Efficient Screening for Politically Exposed Persons

Screening for Politically Exposed Persons (PEPs) can be even more complex than screening for sanctioned names. At least with sanctions, there are formal published lists of names. With PEPs, it is a case of mining the internet for dozens of different sources or relying on list providers to keep their own lists up to date. As a rule, sanctioned names tend to be the less common names – at least compared to a European organisation’s typical customer base. If you look at lists of PEP names, you can see an enormous variety of common – and some more unorthodox – names from every country and language. ​

Understandably, PEP screening can result in huge numbers of false positives. ​

The table below shows the most popular new baby names for girls and boys from a range of European countries and a few selected others, combined with the most common family name for that country*. ​

2020 04 28 09 25 38 Efficient Pep Screening.pptx And 6 More Pages Work Microsoft​ Edge

We ran a very rudimentary exact match lookup – for the exact string of characters in the given name followed by the family name – against a PEP list that we at SQA Consulting use in our testing of screening systems.

Even with this most basic matching technique, there were matches on 16 out of the 30 names. That was 9 out of 15 of the male names and 7 of the 15 female names. The very high number of hits for Maria Fernandez and Maria Silva is due to both a large number of PEP names in the list for Brazil and Argentina and the sheer number of Marias in those countries and elsewhere. A similar result would be observed if this list were expanded to include the most common names in Asian countries such as China and South Korea. ​

More sophisticated screening engines should be able to filter out of some of these matches, but you can imagine the number of false positives that will be created for any organisation with a demographically representative customer base – particularly if fuzzy matching techniques are employed.

This goes to show the importance of tuning your screening engine to try and hit on genuine PEPs and minimise false positives. ​More and more organisations we work with are incorporating secondary attributes in their screening such as country or date of birth.

Another option is to make sure that your PEP list criteria are reviewed and are appropriate for your organisation’s risk appetite. An example of this is to look at the type of roles that your organisation considers to represent a risk to your business. Your list provider should be able to provide you with a list of categories. ​

In addition to this several governments around the world require the identification of specific types of domestic PEPs over and above the more widely accepted definition of PEP. This often includes lower-level individuals and can dramatically increase the number of PEPs for a country. For example, Spain’s definition of domestic PEPs includes mayors of towns with a population of more than 50,000 and Brazil’s includes mayors from all cities, councillors, state representatives, national presidents, and treasurers, or the equivalent of political parties, presidents of courts and councils of municipal auditors, amongst others.​

This map shows the countries that have a higher proportion of additional domestic PEPs. ​

2020 04 28 09 24 42 Efficient Pep Screening.pptx And 6 More Pages Work Microsoft​ Edge

Depending on where in the world your business is based and who you do business with, screening for these additional PEPs may or may not be appropriate. Reducing the number of PEPs that you are screening for can have a significant effect on the resources needed to manage the alerts. ​

SQA Consulting can provide you with assistance in setting up your PEP screening. Please contact us to find out more about this article.

*The data, sourced from a range of Wikipedia pages is not up to date for some countries. New baby names will not necessarily be representative of the current adult population – although they do give an indication of the adult name trends in years to come. However, the data is good enough for the purposes of this experiment.

Get In Touch

Technology Consulting Partners