Counting the Homeless. Searching for Solutions.
During the last 10 days of January, public officials, nonprofit employees, and volunteers in thousands of communities will tromp through alleyways, parks, and culverts to conduct the 2020 annual census of the homeless. Their efforts, combined with tallies of the people in shelters, will produce the “Point in Time” counts that tell us the extent of homelessness in America.
The 2019 census data were released by the U.S. Department of Housing and Urban Development, in the agency’s Annual Homeless Assessment Report to Congress, on January 7. The report found there were about 568,000 homeless people nationwide on a winter’s night in January 2019. Nearly two-thirds of the homeless were in shelters or transitional housing programs. Over one-third were “unsheltered” — sleeping in cars or abandoned buildings, or camping on sidewalks and in parks.
While far from perfect, these data provide the best available picture of regional and national trends in homelessness. Nationally, the trend is troubling: the homeless population increased for the third year in a row, and the 2.7 percent increase in homelessness from 2018 to 2019 is the largest yet.
Unfortunately, there has been significant centralization of the approach to homelessness under a national initiative “to end homelessness.” The centerpiece of this federal solution is the Housing First model, where social workers offer free short or long-term housing quickly and without behavioral conditions pertaining to substance abuse or counseling services.
From 2009 to 2018, targeted federal homelessness spending increased dramatically, from around $2.5 billion to nearly $6 billion a year. Communities receiving this funding have faced strong financial incentives to prioritize the Housing First approach. The funding helped to double the amount of subsidized housing for the homeless. Nationwide, Housing First programs added permanent housing units and social services for 142,000 people and “rapid rehousing” for 100,000 people.
Meanwhile, unsheltered homelessness only declined by 32,000 people.
Researchers, including myself, have studied these trends and found limited policy impact. Kevin Corinth estimated that it takes roughly 10 additional permanent housing units to reduce homelessness by a single person in a community. This implies the need for nearly 5.7 million additional beds to house the 568,000 people who are currently homeless—not including those who may become homeless in the future.
In another study, I found communities that received more federal homelessness funding did not have fewer unsheltered people than localities that received less, although they had more sheltered homeless.
For many experts, these results come as a surprise. They heralded Housing First as an “evidence-based policy,” a “proven” and even “cost-effective” solution that all communities ought to prioritize. Yet, large increases in funding for programs like Housing First have yielded very small reductions in homelessness. How could this be?
For one, Housing First offers a solution that does not consider the many different factors that cause homelessness across communities. For instance, there are generally more unsheltered homeless in states with milder climates. In our research, Corinth and I found the 10 states with the warmest winter temperatures accounted for 67 percent of the unsheltered homeless population in 2017. The unsheltered homeless are different from people who enter shelters; communities need to tailor their efforts accordingly.
We also found significant variation in homelessness in communities with similar climates, suggesting the importance of state and local policy. Restrictive local land use regulations reduce the availability of affordable housing and, in turn, positively predict homelessness. Unfair eviction practices can create housing instability. Addressing these and other local policy issues can to unleash market forces to fill unmet demand for affordable housing.
Housing First emphasizes residential placement — often for the long-term — after people become homeless. But for many families, homelessness is a brief, one-time, and non-recurring event; in these cases, well-targeted prevention programs providing temporary financial assistance have shown promise. Homeless spells can also be preempted through community bonds. Corinth and Claire Rossi-de Vries found that strong social ties to family, friends, and religious communities reduce a person’s risk of experiencing homelessness by as much as 60 percent. Yet, Housing First downplays the importance of relationships and community.
Furthermore, Housing First does not allow organizations to put conditions on program participation. However, not everyone recovers from homelessness in the same way. Many previously homeless individuals believe the behavioral expectations in non-Housing First programs helped them successfully exit from homelessness.
If the Housing First approach were the solution to end homelessness, more money would solve the problem. But it is increasingly clear there may not be a single, universal solution to homelessness in every community.
An alternative to a single, top-down approach is to shift our focus toward innovative solutions tailored to local conditions. Communities might work to loosen restrictive housing regulations. Or they might put existing resources toward novel solutions like New York City’s Doe Fund, which emphasizes work program participation, or Community! First, a “tiny home village” outside Austin. If local organizations are given the freedom, they can support their successes and learn from their mistakes.
We won’t know for many months whether the January 2020 “point in time” count will find more or less homelessness. Yet, we do know that effective solutions to the diverse problems of homelessness are waiting to be discovered.
David S. Lucas is a Research Fellow with the Institute for an Entrepreneurial Society and Assistant Professor of Entrepreneurship in the Whitman School of Management, Syracuse University.