How Can the U.S. End Homelessness?

Giving people access to support services and a place to stay can reduce the number of those living on the streets. But can that be done affordably?

Richard Levine / Corbis / Zak Bickel / Kara Gordon / The Atlantic

A&Q is a special series that inverts the classic Q&A, taking some of the most frequently posed solutions to pressing matters of policy and exploring their complexity.

On any given night in the United States, half a million people are homeless. Some of them sleep in shelters, others on the streets; roughly one-quarter are children.

About 15 percent are so-called chronically homeless, which means they haven’t had a permanent home in years, and often cycle through jails, hospitals and homeless shelters in search of a place to lay their heads.

The government has tried to tackle the problem of homelessness on nearly every level, but comprehensive solutions have proven elusive, despite billions spent over time. The federal government has set a series of goals of ending homelessness for veterans by 2015, chronic homelessness by 2017, and homelessness for families with children and youth by 2020. But reaching these benchmarks appears to be much further off.

Seemingly every policy group that works on this issue has ideas about how to solve it for good. But what really works to help people get—and stay—off the streets? And is there any way to do it that wouldn’t be wildly expensive?


First we need to consider space. We should increase the number of shelters available to the homeless.


Can we count on that as a long-term solution?

Shelters are certainly useful in that they provide beds and roofs to people who don’t have them, especially on cold and rainy nights where sleeping outside could be fatal for some.

But shelters are incredibly expensive to operate. Nationally, the average monthly cost of serving a family in an emergency shelter is $4,819. Providing them with a voucher for housing, on the other hand, is just $1,162. Shelters might be good for emergencies, but does having a bed to sleep in mean that someone has a home?

And quality can be an issue for these shelters: Many homeless people have told advocates trying to get them off the streets that they avoid shelters if they possibly can. They’ve heard about bad experiences there, or have themselves suffered through violence, theft, or other trauma in these ostensibly safer spaces. There were 826 “violent incidents” in New York City homeless shelters last year, including sexual assault and domestic violence, according to the New York Daily News.

People often have to leave food and other belongings behind when they check into a shelter, making it hard to accumulate anything of sentimental or material value. Plus, shelters don’t allow residents to develop a sense of permanency—and it’s permanency that helps people get a job or stay sober, as numerous studies have indicated.


Affordable housing would be a longer-term solution. Let’s just increase the number of these units overall. If more people can afford housing, they won’t be homeless.


If it were that easy to add more affordable housing, cities like New York and San Francisco would be very different than they are now, and far less expensive. It’s costly to build new apartments and homes in cities where land is pricey, and developers want to recoup their investment as soon as possible, which means they have to charge a lot for rent.

That’s not to say cities, states, and the federal government haven’t tried out a few strategies for hastening the construction of affordable housing. Have any of them been effective?

There are federal Low Income Housing Tax Credits that help certain developers build 100 percent affordable housing. But developers compete for those tax credits, and there aren’t enough to held build affordable housing for all the people who need it, much less for those who don’t have homes in the first place.

Inclusionary zoning policies can help create more affordable housing; in places such as Montgomery County, Maryland, for instance, all new apartment buildings with more than a certain number of units have to set aside a few of them to be designated as affordable housing, priced much lower than market rent. But then developers usually have to pass the costs of that lost rent onto the other tenants, which increases market-rate rent.

In most municipalities, inclusionary zoning is voluntary, which means that developers who include affordable units can skirt some regulations, allowing them to build higher, for instance, or make their buildings denser.

Making this kind of zoning mandatory can be tricky, though, because developers argue that they can’t charge enough for market-rate units in low- and middle-income neighborhoods to subsidize the affordable units. In March, New York City made inclusionary zoning mandatory in some neighborhoods, and developers are already complaining that it’s become harder to build affordable housing in the city.


Okay, so if we can’t rely on affordable housing units alone,  it sounds like we should find other methods to give homeless people special access to permanent homes. That would be better in terms of lasting success, boosting their chances of landing and keeping a job, and maybe helping those who struggle with addiction to stay sober.


This is what’s called a housing-first approach, and numerous studies have found it’s much more effective than relying on shelters. Housing-first places homeless people in long-term housing without asking them to get sober or hang onto a job first. After they’re settled in a stable home, they gain access to services such as drug and alcohol treatment, an assigned social worker, or job training. They don’t have to take advantage of those services, but most people chose to do so.

Through housing-first, Utah reduced its chronically-homeless population 72 percent between 2005 and 2014. Just having a roof over their head, a permanent address, and a place to prepare food and store belongings made so much of a difference for people that the director of the state’s Housing and Community Development Division told the Washington Post that the number of chronically homeless was “approaching a functional zero.”

But why should homeless people get a free place to live? There are lots of people who need affordable housing but aren’t currently homeless, after all. And housing-first isn’t cheap—though tenants pay a small portion of their rent, the state or city usually picks up much of the tab. A voucher for a housing program, like Section 8, can cost $1,162 a month, and spending that money means fewer people get rental assistance overall.

When long-term housing is hard to come by, people desperate for help often get abused. As The New York Times pointed out in a heartbreaking story last year, cities such as New York with a large homeless population have seen the growth of three-quarters houses, which cram multiple people into one bedroom while purporting to help them. Often, they’re just collecting these peoples’ money without giving them any services or even a clean place to live.

Not every homeless person will thrive just because they have a place to live. Some have mental or physical problems that make it difficult for them to stay off the streets after getting a home. Others may never be able to support themselves completely without a community to keep them afloat. Jeffrey Nemetsky, who runs Brooklyn Community Housing and Services, says having a social worker knock on the door once a day to say hello can mean the difference between a tenant staying or heading back out onto the streets.


So that’s the answer: provide the homeless with permanent, affordable housing, and wraparound services. Permanent supportive housing might solve this for us.


True, permanent supportive housing can be very effective at helping the chronically homeless get off the streets and stay stable. But is it legal?

Many people who need permanent supportive housing are battling mental problems or drug and alcohol abuse, and would have once ended up in institutions. But since the deinstitutionalization movement began in the 1960s and ’70s, the number of in-patient beds at state or county mental health facilities has declined from more than 400,000 to fewer than 100,000.

Though some institutions still exist, the practice of putting the mentally ill into segregated buildings falls into a gray area. In 1999, the Supreme Court ruled that the segregation of people with disabilities violated the Americans with Disabilities Act. Though the case was about people on Medicaid, homeless advocates interpreted it to apply to some chronically homeless with disabilities. Isolating those homeless people from the rest of society is akin to institutionalizing them, advocates say, and it violates the law.

That’s why some housing developments provide both permanent supportive housing and low-income housing, so that homes can be made available to a larger swath of the population. This kind of mixed-use housing helps create communities; in one building in Harlem, single moms living in affordable housing helped out the ex-cons living in supportive housing, and vice versa. Though the building’s developers worried that low-income moms wouldn’t want to live with the mentally ill, some 2,000 people applied for just a few dozen units when the building opened. This and other experiences suggest that integrating supportive and low-income housing can be successful.

But still, agencies and advocates all over the country are struggling to serve the homeless people with mental illness and addiction. It often takes years for case workers to get people to try out permanent supportive housing and abandon the lives they’ve known on the streets. Some cities and states have started allowing judges to order people who cycle through the system to receive treatment for their illnesses, an approach that’s controversial.

* * *

Across the country, experts on homelessness have solutions they think will work best. The problem is, housing in many cities is getting more expensive every month, and as prices rise, so do the costs of programs to combat homelessness.

Meanwhile, federal funds for affordable housing have stayed at the same levels for years. So as housing costs go up, those funds are spread more thinly and help fewer people.

But if homelessness is really a problem the country wants to solve by 2020, why not increase the amount of money overall that the government spends on programs to help the homeless? Where could that money come from?

Why not stimulate the creation of affordable housing so to assist both the chronically homeless and those who are homeless temporarily?  Such housing could be available to people below certain income levels, and they could qualify whether they are on the streets or are in an apartment they can’t afford.

For some, it’s hard to imagine carving out more money from the country’s budget to address these issues. But solving homelessness can help fix a lot of other problems too, including truancy from schools, food insecurity, drug and alcohol abuse, unemployment. Is it possible that directing more resources toward solving homelessness could actually save society money by helping to fix its other ills at the same time?

Maybe there’s an answer we haven’t considered yet. Drop your thoughts into an email to