality, but at the same time, it is insulated against situations where a traditional contract
329
2015, 51-52 and 56. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last accessed
7 March 2017.
330
Supra, 79.
331
90
would also be legally unenforceable. In the event of future mass adoption of crypto-assets
and smart contract modalities, states may want to facilitate the ex post enforceability of
existing contract law by adopting specific rules on the setup of a weak smart contract on the
one hand and by providing template blocks of smart contract code on the other hand.
93. Jurisdictions may adopt a regulatory framework which guides the arrangement of an
individual smart contract, depending on the contracting parties and on the nature and value
of the smart contract at hand. For instance, a weak setup could be compulsory for a contract
with a converted maximum value over €300 at its conclusion, or with a term surpassing one
year. Additionally, consumer smart contracts could be restricted to an external arrangement
to ascertain the understandability and informed consent by consumers. Such rules could be
enforced through civil or even criminal liability provisions targeting identified contracting
parties and software developers.332
Such laws would favorably be complemented by a state-backed and publicly available

database of template computer protocols, compatible with the major DLT platform
programming languages or in pseudocode form. These templates should specifically address
the structure of the weak contract setup, including access for a designated node or nodes. It
may also go further in promoting the relevance of contract law in smart legal contracts by
formalizing pertinent legal provisions into code which is to be adopted in the smart contracts
in an unaltered way.333 Direct connection to the database would also support automated
updates with changes in the law.334
94. Based on these considerations, we conclude that the weak internal or external smart
contract, when complemented by the necessary state-backed measures, is the only
convincing configuration of a valid smart legal contract; which respects the source of legal
332
2017.
333
As suggested by Kristof VERLSYPE and firstly undertaken by TJONG TJIN TAI, with a template of pseudocode
for withholding performance, see E. TJONG TJIN TAI, “Formalizing contract law for smart contracts”, Tilburg
private law working paper series, 2017, 6-7. Available at
334
91
obligations and the difference between law and code, and, at the same time, embraces and
optimizes the efficiency and trust benefits promoted by the use of DLT.335
335
Partially concurring with the position taken by Kristof VERSLYPE; albeit that VERSLYPE only considers
external smart legal contract setups during his lecture “Bitcoin, Blockchain & Smart Contracts – Inleiding voor
juristen”, Ghent, 26 October 2017.
92
TITLE 5. CONCLUDING CONSIDERATIONS
95. After a deeper analysis of the various implementation schemes for financial DLT
applications, we can establish that there are many ways for this new technology to reshape
the contemporary financial landscape. The most nearby innovations appertain to how
various financial assets are held and transacted with, including currencies; and to the relation
between, or the consolidation of, traditional legal contracts and their smart, coded
counterparts.
Disruption is a popularly used term in relation to the emergent blockchain technology, and
it primarily refers to the highly decentralized applications that pursue the complete
disintermediation of the targeted sector. However, it is now clear that the technology can
become equally relevant in the context of the modernization of pre-existing infrastructures
and legal concepts, through permissioned platform configurations and weak smart contract
setups. This approach combines and reconciles existing trust, regulation and experience with
the main unique capabilities of DLT applications. Indeed, regardless of the question whether
a certain blockchain is highly decentralized and independent, or instead permissioned and
administered to a certain degree; all innovations of this kind still materialize in a context
where the society and its legal normativity prevail – especially in a densely regulated
environment like the financial sector.336 Blockchain is and will remain a regulatable
technology, by one way or another; if necessary, through the use of draconian measures on
the edge of government control abuse.337 As Max RASKIN puts it, “It is a good rule of thumb
that the entity with more guns wins”, and, “governments generally have more guns.”338
96. A more desirable scenario undoubtedly comprises supportive regulation that aims to
maximally turn the implementation of DTL platforms to the advantage of the entire financial
industry, and, by extension, society as a whole. To achieve this objective, regulatory
adaptations specifically addressing this new technology will need to show up on the horizon
in the near future. Many private DLT initiatives come on stage on a daily basis and some of
336
G. JACCARD, “Smart contracts and the Role of Law”, Justletter IT 2017, 23, 9.
337
2015, 51-53. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last accessed 7
March 2017.
338
93
them are based on the false assumption of a legal vacuum. The proposed adjustments will
be built on the present-day legal framework and need not to amount to a whole new domain
of law.339
At the outset of this endeavor, policy makers worldwide, especially in dominant financial
markets, should bear the following key points in mind. Supportive new DLT laws should …
- … appreciate the diversity of possible blockchain configurations and the

consequential differences in their governance and in the suitable regulatory
approach; and,
- … provide for regular re-evaluations as this diversity is continuously growing,
especially with the advent of consensus mechanisms that may drastically extend
the blockchain’s capacities; and,
- … be pragmatic, strict and active where necessary, to deal with outright scams
and manifestly irresponsible decentralized network platforms; and,340
- … also recognize the importance of the protection of individuals with an
information deficit (e.g. consumers), regarding the concrete differences between
highly decentralized crypto-assets and their contemporary counterparts as well
as in the context of strong internal smart contracts; and,
- … be established on international cooperation and harmonized regulatory
standards; and,341
- … leave a certain degree of freedom for ongoing technological creativity,
possibly in a controlled environment to limit risk exposure, depending on the
scale of the project.342
97. Undoubtedly, intriguing times are ahead of us as both the technology and the law
mature together, until maybe, finally a point is reached where both are sufficiently developed
and reliable for the financial sector and for society to abandon or drastically review the
legacy systems and concepts. For the time being, however, whether and when this vision
will materialize is an open question.
339
A new domain called the Lex Cryptographia, supra, 4.
340
E.g. by means of the more drastic measures mentioned supra, 90.
341
With regard to supervision, authentication, applicable law and jurisdiction. See supra, 38; and P. PAECH,
“The Governance of Blockchain Financial Networks”, Mod. L. Rev. 2017, 1110.
342
See for instance the regulatory sandbox approach, supra, 58.
94
BIBLIOGRAPHY
Legislation and legislative documents
European Union
Regulation (EU) No 648/2012 of the European Parliament and of the Council of 4 July 2012
on OTC derivatives, central counterparties and trade repositories.
Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June
2013 on prudential requirements for credit institutions and investment firms.
Regulation (EU) No 596/2014 of the European Parliament and of the Council of 16 April
2014 on market abuse.
Regulation (EU) No 909/2014 of the European Parliament and of the Council of 23 July
2014 on improving securities settlement in the European Union and on central securities
depositories.
Regulation (EU) 2015/848 of the European Parliament and of the Council of 20 May 2015
on insolvency proceedings.
Regulation (EU) 2017/1129 of the European Parliament and of the Council of 14 June 2017
on the prospectus to be published when securities are offered to the public or admitted to
trading on a regulated market.
Second Council Directive 89/646/EEC of 15 December 1989 on the coordination of laws,

regulations and administrative provisions relating to the taking up and pursuit of the business
of credit institutions.
Directive 98/26/EC of the European Parliament and of the Council of 19 May 1998 on
settlement finality in payment and securities settlement systems.
Directive 2009/65/EC of the European Parliament and of the Council of 13 July 2009 on the
coordination of laws, regulations and administrative provisions relating to undertakings for
collective investment in transferable securities.
Directive 2009/110/EC of the European Parliament and of the Council of 16 September

2009 on the taking up, pursuit and prudential supervision of the business of electronic money
institutions.
95
Directive 2011/61/EU of the European Parliament and of the Council of 8 June 2011 on
Alternative Investment Fund Managers.
Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on
access to the activity of credit institutions and the prudential supervision of credit institutions
and investment firms.
Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on
markets in financial instruments.
Directive (EU) 2015/849 of the European Parliament and of the Council of 20 May 2015 on
the prevention of the use of the financial system for the purposes of money laundering or
terrorist financing.
Directive (EU) 2015/2366 of the European Parliament and of the Council of 25 November
2015 on payment services in the internal market.
Conventions
Convention of 5 July 2006 on the Law Applicable to Certain Rights in Respect of Securities
held with an Intermediary.
France
Ordonnance n° 2017-1674 du 8 décembre 2017 relative à l'utilisation d'un dispositif

d'enregistrement électronique partagé pour la représentation et la transmission de titres
financiers. Available at
https://www.legifrance.gouv.fr/affichTexte.do?cidTexte=JORFTEXT000036171908, last
accessed 1 May 2018.
Guidance papers and statements
ECB, “Virtual currency schemes”, report, 2012. Available at

http://www.ecb.europa.eu/pub/pdf/other/virtualcurrencyschemes201210en.pdf, last
ECB, “Virtual currency schemes – a further analysis”, report, 2015. Available at

https://www.ecb.europa.eu/pub/pdf/other/virtualcurrencyschemesen.pdf, last accessed 15
April 2018.
96
ESMA, “ESMA alerts firms involved in Initial Coin Offerings (ICOs) to the need to meet
relevant regulatory requirements”, statement, 13 November 2017. Available at
https://www.esma.europa.eu/sites/default/files/library/esma50-157-
828_ico_statement_firms.pdf, last accessed 7 April 2018.
ESMA, “The distributed ledger technology applied to securities markets”, report, 2017.
Available at https://www.esma.europa.eu/sites/default/files/library/dlt_report_-_esma50-
1121423017-285.pdf, last accessed 7 April 2018.
ESMA, “ESMA Assesses usefulness of distributed ledger technologies”, press release, 2

June 2016. Available at https://www.esma.europa.eu/press-news/esma-news/esma-
assesses-usefulness-distributed-ledger-technologies, last accessed 9 April 2018.
MONETARY AUTHORITY OF SINGAPORE, “Fintech Regulatory Sandbox Guidelines”, 2016.

Available at
http://www.mas.gov.sg/~/media/Smart%20Financial%20Centre/Sandbox/FinTech%20Reg
ulatory%20Sandbox%20Guidelines.pdf, last accessed 20 April 2018.
MONETARY AUTHORITY OF SINGAPORE, “MAS clarifies regulatory position on the offer of

digital tokens in Singapore”, press release, 1 August 2017. Available at
http://www.mas.gov.sg/News-and-Publications/Media-Releases/2017/MAS-clarifies-
regulatory-position-on-the-offer-of-digital-tokens-in-Singapore.aspx, last accessed 16 April
2018.
Case law
ECJ 22 October 2015, Skatteverket v David Hedqvist, C-264/14, EU:C:2015:718.
Literature
Books and contributions
DRESCHER, D., Blockchain Basics: A Non-Technical Introduction in 25 Steps, New York,

Springer Science+Business Media, 2017.
HAUBEN, C., “Bitcoin en EU-recht: de virtuele vreemde eend in de bijt” in STORME, M.E.
and HELSEN, F. (eds), Innovatie en disruptie in het economisch recht, Antwerpen,
Intersentia, 2017.
97
LAM, H., HASHIMI, M. and SCOFIELD, B., “Enabling Reasoning with LegalRuleML” in
ALFERES, J.J., BERTOSSI, L. et al. (eds), Rule Technologies: Research, Tools, and
Applications: 10th International Symposium, proceedings, 6-9 July 2016.
LESSIG, L., Code version 2.0, New York, Basic Books, 2006. Available at
MARINO, B. and JUELS, A., “Setting Standards for Altering and Undoing Smart Contracts”
in ALFERES, J.J., BERTOSSI, L. et al. (eds), Rule Technologies: Research, Tools, and
Applications: 10th International Symposium, proceedings, 6-9 July 2016.
PETERS, G.W. and PANAYI, E., “Understanding Modern Banking Ledgers Through
Blockchain Technologies: Future of Transaction Processing and Smart Contracts on the
Internet of Money” in TASCA, P, ASTE, T. et al. (eds), Banking Beyond Banks and Money.
New Economic Windows, Cham, Springer, 2016.
PROCTOR, C., Mann on the Legal aspects of money, Oxford, Oxford University Press, 2012,
10.
VERHAERT, R., Blockchain, de overheid aan of in de ketting?, Brugge, Vanden Broele, 2016.
VERHELST, J.L., “Zijn cryptomunten munten? Een analyse van Bitcoin” in STORME, M.E.
and HELSEN, F. (eds), Innovatie en disruptie in het economisch recht, Antwerpen,
Intersentia, 2017.
Journals and online research papers
ATHANASSIOU, P., “Impact of digital innovation on the processing of electronic payments

and contracting: an overview of legal risks”, ECB Legal Working Paper Series 2017.
Available at
https://www.ecb.europa.eu/pub/pdf/scplps/ecb.lwp16.en.pdf?344b9327fec917bd7a8fd708
64a94f6e, last accessed 15 April 2018.
ATHEY, S., PARASHKEVOV, I., SARUKKAI V. and J. XIA, “Bitcoin Pricing, Adoption, and
Usage: Theory and Evidence”, working paper, 2016. Available at
https://www.gsb.stanford.edu/faculty-research/working-papers/bitcoin-pricing-adoption-
usage-theory-evidence, last accessed 14 April 2018.
BLEMUS, S., “Law and Blockchain: a legal perspective on current regulatory trends
worldwide”, RTDF 2017, 4.
98
CAYTAS, J.D., “Developing Blockchain Real-Time Clearing and Settlement in the EU, U.S.
and Globally”, Columbia Journal of European Law: Preliminary Reference 2016.
CLACK, C.D., BAKSHI, V.A. and BRAINE, L., “Smart Contract Templates: foundations,
design landscape and research directions”, position paper, 2016. Available at
https://arxiv.org/pdf/1608.00771v2.pdf, last accessed 24 April 2018.
CLAESSENS, S. and KODRES, L., “The Regulatory Responses to the Global Financial Crisis:
Some Uncomfortable Questions”, IMF Working Paper 2014. Available at
https://www.imf.org/external/pubs/ft/wp/2014/wp1446.pdf, last accessed 16 April 2018.
GIANCASPRO, M., “Is a ‘smart contract’ really a smart idea? Insights from a legal
perspective”, Computer Law & Security Review 2017, 33, 825-835.
GIKAY, A.A., “Regulating decentralized cryptocurrencies under payment services law:

lessons from European Union law”, Journal of Law, Technology & the Internet 2018, 9.
HACKER, P. and THOMALE, C., “Crypto-Securities Regulation: ICOs, Token Sales and
Cryptocurrencies under EU Financial Law”, 2017. Available at
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3075820, last accessed 12 February
2018.
HASSAN, S. and DE FILIPPI, P., “The expansion of algorithmic governance: from Code is
Law to Law is Code” Field Actions Science Reports 2017.
JACCARD, G., “Smart Contracts and the Role of Law”, Justletter IT 2017, 23, 7.
JENSEN, M. and MECKLING, W., “Theory of the Firm: Managerial Behavior, Agency Costs
and Ownership Structure”, Journal of Financial economics 1976, 3(4), 305-360.
KARAINDROU, E., “Distributed ledger technology and the future of payment services”, 2017.
Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3078523, last accessed 10
March 2018.
LEE, L., “New Kids on the Blockchain: How Bitcoin’s Technology Could Reinvent the
Stock Market”, Hastings Business Law Journal 2016, 12(2), 81-132.
MANDENG, O. J., “Cryptocurrencies, monetary stability and regulation: Germany’s

nineteenth century private banks of issue”, LSE Institute of Global Affairs 2018. Available
at http://www.lse.ac.uk/iga/assets/documents/research-and-publications/LSE-IGA-WP-5-
2018-Ousmene-Mandeng.pdf, last accessed 11 March 2018.
99
MICHELER E. and VON DER HEYDE, L., “Holding, clearing and settling securities through
blockchain/distributed ledger technology: creating an efficient system by empowering
investors”, Journal of International Banking & Financial Law 2016, 652-656.
MILLS, D., WANG, K. et al., “Distributed ledger technology in payments, clearing, and
settlement”, Finance and Economics Discussion Series 2016. Available at
https://www.federalreserve.gov/econresdata/feds/2016/files/2016095pap.pdf, last accessed
17 April 2018.
MOONEY, C.W., “Beyond negotiability: a new model for transfer and pledge of interests in
securities controlled by intermediaries”, Cardozo L. Rev. 1990, 12.
NABILOU, H. and PRÜM, A., “Ignorance, Debt and Cryptocurrencies: The Old and the New
in the Law and Economics of Concurrent Currencies”, 2018. Available at
PAECH, P., “Securities, intermediation and the blockchain: an inevitable choice between
liquidity and legal certainty?”, Unif. L. Rev. 2016, 21, 612-639.
PAECH, P., “The Governance of Blockchain Financial Networks”, Mod. L. Rev. 2017, 1073-
1110.
RASKIN, M., “The Law and Legality of Smart Contracts”, Geo. L. Tech. Rev. 2017, 305-341.
SZABO, N., “Formalizing and Securing Relationships on Public Networks”, First Monday
1997.
THÉVENOZ, L., “Intermediated Securities, Legal Risk, and the International Harmonisation
of Commercial Law”, Stan. J.L. Bus. & Fin. 2008, 3.
TJONG TJIN TAI, E., “Formalizing contract law for smart contracts”, Tilburg private law
working paper series 2017. Available at
TJONG TJIN TAI, E., “Smart contracts en het recht”, NJB 2017, 92(3), 176-183.
VAN DER ELST, C. and LAFARRE, A., “Blockchain and the 21st Century Annual General
Meeting”, European Company Law Journal 2017, 4, 167-176.
WRIGHT, A. and DE FILIPPI, P., “Decentralized Blockchain Technology and the Rise of Lex
Cryptographia”, 2015. Available at
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2580664, last accessed 7 March 2017.
100
YERMACK, D., “Is Bitcoin a real currency? An economic appraisal”, working paper,
National Bureau of Economic Research Working Paper Series 2013. Available at
http://www.nber.org/papers/w19747, last accessed 14 March 2018.
Whitepapers, blog articles and various electronic sources
DUTCHCHAIN, “The real value of bitcoin and crypto currency technology - The Blockchain
explained”, video, 2014. Available at https://www.youtube.com/watch?v=YIVAluSL9SU,
last accessed 6 March 2017.
FRIEDL, S., “An Illustrated Guide to Cryptographic Hashes”, 2005. Available at

http://www.unixwiz.net/techtips/iguide-crypto-hashes.html, last accessed 15 April 2018.
HEARN, M., “Corda: A distributed ledger”, whitepaper, 2016. Available at

https://docs.corda.net/_static/corda-technical-whitepaper.pdf, last accessed 5 May 2018.
ISDA and LINKLATERS, “Whitepaper: Smart Contracts and Distributed Ledger – A Legal
Perspective”, 2017. Available at https://www.isda.org/a/6EKDE/smart-contracts-and-
distributed-ledger-a-legal-perspective.pdf, last accessed 24 April 2018.
LANTMÄTERIET et al., “The Land Registry in the blockchain – testbed”, report, 2017.
Available at https://chromaway.com/papers/Blockchain_Landregistry_Report_2017.pdf,
last accessed 9 April 2017.
LIM, C., SAW, T.J., and SARGEANT, C., “Smart Contracts: Bridging the Gap Between
Expectations and Reality”, blog, 2016. Available at https://www.law.ox.ac.uk/business-law-
blog/blog/2016/07/smart-contracts-bridging-gap-between-expectation-and-reality, last
accessed 2 May 2018.
NAKAMOTO, S., “Bitcoin: A Peer-to-Peer Electronic Cash System”, 2008. Available at

R3 LIMITED, “Writing a contract”, 2018. Available at https://docs.corda.net/tutorial-

contract.html, last accessed 26 April 2018.
SANTANDER, INNOVENTURES et al., “The Fintech 2.0 Paper: rebooting financial services”,
2016. Available at http://santanderinnoventures.com/wp-content/uploads/2015/06/The-
Fintech-2-0-Paper.pdf, last accessed 17 April 2018.
SCHWARTZ, D., YOUNGS, N. and BRITTO, A., “The Ripple Protocol Consensus Algorithm”,
Ripple Labs Inc 2014. Available at
https://ripple.com/files/ripple_consensus_whitepaper.pdf, last accessed 15 April 2018.
101
STARK, J., “Making Sense of Blockchain Smart Contracts”, 2016. Available at
https://www.coindesk.com/making-sense-smart-contracts/, last accessed 22 April 2018.
SWANSON, T., “Consensus-as-a-service: a brief report on the emergence of permissioned,

distributed ledger systems”, 2015. Available at http://www.ofnumbers.com/wp-
content/uploads/2015/04/Permissioned-distributed-ledgers.pdf, last accessed 7 April 2018.
SWIFT, “gpi Nostro Proof of Concept”, interim report, 2017. Available at

https://www.swift.com/news-events/press-releases/swift-tests-show-blockchain-has-
potential-for-global-liquidity-optimisation, last accessed 15 April 2018.
SWIFT, “SWIFT on distributed ledger technologies”, position paper, 2016. Available at

https://www.swift.com/resource/swift-distributed-ledger-technologies, last accessed 15
April 2018.
SZABO, N., “A Formal Language for Analyzing Contracts”, 2002. Available at

http://www.fon.hum.uva.nl/rob/Courses/InformationInSpeech/CDROM/Literature/LOTwi
nterschool2006/szabo.best.vwh.net/contractlanguage.html, last accessed 24 April 2018.
SZABO, N., “Smart Contracts”, 1994. Available at

http://www.fon.hum.uva.nl/rob/Courses/InformationInSpeech/CDROM/Literature/LOTwi
nterschool2006/szabo.best.vwh.net/smart.contracts.html, last accessed 24 April 2018.
WESTPAC BANKING, “Blockchain Demystified”, video, 2016. Available at

https://youtu.be/LdOcXXB48fI, last accessed 6 March 2017.
X., “Gartner Hype Cycle for Emerging Technologies”, 2017. Available at

https://blogs.gartner.com/smarterwithgartner/files/2017/08/Emerging-Technology-Hype-
Cycle-for-2017_Infographic_R6A.jpg, last accessed 20 February 2018.
Colloquia
MAIJOOR, S., “Regulatory measures to prevent another crisis”, transcript, 19 November

2014. Available at https://www.esma.europa.eu/sites/default/files/library/2015/11/2014-
1382.pdf, last accessed 22 April 2018.
MARCKX, T., “Blockchain: disrupting markets & governments”, Ghent, 25 October 2017.
MUYLDERMANS, C., ZHANG, J. and WAUTERS, N., “Becon Blockchain Conference”,
Brussels, 8 March 2017.
102
RICHELLE, J. and BERGER, P., “Studiedagen Instituut Financieel Recht: Fintech”, Brussels,
16 January 2018.
VERSLYPE, K., “Bitcoin, Blockchain & Smart Contracts – Inleiding voor juristen”, Ghent,
26 October 2017.
Newspaper articles
ARNOLD, M., “Five ways banks are using blockchain”, Financial Times, 16 October 2017.
Available at https://www.ft.com/content/615b3bd8-97a9-11e7-a652-cde3f882dd7b, last
CHAVEZ-DREYFUSS, G., “Sweden tests blockchain technology for registry”, Reuters, 16

June 2016. Available at http://www.reuters.com/article/us-sweden-blockchain-
idUSKCN0Z22KV, last accessed 10 March 2017.
HERN, A., “Bitcoin mining consumers more electricity a year than Ireland”, The Guardian,
27 November 2017. Available at
https://www.theguardian.com/technology/2017/nov/27/bitcoin-mining-consumes-
electricity-ireland, last accessed 8 March 2018.
JEMIMA, K., “Blockchain insiders tell us why we don’t need blockchain”, Financial Times,
2 May 2018. Available at
https://ftalphaville.ft.com/2018/05/02/1525253799000/Blockchain-insiders-tell-us-why-
we-don-t-need-blockchain/, last accessed 2 May 2018.
KAMINSKA, I. and MURPHY, P., “Bitcoin’s surge fuels fears of asset bubble”, Financial
Times, 14 May 2017. Available at https://www.ft.com/content/ce3ef54e-371b-11e7-bce4-
9023f8c0fd2e, last accessed 20 April 2018.
KAMINSKA, I., “Decentralized courts and blockchains”, Financial Times, 29 April 2016.
Available at https://ftalphaville.ft.com/2016/04/29/2160502/decentralised-courts-and-
blockchains/, last accessed 5 May 2018.
MURPHY, H. and STAFFORD, P., “Blockchain explainer: a revolution in its infancy”,

Financial Times, 2 February 2018. Available at https://www.ft.com/content/6c707162-ffb1-
11e7-9650-9c0ad2d7c5b5, last accessed 20 April 2018.
MURPHY, H. and THOMPSON, B., “Law firms look to capitalise on initial coin offering
boom”, Financial Times, 26 March 2018. Available at
https://www.ft.com/content/2ae9154c-1d56-11e8-aaca-4574d7dabfb6, last accessed 20
April 2018.
103
SAMSON, A., “SEC accuses crypto-bank of fraud, halts $1bn initial coin offering”, Financial
Times, 30 January 2018. Available at https://www.ft.com/content/678a75dc-05cc-11e8-
9650-9c0ad2d7c5b5, last accessed 5 May 2018.
SKINNER, C., “Will the blockchain replace SWIFT?”, American Banker, 8 March 2016.
Available at https://www.americanbanker.com/opinion/will-the-blockchain-replace-swift,
last accessed 16 April 2018.
WEINLAND, D. and LUCAS, L., “Asian investors flock to initial coin offerings despite bans”,
Financial Times, 14 November 2017. Available at https://www.ft.com/content/05b2748c-
c903-11e7-aa33-c63fdc9b8c6c, last accessed 20 April 2018.
104

Blockchain Technology and Smart Contract PDF

Uploaded by

Copyright:

Available Formats

Blockchain Technology and Smart Contract PDF

Uploaded by

Document Information

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

Blockchain Technology and Smart Contract PDF

Uploaded by

Copyright:

Available Formats

BLOCKCHAIN TECHNOLOGY AND

SMART CONTRACTS FROM A

BLOCKCHAIN TECHNOLOGY IN FINANCIAL TRANSACTIONS:

Supervisor: Prof. Dr. Michel TISON

Academic year: 2017 - 2018

ABSTRACT (ENGLISH AND DUTCH) .................................................................................................. v

TABLE OF CONTENTS ........................................................................................................................ vii

TITLE 2. BLOCKCHAIN EXPLAINED ............................................................................................... 15

A. INFORMATION TECHNOLOGY ............................................................................................... 15

1. Network access ......................................................................................................................... 18

2. Stages of development .............................................................................................................. 23

3. Consensus mechanisms ............................................................................................................. 26

C. DLT APPLICATIONS ................................................................................................................. 29

CHAPTER 1. CRYPTOCURRENCIES ................................................................................................ 31

A. HIGHLY DECENTRALIZED AND PUBLIC – electronic cash ................................................... 32

1. Qualification by an economist ................................................................................................... 32

2. Qualification by a lawyer .......................................................................................................... 34

B. PERMISSIONED AND PRIVATE – blockchain-based interbank payment solutions .................... 38

1. Interbank financial communication network .............................................................................. 39

2. Interbank money remittance network ......................................................................................... 41

b. Blockchain as an entity ......................................................................................................... 45

A. THE MODERN SECURITIES MARKET – intermediaries, origins and problems ........................ 50

1. Trade lifecycle .......................................................................................................................... 50

3. Deficiencies of the intermediated infrastructure ......................................................................... 54

4. The blockchain alternatives ....................................................................................................... 55

B. HIGHLY DECENTRALIZED AND PUBLIC – token sales, ICOs and … ..................................... 56

1. vs. cryptocurrencies? ................................................................................................................. 56

a. Issuance and market infrastructure......................................................................................... 59

C. PERMISSIONED AND PRIVATE – blockchain-based interbank securities clearing and settlement

TITLE 4. SMART CONTRACTS .......................................................................................................... 71

CHAPTER 2. CODERS’ APPROACH – code is law............................................................................. 73

A. RELEVANT DISTINCTIONS ..................................................................................................... 76

1. Operational vs. non-operational contract clauses ........................................................................ 76

2. Automatable vs. non-automatable contractual rights and obligations .......................................... 78

3. Strong vs. weak smart contract code .......................................................................................... 78

4. External vs. internal smart contract models ................................................................................ 79

1. External smart legal contract ..................................................................................................... 80

a. Setup and use ........................................................................................................................ 80

b. Non-operational and non-automatable distortions .................................................................. 81

c. Smart bugs: disputes and unwanted output............................................................................. 82

2. Internal smart legal contract ...................................................................................................... 83

a. Setup and relevance .............................................................................................................. 83

b. Non-operational and non-automatable distortions .................................................................. 84

i. General non-operational clauses ....................................................................................... 84

iii. Non-automatable elements............................................................................................... 86

c. Smart bugs: disputes and unwanted output ............................................................................. 86

i. Strong internal smart contract ........................................................................................... 86

ii. Weak internal smart contract ............................................................................................ 88

TITLE 5. CONCLUDING CONSIDERATIONS ................................................................................... 93

6. A blockchain network is in fact nothing but a collective, chronological and public

20. The aforementioned proof-of-work procedure undoubtedly is the most recognized

While proof-of-stake is not as energy-costly as proof-of-work, it also features some major

Documenting a supply chain – An interesting private initiative is Modum, a Switzerland-

27. Financial crypto-assets consist of both cryptocurrencies and cryptosecurities. We

29. From an economical perspective, money is defined by the following functions it

- a measure of value and a unit of account;

B. PERMISSIONED AND PRIVATE – blockchain-based interbank

33. An approach fundamentally different from peer-to-peer transferable

1. Interbank financial communication network