During the first 10 months of 2020, nonprofit providers scheduled nearly 3,000 interviews with supportive housing applicants, 560 of whom were outright rejected. The housing shortage allows providers to “cherry-pick” applicants who require the fewest services, a process known as creaming or screening, tenants and advocates say.

David Brand

Supportive Housing Organized and United Tenants (SHOUT) members at a press conference at City Hall earlier this month.

Read more land use and housing news.

Each year, thousands of New Yorkers living in shelters and on the streets apply for permanent supportive housing, a service model championed as a key solution to the city’s homelessness crisis. And each year, the majority of those applicants fail to secure a home.

It’s largely a supply problem. Most of New York City’s 35,000 supportive apartments—affordable units where tenants can also access support services, like counseling and mental health care referrals—are already occupied by tenants and, as nonprofit providers and their trade group point out, there are only so many available units to go around. But some New Yorkers experiencing homelessness, supportive housing residents and their advocates contend that there’s another factor when it comes to who gets selected: the housing shortage allows providers to “cherry-pick” applicants who require the fewest services, a process known as creaming or screening.

Department of Social Services (DSS) records, obtained through Freedom of Information Law (FOIL) requests by the advocacy group Safety Net Project, illustrate the charge. Supportive housing is, by definition, designed for people with mental illness, but on dozens of occasions over the first 10 months of 2020, providers cited an applicant’s “lack of insight” into their mental health needs as the reason for rejecting them.

“Client never lived independently or paid bills. Client had difficulties staying focus[ed] and had no insight in regards to his mental illness, and poor historian,” staff wrote in one April 2020 rejection.

An applicant “has no insight into her mental health diagnosis. At the interview she denied a diagnosis and medication,” wrote another provider in October 2020.

With so many applicants and so few units, rejections are bound to occur. It’s baked into the city’s new multi-agency placement system, which refers three people for each available unit and instructs the nonprofit supportive housing provider to select one.

A four-year-old piece of legislation before the City Council would enable more consistent scrutiny of those rejections, highlight trends in supportive housing placements and, advocates say, allow for closer examination of the gaps in the Coordinated Assessment and Placement System, known as CAPS. The bill’s fate remains up in the air as lame-duck lawmakers approach their final meetings and Mayor Bill de Blasio’s administration attempts to amend key parts of the measure.

The legislation, sponsored since 2018 by Councilmember Stephen Levin, would compel the city’s DSS to publish an annual report listing the number of New Yorkers referred to, accepted to, rejected from and awaiting placement in supportive housing. The report would also detail rejections, creating a database that would allow policymakers to identify and amend trends like creaming by specific providers or a lack of necessary services across the supportive housing sector, Levin said.

“It’s important to have transparency, but not just for transparency’s sake. We want to be able to amend policies based on what the data is showing,” he said. “We put a lot of responsibility onto supportive housing to take care of a lot of issues and I think it deserves a really comprehensive discussion.”

Mayors past, present and future have included supportive housing in their plans for providing affordable homes and services for people with mental illness, HIV/AIDS or other health needs. Policymakers have held up the city’s network of dozens of nonprofits with at least 86 different funding streams as a sort of catchall solution to the homelessness crisis, even if supportive housing isn’t the best fit for many New Yorkers who simply can’t afford a place, or others who need more intensive services.

There are about 19,000 apartments in buildings built or redeveloped specifically for supportive housing. Providers rent out another 16,000 so-called “scattered-site” units from private landlords, with services offered by visiting case managers and social workers. More supportive apartments are in the pipeline, and Mayor-elect Eric Adams has pledged to facilitate the creation of 25,000 additional supportive housing units inside converted hotels.

As it stands, just about 1-in-5 supportive housing applicants get an apartment, according to the Supportive Housing Network of New York (SHNNY) and the Coalition for the Homeless. A total of 1,035 people living in Department of Homeless Services (DHS) shelters moved into supportive housing during the first nine months of 2020, a Human Resources Administration (HRA) administrator testified at a Council hearing last year. Another HRA deputy commissioner said about 5,000 people are awaiting placement at any given time, meaning they have been deemed eligible for supportive housing, but not placed in an apartment

About 16,000 people have moved from shelters and other settings into supportive housing since de Blasio took office in 2014, according to DHS.

Intro. 147 faces little public opposition, but in internal emails obtained through FOIL and shared with City Limits, DSS officials and SHNNY administrators said the legislation would impose a reporting requirement on city-funded supportive housing that does not apply to state-funded units. In emails from 2017 and 2018, they also said the reporting would fail to capture the “nuance” in finding the right housing placement for an applicant. SHNNY said it is no longer involved with the bill and has not seen recent versions.

In recent weeks, DSS and the mayor’s office have attempted to water down the bill’s reporting requirements, allowing providers to choose a predetermined list of rejection reasons from a dropdown menu rather than specify their determination. The legislation’s supporters, including the Supportive Housing Organized and United Tenants (SHOUT), say that change is unacceptable. The group urged the Council to pull the bill after viewing the last-minute changes from the mayor’s office.

A Council spokesperson said the bill continues to go through the legislative process. And a spokesperson for de Blasio said the “administration is committed to making supportive housing data more transparent and accessible.”

But recent data is hard to come by. DSS and the mayor’s office said the acceptance and rejection figures for the past two years are only available through FOIL. For the past seven years, Safety Net Project staff have filed records requests for the numbers from DSS and shared them with City Limits.

The most recent response provides data that would be publicly reported under Intro. 147. During the first 10 months of 2020, nonprofit providers scheduled nearly 3,000 interviews with supportive housing applicants, including some returning for follow-up appointments, according to the DSS records. About 361 of those applicants were definitively accepted for an apartment, according to the data, while at least 340 applicants declined the units they interviewed for.

Another 560 people were outright rejected and at least 720 people were listed as a “no show,” meaning they did not attend their interview, the data shows. A final decision for hundreds of other applicants was still pending based on apartment availability, additional information or other factors, according to the records.

Craig Hughes, a social worker at the Safety Net Project, said his organization has used the data to flag examples of “clear disability discrimination by providers” and brought those to the attention of city agencies.

Nonprofit providers and city officials acknowledge that creaming does occur, but say finding the “right” placement for a tenant can be complicated. Often, they say, the level of services available at a particular site do not go far enough to meet an applicants’ needs. That raises another problem: New York state has cut funding to licensed mental health beds, which serve people who require more intensive services.

“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,” Association for Community Living Executive Director Toni Lasicki told City Limits in 2018, a few months after Intro. 147 was introduced.

Many supportive housing applicants were rejected after interviewing for units because the provider staff determined they needed a “higher level of care”—a phrase used 72* times to justify rejections in the first 10 months of 2020, according to DSS records.

Advocates say Intro. 147 would allow for a better examination of the disconnect between the HRA referrals and the services the nonprofit supportive housing provider offers. “The bill is trying to illuminate what comes out of a referral,” said Homeless Services United Executive Director Catherine Trapani.

She said the reporting requirement could also shed light on the effectiveness of the system that the city uses to make supportive housing referrals and move people out of shelters. CAPS is a federally-mandated system for identifying the most vulnerable New Yorkers and prioritizing them for limited housing and resources. The multi-step CAPS process allows a person experiencing homelessness, usually working with a case manager or social worker, to complete a standardized survey to determine their eligibility for supportive housing or other options, like rental assistance vouchers. If they are found potentially eligible for supportive housing, they complete an application which goes to a team of HRA staff members.

The HRA unit then determines the person or family’s final eligibility based on their application and a Standardized Vulnerability Assessment, which seeks to identify the New Yorkers most in need of housing and resources based on a number of factors, like their use Medicaid, history of incarceration and the amount of time they have been homeless.

HRA staff use this pool of applicants to make referrals to providers with vacant supportive housing units. The team typically refers “three similarly situated individuals for each vacancy” and allows the provider to “select the people that they feel are a good match for their program,” said Jennifer Kelly, deputy commissioner of the Office of Supportive Affordable Housing and Services, at a Council hearing in December 2020.

Trapani, a member of the Continuum of Care coalition that developed CAPS, said the better data reporting would highlight where the system is working well and what problems need to be addressed.

“It’s a way to spot gaps and needs and errors,” she said. “When data becomes available, we’ll notice what if there’s a lack of inventory for a specific housing type or person and need to focus specific supportive housing agreements on a specific population. Wouldn’t that be healthy?”

*Correction: A previous version of this article referred to a higher number of rejections that included instances from late December 2019 and early November 2020.

Read more on this issue: