NEW YORK CITY — Money laundering is a team sport. “The process of cleaning dirty money involves moving funds through an intricate and interconnected series of accounts,” said Katie Levans, marketing director at Tresata, a predictive analytics software vendor.
It’s a team sport that’s paying off — for criminals, at least. Globally, laundered transactions total more than $2 trillion a year, Levans said. And, in the United States alone, billions are spent each year to curtail those activities — investments that have, for the most part, proved rather fruitless. “When it comes to actual seized money from successful AML [anti-money laundering] convictions, that total is less than 0.2% of all global laundered transactions,” Levans said at Spark Summit East, an event named after the open source big data processing engine that came out of the University of California at Berkeley’s AMPLab.
The current anti-money laundering reality has led some financial intuitions to make decisions that meet government regulations but create unintended — even detrimental — consequences, Levans said. Earlier this year, for example, the Merchants Bank of California was pressured by the U.S. government to stop wire transfers to Somalia, a country that’s been known to funnel funds into the hands of what the United States sees as terrorist organizations.
But the all-or-nothing approach created new problems. By cutting off all wire transfers to Somalia, people who depend on funds coming from relatives in the United States were also cut off. “Somalia is one of the poorest countries in the world, relying heavily on remittances for schooling, food, housing and other humanitarian aid,” Levans said. (It should be noted, Merchants Bank is not alone. Wells Fargo and US Bancorp stopped wire transfers to Somalia years ago, and other banks followed suit.)
Merchants Bank made the decision after it determined it was too hard to document who was actually receiving the wire transfer, according to a report by the LA Times. A lack of visibility also happens to be a problem for even the more advanced money laundering schemes, according to Levans. The investigations are highly manual, which also means they’re slow, expensive and sometimes inaccurate; technology products on the market today only analyze at an entity level (an individual or business) or a transaction level, failing to provide visibility into the relationships that create these intricate money laundering networks.
Tresata is attempting to change that. Partnering with Databricks, a company founded by the creators of Apache Spark, Tresata is rolling out a new tool called Teak, an anti-money laundering application that uses Spark for data processing and is delivered in the Databricks Cloud. The goal is to help financial institutions drill into relationships between entities and networks, giving banks a more comprehensive view of how people, businesses and the transactions between them are linked.
According to Tresata CTO Koert Kuipers, Spark was selected because it helps deliver key characteristics needed to detect patterns of fraud at a network level: Spark’s in-memory capability provides speed; it provides search engine capability that processes all data on the graph; and it’s scalable. Most importantly, Kuipers said, it provides “graph traversal,” which helps users quickly explore how entities are connected.