Data accuracy is an important topic when it comes to Legal Entity Identifiers. LEI numbers, when used for trade reporting and KYC or onboarding in the financial industry, need to be accurate in order to ensure that checks and validations are trusted.

In this article, I hope to give you an overview of some mechanisms which exist to ensure greater data quality and accuracy. I also highlight something the GLEIF will be rolling out next year which will also help to address some inconsistencies in LEI records by jurisdiction.

“Data quality doesn’t fall from the sky – it has to be managed”

Stephan Wolf, Chief Executive Officer, Global LEI Foundation (GLEIF) .

When LEIs were first introduced, data quality was already top of mind for the GLEIF. They were already implementing data management programs to offset some of the early data inaccuracies that were seen when LEIs were still new.

First, they check formats and compliance with XML schemes. Local Operating Units (LOUs) who provide a file that is not compliant will be rejected. Implementation of the attributes are also looked at by the GLEIF. Data uploaded into the system is tested and LOUs get a daily report o the findings and issues with their records.

Finally, a data quality report is published by the GLEIF on a monthly basis. Data quality reports look at 12 dimensions of data quality outlined in GLEIF Global LEI Data Quality Report Dictionary.

1. Accuracy – The extent to which the data are free of identifiable errors; the degree of conformity of a data element or a data set to an authoritative source that is deemed to be correct; and the degree to which the data correctly represents the truth about real-world objects.
2. Accessibility  – Data items that are easily obtainable and legal to access with strong protections and controls built into the process.
3. Completeness – The degree to which all required occurrences of data are populated.
4. Comprehensiveness – The degree to which all required occurrences of data are populated.
5. Consistency – The degree to which a unique piece of data holds the same value across multiple data sets.
6. Currency – The extent to which data are up-to-date; a datum value being up-to-date if it is current at a specific point in time, and outdated if it was previously current but then incorrect at a later time.
7. Integrity – The degree of conformity to defined data relationship rules (e.g., primary/foreign key referential integrity).
8. Provenance – History or pedigree of a property value.
9. Representation – The characteristic of data quality that addresses the format, pattern, legibility, and usefulness of data for its intended use.
10. Timeliness – The degree to which data is available when it is required.
11. Uniqueness  -The extent to which all distinct values of data elements appear only once.
12. Validity – The measure of how a data value conforms to its domain value set (i.e., a set of allowable values or range of values).

As Stephan Wolf explained in GFMA webinar on 2^nd April 2020, the LEI records are in very high quality but sometimes the maturity of that data is not always provided by the LOUs. Maturity levels define the evolution of improvements in processes associated with what is measured. The GLEIF is working very hard directly with LOUs like us to ensure training and new processes are put in place to increase the effectiveness of the system and LEI data records.

Despite having 1.5 million LEIs, the GLEIF report 99% data accuracy. This is still 15,000 records and Stephan Wolf was very frank about the fact that even 1% is still considered a discrepancy. Most of these 15,000 records were actually an issue with timeliness as the records were updated at a time where the data in them may have not corresponded with external business entity data. The other top issue was bad data sources. The GLEIF is working with vendors in mapping data (such as BIC data mapping) to increase data quality through sharing data with those vendors.

Finally, the GLEIF do a lot of testing. They started to look at address information and sourced a global postal data file. They have crowdsourcing events where thousands of people look at the records to find discrepancies. This is a huge undertaking but very important in keeping those data records as accurate as possible. All of these activities resulted in 60,000 data challenges in 2019.

Addresses, as an example, are diverse globally so it’s not easy to bring all this data together in one database whilst also being consistent in how that data is displayed. The GLEIF are looking to compare address information with the international postal standard and using website scraping to determine if the street postal code and company match together as well as geocoding to measure distance between addresses.

They continually run programs like this for other data fields and the next is on the names of the legal entities.

LEI Conformity Flags

As mentioned before, LEI records are not always consistent. As the LEI ROC issues policies, jurisdictions create regulations to meet those policies but not all of these are consistent. One example is Canada where counterparties must report in an annual renewal fashion while in other countries, renewal is only mandated for financial institutions and a few countries where renewal is not mandated at all.

To address the discrepancy, the GLEIF put together 5 criteria to find out if an LEI record is conformant or complaint with the ROC policy. This gives people an additional indicator of trust for the LEI records and the GLEIF hope it will have an impact on registrants who will have an additional reason to update their data.

The implementation and testing of the LEI conformity flag is complicated but prototypes are currently in testing with the aim to go live at the end of 2021. More news on this soon we hope!