VA rarely documented why eligible homeless veterans weren't referred
Summary
GAO's 2020 review of veteran homelessness programs was built on interviews at six sites -- local staff told GAO that case-manager shortages might be limiting how many veterans the largest program, HUD-VASH, could serve. Its 2026 follow-up put a number on it: from 2020 through 2024, VA didn't document why it hadn't referred an eligible veteran to HUD-VASH in 151,296 of 174,045 instances -- 87% of the time, even as case managers went unfilled at rates of 14-18% a year and turnover ran as high as 26%.
2020: a shortage flagged by interview, not by data
's 2020 report reviewed seven federal homeless-veteran programs, evaluating how well , , and the Department of Labor collaborated and measured performance. Its findings on staffing were qualitative: local staff and service providers at six sites visited said case-manager shortages might be limiting how many veterans -VASH -- the largest of the programs, pairing a housing voucher with case management -- could actually serve. didn't have the data at the time to say how large that shortage was or what it cost.
View data as table
| January 2009 | 73,367 |
|---|---|
| 2019 | 37,000 |
| January 2024 | 32,882 |
made three recommendations in 2020: two directing to give local staff clearer guidance on collaborating with community partners and understanding how their performance is measured, and one directing the Department of Labor to document its data-quality validation process for a separate homeless-veteran employment program. agreed with both of its recommendations. neither agreed nor disagreed with its recommendation -- and even after supplied additional information about a validation tool it used, found the response still didn't describe what the validation actually checked, and maintained the recommendation.
2026: the shortage, quantified
's 2026 follow-up went straight at -VASH's staffing data. From fiscal year 2020 through 2024, medical centers had 14% to 18% of -VASH case manager positions unfilled in any given year, even as the total number of authorized positions grew 9%, from 3,801 to 4,159. Annual case-manager turnover ran 20% to 26% over the same period. By the end of fiscal year 2024, more than one-quarter of medical centers with two or more case managers had at least 20% of those positions empty.
View data as table
| Authorized (FY2020) | 3,801 |
|---|---|
| Filled (FY2020) | 3,262 |
| Authorized (FY2024) | 4,159 |
| Filled (FY2024) | 3,575 |
The mismatch runs deeper than headcount. The number of -VASH vouchers in active use grew 12% over the same period, from 99,894 to 112,234 -- faster than case-manager hiring kept pace. calculated that nationally, each case manager was effectively responsible for 30 to 32 vouchers. has used hiring flexibilities to help close the gap, but found the department could better use its own staffing data to target where positions are needed most.
That's because still doesn't reliably track why a given veteran wasn't referred to the program in the first place. Of 174,045 such instances from 2020 through 2024, left the reason undocumented in 151,296 of them -- 87%, and more cases than the 112,234 vouchers the program had running nationwide by the end of the period. Without that documentation, can't tell how often insufficient case-management capacity, rather than some other factor, is the reason a veteran doesn't get in.
View data as table
| Reason documented | 22,749 |
|---|---|
| Reason undocumented | 151,296 |
A second, smaller program showed a related gap. 's Tribal -VASH pilot, launched in 2016 to serve American Indian and Alaska Native veterans, had served more than 1,100 veterans as of April 2025. found 's program design followed some leading practices -- it communicated with stakeholders throughout -- but had never clearly defined the pilot's objectives, how it would measure progress, or built a plan to evaluate it.
made two recommendations in 2026: should build a method for medical centers to document why eligible veterans weren't referred to -VASH, including when staffing capacity is the reason, and use that data to inform hiring; should define clear objectives and an evaluation plan for Tribal -VASH. agreed and outlined planned changes to its data systems. , like six years earlier, neither agreed nor disagreed -- though it said it would work with to build a more structured way to assess the pilot.
The takeaway
- A 2020 hunch became a 2026 statistic. 's first review flagged case-manager shortages as something local staff worried about. Its second measured it directly: 14-18% of positions unfilled and 20-26% annual turnover most years, plus 87% of non-referral decisions left unexplained.
- The paper trail is thinner than the program itself. 151,296 undocumented non-referral cases is a bigger number than the 112,234 vouchers -VASH had running nationwide at the end of FY2024 -- can't fully account for demand it isn't meeting.
- Both times, one agency agreed and another didn't quite. has agreed with every recommendation across both reports. in 2020 and in 2026 each neither agreed nor disagreed with theirs, while saying they'd act anyway.
The 2020 findings are from -20-428, 'Homeless Veterans: Opportunities Exist to Strengthen Interagency Collaboration and Performance Measurement Procedures' (May 14, 2020), read directly and in full via an archived copy. The 2026 findings are from -26-107517, 'Veteran Homelessness Programs: Opportunities to Improve Data Collection and Establish an Evaluation Plan' (March 30, 2026), also read directly and in full; it explicitly cites the 2020 report in its own footnotes. Both reports were authored under the direction of the same director. They examine distinct facets of the same - case-management system roughly six years apart -- the earlier report's staffing concern was an interview-based observation; the later report quantified it with 's own data.
Sources(2) ▾
- U.S. Government Accountability Office, Homeless Veterans: Opportunities Exist to Strengthen Interagency Collaboration and Performance Measurement Procedures (2020-05-14) — -20-428, a broad, qualitative review of // interagency collaboration and performance measurement across seven veteran-homelessness programs -- the pre-quantification facet, where case-manager shortages were a stakeholder-interview finding, not yet a measured statistic. Distinct from doc-gao26-107517 (the quantified, -VASH-specific facet), issued nearly six years later and explicitly cited by it. Direct gao.gov fetch returned HTTP 403; fetched via an existing Wayback capture. gao.gov · original document
- U.S. Government Accountability Office, Veteran Homelessness Programs: Opportunities to Improve Data Collection and Establish an Evaluation Plan (2026-03-30) — -26-107517, a report to congressional committees examining -VASH case-manager staffing and the Tribal -VASH pilot in depth -- the quantified facet, six years after doc-gao20-428's qualitative interview-based findings on the same staffing-shortage concern. Explicitly cites doc-gao20-428 in its own footnotes. Fetched directly from files.gao.gov (HTTP 200); a Wayback capture was found to already exist at read time. gao.gov · original document
Comments
Always open. Logged-in readers can annotate paragraphs in place.
The number of veterans experiencing homelessness on a single night has fallen for 15 straight years -- from 73,367 in January 2009 to 32,882 in January 2024, according to . Two reviews, six years apart, examine the program most responsible for that decline. In May 2020⧉, interviewed staff at six sites and reported that case-manager shortages 'may limit the number of veterans they are able to serve' -- an anecdotal finding, not a measured one. In March 2026⧉, quantified it: didn't document why it hadn't referred an eligible veteran to -VASH in 151,296 of 174,045 instances from 2020 through 2024 -- 87% of the time.