Getting to Grips with the Details of LEI Standardization

As soon as it emerged that regulators would allow more than one pre-local operating unit (pre-LOU) to issue pre-legal entity identifiers (pre-LEIs) before detailed standards or a coordinating authority had been established, it was obvious industry participants would face challenges when using the identifier, says Ken Traub, Boston-based standards strategy consultant at GS1. This has proved to be true.

Seven pre-LOUs (and counting) are now issuing pre-LEIs in accordance with ISO 17442. However, there is no central operating unit (COU) to consolidate the data or ensure that all pre-LOUs use the same file format. As a result, industry participants who want up-to-date pre-LEI records potentially have to monitor files issued by all seven pre-LOUs and reconcile differences in the file formats they use.

Fortunately, these challenges were anticipated by a consortium of four organizations: not-for-profit standards body GS1; the Corporation for National Research Initiatives, a not-for-profit organization that supports information infrastructure research and development; FIX Protocol, a not-for-profit standards body; and Tahoe Blue, a data standards and risk management consultancy. After the second pre-LOU went live this year, the four organizations launched P-LEI.org-a website that collects data from all operational pre-LOUs every night, maps it to a common schema and makes the data freely available in a consolidated file. (The London Stock Exchange now also provides consolidated files of pre-LEI data.)

User Friendly

The organizations behind P-LEI.org have been involved in the LEI initiative from an early stage and Traub says the schema used by the website has its origins in the discussions of the LEI Private Sector Preparatory Group, which recognized that ISO 17442 defines the LEI data elements at a high level and that more detail was needed. When rendering the schema in XML, Traub says the consortium wanted to make it as easy to consume as possible. As a result, they avoided creating unnecessary internal structures and made provisions for extensibility.

The resulting schema is free for others to use. Traub says Takasbank, the Turkish pre-LOU, has adopted the schema "pretty much as is" and the P-LEI.org group is keen for other pre-LOUs to do the same.

P-LEI.org performs an important service by making all pre-LEI data available in the same place and in a single format, but it doesn't cleanse data. Traub says there is a good reason for this. "We felt the value in consolidating the files was to make it so that the people didn't have to go to the original sources," he explains. "If we started changing the data and we only produced a file that had data that we had interpreted in some way, then there would be a reason for people to say, ‘That is useful, but I think I still need to go back to the original source because I am not sure what has been done to it.'"

Data Quality

While it doesn't cleanse data, P-LEI.org does publish research on the pre-LEI data it consolidates. Traub says there is "a lot of variation" in data quality. "There are certainly fairly obvious data quality problems-incorrect country codes is one example," says Traub. "We encounter data where names of states in the US are misspelled, where the city and the state are swapped. There are things like that which are fairly obvious, and suggest that there may be other data quality issues that are a bit more subtle."

Traub acknowledges that pre-LOUs "face a challenging problem" to verify the pre-LEI data. "The data submitted by the registrants themselves is not always of high quality, and it is not always easy to verify," says Traub. "Data quality is an area that I think everyone would agree will need more attention. There is definitely work to do there."

More Standards

Traub believes the answer lies in greater standardization. He says ISO 17442 is just the "tip of the standards iceberg" and that more standards are urgently needed.

"The industry needs standards that encompass not only agreed file formats and things like that, but also the provision of more crisp and precise definitions of what each of the data fields means, aligned definitions of status codes, guidelines for data quality and recommended best practices," says Traub. "These are things many participants have raised as well, but what is lacking is any kind of formal standards-setting process to put those [standards] in place. Everybody is hoping the COU gets formed very soon and can hit the ground running and start working towards that, because that is the most critical thing."

For more on this topic, click on this link to see a video interview with Ken Traub.