Adi Talwar

Jamie Powlovich, the executive director of the Coalition for Homeless Youth, is concerned that a new system for prioritizing people for supportive housing will work against younger applicants.

One fell asleep several times during the interview. Another reported daily hallucinations involving his deceased father’s voice. One was “extremely confrontational,” while another was just “very guarded.” The applicant with a history of “noncompliance” denied having a mental illness or substance-abuse problem and one of his fellow applicants spoke “of an electric machine that is trying to harm him.”

“Client blames people and the environment for his anger management issues,” wrote one interviewer. Another client, “describes her mental illness as a ‘pain in her mind’ and states that when she gets upset the muscles in her head sweat and swell.”

“Client is out of touch with reality, has a severe thought disorder and memory loss,” was yet another verdict. “Client appears incapable of living alone.”

Every year, more than 1,000 homeless people in New York City are placed in supportive housing, which combines an affordable apartment with services for dealing with mental illness or substance-abuse problems. Once there, statistics indicate, a person is very likely to stabilize and less likely to return to the streets, making supportive housing “the most successful existing model for ending chronic homelessness among vulnerable populations,” according to a recent City Council report.

But even with 32,000 units of supportive housing operating in the city and more coming online each year, demand far outstrips supply. It’s estimated there are four potential new residents for every opening in supportive housing. In some ways, the program is a victim of its own success—residents become so stable they do not move out, meaning there’s little turnover in beds.

Now there’s new scrutiny of how supportive housing providers—the nonprofits that contract with the city to operate supportive housing—decide who gets a bed.

Prompted by new federal regulations, New York is rolling out a new assessment system meant to identify the people who need supportive housing the most. But some advocates believe supportive housing providers, hoping to conserve resources, are “creaming” or “screening out” harder-to-house people—disqualifying applicants based on their personal appearance, or attitude, or the very symptoms that make them a candidate for support in the first place. The quotes above, from city records of applicants who were rejected for supportive housing, amplify those concerns.

These advocates are pushing a City Council bill, Intro. 147, that would track and make public detailed statistics about providers, applicants and rejections. Providers say the reporting mandate would create unnecessary burdens and foster an overly simplistic critique of their selection process.

Both allies and critics of the supportive housing providers agree on some of the facts in the background: There is too little supportive housing, it’s being run with too few resources and it’s being asked to handle a very broad swath of the people who fall through society’s cracks.

“As a city, we’ve put all our eggs in the supportive-housing basket,” says Jamie Powlovich, the executive director of the Coalition for Homeless Youth. “A lot of pressure is being put on supportive housing to be the be all and end all solution to homelessness.”

* * * *

Supportive housing dates back to the 1970s. It first became a major part of the city’s response to homelessness in 1990 when the city and state inked the first of three agreements to jointly create thousands of units of such housing. More recently, advocates hoped Mayor de Blasio and Gov. Cuomo would team up for a fourth round; instead, the city and state are pursuing separate construction plans. The mayor pledged 15,000 units over 15 years. The governor has spoken of a 15-year target of 20,000 apartments, and has funded 6,000 over the first five years.

Supportive housing is sometimes confused with homeless shelters, and as a result, newly proposed supportive housing facilities are occasionally subjected to the same ugly resistance that proposed homeless shelters often receive. In fact, they are very different housing resources. Shelters are generally meant to be temporary housing and can be pretty bare-boned. Supportive housing is usually supposed to be permanent, and involves not just a room but counseling and other services. While different, the systems do interact, with shelter staff members helping clients apply for supportive housing apartments. Street outreach workers and people who work with runaway youth can also help clients apply for supportive housing.

Around the country, the way supportive housing is selecting tenants is changing. In 2010, President Obama set deadlines for ending veteran and chronic homelessness (by 2015) and family and youth homelessness (by 2020), and—partly as a reaction to those goals—the number of supportive housing units for the chronically homeless more than doubled between 2007 and 2015.

But most of those new units did not actually go to the chronically homeless, according to federal officials. So, in 2015, HUD set new rules for identifying the most needy people and putting them first in line for units. “HUD’s experience,” the agency wrote in a guidance document, “has shown that many communities and recipients of [federal funding] continue to serve persons on a ‘first-come, first-serve’ basis or based on tenant selection processes that screen-in those who are most likely to succeed while screening out those with the highest level of need.”

The new federal mandate is taking shape in New York as CAPS, or the Coordinated Assessment and Placement System. Eventually, this will be applied broadly to homeless programs here, but right now it is being used primarily to manage entry to city- and federally-funded supportive housing.

“The goal is to loop in every single piece of the continuum of care,” says Catherine Trapani, the executive director of Homeless Services United, referring to the system of homeless services. “The purpose of coordinated entry is ‘no wrong door.'”

CAPS introduces two new elements to the process of getting supportive housing. One is an intake survey to identify all housing options for which a homeless person is possibly eligible. The other, for those referred to supportive housing and found eligible, is a standardized vulnerability assessment that aims to triage applicants according to their degree of need, based on a person’s interaction with public systems—Medicaid usage, crisis-shelter stays, jail stints and the like—as well as functional limitations and length of time homeless. The city estimates CAPS will be fully implemented in about a year and a half.

Proponents of CAPS say the system is designed to evolve as lessons are learned. In fact, it has already done so. A New York state law strictly limits sharing of identifying information about runaways, but when it was launched earlier this year, CAPS was pulling in that kind of data from the city’s youth shelter system. Advocates say after they alerted the city to the potential violation, the city moved to change the system.

So far, so good. However, people who work directly with clients who apply for supportive housing aren’t sure CAPS can evolve sufficiently, or do so fast enough, to address deeper concerns about how people are placed in supportive housing—in particular, the discretion that supportive housing providers have over which applicants to accept.

“It’s not as nimble as I would like it to be,” says Trapani. “History has shown us that the system is not really working. Nothing in CAPS is set up to prevent [screening] from happening. I hope that if we see an issue on the placement side that we’ll be able to respond to it. It’s not clear that CAPS can do it.”

* * * *

At a City Council hearing this spring, advocates presented data obtained from the city via the Freedom of Information Law that indicated the outcome of applications to supportive housing from 2013 through 2016. According to Craig Hughes, a veteran social worker who testified, more than 15 percent of people applying for the largest category of supportive housing “were rejected because they were deemed by assessors, on the spot, to need a higher level of care, were struggling with too intense of a mental illness, were experiencing suicidal ideations, or lacked ‘insight’ into their mental illness.” In a handful of cases, people were rejected at least in part because they were “disheveled,” and in one case Hughes cited, because they were “malodorous.”

Joshua Goldfein, a Legal Aid attorney, says that data matches his own observations. “We’ve seen people turned away because of language issues or because they appeared disheveled at the interview,” he says. “It’s the housing agencies. They are acting as landlords and that’s what landlords do. They want to make their lives as easy as possible.”

Some find the “creaming” accusations puzzling. After all, everyone found eligible for supportive housing—by definition—has profound challenges. “All applicants screened and determined eligible by HRA are homeless with serious mental health, medical or substance use disorders,” says Lourdes Centeno, a spokesperson for the Human Resources Administration.

On the other hand, some allies of supportive-housing providers say there’s a difference between “creaming” and exercising discretion. Housing providers need to have some choice over who they admit to their programs, the allies say, to make sure the placement is a good fit, for that applicant and for their other residents. Among the clients whose rejection is detailed in the FOILed data is one who allegedly minimized his “kidnap[ping] of last roommate a year ago” and another who “has [a] history of homicidal ideations from unmanaged anger but is closed to treatment to address this issue.” Concern for other tenants’ well-being could make it hard for housing providers to invite either of those applicants to move in.

The need to make tough choices in who they admit, supportive housing providers say, is driven by the resource constraints they face. Not only are there not enough units of supportive housing to meet demand, the funding for the services providers must offer is inadequate. “It’s becoming increasingly unviable for [supportive-housing] providers to take contracts,” says Giselle Routhier, the policy analyst at the Coalition for the Homeless.

Supportive housing providers also say that the decrease over many years in state-financed licensed mental health beds—which offer a higher level of care than supportive housing—has forced supportive housing to try to serve a clientele it is not equipped to help. “What is not generally understood is that there has been a disinvestment by the state in various types of licensed housing that offers a higher level of care than supportive housing,” says Toni Lasicki, the executive director of the Association for Community Living. And there has been similar budget stagnation in support services for lower-level supportive housing*. “Because increases in financing have been too little over the years, [state-financed mental-health] housing programs have lost up to 70 percent of revenue to inflation over the last 25 years.”

Lasicki adds: “Providers have no choice but to reduce staff while turnover and staff vacancies are high, making admission decisions a risk-balancing exercise. Taking the most difficult to serve in every case is not possible given the resources they have.”

City officials insist they would not refer people who needed more serious care to supportive housing, but say it is possible that the condition of someone referred to supportive housing could, during the period when they are awaiting placement, worsen enough that they are no longer suited to it by the time they get to an interview.

Advocates for tenants do not deny that supportive housing providers face overwhelming demand, a deeply needy population and insufficient resources, but they dispute the notion that those challenges justify screening out. After all, it’s not like people rejected from supportive housing end up with the more intense care they need; instead, they end up with no care at all, in shelters or on streets that are even less equipped to handle them. “If the housing that is meant to serve the most vulnerable can’t house the most vulnerable,” asks Trapani, “then, they go where?”

What’s more, advocates say, screening out runs contrary to the “housing first” model to which the city subscribes. “The city has adopted the premise that where people have stable housing with support that they can access, they’re going to be better positioned to deal with other challenges they’ve been facing,” Sandra Gresl, a staff attorney at Mobilization for Justice, says.

* * * *

Once the vulnerability assessment is fully operational, CAPS will refer three applicants deemed “highly vulnerable” to a provider with an open slot in supportive housing. The providers will pick one and offer them the spot. Whether there will still be room in that system for any screening—in other words, whether providers will find any reason to distinguish among three highly-vulnerable referrals—is an open question.

But some advocates worry about the assessment itself—in particular, whether contact with jails or hospitals really the best indicator of who is more or less vulnerable. “We’re really afraid that the way they are assessing people, especially for young people, isn’t finding the people who most need it,” says Powlovich. “My argument would be that the person most disconnected from services is more vulnerable,” she adds, although she notes that a youth-specific assessment tool might eventually be added to the city’s toolkit.

A particular concern is that applicants have to document that they are chronically homeless by the federal definition of the term, and that can be difficult to do. Routhier from the Coalition for the Homeless says her agency’s clients have been having a hard time being found eligible for supportive housing at all, let alone accepted by a provider. “There should be a level of flexibility acknowledging the vulnerability index is new and there might be some errors in it,” she says.

Intro. 147, which was sponsored by Brooklyn Councilmember Stephen Levin, would not directly address any of CAPS’s supposed shortcomings. It would merely make public the kind of data that could indicate where problems lie, by annually collecting the numbers of people found eligible for supportive housing, interviewed for it, and accepted, broken down by age, gender, the supportive-housing program category they fit into, the agencies they dealt with and how long they have been homeless.

Supportive housing providers say the bill would mean more work for their already overstretched staff. What’s more, the say, CAPS will eventually provide this same kind of data. And besides, they argue, in the complex world of supportive housing, do simple up or down numbers really tell the story?

“I am encouraged that with the implementation of CAPS, the city is moving toward a comprehensive picture of exactly who the most vulnerable homeless people in New York are, what issues they struggle with, and what housing settings are needed to meet their needs,” says Laura Mascuch, executive director of the Supportive Housing Network of New York, and co-chair of the Continuum of Care, a public-private advisory body that helps guide homeless services.

Notably, HRA neither accepts nor denies that screening out is a problem. The agency sees CAPS as a way to detect whether it is or isn’t. “The city consistently focuses on making sure that these vulnerable clients can access the appropriate housing that they need,” says Centeno. “The new vulnerability index and CAPS system … will allow the city to better understand if there is a systemic problem with rejections and to address it.”

One clear downside to CAPS, however, is its results might not be made public. City officials told City Limits they would consider generating some form of reporting from the CAPS system, but as of now, there is no requirement for them to do so.

* Clarification: This sentence was added to clarify that the 70 percent figure applies to non-licensed housing. The parallel figure for reduction in licensed housing is, according to Lasicki, 45 percent.