Understanding LEI Names

A legal entity has an official legal name, and it may also be identified by other names such as a trade name. This is why the LEI common data format requires every LEI to have a legal name and supports other name types. You can use this name data to find the LEI of a party, such as a client, or as a data source for in-house databases. This post provides some tips for using LEI name data.

Legal Name

The legal name is the primary and often only name associated with an LEI. It is a required field when applying for an LEI at an issuer (LOU). If an entity’s legal name changes after its LEI is issued, the registrant is required to update the LEI’s reference data at its LOU to reflect the change.

An LEI’s legal name can be long, up to 500 characters in length. Currently, the longest legal name in the LEI data is 356 characters. However, out of the 1.1+ million LEIs issued, only about 18,000 require more than 80 characters for their legal names.

An LEI’s legal name may contain characters your system isn’t expecting. Legal names, as well as other data elements such as address, appear in the script of the jurisdiction in which the LEI is registered. For example, a Russian legal name will appear in Cyrillic (e.g. Общество с ограниченной ответственностью “АЛЬСТОМ”) and Japanese legal names appear in Kanji (e.g. 明治安田生命保険相互会社).

Unicode

The use of different scripts makes LEI data a bit more complex to process than other reference data.

Unicode is a data representation standard that supports all the world’s writing systems. LEI data uses a particular Unicode character encoding known as UTF-8. Even though it’s popular for web pages, in-house systems do not generally support the full range of characters in UTF-8.

Instead, in-house systems typically expect ASCII, a small subset of Unicode consisting largely of the characters that appear on keyboards in the English-speaking world. Names that use non-ASCII characters like the Russian and Japanese examples above will not display intelligibly in an ASCII-based system.[1]

Transliterated Legal Name

Of course, it is possible to transliterate a word from a non-ASCII character set into ASCII. For example, the Greek word “λογοσ” transcribes to “logos”.

When a legal name can’t be rendered in ASCII, the LEI data contains a transliterated version of the legal name as an additional data field. We can use this additional data to determine that the aforementioned Russian entity written in Cyrillic is “Alstom Limited” and the Japanese entity is “Meiji Yasuda Life Insurance Company”.

There are over 135,000 transliterated legal names available in the LEI data.[2] So, over 11% of LEIs have a transliterated version of their legal name available. Some were provided directly by the registrant and some were auto-transliterated by the managing LOU. Entities domiciled in the US, UK and other jurisdictions where the legal name can be fully represented in ASCII don’t have a transliterated legal name field.

If your system or its users cannot process the legal name in its native character set, consider using the transliterated version if available.

Other Names

Beyond the current legal name and its transliterated version, other LEI names fall into one of three categories:

Trade or Operating Name

A trade, brand or operating name currently used by the entity in addition to, but not replacing, its official legal name. This type of name can optionally be provided when initially submitting or updating an LEI’s reference data.

There are currently over 21,000 trade or operating names in the LEI data, which can be useful when looking up an entity’s LEI by name. You may have an entity name “PIA Wealth Management” in your system while its legal name is “Professional Independent Advisers Limited”. If you searched only legal names, you wouldn’t find a match, but you would by including trade and operating names.

Previous Legal Name

A primary legal name previously used by an entity. As mentioned above, the legal name field must be updated at the LEI’s managing LOU when a legal name change occurs. The old legal name is retained in the LEI data as a previous legal name.

There are close to 70,000 previous legal names recorded in the LEI data. As a group, funds appear to change names with greater frequency than other types of entities.

You can ensure that you have the most current legal names for your clients, counterparties, and issuers by monitoring the daily updates to the LEI data for relevant changes.

Alternative Language Legal Name

Registered name of an entity in an alternative language in the legal jurisdiction in which the entity is registered.

There are over 18,000 alternative language names in the LEI database. Some of these entities are domiciled in jurisdictions such as Canada or Hong Kong that have more than one official language, but the majority are ASCII transliterations of the legal name for entities based in Japan, Finland, Russia and the Czech Republic.

They are not tagged as transliterations due to historical reasons: the ability to tag a name as an alternative language legal name predates the ability to tag a name as a transliterated name by a couple of years.

Related Entities

While this post is about the various names associated with an LEI, we would be remiss if we didn’t mention that LEI data also contains information about the relationship between LEIs. It may reference legal entities that are its ultimate or direct parent, its fund manager or an international branch.

These related entities have their own LEIs with, of course, their own set of names. You can learn more about LEI relationship data here.

Conclusion

The LEI data is a rich source of entity name information. Beyond the required legal name, an LEI may be associated with a trade or operating name, a previous legal name or a transliterated ASCII version of the legal name.

Knowledge of these different entity name types and their nuances is key to effective LEI search as well as making certain your LEI mappings and legal name data are current and correct.

[1] Even if a system supports a larger character set than ASCII, it’s still unlikely to support all the scripts used in the LEI data.

[2] As of early May 2018. All counts in this post are from this time period.