Expanding Lending Opportunities with Alternative Data

In consumer lending, better data often translates to better decisions. Traditional loan underwriting in the U.S. has long relied on limited snapshots of a borrower’s financial picture – credit scores, a couple of recent pay stubs or tax returns, and basic debt ratios. But what if lenders could zoom out and view a 25-year history of a borrower’s income? Access to such long-term, accurate income data can fundamentally improve risk assessment and loan performance.

Leveraging Long-Term Income Data to Lower Default Rates

A longer income history dramatically improves a lender’s ability to improve their underwriting standards. Think about it. If past performance helps you understand future results, it’s a no brainer. During the 2000s housing boom, for instance, “no income, no job” mortgages saw delinquency/default rates of about 27% – more than double the 11.5% rate of conventional loans that verified income. A long term evaluation paints a new picture on stability. Borrowers with reliable, long-term income streams are far less likely to fall behind on payments, even if their credit scores are modest. FinRegLab’s research on cash-flow underwriting confirms that income and expense metrics can predict default risk with high accuracy, achieving AUROC scores up to 0.725 in ranking borrowers’ likelihood of default. The payoff is measurable: if default rates on a portfolio drop even from 5% to 4% (a 20% reduction) due to better screening, the ROI in avoided charge-offs is substantial. According to Point Predictive, over 40% of loans that defaulted early had signs of inflated income on the application.

From Risk to Reward: Long-Term Insights Transform Loan Pricing

Long-term income data also enables more precise risk-based pricing. In traditional underwriting, lenders often rely on coarse factors (credit score bands, DTI ratios at a point in time) to set interest rates. This can lead to mispricing. Incorporating a 25-year income history changes this dynamic. Stable earners with upward income trajectories can be offered lower rates than the traditional model would allow, winning their business, whereas applicants with erratic incomes might be charged a risk premium or required to strengthen their profile. The net effect is pricing that more closely matches true default risk, boosting portfolio yield. Real-world fintech results bear this out: the AI lending platform Upstart (which built their reputation on alternative data lending) was shown to approve 27% more borrowers than a typical model while charging 16% lower average APR to those approved. In other words, by better assessing risk (including via income and employment patterns), Upstart could offer many borrowers cheaper loans and still reduce risk exposure – a win-win of more volume at lower loss rates. Regulators have recognized this benefit; an interagency statement by the Fed, OCC, and CFPB noted that using alternative data like cash-flow (income and expense streams) “may enable consumers to obtain more favorable pricing/terms based on enhanced assessments of repayment capacity” federalreserve.gov. Long-term income analytics let lenders price loans with surgical precision, increasing profitability per loan and delivering savings to worthy customers.

Optimize CECL Reserves with Accurate Income Verifications

Beyond immediate pricing and defaults, long-horizon income data can optimize how much capital a lender must hold in reserves for expected losses. Under the current CECL accounting standard (Current Expected Credit Loss), banks must estimate lifetime losses for loans upfront, using historical data and forward-looking information. If a bank can identify a segment of borrowers with stable 25-year income profiles that correspond to, say, 30% lower loss rates historically, it can justifiably project lower expected losses for that segment and reduce the allowance for credit losses on those loans. This freeing up of reserve capital improves ROI. Conversely, loans to borrowers with volatile income histories would carry higher reserves, appropriately so. By incorporating long-term income patterns (e.g. how borrowers fared through past recessions or layoffs), lenders can produce “reasonable and supportable forecasts” of credit loss that are far more tailored to each loan. For example, a prime auto loan to a customer with 20+ years continuous employment in a stable industry might justifiably receive a lower loss projection than the average prime loan – translating to a lower reserve percentage. Even a modest reduction in loss reserves (for instance, from 1.5% of loan balances to 1.3%) for a low-risk tier can significantly boost return on equity, especially in high-volume portfolios. Moreover, by broadening the data used in loss modeling, banks reduce uncertainty in their estimates.

Financial Inclusion Starts with Better Employment Verification Data

One of the most exciting ROI impacts of using 25-year income histories is the expansion of approval rates and borrower reach without compromising credit quality. Traditional underwriting often declines applicants who lack a strong credit file or have recent income instability, even if they are actually creditworthy. Long-term income data can fill these gaps and paint a fuller picture of an applicant’s true capacity to repay. Many Americans with thin credit histories (or past credit issues) have stable jobs and consistent incomes that span decades – a strength that would go unnoticed in a standard credit report. By recognizing such patterns, lenders can safely say “yes” more often. Studies by the Urban Institute found that cash-flow data (like income records) could expand mortgage access for borrowers with thin or no credit scores, and even lower default risk for those borrowers. In fact, incorporating indicators like consistent rent and utility payments and steady income inflows was shown to re-approve significant numbers of minority applicants who would otherwise be denied, helping to close approval gaps. In the personal loan space, the earlier example of Upstart illustrates expanded approvals: its model nearly doubled the approval rate for applicants in the 620–660 FICO band, thanks to additional signals like employment history that traditional models overlook. For lenders, the ROI comes from writing more loans to good customers who would have been filtered out by rigid criteria. This means higher interest income and market share, without an adverse hit to losses. Put simply, long-term income analytics allow lenders to lend smarter, not just tighter. For example, a borrower with a middling credit score but who has earned a reliable salary for 25 years in the same profession is a far safer bet than the score suggests – and long-term data will flag that. Lenders using such data have reported not only more approvals, but also that many of these newly approved customers perform well. This expanded capacity can be transformative in mature markets: it allows growth in loan portfolios and fosters customer goodwill by offering credit to those who deserve it.

Verify4 is the only firm offering real-time, high-coverage income and employment verifications—covering over 93% of employees and 99% of wage data, with up to 25 years of history. It's fast, affordable, and integrates easily into existing decision systems

To learn more about how Verify4 can help your organization make smarter, fairer decisions, reach out at [email protected] or visit www.Verify4.com