BNY Mellon
Tom Casteleyn of BNY Mellon suggests DLT will not radically transform the business model in the next three to five years, however, it will see small-scale solutions in areas where significant problems exist today
How are DLT technologies changing the custody industry?
It’s interesting because there has been an evolution in how the industry is looking at distributed ledger technologies (DLT). DLT was all the talk of Sibos in Singapore four years ago and everyone thought it was going to completely transform the industry. Now, a few years later, we have all spent a lot of time and energy approving concepts and experimenting with the technology, and over that time we have seen a few strategies come and go. However, we now have a better view of what DLT can and can’t offer.
The first thing to understand is that DLT is just a technology. Initially, we thought it was a technology that was going to completely rip up the business model, but we’re now seeing DLT implementations that actually make sense; often these are a technology solution to an existing business model without having to disrupt it. We see this when we work with clients on concepts; initially, they start off wanting to completely rethink the business model when they are, for instance, putting tokenised bonds on a private blockchain. Once you workshop that through a couple of times, you find that you come back to the existing business model because it is there for a reason, it has been developed over many years as the business evolves. It is what I call a finely-tuned ecosystem of responsibilities and liabilities that are laid off in a chain of people that work together.
So we believe DLT will not radically transform our business model in the next three to five years. However, we will see small-scale DLT solutions in areas where significant problems exist today, and that’s why I think DLT has taken off in trade finance. In securities processing, we’ve experimented with proxy voting and corporate actions. It works on DLT and that’s good, but the benefits are not really there. There are two main reasons why it’s hard to achieve benefits there. Firstly, the problems with standardisation are not solved by the technology—on the contrary, technology makes it even starker that you don’t have standardisation in some of these areas. Secondly, there is the problem of adoption. Even if I’ve built something where 10 percent of my flow goes through, I’ve now built a new system for 10 percent of my flow and I still have my old system for the 90 percent. So instead of making it simpler, easier and more efficient, I’ve actually increased the complexity, cost and amount of reconciliation that I need to do.
Some of the hype over DLT was because people didn’t really understand what it was; they had only heard about it. The potential of the nirvana that’s out there is quite impressive, but how do you get there from where we are today? That is going to be a rocky road because you need full adoption for it to work, and you don’t get full adoption without a big bang—is the business case really there to justify that migration pass for five to 10 years to get there, and who’s going to fund it in the meantime? I don’t think those business cases exist.
We know the endgame, and all the consultants have done work around calculating what that business case looks like, but the adoption plans are going to be very hard.
What developments are you seeing around technology in the custody space, and how do you see it changing in the next five years?
If we talk about technology in the custody space, there is more than just DLT. One of the things that is incredibly important to us is the way in which we deliver data to our clients. Historically, we’ve had SWIFT, our proprietary communications platform, where the clients log in and see the activity.
We are now investing much more into application programming interfaces (APIs). We have many clients live on this already on some of the basic functionality of settled transactions, positions for cash and securities. In addition, we are running API trials with clients for income postings. I see the API side as really growing for us. This area is not the next five years but next year.
Are there any particular countries that you see as ahead of the game with that technology in custody? For those that are lagging behind, what do you feel they need to do to step up their game?
Technologies differ for each client, not necessarily depending on their country but depending on their business profille. That distinction is not around countries, but at the same time for sub-custodian clients, communication is almost exclusively SWIFT. Significantly, some of that communication is migrating from ISO15022 to ISO20022.
It’s really more a case of the type of client, and that’s why it’s important that we, as a global bank, have the right tools available for our clients to communicate with us. For those on SWIFT, we communicate on SWIFT, for those who prefer to use our own proprietary channel, they can do that too. Those that are sophisticated enough to take APIs, we can communicate with them in APIs.
Looking forward, what’s next for BNY Mellon regarding custody, looking to the year ahead?
There are always many projects going on but they can probably be grouped into four themes. One is around efficiency; although we are a people-intensive business, we’re always trying to find ways to further automate the efficiency of processes. We have initiatives on corporate actions, tax processing and client onboarding to make these areas more efficient.
The second theme concerns the areas where we have to adapt to regulation. While the waves of regulation may have quietened down a little bit, there’s still plenty coming down the pipes—for example, the Central Securities Depositories Regulation (CSDR), Shareholders Rights Directive and Money Market Fund Reform.
The third theme is around product development, as there are new markets that are opening up where the ways to connect into them are constantly evolving.
China is a very good example, of course, where we go from stock connecting to bond connecting to the bond market, etc.
That will continue to evolve, and is a continual area of investment for us because we want to give our clients access to those asset classes.
We are also making some big investments into the way that we process funds for clients that want to buy and sell open-ended funds. There is also development in terms of access to new asset classes and new markets, etc.
The final theme concerns the fintech innovation space, where we are working on blockchain concepts. We see significant business potential in some areas but we need to ensure that we understand how we need to adapt our business. For example, bonds and equities on blockchains: we continue to experiment in that space, and we are also investigating the crypto space.
Crypto is a new asset class, but is it an asset class that is ready for custodians to come into that space? Or is it an asset class that we should leave on the side for the crypto websites to deal with?
That is something under investigation that we’ve spent quite a bit of time on because, if you crack that nut, it could be quite significant, but equally if you do it in the wrong way, the fallout could be quite significant as well, so we have to tread carefully.
There is a technology challenge around crypto, a challenge around managing keys and a number of regulatory challenges—our main regulator is the Federal Reserve, but there are many regulators globally that need to agree on the specific and definitions of regulation—and operational risk considerations around all of this.
The main point is that some of the big players in the industry are interested in this space but we have our own list of things that we need to check off before we want to enter into a market, and those things cannot be traded off against each other.
