Digitized securities markets depend on efficient flows of structured data, but the lack of standardized data models could soon start to impede their growth.
Data is now the primary ingredient of the securities services industry. It governs the efficiency of settlement, safekeeping, asset servicing and regulatory reporting. Data is also driving product development. Ultimately, the front-to-back-office outsourcing services global custodians are building to support their global asset management clients are data management platforms.
“Data is extremely valuable, not only in terms of cost savings but also in terms of new services and new revenue opportunities,” said Juliette Kennel, head of standards at Swift, leading a panel of industry experts to discuss the topic at Sibos 2021. “The value of data depends on our ability to share it and exchange it efficiently. Different standards, or data models, obviously make that interoperability much more difficult.”
The key to interoperability is agreement between the owners of different standards to use a common dictionary of standardized data definitions. Kevin Houstoun, executive chairman at Rapid Addition and a Fix board member, says eliminating overlap between data standards such as Swift, Fix and FpML would immediately save securities firms the trouble and expense of supporting multiple data models. Yet unlike payments, where banks have committed themselves to migrate to the ISO 20022 standard as from November 2022, there is no agreed timetable for the securities industry to adopt a common standard.
One reason is that the securities industry has gotten used to managing multiple standards. Another is that the existing ISO 15022 standard is enabling 90% of securities transactions to settle on time already. According to Stuart Warner, head of direct custody and clearing, Europe, MENAT and the Americas at HSBC Securities Services, the remaining inefficiencies are attributable to other causes: complex cross-border transaction chains and poor inventory management. “The results of a survey conducted by ISSA revealed that there isn’t the business case for a wholesale migration from ISO 15022 to ISO 20022,” said Warner.
In addition, the costs of maintaining legacy systems, keeping compliant with regulations and the preference of securities firms for building rather than buying systems devour technology budgets. “Standards migrations are going to be even lower down the pecking order than the challenges and opportunities to innovate,” says Houstoun.
New data demands and APIs
However, market forces are beginning to make this less sustainable. One factor is the rise of digital assets, which are multiplying data sources and flows. “If we go on as we have, the cost of duplication of interfaces and non-standard processes will increase,” says Vicky Kyproglou, head of network and market infrastructure management at UBS.
Pressure to collect and disclose data about the environmental, social and governance (ESG) impact of investing is further increasing demand for access to new sources of information. “It is creating whole new sets of data, which need to be ingested by sell-side and buy-side actors as well, for which there is no standard at the moment,” says Alex Dockx, head of industry development for custody at JP Morgan.
The principal technique for facilitating such data flows is application programming interfaces (APIs). These make it easy for providers to push data to clients and clients to pull data from providers. Yet the lack of an industry standard for the data structure of APIs is causing formats to proliferate. “It does definitely produce extra work and extra effort,” cautions Warner.
So do legacy systems. “There is this incremental cost of sticking to old standards, having to repair things manually, lots of discrepancies, lots of errors in transmission as well, so that is a cost that is not materializing on day one but it keeps on creeping up over the years,” notes Dockx.
Challenging the status quo
Other market forces are increasing the pressure to change the data model as well. One is competition from fintechs unencumbered by legacy technology for new, data-driven business opportunities such as tokenization. Another is adoption of ISO 20022 in areas where new messages have been developed, such as corporate actions, proxy voting, account opening and the messages designed to support implementation of the second iteration of the Shareholder Rights Directive (SRD II). “ISO 20022 allows for greater STP of these messages, improving efficiency while reducing risk,” explains Warner.
He adds that adoption of ISO 20022 by financial market infrastructures (FMIs) such as central securities depositories (CSDs) is also encouraging custodian banks to follow suit, if only to maintain local market connectivity.
Market forces nevertheless act slowly. “A standard only becomes meaningful if it is either mandated by a regulator or market infrastructure or it organically gains critical mass,” says Warner. For Kyproglou, this was too slow. The crucial factor is network effects, which need to be stimulated. “How do we create network effects, be it for ISO 20022 or even a different standard?” she asks.
“At some point, I believe network effects will apply in securities, but in the meantime, I think we need to support our users so they can adopt ISO 20022 when it makes sense for them, which really means at their own pace,” says Kennel. “Co-existence and multiple formats are here for the foreseeable future so I think we need to make this as pain-free as we can.”
Progress depends on industry collaboration
Operational solutions include tools to ease translation between data formats. But Kyproglou reiterates her warning against acceptance of the status quo. “Leadership is required to change this and make the industry focus on the value-add rather than translation,” she says. “This is not a problem that one firm can solve or even one person. We need an industry approach.”
This article was first published in Waters Technology and is published here with kind permission.