So the big question(s)

Will blockchain transform the financial services industry?

Will blockchain be to value as web was to information?
Theres all sorts of solutions proposed
blockchain for core banking
blockchain for identity
blockchain for
not all of which make sense

Will blockchain transform FSI?

Peter Evans-Greenwood, 10 November 2016

Centre for the Edge

Australia

really this comes down to

will Dapps take off?
resulting in a new generation of decentralised infrastructure and solutions
which is the question that Ive tackled instead
the answer, and if you know me this isnt giving anything away, is no
theres too many headwinds against Dapp deployment
however, we forget that there was no new technology in Bitcoin
its emergence was due to an environmental change
this suggests that Bitcoin & blockchain is a sign that something bigger is afoot
rather than replace our platforms with distributed platforms
we might get rid of them entirely

So lets start where most people end: at some stage in their talk, thinkfluences put up a picture like
this.

Money
(payments)

Identity
(accounts)

Reputation
(ratings)

The chain of reasoning is:

its a platform & blockchain is essentially a shared database
If we build the same things on top of blockchain then we have a distributed version of Uber, AirBnb,
etc invent new types of business Dapps as theyre often called
payments -> money, accounts -> identity, ratings -> reputation, catalogue -> marketplace
magic happens

Marketplace
(catalogue)

Blockchain
(database)

At which point we launch into talk of futures where Dapps solve world problems by eliminating rent
seeking corporations from the equation
replace the intermediaries well replace centralised intermediaries with decentralised ones
new decentralised governance models
musicians will have fairer revenue, land titles in the 3rd world are secured, etc.
and head off to the family holiday compound to discuss how to run our new society
This is hugely attractive to many people
It takes something complex bitcoin, side chains &c and transforms it into a web app story.
Solidity is a lot more palatable than Bitcoin scripting and sidechains
The claimed ability to cut rent seeking corporations out of the picture to create a fairer society is
appealing
It all sounds too good to be true: Thats because it is too good to be true
Blockchain is not the next platform, but it is a sign that something bigger is afoot

Pipe
Firm

Platform
Market

The rational behind Dapps is that theyre the logical extension of the old tension between firms and
markets
Coase told us that the choice between firm and market is driven by the cost of discovery, coordination
& contacting

Dapp
Distributed Market

Firms the old value chain stuff goes from left to right
Producers find materials, create value, and deliver to customers
Miserly with communication/coordination as everything is predefined and mandated from above

1924 Model T assembly line

Platform businesses -> the hot thing at the moment

Producers and consumers gather at a venue a market
More choice, but requires more communication

Edward Lloyd's Coffee house,

Londres, par William Holland (1789)

Uni Cal saved ~ 50$ million moving their procurement process to a market
3

The interesting thing about Bitcoin was the lack of new technology
no new maths
Its an environmental change rather than technological innovation
Dapps can be considered an extension of Coases market-firm
Increases in bandwidth / performance-price enable the creation of distributed ledgers -> Bitcoin
Similarly this change enables the creation of Dapps
Bitcoin is sometimes considered the first Dapp.

Trade finance appears to be the canonical example

A complex and fraud-prone process
Many electronic (EDI) and manual (faxes, letters) point-to-point messages
Requires a number of intermediaries, such as correspondent banks
Enables bad actors to play one participant off against another
The cheque is in the mail: release the goods
and so on
Spans juridisctions

Carrier
Freight
Forwarder

Exporter

Importer

Agencies

Bank

Customs
Insurer
Electronic
Manual

One solution is to make all messages public

Create a central location a central ledger
Pass all messages through this ledger
All participants can see all messages (with some privacy limits)

Carrier
Freight
Forwarder

Exporter

Importer

This reduces cost

less hunting for information, less wasted effort
streamline processes
and improves risk management
fewer opportunities for fraud
improved risk management

Agencies

Bank

Customs
Insurer
Electronic
Manual

We know that the benefits are not hypothetical

weve used similar patterns in the past
the first copyright registers in the London printers guild
CBA et al have a real-world PoC in process, not a simulation
We know that the pattern has broad applicability
So why arent we seeing a lot of this given that the of Dapps are already a few years old
the problem is that there are a number of headwinds

first, theres the technical

Scale is challenging

These solutions dont scale well

its the need for global consistency we must compare every pair of transactions, and theres no
escaping this
quote from Ethereum implicit vs explicit support for approving transactions doesnt make any
sense youve gotta compare them all
all the options are being explored
parameter tuning: block size, payload size (SegWit)
centralisation, reduce the overhead: RAFT etc., Chain is centralised
fragmentation/sharding:
always sublinear performance improvement (growing % of capacity goes to overhead)
while cost increases faster than linear (typically exponential) (overhead growing
proportion of cost)
and will max out diminishing returns asymptotes
1 + 1 does not = 2, it depends on cross-talk
off loading: move work (transaction processing) out of the consensus process
lightening networks: interesting but: complicated, and impact overestimated, and only
applies to specific use cases (bar tab)
assuming that some smart person is going to come along and solve this problem is hubris
we also forget that networks dont follow Moores Law
Consequentially we expect blockchain to be limited to low volume, high value applications

Scale is challenging
The economics are unfavourable

Economic problems: expensive to run & use

Betamax vs VHS.
the technology that wins doesnt need to be the best
it needs to be good enough, and cheap enough
We might think blockchain is morally better, but itll be outcompeted by centralised
solutions that are easier and cheaper to build
= ~10$ per transaction with bitcoin
Inherent in the technology as were running many redundant servers
pretty easy to get this cost down by a factor of 10 (parameter tuning etc to increase
throughput)
but thats probably as good as it gets
We need enough to ensure security, and they need to be paid for, and Proof of Stake etc
dont avoid the problem
miners will always invest up to the margin
Currently Bitcoin is paying with this via currency dilution
One day it will hit the bottom line
use of a seperate currency incurs a user cost
Back to the time of free banking
currency risk
exchange costs
users dont want to use your alt coin as a store of value
Its not a feature, its a problem

Then theres the structural

Tangible only has value when it touches the real world -> its not distributed as you think it is
something asserted on the ledger needs an external actor to make it real
these gateways oracles need to be developed and maintained
Conceptually these gateways are responsible for enforcement, not your smart contract
Ditto, we need to interact with the ledger through some sort of UI gateway

Scale is challenging
The economics are unfavourable
Value is created in the real world

The gateways must live somewhere

While a Dapp might control a car, it cant stop it from being impounded Uber tried to escape GST,
but couldnt
digital land titles are just land titles, not something new, as they need to be backed by the
governments monopoly on violence
Gateways must have an address and can therefore be regulated / attacked
Even bitcoin, which was supposed to be anonymous, collapses when it touches the real world:
AML/CTF on-ramps and off-ramps
Blockchains dont enable us to avoid any of these constraints
This is why the majority of existing Dapps focus on redistribuing the native cryptocurrency, rather than
creating value

Governance:
Decentralised governance is slow to change / fix problems
US president can declare war, but needs to go to congress quickly
This is a problem we can see with bitcoin: the block size debate
Its not clear that it enables new governance models
the problem with scaling anarchy is not technology, but people willing to commit time to the
process
this is why direct government is not a thing.
it might enable more complicated governance models (as theyre implemented in software)
this is no necessarily a good thing
(DAO redux)

Scale is challenging
The economics are unfavourable
Value is created in the real world
Decentralised governance is slow

We also forget that centralisation brings other benefits

encouraging people to join the platform via discounts / subsidies
advocate on behalf of the participants
Scale is challenging

Despite the idea of dapps/blockchain being a few years old, were yet to see anything of significance
beyond Bitcoin
Lots of PoCs while the few deployed solutions involve some form of value redistribution using the onchain currency
gambling, binary options, some form of crowd funding
aside from the DAO, but its better not to mention that one as no one involved came out looking
good
theres also a few low scale things Myencyniam, Mediachain, etc that have yet to amount to
anything

The economics are unfavourable

Value is created in the real world
Decentralised governance is slow
Centralisation has other advantages

10

Even Bitcoin, despite finding a stable role for itself, has yet to see widespread adoption
we forget that the killer app for currencies is tax. more on this later
What this means is that theres no silver bullet
Claims that blockchain represents a new paradigm are silly
digital property for example is still just property and relies on the governments monopoly on
violence
But this idea of cutting out intermediaries is good
Do we have other options?

We forget that before the firm and the market we had the village
the baker brought bread around in the morning
the cobbler lived around the corner
we went to the pub at night
the world was distributed, to use the term de jour, as we didnt rely on intermediaries

Village fair, Richard Brakenburgh, 1650-1700, oil on canvas - Villa Vauban - Luxembourg City

11

We didnt need money as we didnt need (as much) formal governance

relationships facilitated via debt and shared value, informally counted
double coincidence of wants & barter is fiction
money only used when we didnt know or didnt trust someone, with damages the most common
use case
it was colonialism and industrialisation (development of the mass market) that pulled the vast majority
of people into the monetary system
Indian pound tax as the killer app
retail & growth of the mass market searching for products and transacting with people we dont
know and will never see again
Similarly with our domestic affairs
a couple were considered married if someone could attest to witnessing them committing to
marriage
Romeo & Juliette
marriage registers were created to
remove inconsistencies (we cant find the witness)
enable the church (1 of 2 major power structures of the time) to control who was married ->
inheritance
Bigamy
Marrying relatives

So we need to consider these on a continuum

Community

Market

Communities are great, but they didnt scale

Markets enable us to create communities around topics, but we need to go looking for them
Firms are efficient, but you can have any colour you want as long as its black

Firm

The amount of administration and formalisation required increases as we move right

The volume of communication/coordination required decreases as we move right <- Coase
Value becomes more narrowly defined as we move right

12

Most of industrial history has been a shift from local markets to global firms
More recently weve seen the shift from global firms to (global) markets, right-to-left
as communication technology has caught up
Its like weve worked our way right, bounced off the end, and now were heading back the other way
The recent emergence of blockchain supports this, as blockchain could only exist when
communication costs are low
Were also seeing consumer shifts
Value is now defined by consumption (consumer) than production (firm/market), consequentally
value is becoming multidimensional and relative
Consumers want direct relationships narratives rather than brands
Etc.
suggest that wed like to return to the village
Rather than build firms or platforms that we try and form communities around
Creating a (virtual) place at great expense, global consistency, blockchain
An intermediary
or find communities, and facilitate the transactions within them
Enabling local consistency
We might even call these Digital Communities or the Digital Village

So lets return to trade finance

A ledger to reduce cost and risk by ensuring global consistency

Carrier
Freight
Forwarder

Exporter

Importer

Agencies

Bank

Customs
Insurer
Electronic
Manual

13

With participants spread around the globe

14

 and part of a much larger community

What were actually doing is putting the entire community on the same page
Were ensuring that the entire population is consistent
when we only need
the participants to the transaction
enough bystanders to ensure nonrepudiation
(think marriage before the church)
We only need local consistency, not global
and global consistency is expensive

15

Global consistency is only required

to ensure transactions are unique
Easy to pick: do we need to compare this transaction to every other
Clearance is a good (the only?) example of needing global consistency
Identity as a counter example
To confirm identity I dont need to compare your identity data to that of others
Similarly:
for my trade to progress, I dont need to compare it to others, only to itself
To determine provenence I only need to see the chain of custody
etc

So lets think about why we use markets and firms

Three basic things:
discovery problem: how do I find what I want / find customers?
ontology problem: how do I communicate (negotiate) with the merchant / customer?
counter-party problem: how can I be sure that I wont be ripped off?
Our current solutions relies on a concept of place, a venue: firm, market, Dapp
Indeed, Dapps might be better thought of as location randomised than distributed
We know where the code was executed / state persisted, but the location was selected randomly

16

Its this idea of place that enables a market/platform owner to extract a rent
Its this rent that were reacting against
Dapps extract a rent, the only difference being that the rent is shared and the redundancy makes the
rent larger
Its also the creation of (virtual) place in Dapps that creates the problems with scalability
Could we build an alternative, a locally consistent solution?
The key is to sort ensure consistency in communication
nonrepudiation: no takesy backsies
non-contradiction: cant say different things to different people

Start with the network

Noting that there is no new technology in bitcoin
theres been an environmental change
bandwidth is plentiful, ubiquitous, and cheap

Network
17

On top of this we create a consistent communication environment

Nonrepudiation
Consistency: say the same things to different people
Blockchain provides globally consistent
Enforces nonrepudiation: durable record
Enforces consistency: everyone sees the same thing
What are our options for local consistency?
Globally
Consistent

Locally
Consistent

Network
18

Theres two
Subjective

Subjective:
base on the messages I receive, and I determine if someone
repudiates
is inconsistent
internal, subjective, data structure

Objective

Objective:
base on the messages others share, can I determine if someone
repudiates
is inconsistent
external, objective, data structure
19

The objective solution is fairly straight forward

Subjective

Messages are distributed via a flooding algorithm

All new messages must sign n previous messages
Ive seen this
A tangle

Objective
Sign each others messages
Reciprocal attestation

Rapidly creates a shared, public data set capturing whos seen what transaction
Anyone who can access the messages can (re)create a graph capturing
who said what, and in what order
Has a problem with liveliness like torrents how do I ensure that the transactions persist
somewhere
20

Subjective is slightly more complex, but more usable

Subjective

Objective

Build our own internal

representation

Sign each others messages

Messages are distribute by gossip point-to-point, pair-wise information exchanges

Pairs exchange
The (signed) record
Who created it
Whom they received the record from
When they first received it

Reciprocal attestation

Who said what to whom, when

Actors, over time, create an internal directed graph capturing

the order the records were created
the provenance of records: how the percolated through the crowd
21

In both cases objective and subjective we could consider consensus to be based on individual
transactions, rather than on the global transaction set
locally consistent, rather than globally consistent
Our report on blockchain in Q1 included a slot for this in the taxonomy
conceptually it might be better to consider both of these approaches to avoid the need for
consensus
In both cases it rapidly becomes obvious if any actor attempts to
repudiate
say different things to different actors

Interestingly, were seeing implementation of both of these approaches in the wild

Subjective

Objective

Build our own internal

representation

Sign each others messages

Reciprocal attestation

Who said what to whom, when

22

So we can add these to the solution

Globally
Consistent

Locally
Consistent

Subjective
Consistency

Objective
Consistency

Network
23

Domain
Concepts &
Lifecycle

Globally
Consistent

Locally
Consistent

Negotiation,
Contracting
etc.

Subjective
Consistency

Next we define the messages we need

CRUD domain concepts and life-cycle
processes such as negotiation, bargaining, resource allocation, coalition formation, etc
we can even do discovery this way
though this doesnt prevent discovery via other means

Discovery,
Yellow
Pages etc.

Objective
Consistency

Network
24

Finally the actors involve sit on top of the stack

Actors (the real world)
Manufacturers, Shippers, QA, Financers,

Domain
Concepts &
Lifecycle

Globally
Consistent

Locally
Consistent

Negotiation,
Contracting
etc.

Subjective
Consistency

Heterogeneous rather than homogenous

Discovery,
Yellow
Pages etc.

Objective
Consistency

Network
25

Returning to trade finance

Freight
Forwarder

Bank

Carrier
Freight
Forwarder

Using this approach we can create a solution where

Parties deal with each other directly
But with a guarantee of consistent and reliable communication
With enough bystanders to ensure that parties to the agreement cant gang up on each other

Exporter

Importer

We have, in effect, removed the need for the platform entirely

and it was the platform and its global consistency that caused our problems

Agencies

Bank

Customs
Insurer
Exporter
Electronic
Manual

Customs

26

So what does this get us?

Scale is challenging
The economics are unfavourable
Value is created in the real world
Decentralised governance is slow
Centralisation has other advantages

27

Our solution is a lot more scaleable, so scale is not a challenge

Only the parties and a few bystanders need to be involved in the transactions
We can rely on eventual consistency to provide increased security over time
Scale is challenging

These approaches have their own limits but

the limits are a lot more generous

The economics are unfavourable

Value is created in the real world
Decentralised governance is slow
Centralisation has other advantages

28

Weve avoided the economic head winds:

No mining, so no cost of mining, avoid the huge expense of global consistency
and no currency, so no wildcat banking, exchange costs, currency risk etc
Scale is challenging
The economics are unfavourable
Value is created in the real world
Decentralised governance is slow
Centralisation has other advantages

29

It acknowledges that value is created in the real world

The real world participants are part of the system, not hanging on the side
Direct enforcement
if contact says make payment then the payment provider makes the payment
credentials are attested by their creators
And so on
Avoid the complexity/overhead of asserting facts on the ledger to be picked up by an oracle
Doesnt do anything about preventing the government seizing a DAOs autonomous car

Scale is challenging
The economics are unfavourable
Value is created in the real world
Decentralised governance is slow

# otherwise

Centralisation has other advantages

30

Theres nothing we can do to speed up decentralised governance

Best approach is to work through the existing centralised governance bodies
Trade associations
Regulators etc
Ditto the advantages of centralisation
If we want to use subsidies to encourage adoption then well need to work with centralised bodies
that have the money
Or we could work with regulators and trade bodies to mandate adoption
And herein lies the problem with getting these distributed solutions working
The larger and more heterogeneous the community, the more value the solution can create
The larger and more heterogeneous the community, the more stakeholders we need to convince,
the harder it is to get the solution up
Its unlikely that there will be any quick wins, just slow and incremental progress

Community

Market

At which point Ill return to this slide and ask the question

Will blockchain transform the financial services industry?

Will blockchain be to value as web was to information?

Firm

The answer is, well, not really, but maybe sort of

Blockchain as a technology is suitable for a narrow set of use cases, ones requiring
Globally Consistent
Multiple writers (decentralised) la IBM: otherwise use a central ledger
Slow & expensive is ok
Unsurprisingly, Bitcoin is a good fit

31

Many of the solutions proposed dont fit these requirements

Identity: no need for global consistency, just sort out key management
Property registers: single writer (the institution creating the property)
Core banking: high performance and complex requirements

Even trade finance: no need for global consistency
Many(?) of these solutions would work well with local consistency
Trade finance is a great example
And smart contracts could just as easily be implemented on a central or locally consistent framework
Enforcement was always vested in the oracles anyway
Its more productive to think of blockchain as a contributor to a larger trend
Moving from firms to markets
And from markets to communities
The thing about the modern technology environment is that its highly entailed
Its not possible / unlikely for a single technology to disrupt / transform the industry
And future will be determined by how we answer the social questions
What does good governance and regulation look like
How do we mobilise our communities?

This publication contains general information only, and none of Deloitte Touche Tohmatsu Limited, its member firms, or their related entities (collectively the
Deloitte Network) is, by means of this publication, rendering professional advice or services.
Before making any decision or taking any action that may affect your finances or your business, you should consult a qualified professional adviser. No entity
in the Deloitte Network shall be responsible for any loss whatsoever sustained by any person who relies on this publication.
About Deloitte
Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited, a UK private company limited by guarantee, and its network of member firms, each of
which is a legally separate and independent entity. Please see www.deloitte.com/au/about for a detailed description of the legal structure of Deloitte Touche
Tohmatsu Limited and its member firms.
Deloitte provides audit, tax, consulting, and financial advisory services to public and private clients spanning multiple industries. With a globally connected
network of member firms in more than 150 countries, Deloitte brings world-class capabilities and high-quality service to clients, delivering the insights they
need to address their most complex business challenges. Deloitte has in the region of 200,000 professionals, all committed to becoming the standard of
excellence.
About Deloitte Australia
In Australia, the member firm is the Australian partnership of Deloitte Touche Tohmatsu. As one of Australias leading professional services firms, Deloitte
Touche Tohmatsu and its affiliates provide audit, tax, consulting, and financial advisory services through approximately 6,000 people across the country.
Focused on the creation of value and growth, and known as an employer of choice for innovative human resources programs, we are dedicated to helping our
clients and our people excel. For more information, please visit Deloittes web site at www.deloitte.com.au.
Liability limited by a scheme approved under Professional Standards Legislation.Member of Deloitte Touche Tohmatsu Limited
2016 Deloitte Touche Tohmatsu.

Peter Evans-Greenwood
Fellow, Centre for the Edge
pevansgreenwood@deloitte.com.au
@pevansgreenwood