Credit risk assessment plays a crucial role in lending as it directly influences the success and profitability of a loan portfolio. While traditional credit scores have been the standard method for evaluating a borrower's creditworthiness, bank transaction-based credit risk underwriting is revolutionizing the industry by revealing new insights into a borrower's ability and willingness to repay a debt.

By leveraging the additional risk-splitting power of cash flow data, underwriters gain a new perspective that complements traditional credit scoring models. This perspective is particularly important for over one-third of the U.S. population, who either have a very limited credit file or no file at all.

In this blog post, we explore the power of bank transaction-based underwriting and focus on strategies to maximize the impact of this data. Additionally, we highlight how EdgeScore, the industry's leading predictive risk score rooted in bank transaction and loan outcome data at scale, excels in accurately modeling risk across the credit spectrum.

‍The Importance of Cash Flow Data

‍Traditional credit scores rely on static information such as payment history and outstanding debt, which often overlooks the dynamic nature of a borrower's financial health and fails to accurately represent those with limited credit history or emerging credit profiles. In contrast, bank transaction-based underwriting overcomes this limitation by leveraging up to 12 months of cash flow data, providing a comprehensive and real-time view of a borrower's financial activities.

By considering cash flow data alongside credit history, lenders gain a holistic understanding of a borrower's immediate financial situation and their ability to meet loan commitments. Traditional scores can reliably predict consumer willingness to pay in the prime or super-prime space but lose predictive power as consumer risk increases. The enhanced perspective of cash flow underwriting allows for better risk splitting, enabling lenders to differentiate between borrowers where traditional methods may fall short.

Analyzing bank transactions provides crucial information on attributes such as actual income, income stability, additional indebtedness, spending habits, risky behaviors, and overall financial management. This additional data empowers lenders with valuable insights to accurately assess the borrower's repayment capabilities and improve the overall accuracy of risk assessment.

‍Limitations of Summarized Data

‍Raw transaction data is difficult to work with and its quality can vary depending on how it was sourced. In many instances it has limited use by human reviewers. Many bank data aggregators and analytic platforms attempt to summarize the data to increase usability and provide insights to lenders. Some aggregation platforms offer thousands of attributes, but unfortunately, these rarely provide the required accuracy to confidently rely on the dataset. Furthermore, extensive uncurated attribute sets may offer slight insights into a consumer’s behavior but don’t directly address the question of their level of credit risk.

Building models to understand which attributes best correlate with risk and then predicting individual consumer ability and willingness to pay with these models is hard. It requires substantial data science resources and expertise that most lenders do not have. Even larger lenders with advanced modeling teams face limitations due to the narrowness of their own datasets, resulting in blind spots outside their historical approval range.

‍The EdgeScore Advantage

‍EdgeScore stands out by crystalizing the true predictive power embedded within bank transaction data and delivering risk-predicting answers to lenders in the form of a three-digit score. Edge is a Credit Reporting Agency and maps EdgeScore to FCRA compliant adverse actionable decline reasons. EdgeScore, and the underlying EdgeEnrich attributes, are built on a rapidly expanding data lake of over four billion transactions and 400,000 loan outcomes.

Edge has the data science expertise and the broad data set to understand what attributes are most predictive and how to combine them into a predictive risk score. EdgeScore captures a comprehensive representation of credit risk across the credit spectrum and especially in emerging and underserved credit profiles. By leveraging this extensive data and Edge's analytics expertise, lenders can confidently assess creditworthiness for a diverse range of borrowers, leading to improved decision-making and reduced portfolio risk.

‍Conclusion

‍Bank transaction-based underwriting revolutionizes credit risk assessment by placing a strong emphasis on the unique perspective of cash flow data. This transformative approach incorporates real-time financial information, significantly improving the accuracy of risk prediction and facilitating better risk splitting among borrowers.

EdgeScore, leveraging a vast proprietary data lake that integrates transaction data with loan outcomes at a large scale, stands out as an exceptional tool for precise risk modeling across the credit spectrum. Embracing the use of bank transaction-based underwriting, particularly with the implementation of EdgeScore, empowers lenders to make better informed underwriting decisions, optimize their loan portfolios, and establish more robust and resilient lending practices.