CRYPTOASSET VALUATION

Identifying the variables of analysis

Working Report v1.0

October 2018

Kary Bheemaiah Alexis Collomb

This report has been written within the Blockchain Perspectives

Joint Research Initiative, which is hosted by Institut Louis
Bachelier with the support of BNP Paribas and CDC Recherche

ACKNOWLEDGMENTS

We would like to thank Anuj Das Gupta (Chief Research Officer at Stratumn) and
Hubert Ritzdorf (CTO at ChainSecurity) for their valuable technical inputs regarding
Smart Contracts. Special mention also goes to Stefan Leins, the author of ‘Stories of
Capitalism: Inside the Role of Financial Analysts’, for his valuable contribution on how
securities are analysed today, and to the team at Autonomous NEXT for providing data
and images from their Crypto Utopia report, which have been republished in this
document.

ABOUT THE AUTHORS

Kary Bheemaiah: Kary is the Director of Research at Uchange, an Associate Research

Scientist at Cambridge Judge Business School and Senior Fellow at École des Ponts
Business School. He is the author of the book, ‘The Blockchain Alternative: Rethinking
Macroeconomic Policy and Economic Theory’ (2017) and his articles on Blockchain and
Fintech have been published on Wired, Harvard Business Review, World Economic
Forum.

www.linkedin.com/in/karybheemaiah/

Alexis Collomb: Alexis is Professor of Finance at Cnam, Paris, where he also heads
the economics, finance and insurance department. He is a scientific co-director of the
Blockchain Perspectives Joint Research Initiative (BPJRI). As a fintech and insurtech
devotee, he writes and lectures regularly on blockchain technology and cryptoassets.
He is also a scientific adviser of PSL/Mines ParisTech’s IHEIE, a leading international
program focused on innovation and entrepreneurship and of Labex Refi, a think tank
dedicated to financial regulation.

www.linkedin.com/in/ alexis-collomb-b2154336/

FOREWORD

It is quite clear that if we look at what has been happening in the ICO space over the
last two years, there has been a mix of great, good, bad, and sometimes very ugly...
In general, valuing crypto-assets is a complex issue. While many different token
classifications have been provided, they all seem to have one common denominator :
tokens are usually hybrid objects with various features—utility, security, etc. Hence
valuing them properly is hard on at least three grounds : firstly, the projects these
tokens support are more often than not very ‘early stage’ ; secondly, understanding
exactly how these hybrid objects will create value, and the non-linearities in the
process, is a difficult exercise ; thirdly, correctly anticipating how their overall
ecosystem will unfold adds an additional degree of complexity. If it is already hard
enough to properly estimate the share value of a ‘traditional’ product-oriented
startup at the seed funding stage, it is easy to figure out that valuing ICO tokens is
even harder. As for traditional startups, many analysts will take shortcuts and focus
on one or two key aspects such as the whitepaper or the team.

To take up this challenge and help us structure our thinking around these questions,
we have asked Kary Bheemaiah to look into this issue of how to properly value
crypto-assets… Could we come up with a universal model to price ICO tokens ? Could
we operate—as is usually done in the financial sphere—by either building discounted
cash flows models or using ‘comparable analysis’ ? How should we organize the
different value components and prioritize them ?

Those who have entered the fray of tokenomics will know how complex these issues
can be and by no means, this report claims to have found the answers. As its first
version is about to be published, it seems that the— irrational, many would say—
exuberance that has fueled the ICO trend—especially at the end of 2017 and
beginning of 2018—has been abating. As expectations are adjusting and the overall
market is maturing, we hope this report will be helpful in helping entrepreneurs,
investors, and researchers alike, to identify the variables and value drivers that
matter—a first stepping stone not to be missed.

Enjoy the read !

Alexis Collomb
Scientific co-director
Blockchain Perspectives Joint Research Initiative

TABLE OF CONTENTS
Executive Summary 6

7
Introduction

12
PART 1: Current State of Token Valuation

1.1: Framing the Problem 12

1.1.1: What does this mean for the future of this

16
investment vehicle?

1.2: Understanding Token Valuation Methods 18

1.2.1: From Stocks to Tokens 19

1.3: Review of Current Token Valuation methods 22

1.3.1: Store of Value Methodology 23

1.3.2: Token Velocity Methodology 24

1.3.3:Crypto J-Curve Methodology 26

1.3.4: Network Value-to-Transaction Ratio 28

PART 2: Variables for building a Token Valuation

31
Methodology

2.1: The issue with the current definition of Technical

31
Analysis

2.2: Variables for Fundamental and Technical Analysis 34

2.3: Detailed analysis of specific Technical Variables 42

2.3.1: Smart Contracts – Code quality and

42
governance of token supply

2.3.1.1: Key Variables for Integrity and

44
Security of Smart Contracts

2.3.2: Level of Decentralization 51

2.3.3: Network Size 53

Conclusion 55

Appendices 56

Appendix 1 56

Appendix 2 57

58
Appendix 3

Bibliography 61

EXECUTIVE SUMMARY

Since the launch of Token Sales or ICOs about the Bias-Variance Trade-off
in the past few years, the notion of which states that complex models tend
developing valuation methodologies to be overfitted, i.e.: they work well on
that can accurately ascertain the future past datasets but poorly on future
of a token’s price has become a subject datasets. This last issue comes back full
of increasing debate, with some experts circle to the first problem of lack of
attempting to retrofit stock valuation empirical evidence and can cause
models in the hope of creating accurate spurious correlations. As a result of
token price prediction models. these issues, investors wanting to
deploy capital in token projects are at a
While these efforts are laudable, and loss as there is no reference framework.
necessary, they suffer from a few flaws:
Firstly, there is a lack of empirical The purpose of this report is therefore
analysis – making any kind of prediction to start at the basics, or more precisely,
model requires rigorous empirical the variables. Having conducted
evidence. As the token market is still extensive research we provide the
nascent we currently lack the necessary reader with a list of variables that need
data to test these models. This issue is to be considered when analysing a
further compounded by the fact that token project. Readers of this report
there is a significant degree of diversity who are interested in developing a
in the token space (work tokens, utility valuation methodology must realise
tokens, asset-backed tokens, etc.). that there is currently no universal
token valuation method. Instead they
This leads us to the second problem, need to follow a modular approach in
which is a lack of consistency in the type which key variables and analysis
of data being used – how are the fields methods need to be selected based on
defined?, where is the data coming the kind of token or cryptocurrency they
from? As tokens have currency-like are interested in evaluating. Thus, our
properties along with functional goal is to provide the variables that
objectives, a clear definition of the type need to be taken into consideration. We
of data being used is crucial to are cognizant of the fact that to truly
developing sensible valuation models. develop a valuation method, we require
to determine the adoption curve of the
Thirdly, there is a large amount of
business model behind a token sale. As
assumptions being made, the key one
most of the ICOs are still in a phase of
being that a model or a formula that is
development, we currently lack the
used for stock valuation can be applied
data necessary to build such a model.
to a new asset class which has very
But by reviewing the models being used
different properties. Very few of the
and by identifying the variables that are
token valuation models being built are
key to understanding the feasibility of a
able to explain the assumptions behind
token project, it is our goal to set the
the formulas being used before applying
stage for building token valuation
them to new situations. Finally, there
methodologies and frameworks in the
are the issues of overfitting and model
future.
complexity – data scientists often talk

INTRODUCTION

Since mid 2016, the subject of token reviews, growth expectations, term
sales/ICOs has become an increasingly sheet proposal, etc.
discussed topic with varying views.
Some regard the rise of this new VCs would then decide on which
investment model as the future of projects were worthy of investing in,
crowdfunding, while others have and it was common to hear stories of
branded it as a vehicle of scam. entrepreneurs being rejected multiple
times before gaining a first round of
The reason for this level of interest funding.
stems from the fact that over the past
two years, the amount of investment But with the rise of ICOs, this process
generated from this investment vehicle has been significantly altered. Based on
has grown at a phenomenal rate all the publication of a white paper (which
across the globe. Few investment is essentially an extensive business
vehicles have been able to generate proposal that includes specifications of
similar amounts of capital raise in such a to-be constructed product or service),
limited time periods (See Figure 1). and by leveraging the hype of the
Blockchain, a number of entrepreneurs
Prior to the launch of ICOs, most have been able to access considerable
entrepreneurs had to follow a laborious sums of funding via token sales.
process to acquire funding – approach
Venture Capitalists (VCs) or similar The sums being raised are quite
investors, and present a project with significant as well. As per recent market
multiple supports - such as a well- statistics, approximately $5.6 billion
defined marketing strategy, a realistic was raised via ICOs in 2017, and by the
yet ambitious business plan, a first quarter of 2018, that sum was
prototype of a product or service already surpassed. Some
offering, summarized test-group approximations state that $12 billion
have been raised since the beginning of
2017. Figure 1 offers some insight.

Figure 1 : ICO Funding versus VC and CVC Funding

next.autonomous.com

Source: Crypto Utopia, Autonomous NEXT, 2018. Image republished with permission.

Key Figures1:

2016: $240 million(M) raised via Token 2018: $6.3B

Sales
Average raised via Token Sales =
2017: $5.6 billion(B) via Token Sales $12.8M
compared to $1B via VC

Summary of waves of investment into crypto-assets (2013 – 2018):

§ First wave of investment from an unprecedented rise in prices for

traditional venture firms in Bitcoin cryptocurrencies. Over $7 billion of
associated companies started in 2013. investment went into the space,
From $200 million in 2013, it reached almost 4x greater than equity
almost $800 million in 2016. investment in crypto companies

§ Second wave of investment came

from corporates. Between 2015 and
§ Many ICOs formed to take advantage
2017, between $250-$400 million was
of the “gold rush” and created
invested annually
questions and issues of quality and
regulation for tokens2.
§ Third wave of funding came from
public crowdfunding into ICOs, with

1,2
Data Source: Tokendata.io, Coindesk and Fabric Ventures

The reason for this spurt in access to funds is two – fold:

Scope: Previously, the domain of sum of many small contributions largely

funding in early stage lucrative start- exceeded the capital furnished by high
ups was a privilege generally reserved net worth investors. Not only could they
to VCs and wealthy investors. Indeed, access a bigger pool of investors, but
in most OECD countries, only 12% of they could also get funded faster – For
the population invests in stocks example: The Bancor ICO raised $153
(Lacalle, 2017) . To partake in this kind Million in less than a day (See Table 1).
of investing, a participant needed
access to significant funds and an ability Utility: While initial Blockchain projects
to deal with risk. were mostly focused on payment
solutions, as the technology has
However, as blockchain technology is matured and as the use of Smart
decentralized and allows investors to Contracts has increasingly proliferated,
contribute small sums , the chance to the application of this technology has
invest in such offers was now open to spread to different sectors and
anyone with access to the internet and industries thus drawing investors in the
with a humble disposable income. process. Figures 2 and 3 respectively
show the industries and investment
Token sellers thus quickly realized that sectors that are being penetrated by
what they could achieve with ICOs was this technology:
vast economies of scale in which the

Table 1: Top 10 ICO raises:

Project Raise

Tezos $230,498,884

Filecoin $200,000,000

Sirin Labs $157,885,825

Bancor Protocol $153,000,000

Polkadot $144,347,146

QASH $108,174,500

Status $107,664,904

Kin $98,500,326

COSMA $95,614,242

TenX $83,110,818

Total $1,378,796,646

Source: ‘State of the Token Market’, Squarespace (2018)

 10

Figure 2: Token Sale investment as per industry

Image Source: Smith and Crown.

Figure 3: Token Sale investment as per sector

Data Source: Fabric Ventures and Tokendata.

 11

As a result of this fast emerging trend, beginning to acknowledge that they

even long term successful VCs like Fred need to adapt their business model in
Wilson of Union Square Ventures are light of these changes:

“[ICOs represent] a legitimate disruptive threat" [to the VC

model],… We are excited about them when they are the right
thing for our portfolio companies and we are encouraging those
companies to use this new approach." (Wilson, 2017)

OBJECTIVES

In light of these fast paced changes, the creating a new crowdfunding model,
first objective of this report is to provide testing a valuation model requires
the reader with an understanding of empirical evidence.
how investors are currently analysing
ICOs prior to allocating their capital. As of today, most of the funds received
by token issuers are being used to build
While equity investing has established the actual product or service they
models and standards, the same cannot described in their whitepapers.
be said for token investing. As a result,
a number of investors have tried Without sufficient data regarding the
retrofitting stock evaluation models in adoption cycle of their products, the
order to attempt to create token switching costs that users will have to
valuation models. bear when switching from an App to a
Dapp, the multi-modal nature of tokens
While these approaches are laudable, (tokens bear currency-like aspects that
we show that owing to the networked are not seen in stocks) and the
nature and diversity of tokens, such changing regulatory environment,
models have limited applicability. Thus, building a generalistic evaluation model
a secondary objective of this report is to would be erroneous today.
show what needs to be taken into
consideration by investors who want to Our main objective is therefore to
create token valuation models. provide some guidelines of what needs
to considered when such an attempt is
It must also be remembered, that while made in the future.
token sales show promise, in terms of

 12

PART 2: CURRENT STATE OF TOKEN VALUATION

1.1: FRAMING THE PROBLEM

While the process entrepreneurs have earnings, there are very few data points
to follow with traditional capital raising that can be used to estimate the value
is laborious, it serves a purpose. It of the company, let alone determine
helps ensure that a project is viable, what the token price should be.
that a product has worth and the capital
being raised will be used appropriately Moreover, a token does not offer a
over a long period. In exchange the dividend (as a stock does) as the
team benefits from the guidance of an company has not yet generated a cash
experienced investor who apart from flow. Most tokens only offer a right to
providing them with capital, can also the future use of a to-be-constructed
furnish technical, legal or business product, under the assumption that the
guidance to help grow their company. company will not pivot the product once
it has received the funds. Recently the
Rigorous due diligence is conducted as concept of Equity Tokens a subcategory
there is a cost of ownership – when a of security tokens that represent
VC invests in a start-up, they gain ownership of an asset, such as debt or
access to a certain percentage of the company stock has begun to gain some
start-ups shares, i.e.: ownership of the attention, but by and large, most
firm, and any associated liabilities as a tokens offer the right to the future use
consequence. Similarly, by accepting of a to-be-constructed product.
VC capital, the entrepreneur is giving up
part of his/her company and open to the The de facto source of reference used
risks that might affect the investor’s by investors interested in investing in
portfolio. an ICO is the whitepaper. Some token
creators will also publish a technical
This mutual sense of responsibility is paper and in some cases, even some
however partially lost with ICOs. A preliminary bits of code would be
token sale generally does not function available for review which would be
as a share/dividend. Contrary to its analysed by serious investors. But by
namesake, the Initial Public Offering and large, there is no concrete
(IPO), an ICO or Token Sale does not investment framework to determine the
involve the sale of equity in (or voting value of a token.
rights pertaining to) a company per se.
Instead, ICO participants are acquiring Thus, token investors are faced with a
an asset— a “token”—which allows the quandary. On one side, they are dealing
holder to use, or govern, a network that with a technology that is yet to become
the token sellers plan to develop using a formal curriculum subject in most
capital raised via the sale of the token. universities. Nevertheless it holds
As the business behind the token is yet tremendous potential and is considered
to be built and has no assets or by a number of tech pundits as the next

 13

internet. The fact that established VCs among others. The market still sets the
are taking this technology seriously price. But by using these evaluation
underlines the impact it could have on models, investors are trying to estimate
the financial industry. what the future price would be based on
a collection of data points that
On the other side, owing to the age of represent concrete elements such as
this technology and its evolving nature, physical assets, existing market share,
there is no formal framework that can IP, product-market fit, existing clientele
aid investors in determining the true population, etc.…
value of a project classified under this
new asset class. This problem is further While these methods allows investors to
compounded by the variety of tokens encapsulate a future anticipated value,
and their built-in functionalities that are they also serve a secondary role of
usually tailored for specific purposes. protecting against runaway market
speculation. When the value of a good
The lack of structure, frameworks and or service is being speculated upon,
regulation means that currently, the often times it is seen that markets can
ICO space is extremely susceptible to go awry creating bubbles and boom
market forces. This statement needs to cycles in the process. Conducting DCF
be further deconstructed in order to and DDM analyses thus allows investors
underline its contemporary significance to determine what a realistic price point
in the context of ICOs. would be for a stock. In the process,
they mitigate against excessive market
History shows us that by and large,
speculation.
markets function as the best price
setting mechanism in distributed, fair, Without tools such as these, the act of
decentralized networks. The actual determining what is a sensible
value of any asset is based on the price investment, and if it fits into your risk
that others are willing to pay for it, and profile, is not feasible and investors
competitive forces in a market allow would be exposed to the full brunt of
different economic agents to set the open market speculation effects. While
price for a product or a service. these evaluation methods are not
perfect, they allow investors to add a
This modus operandi holds true when
certain dose of pragmatism when it
we are dealing with a tangible good or
comes to investing in a to-be-
service that is immediately available.
constructed good or service, or in the
But when it comes to investment,
expected future growth of a company.
agents are forced to compute an
investment price based on its future But when it comes to Token sales, we
anticipated value—usually represented currently do not have a similar control
by its stream of future cash flows. mechanism. As a token is not a stock,
using the same methods of stock
To aid in this future determination, we
evaluation does not offer the same
use a plethora of tools to aid in our
amount of certitude. As we currently do
investment decisions. For example
not have a proper framework that
when evaluating the future value of a
allows us to determine the value of
stock, we use methods such as
certain attributes of a token sale, this
Discounted Cash Flow (DCF) or
leaves the price of tokens open to
Dividend Discount Models (DDM)

 14

market forces speculation. As a result, where cryptocurrency and token prices

the price of tokens is extremely have seen dramatic rises and falls in
volatile. There have been multiple cases short time periods (See Figure 4):

Figure 4: Cryptocurrency volatility index2 movements (May 2017 – Sep 2018)

Image Source: https://www.sifrdata.com/

This susceptibility to market forces and resulting acceptance of token price volatility
has led to a number of negative effects or practices of questionable utility:

§ Pump and Dump Schemes: (such as Big Pump and Alt-Pump) have
resorted to ‘pump n’ dump’ schemes,
As the regulations around this where a group of individuals get
technology are still being set in place, it together on online forums and decided
offers the chance to engage in market to artificially inflate the demand, and
manipulation activities that would be hence the price of a token. If one were
considered illegal with other asset to spend a few minutes scrolling the
classes. Recently, nefarious actors on cryptocurrency content on Telegram,
certain cryptocurrency communities the encrypted messaging app, they

2
The cryptocurrency volatility index is composed of six currencies: BTC, ETH, XRP, LTC, DASH, and
XMR. The volatility index is weighted by the market capitalization of each currency which is updated
daily by the index creators, Sifrdata. See: https://www.sifrdata.com/cryptocurrency-volatility-index/

 15

would soon discover dozens of ‘pump make their platform useable from an
and dump’ scams (Murphy, 2018) (Wall economic standpoint, Telegram decided
Street Journal, 2018). to launch an ICO where investors could
contribute and acquire TON tokens in
Pump and dump schemes (a daily list the process.
can be found on
pumpdump.coincheckup.com) are just While the objectives were aligned with
one example of market manipulation societal issues, nevertheless, the TON
that are rife in this loosely-regulated needed to be protected against
sector. Other practises such as front- volatility. As a result, Telegram decided
running, wash- trading (which involves to keep aside 52% of the capital raised
creating false volume by matching via the token sale in reserve.
trades), and spoofing (where you place
and then quickly cancel orders) are The Telegram ICO aimed to raise $1.2
currently being practised. Unscrupulous Billion in order to create its censorship
agents leverage market forces and proof solutions. 52% of this sum comes
loose regulation to make a quick buck. up to $624 Million, a sizeable sum to
No consideration is given to the say the least. $624 Million that would
business solution or to the problem- not be used in the creation of products,
solving potential of the product behind but kept in reserve and unused. All to
the token. Value creation is scarified at counteract the effects of market
the altar of short get rich quick schemes speculation and the actions of nefarious
(Financial Stability Board, 2018) . In the actors.
words of Asaf Meir, the founder of
It is therefore essential that going
Solidus Labs, which develops market
forward, a solution needs to be found
surveillance tools specifically for crypto
that will allow both token creators and
markets, “[The market] is highly,
token investors with a more efficient
highly, highly manipulated. The extent way to allocate capital raised.
is truly humongous” (Murphy, 2018).
§ Airdrops:
§ Inefficient Capital Usage :
In light of the above mentioned issues
Following the risk to volatility and pump
and the regulatory greyness with
and dump schemes, token creators
regards to ICOs in general, a new trend
have been obliged to hedge against the
has emerged in which potential
token’s price volatility by indulging in
investors are given tokens for free in
capital protection techniques. This the form of an airdrop.
normally involves keeping aside a
sizeable portion of the capital raised in The objective of an airdrop is 3-fold:
reserve to counteract against token
volatility. i. Raise popularity of your token by
engaging early adopters of this tech
For example – The Telegram ICO (which to create a network effect. An
was subsequently cancelled) was airdrop functions both as a
launched in early 2018 and aimed to marketing interface and an on
create censorship resistant file storage, boarding experience.
messaging, decentralized apps and
browsing. To fund the development of
these complex consumer solutions and

 16

ii. Raise awareness of the token once Another issue with airdrops is the
more people start trading it, thus supply of the tokens (a topic that will be
raising the price (if all goes well). addressed in later parts of the report).
While scarcity of a token is often cited
iii. Use (i) and (ii) to get serious as the means of increasing its value,
investors who would be willing to shortage of a token can create its own
contribute capital to the difficulties.
development of the project.
This was seen most recently with the U
While the concept looks enticing and Network (a blockchain publishing
has had certain benefits especially from protocol valued at around $8 million) ,
a legal standpoint, the ability of which in early July 2018, announced
Airdrops to convert into actual that it had run out of its reserve of UUU
investment requires more research and crypto tokens.
documentation – something that is hard
to find today. At the start of the project, U Network
established a 10 billion UUU cap on its
Furthermore, airdrops are essentially token supply (worth approximately
free token giveaways and offer very $15.6 million). An estimated 8 million
little to token holders. Essentially, worth of tokens was given away in
holders of tokens (received via Airdrops airdrops. But as the project gained
or traded later on an exchange) are traction and number of strategic
holding a security without any rights, partners began to increase, the demand
for UUU tokens exceeded the
The development team behind the designated holdings (Milano, 2018).
Airdrop can sell out the business to
anyone without giving token holders The U Network is now attempting buy
any returns. They are also free to, they back some of the supply it distributed to
can take away any rights they do have, early investors through its airdrop in
or decide that it is better for their February. But as it will be seen in later
business to accept USD or BTC next to parts of this report, the supply of tokens
their own token for payment – all of is a key factor that needs to be taken
which lowers the value of the token or into consideration when coming up with
makes it worthless. a valuation model for a token sale.

1.1.1 What does this mean for the future of this investment vehicle?

Token Sales were meant to become an However, a large amount of evidence

investment vehicle that would allow for shows that this objective is not being
democratic investment of capital and met in the way it was initially
allow regular individuals to enter the portrayed.. On the contrary, token
investment space. It was portrayed as sales are being used as a vehicle of
the big push that would allow equity manic speculation on trading markets.
crowd funding and peer to peer lending As providers of a token have little or no
to replace the older system of fiduciary duty to the investors, certain
investment banks and public capital actors have made use of the lack of
markets. regulation to get capital for unrealistic

 17

projects and outright scams - as of § UPS, FEDEX, AND BNSF RAILWAY

today, 46% of ICOs have failed JOIN BITA ALLIANCE to explore
(Sedgwick, 2018) while and close to Blockchain technologies in freight
10% have turned out to be scams transport.
(Cimpanu, 2018) . Considering that § Petroteq creates distributed ledger for
close to Over $20 billion has been raised Pemex (the first petroleum company
by Crypto projects through Initial Coin to accept crypto)
Offerings since the start of 2017 § UBS, BARCLAYS, and CREDIT SUISSE
(Autonomous NEXT, 2018) , 10% create an Ethereum based private
comes up to at least $2 Billion (an platform to cross-reference legal
approximate figure) which is hard to entity identifier (LEI).
ignore. § Government of Singapore has
launched Project UBIN to see how
It is important to affirm at this point, monetary and fiscal policy tools can be
that this does not mean that the better constructed via the Blockchain.
technology is unsound or that it is a § Walmart, Tyson, Unilever, Nestle,
market for making a quick buck. A Kroger, Dole, McCormick, and others
number of projects show tremendous band together to launch a Blockchain
potential to change the existing market pilot
structure and provide better, cheaper § The United Nations explores DLT and
and more secure products and services Blockchains for humanitarian aid and
to consumers all around the world. climate science
§ TEPCO (Japan’s largest utility
This statement can be proved simply by
provider) invested and partnered with
the fact that many of these investments
ELECTRON to explore how energy
have been occurring in the corporate
price matching can be made more
space. The figures below show how
effective with micropayments.
large companies are not just adopting
§ BRAZIL’S MINISTRY OF PLANNING &
blockchain technology but also using it
BUDGET, is piloting a Blockchain
to transform core business elements of
identity application using Uport, a
their business models (CB Insights,
2018): self-sovereign ID platform built by
Consensys
§ MICROSOFT & ID2020 ALLIANCE
This level of interest, the rapid
(Accenture + Avanade) launched a
evolution of this technology and the
project in early 2018 to provide ID
increasing complexity of an ever
services to 1.1B people
growing participation pool, shows there
§ GOLDMAN SACHS plans to open a
is discernible value in this technology
CRYPTOCURRENCY TRADING DESK
and being able to navigate this
(end 2018)
turbulent phase of the technology’s
§ The mining company BHP BILLITON is
evolution requires education, poise,
using the Blockchain for contract work
experimentation and patience. What is
and analysis
required therefore is a need to
§ MAERSK + HYPERLEDGER are uniting
construct a valuation method that:
stakeholders in global supply chains,
to track freight & replace paperwork § Will allow investors to gauge the
with tamper-resistant digital records feasibility of a project,

 18

§ Will have a modular architecture Without such a framework, participants

which is able to adapt to the type of will be speculating on abstract ideas
token being analysed. and rough compositions of turbulent
§ Will allow investors to mitigate against revenue pools of newly founded
the negative effects of speculative companies, without any detailed
market forces ,and understanding of what factors need to
§ Will ensure that they are allocating be analysed when looking at a growing
their capital sensibly based on rational business. This affects not just the
attributes. investors but also the ICO founders
who, under pressure from their
Such a methodology would allow community, are often forced to list their
investors to determine if the market token on an exchange even prior to
price of the token directly or indirectly making any meaningful technological
represents the ‘true’ value of the token, progress or a plausible product/service
and to what extent this price is being offering.
affected by speculative market forces.

1.2 UNDERSTANDING TOKEN VALUATION METHODS

This need to determine the true value to the genesis of financial analysis and
of a new asset class is not a new corporate finance.
requirement. Early financial markets
went hand in hand with fraud (Klaus, Graham went on publish a number of
2014) and since the creation of stocks other books on the subject including the
and trading markets, the ability to well-known classic ‘The Intelligent
discern between intrinsic or the true Investor’. It was on the backbone of his
value of an asset versus the speculative work that the fields of fundamental and
value of an asset, has been an issue of technical analysis of stocks was
pivotal importance to investors. created.

One of the first people to address this One of the repeating themes in the work
need was Benjamin Graham. In 1934, of Graham, was the separation between
Graham published the book, Security Intrinsic and Speculative value of a
Analysis, which offered the first formal security. Intrinsic value, as defined in
approach to determine the intrinsic Security Analysis, is
value of a stock. It is this work that led

“that value which is justified by the facts, e.g., the assets,

earnings, dividends, definite prospects, as distinct, let us say,
from market quotations established by artificial manipulation or
distorted by psychological excesses.” ……..Logically, it must be
based on the cash flow that would go to a continuing owner
over the long run, as distinct from a speculative assessment of
its resale value.”

 19

By driving a wedge between Intrinsic fundamental attributes of a business

and Speculative value, Graham’s work (and its products as proxies), the value
allowed investors to determine what the was measurable, justifiable and stable.
base value of a security was, how to Volatility still played a role as it is
determine it and how to use it in the ultimately the market that sets the final
face of rampant speculation, thus price. But the intrinsic value of a
offering a margin of safety. security remained stable, measurable
and quantifiable, thus making it a
Intrinsic value was a key point in all of reference point when considering an
Graham’s work owing to its stability. As investment as it allows investors to
the intrinsic value was based on discern from market speculation

1.2.1 From Stocks to Tokens

The reason for reviewing the core Secondly a token is used to create,
tenets of financial analysis is because it generate and stimulate value. As it is a
is exactly this bedrock that is missing unit of account and a means of
with token evaluation. In later parts of exchange, a token functioning in a
this report we will review the work done looped and interconnected network
by various entrepreneurs and such as a Blockchain, is also an
researchers who are exploring this Endogenous Monetary system with its
subject with respect to tokens. We will own supply, demand and liquidity
see that while most of these efforts issues Thus, when looking at a
have been made in order to determine tokenized economic system, evaluators
some way to evaluate the value of a need to refer monetary policy aspects
token, most of these efforts have of this networked economy as well the
limited usability as they essentially try underlying business behind it.
to retrofit stock evaluation methods to
create a token valuation method. This aspect of supply is also intimately
linked to the technical aspects of the
But a token is not a stock. While stocks token. The governance of tokenised
share a common framework the same assets or cyptoassets is done by smart
cannot be said for tokens. Not only does contracts. Hence any changes in supply
do tokens not share the same economic – including transfer or lock up plans –
artefacts of a stock, but by themselves need to be hard coded into the smart
tokens are rather unique entities. contract. This makes the governance of
tokens both an economic and technical
Firstly, a token has similar properties to issue simultaneously.
a currency. A token functions not just
as a fund raising mechanism or a utility Lastly, there is the question of variety.
instrument for a new product. It also Today, there are many different types
functions as a means of exchange and of tokens - we have security tokens,
as a unit of account to a certain extent. work tokens, network value tokens,
Thus it shares at least two, if not all asset tokens, utility tokens, payment
three, attributes of a currency (Store of tokens and so on… Figure 5 shows us
Value, Unit of Account and a Means of the current taxonomy of tokens, which
Transfer), which means that it needs to helps us see how diversified this space
be evaluated as a security and a is (Also see Appendix 2):
currency simultaneously.

 20

Figure 5: Taxonomy of Tokens (compiled by Autonomous NEXT)

Source: Crypto Utopia, Autonomous NEXT(2018) . Image republished with permission.

This variety, along with the supply and and vice versa. This is the first lesson
liquidity issues, complicates the manner an investor needs to digest.
in which we can evaluate tokens. There
can be no one-size-fits-all evaluation The second lesson that the investor
methodology as seen with stocks. needs to learn is the kind of variables
that need to be considered. While stock
As a result, investors or academics who evaluation is largely made up of
are interested in developing a valuation financial variables and ratios, tokens
methodology must realise that they are fully digital entities that exist on a
need to follow a modular approach in networked plane. Thus the kinds of
which key variables and analysis variables that need to be analysed are
methods need to be selected based on not just financial but technical as well,
the kind of token or cryptocurrency they especially when analysing smart
are interested in evaluating. contracts. Retrofitting stock valuation
models therefore hold less gravitas as
If we are to allow this technology and they currently do not incorporate such
this new breed of investing model to kinds of variables. Furthermore, when
flourish, then we have to construct a stock analysts use ratios – such as P/E,
toolkit of evaluation methods, where EV/EBIT, Debt/Capital, Debt/Equity,
each evaluation method allows us to ROA, ROE, etc.. – they analyse data
evaluate certain aspects of the token related to these ratios over extended
sale. The evaluation model of a network periods of time. As the token space is
token will not work for a utility token nascent, similar ratios currently do not

 21

exist and any time-series data is over a comes to conducting fundamental and
very short time period. technical analysis of tokens. We will
also explain what we mean by technical
As mentioned, since the variety of analysis, as our definition of technical
tokens is quite widespread, being able analysis does not relate to Chartism.
to evaluate the value of a security token
will be different from evaluating the Hence, the goal of this report is to
value of a utility token. But some provide the key tenet’s of a base
attributes (such as base network value) valuation methodology and introduce
and certain variables will be shared by the concept of a modular token
all tokens which can help us create a valuation approach. As the Blockchain
generalised valuation method to a and cryptoasset space grows, the
certain extent. variety and diversity of products and
services will also grow, making such an
Taking these points into consideration, evaluation model more adjustable to
the remainder of this report is broken the upcoming changes.
down as follows:
Prior to delving further, it is important
Part 1.3: We start with a review of the to highlight that fundamental analysis
current valuation methods being practices that are applied to stock
developed. This provides us with an analysis still apply to tokens. Reviewing
overview of what’s been done and helps the team, the experience they have, a
us ascertain the pertinence of current breakdown of the product/service they
valuation methods and where a model are offering and studying other data
makes sense. points that are today considered normal
practises of due diligence, are classic
Part 2: In this part, we will delve into
analysis techniques that should never
more specific aspects of Token
be forgotten. In this vein, we start by
valuation and introduce the key
looking at what’s already been done.
variables to be considered when it

 22

1.3 REVIEW OF CURRENT TOKEN VALUATION METHODS

In mid-2017, Chris Burniske and Jack cryptoasset valuation, the authors then
Tatar published the book, attempted to use economic metrics
“Cryptoassets”, which has been such as P/E ratio, and offered
described as “The innovative investor’s takeaways from financial analysis
guide to an entirely new asset class”. methods such as the Discounted Cash
Apart from the evident reference to flow method and the Velocity of
Benjamin Graham’s classic book, ‘The circulation in order to underline the
Intelligent Investor’ (1949), similarities between stock evaluation
Cryptoassets also shared a similar and token valuation.
objective – to develop a way to evaluate
the actual value of a token. However, at no point in the book do
they offer a valuation methodology set
In the book, the authors shine light on in stone. Instead the focus is on
some important landmarks. Firstly, establishing linkages between methods
they offered a classification of that are already used to evaluate assets
Cryptoassets into 3 groups – and to highlight the new variables that
Cryptocurrencies, Crypto-Commodities we need to consider when trying to
and Crypto-Tokens (See Appendix 1 for valuate cyptoassets.
more details on this classification).
The book nevertheless marked the
Secondly, they made the first attempt beginning of a formal conversation on
to come up with a token valuation the subject of token valuation methods.
model that would allow investors to Since the publishing of Burniske and
make more informed decisions when Tatar’s book, a slew of blogs, academic
thinking about this asset class. They articles and reports have been
mention, that when examining a published to explore this subject, many
cryptoasset, the fundamental analysis of them referring Cryptoassets and
ought to include: some that have been penned by the
same authors. Since October 2017 an
§ The Whitepaper increasing number of articles on this
§ Decentralization Edge subject have been published
§ Community and developers (Smith+Crown, 2018).
§ Relation to digital siblings
§ Valuation As the number of people exploring the
§ Issuance model subject continues to grow, so does the
diversity of the approaches. Today,
A few technical variables were also there are articles that are attempting to
enumerated, such as: apply Black-Scholes Option Theory
(Antos, 2018) for token pricing, and
§ Hash rates (as a sign of security)
new terms such as Crypto J-Curve
§ Number of miners
(Burniske C. , The Crypto J-Curve ,
§ Company support
2017) seem to be discussion points.
§ User adoption measured by Number
of Users and Number of Transactions Each methodology has its benefits and
challenges. A short review of the
Having established which variables various methodologies helps us
needed to be measured to perform a determine their applicability:

 23

1.3.1 : Store of Value Methodology

One of the first valuation methods that considers that value is based on market
built upon the three functions of money forces and that confidence in a “stable
and stated that a token’s ability to serve coin’’ will automatically translate to a
as a store of value can drive significant stable value.
value to investors.
From the current experience we have,
As per this method, cyptoassets that asset-backed tokens often offer limited
have steady values by design (E.g.: tangible redeeming possibilities (case in
stable coins), or which are expected to point – Venezuela’s Petro). Moreover,
grow in price, make for attractive “store it seems to apply only to asset backed
of value” coins . tokens. Hence, its use is limited both in
scope and technique.
This method is thus mainly reserved to
asset-backed tokens as the valuation However, it could be useful to analysing
process involves determining the total the value of stable coins3 or tokens that
assets attached to a token and dividing are collateralized. This collateralization
it by the number of tokens. can occur in three forms:

As a result, this methodology has little

else to offer and is quite simplistic. It

1. Fiat-Collateralized: A certain amount of fiat currency is deposited as a collateral

and coins are issued 1:1 against this fiat money

2. Crypto-Collateralized: Similar to their fiat-counterparts, with the exception that

the collateral is not an asset in the “real-world” but rather another cryptocurrency

3. Non-Collateralized: Not actually backed by anything other than the expectation

that they will retain a certain value. One oft-mentioned solution to non-
collateralized stablecoins is the seigniorage shares approach. This concept builds
on smart contracts that algorithmically expand and contract the supply of the price-
stable currency much like a central bank does with fiat currencies, but in a
decentralized manner (Schor, 2018).

3
A stablecoin is a cryptocurrency that is often MakerDAO, Basecoin, TrueUSD, Arccy, Stably,
pegged to a stable asset, like gold or the U.S. BitShares, Sweetbridge, Havven, Augmint,
dollar. This gives it lower volatility and some Fragments, Carbon, Kowala, X8X, Globcoin,
practical usage attributes - Sherman Lee, Stronghold USD, etc.
Forbes, (March 2018). Examples– Tether,

 24

Figure 6: Value linkages for stablecoins

Source: “Stablecoins: designing a price-stable cryptocurrency”, Qureshi, (2018)

1.3.2: Token Velocity Methodology

This methodology has gained a lot of this approach is that as the native token
ground in the discussion on evaluation of a smart contract platform becomes
methods owing to its connotations to widespread and sufficiently useful, it
considering a token based economy as will emerge as an independent store of
a monetary system. As a result, its value (Samani, 2018).
immediate application has been with
general purpose cryptocurrencies which Drawing from The Monetary Equation of
function as independent monetary Exchange (MV=PQ), which is often
bases, such as – Bitcoin, Bitcoin Cash, referred to as The Quantity Theory of
Zcash, Dash, Monero, Decred, etc.. Money, this modified version looks at
the token based economy as an asset
Some proponents of this methodology that is being exchanged. The table
also state that it can be used when below offers a comparison between the
trying to valuate native tokens of smart original version of the equation’s
contract platforms such as Ethereum, variables and its crypto equivalent
EOS and Dfinity. The reasoning behind (Lannquist, 2018):

 25

Table 2: MV=PQ in the Crypto Context

Traditional Version Crypto Equivalent

M = Money Supply in the Economy (M1) M = Size of the asset base

V = Velocity of Circulation V = Velocity of the Asset

P = Price of the digital resource being

P = Price level in the economy provisioned

Q = Quantity of the digital resource

Q = Output produced by the economy being provisioned

The method states that velocity is a Thus it is hard to establish the size of
significant driver of token price, and the the asset base, especially when
lower the velocity, the greater token applying it to utility tokens that function
price is via an appreciation of M (size of as proprietary payment currencies such
asset base).. The implication of this as Filecoin, Golem, Civic, Raiden, Basic
method is that tokens with low velocity, Attention Token, etc.…
i.e. those that held (owing to
speculation, asset backed, etc.), will Secondly, and more importantly, the
see prices rise. model is a retrofitted version of an
antiquated equation. Most economist
Prior to going further it would be don’t even use the formula anymore as
essential to note that users of this it requires too many assumptions – for
methodology need to apply it with a example, V is assumed to be a
large pinch of salt, for there are a constant, which is hard to calibrate
number of assumptions needed to make within a functioning economy whose
this method work. natural state is entropic rather than
equilibrium. A number of empirical
Firstly, it is based on having a studies have also shown that the
measurable value of M. If this refers to MV=PQ formula is not supported. Hence
a private Blockchain where all the formula is more of a tautology
quantities held by the networks rather than a method.
participants is declared, this might
work. But the assumption of being able Retrofitting this equation to measure
to calculate M especially. In a public the velocity of exchange of a token
Blockchain, is problematic as holders of continues to prove problematic, as
a token can often store tokens off chain. when velocity changes, the choice to

 26

record the effect in M, P, or Q is white papers. One notable exception

arbitrary and yields different could be the  Basic Attention Token
implications for token price. Further, V’s (BAT), who state the contention that V
relationship and correlation with these is going to be based on natural
factors is dynamic, and assuming a supply/demand dynamics between
steady relationship with P,Q, or M is token users and hoarders such that, an
again arbitrary and problematic. Thus optimal token velocity is ensured. If
many many assumptions have to be such a situation were to exist, then the
made when it comes to using this MV=PQ formula could possibly be used
equation. but alongside a total addressable
market (TAM) analysis4.
It is possibly for this reason that optimal
token velocity is rarely addressed in

1.3.3: Crypto J-Curve Methodology:

This method is a recent idea proposed When a token project is launched, it

by one of the authors of Cryptoassets starts to a develops, and as early
and is an extension of the MV=PQ adopters are excited about the
approach. As per this model, a token’s potential, they drive up the expected
price is based on two components value. DEUV thus dominates the initial
whose contributions to the token’s price growth of the project. . As enthusiasm
evolve over time: wanes or if technical challenges are
discovered, the price declines and CUV
The CUV refers to the current utility now begins to play a role in deterring
value, which represents value driven by the price. As the project matures, the
utility and usage today, token becomes more adopted and CUV
grows. DEUV then catches up as
The DEUV represent the discounted
speculation and excitement follow this
expected utility value, which represents
value driven by investment speculation. new growth. Ultimately in the steady
state, CUV drives token price and this
The model goes on to explain the evolution creates a J-Curve, which is
interlinking between CUV and DEUV – often seen in financial valuation:

4
TAM analysis, is done to determine the The total market price consists of current utility
current utility value of a cryptoasset. A TAM value and discounted future expectations of
analysis is a top-down approach which begins the cryptoasset network’s key drivers in
with the estimate of the market’s total size and subsequent years.
then ascertains what share of the market the
cryptoasset network could potentially obtain.

 27

Figure 7: The Crypto J-Curve by Chris Burniske

Source: The Crypto J-Curve (Burniske C. , The Crypto J-Curve , 2017)

As it can be seen, this approach also Some adopters of the Crypto J-curve,
involves a number of assumptions. have begun using it as a proxy for
Firstly, DEUV is calculated by using a measuring the different stages in the
modified version of the Discounted life of a cryptoasset. For example the
Cash flow formula that is used to New York based VC investment fund,
analyse stocks. Except a Token is not a Placeholder uses the curve as a means
stock as they usually have no assets, of determining which stage the token
earnings or cash flow. Hence using an sale is in - The whitepaper stage is
adapted version of the NPV (Net where the team works to define and
Present Value) or discounted cash flows implement a “minimum viable
has limited applicability. protocol,” which validates the network’s
functionality. The release stage is when
Secondly, it is based on MV=PQ which a crypto network’s token is first made
as we have seen before, is based on available to the public, and the public
antiquated methods and too many stage when the token begins trading on
assumptions. As velocity is an input exchanges (Monegro & Burniske,
value, the model suffers the same 2018).
drawbacks related to the token velocity
model.

 28

Figure 8: Applying the Crypto J-Curve to the lifetime of a token project

Source: Placeholder Thesis Summary, Monegro & Burniske, 2018

Hence, while Private Equity funds often cyptoassets is limited to ascertaining

use the J-Curve to calculate the period the life cycle of the products
over which the return on investment development. In such, it does not
starts to become profitable, the current function as a valuation methodology per
use of the J-Curve in the context of se.

1.3.4: Network Value-to-Transaction Ratio (NVT)

An interesting method that developed The model is interesting as it the first

quite recently, is an adapted version of that looks at Network attributes rather
the stock valuation Price to Earnings than financial models. This is an
ratio (P/E ratio). important point and something that we
will be addressing in greater detail in a
NVT = Network value / Daily later part of the report when we talk
transaction volume. about Metcalfe’s Law.
This valuation ratio compares the Moving forward, the use of this method
network’s value (the market cap) to the will require some formal definition on
network’s daily on-chain transaction what constitutes a valid transaction
volume. NVT may indicate whether a needs to be made as in certain networks
network token is under or overvalued that offer staking rewards - such as
by showing the market cap relative to Dash – would have inflated transaction
the network’s transaction volume, activity resulting from staking. This
which represents the utility that users would increase the denominator,
derive from the network. When the ratio inadvertently causing the ratio to be
becomes very high, it indicates underestimated. However, this effect
potential token over-valuation. could be corrected for by subtracting

 29

staking activity from transaction Propertycoin, Siacoin, 22X Fund, or any

volume. of the tokens hosted on platforms like
Polymath, Harbor, Securitize,
One key point that needs to be SwarmFund and Templum.
considered, is the transaction volume -
Transaction volumes tend to follow Apart from the methods listed above,
changes in price. The higher the price, there have been other approaches that
the greater the tendency to store the have tried to explore adapted models
token and not use them. Thus these that use metrics such as EV/EBITDA,
two elements have reflexive P/E, EV/Sales, Carhart four-factor
relationship, which can be used as an CAPM model (Crypto CAPM), Sharpe’s
adjustment factor to control price. ration and Black-Scholes Options
Theory.
At this point a distinction must be
made. The methods above are primarily As mentioned in the beginning of this
related to the evaluation of section of the report, it is not our
cryptocurrencies or utility tokens. When intention to explore all valuation models
it comes to asset-backed tokens or in all their detail, especially since most
security tokens, the valuation models of stock valuation methods are well
are more traditionalistic. known and well documented. The
objective is to provide an overview of
Security tokens, tokenized securities or the methods being used today, so that
investment tokens, are financial the reader can select the kind of
securities compliant with SEC analyses they wish to perform based on
regulations. These Gen-2 tokens can the type of token or cryptoasset they
provide an array of financial rights to an are interested in analysing. .
investor such as equity, dividends,
profit share rights, voting rights, buy- The models presented till now do not
back rights, etc. Often these tokens offer us a cut and dry method to
represent a right to an underlying asset determine a token’s price. Nevertheless
such as a pool of real estate, cash flow, their contribution has been significant
or holdings in another fund and these as they have helped us realise that a
rights are written into a smart contract new approach needs to be developed
(Koffman, 2018). just as Graham had created back in
1934. Moving forward, along with a
While moving securities onto a modular analysis approach, we also
Blockchain can have advantages in need to be able to identify key variables
comparison to a legacy system in terms that can help in the valuating the
of settlement times, lower fees, intrinsic value of a token whilst
automated service functions and respecting their unique applications.
custodianship, this does not change The image below summarizes the
anything about the nature of the gamut of valuation methods being used
security itself. Hence using the today and sets the stage for the second
evaluation models of traditional part of the report. Figure 9 offers us a
securities, which are widely understood, summary of what we have discussed in
can be applied for these kinds of tokens. this section.
Examples of security tokens include –

 30

Figure 9: Summary of the current models being used for token valuation

Adjusted DCF
methods

Macroeconomic
Community analysis
related metrics
MV=PQ

Network Value and Network Ratios

Metcalfe's Law NVT

Chartism or Options Pricing

Technical analysis methods

 31

PART 2: VARIABLES FOR BUILDING A TOKEN VALUATION

METHODOLOGY

The role of variable selection and the tokens requires that we update our
associated analysis method is of key jargon and kinds of variables to analyse
importance since tokens can function as prior to making an investment decision.
a currency, a commodity, a security or
as a mutualized asset. Owing to this Prior to providing guidelines and
multi-dimensionality, creating a variables for building a framework to
universal valuation framework is measure the intrinsic value of a token,
complicated. However the methods it is necessary to address the latencies
mentioned in the previous part of this of existing valuation practices. This will
section can be used to evaluate certain help us identify the variables that can
kinds of tokenized physical aid in performing fundamental and
assets/security tokens and utility technical analysis with regards to
tokens. As we have seen from the tokens. It also aids us in coming up
attempts of using stock evaluation with a different comprehension and
models to measure token value, the definition of technical analysis in the
fundamental and technical analysis of context of this new asset class.

2.1: The issue with the current definition of Technical Analysis

Technical Analysis, which is “the art of It is important to cite these patterns

gauging markets by looking at patterns used in technical analysis and question
in prices…, rather than the economic their usefulness in making investment
fundamentals of the investments” decision methods for two reasons:
(Arthurs, 2018). A more formal
definition of technical analysis would be 1. The methods are highly
the study of price and volume data to debated and there is no conclusion
predict future direction of stocks and on their actual effectiveness –
other financial instruments. Technical analysts, also known as
“Chartists”, see asset prices as a
This branch of pattern analysis has its function of supply and demand.
own jargon and a community which Chartists believe that price patterns
swears by it. Traders look at patterns in tend to repeat over time and, as a
the market and make investment result, are somewhat predictable. The
decisions if they see a Head and explanation for this belief is that
Shoulders pattern (Bearish market), a repetitive behaviour of markets is the
Death Cross (Very Bearish market), an result of the irrationality of investors.
Ichimoku Cloud (a Range Bound This irrationality manifests itself in
market), a Cup and Handle pattern behavioural biases that are, in their
(Bullish market) or a Vomiting Camel view, exploitable.
pattern (Bearish market).

 32

It is this rule of thumb, that technical researchers analysed 10,452 actively

analysts use to “see” patterns in the managed US equity, global equity, US
market and develop a “feeling” of what balanced, and global balanced
is going to occur in the market. But just portfolios from 1993 to 2012. The
as correlation does not lead to authors found that 55% of those
causation, looking at previous self- disavowing the use of charts were still
derived patterns to predict the future in business, while only 48% of those
seems highly questionable. For one, managers who rated technical analysis
even if a pattern re-emerges, are all the as “very important” had survived
other external, internal and associated (David Larrabee, 2013).
variables the same? The market ,after
all, is a constantly evolving entity. Thus However, they went on to state that,
seeing repetitive patterns with no "Funds using technical analysis appear
understanding of the environment to have provided a meaningful
points to a misconception/no advantage to their investors, albeit in
conception of context. an unexpected way" (Smith et al.,
2013). The following chart, from their
Studies on the effectiveness of stock research compares how institutional
technical analysis highlight this issue. funds whose managers say they make
In a paper titled, “Head and Shoulders some use of technical analysis have
above the Rest? The Performance of performed cumulatively, compared to
Institutional Portfolio Managers who the majority of funds whose managers
Use Technical Analysis” (2013), say they have not:

Figure 9: Cumulative Return Net of Benchmark for Institutional Portfolios

Using Technical Analysis versus Funds that do not

Source: Head and Shoulders Above the Rest? The Performance of Institutional Portfolio Managers Who Use
Technical Analysis, Smith et al., (2013)

 33

What this shows is that the sizeable amount of economic agents

effectiveness of technical analysis (in its take their opinion seriously – which is
current definition) is highly the case when the technical analyst’s
questionable. There may be some forms perspective is transmitted in the form of
of technical analysis that make sense, a widespread sell-side report – the
but it depends on the context and a agents make decisions aligned with this
repetition of a very similar situation. If pattern and thus create the pattern.
a situation were to arise where markets The book cites multiple examples of
have taken leave of fundamentals and how and when this has occurred.
entered bubble territory, there might be
nothing much more to go on than chart To end on this critical point regarding
patterns, which might at least help to technical analysis, it would be useful to
capture predictable mass behaviour in cite, ‘The Vomiting Elephant’ pattern.
extreme situations. But these events The creator of this pattern was Katie
are highly contextual and cannot be Martin, a Financial Times journalist who
generalized as sound investment has covered global foreign exchange
decision references. markets for a number of years. A few
years back, she came up with the
Also there is the concept of Vomiting Elephant pattern as a joke
“Anchoring”, as coined by Daniel (Martin, The truth behind the vomiting
Kahneman and Amos Tversky 5 . If camel graph , 2018). The pattern of a
enough people think that technical vomiting camel was drawn on charts
analysis matters, they will anchor on and tweeted as means of providing a
outcomes that technical analysis deems satirical bent on the state of markets.
important, and in the process execute
group-think actions that manifest Surprisingly, chartists picked up on this
themselves as self-fulfilling prophecies. and began using it increasingly. Even
This concept of self-fulling prophesies CNBC reported in 2013 that a vomiting
has been greatly analysed and pattern formation had appeared in gold
explained from an anthropological (CNBC, 2013). A simple online search
perspective in the book “Stories of shows how this pattern has continued
Capitalism”, (Leins, 2018). Essentially it to be identified and used by technical
shows that technical analysts see what analysts ever since, even in the crypto-
they want to see and then inform the space.
market about their perspective. If a

5
Anchoring or focalism is a cognitive bias that psychology. Kahneman and Tversky carried
describes the tendency for an individual to rely out multiple experiments, whose conclusions
too heavily on an initial piece of information can be found in the book Thinking, Fast and
offered (known as the "anchor") when making Slow.
decisions. It is one of the most well tested
phenomena in the world of experimental

 34

Figure 10: The Dreaded Vomiting Camel Pattern in Gold

Source: Brain Kelly, TradingViews.com (CNBC)

In light of these issues, it would be Today reports explaining how bitcoin’s

prudent to look at stock technical ‘Death Cross’ price pattern might be a
analysis methods with a large pinch of bearish tendency (Godbole, 2018) (Pei,
salt, especially when trying to invest in 2018), video’s explaining the ‘Ichimoku
tokens. Unfortunately, as the point Cloud’ in crypto (Olszewicz, 2016) and
below shows, this is not the case. articles exploring ‘The Vomiting Camel’
pattern of bitcoin are found frequently
2. They are increasingly being (Martin, 2018) (Fadilpašić, 2018). If
used by cryptocurrency or token these articles and videos had remained
investors – Since token sales have in the purview of satire, it would not be
become mainstream, the number of an issue. But based on the arguments
exchanges that trade cryptocurrencies above, the fact that they are being used
and tokens has exploded. A number of by traders to make buy, sell or ‘hodl’,
these exchanges provide real time data decisions means that there is an urgent
of the trading activity, which is need to come up with a more nuanced,
currently being used to create charts educated and scientifically valid
and re-ignite Chartism. definition of technical analysis for this
new asset class.

2.2: Variables for Fundamental and Technical analysis of Tokens:

In light of the issues with using making an investment decision in

Technical Analysis/Chartism methods tokens. As of today, there is no DCF
and retro-fitted stock valuation models model or DDM model equivalent for
for tokens, we propose a list of variables token valuation. One of the objectives
that investors need to analyse when of our report is to therefore enumerate

 35

the main variables that future model We first start by providing a list of
builders need to consider when building fundamental variables. Even though
a model. Tables 3 and 4 enumerate the token sales are relatively new, we can
fundamental and technical variables still borrow a few lessons from
that need to be taken into fundamental stock analysis techniques.
consideration. Below is a table that explains the main
fundamental variables that need to be
reviewed and the purpose for doing so:

Table 3: Fundamental Variables for Token Valuation

Variable/Attribute What to look for How to measure it

§ Track record in the
community
§ Recognized industry
expertise in core team,
§ Mix of tech and business
advisors and board (if
expertise
any)
§ Technical expertise
§ Proven capacity in similar
§ Founders and CTO’s
projects
track records
Team § Core Technical team
§ Experience of team and
members -
advisors in managing
o Aptitude
large scale projects or
o Familiarity with
companies
codebase & language
§ Team Size
§ Work experience of team
§ R&D background in team
§ Have any of the founders
left the team recently?

§ First mover?
§ Number of incumbents
and monopolistic players
§ Ease to enter the market
based on product/service
§ Size of market (Billions
offering
USD)
§ Strategic Partnerships
§ Competition –
Market § Are their centralized
incumbents and Crypto
competitors exploring a
§ Ability to create a unique
decentralized solution? If
presence
yes, how many.
§ Attractiveness and Growth
of market
§ Sensitivity to economic
cycles

§ Infrastructure needed to
§ Problem being
deliver the
solved/Solution being
Product product/service
provided
§ Technology Readiness
§ Feasibility of Proposition
level

 36

§ MVP or Product Status6 § Complexity of

§ Time to achievement product/Service
§ Presentations, videos, § Early adopters client list
whitepaper, technical § Shared repos on Github
paper, code § Presence on social media

§ Role of the token in the

§ Level of detail in
business model
explaining product and
§ Economic and
White paper objectives
Distribution/Issuance
§ Technical description of
model
what is being made
§ Viability of achieving the
milestones in the
roadmap - needs an
§ Often published in
understanding of the
whitepaper
technical difficulties
§ Stages of the project
§ Details of access to funds
§ Budget allocation as per
(escrow) and release as
stage
per milestone
§ Marketing rhetoric &
achievement
Roadmap associated budget – is
§ Transparent reporting on
the token being
development of project
marketed as an
§ Transparent reporting on
investment with future
use of finances7 - Burn
value based on
rate, costs, gain
speculation?
predictions
§ Explanation for soft cap
and hard cap objectives8

§ Github trackers - Forks,

§ Is the code for the smart watchers, stars compared
contracts open source? to other tokens
§ If yes, look for team § Commits unique to the
contributions, project. Comparison to
Code community contributions other similar tokens
and standards being § Security Audits and Bug
respected. Bounties – Bugs found,
§ Vetting of code by Kudos points, Accuracy of
technical experts bug solutions, number of
contributors, rewards as

6
Product Status can range from – Fully working product , Beta version, Alpha version, Prototype /
MVP, Demo only, Just an Idea – to Unknown.

7
Amount being raised needs to be clearly defined and the token sale needs to end on success.
Secondary funding rounds or extension of the ICO after hitting the capital target should be watched
with caution as it be related to corrupt incentives.

8
A number of Token sales will have soft and hard cap targets. Often there is no explanation why an
additional amount of money is needed, if the product can be built with a lesser amount (the sift cap).
Expansion into other areas, vertical or scaling might be the reasons, but the validity needs to be
verified.

 37

per severity9 or CVSS

scores10
§ Independent, 3rd party
reviews of the code
(normally paid for by
token issuers)
§ ICO reviews and diligence
agencies reviews

§ Token Emission Rate

§ Trading Volume § Average trading volume
§ Allocation to founders over 3 months compared
and advisors to similar token’s
§ Sales cycle – Presale, § Average market cap over
Token Distribution Private Sale,, Airdrops, 3 months compared to
Token sale, etc. similar tokens
§ Lock up period and § Value growth since trade
Token Generation Event start date against average
(TGE) total market growth

§ Amount allocated for

marketing compared to
product development
§ Marketing message – is
value based on
speculation?
§ Aggressiveness of
marketing – overly
aggressive marketing § Community size and
raises red flags growth rate
Traction
§ Bios of team on website § Reviews and comments
and interaction mediums on online platforms
§ Community interaction
mediums
o Email
o Slack/Telegram
o Blogs and posting
frequency
o Twitter, LinkedIn,
Facebook presence

9
Bugs in code have 4 levels of severity – Critical, High, Medium, Low. Based on the severity level,
rewards for finding and solving bugs will vary. Eg: Median reward for Critical bugs is $1400 (High =
$500, Medium = $150, Low = $100). Source: Bug Bounty Field Manual by Adam Bacchus.

10
CVSS = Common Vulnerability Scoring System provides a way to capture the principal
characteristics of a vulnerability, and produce a numerical score reflecting its severity, as well as a
textual representation of that score. The numerical score can then be translated into a qualitative
representation (such as low, medium, high, and critical) to help organizations properly assess and
prioritize their vulnerability management processes.

 38

Since Token’s exist on Blockchains and of 3 technical variables – the level of

as Blockchains are distributed decentralization, the smart contract,
networked constructs, most of the and the size of the network – as they
technical variables will refer to network are key variables that concern any type
attributes. Once again, this is our of token resulting in various methods of
definition of technical analysis and we analysis being proposed by a variety of
do not refer to Chartism when talking contributors.
about technical analysis for the reasons
cited in the prior section. Basic Technical Variables - As the
crypto space becomes increasingly
The pertinence of certain technical mature, a number of attacks have
variables will depend on the type of occurred in recent times with smart
Blockchain being used (Public or Private contracts. While the attacks have
- In the past year, we have seen an affected the reputation of the space,
increasing number of Private ICOs or they also offer us guidelines on what
‘PICOs’) and the type of business model the be vary for when looking at a Smart
and the relevant token (Utility Token, Contracts code. As an large number of
Security Token, etc.….). The table ICO’s turn out to be frauds, investors
below offers a summarized collection of have begun asking for snippets of the
a few basic technical variables and lists code to ensure that the project is truly
attacks that have happened in the past viable and that the previous mistakes
that investors should verify when are not being repeated. The table below
looking at a Token Sale. Following this lists a few of these technical waypoints:
list, we present a more detailed analysis

Table 4: Technical Variables for Token Valuation

Variable/Attacks to What to look for

be verified
Basic attributes of the § Coding Language [Eg: Solidity, LLL or Serpent (in
Codebase Ethereum Blockchain), JavaScript and Python...] -
Smart contracts exist primarily in Ethereum and
Hyperledger Fabric environments. Using widely used
programming languages for writing smart contracts has
many benefits: it reduces the learning curve, attracts
more developers, and enables the usage of reliable
existing tools and libraries from these communities.
Almost every program written in Go or node.js can be
run on Hyperledger Fabric.
§ Code Version - Which version of coding language is
being used? For instance if the Smart Contract uses a
complier version from Jan 2017, it might be prudent to
understand why this has not been updated to a later
version considering that more recent versions have

 39

identified and implemented many security fixes and

improvements in the compiler11.

Identity management § Access and Control Layer

system for the Token § Identity Layer
Sale § Public/Private Keys
§ Digital Signatures and Elliptic Curve Digital Signature
Algorithm (ECDSA)

While most of the variables listed above need to be verified

individually, it is important to see what checks and controls
have been set in place with respect to identify verification
and protection (KYX/ALM). Often Token sales will involve
a custodian. Ensuring that the custodian (e.g.: for KYC
processes) is legitimate and follows a strict procedure is
therefore necessary

Scaling factors § Throughput limit

§ Latency limit
§ Cost per Confirmed Transaction (CPCT).
§ Bandwidth
§ Transaction validation

These points will be further explored in later part of the

report along with token supply

Compliance with Does the code implement many interfaces and contain a
ERC20 Standard (or lot of logic that goes far beyond the ERC20 standard? If
similar standards) sections are not in accordance with current best practice
recommendations, this should throw up warning signs. As
smart contracts are still in a development phase, they
should be easily understandable.

Compliance with Code A Smart Contracts’ code should correspond (for the most
Style Findings part) to the recommended Code Style. If a Smart Contract
contains any complex duplicate code it can lead to
diverging program logic.

Presence of negated The negated conditions can cause errors if the condition is
conditions complex and must be avoided. Simplification of the code
is the first step in this direction.

Use of modifiers Modifiers are used for recurring checks and their use
should be explicitly specified – For Eg: The use of modifiers

11
Smart Contracts exist on the Ethereum ledger in a complex, hard-to-read machine language known
as byte code. But they are most commonly written in an intuitive programming language called
Solidity. Solidity hides from developers the internal details of the Ethereum Virtual Machine and the
complex machine language that it processes. Before being uploaded to the Blockchain, a program
called a “compiler” is used to translate the Solidity source code into Ethereum byte code.

 40

in the functions and state variables. This increases the

readability of the contract and makes it more trustworthy.

Return Values of Verify that the return values of your functions are always
Functions within the range of the expected values. EG: If a function
is expected to return numbers bigger than ‘0’, it should be
tested to see that if it is being forced with a ‘0’ return, does
it reject that situation or not.

Limits of the functions If a function is returning a number, test and execute it

with
§ the biggest possible number,
§ the smallest possible number
§ a random value in the middle.
This allows us to see how the functions will react in
unexpected situations.

Format of Return If a function is supposed to return an array of numbers,

Values check if there’s any case where that array returns empty.
This is important because it could break the functionality of
your decentralized application (Grincalaitis, The Ultimate Guide to
Test Your Smart Contract , 2018).

Over and under flows Overflow and Underflow Attacks are similar to the Y2K
problem 12 . An overflow occurs when a number gets
incremented above its maximum value. This can allow a
an attacker to gain more tokens than they actually own or
in worse cases can lead to the breakdown of the entire
system. See the note in the appendix to see how this can
compromise a Smart Contract.

Reentrancy Attack
This attack consists on recursively calling the
(Checks-Effects-
call.value() method in a ERC20 token to extract the
Interactions Pattern)
ether stored on the contract if the user is not updating the
balance of the sender before sending the ether. More
recently, there have been an increasing number of issues
related to this kind of attack with ERC827 tokens13.

Reordering attack
In such an attack, a miner or other party tries to “race”
with a smart contract participant by inserting their own
information into a list or mapping so the attacker may be
lucky in getting their own information stored on the

12
Y2K was a class of computer bugs that was threatening to cause havoc during the turn of the
millennium. To keep it as simple as possible, many programs represented four-digit years with only
the final two digits. So, 1998 was stored as 98 and 1999 as 99. However, this would be problematic
when the year changes to 2000, since the system will save it as 00 and revert back to 1900.
13
ERC827 tokens. is an extension of ERC20. The three functions that are new in ERC827
are: approveAndCall(), transferAndCall(), and transferFromAndCall().
The difference between the ERC827 functions and their ERC20 counterparts is that in addition to what
they do in ERC20, they also call_to.call(_data) on the _to contract to whom the money is being
sent. allowing the attacker to buy as many tokens as he wants, bypassing the individual sales cap.

 41

contract (Grincalaitis, The ultimate guide to audit a Smart

Contract + Most dangerous attacks in Solidity , 2017).

Short address attack If the token contract has enough quantity of tokens and
the buy function doesn’t check the length of the address
of the sender, the Ethereum’s virtual machine will just add
zeroes to the transaction until the address is complete
(See example in the footnote14). This allows an attacker to
gain more tokens than they own.
NOTE: This is a bug of the Ethereum virtual machine.
Hence, when investing in tokens/purchasing tokens it is
important to check the length of the address.

GITHUB/Etherscan § Experience of the contributors - It is not the number of

related variables contributions attracted or retained, but the quality of
the contributions that needs to be analysed.

§ Gross Product Pull Requests (GPPR) - The number of

pull requests are being opened and merged is
considered by some as a more universal health metric
that can work agnostically from a project’s size. GPPR
is defined as the number of pull requests merged in a
month.

§ Regularity of commits and pull requests – Most token

projects will involve a small group of contributors.
Looking at the regality of commits and pull requests
shows the lifecycle of the project. The “latest commits”
can also be used as an indicator of whether a project is
being actively developed.

§ Analysis of the ecosystem – Along with the above listed

GitHub variables, it is also useful to analyse how the
GitHub ecosystem around the project is growing (Refer
Figure 11). As the project grows and the code is
reviewed, users will be attracted and retained if there
is sufficient confidence in. the project. High attraction
and low retention rates signal weakness. Most visitors
who have confidence, and capability, will also become
commentators and might even become part of the
project. Analysing this flow and seeing how the token
issuers deal with the inflow of comments and issues

14
A user creates an Ethereum wallet with a trailing 0,

Eg: 0xiofa8d97756as7df5sd8f75g8675ds8gsdg0

He/she then buys tokens by removing the last zero and affecting the command:

Buy 1000 tokens from account 0xiofa8d97756as7df5sd8f75g8675ds8gsdg

The virtual machine will return 256000 for each 1000 tokens bought. This is a bug of the virtual
machine that’s yet not fixed so when investing in tokens/purchasing tokens it is important to check
the length of the address.

 42

identified by observers over the time period prior to the

token’s launch another variable to worth observing.
Figure 11: Schematic of Interactions in an Open Source Ecosystem

NOTE: Green arrows represent individuals joining the ecosystem while Red arrows represent
individuals leaving the ecosystem.

Source: Modelling Open Source Software Communities. Also refer “Methodologies for measuring
project health” (Eghbal, 2018)

2.3: Detailed analysis of specific technical variables

2.3.1 Smart Contracts – Code Smart contracts are generally touted as

quality and governance of Token the means to execute programs
Supply: securely that suffer from counterparty
risks. The popular quote is “Code is
The underpinning mechanism behind Law” (Lessing, 2000) , which alludes to
any Token Sale is the Smart Contract. the concept that because something
Smart contracts are self-executing coded, and as it has a veneer of legality
programs that run on Blockchains and (steaming from the word “contract”), a
accept digital signatures to record the smart contract operates under a fire-
agreement between stakeholders. They and-forget model.
work on the basis of the IFTTT logic –
aka: the IF-THIS-THEN-THAT logic. But this definition is an overstatement
During the Token Sale, the supply of the which was best exemplified by the DAO
Tokens, the exchange of the Tokens, hack in 2016. It is important to state
the governance of the token in this critical opinion about Smart
circulation and the actual functionality Contracts for two reasons –
of the Cryptoassets they deliver is
managed by the Smart Contract.

 43

Firstly, Smart Contracts are neither This ensures that none less than the
Contracts in the actual sense and their majority of computing members own or
“smartness” is debatable. control the execution of a code running
on it. Neither the developers of the
Secondly, since these contracts manage contract codebase nor the parties of the
the supply and governance of the contract (buyers and sellers) will be
tokens, a gap between what the Token able to influence the contract execution
Sale promises and what their code to their advantage (Das Gupta, 2018)
delivers significantly impacts the
investment decision. To say that smart contracts introduce
self-executing programs to Blockchains
There is nothing specific that smart is therefore false, as the opposite is the
contracts bring to Blockchains. Rather it truth. At best, Smart Contracts can be
is the opposite — Blockchains provide defined as bits of code that interact with
objective code execution infrastructures the underlying Blockchain ledger to
as code is executed in Blockchains only govern the transmission of
if it is approved by a majority of Cryptoassets between counterparties.
computing nodes that it follows the Calling them contracts can even be
protocol (Refer the note “Anatomy of a considered misleading.
Smart Contract” in Appendix 3 to
understand the key components of a The table below showcases our current
smart contract and the link between a misconceptions about Smart Contracts
Blockchain and the different elements and sets the stage for the next level of
of a smart contract). analysis with regards to these
automated control mechanisms.

Table 5: Smart Contracts - Key Misconceptions

Aspect Confusion Reason for it Clarity

Self- Smart contracts Neither the Blockchains are

executing introduce self- developers of a fundamentally
programs executing smart contract infrastructures to
programs to codebase nor the execute code
blockchains. parties of the objectivity, something
contract are able to that is leveraged by
influence the smart contracts; you
contract execution to don’t need a smart
their advantage. contract for this, bitcoin
scripts can do this as
well

 44

The Human Code written clearly Even without bugs, the

Inevitability intervention only can accurately reflect intentions behind a
of Human to fix bugs in the contractual terms code is interpreted
Intervention smart contract differently by different
code people, requiring
humans to clarify
whether the execution
is acceptable

Humans Smart contracts Remove the Humans will always

over over-value inconsistencies of prevail over machines
Machine mechanistic human control so as the later execute at
execution and machines can run the command of
under-appreciate predictably humans, and not for
human creativity themselves.
and change. Predictability ≠
Perfection

Law over Code is Law Code is blind, so is Code is like the codified
Code Law laws of physics, not the
laws of contracts

Source: Smart Contracts: Fulfilling Nakamoto's Dreams (Das Gupta, 2018). Table republished with
permission from the author.

From a valuation context, Smart Second, in terms of how they are

Contracts need to be verified at two managing the supply and exchange of
levels – First, at the level of the quality tokens during the lifecycle of a token
of code to verify the integrity and sale. The following two sub-sections
security of the smart contract; and explore these attributes in more detail.

2.3.1.1: Key Variables for Integrity and Security of Smart Contracts

§ Contract Integrity is sent. Unrestricted upgradability

violates contract integrity, which is a
Regarding contract integrity, a key key feature of smart contract.
factor is unrestricted upgradability: If
present, contracts can be upgraded in If contract upgradability is required a
arbitrary ways and hence no assurance design has to be chosen which
can be given to a user what code will preserves some trust and security
actually be executed when a transaction guarantees.

 45

§ Data Integrity & Contract Security o Undesirable Transaction Orders

o Inconsistent Contract States
General concerns for data integrity and during control flow transfers
contract security are:
Table 6 enumerates the possible issues
o Incorrect Authorization regarding data integrity and contract
o Missing Authentication security in greater detail:
o Insufficient Numerical Precision

Table 6: Smart Contracts - Data Integrity & Contract Security variables

Variable What to look for?

Ether transfers whose execution can be manipulated

Transaction Order Affects
by other transactions must be inspected for
Execution of Ether
unintended behavior.
Transfer

The receiver of ether transfers must not be influenced

Transaction Order Affects
by other transactions.
Ether Receiver

Transaction Order Affects The amount of ether transferred must not be

Ether Amount influenced by other transactions.

Gas-dependent Calls into external contracts that receive all remaining

Reentrancy gas must not be followed by writes to storage.

Reentrancy with constant Ether transfers (such as send/transfer) must not be

gas followed by writes to storage.

Unrestricted write to Contract fields that can be modified by any user must
storage be inspected.

Writes to storage should be used by the contract,

Unused write to storage
otherwise they are unnecessary.

The return value of statements that may return error

Unhandled Exception
values must be explicitly checked.

Division Before The use of division before multiplication may result in

Multiplication incorrect final results due to integer rounding.

The use of division to calculate the amount of

Division influences
transferred ether may be incorrect due to integer
Transfer Amount
rounding.

 46

The execution of selfdestruct statements, which

Unrestricted
remove the associated contract from the blockchain,
Selfdestruct
must be restricted to an authorized set of users.

Method arguments must be sanitized before they are

Missing Input Validation
used in computations.

The origin statement must not be used for

Use Of Origin
authorization.

The execution of ether flows should be restricted to an

Unrestricted ether flow
authorized set of users.

Contracts that may receive ether must also allow users

Locked Ether
to extract the deposited ether from the contract.

Unsafe Call to Untrusted The target of a call instruction can be manipulated by

Contract an attacker.

Unsafe Dependence On Security-sensitive operations must not depend on

Block Information block information.

Unsafe Dependence On Security-sensitive operations must not depend on gas-

Block Gas related information.

Delegatecall dependent The target and arguments provided to delegatecall

on User Input must be sanitized.

Understanding the nuances and this plan? If yes, have these

limitations of smart contracts from a restrictions being encoded into the
technical perspective sets the stage for smart contract?
the determining if the smart contract is
3. Did the token issuers retain the
capable of governing an ICO and is key
to asking the following evaluation power to modify the smart-contract
related questions (Cohney egt al., code governing the tokens they
2018): sold, and if so, did they disclose that
1. Does the Whitepaper state any that power?
restrictions on the supply of tokens
and are these restriction encoded in
the smart contract? Understanding the supply dynamics is
2. Is there a vesting or lock-up plan for of key importance for token valuation.
insiders and are there restrictions to If stock prices reflect (approximately)
transferring the tokens related to the net present value of the expected

 47

future cash flows, then a token’s price § Security risk: It requires the key
should reflect an equilibrium between holder(s) to secure the private key. An
token demand (driven by the present attacker (i.e., a disgruntled employee
value of expected future exchange - A number of Token hacks turn out to
options within the token’s native be inside jobs) could hold the network
ecosystem) and token supply (driven hostage and demand a ransom by
by the token’s monetary policy) . freezing transfers, or short the token
Moreover, this blatant belief in “code is which is sure to drop in value.
law” hinges on the belief that the way
the Smart Contract (the governing This tendency of not disclosing in plain
structure in a Token Sale) is coded will terms that the issuers of the token sale
ensure Trustless Trust. have the power to modify the token
rights is probably one of the biggest
However as we have seen, the current limitations with smart contract based
situation regarding trustless trust in token sales. As of the time of publishing
Smart Contracts is far from optimal. this report, it can be stated that ICO
Furthermore, a review of the 50 top code and ICO contracts rarely match .
grossing ICO’s revealed that a In a report titled, ‘Coin-Operated
significant fraction of issuers retained Capitalism’, researchers from the
centralized control… “and did not University of Pennsylvania explored this
disclose code that permitted the discrepancy was found that many ICO’s
modification of the entities governing failed even to promise that they would
structures” (Cohney egt al., 2018). And protect investors against insider self-
while many think of Ethereum contracts dealing and fewer manifested such
as fully decentralized, nearly half of the contracts in code. The text below
top 20 projects15 can have their token summarizes some of their findings with
transfers completely frozen by an regards to token supply and code in
owner (a single key or a multisig smart contracts (Refer “Anatomy of a
contract) (Que, 2018) . This process Smart Contract” in Appendix 3 in case
known as Pausing can be valuable for you wish to familiarize yourself with
future upgrades, swaps, and disaster some of the basic functions of smart
mitigation. But it also leads to new contracts).
risks:
The researchers obtained a copy of the
§ Trust: It requires all users to trust the Solidity code from etherscan.io 16 or
party in charge of the key, reducing GitHub for the fifty top grossing ICOs.
the degree of decentralization in the Each function of the smart contract was
contract. We will cover how to manually tracked to see how each line
measure level of decentralization in modified the meaning of, or data stored
the next specific technical variable. in, the smart contract

§ With regards to Minting

15 16
These include: EOS, Tron, Icon, OmiseGo, Etherscan.io replicates the byte code present
Augur, Status, Aelf, and Qash on the blockchain, but requires developers to
upload Solidity source code for display

 48

how much can be sold during

o Unlike mining, where each phase and these limits are
contributors need to work to hardcoded in the smart contract.
earn cryptoassets, minting is a The sales process is thus
process in which tokens are automatically executed by the
issued in exchange for another Smart Contract, with the full
cryptoasset. supply being decided at the
offset or depend on how much
o The tokens can be allocated to investment the project receives.
investors in exchange for fiat or
cryptocurrency via a private o However these limits can be
sale. In such a sales process, the modified by the owner of the
supply quantity of tokens is smart contract which poses a
decided at the start of the threat to investors. If there is no
project and investors have a set limit, which is encoded,
fixed quantity of tokens that token issuers can mint a private
they can acquire. stash for themselves or inflate or
deflate the circulating float of
o Another model involves the tokens. The encoded supply
issuance of tokens to the general restrictions are therefore critical
public via mass offerings until a to investors.
predefined target is met. When
that target is reached, the sale o The report from the researchers
stops. found that 90% 17 of the top
projects had stated supply
o We also have a combination of restrictions in their documents
private sale and mass offering. and 75% had coded it. The
Based on the issuing phases – graph below summarizes their
private presale followed by mass results:
offering – a cap is set that limits

Figure 12: Smart Contracts with encoded supply limits (47 of top 50 ICOs)

Data source: Coin Operated Capitalism, Cohney et al., (2018)

§ With regards to Decreasing
Supply or ‘Burning’ –

17
Of the 50 projects analysed, 3 projects values represents these 47 projects. Hence
expressed their code in byte code. Hence 47 90% = 42/47 and 75% = 32/42.
projects were auditable and the percentage

 49

o Not all cryptoassets exist in a tokens from the private presale

state of continuous circulation or public sale.
(e.g.: Bitcoin/Ether).
o Finally, some projects (such as
o They can also be ‘burned’ or Paragon) describe a complicated
destroyed as they are used up. transaction fee structure, where
For example - a token can be half of the transaction fee (in
used to access a right in a this case $0.000000005) is
completed project at a future burned and the rest is used to
date (utility token). replenish the token reserve

o When this occurs, the token o In spite of the importance of this

would be burned and this function in the supply of the
changes the amount of tokens in tokens, the researchers found
circulation, which effects its that 35% (of the 47 reviewed)
price. had not hardcoded the burning
process in their smart contracts.
o Some projects also describe Even some that did, had done so
plans to perform token buy with errors which could cause
backs from holders to burn them the eventual demise of the
and appreciate the price of the network18.
token. Others do the same thing
Figure 12 summarises their
by promising to burn unsold
results:

Figure 12: Smart Contracts with encoded Burning rules (47 of top 50 ICOs)

Data source: Coin Operated Capitalism, Cohney et al., (2018)

§ With regards to Vesting o Vesting is a common investing

practise of providing key

18
Reference made here to Paragon’s PRG burned. After a sufficient number of
token. The researchers found that when they transactions the fee approached the number
modelled the transaction fee system, each of tokens remaining in the supply, causing the
transfer of a PRG token consumed eventual demise of the network
approximately one-six billionth of the total
supply in transfer fees, half of which was paid
to the token issuers and half of which was

 50

members of the entrepreneurial can also be governed offline and

team with equity options/shares enforced using traditional tools
in future profits in order to like corporate charters and
counteract against desertion. It bylaws.
acts as an incentive structure to
act against ‘founder dumps’ and o The researchers found that only
the threat of desertion. 37 of the 47 audited projects
promised vesting in their
o As with supply promises and whitepapers and 31 of these 37
burning, vesting promises are did not even code those vesting
detailed in the whitepaper and rules into their tokens. Figure
ought to be encoded in the 13 summaries their results:
smart contract. However, this

Figure 13: Smart Contracts with encoded Vesting rules (47 of top 50 ICOs)

Data source: Coin Operated Capitalism, Cohney et al., (2018)

§ With regards to Modifiability rights of the token holders can

be significantly impacted by this
o Modifiability is a relatively new (especially for utility tokens), it
phenomenon in ICOs. It is a key component for investors
essentially involves the ability to which should bear heavy when
change the code regarding investing a token with set
certain aspects of the token in functionalities. Nevertheless the
order to provide new features to researchers found that only 10
the token holders. Essentially an of the 50 ICO’s reviewed
upgrade. allowed modifiability in their
code. Only 7 of these discussed
o Modification changes normally modifiability in their
occur through a voting system – whitepapers and out of these 7,
changes are proposed, token only 4 had hardcoded these
holders vote within a given time rules. Figure 14 summarises
frame and then the changes are their results:
executed. As the function and

Figure 14: Smart Contracts with encoded Modifiability rules (47 of top 50
ICOs)

 51

Data source: Coin Operated Capitalism, Cohney et al., (2018)

Just as paper contracts, smart contracts Lifecycle need to reviewed thoroughly

reflect the institutions within which they in whitepapers and in code, to ensure
are produced. The researchers found investors know what they are getting
that “there are systemic differences into.
between code and contract, even within
projects that have attracted significant Some solutions to this problem are
investments”. being provided by firms such as
ChainSecurity who provide automated
While reviewing code can be a audits of the smart contract code.
roadblock for non-technical investors, it Token reviews from 3rd party providers
is important to ensure that the smart need to be considered with care as
contracts are feasible and whether all some rating platforms operate on a
transaction arguments detailed meet pay-to-play model (as seen with the
their desired preconditions. Aspects rating platform ICO Bench (Devoe,
regarding the Token and Crowdsale 2018)).

2.3.2: Level of Decentralization

The level of decentralization of a project private Blockchain or a consensus

is dependent on the nature (Private or mechanism in which only certain parties
Public Blockchain) of the project. If the have the authority to verify and validate
project is on a public Blockchain and is transactions – E.g.: Proof of Authority19
more aligned to the open source – then a lower level of decentralization
principles, then a higher level of might make more sense.
decentralization would be appreciated.
However if the project involves a

identity. By staking this identity to secure the

19
Proof of Authority (PoA) is a form of Proof network in exchange for the block rewards, a
of Stake, in which a set of certain validator is dis-incentivized to act maliciously
“authority” nodes are explicitly allowed to or to collude with other validators. The
create new blocks and secure the Blockchain. advantage of using PoA is very high transaction
In PoA, a validator is not required to hold a rates
stake in the network. He or she is required,
however, to have a known and verified

 52

Understanding the level of i. The Herfindahl Score or the

decentralization is important as it plays Herfindahl–Hirschman Index: The
a key role in the governance structure HHI is a metric used to measure
competition and market
of decentralized projects, as it shows
concentration. It is calculated by
the fragmentation of control. In the taking the percent market share of
context of token evaluation, knowing each entity, squaring it, and summing
the level of decentralization is useful in the squares before multiplying by
analysing the developer community, or 10,000. HHI scores less than 1,500
the client base, or even exchange qualify as competitive, 1500-2000 is
decentralization to see how and where moderately concentrated, and greater
than 2500 is highly concentrated
the token is being traded. Thus
measuring decentralization helps us see ii. The minimum Nakamoto
the size of participants in relation to the coefficient :
ecosystem and reflects the competition In mid – 2017, an article titled
among them. Quantifying Decentralization was
published that proposed a new
However, measuring decentralization is coefficient for measuring
easier said than done. While, some decentralization in crypto networks.
significant advancements have been Inspired by income distribution and
economic inequality measuring
made by a few researchers (notably
methods, the authors used the Lorenz
Gencer et al, (2018) and Srinivasan, Curve and the Gini Coefficient to come
(2017)), they have been done in the up with a cumulative scoring
context of cryptocurrencies and not methodology.
necessarily for tokens. But the same
variables that are used to measure the First, a crypto economic network was
decentralization of cryptocurrencies broken down into its independent
subsystem components [Mining
(Provisioned Bandwidth, Network
(measured by reward), Client developer
Structure, Distribution of Mining Power, (measured by codebase and commits),
Mining Power Utilization) cannot be Exchanges (measured by Volume),
used for evaluation of tokens as most Nodes (measured by Country),
tokens run on top of an existing Ownership measured by Addresses)]
Blockchain (e.g.: Ethereum and ERC20 and a Gini score was given to each
component to measure their level of
type tokens). It could be useful to know
decentralisation. The cumulative score
the level of decentralization of the
of all components thus gives a
Blockchain being used. But the same ‘Maximum Gini coefficient”.
variables cannot be used to evaluate
the level of decentralization concerning The Nakamoto coefficient then acts as a
the token. comparative score. It measures the
minimum number of entities in a given
Hence, when looking at a token project subsystem required to get to 51% of
knowing how distributed the developer the total capacity. Aggregating this
measure by taking the minimum of the
community is, the spread of the client
minimum across subsystems provides
base and the variety of exchanges it is
the “minimum Nakamoto coefficient”,
being traded on, can aid in providing which is the number of entities needed
some conception of the diversity of the to compromise in order to compromise
project. Measuring the decentralization the system as a whole.
of such market-based variables can be
addressed using 2 methods:

 53

Both decentralization measuring trying to assess the level of

techniques are limited in usefulness but fragmentation in decentralized projects.
nevertheless can act as guidelines when

2.3.3 Network Size:

Our final technical variable is the size of Like a physical coin, a cryptoasset is
the network. The reason for paying scarce, and its ownership is
special emphasis on this variable stems transmittable. But while physical coins
from the fact that there exists a large are transmitted from hand-to-hand (or
amount of empirical evidence that hand-to machine), changes in control of
establishes the relationship between cryptoassets occur through the
Network size and Network Value networks that host them.
(Metcalfe, 2013) (Zhang et al., 2015).
This is especially pertinent for Utility Cryptoassets with a utility function
therefore resemble micro-economies,
tokens which make up a significant
as the token can be seen as another
portion of ICOs. currency used in the economy that
specializes in the exchange of a
While the previous attempts at valuing particular service. Token sales thus
utility tokens have made important enable the creation of private
contribution to this new asset class, transactional economies (Mougayar,
more focus needs to be given to the 2017) and as stated by Primoz Kordez,
activity of the network, represented by co-founder of D2 Capital,
the velocity of token’s exchange.

“This turns classic enterprise valuation upside down, since we

don’t value cash flow to investors but use of the token by
consumers. In token evaluation, we want to measure how large
and active the exchange of service is, reflected in the
aggregate value of transactions (can be in $ terms) and the
number of transactions of each token represented by its
velocity….”….Thus, when valuating utility tokens, we should
value the network and “account for the velocity of tokens. The
more tokens circulate between users, the more interactions,
the more network effect and healthier network”. (Kordez, 2017)
When considering the sensible operational and the token is being used
statement above, it is important to to facilitate the underlying business
remember that while a number of utility model. Currently, the token velocity
token ICOs have been funded, the being measured is based on future
actual product or service that is being speculations of expected utility.
provided will only be available at a
future date. Hence measuring the token Hence, when a token-based product or
velocity in the context of usage of the service will be launched, then the
associated product or service, will only velocity of this token’s exchange will
be truly measurable and representative provide us with a way of estimating it
of its usefulness when they are usage and value to the end-user. When

 54

this occurs, we would estimate the from the most valuable to the least,
value of the token, based on the weighting the links by their inverse rank
estimation of the value of the service. would yield an average number of
The latter has value only if made connection of log(N) for each node and
accessible. Thus, connections are therefore Nlog(N) connections in total.
paramount for making this estimation The latter representation of a network
as tokens are network native entities. If is often referred as the Zipf’s law.
each participant using the service can
also offer it to others, its value will be Both these network models are extreme
tightly linked to the number of scenarios. If we are to adhere to
participants and their interactions. Metcalfe’s Law, each new participant in
the network instantly connects with all
Using this approach, one could model other nodes. If we are to adhere to
the value of a token in the following Zipf’s law, the average number of
way: connections per node grows
logarithmically, which means that the
(The value of the service) * (The increase in connectivity for an
accessibility of the service) additional participant in the network is
the least positive one.
where accessibility is a function of the
network of participants in the Depending on the maturity of the
blockchain. network, its average number of
connections per node can thus be
Thus, valuating a token based on its
modelized as a function of the number
velocity requires us to evaluate the
network and the service being provided. of participants as Na where the scale
There are various methods that are factor a would lie between 0 and 1. The
currently used for valuating services. To total number of connections in the
conclude our explanation regarding this Network will therefore be of the order
technical variable, we offer some 1+a : N1+a.
insight on estimating the value of a
network. Hence, estimating the connectivity of
the network would come down to
Network Valuation: estimating the exponent of the average
network connections, N1+a, to match
A network is a connected graph with the market capitalization evolution
nodes linked to others via the network. (using transactions a proxy) to a
If a network were to be completely function of the number of market
linked, such that each node is participants (i.e. the N nodes).
connected directly to all others, as per
Metcalfe’s Law, for N nodes, there NOTE: A few papers published in 2017
would be N(N+1)/2 links. (e.g.: See Alabi, 2017) have analysed
historical BTC data to come up with an
Nevertheless, in any network, all links exponent estimation of approximately
do not share equal importance. N1.5. However, these results are
Therefore, to take this into account we pertinent to cryptocurrency networks
would need to give each connection a and the same estimations cannot be
weight to reflect its importance. transposed to tokens owing to their
multi-functionality. We conclude by
For instance, a fully connected network stating that while these methods offer
would have N(N+1)/2 links of equal insight, without empirical data, it would
importance, in a hierarchical network be hard to come up with an token value
where each node is connected to all estimation model based on this
others but with an importance ranking approach.

 55

CONCLUSION

As we have seen in this report, the exchange mechanisms, they can act as
current situation regarding token accounting units in the creation of
valuation models is a situation that is economic systems. As we have seen,
clearly in need of greater study and the supply of tokens and the
testing. Moving forward we see three governance of their circulation is a key
topics that need to be explored. element when evaluating a project.
Thus moving forward, advances from
First, there is the subject of the endogenous monetary systems need to
adoption curve. While the promises and be integrated into token valuation
ideas stated in whitepapers are models to respect their fundamental
ambitious and pioneering in some nature. Insights from monetary theory
cases, history has shown us that and monetary and fiscal policy should
building an innovative new product or play an increasingly important role in
service that resolves a problem does the token valuation methods of the
not always translate to large scale future.
adoption. Indeed, even applications
such as WhatsApp, Facebook and Lastly, the variety of tokens and their
Telegram had to wait for three years on multi-functionality needs to be given
average to achieve their mass adoption more importance. As every token
hockey stick curves, in spite of the fact project is different, valuators need to
that consumers were used to mobile develop a modular approach in which
texting for many years. Most ICO they build models based on the type of
projects are still in the phase of project, the function of the token and
development. Once these decentralized how this relates to its supply dynamics.
products and services are made There is no one size fits all model for
available, the furnishers of the company token valuation. Based on the type of
behind the project will have to deal with token being analysed, model builders
switching costs, customer feedback will need to select the most pertinent
loops and other such factors that will variables and create models that
determine their success and give us a respect the nature of the project.
true gauge of the potential behind these
innovative solutions being provided. As As we have seen, the current practise of
this process occurs, there will some retrofitting stock valuation methods has
variables that hold more gravitas limited applicability. It has been our
compared to others and it is only by attempt to go back to the basics and
continuously analysing this evolution start the conversation on token
that model makers will be able to valuation with a fresh lens by focusing
determine what needs to be considered, on the variables of analysis. We are
and to what extent, when building a certain that moving forward, this list of
token valuation methodology. variables will grow owing to the
complex nature of this technology and
Secondly, as tokens function on this new investment vehicle.
blockchains which can be used as value

 56

Appendix 1

Taxonomy of Tokens

Source – Cryptoassets by C. Burniske and Tatar (2017)

1. Crypto – currencies:
§ Perform the 3 functions of money
§ Is network specific and can be forked
§ Differences in supply schedules, Proofs (Work, Stake, Existence, Elapsed Time,
Process, etc…)
§ Public or Private == Traded on exchanges
§ Examples – Bitcoin, Ripple, Dash, Ether, Monero, ZCash, etc…

2. Crypto – Commodities:
§ Represent digital commodities used to make digital goods & products.
§ The main digital goods and services that are considered crypto-commodities
include – Computing Power, Storage Capacity, Network bandwidth, Transcoding
and Proxy Re-encryption.
§ Examples -
o Computing Power (Ethereum)
o Storage Capacity (Storj)
o Network Bandwidth (Privatix)
o Transcoding = MP4 to MP3 (Transcodium)
o Proxy Re-encryption =email forwarding (NuCypher)

3. Crypto – Tokens:
§ Built on a robust crypto-currency and crypto – commodity infrastructure
§ Can issue a token which is intrinsically linked to a digital service or product
(dAPP / ERC20)
§ Can be part of its own Network and Blockchain Eg: Waves)
§ If based on a network (eg: Ethereum), will pay the network for certain kinds of
transactional operations.

 57

Appendix 2: Framework showing the different types of tokens

Image Source : The Token Classification Framework: A multi-dimensional tool for understanding and classifying
crypto tokens (Euler, 2018)

 58

Appendix 3: Anatomy of a Smart Contract

Source: The Anatomy of ERC20: What’s on the Inside of Ethereum’s Most Popular
Contract (Nash, 2017), Anatomy of a Smart Contract (Jones, 2017)

Ethereum Request for Comments 20, or ERC20, is an Ethereum Improvement Proposal

introduced by Fabian Vogelsteller in late 2015. It’s a standard by which many popular
Ethereum smart contracts abide. It effectively allows smart contracts to act very
similarly to a conventional cryptocurrency like Bitcoin, or Ethereum itself. In saying
this, a token hosted on the Ethereum blockchain can be sent, received, checked of its
total supply, and checked for the amount that is available on an individual address.
This is analogous to sending and receiving Ether or Bitcoin from a wallet, knowing the
total amount of coins in circulation, and knowing a particular wallet’s balance of a coin.
A smart contract that follows this standard is called an ERC20 token.

Smart Contract code (Solidity in this case) contains four major types of entities:
variables, functions, events, and modifiers.

§ Variables are the data storage component of any smart contract and, in the case
of a token’s smart contract, store balances for each user-address, along with
other data required for the smart contract to operate.
§ Functions describe the rules by which the smart contract operates, storing
discrete chunks of code that perform specific tasks. Functions are executed (or
“called”) by sending a specially formatted transaction to the Ethereum network.
Functions are identified by a name and a set of parameters or “arguments,” that
are the inputs to the function.
§ Events are signals that a smart contract sends to other applications or smart
contracts programmed to receive them—acting as a form of logging.
§ Modifiers allow a developer to easily restrict the execution of a function under
certain conditions. For example, a developer may restrict the ability to mint new
tokens to the smart contract owner alone.

Analysing each function of the smart contract allows us to track how each line modifies
the meaning of, or data stored in, the smart contract. ERC20 defines the functions
balanceOf , totalSupply , transfer , transferFrom , approve , and allowance . It also
has a few optional fields like the token name, symbol, and the number of decimal
places with which it will be measured.

§ totalSupply(): Although the supply could easily be fixed, as it is with Bitcoin, this
function allows an instance of the contract to calculate and return the total amount
of the token that exists in circulation.
§ balanceOf(): This function allows a smart contract to store and return the balance
of the provided address. The function accepts an address as a parameter, so it
should be known that the balance of any address is public.
§ approve(): When calling this function, the owner of the contract authorizes, or
approves, the given address to withdraw instances of the token from the owner’s
address. Here, and in later snippets, you may see a variable msg . This is an implicit
field provided by external applications such as wallets so that they can better
interact with the contract. The Ethereum Virtual Machine (EVM) lets us use this
field to store and process data given by the external application.
§ transfer(): This function lets the owner of the contract send a given amount of
the token to another address just like a conventional cryptocurrency transaction.

 59

§ transferFrom(): This function allows a smart contract to automate the transfer

process and send a given amount of the token on behalf of the owner. Seeing this
might raise a few eyebrows. One may question why we need both transfer() and
transferFrom() functions.

Consider transferring money to pay a bill. It’s extremely common to send money
manually by taking the time to write a check and mail it to pay the bill off. This is like
using transfer() : you’re doing the money transfer process yourself, without the help
of another party. In another situation, you could set up automatic bill pay with your
bank. This is like using transferFrom() : your bank’s machines send money to pay
off the bill on your behalf, automatically. With this function, a contract can send a
certain amount of the token to another address on your behalf, without your
intervention.

Token Name, Token Symbol (Ticker) and Number of Decimals (normally 18 with ERC20
Tokens) are optional. The image below summarizes the points above and shows us
how an ERC20 contract looks like:

Source: What Is ERC20? | Everything You Need To Know About ERC20? (Singh, 2018)

Smart contracts exist on the Ethereum ledger in a complex, hard-to-read machine

language known asbyte code. But they are most commonly written in an intuitive
programming language called Solidity (Serpent and others are supported as well)
where data structures, functions for business logic, and authorization based on
addresses are checked. The source code is compiled into bytecode, and deployed to
all nodes on the blockchain for execution. When a DApp is configured properly, it sends
a message or transaction to a function of the corresponding smart contract. To do
that, it needs the ABI (Application Binary Interface) to correctly format the message
and digitally sign it for submission. Once the message is received by a node on the
network, it is replicated to all the nodes on the network for execution.

Unfortunately, the initial approach presents challenges that are often difficult for DApp
(Distributed Application) developers. A DApp’s presentation logic has dependencies at
runtime, such as an address of a node on the network (DNS, IP, URI), as well as a
port to communicate with. The DApp also needs to know the Ethereum address of the
smart contract that is deployed on the blockchain, which is not easily discoverable.
Finally, it also needs access to secure private keys, which can be manually inserted by
using a file, a blockchain wallet, or a secure device.

 60

To pull business logic up above the blockchain to a separate middle layer, the logic
code needs access to a variety of services, including secure execution, attestation,
identity, cryptographic support, data formatting, reliable messaging, triggers, and the
ability to bind that code to schema in specific smart contracts on any number of
blockchains. Those services can be provided in a fabric, where the individual pieces of
code that support the smart contracts can execute, send transactions to Blockchain
nodes, and be bound to the schema in the data tier.

To get a clearer picture of how this separation of concerns is achieved, we can separate
out the different portions of a smart contract into discrete components. These basic
components are the properties (static and variable), the logic and the ledger. Each of
these components can be mapped directly into technical concepts. Properties
represent a data schema, logic represents code, and the ledger corresponds to a
database. Once each of these components are defined, they can be deployed to
environments that are optimized for their function.

The smart contract is now a composite of the on-chain Solidity smart contract that
defines the data schema on the blockchain, and a Cryptlet20 that hosts the logic for
the smart contract. These Cryptlets can be run on a different computer or the cloud,
rather than the actual nodes, and as a result, do not need to be executed by every
node on the network. Cryptlets execute in a secure computational environment, and
have the cryptographic primitives that allow them to work directly with blockchains,
thereby extending smart contracts off the blockchain within the same security
envelope.

20
To pull business logic up above the Blockchain to a separate middle layer, the logic code needs
access to a variety of services, including secure execution, attestation, identity, cryptographic
support, data formatting, reliable messaging, triggers, and the ability to bind that code to schema in
specific smart contracts on any number of Blockchains. Those services can be provided in a fabric,
where the individual pieces of code that support the smart contracts can execute, send transactions
to Blockchain nodes, and be bound to the schema in the data tier. We refer to these code blocks as
Cryptlets, and the execution environment they run is called the Cryptlet Fabric

 61

BIBLIOGRAPHY
Adem Efe Gencer, S. B. (2018). Decentralization in Bitcoin and Ethereum Networks. Initiative for Cryptocurrencies
and Contracts (IC3) Computer Science Department, Cornell University & Electrical Engineering
Department, Technion.

Alabi, K. (2017). Digital blockchain networks appear to be following Metcalfe’s Law . Electronic Commerce
Research and Applications, 23-29.

Antos, J. (2018, March 6). An Efficient-Markets Valuation Framework for Cryptoassets using Black-Scholes Option
Theory. Retrieved from Medium: https://medium.com/therationalcrypto/an-efficient-markets-
valuation-framework-for-cryptoassets-using-black-scholes-option-theory-a6a8a480e18a

Arthurs, J. (2018, June 6). Authers’ Note: Vorsprung dürch Technical Analysis. Retrieved from Financial Times:
https://www.ft.com/content/52b3ccea-6930-11e8-b6eb-4acfcfb08c11

Autonomous NEXT. (2018). Crypto Utopia.

Burniske, C. (2017, August 12). The Crypto J-Curve . Retrieved from Medium:
https://medium.com/@cburniske/the-crypto-j-curve-be5fdddafa26

Burniske, C. (2017, August 12). The Crypto J-Curve . Retrieved from Medium :
https://medium.com/@cburniske/the-crypto-j-curve-be5fdddafa26

Burniske, J. M. (2018). [placeholder] Thesis Summary. Retrieved from Placeholder:

https://ipfs.io/ipfs/QmZL4eT1gxnE168Pmw3KyejW6fUfMNzMgeKMgcWJUfYGRj/Placeholder%20Thesi
s%20Summary.pdf

CB Insights. (2018, Feburary 22). 19 Corporations Working On Blockchain And Distributed Ledgers . Retrieved
from CB INSIGHTS: https://www.cbinsights.com/research/organizations-corporates-test-blockchains-
distributed-ledgers/

Cimpanu, C. (2018, March 29). 81% of Recent ICOs Were Scams, Research Finds . Retrieved from Bleeping
Computer: https://www.bleepingcomputer.com/news/cryptocurrency/81-percent-of-recent-icos-
were-scams-research-finds/

CNBC. (2013, Nov 3). BK explains gold's 'vomiting camel' pattern . Retrieved from CNBC:
https://www.cnbc.com/video/2014/11/03/bk-explains-golds-vomiting-camel-pattern.html

David Larrabee, C. (2013, March 6). Can Technical Analysis Boost Portfolio Returns? . Retrieved from Enterprising
Investor - CFA Institute : https://blogs.cfainstitute.org/investor/2013/03/06/can-technical-analysis-
boost-portfolio-returns/

David M. Smith, C. F. (2013). Head and Shoulders Above the Rest? The Performance of Institutional Portfolio
Managers Who Use Technical Analysis . Money Science: Finanical Intelligence Network.

Devoe, R. (2018, June 19). Bogus ICO Ratings Easy to Buy, Reveals Alethena . Retrieved from Blockonomi:
https://blockonomi.com/alethena-fake-reviews/

Eghbal, N. (2018, July 18). Methodologies for measuring project health. Retrieved from
https://nadiaeghbal.com/project-health

 62

Euler, T. (2018, January 18). The Token Classification Framework: A multi-dimensional tool for understanding and
classifying crypto tokens. Retrieved from Untitled INC: http://www.untitled-inc.com/the-token-
classification-framework-a-multi-dimensional-tool-for-understanding-and-classifying-crypto-tokens/

Fadilpašić, S. (2018, April 19). The Vomiting Camel Spotted in the Market . Retrieved from cryptonews:
https://cryptonews.com/news/the-vomiting-camel-spotted-in-the-market-1616.htm

Financial Stability Board. (2018). Crypto-assets: Report to the G20 on work by the FSB and standard-setting
bodies. Financial Stability Board.

Godbole, O. (2018, March 28). Why Bitcoin's 'Death Cross' May Be a Bear Trap . Retrieved from Coindesk:
https://www.coindesk.com/why-bitcoins-death-cross-may-become-a-bear-trap/

Grincalaitis, M. (2017, September 17). The ultimate guide to audit a Smart Contract + Most dangerous attacks in
Solidity . Retrieved from Medium: https://medium.com/@merunasgrincalaitis/how-to-audit-a-smart-
contract-most-dangerous-attacks-in-solidity-ae402a7e7868

Grincalaitis, M. (2018, May 9). The Ultimate Guide to Test Your Smart Contract . Retrieved from Medium:
https://medium.com/@merunasgrincalaitis/the-ultimate-guide-to-test-your-smart-contract-
ddc65fbb5ba5

Gupta, A. D. (2018, July 3). Smart Contracts: Fulfilling Nakamoto's Dreams . Retrieved from Stratumn Thinking:
http://blog.stratumn.com/smart-contracts-the-nakamoto-way/

Jones, S. (2017, Feburary 14). Anatomy of a Smart Contract. Retrieved from Blockchain Expo World Series:
https://www.blockchain-expo.com/2017/02/blockchain/anatomy-smart-contract/

Klaus, I. (2014). Forging Capitalism: Rogues, Swindlers, Frauds, and the Rise of Modern Finance. Yale University
Press.

Koffman, T. (2018, April 13). Your Official Guide to the Security Token Ecosystem . Retrieved from Medium:
https://medium.com/@tatianakoffman/your-official-guide-to-the-security-token-ecosystem-
61a805673db7

Kordez, P. (2017, September 4). Network Output and Velocity of Tokens . Retrieved from Medium:
https://medium.com/d2-capital/network-output-and-velocity-of-tokens-da7e800ca4c0

Lacalle, D. (2017). Escape from the Central Bank Trap: How to Escape From the $20 Trillion Monetary Expansion
Unharmed. Business Expert Press.

Lannquist, A. (2018, March 7). Today’s Crypto Asset Valuation Frameworks. Retrieved from Medium:
https://blockchainatberkeley.blog/todays-crypto-asset-valuation-frameworks-573a38eda27e

Lessing, L. (2000, January 01). Code Is Law . Retrieved from Harvard Magazine:
https://www.harvardmagazine.com/2000/01/code-is-law-html

Martin, K. (2018, April 19). The truth behind the vomiting camel graph . Retrieved from Financial Times:
https://www.ft.com/video/2452d265-ef0e-4f7f-b0b2-ca1dc076ff33

Martin, K. (2018, April 19). The Vomiting Camel has escaped from Bitcoin zoo . Retrieved from Financial Times:
https://ftalphaville.ft.com/2018/04/19/1524110400000/The-Vomiting-Camel-has-escaped-from-
Bitcoin-zoo/

Metcalfe, R. (2013). Metcalfe's Law after 40 Years of Ethernet . ACM Digital Library, 26-31.

 63

Milano, A. ( 2018, July 23). An $8 Million Airdrop Ran Out of Tokens – What's Next Is Anyone's Guess. Retrieved
from Coindesk: https://www.coindesk.com/8-million-airdrop-cryptocurrency-run-out-tokens/

Mougayar, W. (2017, July 17). Retrieved from Twitter:

https://twitter.com/wmougayar/status/885444010550210561

Murphy, H. (2018, July 23). Pumping and dumping in crypto markets . Retrieved from Financial Times:
https://www.ft.com/content/61ddfea4-8e72-11e8-bb8f-a6a2f7bca546

Nash, G. (2017, August 25). The Anatomy of ERC20: What’s on the Inside of Ethereum’s Most Popular Contract.
Retrieved from Medium: https://news.earn.com/the-anatomy-of-erc20-6ab09d4206a5

Olszewicz, J. (2016, November 3). My Ichimoku Cloud Settings For CryptoCurrency . Retrieved from Youtube:
https://www.youtube.com/watch?v=5x0r-qcGoQQ

Pei, A. (2018, March 28). Bitcoin is nearing a 'death cross' on the charts. Here’s what it means . Retrieved from
CNBC Markets: https://www.cnbc.com/2018/03/28/bitcoin-is-nearing-a-death-cross-on-the-charts-
heres-what-it-means.html

Que, D. (2018, March 28). What we learned from auditing the top 20 ERC20 token contracts . Retrieved from
Medium: https://blog.cryptofin.io/what-we-learned-from-auditing-the-top-20-erc20-token-contracts-
7526ef3b6fb1

Qureshi, H. (2018, Feburary 19). Stablecoins: designing a price-stable cryptocurrency. Retrieved from haseebq:
https://haseebq.com/stablecoins-designing-a-price-stable-cryptocurrency/

Samani, K. (2018, Feburary 13). New Models for Utility Tokens. Retrieved from Multicoin Capital:
https://multicoin.capital/2018/02/13/new-models-utility-tokens/

Schor, L. (2018, March 15). Stablecoins Explained . Retrieved from Medium :

https://medium.com/@argongroup/stablecoins-explained-206466da5e61

Sedgwick, K. (2018, Feburary 23). 46% of Last Year’s ICOs Have Failed Already . Retrieved from
https://news.bitcoin.com/46-last-years-icos-failed-already/

Shaanan Cohney, D. A. (2018). Coin-operated Capitalism.

Singh, N. (2018, January 22). What Is ERC20? | Everything You Need To Know About ERC20? . Retrieved from
Cryptoniam: https://www.cryptoniam.com/what-is-erc20/

Smith+Crown. (2018). Cryptoasset Valuation. Retrieved from

https://www.smithandcrown.com/list/cryptoasset-valuation/

Srinivasan, B. S. (2017, July 28). Quantifying Decentralization . Retrieved from

https://news.earn.com/quantifying-decentralization-e39db233c28e

Wall Street Journal. (2018, August 5). Some Traders Are Talking Up Cryptocurrencies, Then Dumping Them,
Costing Others Millions. Retrieved from Wall Street Journal:
https://www.wsj.com/graphics/cryptocurrency-schemes-generate-big-coin/

Wilson, F. (2017, 06). ICOs and VCs . Retrieved from AVC: https://avc.com/2017/06/icos-and-vcs/

Xing-Zhou Zhang, J.-J. L.-W. (2015). Tencent and Facebook Data Validate Metcalfe’s Law . Journal of Computer
Science and Technology , 246–251.