In the Banking Capital of the South, Some Communities Find Challenges in Accessing Mortgage Loans

Low-cost homes valued at $100,000 or less are an important wealth-building vehicle for low- to moderate-income households in the United States. As my colleague noted in a previous post, low-cost homes do exist—but the vast majority are purchased with cash. Buyers looking to purchase these homes through the traditional mortgage process often face higher rates of application denials than buyers of more expensive homes, despite similar levels of credit worthiness.

While much of the existing research and reporting on small dollar mortgages is focused on a national scale, I wanted to know what publicly available data reveals about the market for low-cost properties at the state and neighborhood level. As part of our research on small dollar mortgages with the Center for the Study of Economic Mobility at Winston-Salem State University, I conducted a preliminary analysis using data from the Home Mortgage Disclosure Act (HMDA).

I focus on the state of North Carolina, a hub for banking institutions, and Forsyth County in particular, to understand the spatial patterns of mortgage denial rates for low-cost properties. Are there areas that experience chronically high mortgage loan denial rates? What are the implications of these findings for communities of color, particularly Black families who have been historically locked out of homeownership?

Enacted in 1975, the HMDA requires that lending institutions report loan-level characteristics about mortgage applications, including applicant demographics (i.e., race/ethnicity), loan amounts, and application outcomes (i.e., whether a mortgage originated or was denied). Complex institutional reporting requirements—governing which financial institutions are required to report data—means that this data does not encompass all mortgage loan applications made. However, according to the Consumer Financial Protection Bureau, this data is still the “most comprehensive source of publicly available information on the U.S. mortgage market.”

Additionally, in this analysis, I only consider mortgage applications for conventional loans, excluding federal mortgage products like FHA-insured loans (see data note 1). Because the data is time-limited, and only includes application records over the twelve-year period between 2007 and 2019, I look at denial rates that are summarized across these years.

To provide context on the small dollar mortgage market—defined in this analysis as loans $100,000 and below—we first assessed the denial rates of all mortgage applications in North Carolina, regardless of the loan amount (see data note 2). Figure 1 visualizes average mortgage denial rates at the census tract level in North Carolina between 2007 and 2019. Over this period, 14 percent of mortgage applications were denied on average.

While a 14 percent denial rate may seem like fairly decent odds, Figure 1 shows that different regions of the state experience vastly different denial rates. Charlotte and Raleigh, the state's major economic centers, include neighborhoods with some of the lowest mortgage denial rates in North Carolina.

Some of these rural areas experience denial rates that are over 50 percent—meaning the chances of mortgage approval are essentially a flip of a coin.

Conversely, rural communities in the eastern part of the state primarily experienced some of the highest mortgage denial rates. The average denial rate in rural census tracts was 18 percent while the denial rate in urban census tracts was only 11 percent (see data note 3). These disparities are another example of the widening urban-rural divide in North Carolina. The data shows that some of these rural areas experience denial rates that are over 50 percent —meaning the chances of mortgage approval are essentially a flip of a coin.

Using American Community Survey data from the U.S. Census Bureau, we found that in rural and majority-Black census tracts, the average denial rate was 32 percent—more than twice the state average. In comparison, the average denial rate of rural, majority-white census tracts was only 17 percent (see data note 4). These glaring disparities in access to credit are likely creating additional barriers to wealth building for rural populations, particularly rural Black households, despite living in an area where housing costs are much lower.

To shed light on how the mortgage market for low-cost properties differs, we assessed the denial rates of small dollar mortgages statewide over the same period, 2007 to 2019. Over this period, small dollar loans accounted for 24 percent of all mortgage applications.

Figure 2 maps the average mortgage denial rate of loans under $100,000, also at the census tract level. The data reveals that the average statewide denial rate for sub-$100k loans is 27 percent, nearly two times the average overall denial rate. The higher rate of mortgage denials for low-cost properties confirms previous research and reporting on lower loan accessibility for the sub-$100k market.

At the same time, spatial patterns show that economic centers like Raleigh and Charlotte remain somewhat resilient to high denial rates, especially in the urban core. This could be for a myriad of reasons, including the fact that urban low- to moderate-income households are more likely to earn higher wages than their rural counterparts. This makes it easier for these households to save for a down payment, lower their debt-to-income ratio, and improve their overall chances in accessing mortgage credit. This could also be due to greater access to homeownership programs (like down payment assistance) in urban areas relative to rural ones.

Ultimately, limited access to small dollar loans affects low- to moderate-income households in both rural and urban areas, and particularly Black and brown families which were excluded from receiving publicly subsidized, federally-insured mortgage loans after World War II. These systemic disadvantages have resulted in today’s lower rates of homeownership and little intergenerational wealth for these communities.

While the statewide maps raise concerns around access to mortgage credit for rural communities, a closer look reveals that disparities extend beyond the urban-rural divide. The focus of our larger research is in Forsyth County, North Carolina—home to the city of Winston-Salem. The county is racially and ethnically diverse: about 40 percent of residents identify as Black or Hispanic/Latino. However, it’s also ranked one of the lowest in the country for upward economic mobility and suffers from high rates of concentrated poverty. So, who is impacted by lower loan accessibility for low-cost homes in Forsyth County?

Figure 3 visualizes denial rates of sub-$100k loans in Forsyth County, and shows that they are especially high in particular census tracts. While the average small dollar mortgage denial rate in the county is 18 percent, some census tracts, notably those encompassing Winston-Salem’s historically Black neighborhoods, have a much higher denial rate. One tract had an average denial rate of 57 percent—more than three times the county average. Lower accessibility to mortgage credit for these relatively affordable homes hinders upward mobility and opportunities to build wealth in these communities, especially as rental prices soar in Winston-Salem.

These preliminary findings reveal the importance of location when analyzing mortgage credit access for low-cost properties. Many investigations on mortgage denial rates generalize findings at the metropolitan, regional, or national level (see data note 5). However, this obscures the disparities that occur at the census tract or neighborhood level—a scale which allows local policymakers, housing advocates, and community development financial institutions to better understand the experiences of the communities they serve and target resources which aim to increase access to homeownership. Further, these insights can also inform future research on how decreased access to credit in certain communities affects property values, social capital development, and economic mobility.

For more information about our research on small dollar mortgages, please see our project summary and a post summarizing prior research on the issue.

To stay up-to-date on all our research, subscribe to our monthly newsletter here and follow us on Twitter at @FLHatNewAmerica.

Data note 1: This analysis only includes conventional loan applications for the purpose of purchasing an owner-occupied home. Loans for home renovations and refinancing were excluded.

Data note 2: Denial rates were calculated by dividing the number of denied applications by the total number of applications per census tract (excluding records where loans were purchased by institutions or mortgage pre-approvals). The total number of applications include records where loan applications were: approved but not accepted by the applicant, denied by the financial institution, withdrawn by the applicant, closed for incompleteness, or originated.

Data note 3: We define rural census tracts according to the USDA Food Access Research Atlas: A census tract is urban if the geographic centroid of the tract is in an area with more than 2,500 people; all other tracts are rural.

Data note 4: 32 percent of all census tracts in North Carolina are rural, majority-white while only 1 percent are rural, majority-Black.

Data note 5: We use loan amounts as a proxy for low-cost homes, however, it’s important to note that property values often differ from mortgage loan amounts due to upfront costs (i.e., down payments).