“The magnitude of the racial and ethnic disparities is striking,” said Vincent Fusaro, a co-author of the study and a professor of Boston College, in an email.
Homeless advocates and activists have for years attacked the federal government’s estimates of the national homeless population, accusing it of dramatically understating the scope and severity of the problem.
The Department of Housing and Urban Development conducts an annual study that captures “point in time” estimates of America’s homeless population — a snapshot of how many people are living in homelessness when the study is conducted. Fusaro and his colleagues say their work is the first since the 1990s that captures whether someone has ever been homeless, using a recent survey of older Americans by the Institute for Social Research at the University of Michigan.
The latest HUD survey found that about 550,000 Americans — roughly 0.17 percent of the population — were homeless in 2017.
The 6 percent number suggests that about 19 million Americans today will be homeless at least once in their lives, assuming the experience of the baby boomers is roughly comparable to other generations, creating a starkly different impression of the pervasiveness of the nation’s homeless population.
Research from Dennis Culhane, a professor at the University of Pennsylvania, suggests that the baby boomers had higher rates of homelessness compared to other generations, according to the study’s authors. But is unclear if homeless rates have gone up or down among post-boomer generations, given the mortgage crisis of the mid-2000s that cost tens of thousands of Americans their homes, the researchers said.
The HUD numbers and the researchers' new study use different methodologies. The HUD survey estimates how many people on one night are sleeping outside, in places like cars or abandoned buildings, or in a shelter. By contrast, the researchers’ new study is based on asking respondents if they have “ever been homeless or lived in a shelter,” relying on their understanding of those terms.
“A one-night snapshot tells us only how many people are experiencing homelessness at one time, not how many have ever experienced homelessness,” said Fusaro, who co-authored the study with Helen Levy and H. Luke Shaefer, two researchers at the University of Michigan. Fusaro pointed out that the formerly homeless can suffer adverse physical and mental health consequences long after they have secured housing. “If homelessness has longer-term consequences, a lifetime estimate is an important complement to our homelessness numbers.”
Others worried that the definition used by the researchers was overly broad, and risked over-estimating the extent of homelessness.
“Most literature would define homelessness as either being on the street or being in a shelter, and they should not rely on how respondents define it," said Robert Rector, a research fellow at the Heritage Foundation, a conservative think-tank. "You don’t know what you’re getting there.”
The study looked at those born between 1946 and 1964, aged 47 to 68, but is essentially consistent with other attempts to get a more comprehensive picture of national homelessness rates — including a 2009 paper looking just at lifetime homeless rates among young adults, as well as a 1994 study that found slightly lower estimates of lifetime homeless rates.
But there are some limitations to these researchers’ approach as well, including the most glaring: It omits everyone who was homeless but died before reaching the late-middle age population studied by the researchers. A study from 2006 found that homeless people on average die at 50, while people who are not homeless die on average at 78.
The consequences of homelessness are particularly acute for black Americans, and appear to correspond at least in part with higher poverty rates among black Americans. Roughly one in four black Americans are in poverty compared to roughly one in 10 white Americans, according to the U.S. Census.
“This study’s findings create new urgency to identify federal, state, and local policy responses to alleviate and prevent such racial disparities,” said Diane Yentel, president and CEO of the National Low Income Housing Coalition.
No other variables tested by the researchers found as dramatic of a gap, though level of education — often used as a rough stand-in for class — was unsurprisingly also an important factor. About three percent of college graduates in the baby boomer cohort had been homeless at some point, compared to eight percent of those without a college degree, the researchers found.
About 10 percent of U.S. military veterans in the baby boom cohort have been homeless, almost double the national rate, according to the study.
International comparisons of homelessness are difficult to draw, because each country has its own methodology for reporting homeless rates and its own definition of homelessness. Austria, for instance, only counts “registered homeless” — those staying in homeless shelters — while some but not all countries include those temporarily living with friends and relatives.