Regardless of economic climate, risk management remains an important factor for lenders to consider during the approval process. Its importance in the current climate though, is greater than ever.
Luckily, there are advanced credit risk management technologies and services that banks, credit unions, and fincos can employ to obtain a better understanding of an applicant’s financial position and boost approval rates without increasing risk.
For many lenders, big data, artificial intelligence (AI), and machine learning remain untapped resources. However, some of the savviest lenders are realizing that the wealth of new consumer data coupled with the capabilities of AI and machine learning when applied to their scorecards can deliver substantial, quantifiable benefits, particularly in the area of risk management.
Credit Risk Management Technology and Trended Credit Data
While a traditional credit score condenses a lot of consumer information into a single number, it doesn’t always provide the most accurate and current assessment of a borrower’s financial standing. Lenders are beginning to realize that additional relevant consumer data can better equip them to assess credit risk.
Trended credit data provides as much as a 24-month window into a borrower’s credit card charges and payment history. Analyzing the trends over a 12 to 24-month period can reveal an applicant’s current financial position and its effect on perceived creditworthiness. In light of the challenges that COVID-19 presents in assessing an applicant’s current financial strength, trended credit data may be one of the more practical credit risk management technologies to employ. To better illustrate how trended credit data can benefit lenders, read through the three scenarios below:
|
Charges and Payments Trend |
Effect on Creditworthiness |
| Improving: Increasing monthly charges with full payments made each month. | Offer applicant better terms than would be indicated by traditional credit score alone. |
| Unchanged: Consistent monthly charges with full payments made each month. | Offer applicant terms based on traditional credit score. |
| Declining: Increasing debt with missed or partial payments each month. | Depending on credit score, offer applicant risk-adjusted terms or issue adverse action. |
Lenders that incorporate trended credit data into their credit risk management methodology can boost approval rates by offering more competitive terms for applicants with good credit scores and credit payments that demonstrate improving financial strength. At the same time, trended credit data allows lenders to decline applicants whose credit score and credit trends together indicate unacceptable risk.
Credit Risk Management Technology and Machine Learning
With increasingly digital transactions across all channels of commerce, new sources of data provide valuable insight into a consumer’s financial behavior. Bureau scores, utility payments, checking accounts, employment records, income data, and other alternative data sources are being used to better assess applicants’ creditworthiness. Although these data are readily available, the biggest challenge is determining how these data can be used to better assess creditworthiness and predict loan or lease performance.
Machine learning is addressing this challenge, allowing lenders to analyze their application and portfolio data to better understand the data relationships and their impact on portfolio performance. The power of machine learning identifies patterns in these data that would otherwise be overlooked by traditional statistical methods.
Based on the analysis, machine learning then builds a custom credit model, providing a more nuanced picture of default risk that can be used to evaluate loan or lease applications. A more accurate model of credit risk gives lenders greater confidence in all lending decisions. It allows lenders to identify borrowers who may have been overlooked using traditional credit models, particularly consumers in the near-prime or subprime segments.
At the same time, the machine learning credit models better identify risky borrowers who previously would have been granted credit. The use of machine learning credit models has typically resulted in a 15% increase in loan or lease approval rates without taking on any additional risk.
Don’t Pursue Lending Efficiency Without Improving Risk Management
There’s no denying that technology continues to improve the efficiency of nearly every aspect of the lending process, from mobile loan or lease originations to servicing. Equally important in lending efficiency is credit risk management. No amount of increased efficiency can compensate for credit decisions that result in defaults. Lenders employing technologies such as workflow, decision rules, and auto-structuring to increase the efficiency of their lending processes should also invest in credit risk management technology that boosts approvals without increasing risk.
Getting Started
defi SOLUTIONS provides configurable loan or lease originations systems pre-integrated with advanced credit risk management technology, including trended credit data and machine learning credit models. If you’re struggling with the limitations of your current lending technology solutions, take the first step in realizing the benefits of modern technology. Contact our team today or register for a demo.
