The U.S. homeownership rate remains lower than it has been for more than 20 years, even though housing markets have largely recovered from the Great Recession (U.S. Census Bureau 2017). Most of the drop in homeownership is due to fewer renters choosing to purchase first homes than prior to the crisis. Researchers and policymakers have posited several possible reasons for the apparent shift in behavior, including:
1. Increased regulation of mortgage lending and stricter underwriting criteria.
2. Weak labor markets for young workers, leading them to delay household formation and homeownership;
3. Millennials’ lower preferences for owning rather than renting; and
4. High levels of student loan debt among younger households.
(Dettling and Hsu 2014; Gabriel and Rosenthal 2015; Larrimore et al 2016)
To understand the causes and policy implications of lower homeownership rates, we need to understand the decisions and behaviors of individual households, especially those of prospective first-time homebuyers. In this analysis, I describe the demographic and economic characteristics of prospective homeowners, using household-level data from the most recent American Housing Survey (AHS 2015). Because we can’t identify “true” prospective owners until after they purchase a home, I use new homeowners as a proxy: households that purchased their current home within the last two years and previously were renters. For context, I compare them to more established homeowners and to renters. To give some historical perspective, I also compare new owners in 2015 to new owners in 2001 (a reasonable benchmark year for housing markets, pre-dating the housing boom and bust). To conclude, I discuss three possible policy implications.
New homeowners mostly fit the “family with kids, house with yard” model
Conventional real estate wisdom is that many renter households choose to purchase their first home around the time they experience major life changes, especially getting married and having children. Consistent with these predictions, new homeowners in 2015 had larger household sizes and were more likely to have children under age 18 than either renters or established homeowners (Table 1). The median new homeowner was 42 years old, similar to renters and significantly younger than established owners.
Table 1: Household type and housing unit characteristics, 2015
|New owners||Established owners||Renters||All households|
|Number of people||3.09||2.49||2.37||2.47|
|Percentage of kids under 18||42%||27%||33%||30%|
|Median age (years)||42||57||41||51.77|
|1-4 family buildings||88%||90%||56%||78%|
Source: American Housing Survey
The housing units occupied by new homeowners are quite similar to those of established owners. Both new and established owners overwhelmingly live in buildings with fewer than four housing units, while renters are more likely to live in multifamily buildings. Matching the somewhat larger family size, new owners live in larger housing units than renters, measured by the number of bedrooms.
Like the U.S. population, new homeowners are older and less white than they used to be
New homeowners in 2015 were noticeably older than those in 2001, when the median age of new owners was 34 (Figure 1). Some of this is due to the general aging of the U.S. population–renters and established owners were also older in 2015 than the same groups in 2001–but the age distribution has changed more dramatically for new owners. In particular, the share of new homeowners under age 30 declined from 29 percent in 2001 to about 15 percent in 2015. During the same time, the share of all households under age 30 declined slightly from 13 percent to just under 10 percent.
The race and ethnicity of new owners has also changed over time (Figure 2). Between 2001 and 2015, the share of new owners who were non-Hispanic whites dropped by four percentage points, from 71 percent to 67 percent, while the share of Hispanic new owners rose from 11 percent to 16 percent. Renters were the most racially and ethnically diverse group in both time periods, and have become more Hispanic over time. Established owners in 2015 remained mostly non-Hispanic white, although the group saw small increases in the share of blacks, Asians, and Hispanics since 2001.
Most new homeowners don’t live in newly built housing
The age of housing units occupied by new owners–and by all households–has changed substantially since 2001 (Table 2). New owners live in slightly newer houses than established owners or renters, although most new owners do not live in newly built homes (defined here as less than five years old). In 2015, 7.5 percent of new homeowners and 3.3 percent of all households lived in newly built housing. By contrast, in 2001, 25 percent of new homeowners lived in newly built housing, as did 8.5 percent of all households.
Table 2: Housing unit age, 2001 and 2015
|New owners||Established owners||Renters||All homeowners|
|Year built (average)||1971||1969||1965||1968|
|Housing built last five years||7.5%||2.8%||3.4%||3.3%|
|Year built (average)||1970||1963||1959||1962|
|Housing built last five years||24.9%||10.7%||6.2%||8.5%|
Source: 2015 American Housing Survey
Two factors contribute to this change over time. First, new housing production slowed to a crawl during the Great Recession and recovery. From 2010-2012, total housing starts nationwide fell below levels observed since 1960, before gradually picking up (Federal Reserve Bank of St. Louis). Second, many of the housing units built between 2010 and 2015 were in apartment buildings intended for renters, while in the earlier period, single-family homes represented a larger share of new construction. To illustrate how this changes the occupancy of new housing, consider that, in 2001, 16 percent of newly built housing units were occupied by renters. In 2015, 38 percent of newly built homes were occupied by renters. If most prospective homebuyers still want to live in single-family homes, then increasing homeownership rates may require higher levels of production of single-family homes–particularly at prices within reach of first-time buyers.
New homeowners remain more affluent and more college-educated than renters
By income and educational attainment, new homeowners are much more similar to established homeowners than they are to renters (Table 3). The median income of new homeowners ($69,000) was nearly double that of renters ($34,000), and slightly higher than that of established owners. Similarly, new homeowners had completed more education than most renters, with nearly 40 percent of owners having a college degree, compared to 26 percent of renters. These differences are not surprising, given that households with higher, more stable incomes and more assets can more easily accumulate down payments and qualify for mortgages. It is notable that about one-third of renter households had incomes over $50,000, and one-fourth had college degrees; these could represent near-term prospective owners.
Table 3: Income and educational attainment
|New owners||Established owners||Renters||All homeowners|
BA or above
Source: 2015 American Housing Survey
Income and educational differences between new owners, established owners, and renters have not changed significantly between 2001 and 2015 (inflation-adjusted incomes of all three groups have remained essentially the same).
Three policy implications
Encouraging homeownership has long been an explicit policy goal of the U.S. government, mostly implemented through the federal tax code and regulation of residential mortgage markets. The changing characteristics of prospective homeowners presented here raise several questions about whether current policies will be effective at increasing homeownership–and whether they adequately support the growing share of renter households.
1. Because homeowners are wealthier than renters–a trend that has not diminished over time–policies that subsidize homeownership are fundamentally regressive. Both the mortgage interest deduction and the federal income tax deduction for local property taxes primarily benefit higher-income, better-educated households. By contrast, tax deductions for housing expenditures irrespective of tenure (and capped at some dollar value or percentage of income) would provide greater benefits to middle income families.
2. Housing is by far the largest asset for most U.S. households, and homeownership is often promoted as a wealth-building strategy. The later in life that people become homeowners, the shorter the period of time they have to build equity in their homes. Crafting policies to explicitly reduce the age of first-time buyers will likely be difficult. Alternatively, we should develop and support other mechanisms for wealth-building, which would benefit both young households and long-term renters.
3. Despite much media attention to millennials’ supposed preference for high-density urban living, the data suggests that most new homeowners still purchase single-family houses. The dearth of new housing development during the Great Recession and recovery–and the scarcity of new single-family homes in particular–may constrain both first-time homeowners and established homeowners looking to trade up.
Dettling, Lisa J. and Joanne W. Hsu. 2014. Returning to the Nest: Debt and Parental Coresidence Among Young Adults. Finance and Economics Discussion Series 2014-80. Washington DC: Board of Governors of the Federal Reserve System.
Federal Reserve Bank of St. Louis. Housing Starts: Total: New Privately Owned Housing Units Started. https://fred.stlouisfed.org/series/HOUST
Gabriel, Stuart A. and Stuart S. Rosenthal. 2016. The Boom, the Bust, and the Future of Homeownership. Real Estate Economics.
Larrimore, Jeff, Jenny Schuetz and Sam Dodini. 2016. What are the Perceived Barriers to Homeownership for Young Adults? Finance and Economics Discussion Series 2016-021. Washington DC: Board of Governors of the Federal Reserve System.
U.S. Census Bureau. 2017. Quarterly Residential Vacancies and Homeownership, Second Quarter 2017. https://www.census.gov/housing/hvs/files/currenthvspress.pdf