For example, the Office of Financial Research (OFR) was established under the Dodd-Frank Act to provide policymakers with more comprehensive financial data and to better assess the risks to the U.S. financial system. To meet its charter, the OFR may request specific datasets, such as transactions, contracts and legal entities' data.
To handle these requests promptly, banks will need to quickly identify the data requested, isolate it from other information, and consolidate it in a format accepted by regulators. Banks could first consider aggregating their data repositories into centralized data warehouses that integrate data most likely requested by the OFR. Following a process of data cleansing and filtering, banks can use analytics to isolate highly specific information within millions of records. To help enhance the data mining process, centralized data warehouses could categorize transactions by legal entity to address requests by regulators looking for entity specifics. They may also contain digitized and indexed paper-based contracts to help banks speed their ability to provide additional documentation.
Due to the complexity of working with unstructured data, banks may consider using text analytics to gain insights from counterparty contracts and other text-based datasets. For example, term association and clustering can help improve search accuracy by grouping words with similar meaning and categorizing keywords relating to a specific topic. This can be particularly helpful when regulators have some existing information that leads them to request further details. As banks become familiar with the type of information requested by the OFR, they can build custom applications that are specifically designed to automate this process. Such applications can offer dashboard interfaces to help legal specialists better manage the lifecycle of each request and can summarize results in pre-formatted reports based on regulators' needs.
The scenarios described in this article provide just a few of the many ways that executives can leverage data analytics to tackle the compliance challenges of the Dodd-Frank Act. In fact, banks can also use advanced analytics to enhance continuous monitoring and proactively seek out evidence of fraudulent activities before regulators or internal whistleblowers detect and report them. For example, predictive analytics can help improve fraud detection accuracy based on historical trends and risk assessments, reducing the number of "false positives" for review by compliance professionals. Since many of these challenges involve a range of functional departments (see infographic below), banks might consider defining and implementing an enterprise-wide analytics strategy. Perhaps most significantly, a thoroughly integrated analytics strategy can help banks turn the ever-growing volumes of data generated by their businesses into assets that help them efficiently respond to regulators, better measure risk and uncover violations across the bank.
Dan Krittman is a director in the Analytic and Forensic Technology practice and the data analytics national leader at Deloitte Financial Advisory Services. Mike McCabe is a director and Mohamad Said is a manager in the Analytic and Forensic Technology practice at Deloitte Financial Advisory Services.