The Blockchain Fallacy

Financial institutions are in dire need of interbank payments and settlement systems that are less expensive, quicker, and less error prone than the traditional systems currently in use. The blockchain promises to make reconciliation of individual ledgers obsolete, client transactions can be processed and settled within minutes or even seconds, and counterparty risks can be eliminated entirely. This is banks’ heaven.

It is a common misconception that blockchain technology inherently is about anonymity. The opposite is the case. Once a digital identity has been mapped to a real-life identity, all their previous and future transactions can be traced back reliably, leading to full transparency. This is regulators’ heaven.

It is not surprising that the financial industry has caught fire and is currently exploring all possible ways to leverage blockchain technology in order to provide a better client experience and deliver better services at a lower cost.

Estimates say that over the past couple of years approx. $1b has been invested in blockchain related initiatives and start-ups, and projections say that it will be another $1b over the coming two years. [1] All major players in the financial industry have set up innovation labs or are participating in blockchain related industry consortiums such as R3 CEV. [2]

At the same time, one of the key figures in the blockchain scene, Mike Hearn, has just recently turned his back on Bitcoin, the very birthplace of the blockchain concept, publicly declaring Bitcoin a failed experiment. [3]

So, if Bitcoin has failed – has the blockchain as such failed, too?

Certainly, some of the reasons for Bitcoin’s alleged failure can be fixed, and probably will be fixed, in alternative blockchain implementations: Scalability issues can be overcome by designing system parameters differently, such as the maximum number of transactions that can be processed in a given time period or the maximum block size, for instance.

A few characteristics of the technology that contributed to its failure, however, sit right at its very core.

Can these be fixed, too?

Let’s take a closer look.

What’s so special about the blockchain?

Bitcoin, a so called crypto-currency, has been the first and most prominent example of a blockchain implementation and is considered the birthplace of the blockchain concept. [4] Since the Bitcoin’s invention in 2008, numerous other examples of blockchain based systems have become available, and many more are in the process of being built.

Conceptually, a blockchain is a distributed, shared, trusted ledger that everyone can inspect, which no single participant or authority can control, which is kept up-to-date by all participants in a blockchain system collectively, and which is protected against manipulation.

In such a blockchain ledger system, transactions are put into blocks which are then cryptographically ‘sealed’ – upon which the transactions become valid. Each block contains a reference to the previous block, forming a block chain. Hence its name.
To visualize the concept, think of transactions as individual entries in a ledger book, think of a block as one page in the ledger book, and think of the blockchain as the ledger book as a whole.

Linking items into chains with pointers to each other is nothing new per se. Linked lists have been around since the invention of computer memory address pointers 52 years ago.

What is revolutionary, though, is the blockchain’s inherent robustness against illegitimate manipulation in an environment that is controlled by no single authority but is maintained by an open global network of computer systems that are connected through the internet.

But what is it that makes the blockchain so secure against tampering in a hostile environment?

The proof-of-work

A system that is based on blockchain technology needs to avoid a number of problems, such as: double spending, blockchain forking (prevent the blockchain from splitting into branches), the creation of fraudulent transactions, and fraudulent manipulation of previous transactions.

A concept called proof-of-work addresses these challenges and is used to legitimize a block, and with that the transactions it contains.

The following examples are based on the Bitcoin specific blockchain implementation. In principle, they are true also for other blockchain systems.

Special types of computer systems that are connected to the blockchain system, so called miners, carry the burden of validating transactions, bundling them into blocks, and making these blocks legitimate by creating a valid proof-of-work.

The proof-of-work is the evidence that a difficult mathematical problem has been solved – a problem which can be solved through trial-and-error only, just like throwing dice. All miners in the network are ‘throwing dice’ until one of them finds a valid proof-of-work.

This requires a significant amount of computing power, a resource which at any given point in time is available in limited quantities only.

The proof-of-work therefore is the evidence that a significant amount of computing power has been put into the creation of each block. If this evidence is missing then the block is void and will be ignored by the system.

How does this make the blockchain secure against manipulation?

The more computing power becomes available in the network, the more computing power the system requires to be put into each block’s proof-of-work.

Specifically, the Bitcoin blockchain will automatically adjust the difficulty requirement of the proof-of-work so that the entire network of miners is able to find a valid proof-of-work only every ten minutes in average. This means that, by design, the entire computing power available to the Bitcoin network will be burned in order to create one block every ten minutes.

More work means more trust. If there are ties and two blocks are produced simultaneously, the system will stick with the computationally most 'expensive' one. The system requires everyone to trust the ‘longest’ chain, the one which was the hardest to create.

A fraudster would need to create a blockchain that is ‘longer’ than the legitimate one – all by himself. He would need to outrun the whole system’s computing power combined, over a prolonged period of time. This is an uphill battle one can only lose.

That’s the theory.

Concentration of power

Miners are providing significant amounts of resources and they need to be given an incentive to do so. In the Bitcoin system, for instance, they earn a number of Bitcoins for each valid block they produce.

All miners are competing against each other – as soon as one miner has produced a valid proof-of-work he wins this round and all other miners have lost. The race for the next block begins. No miner can know upfront whether or not he will win the current race. This is what makes the system robust against fraudulent activity.

This mechanism of competition is intentional and part of the blockchain concept. In the example of the Bitcoin blockchain this has led to failure, though:

Bitcoin miners have started to add massive amounts of computing power to the network in order to outrun the others and to increase their chances to earn more Bitcoins. The system has, by design, responded to this by making the proof-of-work more and more difficult. As a result, massive amounts of computing power are now required for new blocks to be created, turning vast amounts of electrical energy into hot air. A major waste of energy. [5]

Even worse, though, there seem to be clear winners of this ‘arms race’. The blockchain’s original idea – that it can be controlled and maintained by no single authority but only in a collective effort – has been undermined. According to Hearn, by autumn last year 95% of Bitcoin’s collective mining power has been represented by 10 participants, with just two Chinese individuals accounting for more than 50% of it. [6]

While the Bitcoin blockchain is protected against manipulation, it is not protected against concentration of computing power resulting from such an ‘arms race’.

This dilemma cannot be easily overcome.

On one hand, in order for the blockchain to be robust against illegitimate manipulation, two things are required: 1) There needs to be a sufficiently large number of independent miners in the network. 2) It needs to be entirely unpredictable which miner will find a valid proof-of-work to legitimate a block. There must be no concentration of mining power.

On the other hand, the proof-of-work needs to be difficult and miners are dedicating significant resources into solving the proof-of-work problem – and they need to be given an incentive to do so. Miners never know whether they will succeed in finding a valid proof-of-work and earn the reward before someone else does. Hence, they will try to outrun the others by dedicating more resources in order to have a better chance of winning the incentive. This will lead to an ‘arms race’ and, subsequently, to a waste of resources and to concentration of power – which is precisely what mustn’t happen.

This is a fundamental problem of the blockchain concept and is not limited to the Bitcoin implementation.

The technology is at a breaking point. A solution needs to be found to harden the blockchain while preventing the waste of resources and the concentration of mining power.

So, where do we stand?

The blockchain fallacy

The proof-of-work is a key element of the blockchain concept. The Bitcoin specific implementation of the proof-of-work seems to have led into a dead end. What if we remove the proof-of-work concept from the blockchain altogether? What remains? A linked list, albeit a potentially very smart one?

And will this be sufficient to revolutionize the financial industry?

Let’s remember what the original motivation behind Bitcoin was: To create a self-sustained shadow financial system that operates outside the boundaries of the regulated financial industry, for which no one will ever assume ultimate accountability. Admirable as it may be for its intellectual brilliance, ultimately it’s been a subversive act of rebellion against fractional-reserve banking, today’s predominant form of regulated banking. [7]

If we have learned one thing from recent history, it is this: Leaving a financial system open and unregulated will inevitably lead to excess and, ultimately, to its failure.

If this is supposed to become the future backbone of the financial industry then we don’t want this to be a self-sustained and unregulated system for which no one will be accountable.

In such a scenario, the proof-of-work mechanism might actually become quite irrelevant. It could for instance be simply replaced by an officially appointed, regulated financial authority, such as for instance a certified industry-wide service provider, which will simply digitally sign each transaction once it has been verified for its legitimacy.

If we take away the one core concept of the blockchain that separates it from mere linked lists, though, can we still call it a blockchain? The answer is no.

As a matter of fact, eliminating the proof-of-work makes the whole blockchain construct collapse like a house of cards. Grouping transactions into blocks is only meaningful if these blocks are used as processing units for the proof-of-work. And linking these blocks into a chain is only required in order to ensure that no single individual can outrun the system and create fraudulent blocks.

If, however, there is a central authority that validates and signs transactions then there is no more need to put transactions into blocks and linking these blocks into chains in the first place.

In conclusion, does it turn out that the blockchain hype in the regulated financial industry might not even be about the blockchain as such?

Doing the math

On one hand, there is massive demand for what the blockchain promises. The financial industry needs to renovate its underlying infrastructure in order to serve its clients better, quicker, and at lower cost and risk. The blockchain makes reconciliation of banks' records, kept separately from each other, obsolete. Client transaction settlement will come down from days to minutes or seconds, allowing banks to provide a better client experience. According to The Economist, Santander reckons that blockchain technology could save banks up to $20 billion a year by 2022. [8]

On the other hand, a massive research & development machinery has been set in motion. There is a push from FinTechs that are actually attacking traditional ways of banking. Hundreds if not thousands of the world’s brightest minds are being given a platform and the financial support to develop new ideas, even beyond pure FinTech applications. Some fundamental issues will need to be solved, though.

In conclusion, there is demand and there will be supply. Inevitably, some great new ideas will come out of this and there will be change.

There are two realistic scenarios – which might well happen in parallel:

1. Disruption of traditional banking

We might see innovative and potentially disruptive solutions based on blockchain technology in certain very specific areas, such as for example:

• Quick and reliable international payments for unbanked individuals at low cost (by the end of 2014 there have been approx. 2b unbanked individuals worldwide [9]).
• OTC derivatives are automatically listed in a central trade repository.
• Smart bonds that automate coupon payments and reimbursement.

Potential use cases go way beyond the financial industry and are limited by imagination only:

• Smart contracts that enforce themselves will for instance let you start the motor of your car only if you’ve paid the insurance.
• Ownership of physical goods such as diamonds or pieces of art can be tracked globally.

NGOs such as Identity 2020 have an ambition to give those who are at risk of violence and exploitation and who lack basic social needs a self-sovereign identity and digital inclusion. [10] Is it a coincidence that a world-renowned blockchain engineer has been appointed as Identity 2020’s Technical Advisor?

As a prerequisite for all this, a solution is required to address the problems related to the proof-of-work concept as previously discussed.

Articles on this topic such as those published by Ethereum’s Vitalik Buterin [11] and many others shed light on where the journey might be going:

• Substitutes for the proof-of-work’s underlying resource, for instance using computer memory instead of computing power;
• Making the proof-of-work useful by doing some actual work rather than performing mere number crunching just for the sake of solving a difficult mathematical, but otherwise useless problem;
• Proof-of-stake asks users to prove ownership of a certain amount of what the blockchain provides (e.g. their "stake" in a digital currency);
• Other ideas are based on the notion of a network of trust.

Each of these ideas have their own specific challenges. There are no conclusive answers yet.

The solution may lie in either finding an ingenious new way for a proof-of-work alternative which is energy efficient and truly constrains excess, or have a community with a real stake in it regulate it.

Or, contrary to the blockchain paradigm, regulate and manage this through some sort of authority. Which leads to the second realistic scenario.

2. Renovation of the regulated financial industry infrastructure

Separate from potentially disruptive developments, we will probably also see a much more mundane shift from traditional message based interbank payments and settlement systems with slow and cumbersome reconciliation processes to a globally shared ledger system, presumably in the form of some new type of globally shared data base management system, potentially leveraging cloud technology.

If such a global ledger system is controlled by some official authority and operated by a regulated and certified service provider – rather than a network of independent and inherently untrustworthy computer systems in the internet – then there is no need for this to be built on the basis of blockchain technology.

It will be this (new) service provider’s duty to validate and sign transactions. Key features of the blockchain can be easily preserved, such as transaction traceability, elimination of counterparty risk, and instantaneous transaction settlement.

As a prerequisite, there needs to be an agreement on an industry-wide, commonly accepted set of standards for data exchange and validation, and financial institutions will need to define how to represent their data models in such a globally shared ledger system.

Maybe such an infrastructure will not be based on blockchain technology as such. But the blockchain certainly will have contributed to its development and some of its features, ideas, and concepts will be adopted.

APPENDIX: References

[1] insidebitcoins.com - Bitcoin Venture Capital Funding on Pace for $1 Billion in 2015 / fx-mm - New report: top global financial institutions will invest more than $1bn in blockchain-related projects in the next 1-2 years / infosecurity magazine - Global Financial Leaders to Invest More Than $1bn in Blockchain Projects in Next 1-2 Years

[2] R3 CEV

[3] The resolution of the Bitcoin experiment

[4] The invention of the blockchain

[5] Bitcoins are a waste of energy - literally

[6] High-profile developer declares Bitcoin doomed, leaves currency project

[7] Genesis block: comment on the instability caused by fractional-reserve banking

[8] The Economist / The trust machine / Oct 31st 2015

[9] Global Findex

[10] Identity 2020 NGO

[11] Long-Range Attacks: The Serious Problem With Adaptive Proof of Work