Asset 1
Investment Banks Can Expect More and Bigger AML Fines until KYC is Fixed

Regulators have lost patience with institutions that fail to confirm customer identities

Banks have argued hard against plans, drawn up in the wake of the financial crisis, to divide their investment and retail operations. A recent $600 million fine for anti-money laundering failures may give them pause. This time Deutsche Bank was in the regulators cross hairs, following a serious compliance breach that facilitated a $10 billion Russian money-laundering scheme.

Investment banks are not bound by the same strict know-your-customer or KYC rules that require retail banks to perform checks on customer identities and intentions. However, funds that are traded must first be deposited in accounts that are subject to KYC checks. Despite the misconduct taking place in Deutsche Bank’s investment arm, the fines levied by American and British authorities demonstrate that liability for KYC failures extends across a bank’s entire operations.

Too big to ignore

The investigation into the misconduct revealed that a group of bank executives arranged a “mirror trading” scheme for wealthy Russians. This enabled them to order stock trades which, in the words of Maria T. Vullo of the New York State Department of Financial Services, “lacked economic purpose and could have been used to facilitate money laundering or enable other illicit conduct.” The transgression suggests a lack of concern in investment banking about why some customers might wish to use spurious trading techniques, especially when they provide a means to move money internationally.

Unlike the billions in fines levied in the wake of the financial crisis, these latest penalties are unlikely to be an existential concern for Deutsche Bank. However, $600 million is not a figure that can be brushed off lightly. In the past, banks charged with anti-money laundering failures faced the equivalent of a gentle slap on the wrist. Now, regulators have signaled that the punishment will fit the crime.

Pressure to get tough has intensified following the Panama Papers exposure, which pointed to crimes such as terrorism, tax evasion, and child prostitution as sources of illegal funds. The patience of legislators and the public has run out. Regulators appreciate that inherent weaknesses in KYC processes are not the fault of banks, but big fines suggest that it is the banking industry which is on the hook to find a solution.

The KYC conundrum

Traditional KYC has repeatedly proven to be inadequate because it is difficult and time consuming. Banks face a massive task of analyzing and cross-comparing huge and multiple sources of internal and external data. Insights gleaned from social media, news media, electronic communications, credit and criminal records, and other consumer and public records can help to uncover hidden identities, relationships, and intent. However, legacy analytics technologies cannot make sense of human communications (which makes up the bulk of data) or easily resolve what is discovered to profiles about specific customers.

Banks have compensated with intensive processes that score customers for risk and task knowledge workers with making manual checks on those deemed to be of most concern. The process is slow, expensive, and overlooks those who have the skills to disguise their identity and intent.

Using AI to join the dots

This KYC problem is common to all major banks, but it is precisely the kind that cognitive analytics, a form of artificial intelligence (AI), is good at solving. Solutions that use AI have been widely adopted by banks to monitor employee communications for noncompliant or illegal behavior, such as insider trading or market fixing. Now, some banks are applying the same approach to KYC.

Cognitive analytics holistically analyzes multiple data sources, including human communications, semantically and in context. Able to capture deeper insights, findings are accumulated into individual customer profiles that automatically join the dots to disclose details that would not otherwise be apparent: concealed identities, criminal involvements, and hidden networks and relationships. Greater automation means every customer can be assessed to the same exacting degree, both at the time of account opening and on an ongoing basis. Human resources are freed to perform more thorough investigations on customers where evidence has raised a concern.

The threat of more punitive fines has changed the calculation about what can and should be done to prevent money launderers exploiting systemic weaknesses. Expanding the use of AI to strengthen KYC provides banks with stronger defenses against both money launderers and any employees who might be willing to facilitate their activities. Regulators know this and they will be looking to major banks to react accordingly.