Homeless Crisis: How to solve a problem we can’t quantify?

Denver’s failed Issue 300 to overturn a ban on “urban camping” has galvanized non-profits, developers, investors, service providers, and advocates to find better ways to address the homeless problem in the Mile High City. I opposed Issue 300 because it didn’t address the issue of homelessness while ceding unlimited public space for the problem to grow, but frankly struggled to explain policy technicalities to friends or acquaintances whenever it came up. Within the broader umbrella of the planning field, I have specialized in housing and community development ever since my AmeriCorps service 7 years ago, but find the homelessness issue more vexing to explain than any other subject matter.

Housing as a Continuum


Above all, housing is a continuum. This means that everything from homeless housing to market-rate housing are connected. We’re all humans with similar needs, and conceptually-speaking, home-based solutions for those need has always been proven cheaper, more efficient, with better outcomes. If you can keep a senior in her accessible apartment for 10 years longer than she would need a nursing/assisted living facility without that accessibility, you could save the Medicaid program a million dollars and give that senior a better life according to the Kaiser Foundation. Similarly, if you can house a homeless person, you give them a home in which to deal with their issues, which they would otherwise have to do on the streets, jail, or the hospital ER. If you think homeless housing is expensive, wait till you see the bill for the social service system. Housing this person costs 1/53rd the cost of the hospital ER, or 1/18th the cost of a mental healthcare facility. Life on the streets, in tent cities, or encampments often includes frequent hospital visits, re-incarceration, and other bad human outcomes.

Despite the headline-grabbing growth in the problem, there is much hope due to the innovation and dedication most cities put toward this issue as the growing economy has created a nationwide housing crisis. Wherever you’re from, your city is likely discussing its own housing crisis as I type. To be clear, Denver is deploying more tools to combat the scourge of homelessness than most cities, but there’s no doubt that many cities offer great case studies and innovations right now.

All eyes are on city hall(s) right now. After strong national leadership from HUD during the Obama administration, it seems plausible that municipalities themselves will have to fill the void of research, coordination and innovation in this space for a while. Even more confusing, HUD frankly uses its power of purview over Participating Jurisdictions (PJs, or cities that recent direct funding from HUD) to embarrass progressive leadership or inhibit cities’ ability to upgrade/improve/resolve projects or programs with historic HUD funding. Such was likely the case when HUD withheld $80 million in CDBG and HOME funds from LA, citing widespread monitoring and compliance issues that probably exist in every PJ. Ask yourself how that helps the city with the nation’s fastest-growing homeless crisis, reportedly growing at 16% last year, but likely much more than that. In fact, homeless counts have always been notoriously inaccurate, often due specifically to mandates from HUD.

Data is Power

One thing a strong, advocate-driven HUD can do is bring together academic and analytic capacities to tackle data gaps. Right now, homeless counts are done through HUD’s Continuum of Care (CoC) network, with a coordinating agency in charge of each regional CoC Point in Time count. While it’s great to have a system in place for Point in Time counts, there are many flaws with the present system of collecting data beginning with the date on which the count occurs, in January for almost all cities, every year. Furthermore, street canvassing is always difficult and imperfect.

In 4/5ths of the nation, the count of people living on the streets will be significantly impacted by the dead of winter. As obvious as it seems, I can’t fully explain how the count would be impacted, except that it is. I look forward to volunteering in the upcoming PIT count and learning more about this first-hand.

Theoretically, the majority of homeless counted on that night are going to be in dedicated homeless shelters, where occupancy rises during inhospitable weather. The urban camping ban in Denver and many other cities is driven by the fact that when able, many homeless actually prefer the streets over the shelter. As mind-blowing as that may seem to most people, the reasons for this include the following:

  • Some shelters (but certainly not all) are described as crime dens where criminals prey on weaker homeless people
  • Some faith-based organizations (also certainly not all) require folks to go through sermons/life classes/etc in order to receive shelter and food for the night; these sermons can seem extremely prejudiced against the mental health, social, and addiction challenges that many homeless face
  • Shelters do not allow dogs because they are not dog pounds; many homeless have dogs
  • Most shelters do not allow unmarried couples to stay together and separate homeless by sex
  • Many Continuum of Care regions lack dedicated shelter space for trans people or non-conforming genders, of which the homeless community is disproportionately comprised
  • Shelters usually kick people out during the day, forcing them to go somewhere during the day, and moving them around the city based on shelter capacity or the services someone needs
  • The shelters are heavily overcrowded, combined with building code issues, making them a “deathtrap”

Those are just the reasons I can conceptualize for someone preferring the streets over a shelter, which is important to understand because you then realize solving this is not just an issue of shelter capacity. Also important, is realizing that I can’t really, fully conceptualize being homeless, and try as we might none of us who’ve never had an experience of homelessness will be able to make these people whole. In forming an action plan, you have to be honest and realistic toward what you can promise. I think many non-establishment homeless advocates get adrift from solutions in focusing on intrinsic dignity, being valued, and made whole – possibly from reparations. While local government can never solve all of the problems these individuals face, just understanding how they feel about these “solutions” for their plight is even more important than it obviously sounds.

As any plan must start with the homeless population, all of these programs and projects start with the official Point in Time count of the homeless population they intend to serve. In Denver, planning a program or project first starts with the 3,445 homeless “households” you intend to serve. The obvious problem with that everyone knows there are more than 3,445 homeless on the streets of Denver during the day.

When accounting for the 1,000+ units of permanent supportive housing built in Denver in the last two decades, it is difficult for a homeless project to boast an attractive “capture rate” to get funded through tax credits, grants, and other funds. In order to get funded, projects typically need capture rates below 25% which means that the need/demand for your project exceeds your capacity by at least 4-to-1. This makes sense in solving affordable housing challenges where the demographic has other options, but doesn’t help homeless efforts for which the demographic has no other option for housing. If there were more accurate PIT counts, the numbers would make it easier to get these projects funded (albeit at the expense of other affordable housing initiatives like workforce housing).

This gets me back to the data challenges that are going unsolved right now. This is no knock on the current system, which must be maintained to support the current level of action. Housing officials working on these challenges know all about the flawed data, the spiraling growth of the problem, and the inherent challenges in scaling up solutions to serve the human problems of every individual. We attack housing providers for providing too much focus and care to 40-50 people at a time and failing to serve the thousands; simultaneously, we attack housing policy makers for the utilitarianism of massive shelter and service facilities.

I say to all the advocates out there, instead of attacking city hall and developers and the system, work together to come up with a better count of the local homeless population. It’s not a silver bullet, and doesn’t come with resource assurances, but it is as good a starting point as any. City halls have been attacked to death over this issue from all sides, when they’re the only entities even trying to solve homelessness. Maybe the issue is that they can’t do it alone. Maybe LA is deplorable for allowing homelessness to spiral out of control, or maybe LA is actually a national model for supportive housing. Same goes for Seattle, San Francisco, Denver, Portland, and all big cities. HUD is the one big player in all of this that exists above the fray, shielded from public outrage. All of these cities are now bringing more dollars to the table than the federal department that exists to solve housing crises (plural), but still must conform to HUD’s prescriptive mandates that paralyze projects that get federal funding that no city can afford to decline.

HUD trusts academics any policy “thinktanks” more than it trusts practitioners such as city officials, program/providers, and developers. I know it may seem obtuse to focus on statistics when there are life-and-death needs not being met, but there is no other way to hold HUD accountable than through data and policy white papers. City halls, on the other hand, have always been hyper-reactive to conditions and experiences on the ground. 


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s