The U.S. homeownership rate stood at 64.2 percent in the first quarter of 2019, an improvement from its level between 2015 and 2018, but still significantly below its peak of 69.2 percent in 2004. The weaker homeownership rate is partly due to housing affordability challenges facing the market today. But it could also be the case that more households prefer to rent than before, or they are moving in and out of homeownership over shorter periods of time to meet their needs. Considering this environment, an alternative measure of homeownership demand may be the lifetime homeownership rate, which measures the rate at which people have been a homeowner at any given point in their lives.
To approximate the lifetime homeownership rate and examine how long homeowners stay in homeownership, this special study uses the National Longitudinal Survey of Youth (NLSY) 1979, a survey administered by the Bureau of Labor Statistics. The NLSY 1979 survey is a nationally representative sample of 12,686 men and women who were ages 14 to 22 in 1979, and tracked them on an annual basis until 1994, and a biennial basis after 1994. The last year in which survey data is available from the NLSY 1979 is 2016. The purpose of the survey is to gain insight into the population’s labor market patterns, as well as significant events that happen over the course of their lives.
Included in the NLSY 1979 is a marker that flags homeownership on a once every four-year basis starting in 1980 and running through 2016. To approximate the lifetime homeownership rate, we use the NLSY 1979 by examining homeownership markers between 1980 to 2016. Almost all of the NLSY cohort homeowners – 91 percent – were between the ages of 18 and 22 in 1980, and range from age 51 to 59 in 2016. Although not the perfect measure of a lifetime homeownership rate, studying the NLSY provides the best approximation as it tracks individuals for nearly 40 years.
Figure 1 (click on image to access graph) shows that 90.5 percent of individuals in the NLSY 1979 cohort were homeowners at some point between the period of 1980 and 2016, while 9.5 percent were not. The lifetime homeownership rate provides evidence that, even with a homeownership rate currently hovering around 64 percent, a much higher share of the population has owned a home at some point in time.
Analyzing the lifetime homeownership rate by demographic groups shows a different story, however. Figure 2 (click on image to access graph) displays lifetime homeownership rates from the NLSY 1979 by race and ethnicity. Black individuals in the survey have a lifetime homeownership rate of only 74.5 percent, compared to 87.8 percent among Hispanics/Latinos, and 93.9 percent of whites.
The low black lifetime homeownership rate indicates that homeownership has been out of reach for a significant share of the black population, due to factors such as lower incomes, marriage rates and discrimination. A notably larger share of Hispanic/Latinos have attained homeownership, but have dropped out at higher rates, as illustrated by comparing the large percentage point gap between their lifetime rate of 87.8 percent and their homeownership rate, which has hovered around 44 to 50 percent for the last 15 years. It is important to point out that almost all of the white population in the NLSY 1979 has achieved homeownership, and has the smallest gap between their lifetime rate of 93.9 percent and their homeownership rate of 73.1 percent (from the second quarter of 2019).
Differences in lifetime homeownership rates, although less extreme, can also be seen when analyzing a mother’s level of education (Figure 3, click on image to access graph). Previous research has shown that maternal education has a significant impact on a child’s (and future adult’s) socio-economic, cognitive, and health outcomes.A mother with lower levels of education is more likely to experience higher levels of economic insecurity, and suffer from mental illness, such as depression, making it harder to properly care for and achieve better outcomes for a child.
Figure 3 shows that 86.2 percent of those with mothers who never started high school have been homeowners at some point between 1980 and 2016, compared to 88.5 percent of those with a mother who has less than a high school degree, and 92.7 percent of those with a mother who has a high school degree or more, indicating a positive relationship between maternal education levels and a child’s lifetime homeownership rate.
Survival in Homeownership
The concept of lifetime homeownership rates leads to the question of once one becomes a homeowner, how long do they sustain homeownership? The second part of this homeownership analysis uses survival analysis to track how long the NLSY 1979 cohort lasts in homeownership by the year in which homeownership was first detected (Figure 4, click on image to access slideshow).
A notable finding from Figure 4 is that, survival rates gradually increase, starting with the 1980 group up until the 1992 group (survival curves shaded in blue). For example, among homeowners who started homeownership in 1980, only 53 percent survived in it after eight years (1988). But larger shares of groups that become first-time homeowners in later years maintain homeownership for longer periods. For example, the share who sustained homeownership after eight years increases to 67 percent among those who started homeownership in 1984, to 77 percent among those who started in 1988, and to 93 percent of those who started in 1992.
This gradual increase in survival is most likely associated with age. In 1980, the 1979 NLSY cohort were 23 years old at most, far younger than the typical age of first-time homeowners at the time (first-time homeowners had a median age of 30 between the years of 1980 to 1984). A homeowner at the age of 23 most likely does not have the income level or other resources to last long in homeownership compared to an older first-time homeowner.
It is also important to note that the U.S. experienced a recession in 1980, brought on by unusually high real interest rates engineered by the Federal Reserve to fight high inflation. This most likely had a negative effect on survival rates among those who started homeownership in 1980.
Homeownership survival rates start to fall beginning with those who entered homeownership after 1992 (survival curves highlighted in red). For example, after one period (four years) in homeownership, 84 percent of those who started homeownership in 1996 survived, compared to 83 percent of those who started in 2000, and only 58 percent of those who started in 2004.
It is important to note that this time period – roughly 1996 to 2004 – coincided with larger shares of minority groups entering homeownership. For example, NLSY data show that 17 percent of all homeowners who entered homeownership in 1992 were either black or Hispanic, but the share grows to 32 percent among those who entered in 2004. It may be the case that during this period fewer black and Hispanic/Latino homeowners were able to survive in homeownership, thus driving down the overall survival rates.
It is also important to point out the steeper survival curve slopes between the period of 2004 and 2012 in Figure 5 (a different view of Figure 4, but by actual year) particularly for groups who entered homeownership later (survival curves highlighted in red). Groups that entered homeownership later were most likely hit the hardest by the Great Recession, as they had less opportunity to build equity in their homes compared to groups that entered homeownership at earlier dates (survival curves highlighted in blue). For example, groups that entered homeownership before, such as the 1984 and 1988 groups, have relatively flat slopes, indicating that they were better able to weather the storm than groups who entered homeownership closer to the crisis.
Moreover, home prices started to fall in 2006, after rising significantly in the early 2000s. Individuals who entered homeownership just prior to 2006 probably experienced distress as price declines reduced much of their equity.
Another point of interest in Figure 5 is that the 2004 and 2012 groups have relatively steep curves, compared to 2008 group, which has a flatter curve. It could be the case that looser mortgage underwriting standards contributed to the steep curves among the 2004 and 2012 groups, but tighter underwriting standards immediately after the financial crisis affected the 2008 group the most. Another possible reason is that individuals simply changed their tenure habits over this time period.
Homeownership Survival Rates by Race and Ethnicity
Figure 6 breaks down homeownership survival rates by race and ethnicity, as well as the year in which homeownership was first registered (click on image to access slideshow). In general, black homeowners in the NLSY 1979 are the group least likely to survive in homeownership over time, while white homeowners have significantly higher survival rates. Hispanic/Latinos homeowners, for the most part, fall somewhere in between, except for Hispanic/Latinos who entered homeownership in 2000. That group had survival rates significantly below that of both black and white homeowners.
Differences in survival rates by race and ethnicity are relatively extreme for those who entered homeownership in 1980, but the gap narrows somewhat for those who start homeownership in 1988. For example, among those who started homeownership in 1980, 52 percent of white homeowner last in homeowners maintained homeownership after 16 years (1996), 40 percentage points above the share of Hispanic/Latinos (12 percent), and 44 percentage points above black home owners (eight percent). But when examining survival rates after 16 years (2004) among those who started homeownership in 1988, the gap narrows to 20 percentage points at most between white homeowners (72 percent), and Hispanic/Latino (53 percent) and black homeowners (52 percent).
After the 1988 start year, survival rate gaps by race and ethnicity started to diverge again. The wider gap between white and black homeowners is particularly noticeable for some groups when looking at the period of 2004 to 2008, during which the financial crisis occurred on the tail end. For example, among those who entered homeownership in 1996, white homeowners experienced only a two-percentage point decline in survival between 2004 and 2008 (81 percent to 79 percent), while black homeowners saw a 12-percentage point drop (51 percent in 2004 to 39 percent in 2008).
It has been documented that black homeowners were more likely to buy homes in areas with larger declines in property values during and after the recession. This put them at higher risk for foreclosure than white homeowners during the Great Recession and its aftermath, which is most likely one of the reasons contributing to the lower survival rates among black homeowners.
Almost all of the NLSY 1979 cohort – 90.5 percent – have been a homeowner at some point between 1980 and 2016. But analyzing lifetime homeownership rates by race and ethnicity shows that black and Hispanic/Latino individuals have lower lifetime homeownership rates compared to the white population. Individuals with mothers who have less than a high school degree, or never started high school have lower lifetime homeownership rates, compared to individuals with mothers who have a high degree or more. This study also shows that black and Hispanic/Latino homeowners have lower homeownership survival rates, compared to white homeowners.
The NLSY codes uses three race/ethnicity codes: black, Hispanic; and non-black/non-Hispanic. NAHB analysis of the 2017 American Community Survey (1 year) shows that a majority of non-black, non-Hispanic homeowners – 83 percent -- are white. Therefore, for the purposes of this survey the non-black, non-Hispanic homeowners will be referred to as white homeowners.
This analysis only tracks how long they stayed in homeownership from when they first became homeowners until when they dropped out for the first time; it doesn’t not track if they repeatedly entered and dropped out of homeownership.