By James Dysart, Associate Policy Director
The most recent Annual Homelessness Assessment Report to Congress (AHAR) raises important questions about data in the current landscape.
Although the AHAR estimates that the number of families experiencing homelessness in the United States fell by 11% from the prior year, there were still an estimated 230,366 people in families with children who experienced homelessness on a single night in January 2025.1 This estimate understates the true scale of the crisis.
The estimates for both sheltered and unsheltered people are based on the Point-in-Time (PIT) counts. Families, especially those who are unsheltered or staying with others, are particularly difficult to record in the PIT counts.2 Sheltered counts reflect the current shelter population, so they often demonstrate local system usage and, when paired with the Housing Inventory Count (HIC), capacity rather than overall need or how accessible shelter is to families. When counts miss unsheltered families, systems allocate fewer resources and limit capacity, which creates even greater barriers to shelter access.
For example, Continuums of Care (CoCs) in Massachusetts attributed the 7% decrease in the number of people in families with children to a reduction in shelter capacity, stricter eligibility rules, and length-of-stay limits.3 This observation reflects a change in how the need for and provision of shelter is determined in the state; it does not clearly demonstrate that the housing needs of families have been better met. In fact, a quick comparison of the AHAR estimate of the overall number of people in families experiencing homelessness and the total available family beds enumerated in the HIC shows that in 2025 and 2024, Massachusetts estimated a need for over 5,600 more family beds in order to be able to accommodate the total estimated number of homeless individuals in families with children.4
The PIT counts as summarized in the AHAR and other local publications can still be a useful tool for comparing information over time. However, they do not represent the entire scale of homelessness. That is an important distinction to make as we try to think about solutions that are meaningful to children, youth, and their caregivers. Whatever these trends might suggest about federal, state, or local policy, they also point back to a central concern: how we use, communicate, and interpret data.
In New York City, recent Department of Homeless Services (DHS) data show that family shelter use remains high while showing a gradual decrease in the total family census.5 Any reported stabilization or decline should be interpreted carefully. It may reflect changes in shelter use, management of asylum seekers, eligibility, reporting, or timing rather than a clear reduction in need.
Families have experienced a local environment shaped by persistent housing unaffordability, pressure on shelter capacity, and a stressed prevention system—all factors not unique to New York. These are concerns we hear as we talk to providers, advocates, and policy makers in all corners of the nation.
- U.S. Department of Housing and Urban Development. “The 2025 Annual Homelessness Assessment Report (AHAR) to Congress Part 1: Point-In-Time Estimates of Homelessness.” Huduser.gov. May 2026. pp. 29,39. https://www.huduser.gov/portal/sites/default/files/pdf/2025-AHAR-Part-1.pdf.
- U.S. Department of Housing and Urban Development. “The 2025 Annual Homelessness Assessment Report (AHAR).” May 2026. pp. iii,29.
- U.S. Department of Housing and Urban Development. “The 2025 Annual Homelessness Assessment Report (AHAR).” May 2026. pp. 37.
- U.S. Department of Housing and Urban Development. “2007 – 2025 Housing Inventory Count by CoC.” May 2026. https://www.huduser.gov/portal/datasets/ahar/2025-ahar-part-1-pit-estimates-of-homelessness-in-the-us.html.
- NYC OpenData. “DHS Daily Report.” Accessed June 3, 2026. https://data.cityofnewyork.us/Social-Services/DHS-Daily-Report/k46n-sa2m/about_data. Note: ICPH analysis of DHS data and trends.