New research indicates that FinTech lending has not been as ‘disruptive’ in risk-based pricing as claimed. While FinTech has provided increased loan access to some individuals, reliance on traditional credit scoring and spillovers from banking regulations leads to mispricing and cross-subsidization of borrowers. The authors suggest alternatives to allocate capital efficiently and improve financial inclusion.
FinTech lending has exploded in popularity over the past decade, promising to revolutionize the credit market by utilizing big data and advanced statistical techniques to improve underwriting and credit allocation. These digital lenders have positioned themselves as game-changers, pledging to increase the efficiency of credit allocation and better serve underserved populations with weak credit scores. As researchers, we sought to investigate whether FinTech lenders have indeed delivered on their promises, breaking away from traditional lending practices and ultimately improving consumer welfare. What we find in the data, however, is that FinTech lenders still heavily rely on traditional credit scoring methods in their loan pricing.
Interest rates are largely determined by FICO scores with large disparities in interest for borrowers that have similar expected risks. This apparent emphasis on FICO scores leads to mispricing and cross-subsidization across borrowers. We attribute some of the mispricing to segmentation in consumer credit markets that arise from bank regulations which limit availability of low-cost loans to high-risk borrowers. This unexpected truth underscores the unintended consequences of bank regulations and the need for FinTech lenders to adopt more sophisticated pricing models.
The Study
In our study, titled “FinTech Lending with LowTech Pricing,” we analyzed a dataset of unsecured FinTech personal loans made in the U.S. from 2014 to 2020. We examined the pricing models used by these lenders and compared them with a counterfactual pricing model to assess the level of innovation and disruption promised by the FinTech industry. Our research reveals some startling findings.
Findings
First, we found that FinTech lenders continue to heavily rely on traditional FICO scores as the primary determinant of loan pricing, resulting in nonprime borrowers (those with FICO scores less than 660) paying significantly higher interest rates than prime borrowers with the same default risk. This finding suggests that FinTech platforms have not yet broken away from the long-standing pricing regularities of traditional lending markets, contrary to their promises of innovation and disruption. The patterns we observed appear to be driven by limited competition and/or collaboration with traditional banks in funding nonprime borrowers, where regulations restrict risky lending for FDIC insured banks, leading to spillover effects on the industry.
Second, the over-reliance on FICO scores in FinTech pricing decisions has led to economically significant cross-subsidization or mispricing across and within market segments. Nonprime borrowers, who typically lack access to traditional unsecured and secured credit, end up subsidizing prime borrowers. Moreover, within each segment, low-risk borrowers subsidize riskier borrowers. Our research shows that more than 83% of nonprime borrowers overpay relative to counterfactual rates, indicating that they pay more than they would have under ideal risk-based pricing.
Implications for Borrowers, Investors, and Regulators
This revelation about the simplicity of FinTech loan pricing has important implications for borrowers, investors, and regulators. Nonprime individuals, who previously had little access to unsecured credit, have gained access through FinTech lending but appear to overpay relative to their credit risk level. The persistence of these pricing patterns highlights the need for reassessing banking regulation, as they have unintended spillover effects. Additionally, FinTech lenders must improve their pricing models by incorporating alternative data sources and known risk factors to achieve more accurate and fair risk-adjusted pricing.
For investors, the current pricing model creates a suboptimal allocation of capital, as the mispricing leads to inefficient risk-reward trade-offs. Adopting more sophisticated pricing models would benefit investors by enabling better assessment of risk and returns, potentially unlocking greater value in the FinTech lending market.
Regulators should also be concerned about the current state of FinTech lending, as it falls short of providing fair and transparent access to credit. Ensuring a level playing field for all borrowers, irrespective of their credit history, is crucial for promoting financial inclusion and maintaining the stability of the credit market.
The Path Forward
Our findings provide a silver lining for the FinTech industry: while loan pricing appears simplistic by the end of our sample period, there is significant potential for improvement. By adopting more sophisticated pricing models and leveraging alternative data sources, FinTech lenders can increase access to fairly-priced credit for households, especially those underserved, and provide fair risk-adjusted returns to investors. This direction will likely enhance capital allocation across borrowers and boost consumer welfare, fulfilling the promises that the FinTech sector initially set out to achieve. As we look to the future, here are some key areas in which FinTech lenders can improve:
Alternative Data and Advanced Analytics: FinTech lenders should explore the use of alternative data sources to supplement traditional credit scores. These sources can include non-financial information such as utility payments, rental history, and social media data. By incorporating these unconventional data points, lenders can better assess the creditworthiness of borrowers who may not have a strong credit history. In conjunction with new data sources, FinTech lenders have the potential to leverage machine learning and advanced analytics to develop more accurate and dynamic pricing models. These models can identify patterns and relationships between various risk factors, allowing lenders to better segment borrowers and adjust pricing accordingly.
Financial Inclusion: By refining their pricing models and incorporating alternative data sources, FinTech lenders can enhance financial inclusion and extend credit to underserved borrowers. This access to fairly-priced credit can empower individuals and businesses to invest in their futures, fostering economic growth and development.
Collaboration with Regulators: FinTech lenders and regulators should collaborate to establish guidelines and best practices for the use of alternative data and advanced analytics in credit pricing. This collaboration can help ensure that new lending models are transparent, fair, and compliant with existing regulations while also promoting innovation and growth in the sector.
Consumer Education: FinTech lenders should invest in consumer education initiatives to help borrowers understand the factors that influence their loan pricing. By promoting financial literacy, lenders can empower borrowers to make informed decisions and potentially improve their creditworthiness over time.
Conclusion
While our research reveals that FinTech lending has not fully delivered on its promise of revolutionizing the credit market, the potential for change remains strong. Policymakers should consider revisiting stringent banking regulations, moving away from heuristics like FICO score cutoffs as determinants of risk. FinTech lenders, on their part, can embrace more sophisticated pricing models, leveraging alternative data sources, and collaborating with regulators. By disrupting traditional lending practices, FinTech lenders can promote financial inclusion, and ultimately improve consumer welfare. As we continue to examine the evolution of the FinTech lending market, it is our hope that the industry will rise to meet the challenges and fulfill its potential as a force for positive change in the credit landscape.
The Securities and Exchange Commission disclaims responsibility for any private publication or statement of any SEC employee or commissioner. This article expresses the authors’ views and does not necessarily reflect those of the Commission, the commissioners, or staff members. Ben-David is a member of the Academic Research Council at the Consumer Financial Protection Bureau (CFPB). This article does not necessarily reflect the views of the CFPB or its staff.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.
