The backlog of money-laundering cases at Germany’s Financial Intelligence Unit stood at 30,000 at the start of February. Every month, another 5,600 cases on average are added to the pile. It probably didn’t help that, until November of last year, banks and insurance companies were submitting suspicious cases to investigators by fax.
Nor does the FIU have automatic digital access to police data and other crime authorities. Information on things like visas and bank accounts has to be requested by phone or submitted in writing. It’s the same thing the other way around – police don’t have access to any of the FIU’s findings.
It’s a problem that has been swept under the carpet for years, according to Lisa Paus, a financial policy spokesperson for the opposition Green Party. She calls the backlog a “ticking time bomb” that should be top of the priority list for incoming Finance Minister Olaf Scholz. Outside events may actually force Mr. Scholz’s hand: The closure of three Baltic banks over money-laundering cases this year has put the entire European banking system in the spotlight.
Banks have partly faulty data in their systems, which in turn sets off false alarms. Jan-Alexander Huber, Bain & Company
The sordid state of the FIU, which has built up since last summer, was detailed in a rather frank report by Michael Meister, a deputy finance minister. Technological upgrades, already in the works for a year, are “still being implemented” and can be expected by the end of 2018, he wrote. But it’s not just the government that has fallen behind. Banks themselves could sorely use an upgrade of their technology, too.
It isn’t for a lack of manpower. Though about one-third of the financial crimes agency’s 165 permanent positions are unfilled, some 180 people have been drafted in to handle the backlog. German banks, frequent targets of regulators, have also massively stocked up their compliance departments over the past few years since the financial crisis. Deutsche Bank, for example, has doubled the size of its compliance division since 2015.
The problem isn’t really a lack of money, either. German banks are spending about $46 billion (€38 billion) per year to combat money laundering, according to a survey of 51 banks by consultants Lexis-Nexis Risk Solutions. Overall spending has jumped about 20 percent each of the last two years and will climb another 20 percent by the end of 2019, according to Seyfi Gulney of Lexis-Nexis’ financial crimes department.
Some point the figure at regulation. Banks complain that costs are rising in part because the rules keep changing in each country. The fourth set of guidelines in Germany was adopted in June 2017. Since December there’s been talk of another updated set at the EU level.
The biggest challenge, most experts agree, is a lack of efficiency. Like the German government’s supervisory agency, many banks are dealing have a surplus of people working with faulty or outdated technology. “The core problem that banks have in preventing fraud is that they have partly faulty data in their systems, which in turn sets off false alarms,” said Jan-Alexander Huber of the consulting firm Bain & Company, which conducted a study of five major global banks. Consumer data is often incomplete and – again like the FIU – financial crimes often have to be filtered out manually rather than flagged by an automatic system.
That’s part of the reason banks are putting their money in artificial intelligence. The hope is that smarter computers will be able to automatically flag suspicious clients or transactions. For example, a smart computer might flag if a customer based in Lübeck is trying to open an account in Switzerland, says Mr. Günay of Lexis-Nexis.
Philippe Vollot, who heads up Deutsche Bank’s financial crimes division, says using robots and artificial intelligence could be a “medium-term solution” to confronting financial crime. The idea would be to look for anomalies among clients and transactions “to actively detect concrete dangers and recognize patterns more effectively,” he wrote in a guest op-ed for Handelsblatt.
John Cryan, Deutsche Bank’s chief executive, also highlighted the problem in a recent speech at the SXSW startup conference in Austin, Texas. Finding patterns is critical to fighting financial crime, and computers can identify patterns far more quickly than humans, he said.
Mr. Huber of Bain agrees. In the future, banks will still need compliance managers, “but not hundreds.” Instead, they’ll need a few specialists to monitor the work of technical systems. Now, there's a pattern that extends well beyond the banking world.
Jan Keuchel is a financial editor for Handelsblatt in Düsseldorf. Yasmin Osman is a senior financial correspondent for Handelsblatt in Frankfurt. Christopher Cermak adapted this story for Handelsblatt Global. To contact the authors: [email protected] and [email protected]