Development of a global legal entity identifier (LEI) system designed to curb systemic risk is making progress under the auspices of the Financial Stability Board (FSB), but major challenges remain as the industry and individual firms tackle the complexity of implementing the LEI, consider control mechanisms to maintain its integrity and find budget to fund a costly systems change. With an FSB proposed March 2013 deadline for a global LEI system, but no final agreements yet in place on exactly how it will work, the challenge is complete.
These issues and some potential solutions were discussed this week during an A-Team Group webinar, What the LEI Means for Your Firm. The webinar was hosted by A-Team Group editor-in-chief, Andrew Delaney, with LEI experts from sponsors Bloomberg, Thomson Reuters and Depository Trust and Clearing Corp (DTCC) joining the debate.
Delaney kicked off with an introduction to the LEI, describing the need for a system that can avoid a repeat of the 2008 financial crisis and the Commodity Futures Trading Commission’s (CFTC) forerunner to the LEI, the CFTC Interim Compliant Identifier, or CICI. He outlined the federated framework for the LEI that is being developed by the FSB and includes a regulatory oversight committee (ROC), a central operating unit (COU) and local operating units (LOUs), and noted that the LEI is likely to be a unique, non-intelligent 20-digit identifier structured in line with the ISO 17442 standard that has been endorsed by the FSB. Finally, Delaney called on firms to act now if they are to meet the 2013 deadline, commenting: “There is much work to be done and great need to understand the work.”
On this basis, Delaney opened the debate to the panel and webinar audience, asking how the LEI registration process could work in the proposed federated framework. Peter Warms, head of product development for global data and symbology at Bloomberg, suggested existing National Numbering Agencies could act as LOUs, registering legal entities and disseminating their LEIs. He pointed our that the federated approach varies from the initial proposal of a single facilitator and registration authority made by DTCC and Swift before the FSB changed tack and decided on a federated system. “The system will be quite dynamic in how it is organised, but it raises issues of how LEI databases at LOUs will be synchronised, who will manage the content in the central public utility and who will make changes that are required as a result of corporate actions,” Warms said.
Mark Davies, business development director at DTCC, compared the centralised model of the CFTC’s CICI, which has a single point of entry and single registration function, to the federated model that will ultimately be used to issue LEI codes. He said: “The federated model will be agreed in consultation and over time. There will not be a big bang change. The LEI system will start as a centralised system and federated modules will develop form there. We are in consultation with the Association of National Numbering Agencies about federating its existing structure, managing corporate actions, dealing with many languages, understanding legal structures in different jurisdictions and more.”
Agreeing that the CICI model and its implementation by DTCC will offer some guidance for the development of the LEI system, Tim Lind, global head of legal entity and corporate actions in the enterprise content business of Thomson Reuters, hammered home the complexity of the LEI compared to the CICI, saying: “There will need to be interaction between facilitators. How will LOUs be selected and who will step in where there is not an LOU in the local market or no organisation that wants to be an LOU? The LEI system is not for profit, so who will want to step in and who will have legal responsibility in cases where thousands of entities across jurisdictions are tied to one holding company? By comparison, the scope of the CICI is much more controlled. It can provide some learning, but is not nearly as complex as the LEI.”
Considering practical matters around the LEI, Delaney asked the panel to comment on the LEI structure and how the system could be rolled out. The expectation that the LEI will be a 20-digit code created randomly and with no intelligence has caused concern in the market and raised the issue of whether it is possible for more than one LOU to enter a market if codes are generated randomly. Davies proposed two potential solutions to the problem, one being multiple LOUs each with an algorithm to generate unique codes and the other being large batches of codes dedicated to different LOUs.
In terms of roll out of the LEI system, the consensus was that a few firms are taking the early adopter route and are beginning to adapt systems to include the LEI, but many more are waiting for the final LEI process to be defined before making any major moves. If the LEI will not be driven from the bottom up, it will be driven from the top down by regulations requiring the use of LEIs in reporting.
Davies commented: “The code will have other uses, but its adoption will be driven by regulatory activity. The only entities that will not receive LEIs are sole traders and individuals. Others in the scope of the system will have to roll out solutions as regulation requires.” Answering an audience question on the requirement for LEIs in the case of pension funds with many mandates, Davies said the pension fund and each mandate would need an LEI. Company branches and trading desks will not need their own LEIs.
Lind added: “Boiling the ocean and assigning an LEI to every entity on the earth won’t work. We need to identify those that could cause systemic risk. These are entities in capital markets and large entities in the supply chain. We need to focus on these.”
Considering how buy and sell side firms might use and even benefit from the LEI, Lind said: “The LEI itself does nothing in terms of risk, the heavy lifting is attaching data to the LEI and being able to roll it up to see risk exposure. The LEI is a foundation to joining data and creating monitoring activity for purposes such as know your customer (KYC) and anti-money laundering. The aim must be to attach data to an entity and be predictive in risk policies.
“While there is still confusion about the scope of use of the LEI, some of our clients are taking a tactical approach, perhaps fitting in with the CICI rules for swaps until there is more regulatory pressure to use LEIs. Others are more strategic, considering the LEI as a catalyst to look at standardisation and governance of entity data with the LEI being part of a larger programme.”
If financial firms have much to do, so too do market data vendors, but there is little hope that the latter’s systems will match up perfectly and make data consumption painless. Warms explained: “Integration of the LEI is critical for data vendors. There must be full coverage of entities globally and the biggest challenge will be mapping entities to the underlying securities they represent.”
Davies noted that the scope of the CICI will never be broad enough to provide a mapping table for existing identifiers and instead suggested the primary means of mapping LEIs to existing identifiers will be handled by data vendor or organisations with experience of extract, transfer and load tools. Unlike identifiers such as ISINs and Cusips, the LEI adds complexity in the need to match meta data around the central identifier to existing codes. Noting that each LEI is likely to have seven or eight data points that help distinguish entity names in the central repository, Warms said: “At the end of the day, clients will ask data vendors to integrate the LEI, but if they think entity hierarchies from Bloomberg and Thomson Reuters will be identical that is unlikely. This means rolling up data for risk reporting could be flawed as there are two different data sets. There are a lot of challenges around building hierarchies and understanding relationships within them.”
The cost of putting the LEI in place escaped no-one. As Warms put it: “The fee schedule – $200 to register a CICI and $100 annually to maintain it – is minimal, the real cost is in database development and integrating the LEI. Many firms have a lot of information in silos and they need to build out a database on top of all the assets they cover, which is how they work with KYC and onboarding. The cost of the build between now and March 2013 will be significant.”
A stumbling block identified by Lind – and one which he feels the next FSB LEI Private Sector Preparatory Group (PSPG) meeting on 15th and 16th October should address – is control around getting an LEI through a portal. He explained: “This could be done fraudulently when the need is for bona fide status of entities. There must be a quality control process, this is the single biggest barrier to moving forward.”
If this is one issue the PSPG must turn its attention to, there are many more. As Davies concluded: “The next PSPG meeting will not expand the mandate. It is designed to provide feedback from PSPG work streams and an opportunity to understand some providers potential solutions for the LEI. I expect the outputs to be advances by regulators in putting a charter for the LEI system in place by November and regulators then signing up and sitting on the ROC. This will establish the principles of the LEI, but there is still a lot to do around the definitions of the COU and LOUs, how they will hand off and how the LEI numbering system will work. Everyone is working towards March 2013. By then, there should be a COU with core processes in place. This will be the genesis of the LEI system.”