Please rotate your device to landscape mode to view the charts.

Lender Automation and Racial Disparities in Credit Access

Journal: Journal of Finance

Date: 20240401

Author: Howell, Sabrina T.; Kuchler, Theresa; Snitkof, David; Stroebel, Johannes; Wong, Jun

Abstract:
Process automation reduces racial disparities in credit access by enabling smaller loans, broadening banks' geographic reach, and removing human biases from decision making. We document these findings in the context of the Paycheck Protection Program (PPP), where private lenders faced no credit risk but decided which firms to serve. Black-owned firms obtained PPP loans primarily from automated fintech lenders, especially in areas with high racial animus. After traditional banks automated their loan processing procedures, their PPP lending to Black-owned firms increased. Our findings cannot be fully explained by racial differences in loan application behaviors, preexisting banking relationships, firm performance, or fraud rates.

Link: Google Scholar

Background and Context

The Program

The Paycheck Protection Program (PPP) provided $800 billion in government-guaranteed loans through private lenders to help small businesses during COVID-19, with no credit risk to lenders.

Research Focus

This study examines how automation in lending processes affected racial disparities in PPP loan access by analyzing data from 11.8 million PPP loans made between April 2020 and May 2021.

Methodology

Researchers analyzed lending patterns across different types of lenders (fintech, large banks, small banks) and tracked changes when banks automated their loan processing systems.

Fintech Lenders Served the Highest Share of Black-Owned Businesses

  • Fintech lenders made 26.5% of their PPP loans to Black-owned businesses, far higher than traditional banks
  • Small banks had the lowest share at 3.3% of loans going to Black-owned businesses
  • This pattern aligns with differences in automation levels across lender types

Bank Automation Led to Increased Lending to Black-Owned Businesses

  • Banks that automated their loan processing saw their share of lending to Black-owned businesses increase from 4.4% to 12.0%
  • This represents nearly a tripling in the share of loans to Black-owned businesses
  • The increase occurred shortly after automation implementation

Automation Impact Stronger in Areas with Higher Racial Animus

  • The effect of automation was stronger in areas with higher measured racial bias
  • Across different measures of racial animus, automation had a larger positive effect on Black-owned business lending
  • Suggests automation helps reduce the impact of human bias in lending decisions

Racial Disparities in PPP Loan Approval Rates Through Lendio Platform

  • No racial disparities in approval rates at fintech lenders
  • Conventional lenders were 3.9 percentage points less likely to approve Black-owned business applications
  • Demonstrates that automated fintech lending processes treated applications more equally

Average PPP Loan Amounts by Lender and Race

  • Fintech lenders made smaller loans on average ($31,228) compared to small banks ($88,083)
  • Lower fixed costs allowed fintech lenders to profitably serve smaller businesses
  • This benefited Black-owned businesses which tend to be smaller on average

Contribution and Implications

  • Process automation in lending can substantially reduce racial disparities in credit access by enabling smaller loans, broadening geographic reach, and removing human biases
  • The findings suggest that technological innovation in financial services can help address long-standing inequities in access to capital
  • Results highlight the importance of automated lending processes in achieving policy goals around financial inclusion

Data Sources

  • Chart 1: Based on Table I Panel B showing share of loans to Black-owned businesses by lender type
  • Chart 2: Based on Table III showing automation effects on lending at small and medium-sized banks
  • Chart 3: Based on Table V Panel B showing interaction effects between automation and racial animus measures
  • Chart 4: Based on Table VII showing analysis of Lendio loan application data
  • Chart 5: Based on Table I showing average loan amounts by lender type