JP Morgan DLT exec: Settlement rails needed for digital bonds to gain traction

Will distributed-ledger technology (DLT) take off as an infrastructure for financial markets? This was the question that dominated conversation at the Association for Financial Markets in Europe’s (Afme’s) Bond Trading, Innovation, and Evolution Forum on Tuesday, February 27. Market participants from the buy side and sell side turned to issuances of digital bonds to illustrate the virtues and limitations of the technology.

Scott Lucas, head of markets DLT at JP Morgan, said these issuances had demonstrated DLT’s potential to deliver faster settlement, new approaches to distribution, and different methods of custody. “But cash is a key problem. And until there are the appropriate settlement rails to effectively manage the purchase, sale, and payment of coupons, I think we’re going to bump up against a boundary of efficiency that can be achieved or experimentation that can be done,” he said, speaking on a panel at the Afme conference.

Getting a significant proportion of the cash market to move will require the technology to continuously demonstrate its utility, Lucas said. But he added that it also requires investors to see the possible trading outcomes resulting from new functionalities—like calculating coupon payments to the second and resolving questions around clean vs. dirty prices.

“If you’re a capitalist, and you believe that better technology will win, you believe that at some point, DLT will be the technology we use. But there’s been a litany of better technologies that have never quite made it, because the investment threshold to get over is too high,” Lucas said.

Digital bond issuances could cut trade processing costs and free up liquidity that is currently locked up in settlement procedures. Societe Generale’s digital assets subsidiary, Forge, issued the first covered bond on a public blockchain in 2019.

But progress since then has not been as fast as some proponents of DLT had hoped.

“It certainly feels as though there’s increasing size, sophistication and frequency. But it’s still tens of issuances compared to the regular market,” Lucas said. “So there’s a considerable gap to close.”

Christoph Hock, head of multi-asset trading at Union Investment, agreed that DLT needs to demonstrate its value to end users.

“What’s in it for the investor? What’s in it for the client? It’s not implementing DLT for the sake of new technology. It really has to prove that it delivers benefits for our clients,” he said.

But Hock added that there was no lack of use cases where DLT could play a role in the lifecycle of bond issuance and trading, which is highly manual.

For example, he said that issuers typically have T+5 to T+7 settlement timeframes with new issuances, which increases a firm’s risk. Additionally, when issuing a physical global document, the market standard is that these documents are 80 to 100 pages, and they need to get delivered by a courier from the issuer to the central custodian.

There are very few that are fully digital end to end. Because the markets aren’t ready for fully digital end to end

Scott Lucas, JP Morgan

He said that this process is “quite painful” for asset managers to get the static data as a part of the new issuance process into their order and execution management systems. 

“When we look at trading platforms, we still have an oligopolistic structure as of today, which leads to, from our perspective, too high costs of trading,” Hock said. “With settlement, we have typically T+2, so when we’re looking at DLT-supported solutions, we are talking about atomic settlement. Also lifecycle events … the coupon payment or the payment at maturity is still a lengthy process: With some Isins, it takes up to five or six days before we get the credit from the CSD via our custodian.” 

Still, panelists recognized that there have so far been few examples of bond issuances that were able to fulfill all of these use cases.

“There are very few that are fully digital end to end. Because the markets aren’t ready for fully digital end to end,” Lucas said.

Nor is it clear how markets will reach that fully digital stage, given that DLT is used for such a small portion of bond issuances currently. He noted that industry participants are reticent to embrace unproven, new technologies because of the cost associated with such moves. Lucas hoped, though, that while DLT’s expansion in the capital markets has been marked by fits and starts, a foundational layer is being created whereby new tools can more easily be built on top.

The reason for hope, as Lucas put it, is because—in theory—when you trade on a DLT platform, the coupon is calculated to the second; the price is “clean”; people know what time the trade was settled; and “you can still net them all down and settle them at precisely [9 am] tomorrow morning. And in some window between [9 and 11 am], you’ll see what comes out at the end and failed.”

Lucas added a caveat: “It will take investment. Are we going to realize the value on investment in enough time for market participants that typically have a shorter horizon than two to three years to make that investment? That’s the question.”