In the early 1980s, credit unions lived in a world of manual lending decisions. Members completed loan applications at their local credit union and based on that information, an underwriter would make a decision to approve or reject that loan. Credit bureau data was sometimes thrown into the decisioning mix, but at the time, bureau data was often regional at best and inaccurate at worst. By the 1990s, the lending landscape had changed dramatically and technology, bureau data, credit scoring, and centralized decisioning were key features ─ all with the goal of managing credit risk.
With lending’s growing role as a revenue generator and market expander, so too came an increasing reliance on auto-decisioning for many lenders. Thanks to its ability to boost operational efficiency and help meet customer demands for quick loan decisions, it’s easy to see why credit unions came to depend on this technology. It’s also not too hard to see how auto-decisioning – when not used properly – can expose a credit union to unexpected risk and severely degrade the institution’s portfolio quality.
However, credit unions should not let their own negative past experiences or the negative experiences of peers dissuade them from utilizing a useful tool that will help their institutions remain competitive and grow. Auto-decisioning wasn’t the root of the problems financial institutions faced in their loan portfolios at the onset of the recent financial crisis. The real enemy was that no one was minding the store. Auto-decisioning was put into place, but not enough tracking, measuring, or monitoring was done to ensure that it was working appropriately. So when consumer behavior changed, risk ran rampant.
When executed properly, auto-decisioning offers credit unions some big benefits, especially when it comes to increasing efficiency. Auto-decisioning can greatly improve time to decision, enhance member relationships, lower underwriting costs, allow lending staff to focus their time and resources on more difficult decisions, and increase approval, look to book, and capture rates.
Perhaps surprisingly, one of the biggest benefits of auto-decisioning is also one of its biggest ironies – that automated decision making can actually help credit unions mitigate the very same risks they fear it would make them vulnerable to. The NCUA identifies seven categories of risk: credit risk, interest rate risk, liquidity risk, transaction (or operational) risk, compliance risk, strategic risk, and reputation risk. To further complicate things, these categories are not mutually exclusive and they are codependent. Auto-decisioning can aid a credit union significantly in identifying, measuring, monitoring and controlling:
Credit Risk – By minimizing the human element in the lending process, auto-decisioning mitigates human error while increasing consistency, thus lessening a credit union’s exposure to credit risk. It also enables more effective tracking and monitoring, and gives credit unions faster responses with the right offers to help compete and capture profitable applicants.
Compliance Risk – Auto-decisioning allows credit unions to build an established risk management business process that provides fair treatment of consumers, which is a key part of mitigating compliance risk.
Operational Risk – By reducing human intervention, auto-decisioning reduces errors and allows measurement and monitoring to validate that all models and scores are implemented and used the way they were intended. In addition to mitigating risk, reduced human interaction also reduces the per loan application cost.
Strategic Risk – Auto-decisioning can help credit unions demonstrate that underwriting decisions are in alignment with their enterprise risk policy and appetite, thus allowing for consistent deployment of their strategic vision and direction for the future.
While properly managed, auto-decisioning is clearly beneficial. But there’s also the need for caution: Automated decisioning does not equal better decisions. To arrive at better lending decisions, a credit union also needs to utilize analytics.
An integrated lending analytics solution allows for examination and alignment of a credit union’s entire underwriting policy including generic scores, custom scores, and cutoffs. It also addresses policy rules and data, including application, credit bureau, and any alternative data. This gives institutions a thorough understanding of what is and isn’t working in their policy, as well as probable outcomes that may result from policy changes. And it also provides the information needed to gain organizational buy-in on these changes, from the underwriter level all the way up to the board.
Even when employing the insights of analytics, implementing auto-decisioning is often met with some trepidation by credit unions. Learn how one institution overcame those anxieties and embraced new opportunities with auto-decisioning – download the case study now.
With over 35 years of combined experience in decision analytics, both domestically and abroad, Kathy Kieffer and Kristina Corts use their analytics expertise to serve the clients of CRIF Achieve, the decision management division of CRIF Lending Solutions. To learn more about CRIF Lending Solutions’ automated lending solutions, visit www.criflendingsolutions.com.