Are public blockchains really a viable option for guaranteeing settlement finality for financial instruments?
Tim Swanson, director of market research at R3, claims public blockchains by design cannot definitively guarantee settlement finality. For this reason, he says they are not presently a reliable option for clearing and settling financial instruments.
Swanson aired his views on tabbforum in response to the view among some crypto enthusiasts and investors who see public blockchains as a settlement mechanism for financial instruments.
Swanson claims they ignore the regulatory, legal and commercial requirements and regulations concerning the need for definitive settlement finality.
The European Commission’s directive 98/26/EC was motivated by the need to minimize systemic risk by ensuring any payment considered final according to the system rules was, in fact, final, even in insolvency proceedings.
Minus definitive finality, one participant’s insolvency could undermine transactions considered settled and unleash a host of liquidity and credit problems for other participants in the payment system.
By ensuring definitive settlement, finality creates trust in the system and minimizes risk. Hence, it is one of the most critical concepts in payments, one that applies to all settlement and clearing systems such as bulk SEPA clearing system STEP2 and high-value payment system Target2.
Cyrpto proponents think immutability is unique to public blockchains, Swanson noted. But strong, one-way cryptographic hashing gives immutability to any data it is hashed by. If a person changes one part of a transaction, the hash changes and another person knows it has changed.
Proof-of-work that many blockchains use gives a way to vote on the inclusion and ordering of transactions in a block in an environment where one does not know who is voting. If one knows who votes, there is no need for proof-of-work.
Hence, proof-of-work-based blockchains like bitcoin offer no way to predict their future security level, or settlement, since it is directly proportional to the token’s future value, which is not knowable.
If a native token’s market value rises or falls, the amount of work created by miners competing to receive the seigniorage and contract or expend capital outlays also changes in proportion to the token’s marginal value. This leaves open the possibility that Byzantine actors will create block reorganizations with no legal recourse.
The miners can remove a transaction from the history so that a payment someone thought they made is suddenly unmade.
With public blockchains, miners have complete discretion on ordering and reordering transactions. While mining pools can’t reverse one-way hashes like a public key, they can make it so that in any transaction, despite its value, can be blocked, reordered or censored.
If two conflicting transactions can be included in a blockchain, the payment to one party can be mined and replaced at any time in the future by the payment to another party.
In Ethereum and bitcoin, mining pools have the complete discretion of organizing and reorganizing blocks. As there is a cost to such rewriting of history, there are also tradeoffs in making censorship-resistant systems like bitcoin.
One tradeoff is that entire amounts of value can be reorganized or removed with no recourse since public blockchains were designed around securing pseudonymous consensus. This is a key characteristic that cannot be eliminated without destroying the core utility of a public blockchain, which is censorship resistance.
Hence, as long as bitcoin miners have complete discretion over the process of transaction validation, a risk of a reorganization exists.
If you remove censorship resistance but still utilize proof-of-work, a public blockchain no longer exists. Instead, there is a very costly hash-generating gossip network.
Such a quasi-anarchic system can be useful to some, but it is not a desirable attribute for regulated financial institutions that have invested in removing risks from the settlement process.
Putting aside the regulatory structures surrounding the settlement of financial instruments, participants today realize that it makes sense to have definitive settlement finality.
Considering how the mining process works, miners can and have reorganized history. A public blockchain cannot serve as a definitive guarantee of a final settlement.
Uncertainty is not desired by markets. Hence, removing systemic risks has been a driver for global settlement platforms.
Public blockchains can be alluring because of the way they are marketed, but they are not a viable solution for organizations that want to provide certainty. Nor are they reliable for settling and clearing financial instruments.
Solutions are evolving to address this issue that do not rely on public blockchains for settlement. Private blockchains are designed specifically to give users definitive legal settlement finality. This is needed for regulators and regulated institutions.
For the past 18 months, banks have examined more than 150 pilots and proof-of-concepts and denied nearly all of them. The banks have done so not because they oppose cryptocurrency, but because public blockchains do not meet their needs.
If researchers observe the failure risks connected with public blockchains falls within an accepted risk, some institutions will consider conducting more proof-of-concepts.
Tradeoffs in designing permissioned ledgers and public blockchains do exist. It is self-defeating to create a network that is censorship resistant from legal infrastructure and at the same time compliant with legal settlement requirements. Nonetheless, both networks will coexist, and their communities will learn from each other.
If the objective for fintech startups is to develop a new commercial rail for securing different types of financial instruments, then delivering products that meet the needs of market participants is more important than trying to link everything into a pseudonymous network that doesn’t have the characteristics institutional customers need.
Venzen Khaosan, a market analyst and writer of CCN.com’s daily bitcoin price analysis report, takes issue with some of Swanson’s points.
1) Swanson states as a fact that a single miner (or mining pool) can remove a transaction from history. This is only true in a very specific case: the attack vector of removing a transaction from history requires that the miner (or pool) commands more than 51% of network hashrate. He’s speaking about a 51% attack. The entire point of a proof-of-work blockchain is that it is immutable – the more decentralized (healthy) the network, the truer that becomes.
2) He states as a fact that one party’s payment can be replaced by another’s at some point in the future. This is incorrect. Transaction malleability only allows tx replacement within a very small time window (less than 10 minutes) but does not allow arbitrary tx replacement once a tx is confirmed. Even so, transaction malleability will be “fixed” by SegWit, along with the introduction of additional fraud proofs.