Auto lending has been on a positive trajectory for the last few years. Delinquencies are down. Asset values remain solid. Pent-up demand is still emerging. Used car lending is on the rise. With all that going on, it’s a good time to ask, “What could possibly go wrong?”
In other words, where might new risk lurk? Of all the things that have become both harder and more critical for financial institutions in the current environment, foremost among them is the ability to predict, assess, monitor and provision for risk. As we view the auto lending landscape, and incorporate the lessons of the recent past, it seems timely to ask, “How well do yesterday’s risk management tools serve today’s needs?”
Or, as one lender put it to us, “They tell me my job is still the same – make a lot of loans, go where the growth is. But when they add that I also need to look out for new risks in the new environment, I have to agree: our business is riskier in ways that old risk management tools won’t reveal.”
Among the new facets of risk are these:
Risk scores. One cause of high delinquencies in the past was lenders’ overreliance on the borrower’s risk score. Some lenders looked at nothing else. But as the economic crisis took over, it became apparent that many borrowers could not afford the debt they had taken on. If lenders had looked at the borrower’s income and expenses, their inability to afford the loan would have been evident, good credit score or not.
Credit worthiness. In that same vein, now that the worst of the credit crisis is behind us, borrowers often have more complicated profiles due to their experience in the recession. They may have seen their assets depleted, for example, and found other credit harder to get or possibly experienced delinquencies. They are either less credit-worthy than they were before or possibly fully recovered. Lenders need tools that can assess risk on multiple dimensions that were not part of the equation a short time ago. Fewer people are as credit-worthy as they were before but how can you tell?
Compliance risk. “Compliance risk is growing” could be the lead statement in just about any financial venture and it’s true for auto lending, too. Regardless of the merits of the new rules contemplated by the Consumer Financial Protection Bureau, the general climate is one of heightened regulatory scrutiny, an appetite for penalties vs. wrist-slaps. Along with the additional risk, this environment adds complexity to the business of auto lending.
M&A. Many post-recession mergers and acquisitions have yielded mixed portfolios that can confound the models of the acquirer, thus creating risk vulnerabilities that are not easy to spot.
Overdependence. When there aren’t a lot of bright spots in lending, which is true today, there can be a temptation to load up, over-compete, over-commit and end up with a less-than-desirable portfolio.
Overconfidence. They say generals always fight the last war, and what they mean is that’s what generals have confidence in – the hard lessons learned from the last war. But the next credit setback might not look at all like the recent credit crisis. Auto lenders need to be careful to keep looking around the corner, not just over their shoulder.
With all those factors raising the risk bar for lenders that wish to maintain the value of their current auto lending business and ride the wave without getting swamped by unexpected risk, we have four suggestions for improved risk management in this area:
Better assessment of after-the-deal portfolio risk. There’s no doubt that lenders that use analytics after the deal to assess the risk of their resulting portfolio will be better able to manage the risk they have and make better credit decisions in the future. When they drill down into the portfolio, they can determine if the risks are consistent across the portfolio. Often the best warning signs are those that show that risk is being driven by particular pockets of the population, so lenders can make better decisions about those populations.
Analytics can give lenders a full view of customers, not just their credit score and indebtedness but other critical attributes like income and expenses. Analytics can enable the financial institution to more accurately value assets and their related exposure, and employ behavioral scoring based on the borrower’s payment history and payment trends. Analytics also enable lenders to more quickly identify customers that exhibit signs of imminent delinquency so they can reach out, provide coaching and thereby reduce delinquencies and losses.
For an example of managing portfolio risk, one can look at how some lenders responded to the rising costs of collections to cover the increased delinquency volume experienced during recent periods of high economic volatility. The challenge was to increase collections without a corresponding increase in collections costs. The solution involved analyzing loan portfolios to predict which accounts were most likely to become delinquent and predict which accounts were most likely to respond favorably to low-cost collections efforts. The result was an improvement in collections with a decrease in associated collections costs. Auto loan lenders may benefit from similar portfolio analytics.
Dig deeper into certain elements of an individual borrower’s actual credit profile. One lesson learned from the sub-prime-mortgage-driven financial crisis is that lenders were overly trusting in the ratings provided by the ratings agencies. As became all too apparent, many AAA-rated securities were in fact not high quality. While there is no reason for a lack of confidence in FICO scores, heavy dependence on FICO sets the auto lending industry up for the same kind of risk.
To mitigate the risk of overdependence on FICO scores, lenders need to routinely draw on independent measures of credit risk, such as income and expense, as at least a check-and-balance on the FICO score. After all, credit scoring is predictive only in terms of historical data. In the past couple of years, almost every historical attribute pattern has changed immensely and has not returned to “normal.” Both generic and empirical data have been impacted this way, so lenders are correct in placing less confidence in them.
Steady refreshing of loan loss estimates/provisioning. The top lenders are beginning to incorporate loan loss estimation and loss provisioning into their auto lending and leasing capital management programs. They begin by building models of expected losses by levels of credit worthiness and then use analytics to factor in pricing and business management strategies. They also conduct back-testing of the models against historical data to confirm how well they actually forecasted loss rates.
Banks, in general, struggle to determine the proper amount that they need to set aside in their Allowances for Loan and Lease Losses, and the same is true for auto loan lenders. Historically, approximately 1% of all consumer loans will not be repaid and will eventually have to be written off. This 1% is often used as the basis for a bank’s loan loss reserve. However, analysis of the loans that have defaulted over the last few years shows that 80% to 90% of the defaults and eventual charge-offs involve loans to individuals whose FICO scores dropped significantly after the loan was written. Thus, a drop in FICO scores over the loan portfolio is an indicator that the bank’s loan loss reserve should be increased.
More holistic view of compliance.Compliance is not the job of the compliance officer. Today, regulators expect all practitioners to know and practice the rules – no excuses. Of course, it is impossible to anticipate and prevent all “black swan” events, but usually there are early indicators that indicate the potential for compliance problems. Those early indicators need to be identified and included in the lenders’ management program. Further, all levels of the organization need to be fully aware of the internal and external compliance requirements for risk management. Being upfront and transparent about these matters allows the lender to better understand how bad a storm they could weather.
If bankers learn from the risk management lessons of the recent past and continually watch out for what might go wrong, they might well find that things continue to go well in the auto lending business.
Mr. Mattiacci is global business team head of lending and Mr. Clemens a director in the Risk Analytics Global Business Team, at Redwood City, Calif.-based Oracle Corp. They can be reached at [email protected] and [email protected].
Financial institutions are facing new challenges to keep their customers both happy and loyal. Download the Verint Experience Index: Banking Report to learn more about the consumer trends that can help you improve and grow. Download Now...
Holly Hughes, BAI CMO, will share BAI’s latest banking channel research and host a conversation with Colleen Wilson, Vice President, Product at MANTL, on what the trends mean for financial services leaders....
Providing accurate consumer information to credit-reporting agencies can be challenging for financial services organizations due to the volume and complexity involved.
Establishing a Fair Credit Reporting Act (FCRA) center of excellence can help ensure accuracy and reduce regulatory risk. It can...
Compliance training and professional development courses that are efficient, effective and on-point. Give your people the latest industry-approved tools they need to improve performance, reduce operational risk and better serve your customers.