AML Name Screening – Not Simple
AML Name Screening mathematical equation:
Z = (spelling variations x transliteration x name structures) x 170,000,000 data entries
Sanction & PEP lists – 6,500,000 names X 20 Aliases = 170,000,000 entries
Spelling Variations & Transliterations – can go beyond hundreds of different variations!
A fundamental requirement in AML (CTF/BSA) compliance is name screening against “Black Lists” as part of the KYC process, payment transactions, and ongoing monitoring of the entire customer database. The so-called Black List are comprised of various national sanction lists, Publicly Exposed Persons (PEP), most wanted list and all in all there are about 6.5 million names with an average of 20 aliases, meaning there is a need to screen against approximately 170 million names.
The complexity in Name Screening becomes even greater when implementing the regulators guidance, demands and rulings. US Regulators FinCen guidance regarding Name Screening gives an example that an automated system must be able to match between Havana and Habana, or the requirement in understanding the structure which could be that a part of the name structure matches, raising the question is it common meaning a likely to match or not.
Today the global market brings people together from many countries speaking different languages having different to sometimes long and/or exotic names. These names must be translated into an English format and screened as required, but the translation of a name is very dependent on the origin of the person translating and the language itself. These factors cause name variations, transliterations changes, and most importantly as a rule “there is no correct way in spelling of a name”.
AML guidelines and rulings are pushing for what would be called a “Wide Net”, meaning catching all probable matches. Hence if there is a probable match, that triggers the financial institution to commence a costly Enhanced Due Diligence (EDD). But what is a probable match triggering an EDD process and what is a probable False Positive miss that can be resolved immediately at no or minimal cost. This is the dilemma: if the system does not alert probable matches that would be triggering EDD then the financial institution is noncompliant which they could be heavily fined and if the system triggers too many alerts the operational overhead, slowdown, and cost are all too great of a strain on any financial institution – as noted AML Name Screening NOT simple at all.