SAN DIEGO – Data and analytics go hand in hand with managing risk in today’s digital-driven auto finance market, making it crucial for auto lenders to rely on the right data for their decisions, Automatic flush Risk Director Ajit Nallasaid Tuesday during a fireside chat at the Auto Finance Risk Summit.
“Having the best data is important,” Nalla said, noting that lenders should be aware of consumers’ obligations to other credit issuers with data from credit bureaus.
Internal data, however, is “more powerful than what you get from a desktop,” he said. JPMorgan Chasefor example, leverages consumer deposit data and runs algorithms to understand customer spending patterns, cash flow, income, and where a consumer transacts.
Alternative data, such as rent and utility obligations, can also paint a valuable picture, especially for consumers with little or no credit, Nalla said.
Lenders should also take note of what information they need from a consumer at a specific point in the decision and tailor borrower questions accordingly, he said. “If we randomly ask consumers a lot of questions, they ask ‘why do I have to give it away’, but if you do it more in the context of your vision, you get [what you need].”
When it comes to leveraging data for risk management, strategies need to be simple and make the most of the information, Nalla added. “You can’t have very complex strategies with thousands of rules because what happens is you don’t know where your margins are and then you end up having layered risks and concentrations and you don’t can’t rule them,” he said. “This is where taking all the data and putting it into meaningful scores that can help you rank order is important.”
Risk strategists also need to design tests in such a way that “you can learn and feed data back to the algorithms and improve,” Nalla said.
“Models are going to have blind spots; they are only as good as the data they have seen in the past,” he noted.