There is a range of future applications of behavioral analysis in banking risk management, said Clifford Rossi, Professor of Finance at University of Maryland. Its true potential has yet to be tapped.
As an example, he described how hehavioral analysis can be deployed in credit risk assessment of non-traditional borrowers. The borrower willingness-to-repay preference used to always prioritize the family home. Since the 2007 financial crisis, there’s been a shift in mortgage borrower attitudes. In some areas, borrowers prioritize their car. The change in this sentiment can be detected using behavioral analysis.
“It was virtually impossible using the proxy variable of FICO scores to accurately measure this critical behavioral shift,” Rossi noted. There is a new paradigm for mortgage borrower willingness-to-repay.
There are also applications in operational risk management. Rossi noted that “use of bank intranet, e-mail and other communications, as well as tracking trade ticket information and patterns could improve detection of internal fraud events such as rogue trading.”
Reputational risk is very hard to measure, but behavioral analysis permits sentiment analysis through data collected on social media platforms such as Facebook and Twitter.
Blind Spots
Privacy concerns regarding use of unstructured digital media are growing. Also, machine learning is difficult to track explicitly and thus does not lend itself to the transparency needed to ensure fair lending practices.
Joe Mattey, Vice President of Enterprise Risk Analytics, and Chief Risk Officer at USAA, said that financial institutions must make sure practices are fair and defensible. “Companies must invest in the logistics.” Upper management will require a change of mindset.
Logistics will continue to be a concern. “Data management was not formerly seen as a core competency, but now it is because the effort to identify, translate, and manage unstructured data is significant.”
Resource intensity is growing. “IT systems and storage requirements are daunting,” he said. With larger datasets involving cloud providers and cloud platforms, new security concerns will arise.
Joe Mattey and Cliff Rossi were two panellists speaking about how banks can leverage behavioral analytics. The webcast on March 13, 2018, was organized by the Global Association of Risk Professionals (GARP). ª