Arjun Jayaram, CEO of Baton Systems, talks to Profit & Loss about how distributed ledger technology (DLT) is already helping the FX market, why the hype cycle around this technology is coming to an end and who he expects to win the “blockchain wars”.
Profit & Loss: What is the problem in the FX market that you think DLT is a good solution for?
Arjun Jayaram: The problem that we see is that large amounts of money have to be pre-funded by each party to do an FX settlement, for anywhere from a few hours all the way up to two days. The reason why this is such a big problem in FX is because of the very large notional values that are being settled every day. With trillions of dollars being traded, tens or hundreds of billions actually have to be pre-funded to facilitate the settlement of these transactions. There’s a real cost to this, both on the balance sheet and in terms of settlements costs; it also directly impacts risk for all the counterparties involved. So we looked at this problem and decided that it could be solved using DLT without the need for tokenisation.
We use DLT technology to facilitate payment versus payment with settlement finality between two parties by eliminating the need for pre-funding up to the point of time of settlement, the moment when the parties or the principles of the transaction need to come up with the money. Throughout the process, they have complete visibility, from the trade all the way up to settlements, on a netted or gross basis, across multiple currency pairs. It’s seeing rapid pick-up, so that we have already had double digit billions of dollars settled by our technology.
P&L: Is there a scalability issue here? Or is it the case that once the technology is settling one transaction this way does it then not matter how much is pushed through the pipe?
AJ: What matters is that as you get more and more trades from each counterparty, you need to be able to scale up to meet the level of demand, because these trades are coming in at a very, very fast pace.
We believe that for any solution to scale to the needs of the capital markets, it has to handle at least a billion events in a day; the traditional blockchain or distributed ledger technologies are nowhere near capable of that. That’s why we built a DLT solution from scratch using commodity hardware that enables us to handle up to 83,000 events per second, per channel.
P&L: Can you go into more detail then about how your technology is different to “traditional” blockchain?
AJ: There’s a technology piece and a market infrastructure piece here to consider, and I think that you have to separate the two. On the market infrastructure side there needs to be someone to run the blockchain, to be responsible for deploying software onto the chain – there’s a governance around the infrastructure that is required.
On the technology side, we came to realise early on that blockchain does have some really interesting properties, such as the transparency, the auditability of the data, the fact that the data is tamper resistant, the ability to do hashing and the linking of transactions, etc.
“I think we’re going to see what I call the “blockchain wars”, and I do not think that small companies are going to win this”
All of that is good, however, there were some problems around tokenisation and around the scalability of the technology. So we wanted to keep these good properties of the blockchain but bring it down into traditional technologies that are well proven.
That’s why Baton uses DLT but within that, there are two parts: the logical layer and the physical layer. The logical layer provides that tamper-resistant data with a complete audit, but then the different nodes on the network, the underlying data sources, can have entirely different technologies. So you could have Ethereum on one side and Corda on the other, and then have a Db2 or Oracle on the third side and essentially they could all still operate on top of the Baton technology. That’s a big distinction because now you have technologies that are capable of scaling up to capital market needs, it’s just that we sit on top of that.
P&L: When it comes to implementing this technology, what’s the biggest challenge that you come up against?
AJ: There are technology challenges and there are non-technology challenges, and I don’t want to trivialise either. The technology challenge is the interoperability. This is important because everybody who’s coming in and talking about these newer technologies want to rip up and replace existing systems and that is extremely hard – in fact, almost impossible to do simultaneously – in a large bank. That’s why we felt it was important to be able to interoperate with the existing technologies.
The non-technology part is just the intrinsic issue of being a small company trying to solve a big problem. We have to get through the vendor onboarding process, which just takes a lot more time than we would like.
P&L: When it comes to DLT there was this hype cycle where at one point, people were touting DLT solutions for everything under the sun. A lot of that hype seems to have burnt away a bit and a lot of the proposed DLT projects haven’t come to fruition – why do you think there’s been such a stark disparity between the enthusiasm for this technology and the number of firms offering solutions using DLT and the lack of actual real live implementation of it in financial markets?
AJ: That’s a great question. We’re technologists, but before we even wrote a single line of code, we spent 18 months just learning the problems that existed. We looked at which specific points payments slow down and why, what exactly happens when you move an asset from point A to point B, how it works when these points are on different continents and what existing technology the banks were currently using for all this.
So we had a very deep understanding of the problem that we were addressing, and also the constraints around the problem – from a technology perspective, from a regulatory perspective and from a market dynamics perspective. Not everyone in the blockchain space did this, and I also think that when there’s any hype cycle, whether it’s the Internet and the dotcom days or the big data space, there’s just a lot of over promising as the investment dollars are pouring in. This is exactly what happened with blockchain too, with people looking at a solution trying to see what problem it could solve.
“Blockchain itself will become a commodity. The value is in the smart contracts or the workflows that you deploy on top of that”
But I believe that we’re coming out of that hype cycle now, some of the pretenders in this space are already stumbling and some of them will go away, and then we’ll be left with a much more solid set of small but maturing companies that actually offer real solutions. I think that this is a very important phase because you’re beginning to see some systems using this technology that are live in production now.
I think we’re going to see what I call the “blockchain wars”, and I do not think that small companies are going to win this. The blockchain wars from an infrastructure perspective, are going to be won by the likes of Microsoft, Google, Amazon or Oracle, not by a small firm like Baton – our value proposition is in the vertical, such as FX settlements or collateral settlements. If you’re small, you need to target a niche, at least initially.
P&L: Why do you think those big firms will win? Does it just become a scale game?
AJ: It becomes a scale game, but it also becomes a legacy technology game. If you’re a bank and the established firm that you use to store your data with says that they can migrate everything to their blockchain, why would you go and implement another blockchain on top of that? Amazon is already offering blockchain-as-a-service on Amazon Cloud. That is why at Baton we separated the physical data store from the logical DLT, because the blockchain itself will become a commodity. The value is in the smart contracts or the workflows that you deploy on top of that.
P&L: Do you think there’s still a lot of misconceptions about DLT with regards to what it’s good for and what it’s not?
AJ: Yes. DLT or blockchain-based technologies shine only when there is a market need for multiple parties to interact with no central party in-between. If there is no need for multiple parties to interact, then there is no need for DLT.
So if it’s a trade between two parties, there is golden copy of that now and you don’t need this technology. Whereas if you look at something like trade finance, where there is an exporter bank and an importer bank, there are customs officials, insurance, etc. That makes sense. A syndicated loan would make sense, because you’re distributing it to multiple firms. This is where the technology shines. But we shouldn’t over emphasise this point and say that it will solve all the problems. DLT will solve the data distribution problem and transparency problems, but it won’t solve everything.