Since the advent of bitcoin, there has been a lot of speculation that DLT (distributed ledger technology) would ultimately disintermediate the entire financial industry, including in trade and securities settlement.
DTCC (the Depository Trust & Clearing Corporation) has been studying the technology, and starting to apply it to upgrade some of its infrastructure, such as in an initiative to re-platform its Trade Information Warehouse – which serves as a golden record for bilateral credit derivatives and performs lifecycle events, payment calculations and settlement.
Regulation Asia sat down with Rob Palatnick, Managing Director and Chief Technology Architect at DTCC to discuss his views on how far DLT will go in disintermediating the financial industry.
Regulation Asia: Good morning Rob. I understand you are involved in real-world applications of DLT, a topic that continues to dominate fintech discussions. To start, can you tell me a bit about your background?
Rob Palatnick: I am responsible for the global IT strategy, architecture, standards and engineering design of the applications and systems that support DTCC’s products. I also sponsor DTCC’s Office of Fintech Strategy, which defines how fintech is introduced within the firm. In support of these responsibilities, I have been very active around our effort to re-platform our credit derivatives Trade Information Warehouse onto DLT and Cloud.
RA: What are your general thoughts on DLT and its application to the financial industry?
RP: I am hopeful that in 20 of 30 years from now, we’ll be able to look back on this period and see that we are in the middle of a large transition of the entire financial industry to new version of the industrial age – with information, decentralisation, distribution, machine learning and automation, etc.
There was this myth a few years back – maybe not widely embraced but certainly evangelised – that the whole world was going to move to some form of a decentralised model, without fiat currency, without a need for financial institutions, and the entire financial industry would be disintermediated.
I think people now recognise that it’s very complex, that regulations are there to protect investors and they’re purposeful. Starting up brand-new technologies and asset classes don’t always come with the same level of protections. I think we’re proceeding in an appropriate path towards what the new models should be.
RA: What aspects of DLT do you see as being the most useful?
RP: A few years ago, there were a lot of anecdotes and ideas about what DLT is. Among the many attributes that we have seen is that it does answer a really complicated question about data distribution, which has been a hard problem to solve, even from a one-company resiliency perspective, such as copying and backing up data and replicating multiple online copies – it’s complicated!
DLT also addresses reconciliation challenges, where currently each firm maintains its own books and we spend a lot of effort and energy passing information back and forth, and making sure our data rules, business rules, taxonomies and interpretations are all the same – and then going back and forth to make sure we still agree on a version of the truth.
In addition, DLT has an element of a built-in workflow – its’s a multi-party workflow platform that is distributed among multiple parties in a community. There is a lot you could build on top of such a model.
We have observed the way DLT has manifested itself and we think we have developed a real and practical view. We see the technology has continued to mature, with vendors, industry and open source communities moving forward at an impressive pace. The technology is becoming mature enough to be operational for enterprise environments.
All of this is very positive in DLT’s evolutionary path.
RA: Given that you deal with trade infrastructure as part of your main role, and DTCC has been fairly forward-looking in exploring various new technologies, what are your thoughts on the use of DLT to enhance trading processes?
RP: For trade, payment-versus-value first means that every asset and every payment mechanism has to be on the same exact ledger if a DLT model is going to work, and that’s probably not going to happen. The whole world is not going to move to a single ledger.
The way it works currently is you can perform trade order and execution (including order routing) on over 50 different marketplaces in milliseconds in the United States. This means you get finality within milliseconds.
What we’ve seen in many instances of DLT is that there are a lot of questions about finality. In some models you could execute a transaction, and then you may have to wait an hour, or for some series of transaction blocks before your transaction is final. You don’t have immediacy.
There was also this notion of DLT being ‘frictionless’, where you’d get rid of intermediaries. Instead, we see DLT systems as full of friction; the technology has added a whole new world of intermediaries.
RA: Some have argued – and there are currently a few pilot projects ongoing that are exploring this – that DLT would be useful in trade settlement. Can you see any benefits in this area from an efficiency or cost standpoint?
RP: The most optimal near-term opportunities are in asset classes and transactions that are long-lived, that are multi-party, and that are today very manual or not well automated.
When you look at the equity markets, as an example, it has already been very well automated. The settlement cycle is not immediate, but the reality is people buy what they can’t pay for, and they sell what they don’t own, so liquidity management and financing are incorporated into today’s post trade process.
An important aspect to understand is that most equity trades are done by some type of financial manager. Even for individual investors, they have retirement accounts, pension funds, annuities and investments in other accounts that buy and sell large blocks of equities.
Most trades are executed this way, which means the reality of block trades is that you need to break them down and allocate portions to different funds – this is not going to go away, as it’s the majority of the industry. The retail segment is not.
So, what you have is this large, wholesale, institutional set of processes that is not easily accommodated by this notion of end users having wallets, beneficial owners, or companies communicating directly with individual shareholders.
In the equity market, there are processes that have been highly automated over decades; and replacing these is going to require a lot of thought because they are already very well wired and quite efficient.
From a cost standpoint, for clearing and settling well over a hundred million trades a day, DTCC’s added cost to any particular trade is under a penny. So, the economic benefits that DLT proponents promised are just not there for equities at this point.
RA: There has also been talk of using DLT to shorten the equities settlement cycle. What is your view on this? Is there a benefit to moving to T+1 or T+0 settlement?
RP: I think it’s important to state that, for decades, every day, DTCC has handled tens of thousands of transactions that are submitted to us for same-day settlement. T+0 already occurs today, in today’s markets, through DTCC’s existing systems.
I think the T+2 regulation is sometimes misunderstood. T+2 is essentially a guideline on what market practice should be, i.e. you should not take any longer to settle, but you can settle in a shorter time frame.
Settlement rules create market practice so that participants know they have two days to settle, and this allows them to deal with things like allocations, communications, payment, funding, accessing liquidity, retrieving the asset you’re selling, and so on.
Are there occasions where T+0 is appropriate? Yes, and it happens every day, but it doesn’t need DLT to occur.
Instant settlement makes a lot of sense in the payments world, but when it’s in relation to a financial contract or the exchange of an asset, but not every single transaction wants T+0 settlement.
Generally speaking, the industry wanted to be T+2. There may be parts of the industry that are interested in moving market practice to a shorter or longer settlement cycle, and there are other parts of the industry that want optionality.
But there are a lot of implications to settlement timing that have to be built into a market – market conventions, market practice, market structure, market rules, legal considerations, regulations, financing, pre-funding – all of these are issues to consider.
But simply put, the technology is not the problem – our technology already does T+0 settlement every single day.
RA: Thank you Rob. We started this conversation talking the aspects of DLT you see as most useful – you mentioned data distribution, reconciliation, as a multi-party workflow platform, in enterprise environments. You also said the most optimal near-term opportunities are in asset classes and transactions that are long-lived or where processes are not well automated.
With this in mind, could you pick out certain areas in financial services where you think DLT could make the biggest impacts?
RP: I would say mortgages – processes such as managing land titles, selling property, creating a mortgage, securitising mortgages, secondary trading in mortgages – these are largely manual today, and very paper-heavy, maintained in multiple places and in multiple books.
Insurance is another area – also very paper-heavy, where large documents need to be copied and transmitted among multiple parties.
But a real sweet spot is currently in global trade, trade financing and supply chain management, where there are a lot of manual processes, multiple parties, large documents, a relatively long life to a particular transaction.
DLT could really simplify processes and remove a lot of costs that are just associated with managing and reconciling paperwork.