Skip to main content

The Price of Mortgage Financing for Native Americans


We demonstrate that mortgage loans made to Native American primary borrowers are systematically more likely to be higher-priced than other mortgage loans. This difference is pronounced for mortgage loans on reservations: 27% of mortgage loans for properties on reservation lands with Native Americans as the primary borrower are higher-priced, while only 8.7% of mortgage loans nearby reservations made to non-Native American borrowers are higher-priced. Conditional on having a higher-priced loan, the rate spread facing Native Americans is also higher than non-Native Americans. These findings come from the 2010 to 2017 Home Mortgage Disclosure Act Data which is the most comprehensive, consistent series of mortgage loan data in the USA. We discuss the factors that may drive these outcomes.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2


  1. See Community Development Financial Institutions Fund ((Community Development Financial Institutions Fund)), United States Congress House. Senate Committee on Indian Affairs ((United States Congress House. Senate Committee on Indian Affairs)), Jorgensen ((Jorgensen)), Jorgensen and Akee (2017).

  2. We use the standard definition of “higher-priced mortgage loans,” where higher-cost loans are loans that have an annual percentage rate (APR) that is 1.5 percentage points higher than the average prime offer rate (APOR) if it is a first-lien home purchase loan (Consumer Financial Protection Bureau (Consumer Financial Protection Bureau)).

  3. Specifically whether it is VA-guaranteed (Veteran Administration), FSA/RHS-guaranteed (Farm Service Agency/Rural Housing Service), or FHA-insured (Federal Housing Administration) mortgage.

  4. The rate spread is effectively the difference between the annual percentage rate on the loan and the average prime offer rate in the market at the time the loan was originated.

  5. The Section 184 program is studied bundled with the Federal Housing Administration Section 248 Mortgage Insurance Program in the work of Cyree et al. (2004).

  6. Using HUD 184 data, Wellhausen et al. (2017) also studies the cost of mortgage loans as an outcome of interest using the interest rate as a measure, but focuses on differences in the interest rate heterogeneity across reservations rather differences on and off reservation lands.

  7. There are numerous literature reviews of this area that include, but are not limited to Yinger (2018) and Dymski (2006).

  8. For example, see Oliver et al. (2006) and Beeman et al. (2011).

  9. If it is a jumbo loan, it is 2.5 percentage points or more higher, and if it is a subordinate-lien mortgage, it is higher price if the APR is 3.5 percentage points or more higher than the APOR (Consumer Financial Protection Bureau (Consumer Financial Protection Bureau)).

  10. Less than 1% of census tracts are classified as both a “nearby tract” and a “reservation associated tract” since a smaller number of reservations boarder each other within the same state.

  11. See

  12. Since we cannot perfectly proxy for whether a loan is in reservation land, a significant coefficient on this could imply that Native peoples in census tracts are far more likely to live on tribal land and thus it is a better proxy for the relative cost of a loan on reservation lands. However, to the extent non-Native individuals live in the same locations within a given census tract, it is suggestive about the importance of race even conditional on being on reservation lands.

  13. We follow the recommendations in Oster (2018) and assume a maximum R-squared of 1.3 times the R-squared from the controlled regression.

  14. Excluding loans in census tracts associated with the Navajo reservation from the sample has no qualitative impact on the results. Oklahoma Tribal Statistical Areas, which are institutionally distinct from reservation areas, are also excluded from the analysis.

  15. The sample size changes between Tables 1 and 2 because some counties fixed effects are perfectly correlated with the probability of having a higher-priced loan and these observations are excluded from the models in order to have consistency across specifications within Table 2.

  16. These specifications have a fewer number of observations because bank presence is only available up to 2016. However, the results are qualitatively and quantitatively similar to using the full sample as seen in columns (1) and (3).

  17. Again, specifications that control for bank presence, presented in Table 6, have little impact on the results.

  18. We discuss these issues in more detail in Section “Discussion.”

  19. Only 21.8% of individuals in Hawaii identify as “white alone, no Hispanic or Latino” (United States Census Bureau (United States Census Bureau)).

  20. Future research should evaluate the underlying causes of this sharp increase.

  21. We thank an anonymous referee for this point.

  22. This is consistent with no premium on non-Native American mortgage loans on reservation since Tribal jurisdiction largely only covers individuals identified as “Indians” in the legal sense.

  23. This exercise showed that differences in credit histories could potentially explain differences in the probability of a loan being higher priced and the rate spread under reasonable assumptions.

  24. We believe this classification of Native peoples also fails to recognize non-trivial migration of Native peoples to and from reservation lands (Feir and Gillezeau 2018).

  25. This dollar amount is the average loan amount for AIAN-reservation loans.

  26. This assumes an average rate spread of 2.5% for non-AIAN nearby loans and 5.5% for AIAN-reservation loans.


Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Donna Feir.

Ethics declarations

Conflict of Interest

The authors declare that they have no conflict of interest.

Additional information


The views expressed here do not necessarily represent those of the Federal Reserve Bank of Minneapolis or the Board of Governors.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.



Fig. 3
figure 3

Number of loans originating for tribal lands and neighboring tracts

Fig. 4
figure 4

Proportion of mortgage loans that are higher-priced by race of the primary borrower for all loans within the 2010–2017 Home Mortgage Disclosure Act data in the 48 contiguous states

Fig. 5
figure 5

Mean rate spread, conditional on a loan being higher-priced, by race of the primary borrower for all loans within the 2010–2017 Home Mortgage Disclosure Act data in the 48 contiguous states

Table 6 Probability of higher-priced loan and rate spread conditional on having a higher-priced loan: Accounting for bank branch presence in county
Table 7 Probability of having a higher-priced loan and rate spread conditional on having a higher cost loan: Linear models

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Cattaneo, L., Feir, D. The Price of Mortgage Financing for Native Americans. J Econ Race Policy 4, 302–319 (2021).

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Indigenous peoples
  • Native American
  • Mortgage financing
  • Home ownership
  • Capital access