Anti-Money Laundering Model Risk Management

The 2020 AML Act encourages financial institutions to explore innovative solutions for suspicious activity monitoring. Many institutions either are already or are considering automated solutions to aid in this process. However, in spite of what a salesperson may tell you, these systems are not “plug and play.” Guidance from regulators make it clear that financial institutions are expected to evaluate monitoring parameters, risk rating methodologies, and OFAC filtering models to ensure that they are calibrated based on your unique risk profile. Recently, a community bank was fined $8 million for missing suspicious activity in part due to misuse of its automated system. In this session we will cover the requirements of developing and maintaining a suspicious activity monitoring model.

As financial crimes grow in complexity, the systems needed to monitor, identify, and report suspicious activity must keep pace. The key to a compliance BSA Program is developing adequate policies, procedures, and processes for model risk management, periodic testing of the system to ensure that data feeds are working, and evaluation of the model’s performance to ensure outcomes are generated as expected.

When considering which vendor to utilize, financial institutions should carefully evaluate the system’s capabilities and compare those to the institution’s needs. System limitations may necessitate that some customers and transactions are still monitored manually or require separate monitoring systems/ The process for selecting the right vendor for your institution is just as important as managing the software once it is implemented.

The most common frustration with model risk management is the volume of alerts systems can generate. Proper evaluation and calibration of parameters is necessary to ensure that you are not missing suspicious activity because the model is too narrowly defined, but also that you mitigate the risk of missing suspicious activity because the model is so broad that you cannot keep up with the volume of false alerts.

In this two-hour session, Brian will cover:

• Model Development
• Model Governance
• Model Risk Assessment
• Preparing a gap analysis
• Data Integrity Analysis
• Parameter testing
• Model Validation

Written materials will be provided for the listener to refer to during the presentation.

WHO SHOULD ATTEND:

BSA Officer, BSA Analysts, BSA Auditors