Are Loans to Minority Firms Mispriced?

Summary

A research presentation titled Are Loans to Minority Firms Mispriced?, by researchers at the University of Washington, Virginia Tech and the Public Policy Institute of California, dated January 29, 2025. It uses a survey of small business loans, conducted by the Foster School of Business and the Social & Economic Sciences Research Center at Washington State University, to test whether interest rates and collateral differ across racial groups and by gender. The presentation reports statistically significant mispricing resulting in higher interest paid by Asian American-, Black- and Hispanic-owned firms and woman-owned firms. It estimates that minority business borrowers pay over $8 billion in higher interest annually. It also compares results across lender types, summarizes borrower opinions and describes weighted robustness tests on a sample of 2,784 firms.

Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used

Full text

ARE LOANS TO
MINORITY FIRMS
MISPRICED?
Measuring the Risk Mispricing for Firms Owned by
People of Color and the Lost Economic Opportunity
That It Creates

William D. Bradford University of Washington
Michael Verchot University of Washington
Chunbei Wang Virginia Tech
Magnus Lofstrom Public Policy Institute of California
Credit mispricing occurs when lenders adjust
the terms for borrowers based on factors
unrelated to their financial risk, such as the
race of the business owner, rather than
economic and creditworthiness indicators.

This research explores whether interest rate
and collateral mispricing are evident across
racial groups using survey data on small
business loans.
Overview
                                               Findings
                                               Interest Rates                     Collateral
                                               • Our data indicates               • Co-signatures from third
                                                 statistically significant          parties are required for
                                                 mispricing resulting in higher     Asian American-, Black- and
• Nationwide Survey on Small Business            interest paid by Asian             Hispanic- and Native-owned
  Lending                                        American-, Black- and              firms more frequently than is
                                                 Hispanic- owned firms, and         justified by our economic
                                                 woman-owned firms.                 analysis.
• Assessment of loan terms—interest rates
  and collateral—for loans granted to small    Impact                             Impact
  businesses across the U.S. from Jan 2022
                                               • This mispricing leads to         • The opportunity cost due to
  to Jun 2023. Similar data has not been
                                                 minority business                  excessive collateral
  available since the Federal Reserve’s 2003                                        requirements remains
                                                 borrowers paying over $8
  Survey of Small Business Finances (SSBF).      billion in higher interest         unclear due to limited data.
                                                 annually.
II. Background
                                             Description of Surveys
• Conducted By:                              • Supplier.IO provided the authors with a nationwide list of over 300,000
   ⚬ The Foster School of Business,            firms, to which we sent an internet survey.
     University of Washington
                                             • The firms were potential or actual government or business suppliers,
   ⚬ Social & Economic Sciences                some of which also served personal consumers.
     Research Center (SESRC),
     Washington State University             • The survey collected firm/owner/lender information on non-
                                               government guaranteed borrowing by small firms (< 500 employees)
                                               from January 2022 through May 2023.
• Focus Areas:
   ⚬ National survey on business loan        • Duns credit score information was added for a subset of firms that
     and credit accessibility, equity, and     completed the survey.
     related challenges
                                             • To be eligible for the survey, a business must be operating at the time
                                               of the survey and a for-profit firm with employees.
• Sponsor:
   ⚬ The Schultz Family Foundation           • The firm must have borrowed during the specified period through a line
                                               of credit or loan. Following previous studies, firms in the finance
                                               industry were ineligible.
FINDINGS
Exhibit 1A
Loan
Outcomes                                                                   Race                                        Gender


Exhibit1A
Measures:
Race and
Gender




             *The survey asked the firm its location in six intervals of collateral relative to the amount of credit
             granted: 0, 1 – 50%, 51 – 99%, 100%, 101 – 150%, or greater than 150%.
             Loan Outcomes
                      Race   Gender
Exhibit 1A
Loan
Outcomes
(Cont.)

Exhibit1B
Measures:
Collateral
SUMMARY   White-owned firms

          • Paid the lowest interest rates.

          • Committed the least collateral relative to
            the loan amount, were least often required
            to commit more collateral than the amount
            of the loan, and least often required by the
            lender to have an external party sign for
            the loan.

          • Were the most satisfied with the credit they
            received.
Comparison
Process
Minority-Owned vs. White-Owned Firms
in Loan Pricing
 • Separate Comparisons:
   Each minority group is compared individually to white-owned firms.

 • Indicator Variable:
   An indicator variable for each minority group is included, capturing the
   loan rate difference between the minority and white-owned firms.

 • Sequential Analysis:
   The minority-white loan pricing difference is analyzed by sequentially
   adding control variables to examine the impact on loan rate disparities.
Control
Variables
                  Race   Gender
Selected Firm
Characteristics


  Exhibit
  2
Firm Condition   White Firms
                  • Report the highest fraction of firms with the same or higher revenue in
                    2022 vs. 2021.
                  • Have the highest fraction of firms with the same or more employees in
                    2022 vs. 2021.
                  • Show the highest credit rating scores.
                  • Report the lowest fraction of firms in good or better condition at year-end
                    2022.


                 Black and Hispanic Firms
                  • Have the lowest credit score profile



                 Other Traits
                  • Mixed across groups; no group stands out as significantly better or worse in
                    other firm measures
Quantifying Impact of Race and Gender on Interest Rates and Collateral

       I = interest rate charged to borrowing firms
       I is based on a set of controls (attributes) that gauge the lender’s risk:

       I = 𝛽₀ + 𝛽₁X₁ + 𝛽₂X₂ + 𝛽₃X₃ + …. + 𝛽NXN + 𝛽RRace + 𝛽GGender                           (1)


       𝛽₀ is the base interest rate charged

       𝛽j (j = 1 through N) is the coefficient reflecting the effect of attribute j on the interest
       rate. X j is the firm’s quantity of attribute j.

       Mispricing indicated if 𝛽R or 𝛽G is statistically different from zero.

       Estimate each 𝛽i through ordinary least squares regression.

       For Collateral, change I to C in the model above. C and 𝛽j are estimated using model
       (1) with a different form of regression.
Exhibit 3
                         Hispanic

Regressions
Controlling for
                          Black
Different
Characteristics

Minority v.               Asian

White Firms
and Male-                 Native
Owned vs.                American
Female
Owned Firms              Women

• Significance levels:
    • p < 0.10 (*)
    • P < 0.05 (**)
    • P < 0.01 (***)
Exhibit 4                     A



Split Sample
Regressions:

Minority -
White                         B
Interest Rate
Differences


Notes:
• Standard Errors in
  Parentheses                 C
• Statistical Significance:
    • p < 0.10 (*)
    • p < 0.05 (**)
    • p < 0.01 (***)
Interest
Rates                        Minority-White
                             difference at
Minority-White               large banks
Interest Rate
Differences
Across
Lenders
                             Lender i-large
                             bank difference
                             in minority-
                             white interest
Notes:                       rate
• Dependent Variable: Loan
  Rate
• Significance levels:
    • p < 0.10 (*)
    • P < 0.05 (**)
    • P < 0.01 (***)
SUMMARY:         Black-owned firms

Interest Rate     • Paid higher rates than their white counterparts at credit unions, fintech
                    lenders and nonbank finance companies. Black owned category was not


Differences
                    statistically significant in loans from large banks, small banks and CDFIs.




Across Lenders   Hispanic-owned firms
                  • Paid higher interest rates than their white counterparts at all of the six lender
                    types except CDFIs.


                 Asian-owned firms
                  • Paid higher interest rates than their white counterparts at large banks. There
                    were too few cases in the sample to make inferences about the other lenders
                    to Asian firms.


                 Women-owned firms
                  • Paid higher interest rates than male counterparts at credit unions, CDFIs, and
                    fintech lenders. We did not find that gender affected the interest rate at large
                    banks, small banks and nonbank finance company loans
Collateral         (cont.)   Race   Gender


Minority vs.
White Firms
& Women-
Owned vs.
Man-Owned
Firms



Notes:
• Dependent Variable: Loan
  Collateral Requirements
• Standard errors in
  parentheses
• Significance levels:
    • p < 0.10 (*)
    • P < 0.05 (**)
    • P < 0.01 (***)
IV. Conclusion
A.A: Discussion
  Discussion

Disparities in borrowing experiences extend
beyond economics, fueling broader inequalities and
perpetuating historical discrimination.

Addressing these issues requires joint efforts from
policymakers, regulators, lenders, and other
stakeholders to improve credit access and reduce
discrimination in lending.
    Conclusion B                              Total Annual Difference in Interest $8 Billion
                                                       A             B              C           D              E             F             G
    Additional                                                                                A*B*C*                                     D*E*F
    Interest Paid                                  2021          2022       2022               2022                     Survey
    From                                           Census ABS Fed Res SBCS Fed Res                                      Estimated     $ Total Annual
                                                   U.S. Total    Fraction   SBCS             Number of $ Estimated      Mispriced       Difference
    Mispricing                                     Employer      Applied    Fraction         Approved Average           Interest Rate   in Interest
                                                   Firms         For Credit Approved#        Loans      Loan Size##     Difference %       (000)
                                            White      4,409,715      0.40      0.820         1,446,387      414,796
                                          Hispanic       406,086      0.44      0.680           121,501      336,158        3.34          1,363,765
Notes:                                       Asian       637,539      0.48      0.690           211,153      776,816        3.96          6,502,357
 • Sources: Census ABS (2021),               Black       149,326      0.40      0.500            29,865      276,518        2.82            232,682
   Fed Reserve SBCS (2022)                  Native        28,338      0.46      0.560             7,300      489,811        0.00                  0
 • Calculation Details:
      • Column D, E, F contribute
                                                                                                                                          8,098,804
        to final total in G                             #Partially approved plus fully approved
      • # Partially approved plus fully                ##Fiscal year 2023 average SBA 7(A) loan for each racial group
        approved applications
        included.
      • ## Fiscal year 2023 average
        SBA 7(a) loan size for each
        racial group.
IV. Appendix
Opinions of
Borrowing
Firms
Q35:                   Race   Gender
“Overall, how
satisfied were you
with the credit
services offered by
lender that provided
you this credit?”
Opinions
Lender of
Borrowing
Firms (cont.)
Q37:                    Race   Gender
“Compared to the
average business
owner with the same
business risk factors
and other
attributes, how
would you describe
your experiences in
applying for a loan
or line of credit?”
SUMMARY:          Minority and women-owned firms

Opinions of        • Found going through the credit process
                     more difficult
Borrowing Firms
                   • Were less satisfied with their loan terms

                   • More often felt that their loan terms were
                     mispriced, than were White and men
                     owners.
Opinions
Lender of
Borrowing
Firms (cont.)
Q39:
“Mispricing occurs
when the lender’s
terms (interest rate,      Race   Gender
collateral, etc..) do
not objectively match
the risk and other
attributes of the
business being lent
the money. The terms
are more onerous or
less onerous than
those in a fairly priced
market. Everything
considered (interest
rate, collateral, terms
of loan, etc.), do you
feel you have
experienced
mispricing?”
Robustness    Sample Composition

Tests Using
               • Our sample of 2,784 firms includes a higher
                 concentration from more populated states.
               • However, it does not fully mirror the U.S. distribution of

Weights          minority and white-owned firms by state and industry.


              Weighted Analyses
               • We conducted two separately weighted analyses to
                 enhance representativeness.
               • In both analyses, weights were applied to reflect the
                 national distribution of white-owned and minority-
                 owned firms across different states.
Excerpts                     Race   Gender

from
Weighted
Regressions
Using Race-
State-Industry
of Gender-
State-Industry
Weights
Notes:
• Dependent Variable: Loan
  Rate and Requirement for
  External Party Signature
• Standard errors in
  parentheses
• Significance levels:
    • p < 0.10 (*)
    • P < 0.05 (**)
    • P < 0.01 (***)
THANK YOU
For More Information, Contact:


Michael Verchot
Director, Consulting and Business Development Center
Foster Scholl of Business, University of Washington
Email: mverchot@uw.edu
Phone: 206.572.9521


File and source

File

2025-01-29_a32584_d276684_uw-foster-school-of-business-interest-rate-and-collateral-di.pdf

Size

962,616 bytes

SHA-256

41fcbb85a99ce727e358072056cd7015e24f56f4cb8cd2aaddc3ca2371d694c1

Our copy

2025-01-29_a32584_d276684_uw-foster-school-of-business-interest-rate-and-collateral-di.pdf

Original

app.leg.wa.gov