How Many Households in the Southeast are Low- to Moderate-Income

How Many Households in the Southeast are Low- to Moderate-Income?

October 14, 2026

• Partners Update
• Publication
• Regional Economy
• Community and Economic Development

The Atlanta Fed's Community and Economic Development function supports the central bank's mandate of stable prices and maximum employment by helping to improve the economic opportunity of low- and moderate-income (LMI) individuals and underserved places through research and convenings. As such, accurate and timely estimates of LMI populations allow the Atlanta Fed to better understand the communities that we serve in the Sixth District.¹ The share of LMI households can vary considerably from place to place.² Identifying this variation across geographies provides crucial context for understanding regional economies and community conditions. By providing timely and comparable estimates across states and metro areas, this analysis fills an important data gap and offers clearer context for understanding how local economic conditions differ.

In this Partners Update, we provide estimates of LMI households using data from the 2024 American Community Survey (ACS). The results of this work answer the question, "What percentage of households in my area are LMI?" for each micropolitan area, metropolitan area, and state. Our LMI estimation methodology is unique because it incorporates a fractional allocation technique to address a geographic inconsistency, without discarding observations or geographies from smaller regions. Thus, our methodological approach addresses this alignment challenge in a way that enables complete and comparable LMI estimates for all metropolitan and micropolitan areas.

The resulting estimates offer a practical lens for understanding how households experience their regional economies. Over half of LMI households are cost-burdened, spending more than 30 percent of their income on housing costs, while households with incomes above the LMI threshold are less likely to face affordability constraints.³ Further, research shows that workers in LMI tracts face additional barriers and disadvantages in seeking employment, and LMI individuals are more likely to work in occupations with lower levels of educational attainment and below-median earnings.⁴ LMI status is also commonly utilized in federal housing programs and community development policy and is often referenced for purposes of eligibility and needs assessments.

Estimates of LMI populations are also valuable because, as a relative measure of economic need, they assess a household's standard of living compared to all households in its community, rather than against an absolute benchmark.⁵ Some assessments of economic vulnerability rely on absolute measures of financial need, which evaluate a household's standard of living against a fixed income benchmark like the Federal Poverty Level (FPL), which does not vary across state or locality.⁶ However, economic vulnerability extends to a larger part of the income distribution than what the FPL encompasses. Thus, estimates of LMI populations provide a more expansive and contextually relevant description of a given communities' lower-income residents. We define an LMI household as one with an income at or below 80 percent of area median income (AMI).⁷

Chart 1: Share of Low- to Moderate-Income (LMI) Households in Sixth District States

Data insights

Nationally, an estimated 41.6 percent of households are low- or moderate-income. Across Sixth District states, LMI household shares range from 39.1 percent in Tennessee to 42.7 percent in Louisiana. All six states in the Atlanta Fed's District have LMI household shares within 6.1 percent of the national average.

Chart 2: Share of LMI Households in Sixth District Metropolitan Areas

Across all metropolitan areas nationally, the average LMI household share is 41.6 percent. There are seventy-one metropolitan statistical areas that are within or intersect with Atlanta Fed District states. Twenty-two of these metropolitan areas have an LMI household share above the national average. Among these metro areas in Sixth District states, Houma-Bayou Cane-Thibodaux, Louisiana, had the highest LMI share at 48.1 percent, while Hinesville, Georgia, had the lowest, at 29.5 percent. Zooming in on the six metro areas in the Southeast where Atlanta Fed branch offices are located, Miami-Fort Lauderdale-Pompano Beach, Florida, had the highest LMI household share, at 46.5 percent, while Nashville-Davidson-Murfreesboro-Franklin, Tennessee, had the lowest LMI household share, at 37.7 percent. Two of the six metro areas where Atlanta Fed branch offices are located have an LMI household share above the national average for metro areas.

Chart 3: Share of LMI Households in Sixth District Micropolitan Areas

Turning to micropolitan areas within Atlanta Fed District States, Oxford, Mississippi, had the highest LMI share of all micropolitan areas in Sixth District states (52.4 percent), while Russellville, Alabama, had the lowest LMI share (31.9 percent). The average LMI household share across all micropolitan areas nationally was 41.5 percent, similar to the national share in metropolitan areas. Forty-four of the eighty-three micropolitan areas in Sixth District states have an LMI household share that exceeds that national average.

Chart 4: Share of LMI Households in US States

To help explain these patterns, it is informative to consider the mix of incomes present within a community. The percentage of households that fall below the LMI threshold is directly related to an area's income distribution. Places with flatter income distributions, with larger middle classes and fewer households with very high or very low incomes, will have lower LMI household shares. This occurs because relatively few households fall below the LMI income threshold for such places. Conversely, in places with very unequal income distributions, where many households have either very high or very low incomes, the LMI household share will be higher. The large population of high-income households will drive area median income to be higher, leading the large populations of low-income households to fall below the threshold. This can be seen in areas with high concentrations of certain industries such as finance and tech, which create a large group of very high earners as well as a large base of workers in service occupations. This relationship may contribute to why New York, which has the highest level of income inequality among US states, has the highest state-level LMI household share.⁸ In areas with high rental housing costs, such as New York City, high LMI household shares can partly reflect HUD's Fair Market Rent-based adjustments to income limits, which raise LMI thresholds independently of the underlying income distribution.⁹ Some areas have LMI shares above 50 percent because both local income distributions and HUD's housing-cost adjustments can result in a relatively high LMI threshold, which a majority of households may fall below.

While we will not be fully examining the factors that may cause variation in LMI household share in this Partners Update, a number of characteristics may offer insight into why LMI household shares differ across metropolitan and micropolitan areas. For example, areas with a sizeable college or university presence may have higher LMI household shares, as student populations tend to report lower incomes, influencing the income distribution of an area.¹⁰ Similarly, the demographic composition of a place may also have an impact; places with larger shares of elderly residents or single-parent households may have higher LMI household shares.¹¹ Finally, areas with higher poverty rates will naturally have higher shares of LMI households, as LMI populations are inclusive of households in poverty. Ultimately, variation in LMI household share is likely shaped by a range of factors, extending beyond those considered here, which reflect the local dynamics of individual communities.

Opportunities, applications, and conclusions

These data offer timely, localized estimates of the share of a geography's households that are low- and moderate-income. Alternative datasets are limited by infrequent updates or the absence of household-level estimates.¹² The estimates in this analysis address these constraints by using the most recently available microdata to generate LMI household shares that are comparable across geographies. These data can be valuable for communities looking to make data-driven decisions. Conversations related to program targeting, policy analyses, and funding allocation can be enriched with a greater understanding of where LMI households are located.

In a forthcoming Partners Update, we'll look to tract-level datasets to uncover additional insights about LMI households and their communities. These smaller area estimates allow for even more local and granular analysis, offering a more specific look at variation within larger urban areas and across rural areas. These more localized insights will provide a complementary lens for understanding how household conditions unfold across different community contexts.

Data and methods

We use two datasets in this analysis. First, we use the Department of Housing and Urban Development (HUD) 2024 Section 8 Income Limits to calculate a household's LMI designation. HUD adjusts this dataset for cost of living and family size.¹³ Using this data also allows the resulting estimates to be consistent with policy around housing programs that impact LMI households and communities including Public Housing, Section 8 project-based and Housing Choice Voucher programs, and programs providing housing for the elderly and persons with disabilities. Second, we use 2024 American Community Survey (ACS) Public Use Microdata one-year estimates for data on household size and income, as well as where a household resides at the Public Use Microdata Area (PUMA) level.¹⁴ We restrict the analysis to households with eight or fewer members, matching the household-size categories available in the HUD income limits data, and accounting for 99 percent of the ACS sample.

However, using these two datasets together creates a geographic mismatch: the ACS microdata used in this analysis are reported at the PUMA level, while the HUD Income Limits data are reported at the county level.¹⁵ PUMAs and counties are not directly analogous. While some PUMAs are delineated in a way that aligns with county boundaries, some PUMAs comprise multiple counties, some counties comprise multiple PUMAs, and, in some instances, PUMA boundaries partially divide counties. Thus, within the ACS, while it is known what PUMA each household resides in, it is often not known which counties within that PUMA the household resides in and, therefore, what county-level HUD income limits to apply.

To remedy this discrepancy, we fractionally allocate each household across the counties in its PUMA, using a PUMA-to-County Census crosswalk. Fractional allocation is a method that allows us to estimate the probability that a household is located in a county that intersects with a PUMA, based on the crosswalk's allocation factor.¹⁶ This method allows us to compare each household's income to the income limits of each county that is represented within their PUMA, and assign LMI status if household income is less than the 80 percent AMI threshold.

To produce state-level estimates, we use the allocation factors to create a fractional LMI measure for each household, then apply ACS household survey weights to calculate the state-level LMI household share. To produce metropolitan and micropolitan area-level estimates, we use the same approach but add one step: when a household's PUMA overlaps more than one metropolitan or micropolitan area, we split that household's ACS survey weight across the relevant areas using crosswalk allocation factors. This process ensures each household's survey weights continue to reflect the correct number of households. After this adjustment, we calculate the LMI household share for each area as a survey-weighted average of the household-level LMI measures. The results of these calculations are a continuous LMI measure, which preserves the full ACS microdata sample while accounting for the geographic uncertainty of residence introduced by the geography mismatch. From there, we calculate LMI household shares at the metropolitan, micropolitan, and state level, as described above. By focusing this analysis on LMI households, rather than LMI neighborhoods or communities, we can identify LMI households living in places that are not classified as LMI, a nuance that area-level indicators often obscure.

The methodological approach outlined here differs from prior research on this demographic group. For example, in "Understanding the Employment Patterns of Workers in Low- and Moderate-Income Households," the authors construct income thresholds from survey data, rather than using published income thresholds, and in some instances, use PUMAs as the geographic area for calculating AMI.¹⁷ Other methodologies using PUMS data, such as that presented in the Southeastern Rental Affordability Tracker, require imperfect recreation of combined statistical areas, which can necessitate the exclusion of entire PUMAs.¹⁸ We use fractional allocation to address limitations of both approaches. This method allows us to create meaningful economic areas without disposing of any PUMAs, meaning our statistics are available for geographic areas that users are familiar with. The resulting calculations for metropolitan and micropolitan areas are accurate and complete. Using published income thresholds allows these estimates to be meaningful and relevant because they align with the real programs, resources, and policies that community development professionals and LMI households touch and access.

The views expressed here are those of the author's and not necessarily those of the Federal Reserve Bank of Atlanta or the Federal Reserve System. Any remaining errors are the author's responsibility.

The Federal Reserve Bank of Atlanta's Community and Economic Development function supports the Central Bank's mandate of stable prices and maximum employment by helping improve the economic opportunity of low- and moderate-income (LMI) individuals and underserved places for a stronger economy for all Americans. Community development is one of the Federal Reserve's core functions and this responsibility is rooted in its mandates from Congress. Partners Update articles address community and economic development trends, issues, and events. Find more research, use data tools, and sign up for email updates at Community Economic and Development.

• ¹ The Atlanta Fed's District encompasses Alabama, Florida, and Georgia, as well as parts of Louisiana, Mississippi, and Tennessee. In this Partners Update, we present results for the totality of each of these states, not just the areas that fall within our District's boundaries.
• ² Under the Community Reinvestment Act (CRA), income classification for a census tract or a borrower is defined relative to either the Metropolitan Statistical Area (MSA) or Metropolitan Division (MD) median family income or the statewide nonmetropolitan area median family income. Classifications of census tracts change periodically based on updated census data and changes to MSA and MD designations. Classifications of borrowers are based on the Federal Financial Institutions Examination Council's median family income that is adjusted annually using US Department of Housing and Urban Development (HUD) adjustment methodology.
• ³ The National Low Income Housing Coalition found, in 2024, that 87 percent of households earning between 0-30 percent of Area Median Income (AMI), 79 percent of households earning between 30-50 percent of AMI and 53 percent of households earning between 50-80 percent of AMI were cost burdened. The rate of cost-burdened households quickly dropped beyond that threshold to 27 percent for those with incomes between 80 percent and 100 percent and those earning beyond 100 percent of AMI (9 percent). Dan Emmanuel, Raquel Harati, Esther Colón-Bermúdez, and Tyler Eutsler, The Gap: A Shortage of Affordable Homes (Washington, DC: National Low Income Housing Coalition, March 2026).
• ⁴ Kelly D. Edmiston, "Why Aren't More People Working in Low- and Moderate-Income Areas?," Economic Review (Federal Reserve Bank of Kansas City, January 01, 2019); Elizabeth Kneebone and Natalie Holmes, "Understanding the Employment Patterns of Workers in Low- and Moderate-Income Households," Federal Reserve Bank of San Francisco Community Development Research Brief 2025-01 (Federal Reserve Bank of San Francisco, September 24, 2025).
• ⁵ David Brady, "American Poverty Should Be Measured Relative to the Prevailing Standards of Our Time," The Century Foundation (April 27, 2021).
• ⁶ Sierra Stoney, "Who's in Need? Measuring Low and Moderate Income Status and Poverty Status in the Fifth District," District Digest, Econ Focus (Federal Reserve Bank of Richmond, First Quarter 2023).
• ⁷ In performing these calculations, households' incomes are compared to the area median income of their county. The county is the smallest and most local area of geography in this analysis. See Data and Methods.
• ⁸ US Census Bureau, "Income Inequality (Gini Index), 2024: Table B19083," accessed June 12, 2026.
• ⁹ US Department of Housing and Urban Development, Methodology for Determining FY 2024 Section 8 Income Limits (Washington, DC: HUD, 2024).
• ¹⁰ Craig Benson and Alemayehu Bishaw, "When Many College Students Live Off Campus, Poverty Rate Goes Up," America Counts: Stories (US Census Bureau, October 22, 2018).
• ¹¹ Emily A. Shrider and Christina Bijou, "Poverty in the United States: 2024," Current Population Reports, no. P60 287 (Washington, DC: US Census Bureau, September 9, 2025).
• ¹² Data sources such as the Low and Moderate-Income Summary Data (LMISD) and Comprehensive Housing Affordability Strategy (CHAS) Data produce estimates for the population and households, respectively. US Department of Housing and Urban Development HUD Exchange, "ACS 5-Year 2016-2020 Low- and Moderate-Income Summary Data,"; HUD Office of Policy Development and Research, "Comprehensive Housing Affordability Strategy (CHAS) Data and Tables," (Updated December 23, 2025).
• ¹³ HUD's cost of living adjustment adjusts area median incomes upward or downward based on local housing costs. As such, our LMI estimates reflect not only household incomes, but the relative affordability of local housing markets. Higher cost areas will thus tend to show higher LMI household shares than lower-cost areas when both areas have similar median incomes and income distributions. US Department of Housing and Urban Development, "Methodology for Determining FY 2024 Section 8 Income Limits" (Washington, DC: US Department of Housing and Urban Development, 2024).
• ¹⁴ HUD income limits are set using Median Family Income, while the ACS data used in this calculation utilizes household income in order to capture more households, such as those occupied by single individuals and roommates. As single-earner households tend to have lower incomes than families, and because HUD's scaled one-person income limit is relatively high, they are more likely to fall below the income threshold determined by Median Family Income. Accordingly, places with large populations of non-family households, such as those with many single young professionals or college students may therefore see higher LMI household shares.
• ¹⁵ The Census PUMA geography is the smallest geographic area for which microdata is reported. Each PUMA has a minimum of 100,000 people.
• ¹⁶ John R. Logan, Zengwang Xu, and Brian Stults, "Interpolating US Decennial Census Tract Data from as Early as 1970 to 2010: A Longitudinal Tract Database," The Professional Geographer 66, no. 3 (July 1, 2014): 412-420. Our fractional allocation approach closely follows that of Logan et al.
• ¹⁷ Elizabeth Kneebone and Natalie Holmes, "Understanding the Employment Patterns of Workers in Low and Moderate Income Households."
• ¹⁸ Federal Reserve Bank of Atlanta, "Southeastern Rental Affordability Tracker: Methodology" (2026).