The global legal entity identifier initiative is well underway with a Regulatory Oversight Committee providing governance and pre-Local Operating Units assigning legal entity identifiers, but the ultimate success of the system depends on the as yet unformed Central Operating Unit, the quality of entity data and its ongoing maintenance.
These issues and more related to legal entity identifiers (LEIs) were discussed at an A-Team Group webinar this week and you can read more on this hot topic by downloading an A-Team report here.
Andrew Delaney, editor-in-chief at A-Team Group, moderated the webinar entitled Practical Applications of the Global LEI – Client On-Boarding and Beyond, and was joined by panellists Peter Warms, head of product development for global data and symbology at Bloomberg Data Solutions; Scott Preiss, vice president at Cusip Global Services; and Tim Lind, global head of middle office and entity, enterprise content at Thomson Reuters.
Delaney set the scene for discussion by reporting on recent A-Team Group research into use cases for the LEI. Entity and client data managers at 10 tier one sell-side and custodian organisations took part in the research, which found most suggesting a standard LEI could be useful as a means of getting control of entity and client data, but that it raises serious operational challenges in terms of integrating entity data with adjacent datasets. The research participants noted the potential of the LEI in high-cost operations such as client on-boarding and know-your-customer (KYC), and a surprisingly large segment of 70% said they had significant budget to reorganise entity data management. The idea of a utility for entity data was also mooted during the research and while such concepts are usually popular, the research found ongoing conversation but caveats around issues including data quality and service levels.
Kicking off the webinar discussion, Delaney asked panel members how they viewed the March deadline set for the introduction of the global LEI system by the Financial Stability Board (FSB), which led the initiative until handing over to the LEI Regulatory Oversight Committee (ROC) in January 2012.
Lind commented: “There was no champagne and no cake. This is a marathon, not a sprint. A lot of development of the LEI system will occur in its roll out, the March deadline date was just part of the momentum.” With the ROC in place providing governance and infrastructure for the system, Lind added: “This is important, but ultimately it is just the framework. Much more difficult is putting in place the Central Operating Unit (COU).”
Warms agreed, noting that while the ROC has been set up, the COU is not expected to be established until later this year and that, to date, there are only seven Local Operating Units (LOUs) authorised around the world.
These may be early days for the LEI, but Preiss noted the Commodity Futures Trading Commission’s (CFTC) April deadline for all OTC derivatives trades to be reported using CFTC Interim Compliant Identifiers (CICIs) that are intended to morph into LEIs as the system develops. He also described the required use of LEIs by the US National Association of Insurance Commissioners as a mandate that will accelerate and spread the use of LEIs.
Turning to those who must take LEIs into their data management systems, Delaney questioned how the panel members’ clients are approaching the problem. Lind said the LEI will initially be limited in use to credit and counterparty risk reporting and noted that the CFTC reporting deadline is the primary interest around the LEI. But he cautioned: “The CFTC is driving this at the moment, other regulators are behind. The system could be fragmented if other regulators do not follow.”
Warms concurred, saying he expects the LEI to be used on the reference data side to roll up exposures to an LEI, and in the front office for client on-boarding and KYC. While these use cases are historically separated within a financial institution, he said some large banks are developing more sophisticated models that bring them together to share data.
While this can be the case, Delaney referred to A-Team Group’s research, pointing out that most participants managed entity and customer data separately, leading to the question of whether the LEI could bring the data silos together. Lind responded: “Some organisations are centralising the capture and validation of all entity data, but this creates a huge compliance issue. A centralised approach needs a corporate standard, domain expertise and the ability to convince internal clients that data quality and service will be better than that provided by silos. This is more of a challenge than centralising securities data.”
Mapping vendor and internal proprietary entity codes to the LEI is also a major task, although panel members suggested embedded schemes such as BICs and Swift messaging will continue to be cross-referenced and not replaced by the LEI for a long time. That said, Lind commented: “The LEI provides an open system of entity identifiers. This is a new era. There are places where LEIs will be useful and places where they won’t be, and it is premature to architect systems around LEIs. The use of standard entity identifiers in swaps is sure, but we don’t yet know the next step.”
In terms of client on-boarding and KYC, panel members reflected Lind’s view on the usefulness of the LEI, with Warms suggesting the LEI could provide commonality for entity and on-boarding data, although using the LEI as one comprehensive means of entity identification will take time. Preiss pointed out that the limitation of the LEI to financial transactions may restrict it from meeting the full needs of on-boarding and KYC, which often involve organisations and individuals that are not involved in financial transactions. The LEI also fails to challenge the need to gather non-public information for KYC tasks.
Whatever its use, panel members agreed that LEI data must be of high quality and must be maintained. The issue of maintenance has yet to be settled by the LEI ROC and could cause a breakdown in the global system. Warms explained: “At the moment, conversation focuses on registration of entities, but what happens next? The LEI system is all about data quality, but will those registering entities update them in a timely way? I doubt this will happen without regulatory compulsion.”
Warms also noted discrepancies in entity data records from two operating LOUs, DTCC’s Avox in the US and WM Datenservice in Germany. He questioned: “Will market data vendors use data from various LOUs or will there be central determination from the COU. It is hard to take data daily from even seven LOUs and deliver it to clients.”
Lind added: “Multiply the number of LOUs and if each one publishes data to customers the customers will have to build integration to each LOU to take in the data. This would cause chaos. This part of the system needs to be centralised.”
With such fundamental issues still to be resolved, Delaney asked how the global LEI system will develop over the next year. Warms said: “It is notable that the Private Sector Preparatory Group [set up by the FSB to work on LEI development] has been working on recommendations for some time, looking at details around standards, data hierarchies, thresholds and more. It has provided recommendations to the ROC and its Committee for Evaluation and Standards, but when will we hear the next steps based on these recommendations? The key component of the system, the COU, is also slow in development, leaving open to question issues such as whether LOU databases will be tied together and whether all data will be linked to the COU? At the moment, checking for LEI duplication in other LOUs is a manual, archaic process that doesn’t work. There is a lot to do, but it could work through the COU.”
Lind took a similar view, concluding: “When it was decided that the global LEI system should be based on a federated model it became clear that it would be hard to govern and operate. We need the COU to act to avoid the potential of LOUs duplicating LEIs. There is an upside to making LOUs local, but the downside is a very complex requirement for full integration and transparency with the COU. The COU will make or break the system.”