Introduction

Today's first-time buyers face the toughest path to homeownership in decades. Post-pandemic increases in home prices and interest rates led to a rapid deterioration in housing affordability. In 2019, only one-in-five people thought that home prices would rise by 20 percent or more over the subsequent 5 years, according to a New York Fed survey of consumers.¹ Over the next 5 years ending December 2024, nationwide home price indices rose about 50 percent, indicating how far the housing market moved relative to expectations.² Meanwhile, the 30-year fixed mortgage rate rose from 3.7 percent at the end of 2019 to 6.9 percent at the end of 2024. This combination of higher prices and rates caused monthly mortgage costs to double for would-be buyers.

Incomes have not kept up with housing cost increases—even for younger individuals, who historically have higher wage growth on average. Median income growth from December 2019 to December 2024 was 41 percent in nominal terms for the cohort of individuals aged 25−44 in 2019.³ After adjusting for overall consumer price inflation, this performance was below the pre-pandemic trend.⁴ The average age of first-time homebuyers rose to an all-time high of 38 years old in 2024, according to a survey from the National Association of Realtors, as individuals needed to climb further up their career ladder to afford a home.

The broad decline in housing affordability can be quantified using a variety of public data sources. This analysis seeks to complement recent studies by using granular data showing how budgets—based on changes in take-home pay of millions of individuals—have evolved for the typical individual over time. This lens helps remove a key source of bias affecting public data on housing affordability: since people tend to move to areas in which they can afford a home, local income levels may, in part, reflect home prices themselves. By looking at individual-level income changes, we can better capture how the pandemic and post-pandemic shocks to the housing market led to divergence between incomes and home prices in different types of neighborhoods.

Taking both national and local perspectives, our analysis is organized around the following findings:

In data covering millions of individuals’ take-home pay, the median increase in income among 25−44-year-olds (typical first-time homebuyer age) was 41 percent from 2019 to 2024.⁷ While this growth outstrips economy-wide aggregates—because career and skill advancement is faster for younger people—it falls short relative to the doubling in the monthly expense of a mortgage.⁸ Many people that have been saving for a home over the past several years are finding the reality of what they can afford falling far short of their expectations.⁹

The simplified example depicted in Figure 2 translates these metrics to a hypothetical household budget of a renter seeking to become a homeowner. A mortgage with monthly expenses equal to 40 percent of take-home pay in 2019 would take 58 percent of 2024 take home pay, assuming median growth in incomes.¹⁰ This implies that an individual would need to cut non-housing expenses equivalent to a fifth of their take-home pay to buy a home and maintain a balanced budget. This is a substantial challenge, made worse by rising housing-related costs beyond principal and interest. Higher home prices increase downpayments and property taxes, while closing costs and other expenses, like insurance, represent additional hurdles to homeownership.

The 2019−2024 period comprises sub-episodes of economic volatility that had distinct effects on affordability. Figure 3 shows change over time in the budget share required to support an equivalent home purchase from a starting point of December 2019. The figure compares the 2019−2024 trend with the same measures applied to 2014−2019 for reference. 

During 2020 and the first half of 2021, affordability improved—income gains outstripped monthly mortgage expenses—supported by falling interest rates from stimulative Federal Reserve policy and fiscal income supports during the pandemic. Deterioration began quickly thereafter, as government income supports faded, while inflation and interest rates rose sharply. The implied mortgage expense budget share rose sharply from mid-2021 through 2023. Since the start of 2024, affordability metrics improved marginally, but stood at very challenging levels relative to 2019. Appendix I provides additional detail, jointly showing both income growth and mortgage cost growth over time. 

Over the comparison period of 2014−2019, the monthly cost of a mortgage payment roughly matched median growth of income for individuals in our sample. In effect, individuals attempting to save for years to buy a home ended up with a housing market that had neither surprised to the upside nor downside, on balance. 

The increase of about 50 percent in home prices 2019−2024 implies a proportional rise in down payments, all else equal. This is a higher hurdle to clear for first-time homebuyers and lower-income individuals generally: they are less likely to have buffers built from previous increases in the housing or stock markets, and they have less wiggle room to lower down payment percentages.¹¹ While substantial growth in retail investing during the pandemic lessens these disparities, purchases of stocks by lower-income individuals came relatively later, meaning they benefited to a lesser degree from the roughly three-fold rise in the S&P 500 index over the 2010s and the near doubling from December 2019 to December 2024 (Wheat and Eckerd 2024b). These patterns echo recent findings on rent inflation, where rising housing costs similarly constrained spending and worsened financial vulnerability, especially among younger, lower-income households. 

Figure 4 shows the share of individuals aged 25−44 that save a portion of their take-home pay in financial assets, mainly stocks, across the income distribution.¹² The investing share over 2019−2024 was just under 35 percent for individuals in the middle-income quintile (40−59.9 percentile), compared with approximately 70 percent for top-5-percentile earners—a 2X difference. The investing shares for the earlier 2014−2019 period were 15 percent for the middle-income quintile versus 55 percent for the top—a more than 3.5X difference. 

As noted in an Institute note on wealth inequality, the reduced gap in stock market participation observed more recently may reduce the potential for further rises in stocks to exacerbate wealth disparities. However, the broadening in stock market participation since December 2019 comes during a time of relatively high valuations. Earlier investors—which were even more tilted towards high-income individuals—benefited to a greater degree from the expansion in stock market valuations.¹³ In short, lower-income individuals are relatively less likely to keep up with rising downpayment needs by using past stock market gains, either because they don’t own any stocks, or they made purchases at higher valuations.

Market forces and policy have long created differences in housing markets across urban, suburban, and rural neighborhoods.¹⁴ For would-be first-time homebuyers, shocks to local housing demand are a source of risk: an unforeseen demand increase in a person’s preferred neighborhood, for example, can lead to price increases that make the location unaffordable. The pandemic-era provides a prime example. Densely populated urban centers faced decreases in demand, while certain suburban and rural housing markets saw demand increases, according to academic research (Ramani and Bloom, 2022; Mondragon and Wieland, 2022).¹⁵ Changes in relative home prices in the wake of the pandemic have persisted. 

Given the importance of local factors in driving differences in affordability over time, Figure 5 shows changes in home prices relative to median income growth broken out across categories capturing urban city centers, suburbs, and rural areas. We use ZIP code-specific measures of median income growth and compare against data on local home price changes from Zillow. This method better captures welfare shocks that come with changes in local housing market conditions. For example, aggregate local statistics may mistake a relative shift in incomes of people moving to an area for income growth. 

Changes in affordability by area types are somewhat larger than what can be known from prices alone. Gaps are largest between the urban core of large cities and the rest of the housing market. For metro areas with over one million residents, the urban cores experienced the highest income gains but the lowest home price growth. The less-densely-populated areas (suburbs and rural areas) saw both higher home price growth and slightly lower income growth. Small metro areas (500,000 to 1 million people) experienced the worst declines in affordability. Despite similar income growth to the rest of the country, small metros saw the largest increases in home prices. Outside of major MSAs, suburbs experienced relatively worse trends in affordability than the other ZIP code categories.

Appendix II shows results from a regression framework that uses additional factors—including IRS data on shifts in income and wealth of tax filers in each ZIP code—to explain differential trends in home prices across locales. Our findings are consistent with Ramani and Bloom (2022), which explores remote work as a driver of housing demand that led to increases in home prices in certain areas. Additionally, home prices relative to local incomes, an indication of overvaluation, played a significant role. Areas with inflated home values compared to incomes tended to experience relative declines in home prices, all else equal. 

Conclusion and implications

Housing affordability reached a historically low level due to increases in home prices and interest rates relative to incomes from 2019 to 2024. Unique data used to measure income changes at the individual level more precisely capture the financial challenge facing young individuals seeking to become first-time homebuyers. To buy a comparable home, a household would have to increase its budget allocation to a mortgage payment by 45 percent as of December 2024, compared to December 2019. Lower-income individuals, who have lower rates of homeownership and stock market participation, have been relatively disadvantaged by the rise in asset prices. Geographic differences in affordability dynamics have been notable since the pandemic. The densely populated urban centers of large cities have experienced notably less home price appreciation relative to the suburbs and more rural areas.

The decline in housing affordability is feeding into disparities in financial health. A study of retail spending broken down by homeownership status shows divergent trends (Hacıoğlu et al., 2024). Those who owned homes prior to the pandemic have seen higher growth in spending since the start of 2023 than renters or those that became homeowners during 2022, when prices and interest rates were already rising. Perceptions of overall economic conditions may also be influenced by housing affordability challenges, as suggested by recent academic research.¹⁶ However, it’s challenging to disentangle housing from related factors, such as higher consumer prices and interest rates, generally.

Finally, shocks to home prices and interest rates described in this report carry implications for wealth inequality. High inflation and rising asset prices, as seen during the 2019–2024 period, can widen gaps due to balance sheet differences across the population. Those that invest a significant amount of their wealth in stocks and housing tend to benefit, while households preserving savings in lower-risk assets, like cash, often see the value of their holdings depreciate in terms of purchasing power. The specific nature of the pandemic added another fault line: existing homeowners in the suburbs benefitted relative to homeowners in dense urban centers. These findings can help make sense of the differences in households’ experiences in the wake of the substantial volatility in recent years.

Appendix I: Timeline of income and mortgage cost growth over 2014−2019 and 2019−2024.

Figure 6 below provides additional detail, tracking separately the paths of the primary components of affordability: mortgage costs and incomes. Analogous to the discussion in Figure 3 from Finding 1, the figure shows a favorable trend from December 2019 through December 2021—when nominal incomes were up 25 percent against a 20 percent increase in implied mortgage expenses. In 2022 and 2023, the cumulative growth in housing costs rose about 100 percentage points against a further rise in the income growth rate of just over 10 percentage points. By contrast, the 2014−2019 period featured minimal gaps between income and mortgage cost growth. 

Appendix II: Factors associated with local affordability changes

This section presents results from ZIP code-level regression analysis of factors that are associated with home price growth. In addition to the new perspective available with within-person income growth measures, ZIP code-level data from the IRS Statistics of Income provide additional clarity on affordability changes.¹⁷ We use a regression framework to quantify the relationships between home prices and income growth, in conjunction with public data aggregated at the ZIP code-level. Specifically, we model changes in local home prices from Zillow as a function of the following: JPMC data on median individual-level income growth of residents at the start of the period and IRS data showing net changes in tax filers and changes in the income and wealth level of residents. We also use a proxy for housing market valuation at the start of the period using home prices (Zillow) relative to average per-person income (IRS). As a caveat, the variables used in the regression exercise may be noisy indicators of the factors that really matter for home prices.

Individual-level income growth, the income variable used in the main analysis of affordability, shows up with the expected positive relationship with local home prices. The regression explains roughly 45 percent of the variation in home price changes 2019−2024. However, this represents roughly a 15-percentage point drop from the baseline 2014−2019 period, suggesting an increased role of other factors in driving recent changes in home prices. Still, the regression may be useful for gauging the relevance of the following dynamics, summarized in Figure 7 and the regression table below:

• Population shifts. The regression shows an intuitive connection between population change (interpreted as a demand signal) and home prices: areas with population increases saw larger home price appreciation.
• Affluence of movers. Areas shifting in composition toward individuals with relatively higher incomes and higher wealth experienced larger price increases, all else equal. As a proxy for shifts in the income of individuals living in an area (apart from individual-level growth), we use the percent change in per-person gross incomes, after controlling for our internal data on individual-level changes in income. Similarly, for changes in wealth of residents, we use the change in the share of the tax filters reporting dividend income.
• Valuation. Home values increased by less in areas with high prices in 2019 relative to the average income of residents at that time. That is, relatively unaffordable areas—using pre-existing prices and incomes—had lower subsequent home price appreciation, after controlling for other factors.