Many banks are wasting the loans data they capture, according to David O’Connell, Senior Analyst, Aite Group, a financial services consulting group. This posting summarizes the second half of his webinar organized by the Global Association of Risk Professionals on February 20, 2014.
O’Connell contrasted marketing teams with underwriting teams. Marketing teams use predictive analytics to decide which customers are most likely to respond to certain campaigns. They are very forward-thinking in devising the “customer next best action,” he said.
O’Connell encouraged the credit and underwriting teams to have a similar outlook—to also make use of predictive analytics to determine “borrower next best action.”
Underwriting Survey Question. Why doesn’t your bank use predictive analytics to monitor loan portfolio quality or identify problem assets? The obstacles most commonly cited were lack of internal skills to build predictive models, and staff to deploy the models.
O’Connell said the lack of predictive analytics in underwriting was a “missed opportunity.”
Q. What would convince senior management at your bank to invest in predictive analytics? The majority agreed the most convincing argument was improved ability to identify deteriorating loans (either groups or individual) in time to take corrective action.
“Banks tend to apply underwriting to do home-grown due diligence and then go to credit agencies,” he said, noting that very few banks examine the proprietary data any further. This is a waste. Ignoring their own data can lead to delays in diagnosing loan deterioration. Delays lead to settling for a steeper discount from defaulting customers. “The principals by that time are running around with their hair on fire” he said in one memorable metaphor.
O’Connell has spoken to clients about a host of points connected to commercial credit analytics. He examined what capabilities are available in the IBM Algorithmics Credit Lifecycle Management package, and looked at how these would help individual banks.
Underwriting and transaction analysis is a given. The risk governance that pertains to a deal can be easily tied in to the data already collected. Aggregated views of loans (for example, by geography or industry sector) are easy to prepare. Management of post-approval loans, terms, and collateral is an important feature because some items might not get followed up on in timely fashion. “Structuring a deal is one thing… keeping the structure as intended is another,” said O’Connell.
Overall, there was better loan conformance, and improved credit-event management. O’Connell referred to a “system-of-record effect,” noting that there was enthusiastic response from the C-suite. “Senior folks liked having a central hub … if it wasn’t in the Credit Lifecycle Management system, it didn’t happen.” ª
Click here to go to Part 1, Under-Utilized Tools in Commercial Credit Analytics. ª
Click here to go to webinar presentation slides on Commercial Credit Analytics.
The Twitter feed for David O’Connell is at: https://twitter.com/davidpoconnell