An inaugural city report groups together shelter headcounts from five municipal systems, revealing a truer—and larger—picture of the homelessness crisis than the often-cited Department of Homeless Services count. 

New York City is one step closer to reporting an accurate shelter census, amid a homelessness crisis of staggering proportion.

A total of 89,951 people slept across five city-administered shelter systems on the last night of May, according to a first-of-its-kind report released this month by Mayor Eric Adams’ Office of Operations. The report answers the calls of advocates who for years pressed City Hall to overhaul how it publicly shares data on unhoused New Yorkers.

Still missing from the count are tens of thousands of asylum seekers currently staying in emergency facilities across the city, mostly run by the Office of Emergency Management (OEM) and the Health + Hospitals Corporation. All told, the city said it had 105,800 people in its care as of July 19, including asylum seekers and longer-term residents.

However, despite some emergent questions about methodology, advocates say the new census offers a more comprehensive picture than a daily Department of Homeless Services (DHS) report often cited by journalists and policymakers. That count was about 83,000 at the end of May.

“The report is a devastating index of human misery and suffering more than it is anything else,” said Craig Hughes, a social worker at MFJ Legal Services and advocate for an accurate shelter census.

Truer picture

Experts have cited a perfect storm of factors fueling New York City’s booming homelessness population, including the ongoing economic fallout of the coronavirus pandemic, the phasing out of associated eviction protections and a shortage of deeply affordable apartments, in addition to an influx of asylum seekers.

Required by 2022 legislation called Local Law 79, the new census adds up not only people sleeping in DHS shelters—asylum seekers among them—but thousands more in smaller shelters run by other city agencies. 

These include young people in Department of Youth and Community Development (DYCD) shelters; domestic violence survivors and people living with HIV/AIDS in shelters managed by the Human Resources Administration (HRA); and people displaced by flood and fire in Housing Preservation and Development (HPD) shelters. 

Comprehensive reporting is crucial for smaller shelter systems to get the resources they need, according to Gabriela Sandoval Requena, director of policy and communications for New Destiny Housing, an organization that builds housing for survivors of domestic violence. 

Certain city-financed apartments are set aside only for DHS shelter residents, for example, while those in the domestic violence shelter system are “consistently left out of special programs,” she said. 

Since early last year, City Limits has published its own more accurate shelter census effort—excluding headcounts from emergency migrant shelters—drawing from a predecessor monthly report published under Local Law 37 of 2011. 

But that report did not provide aggregate numbers spanning the five municipal shelter systems managed by DHS, HRA, HPD and DYCD, requiring extra work to combine data. 

City Limits has also tracked the number of unique—or “unduplicated”—individuals in these shelters each month, offering a different picture than a nightly census. April, the most recent month available, saw 95,033 people staying across these systems—up nearly 57 percent since Mayor Adams took office. 

The inaugural Local Law 79 report includes an unduplicated count of 97,318, though the descriptive language does not match the Local Law 37 reports. City Hall did not clarify if an apples-to-apples comparison was appropriate.

Exploring exits 

Local Law 79 also requires the city to start reporting how many people are exiting the various shelter systems each month and where they are going next—an important indicator of its ability to move homeless New Yorkers into stable housing. 

However, the underlying legislation compels agencies to report this information “if known”—language advocates fear has let the city off the hook. Numerous fields in the May report simply say “N/A,” meaning not applicable. 

“All those zeros and ‘N/As,’ it raises a bit of questions,” Sandoval Requena said. Ideally, she added, the City Council would hold a hearing so that the administration could walk through the new census methodology and make adjustments as needed. 

Reached for comment, an HPD spokesperson said that its exit data is based on self-reporting from people leaving its emergency shelters. The Department of Social Services, which encompasses DHS and HRA, said it reports all exits from its shelters, though it does not break them down by subsidy type, accounting for some blank fields. 

Perusing the exit numbers the city has provided so far, advocates drilled down on the number of people leaving DHS shelter for supportive housing, a count they described as paltry given the number of vacant apartments in that system. In supportive apartments, tenants can access special services such as counseling and health care referrals.

The DHS exit data—which lags one month for families and two months for single adults—shows that 135 single adults moved into supportive housing, along with 25 families with children and fewer than 10 adult families, during the May reporting period. 

“This data shows what happens when bureaucracy stands in the way of housing, and providers get unlimited discretion on who they will—and won’t—house, using primarily public funds,” stated Meg Floss, an advocate with the group Supportive Housing Organized and United Tenants, or SHOUT. 

As of July 17, DSS reported 2,384 vacant supportive housing apartments, including 1,825 that are available, 202 that are linked with someone in the process of moving in, and 357 new units poised to come online. 

A spokesperson for the department stressed that the exit data only covers one month, and that DSS’s supportive housing portfolio, with more than 30,000 units, has an occupancy rate over 90 percent. According to the city, the year ending in June saw a 20 percent increase in shelter residents moving into supportive housing from the year prior.  

Still excluded 

The City Council passed Local Law 79, sponsored by General Welfare Committee Chair Diana Ayala, in July of last year, after a similar measure floundered in 2021. At the time, then-DHS Commissioner Steve Banks said updating the census would have held former Mayor Bill de Blasio’s administration to a different standard than prior mayors, who’d been “measured historically” by the smaller DHS count.

An earlier version of Ayala’s bill was supposed to phase in over two years, but the ramp up was then shortened. 

Tens of thousands of asylum seekers have traveled to New York City from the southern border and other U.S. cities in the intervening months, prompting conversations about expanding the new census even further to cover emergency migrant shelters run by OEM, Health + Hospitals and HPD.

Brooklyn Councilmember Shahana Hanif, chair of the Immigration Committee, said that she and her colleagues have struggled to get consistent data on how asylum seekers are distributed across all shelter systems outside of occasional hearings, and that this data should be incorporated into Local Law 79 reports. 

“There’s no reason why there are carve-outs for some shelters,” she said. “Even if the administration is saying ‘Well, you know, this is a facility that will operate for a month,’ we still need that record. We need transparency.”

At a June 21 hearing, city officials reported about 13,000 migrants staying at Humanitarian Emergency Response and Relief Centers (HERRCs) run by Health + Hospitals and HPD, and another 3,000 at emergency respite centers.

Positive step 

Jamie Powlovich, executive director for the statewide Coalition for Homeless Youth,  helped advocate for the passage of Local Law 79. She acknowledged the data collection challenges, including some unique to the young people her members serve. 

For example, state privacy law prevents youth drop-in centers from sharing unduplicated data, meaning some youth could be counted more than once. 

Powlovich also noted that any figure less than 10 in the report is marked “<10,” without more specificity—an approach HPD said is also privacy related. “It’s very different if all of those are really 1 or if they’re 9, because they do add up,” Powlovich said. 

Still, she views the new census as a victory for unhoused New Yorkers and their allies.  

“Now that the report’s out we can recognize where we need to maybe make some tweaks or clarify methodology behind some of these numbers to ensure accuracy,” she said. “But regardless, even with the numbers produced in this first report, it’s definitely a more true representation of the scope of the population than we had before.”