How leading banks and fintechs are verifying thin-file applicants

“A growing number of FIs utilize alternative verification methods to achieve this goal. Some FIs utilize data from a third-party credit bureau and authenticate it based on the local country’s requirements. Others use document verification tools to verify an applicant’s foreign passport or driver’s license and behavioral biometric data to identify potentially fraudulent applicants.” (Source: Forbes)

“Newcomer identity verification can be automatically handled by matching and authenticating a newcomer based on their home country credit bureau. Each international credit bureau that we partner with has a unique process for validating the identity of an applicant before they release the report to Nova Credit.

These processes include having the applicant answer knowledge-based questions (for example, streets they’ve lived on), satisfying one-time passwords, providing additional information such as a Government ID number, or scanning a passport for validation.”

“Nimble lenders may seek other data sources to understand an applicant’s financial health. Transactional bank data provide a highly-detailed source of truth. From this, a lender can understand the nature and reliability of income, the types of recurring transactions, past interactions with lenders who may not actively furnish data to credit bureaus, and a measure of the consumer’s debt capacity.

In addition to bank data, many lenders seek to verify an applicant’s employment or income, relying on a mix of online employment verification services and document analysis. This can be relatively straightforward for traditionally employed borrowers, whereas, for non-traditionally-employed borrowers, it often requires greater documentation and a more complex set of calculations. This is all the more true in mortgage lending, where the procedures for income calculation are tightly specified.

Overall, lenders using non-standard data sources to evaluate applicants need to apply the same amount of risk management rigor they would apply to credit bureau data. It is essential to understand the information's completeness, accuracy, and reliability and assess the impact of underwriting strategies across multiple time periods and economic conditions. Finally, it is critical that lenders offer a high-quality, consistent, and transparent customer experience regardless of the specific data sources used to evaluate risk.”

“Fortunately, there are options for underwriting applicants with little or no traditional credit history. These include things like consumer-permissioned bank transaction data, income and employment verification data, telco and utility payment history data, and specialty finance data (such as short-term lenders, lease-to-own, and other markets). These have been found to be helpful in expanding the viable credit universe beyond the traditional credit file. For example, by adding alternative data into scoring, lenders can now score an estimated 21 percent more - or 8.8 million credit-seeking consumers - than when compared to traditional scoring models. And, approximately 15 percent more or 6.3 million applicants that are considered subprime, no hit or thin file could be approved for a near-prime or prime offer without increasing risk when using alternative data in combination with a traditional risk score. Alternative data can also help improve the standing of credit consumers whose limited traditional credit history would make them appear to be greater credit risks.”

“Traditional credit scoring methods don’t fully serve the thin file population very well. However, financial institutions have a plethora of resources beyond traditional credit scoring methods when it comes to underwriting thin-file applicants. Today’s technology affords financial institutions the benefit of allowing each individual’s financial journey to speak for itself. AI affords greater flexibility in credit risk assessment by not requiring fixed relationships between inquiries, payment patterns, types of accounts, and utilization like traditional risk assessments.”

“In markets where access to open banking data is more established, we’ve seen a growth of innovation around using this data to augment traditional credit scores. For example, in the UK, Credit Kudos was one of the early companies to improve open banking data’s usability for affordability and credit risk assessments. Now, we have a host of other companies iterating on this theme, such as ClearScore using open banking data to pull out underwriting risk indicators based on the financial behaviors within transaction data. FIs now have a variety of options of capitalizing on this data and enriching their credit underwriting decisioning.”

“Recently released government projections estimate that the U.S. population will grow by about 40 million people over the next 30 years, with nearly 100% of that growth coming from immigration. Capturing this segment is critical to any FI looking to grow their business in the coming decades. Nova Credit’s Credit Passport provides lenders visibility into newcomers’ identity and complete global credit history, unlocking a significant and growing market segment for FIs.

More than 60 million U.S. consumers are excluded from accessing credit due to limited US credit history, and many of those individuals are creditworthy consumers not targeted by other FIs. We recommend instituting data-driven processes to tap into datasets outside of the traditional US credit bureaus and in doing so be able to confidently understand the identity and financial health of applicants. These data sources include overseas credit history, bank transaction data and so much more.”

“Ultimately, consumers will select the financial institutions that make it easy to work with them, which can mean having multiple ways to evaluate borrowers with different levels of credit history. FIs who don’t have a strong program to evaluate, back-test, simulate, and pilot the use of new data sources may get left behind and miss out on the opportunity to earn the loyalty of millions of non-traditional borrowers.”

“Strong fraud prevention includes identification and verification points across a network of different contributors. Some industries, such as financial services, contribute valuable information about potential fraud. Even there, though, depending upon only one company or source will almost certainly open blind spots in fraud detection. However, the power of a network of fraud detection sources that includes sources outside of financial institutions closes those blind spots and would be the best practice for detecting fraud for thin-file consumers even for those with deeper credit history, as well.”

“Fraud is an ever-evolving risk to financial institutions, and it’s crucial to stay educated on the latest advancements in fraud detection concerning KYC requirements. We have seen synthetic fraud be prevalent with thin-file applications, and more data products and consortiums continue to tackle the mitigation of that risk. Collaboration amongst financial institutions in reporting payment histories to consumer reporting agencies has allowed greater resolution into credit risk assessments. Similar consortiums for mitigating various types of fraud can also be very powerful.” 

“FIs can utilize progressive onboarding which allows applicants who pose a higher risk or may be more difficult to verify (for any reason) to access some products or services at account origination, but not all. For example, once an FI has obtained the required CIP information to verify an immigrant applicant’s identity, they can start the immigrant off on a really low credit limit to reduce any potential fraud risk, then slowly ramp up that credit limit as the applicant proves themself to be a genuine customer.“ (Source: Forbes)

“With respect to newcomers, fraud can be mitigated by relying on an international credit bureau check and the authentication standards that international bureaus require. In addition to traditional KYC safeguards, the unique process of consumer-permissioned data provisioning - for example, entering login credentials before accessing a consumer’s bank transaction data - provides an additional validation layer that can make fraud more difficult for thin-file applications.” 

“Very similar to the ways small businesses and sole proprietorships, and single-member LLCs are underwritten: by looking at the credit risk of the principal responsible for running the business. Student loans are underwritten and often co-signed with the applicant’s parents. The same consideration should be given to other lines of credit as well, where the greater opportunity lies in using underwriting methods that are smarter, more inclusive, and more efficient, as discussed above.”

“Banks under federal law in the US are not prohibited from opening a bank account or from providing other financial services (like credit cards) to minors. However, state laws generally identify that a minor cannot legally enter into contracts, like account agreements or terms and conditions for credit. Therefore, minors need a Sponsoring Adult (a parent or legal guardian) to co-sign with them on the account. Banks and fintechs should then run both the minor and the Sponsoring Adult through their typical KYC/CIP processes. State laws also determine at what age a person becomes an adult and can legally consent — so be sure to check that for whatever state you’re operating in as well.”