Galen Stops takes a closer look at the distributed ledger technology solution that HSBC has deployed for settling FX transactions.
One of the long-standing accusations levelled against distributed ledger technology (DLT) within financial services is that it is a solution looking for a problem, an accusation that appears to have gained some weight over the past few years given the vast disparity between the initial level of hype around this technology and the actual implemented use cases for it.
Which is why it is significant that HSBC recently announced that it has settled more than three million FX transactions and made more than 150,000 payments worth $250 billion using DLT. If this announcement caught FX market participants by surprise, this is probably because this DLT solution, called HSBC FX Everywhere, has thus far only been used to orchestrate payments across HSBC’s internal balance sheets.
The project to develop this DLT solution began in late 2017, when staff at the bank started fleshing out ideas regarding how the technology could help its businesses and then created a proof of concept (PoC) regarding how to execute these ideas in reality. In early 2018, HSBC FX Everywhere went live on a very limited basis, with a handful of trades manually being put through, while the rest of the year was spent scaling up the number of settlements on the network.
In developing this technology, HSBC identified three main benefits that could be derived from it. The first is that it provides a shared, single version of “the truth of intra-company trades”, from execution through to settlement, which reduces risk of discrepancy and delay. Secondly, it means that confirmation and settlement is automated by matching and netting transactions, which reduces costs and reliance on external settlement networks. A third benefit is that DLT provides a consolidated, global view of forward cash flows, and certainty of funds throughout the funding cycle, supporting greater balance sheet optimisation.
One of the barriers to the adoption of DLT or blockchain solutions by the very largest financial institutions is that the cost of replacing existing technology or systems is often vast, and to make a business case for doing so, any new systems need to more than offset these costs. Indeed, Profit & Loss previously quoted one banker stating that his team would only look at DLT solutions where it was projected that there would be a 50% or more cost in reductions or capital savings from the current systems in place.
HSBC though, was able to prove the cost-savings necessary to justify this FX-focused DLT project, in large part, because of its structure. The firm is effectively organised as a conglomerate of different banks with local balance sheets in their own jurisdictions. As a result, these banks have traditionally operated somewhat at an arm’s length from one another, meaning that they process and settle with one another just as they would with another bank on The Street.
“What we looked to do was not to change the local bank systems in each operating entity, but rather to overlay new technology so that we can get more transparency on what the settlement activities are across the forward ladder, get greater certainty around what the net settlement amounts are, or the cash calls that are happening between the local entities. Then we’re using a workflow tool within the DLT to coordinate the settlement amongst the entities, which allows us to do safe settlement and payment versus payment between entities, because they have visibility and transparency of where the cash is at any moment in time between the two legal entities that have to settle,” explains Jesse Drennan, a global business process engineer at HSBC.
He adds that if HSBC was a single balance sheet entity without all these different local banks acting under the brand as nodes on the distributed ledger network, it would have been hard to achieve the scale necessary to justify the underlying business case behind the technology roll-out.
According to Drennan, HSBC is already seeing the operational efficiencies and settlement efficiencies that were projected ahead of the DLT launch and now the next step is to continue growing the solution within the bank’s internal network and to look at ways to leverage this new capability for its clients. In particular, the focus is going to be on how HSBC can use this technology to help its corporate clients, which remain at the heart of the bank’s franchise.
“What we think is interesting about what we’ve done is that if you think about cash today on the ledger it’s very anonymous, you don’t necessarily know where it’s come from and what it’s doing there – if a firm needs to hedge £100 million, they just do so and that amount pops onto the ledger. By contrast, we think that DLT offers an opportunity to revisit the workflow and also to look more holistically at new technologies to see how they are interacting and working. So we believe that by packing metadata and context around the cash flows we can provide real world awareness as to where those cash flows are coming from and what’s driving them, allowing firms to react accordingly,” says Drennan.
Giving an example of this, he states: “Say you have a container ship that’s out to sea, there will be some cash on the ledger that reflects the commercial activity of goods on the ship. Now if the ship is delayed because of a storm and then re-routed, subsequently delaying it by a week, I want to dynamically hedge based on the real world inputs that are based on what’s happening to that boat in real time as reflected on my ledger and then bring that back to the marketplace in terms of how I’m situated in regards to my FX hedging.”
Discussing the challenges associated with getting such a significant DLT project up and running, Drennan says that getting the legal aspects of the solution determined consumed the greatest amount of time. In part, this is because HSBC wanted to ensure that there was irrevocable settlement finality on these FX trades, and knew that simply having settlement reflected on the distributed ledger itself wasn’t enough to satisfy the legal requirements in all the various jurisdictions where the bank is active.
“We were looking operationally at how we implement this technology, but working closely with the legal team on the settlement finality in the currencies that we were focused on. So the legal team would provide real-time feedback like: ‘If you do ‘x’ then these are the legal risks that are created, or this analysis is no longer applicable, is that what you want?’ And this created a feedback loop between the legal outcomes and legal certainty that we were after and the operating model that we were implementing,” says Drennan.
Another legal issue that had to be considered is data residency: where the data actually sits and who has access to it. The idea of settlement netting – taking a population of trades, netting them down and then exchanging values – can be very inefficient. For example, if two parties disagree on the final figures, then unpacking potentially thousands of trades to find the discrepancy can, as Drennan puts it, be a “nightmare”. Being able to centralise this process and have a central point of reference is therefore clearly beneficial, but historically one of the challenges to doing so has been that different jurisdictions have different data restrictions and limitations on offshoring data.
“The good thing about permissioned ledgers is that we can encrypt the data in tranches. This means that customer data and the data that regulation are seeking to control can be encrypted natively at the local site, the local entities can hold the key and can demonstrate complete control over those data elements and who has access to them. This allows us to consolidate the common data elements together so that we can implement common processing. This is very powerful given our large footprint, because it enables us to stay compliant locally whilst also consolidating our data set,” says Drennan.
A catalyst for change
FX isn’t the only business segment where HSBC has been implementing DLT. In March 2018, the bank announced that it had completed a trade finance transaction using this technology – but the structure of FX transactions means that in some regards, it was well suited to it.
“In FX, and in particular cross-border payments, there is this multi-party nature of the settlements. This is where the DLT gets interesting, because you have the counterparty to the transaction, you have the nostro agent potentially doing ‘on behalf of’ settlement with the counterparties, so this means that there are third or fourth parties involved. The ability to link these parties together, extend permissions to see the data and manage those permissionings natively at the data layer is an important feature of DLT, and it’s one of the reasons why we looked more closely at it for FX,” comments Drennan.
Hearing about the use of DLT to settle FX transactions logically suggests questions about the role of CLS, which also has a DLT project that went live in November 2018, which is a bilateral payment netting solution for the FX industry. Drennan though, is quick to insist that CLS still plays a “crucial” role in the market and says that its launch in 2002 was central to the expansion of the FX market that subsequently occurred. However, he does add that growth of fintech is increasingly challenging certain underlying business assumptions for financial services firms, and that ultimately, these firms may have to adjust their operating models in response to changing customer expectations.
This, says Drennan, may just be a natural part of the continuing evolution of the FX market.
“[Fintech] is changing our customers’ service expectations and therefore transforming our business model. But the FX market has changed a lot in the past 20 years and so there’s questions about whether this is part of an evolution that would have come about regardless. Maybe DLT is just a catalyst for something that’s always been the direction of travel,” he says.