Ghostbusters: How investigators used data to catch a ‘ghost’

Robin Wagner, Vice President: International Insurance at TransUnion.

A ‘Ghost’ wreaked havoc in South Africa’s insurance industry for eight years by hitting different insurers with fraudulent claims worth over R2 million. Ghost was the codename used by the South African Insurance Crime Bureau (SAICB) in their investigation into this fraudster, who mostly abused fast track and vehicle insurance claims.

However, thanks to the Insurance Data System (IDS) – an information-sharing platform – investigators were able to detect Ghost’s network of fraudulent claims through industry data and track him down. He is facing an ongoing criminal case following his arrest in March 2015.

IDS was created by risk and information solution provider TransUnion, under the guidance of the South African Insurance Association (SAIA).

“Fraudsters such as Ghost often act over time and across multiple channels and insurers, which makes their scheme so complex that it often can’t be detected by a single insurance company,” said Robin Wagner, Vice President of International Insurance at TransUnion.

“When data shared by a group of insurers is centralised on IDS, however, there is a greater likelihood of being able to detect these patterns and use this information in subsequent criminal investigations.”

Garth de Klerk, CEO of SAICB, said that his organisation utilises data from the IDS in their criminal investigations once it has been correctly stored, cleaned up and interrogated by TransUnion to meet certain quality standards. This information is then run past several databases, including the South African Police Service’s wanted list and TransUnion’s Auto Information Services Bureau to connect cases and detect patterns.

“Currently we have over 55 ongoing insurance fraud related investigations, worth in excess of R250 million in claims, and this indicates why we are dependent on receiving data of the highest standard. Without a partner like TransUnion, which is equipped to handle these volumes of data through their technical expertise, it would be difficult to gather the evidence needed to combat large-scale insurance fraud,” de Klerk said.

He said another example of the success of IDS is highlighted in another SAICB investigation codenamed ‘Android’. Android was a criminal network of 80 people who would fraudulently claim on lost cellphones, moving tactically from different broker insurance to the underwriting manager market and across providers to avoid suspicion.

“One of the strengths of the IDS system was that we were able to identify the ID numbers attached to the claims and cross reference these to bank accounts used in order to link the suspects and close the net on them,” said de Klerk.

Wagner said it is not only about the quantity of data that insurers supply to IDS, but also ensuring the quality of data is up to standard. “Insurers can play their part in further improving the efficacy of IDS in insurance fraud investigations by ensuring the data they supply to us is of the highest standard. Ensuring all information is captured and cross-checked accurately throughout all customer interactions and claims is the first step in building a reliable database.”

Wagner concluded, “IDS is a prime example of information being used for good. We are confident the System’s role in combatting insurance fraud will only grow as insurers see the value it provides in criminal investigations.”