Mortgage
Census tract averages are hiding who actually cannot get a mortgage
By Real Estate Wire Staff, . Real Estate Wire.
Mortgage lenders who use neighborhood income data to assess borrower access are working from a distorted picture, according to an analysis published by the American Land Title Association drawing on research from iEmergent. The firm examined five years of Home Mortgage Disclosure Act data, comparing individual applicants' incomes against the median family income of the census tracts where they sought to buy or refinance.
The core finding is a mismatch between where low-income borrowers live and where low-income neighborhoods are. Applications tied to low-income tracts represent only 3% of all mortgage applications, while applicants who are themselves low-income account for 11%. That gap means a substantial share of lower-income borrowers are applying in higher-income neighborhoods, and tract-level averages erase them from view.
"A census tract's income level is a neighborhood average, and averages smooth over the households facing the steepest barriers," iEmergent chief executive Laird Nossuli said. "If lenders rely only on tract-level data, they can conclude that access for low-income borrowers is better than it actually is and build strategies around the wrong problem."
The distortion shows up in approval rates too. Geography-based figures imply that around half of low-income applicants receive loans. Measured by the borrower's own income, the actual rate is closer to four in ten.
Loan size tells a similar story. The average purchase loan in lower- and moderate-income areas runs 31.6% below the average in higher-income areas. But when sorted by the borrower's income bracket rather than the neighborhood, lower- and moderate-income borrowers receive loans that are, on average, 50.7% smaller than those of higher-income borrowers. The neighborhood comparison understates the gap by nearly twenty percentage points.
iEmergent's conclusion is that lenders who want to serve this segment need individual-level data, not tract-level proxies, and that effective outreach has to address both financial readiness before application and the lending process itself.
The supply side compounds the problem. Listings a household on $75,000 or less could actually afford are down 60% since March 2019, the ALTA report says, measured to March 2026. The median down payment in the first quarter of 2026 was $23,400, down 19% from a year earlier but still nearly double the $12,500 recorded in the same quarter of 2019.
Owners equivalent rent was up 3.1% year on year as of August 2026, according to the U.S. Bureau of Labor Statistics. For lower-income households already stretched by rising shelter costs, that pressure on monthly budgets makes accumulating a down payment harder even as listed prices have moderated slightly from their peak.
The practical implication for lenders is that fair-lending analysis and community reinvestment strategies built on area-level income classifications may be measuring the wrong population. A lender whose data shows adequate approval rates in low-income tracts could still be systematically underserving low-income borrowers who happen to live or buy elsewhere. Whether regulators will push for individual-level reporting standards as a result of findings like these is an open question the iEmergent analysis does not address.