What’s the Magnitude of the Housing Crisis? It Depends on Your Definition of ‘Affordable’

The United States is in the grips of an affordable housing crisis. However, how we choose to define “housing affordability” can shape our understanding of the scale of the problem—and what policymakers’ subsequent responses should entail.

Typically, housing is considered affordable if total housing costs are less than 30 percent of total household income, but some experts criticize this measure. A recent study by Matthew Brooks produced a series of alternative estimates of affordable housing among American households with low incomes and discusses how different measures of income, household composition, and affordability substantially change the apparent scope of the housing affordability crisis.

The author used household microdata from the 2019 American Community Survey for information on housing costs, incomes, and household sociodemographic characteristics. He limited households to those with incomes below the 40th percentile ($60,000) and used the TaxSim32 (a tax-based model) to estimate household income tax liabilities. The author then focused on the following three variables, analyzing the different ways they can be measured:

1. Sources of income. Brooks focused on pretax total income, post-tax total income, wage income after tax, and wage income before tax.
2. Providers of income. The author analyzed incomes from all household members, heads of household alone, heads of household and spouses, and household members between ages 25 and 64.
3. Threshold of affordability. Brooks introduced multiple thresholds for how to define affordability, such as spending a certain percentage of household income on housing or measuring a household’s residual income after housing payment.

Interacting these variables, the author created 80 unique measures of housing affordability. Each measure produced a different estimate of the percentage of people with low incomes living in affordable housing. The author then compared these estimates, noting differences in the overall percentage, the prevalence of ethnic and racial disparities, and the share of people with low incomes living in “shelter poverty”—defined as the threshold where housing costs are high enough after considering income to put a household into poverty. The author also finds that the traditional measure of housing affordability—spending no more than 30 percent of household income on housing—may overstate ethnic and racial disparities.

Key findings

• Different measures produced widely varying estimates of housing affordability, from as high as 69.8 percent of households with low incomes living in affordable housing to as low as 20.2 percent.
• Across all measures, the average estimate of housing affordability was 44.2 percent, meaning 55.8 percent of households with low incomes live in unaffordable housing.
• 26.2 percent of households with low incomes do not have affordable housing under any measure, while 17.5 percent have affordable housing under all 80 measures.
• The majority of households with low incomes (66.6 percent) live in shelter poverty.
• The share of households living in affordable housing decreased by 2.9 percentage points after accounting for household composition, suggesting that households with children may struggle to attain affordable housing.
• Setting an affordability threshold based on residual income (household income after housing costs) rather than housing cost burden (share of household income devoted to housing) substantially increased the estimated number of people living in affordable housing.
• Measurement choices also influence estimates of housing affordability and disparities in housing affordability for specific ethnic and racial groups.
• When comparing white, low-income households with Black, Hispanic, and Asian low-income households, the standard measure of affordability produced a larger difference between the three pairs than did the mean of the 80 measures.
  • Under the standard measure, the share of white households living in affordable housing (45.7 percent) is 11.7 percentage points greater than the share of Black households living in affordable housing (34.0 percent). Using the average of the alternative measures, this difference falls to 9.1 percentage points. For Hispanic households, the difference fell from 12.7 to 10.1 percentage points; for Asian households, the difference fell from 17.5 to 12.9 percentage points.
  • Measures focusing on heads of household reduce disparities between white and nonwhite households, which may be explained by age and household structure.
  • Residual income methods increase the shares of people living in affordable housing and reduce white-nonwhite disparities. However, Black, Hispanic, and Asian households are still more likely to spend a higher proportion of their incomes on housing and have a harder time meeting all other household needs.
  • White households had yearly gross housing costs of $11,366, compared with $11,310 for Black households, $13,156 for Hispanic households, and $16,504 for Asian households. The author largely attributed these differences to the groups’ geographic distribution and disparities in homeownership.

Policy implications

• Researchers and policymakers should be mindful of the underlying assumptions of the standard measure of affordable housing and consider modifying those assumptions to reflect local community circumstances. For instance, when assessing housing affordability for a community with a high number of multifamily households, a measure that only includes head-of-household income may underestimate affordability.
• The study’s findings underscore the importance of ensuring affordable housing for diverse household types, including those with only one income earner or those with children.