Nan Roman’s Keynote Address from the 2020 Solutions for Individual Homeless Adults Conference

Welcome to the National Alliance to End Homelessness Conference on Solutions for Individual Homeless Adults.

We acknowledge the traditional owners of the land, the Ohlone group, and offer recognition and respect to the indigenous people past and present on whose land we meet today.  We thank the tribes of the Ohlone group, and the over 200 Native American tribes whose home is now California. 

Thank you all so much for being here – 1,200 of you from around the country — and nearly half from here in California.  Clearly, we are all looking for better ways to help individuals experiencing homelessness, so I know it will be great to spend the next few days learning from each other and figuring out how we can make more of a difference.

I have a lot I want to cover today: causes of individual homelessness; what the numbers look like; the emerging trends in the population; what is working, and what is not; what messages we could be using to get the public behind our efforts; and are there any opportunities to change the dynamics around homelessness, altogether.

To begin, what causes homelessness among individuals?

Let me start by saying that homelessness is not the fault of homeless people. And homelessness is not the fault of the homeless system. We could do better, but you are getting more and more effective and efficient by using data and experience to improve your work.

Homelessness is caused by structural economic and social factors and the decisions of public and elected officials. 

The major driver in homelessness is the mismatch between the cost of housing and what people earn. Rents are rising much faster than incomes. Here in California, a recent McKinsey report said that real estate prices have risen three times faster than household incomes and that more than half of the state’s households cannot afford the cost of housing. As the rental housing crisis, nationally, pushes farther up the income ladder, the competition for less expensive units becomes greater, and housing available to the lowest income people gets more and more scarce.

There is a racial dimension to this. The Joint Center for Housing Studies at Harvard points out that black renters had the highest rent burden rate in 2019 with 55 percent paying more than 30 percent of their income for housing. In contrast the cost-burdened share of white renter households was 43 percent. Perhaps not surprisingly, then, poor Black renters are almost twice as likely as poor White renters to have been evicted.  

On the income side, wages for poor people are going up at a much slower rate than rents. There is a racial dimension to this as well, as a median income black household earned just 59 cents for every dollar that a median income white household earned. 

Additional causal factors of homelessness are the behavioral and physical health problems of individuals, and the lack of treatment for them. People with disabilities are disproportionately poor and discriminated against. Of course, health problems are both a cause and a result of homelessness.   

Racial inequality overall is a major factor in homelessness. We know that African Americans in particular are disproportionately homeless. A recent paper by Dr. Margot Kushel found that criminal justice discrimination, employment discrimination, exposure to violence, premature death, and limited family wealth were the key factors of structural racism that affected people’s homelessness. It seems to me that knowing the things that cause homelessness should influence our understanding of what we need to do to help people end their homelessness. People with different causal experiences based on race might need different solutions based on race. 

At present many communities are beginning the process of examining their homelessness data for racial disparities. Usually this is in terms of who gets admitted to programs, how long they stay, who is placed in housing, and how many return to homelessness. Perhaps we should also be looking at the institutional factors that cause homelessness for clues as to what might help end it for people. Are we addressing, for example, the fact that poor mental health treatment for African Americans could result in them not identifying as having mental health issues during coordinated assessment, and therefore receiving low vulnerability scores and improper assignment to solutions? We are at the beginning of this critically important work, and there is much more to be done. 

I should also add that people with disabilities, LGBTQ people, and Native Americans, Alaska Natives and Native Hawaiians are disproportionately affected by these issues, and therefore disproportionately homeless – and possibly receiving disparate treatment in the homeless system.

It is these structural, social and political issues that drive homelessness. 

How are we doing on solving the problem of homelessness among individuals? Are the numbers going up, or down? 

The major source of information about trends is the Point in Time (PIT) count and the PIT has been in for a lot of criticism lately. Let me just say that while the PIT count is not perfect, it is very useful if we recognize what it does and does not do.

The PIT does not count of every single homeless person. It is not the best way to identify people’s characteristics. Jurisdictions don’t all use the same methodology, and sometimes they change their methodology from one year to the next. It only tells us how many people are homeless on a night, not the full scope of the problem and how many people are affected. 

So, it is not perfect. However, no count is perfect. And we have other data, such as from Homeless Management Information Systems (HMIS), to help us address those shortcomings.

What the PIT count is useful for is helping us see trends. What does the 2019 PIT count tell? It tells us that homeless individual adults on their own are the majority of all homeless people — 70 percent — and that the number of individuals went up last year (by 6 percent). Half of all homeless individuals are unsheltered. The typical homeless individual is a White male, over 25 years old. About 30 percent are women. A third of individuals are African American, and 19 percent are Hispanic or Latinx.

For individuals who are chronically homeless, veterans, and increasingly youth, there are specific strategies, approaches or programs to end their homelessness. The approximately half of individual homeless adults who are not in these categories really get whatever is left over after others are helped. And that is often not very much. Which is undoubtedly one reason their number is growing. So, we need to figure out a clear path for individuals, since getting what’s left over does not seem to be working.  There needs to be a solution for every person who experiences homelessness.

There are some important trends that we are seeing in the individual homeless population that we need to think about as we work to do better.  First is the increasing number of individuals who are entirely unsheltered. This number had been trending downward, but starting in 2015 it began to go up, and has now returned to 2007 levels of nearly 200,000 people nationally.

The Administration says that the recent increases in unsheltered homelessness were caused by a “switch” from transitional housing to Housing First.  If this were true, it would be important to understand because if we did something wrong, we need to fix it.

The steady reduction in transitional beds began in 2010, but it did not spark an increase in unsheltered homelessness, which stayed pretty steady until 2015 when it started to go up slightly.  In fact, it wasn’t until 2019  that there were actually more unsheltered individuals than there were when transitional housing first began decreasing in 2010.  In other words, there doesn’t seem to be any correlation between the availability of transitional beds and an increased number of unsheltered people.  Unless you consider families and veterans, where there is a very strong correlation, but in the other direction.    

Much of the transitional housing that has shut down has been for families.  During the time that family transitional housing was going down, there was no increase in unsheltered family homelessness. In fact, unsheltered family homelessness went down nearly 40 percent.  The switch from transitional to Housing First approaches (mostly rapid rehousing) for families reduced unsheltered homelessness.  It didn’t increase it.

There is also the question of how unsheltered veteran homelessness was cut in half.  Was that accomplished by increasing transitional housing?  No.  The US Department of Veterans Affairs (VA) committed to Housing First, additional federal resources were invested primarily in Housing First approaches, and unsheltered homelessness among veterans was reduced by half.      

Now, unsheltered homelessness among individuals has recently increased.  And this is unacceptable.  But that increase doesn’t seem to have anything at all to do with the switch from transitional housing to Housing First.  Housing First actually seems to reduce unsheltered homelessness.  So, something else is happening.

Why is unsheltered homelessness trending up in some places like California?  It seems to be a combination of lack of affordable housing; low wages for the poorest people that don’t come close to keeping up with rising rents; an aging and more vulnerable population of very poor people who become detached from family, employment, etc.; in some places very few shelter beds coupled with high barrier policies in the shelters that do exist; and lack of behavioral health and health care.  Neither the homelessness system nor homeless people are responsible for increasing unsheltered homelessness.  The failure of public and elected officials to address people’s income, housing and health needs plus factors such as racism are the cause. 

What do unsheltered homeless individuals need? 

Last year Iain deJong of OrgCode gave the Alliance, USICH and HUD matched data on 32,000 unsheltered and 32,000 sheltered people from around the country.  This data is not a representative sample, it may be biased toward higher need people, and it is self-reported.  So, it is not terribly strong, but it is the most information we have had on the unsheltered population, and it is quite grim.

Unsheltered individuals are homeless on average just short of 3 years, versus less than a year for sheltered people.  The older they are the longer they are homeless, but there are very few unsheltered people over 65. 

Their health is terrible.  They were sick when they became homeless, with nearly half saying that they became homeless because of a physical health problem; nearly half reported a chronic health issue in their hearts, lungs or other major organ.  They were more than twice as likely as sheltered people to report that they had a physical disability and that this has limited their housing options.  They are 25 times more likely than sheltered people to be trimorbid, and 80 percent of unsheltered women are trimorbid. 

What should we be doing to reduce unsheltered homelessness? 

Jurisdictions with growing unsheltered populations and encampments are struggling with how to balance the need for more emergency assistance – shelter – with the need for more housing and services.  To get unsheltered people into housing we will need to reach them much faster with diversion or housing-focused outreach.  We will need to quickly increase the supply of shelter, but only as a waystation, not as a home.  Shelter will need to be a pipeline to housing, and it will have to be decent, low barrier, and housing-focused.

And then, because we don’t have enough of it, we need to be a lot more creative about housing.  Many people who are homeless – especially adults living on their own – are not going to be able to afford the rent on their own apartments.  The median rent for a one-bedroom in the US is over $1200/month.  Supplemental Security Income is less than $800/month.  Someone working full time for minimum wage earns $1300/month.  They are not going to be able to cover the rent on those incomes. We need to look at shared housing, renting rooms, single room occupancy housing, boarding house, group homes, anything that will help people exit homelessness and get a place to live. 

We do a lot of things to deal with unsheltered people that fall short of housing, but that we hope will work in the short run even though we have very little evidence on their outcomes.  Such strategies include sanctioned encampments, safe parking, command centers, tiny homes, cabins, RV lots, and campuses.  We also try various housing subsidy strategies that might speed up the process, again with little evidence as to their outcomes.  That is not to say that some don’t work, and the Alliance and the Benioff Initiative on Housing and Homelessness expect to be working together to quickly assess the impact of some of these strategies. We encourage you to use your data to do the same. 

The Administration has also proposed some strategies to address the problem of unsheltered homelessness. The White House Council of Economic Advisors’ State of Homelessness paper proposed that the answers to unsheltered homelessness are reducing zoning and code barriers to bring down the cost of housing development; aggressive policing and enforcement of public space laws to make it harder for people to be homeless; reducing use of Housing First and increasing use of transitional housing; and conditioning housing on compliance with services. Reducing the cost of development is a good idea, but it will take decades for that to trickle down to homeless people. We did aggressive policing and transitional housing for decades during which time the number of homeless people just increased; it was only when we switched to more housing-focused strategies that the number began to go down. For some people the carrot and stick approach can work, but for many it does not, and housing is a floor beneath which a country like the United States should not go.  We should not condition a life necessity on people’s behavior. Everyone needs a home.

So, I do not feel confident that these strategies will work. I must also note that the Administration’s latest budget did not propose to provide more housing to address the unsheltered crisis – which would certainly have helped.

Another trend that we need to be attending to is the aging of the homeless population. 

Dennis Culhane at University of Pennsylvania and colleagues have recently completed a study on how many homeless people will be older and have health needs in the coming years, and what will it cost. The sobering findings are that the population of older homeless people will double in the next five years and triple by 2030. This will be accompanied by significantly more health costs, particularly from nursing homes. The data show that housing, in particular, can significantly reduce shelter and health care costs for this group, with a housing subsidy creating per capita savings ranging from $8,000 for lower need people to $50,000 for higher need people. 

Those are some of the emerging trends.

Looking now at what we are doing about homelessness among individuals, what is working and what isn’t? 

Housing First has worked.  By Housing First, I mean getting every homeless individual into a home and connected to services they may need as fast as possible.  This includes permanent supportive housing, rapid rehousing, and rental assistance.

I want to be clear that Housing First is not Housing Only, as it has recently been characterized.  People can get services before, during and after being connected to a home.  Services are not mandated, nor is housing conditioned on participation in services. But services are a huge part of Housing First, and traditional Housing First programs like Pathways are among the most sophisticated in the country in their approach to services. 

I should mention that Housing First approaches include rapid re-housing, and individuals actually get only about a quarter of the rapid re-housing resources, even though they are 70 percent of the homeless population.  That is a shame.  The use of rapid rehousing money for families has resulted in a significant decrease in that population – down nearly 30 percent since its high in 2010. 

Rapid rehousing has also worked great for individuals in the VA’s Supportive Services for Veteran Families program, run by John Kuhn here with us today.  We should do more rapid rehousing for individuals and make sure it also connects them with any services they need. 

Newer strategies that are showing great promise are diversion and so-called problem-solving. The idea here is to get to people faster and use flexible approaches to solve their immediate problem. If successful, this helps them avoid longer stays in homelessness, which are debilitating, expensive, and make the problem worse.

There are some things that we are learning about what works for youth aged 18-24 and experiencing homelessness, especially now that the Youth Homeless Demonstration Act programs are kicking in. First, that rooting the work in a Housing First approach also works well for youth.  We have learned from the youth homelessness movement the importance of the consumer voice.  The youth programs are much more likely to include youth in planning and designing and implementing solutions, and in policy development.  The result is that strategies are more successful. Finally, a lesson for homelessness across the board is that partnership improve outcomes.  No one system has the resources or know-how to solve the whole problem. 

The bottom line then is that we should stick with what we know, from evidence and from data, and from your experience across the country, works to end homelessness for individual adults. This includes Housing First approaches, including permanent supportive housing, rapid re-housing, rental subsidies, etc. that return people to housing as soon as possible.  It includes flexible strategies, not one size fits all approaches. It includes services and service connections for those who need them.  And it includes rapid problem-solving responses at the front door.  And we agree with USICH that all programs for individuals experiencing homelessness should be trauma-informed.

I would just point out that trauma informed care is not actually treatment or services, but rather a way of delivering treatment or services.  It is defined by SAMHSA as the adoption of principles and practices that promote a culture of safety, empowerment and healing for clients.  To me, mandatory services, conditioning shelter on compliance with services, and aggressive policing of unsheltered people do not seem trauma-informed – although these have been promoted by the Administration.

What doesn’t work? 

For most people, transitional housing is neither necessary nor effective.  Conditioning shelter or housing on participation and success in services might work for some, but at the end of the day there has to be a home – not a jail cell or a mat outdoors – for every person.  Lots of rules and barriers in shelter does not work. 

I know that in many places, the public’s support for homeless people is waning.  The public wants the problem solved, and instead it sees more money being spent, and the problem getting worse.  We know that we do have solutions and are housing more and more people.  But we’re bailing a leaking boat as the number of people becoming homeless is more than we can house every day.  

So, when we say we know what works, people don’t always believe us; and who can blame them?  Possibly why the idea of pushing the problem behind a closed door – whether a campus or a jail – starts looking better and better to the public and to a lot of mayors.

Recently the National Low Income Housing Coalition and the Alliance hired Lake Research to do some focus groups for us.  I thought you might want to know some of the things we learned in the first ones, which were conducted in Chicago and Orlando.

People did understand the connection between homelessness and affordable housing – this was heartening because it has not always been the case.  While they definitely saw the problem as housing, they did not talk about it in terms of housing, but rather in terms of income.  So the “affordable housing crisis” and shortage of supply didn’t resonate.  They said that what people earn is not enough to afford the housing that’s out there.

While they felt that all of society had some responsibility, they felt that the government caused homelessness, and that the government should pay to fix it.

While they understood the primacy of housing, they also felt that mental illness and substance abuse disorders were a factor, and that it was important to recognize that. 

They felt that the current approaches must not be working because the problem is still there.  But they did not agree with the Administration that the answer was to make it less tolerable to be homeless, or to arrest homeless people.  They saw housing as the solution.  But they really did not see the “story” of how it would be solved, and we’re not doing a good job of communicating that.

The top tested message, the message they liked best, was, “People deserve a safe, secure and affordable home.” 

An observation of the researchers was that, as with other issues, cynicism about the government’s ability to solve a problem, as well as the politicization of problems, was more of a barrier to getting people to support our solutions than opposition to our solutions. This is why it is so tragic that the issue of homelessness – which has always been bipartisan – is increasingly becoming politicized in the Congress, and now between the Administration and local governments.

What would change the dynamics around homelessness altogether? 

I often observe that when I first started working on urban housing issues, there was not widespread homelessness in the US.  There was plenty of poverty.  There was a lot of racism.  There was mental illness, substance abuse, alcoholism, family dysfunction, criminal justice issues, poor outcomes from foster care. If anything, these things were worse.    And yet, there was not widespread homelessness. 

Why?  The reason is that there were more affordable and available homes than there were low income households that needed them.  No matter what a person’s characteristics, they could find and afford a place to live.  Today, for every 100 low income household, there are only 37 affordable and available homes.  The country is over 7 million units short of enough places for them to live.  If people today could afford the homes that are out there – notwithstanding all the other problems they might have – we would not have widespread homelessness.          

While it is a difficult moment in terms of homelessness, especially on the West Coast, there are a lot of reasons to be optimistic.  You are really improving your approaches to homelessness – getting to people faster, getting more and more people into housing, doing smarter service delivery.  Last year, 20 states saw increases in homelessness, but 30 states saw decreases.  You are on the right track.

To make progress, we need a short-term strategy, and a long-term strategy.

In the short run, we can keep trying to improve the underfunded emergency homeless assistance system.  In particular, we need to get to people a lot faster – as soon as they become homeless – and try to solve their problem, then.  We are doing an especially poor job of this with unsheltered individuals.  We need more flexible resources to solve problems.  In some places, like California, we need more shelter beds and fast.  And everywhere, we need to reduce shelter barriers and make shelters much more housing-focused, moving through shelter to housing faster with more creative and varied housing strategies.  These are things we can do.

In the long run, we are going to have to do something about the cost of housing. And for the first time in my memory, affordable housing is on the national agenda.  The majority of presidential candidates have housing and/or homelessness plans and positions, many of them quite ambitious. Other public and elected officials are prioritizing these issues as well.  We must continue to elevate the issue and take advantage of the moment.   And of course, not just on housing — living wages, adequate incomes for those who cannot work, and health and behavioral health care for all are also targets we should shoot for.  Those are the real solutions.  We must also devote ourselves to these goals. 

How can we get there?  I want to ask you to do three things.

First, make sure your Members of Congress know you and your organization and come to you when they need information on the issue.  You can do this through site visits; sending the them your annual reports, invitations to events, and promotional materials; etc. 

Second, we ask you to sign up for Alliance Advocacy Alerts so that you can communicate with your Members on Congress on important votes – like for money.  Just go to  An alert will go out next week, and this website will make it so easy to let them know your views.  So please sign up for alerts.

Finally, we ask you to make sure that all your consumers are registered to vote.  We have information on how to do this at the front desk and on the conference app — prepared by our colleagues at the National Coalition for the Homeless.   Who is elected matters, all the more so for people who have to depend on public policy to meet their daily needs, including for housing. 

Thank you again for joining us here.  Doing a better job with individuals experiencing homelessness is critical.  We are honored that you are here, and to be able to work with you to find and scale up solutions.  Please enjoy the conference, and don’t hesitate to ask any Alliance staff if you need help or guidance of any kind.